v>EPA
EPA/600/R-22/042F | July 2023 | www.epa.gov/risk
United States
Environmental
Protection Agency
ORD Human Health Toxicity Value
for Perfluoropropanoic Acid
(CASRN 422-64-0 I DTXSID8059970)
Office of Research and Development
Center for Public Health and Environmental Assessment
-------
4F\ Unitad States
Environmental Protection
Lai •» Agency
EPA/600/R-22/042F
https://www.epa.gOY/
ORD Human Health Toxicity Value for
Perfluoropropanoic Acid
(CASRN 422-64-0 | DTXSID8059970)
July 2023
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
DISCLAIMERS
The human health assessment is intended to inform fit-for-purpose decision contexts
(e.g., site-specific screening, prioritization, and assessment) and was developed in response to a
request for site-specific technical support. The document provides hazard and dose-response
information about the adverse effect(s) of the chemical and derived toxicity value(s) based on the
available evidence, including the strengths and limitations of the available data. All users are
advised to review the information provided in this document to ensure that the value(s) used is
appropriate for the types of exposures and circumstances at the site in question and the risk
management decision(s) that would be supported by the assessment.
This human health assessment was developed by EPA's Office of Research and
Development (ORD) to respond to a request from the EPA Office of Enforcement and
Compliance Assurance (OECA), in support of site-specific decision-making under the purview
of the Safe Drinking Water Act. Other EPA programs or external parties who may choose to use
values derived in this fit-for-purpose assessment are advised that EPA resources will not
generally be used to respond to challenges, if any, pertaining to content/values used in a context
outside of the specific purpose or applicability domain for which it was derived.
This document has been reviewed in accordance with EPA policy and approved for
publication. The human health assessment has received internal peer review by at least two
EPA/ORD/CPHEA scientists and an independent, external peer review by at least three scientific
experts outside of EPA. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chemical Managers/Authors
Jason C. Lambert, PhD, DABT
Lucina Lizarraga, PhD
Elizabeth Oesterline Owens, PhD
U.S. EPA, Center for Computational
Toxicology and Exposure
U.S. EPA, Center for Public Health and
Environmental Assessment
U.S. EPA, Center for Public Health and
Environmental Assessment
Contributors
Avanti Shirke. MPH
Christine Cai, MS
Elizabeth G. Radke, PhD
Production Support
Dahnish Shams
Jessica Soto-Hernandez
Ryan Jones
Samuel Thacker
Vicki Soto
U.S. EPA, Center for Public Health and
Environmental Assessment
U.S. EPA, Center for Public Health and
Environmental Assessment
Lauren Johnson
Student Services Contractor, Oak Ridge
Associated Universities (ORAU)
Primary Internal Reviewers
Andrew Kraft, PhD
J. Allen Davis, MSPH
U.S. EPA, Center for Public Health and
Environmental Assessment
111
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Primary External Reviewers
Gloria B. Post, PhD, DABT
New Jersey Department of Environmental
Protection
Rutgers University
Panagiotis G. Georgopoulos, PhD
Penelope A. Rice, PhD, DABT
U.S. Food and Drug Administration
Executive Direction
Wayne Cascio, MD
V. Kay Holt
CPHEA Center Director
CPHEA Deputy Center Director
CPHEA Associate Director
CPHEA Chemical and Pollutant Assessment Division
Samantha Jones, PhD
Kristina Thayer, PhD
Director
Questions regarding the content of this assessment should be directed to the EPA Office
of Research and Development (ORD) Center for Public Health and Environmental Assessment
(CPHEA) website at https://ecomments.epa.gov/.
1 Dr. Post and Dr. Rice conducted this review as independent consultants and not as a representative of the New
Jersey Department of Environmental Protection and the U.S. Food and Drug Administration, respectively.
IV
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
TABLE OF CONTENTS
BACKGROUND 1
PFPrA Quality Assurance 1
INTRODUCTION 3
METHODS 5
RESULTS 6
Literature Search and Screening Results 6
Human Studies 6
Animal Studies 10
Other Data 15
DERIVATION 01 REFERENCE VALUES 16
Derivation of Oral Reference Dose 16
Derivation of Inhalation Reference Concentrations 21
Summary of Noncancer Reference Values 21
CARCINOGENICITY ASSESSMENT 22
REFERENCES 23
APPENDIX A. SYSTEMATIC LITERATURE SEARCH METHODS AND RESULTS 27
Methods 27
Populations, Exposures, Comparators, and Outcomes (PECO) Criteria and Supplemental
Material Tagging 27
Literature Search and Screening Strategies 27
Database Searches 28
Data Extraction of Study Methods and Findings 32
Study Evaluation 33
APPENDIX B. BENCHMARK DOSE MODELING RESULTS 38
Modeling Procedure for Continuous Noncancer Data 38
Modeling Procedure for Dichotomous Noncancer Data 38
Model Predictions for Increased Relative Liver Weight in Male Rats 39
Model Predictions for Increased Serum Alanine Aminotransferase (ALT) in Male Rats 42
Model Predictions for Increased Serum Alkaline Phosphatase (ALP) in Male Rats 45
Model Predictions for Increased Hepatocyte Hypertrophy in Male Rats 48
v
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
TABLES
Table 1. Physical and Chemical Properties of PFPrA 4
Table 2. Associations Between PFPrA Concentrations and Health Effects in Human
Epidemiological Studies 9
Table 3. Available Experimental Animal Oral Toxicity Data for PFPrA 14
Table 4. Summary of PFPrA Genotoxicity Studies 15
Table 5. Data for Liver Effects in Adult Male Cij :CD (SD) IGS (SPF) Rats Exposed to
PFPrA for 28 Days via Gavage 17
Table 6. Candidate PODs for Derivation of the Chronic RfD for PFPrA 18
Table 7. Uncertainty Factors for the Chronic RfD for PFPrA (CASRN 422-64-0) 20
Table 8. Confidence Descriptors for the Chronic RfD for PFPrA (CASRN 422-64-0) 21
Table 9. Summary of the Noncancer Reference Values for PFPrA (CASRN 422-64-0) 21
Table A-l. Populations, Exposures, Comparators, and Outcomes (PECO) Criteria 34
Table A-2. Major categories of "potentially relevant supplemental material" 36
Table B-l. BMD Modeling Results for Increased Relative Liver Weight in Adult Male
Cij :CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 39
Table B-2. BMD Modeling Results for Increased Serum ALT in Adult Male Crj :CD (SD)
IGS (SPF) Rats Exposed to PFPrA for 28 days via Gavage 42
Table B-3. BMD Modeling Results for Increased Serum ALP in Adult Male Cij :CD (SD)
IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 45
Table B-4. BMD Modeling Results for Increased Hepatocyte Hypertrophy in Adult Male
Cij :CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 48
FIGURES
Figure 1. PFPrA Literature Search Flow Diagram 6
Figure 2. Summary of Study Evaluation for Human Epidemiological Studies of PFPrA and
All Health Outcomes 10
Figure 3. Summary of Study Evaluation for Experimental Animal Toxicological Studies of
PFPrA and All Health Outcomes 11
Figure 4. Liver Weight Changes in Rats Following Oral Expose to PFPrA 12
Figure 5. Liver Histopathology in Rats Following Oral Expose to PFPrA 12
Figure B-l. Fit of Hill Model to Data for Increased Relative Liver Weight in Adult Male
Cij :CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 40
Figure B-2. Fit of Hill Model to Data for Incrased Serum ALT in Adult Male Crj :CD (SD)
IGS (SPF) Rats Exposed to PFPrA for 28 days via Gavage 43
Figure B-3. Fit of Exponential Degree 5 Model to Data for Increased Serum ALP in Adult
Male Cij :CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 46
Figure B-4. Fit of Multistage Degree 1 Model to Data for Increased Serum ALP in Adult
Male Cij :CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage 49
VI
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
COMMONLY USED ABBREVIATIONS AND ACRONYMS
AIC
Akaike's information criterion
ALP
alkaline phosphatase
OECD
Organization for Economic
ALT
alanine aminotransferase
Cooperation and Development
AST
aspartate aminotransferase
ORD
Office of Research and Development
BMCL
benchmark concentration lower
PBPK
physiologically based
confidence limit
pharmacokinetic
BMD
benchmark dose
PECO
populations, exposures, comparators,
BMDL
benchmark dose lower confidence limit
and outcomes
BMDS
Benchmark Dose Software
PFAS
per- and polyfluoroalkyl substances
BMR
benchmark response
PFOA
perfluorooctanoic acid
BW
body weight
PFOS
perfluorooctane sulfonic acid
CASRN
Chemical Abstracts Service registry
PFPrA
perfluoropropanoic acid
number
POD
point of departure
CBI
nonconfidential business information
PODhed
human equivalent
CERI
Chemicals Evaluation and Research
POD
Institute
QA
quality assurance
CPHEA
Center for Public Health and
RD
relative deviation
Environmental Assessment
RfC
inhalation reference
DTXSID
DSSTox substance identifier
concentration
EPA
Environmental Protection Agency
RfD
oral reference dose
FT3
free triiodothyronine
SD
standard deviation
FT4
free thyroxine
TGAb
thyroglobulin antibody
GGT
y-glutamyl transferase
TIAB
title or abstract
GLP
Good Laboratory Practice
TMAb
thyroid microsomal antibody
HbAlc
form of hemoglobin linked to sugar
TSH
thyroid stimulating hormone
HED
human equivalent dose
UF
uncertainty factor
HERO
Health and Environmental
UFa
interspecies uncertainty factor
Research Online
UFC
composite uncertainty factor
IRIS
Integrated Risk Information System
UFd
database uncertainty factor
LOAEL
lowest-observed-adverse-effect
UFh
intraspecies uncertainty factor
level
UFl
LOAEL-to-NOAEL uncertainty factor
NLM
National Library of Medicine
UFs
subchronic-to-chronic uncertainty factor
NOAEL
no-observed-adverse-effect
U.S.
United States of America
level
WoS
Web of Science
NTP
National Toxicology Program
vii
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
BACKGROUND
The U.S. Environmental Protection Agency (EPA) Office of Research and Development
(ORD) under the Health and Environmental Risk Assessment National Research Program has
developed a human health toxicity value for perfluoropropanoic acid (PFPrA; Chemical Abstract
Services Registry Number [CASRN 422-64-0]). The assessment was developed in response to a
request for site-specific technical support and scoped to help meet site-specific public health
goals. The assessment provides qualitative and quantitative toxicity information that can be used,
along with exposure information and other important considerations, to assess potential health
risks to determine if, and when, taking action to address this chemical is appropriate.
The express purpose of this assessment is to provide support for risk-based decision-
making pertaining to chronic exposures to PFPrA at sites or geographic locations or in specified
environmental media (e.g., water, soil, air). Several factors are considered during scoping and
problem formulation activities to ensure the assessment appropriately fits the intended purpose.
These factors include the anticipated end-user need, peer-review and public comment
requirements, and anticipated availability of hazard and dose-response evidence. Factors are
assessed and informed through direct conversations with the requesting office(s) (e.g.,
EPA/OECA).
Noncancer and cancer toxicity values are derived (when supported by data) after a
systematic review of the relevant scientific literature, an evaluation of available hazard and dose-
response information using established EPA guidelines on human health risk assessment, and
appropriate internal EPA and external independent peer reviews. To the extent possible based on
the currently available evidence, the objective of this assessment is to present the major
conclusions reached in the hazard identification and derivation of human health toxicity values
and to characterize the overall confidence in these conclusions and values. This assessment is not
intended to represent a comprehensive treatise on the chemical. For example, less emphasis is
placed on providing definitive judgments of the integrated weight of evidence.
PFPrA Quality Assurance
This work was conducted under the EPA Quality Program to ensure data are of known
and acceptable quality to support their intended use. Surveillance of the work by the assessment
managers and programmatic scientific leads ensured adherence to quality assurance (QA)
processes and criteria and to quick and effective resolution of any problems. The QA manager,
assessment managers, and programmatic scientific leads have determined this work meets all
EPA quality requirements. This human health assessment was written with guidance from the
Center for Public Health and Environmental Assessment (CPHEA) Program Quality Assurance
Project Plan, the Quality Assurance Project Plan titled Umbrella Quality Assurance Project Plan
for CPHEA Fit-For-Purpose Toxicity Assessments (L-CPAD-0033369-QP-1-2), and the
contractor-led Quality Assurance Project Plan, General Support of CPHEA Assessment
Activities (L-CPAD-0031961-QP-1-2). As part of the QA system, a quality product review is
completed prior to management clearance. During assessment development, a Technical
Systems Audit was performed on December 15, 2022, with no major findings.
1
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
This assessment received internal peer review by two EPA/ORD/CPHEA scientists and
an independent, external peer review by three scientific experts outside of EPA. External peer
review was managed by Eastern Research Group, Inc. (110 Hartwell Avenue
Lexington, MA 02421) under contract EP-C-17-017. The reviews focused on whether studies
were correctly selected and interpreted and adequately described for the purposes of this ORD
assessment. The reviews also covered quantitative and qualitative aspects of the toxicity value
development and addressed whether uncertainties associated with the assessment were
adequately characterized.
2
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
INTRODUCTION
Per- and polyfluoroalkyl substances (PFAS) are a large group of anthropogenic chemicals
that include the well-known C8 species, perfluorooctanoic acid (PFOA), perfluorooctane
sulfonic acid (PFOS), and thousands of other structurally diverse fluorinated species. The
universe of environmentally relevant PFAS, including parent chemicals, metabolites, and abiotic
degradants, includes more than 12,000 substances.2
PFAS have strong, stable carbon-fluorine (C-F) bonds, making them resistant to
hydrolysis, photolysis, microbial degradation, and metabolism (Ahrens. 2011; Buck et at.. 2011;
Beach et at.. 2006). The chemical structures of PFAS make them repel water and oil, remain
chemically and thermally stable, and exhibit surfactant properties. These properties make PFAS
useful for commercial and industrial applications and purposes but also make some PFAS
persistent in the human body and the environment (Calafat et at.. 2019; Calafat et at.. 2007). Due
to their widespread use, physicochemical properties, persistence, mobility, and bioaccumulation
potential, many PFAS occur in exposure media (e.g., air, water, ice, sediment) and in tissues and
blood of aquatic and terrestrial organisms and humans.
Humans are widely exposed to PFAS (Sunderland et at., 2019), and PFAS have been
shown to pose ecological and human health hazards (Fenton et at.. 2011,1 v H \ 2021c. d;
DeWin JO I ^ Hekster et at.. 2003). The available toxicity data, however, are limited to
relatively few, well-studied PFAS (e.g., PFOA, PFOS, GenX chemicals, PFBS and others). Most
of the PFAS structures listed in EPA's CompTox Chemicals Dashboard1 are data poor, having
little to no toxicity data that might inform potential hazards to human health. One of these PFAS,
PFPrA (CASRN 422-64-0), has been detected in surface and ground waters in or around
manufacturing facilities. PFPrA, and its related salts, are all members of the overall PFAS class.
This assessment applies to the desalted acid form of PFPrA as well as salts (including non-metal
or alkali metal salts) of PFPrA that would be expected to fully dissociate in aqueous solutions of
pH ranging from 4-9 (e.g., in the human body). The synthesis of evidence and toxicity value
derivation presented in this assessment focuses on the forms of PFPrA with currently available
toxicity data. PFPrA is a short-chain, three-carbon perfluoroalkyl carboxylic acid and is a clear
and colorless liquid. The molecular formula and experimental or predicted physicochemical
properties of PFPrA are presented in Table 1.
2fattps://eomptox.epa.gov/dasfaboard/efaeniieat lists/PFASMASTER.
3
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
Table 1. Physical and Chemical Properties of PFPrA
Property or Endpoint (unit)
Value
Reference
Structure
F
F
0^
F
F
^OH
U.S. EPA (2021a)
CASRN
422-64-0
DTXSID
8059970
Synonyms
Perfluoropropanoic acid
Pentafluoropropanoic acid
Propanoic acid, 2,2,3,3,3-pentafluoro
Propanoic acid, pentafluoro-
Propionic acid, pentafluoro-
2,2,3,3,3-Pentafluoropropanoic acid
2,2,3,3,3-Pentafluoropropionic acid
Pentafluoropropionic acid
acido pentafluoropropionico
Pentafluorpropionsaure
Acide pentafluoropropionique
PFPA
PFPrA
Molecular formula
C3HF5O2
Molecular wt. (g/mol)
164.031
Physical description
Liquid, clear and colorless
CERI (2002c)
Odor
NA
Melting pt. (°C)
-11.0 (predicted average)
U.S. EPA (2021a)
Boiling pt. (°C)
96.4 (experimental average)
Density (g/cm3)
1.59 (predicted average)
pH (unitless)
NA
pKa (unitless)
NA
Vapor pressure (mm Hg)
19.4 (predicted average)
Henry's Law constant (atm-m3/mole)
3.64e-6 (predicted average)
Water solubility (mol/L)
0.291 (predicted average)
Octanol-water partition constant (log Kow)
1.79 (predicted average)
Bioconcentration factor (unitless)
3.57 (predicted average)
NA = not available.
4
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
METHODS
For this assessment, the general systematic review steps common to other assessments
(e.g., in Integrated Risk Information System [IRIS]) were applied, including the development of
populations, exposures, comparators, and outcomes (PECO) criteria for inclusion and literature
search and screening of multiple databases for relevant articles.
Briefly, methods used here are consistent with the ORD Staff Handbook for Developing
IRIS Assessments (Version 2.0, referred to as the draft "IRIS Handbook") ( 322b).
These methods were reviewed by the National Academy of Sciences (NASEM. 2021. 2018) and
used in other peer-reviewed systematic reviews (Yost et at.. 2019; Radke et at.. 2018). The
purpose of this human health assessment is to develop scientifically supported chronic toxicity
values, where data are available. Less emphasis is placed on providing definitive judgments of
the integrated weight of evidence.
The systematic review methods used to collect epidemiological and toxicological
evidence for PFPrA are described in detail in Carlson et al. (2022) and Radke et al. (2022)-. as
well as Appendix A. PFPrA was included in the list of 150 PFAS described in those materials,
and for the purposes of this summary, the PFPrA-specific results found were isolated as a result
of the outlined processes. In addition to database searches, nonconfidential business information
(non-CBI) industry studies were identified that included toxicological evidence for PFPrA. Since
February 2020, EPA has requested, pursuant to section 308 of the Act, 33 U.S.C. § 1318, that
3M provide information on its use and possible release of certain PFAS, such as PFPrA. Under
that agreement, 3M shared internal files of memos, reports, interim or final data summaries or
studies, correspondence, etc. that included non-CBI content for PFPrA. The available
information for PFPrA received from 3M was screened for relevance to the PECO criteria (see
Figure 1). Literature identified as relevant from the database searches or non-CBI data
repositories underwent data extraction and study evaluation (see documentation in Health
Assessment Workspace Collaborative [HAWC]:
https://hawcprd.epa.eov/assessment/100500281/).
3 A literature search update has been conducted for PFPrA in December 2021 for the purposes of this assessment
since the publication of the evidence map; therefore, the literature search and screening results presented here may
differ from those described in Carlson et al (2022) and Radke et al. (2022).
5
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
RESULTS
Literature Search and Screening Results
The database searches yielded 352 unique references for PFPrA. As shown in Figure 1,
three human, one animal, and two genotoxicity studies from the 352 initial references with
information relevant to understanding the potential health effects of PFPrA were identified.
Perfluoropropanoic Acid (PFPrA) (CASRN 422-64-0)
Literature Searches (June-Aug2019, December 2020, December 2021)
PubMed
(n = 312)
WOS
(n = 144)
ToxLine
(n = 22)
Other
Selected non-CBI 3M
Studies Submitted to
EPA (n = 3)
T
REVIEWED FOR PECO
Title abstract and full-text Screening | >
(352 records after duplicate removal)
Studies Meeting PECO (n = 4)
Human health effects studies (n = 3)
Animal health effect studies (n = 1)
Excluded (n = 342)
Not relevant to PECO (n = 342)
Tagged as Supplemental (n = 6)
Genotox (n = 2)
Exposure only/no original data (n = 2)
Non-mammalian models (n = 1)
Mechanistic (n = 1)
Study Summary {n = 6)
Human (n = 3); Animal (n = 1); Supplemental (Genotox, n = 2)
Figure 1. PFPrA Literature Search Flow Diagram
Human Studies
Three studies (Duan et al.. 2020; Song et al.. 2018; Li et al.. 2017) examining
associations of health effects with PFPrA blood concentrations in humans were identified (see
Table 2). All three studies were general population cross-sectional analyses conducted in China.
Adults were the primary subject in each case; a portion of the study sample in Li et al. (2017)
was younger than 18 years, but the authors did not conduct sub-analyses by lifestage. One study
6
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
was rated at an overall medium confidence (Duam et at., 2020), and two were rated low
confidence (Sons et at.. 2018; Li et at.. 2017) (see Figure 2 and H.A.WC link for details on study
confidence ratings). All three studies were rated as deficient for study sensitivity due to narrow
concentration contrasts (Duam et at.. 2020; Sons et at.. 2018; Li et at.. ) or small sample size
(Duam et at.. 2020; Some et at.. 2018; Li et at.. 2017). so null findings should not be interpreted as
evidence of no effect. Further details on the specific studies, concentration measurements, and
chemicals evaluated are available in the interactive HAWC link.
Specific Studies
Duanetal. (2020)
One medium confidence cross-sectional study of nondiabetic Chinese adults examined
the association between serum PFPrA concentrations and glycemic indicators (Duam et at..
2020). Participants provided an overnight fasting blood sample. Serum fasting glucose and
HbAlc (a form of hemoglobin linked to sugar and a biomarker for prediabetes or type 2
diabetes) were measured. Concurrent measurement of serum PFPrA concentrations and outcome
was considered adequate due to the potential for short-term responses in these outcomes. In
unadjusted linear regression models, serum PFPrA was significantly associated with decreasing
HbAlc levels. After adjustment for potential confounders (sex, age, body mass index, smoking
and alcohol use, exercise, education, and family history of diabetes), however, the effect was no
longer significant (P [95% confidence interval]: -0.012 [-0.026, 0.002]). A significant
interaction (/^-interaction = 0.024) with body mass index was observed for the association
between serum PFPrA levels and HbAlc. When stratified by age (<55 and >55 years), the
association for HbAlc was not significant for either group, although the direction of effect was
similar to that of the combined analysis. The association for fasting glucose was in the same
direction as for HbAlc but was not statistically significant before or after adjustment. The
biological significance of decreasing HbAlc levels in association with serum PFPrA
concentrations is unclear.
Li ei al (2017)
One low confidence cross-sectional study examined the association between PFPrA
concentrations and thyroid hormones (Li et al.. 2017). Adult and child participants with normal
thyroid function and with thyroid disease (i.e., hyperthyroidism, hypothyroidism, and
Hashimoto's disease) in China provided serum samples for analysis of PFAS and thyroid
hormones. The analysis included thyroid stimulating hormone (TSH), free thyroxine (FT4), free
triiodothyronine (FT3), thyroglobulin antibody (TGAb), and thyroid microsomal antibody
(TMAb). This study was low confidence due to multiple concerns for risk of bias. Details on
recruitment and participation were limited, and, for those participants without thyroid disease,
the reason for presentation to the hospital was not clear. Potential confounding by socioeconomic
status and other factors was a concern. Additionally, timing of outcome assessment was
unaccounted for in the analysis or design. Inconsistent timing of outcome assessment could lead
to outcome misclassification due to the diurnal variations in thyroid hormones. Bivariate
correlations of PFPrA concentrations and thyroid hormones and antibodies revealed no
statistically significant associations (all correlation coefficients ranged from -0.05 to -0.1,
except for FT3 in participants with hypothyroidism, r = 0.4). Only statistically significant
associations were carried forward to linear regression analysis, which included adjustment for
7
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
confounding. Quantitative results for multivariable linear regression were not reported for PFPrA
due to lack of statistical significance.
Sons et al (2018)
One low confidence cross-sectional study examined the association between PFPrA
concentrations and semen parameters (Song et al.. 2018). Men at an infertility clinic in China
were recruited; reasons for visiting the infertility clinic were not provided, but only men without
genital damage, venereal disease, or azoospermia were included in the study. The association
between PFPrA concentrations and semen parameters was examined separately for PFPrA
concentrations in serum and semen. This study was low confidence due to lack of adjustment for
potential confounders and limited information on participant selection and semen collection and
analysis. No statistically significant effects were identified for semen quality parameters using
concentrations from either biomonitoring matrix, and the direction of association was
inconsistent in serum and semen for both sperm motility and concentration.
Summary
In summary, evidence on the health effects of PFPrA in human epidemiological studies is
limited. No clear associations were observed with glycemic indicators, thyroid hormones, or
semen parameters. No effects of PFPrA concentrations were reliably identified in the available
human studies; due to poor sensitivity across the available studies, however, this should not be
interpreted as evidence of no effect.
8
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Table 2. Associations Between PFPrA Concentrations and Health Effects in Human Epidemiological Studies
Reference,
Confidence
Location, Years
Design
Population
Ages (N)
Concentration Matrix
and Levels (ng/mL)
Outcome
Comparison
Select Results3
< icncrul Population
Duan et al. (2
Medium (see
HAWC link for
details)
(lima. : 0.05
Sons et al. (2018)
Low (see HAWC
link for details)
China, 2012-2013
Cross-
sectional
Men
(N= 103)
Median
(5th-95th percentile)
Serum:
0.62 ng/mL (0.21-2.1)
Semen:
0.95 ng/mL (0.29-4.1)
0% BLOD in both
matrices
Semen
concentration
(106/mL),
progressive
motility (%)
Spearman
correlation
coefficients
Semen concentration,
Serum: -0.112
Semen: 0.114
Progressive motility,
Serum: 0.176
Semen: -0.180
BLOD = below the limit of detection; FT3 = free triiodothyronine; FT4 = free thyroxine; HbAlc = form of hemoglobin linked to sugar; LOD = limit of
detection; TGAb = thyroglobulin antibody; TMAb = thyroid microsomal antibody; TSH = thyroid stimulating hormone.
aResults reported as effect estimate (95% confidence interval) unless otherwise specified. In some cases, presented results are constrained to those models or
comparisons that most fully addressed major sources of potential bias.
9
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
-,e
,Ce
Duan, 2020, 5918597-
+
+
+
-
+
+
-
+
Li, 2017, 3856460-
-
+
-
-
+
+
-
-
Song, 2018, 4220306-
-
++
-
-
-
+
-
-
Figure 2. Summary of Study Evaluation for Human Epidemiological Studies of PFPrA and
All Health Outcomes
Interactive figure and additional study details available on HAWC.
Animal Studies
A single repeat dose study is available that evaluates the toxicity of PFPrA after oral
exposure (CERI 2002c) (see Table 3). The available study is a 28-day exposure in 5-week-old
male and female Crj:CD (SD) IGS rats (SPF). To determine a dose range for the 28-day study, a
14-day study was first performed that included doses of 50, 250, or 1,000 mg/kg-day. Effects on
hematological parameters and organ weights were found in all dose groups after 14 days of oral
exposure. Clinical chemistry abnormalities and necropsy findings were reported in the mid- and
high-dose groups and clinical signs of toxicity, changes in body weights, and histopathology
were reported in the high-dose group. No other details regarding the 14-day study design (e.g.,
number and sex of the animals) methods of endpoint evaluation, or quantitative exposure-
response data were provided. The results of the 14-day study informed the authors' selection of
doses for the 28-day study: Male and female rats (n = 6 per dose group and sex) were exposed
via daily oral gavage to 0, 5, 20, 80, or 320 mg/kg-day PFPrA in water for 28 consecutive days.
Additional control and high-dose animals were maintained for a 14-day recovery period. The
study was conducted according to Organization for Economic Cooperation and Development
(OECD) and Good Laboratory Practice (GLP) guidelines and evaluated clinical signs, body
weights, food intake, hematology, blood chemistry, urinalysis, organ weights (liver, kidneys,
10
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
testes, ovaries, brain, spleen, adrenals) and histopathology (forestomach, glandular stomach,
intestine [duodenum, jejunum, ileum, cecum, colon, rectum], liver, heart, kidneys, spleen,
adrenals). Confidence is high in the study for all endpoints evaluated, with no significant
concerns for potential bias or insensitivity (see Figure 3).
CERl 2002. 8728368
¦
-Hh
NR
++
++
+-*-
++
HH-
++
4-+
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (m etric) or Uninformative (overall)
NR Not reported
Figure 3. Summary of Study Evaluation for Experimental Animal Toxicological Studies of
PFPrA and All Health Outcomes
Interactive figure and additional study details available on HAWC.
No significant effects on body weight or food intake were reported in rats with doses up
to 320 mg/kg-day. Clinical signs of toxicity occurred in both males and females at the high dose
and included decreased movement (8/12 males and 4/12 females) and increased salivation (9/12
males and 6/12 females). The salivation was noted to occur right after dosing and was associated
with mucosal irritation and hyperplasia of squamous epithelium in the limiting ridge of the
forestomach in females (4/6 animals). Loss of hair and exude/scab formation were observed in
1/6 males in the 5 and 80 mg/kg-day dose groups.
Dose-related increases in liver weights (absolute and relative) were reported in male rats
(see Figure 4). Relative liver weight, the preferred metric for this organ based on its proportional
relationship to body weight ( jalley et al, 2004), increased 14%-36% at >20 mg/kg-day in
males, reaching statistical significance at 80 and 320 mg/kg-day. Marginal, non-statistically
significant increases in relative liver weight were observed in the females of the same dose
groups (7% and 9% at 80 and 320 mg/kg-day, respectively). Increased relative liver weights
persisted after the recovery period in the 320 mg/kg-day males.
11
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
Study Name Study Endpoint Name animal description (with N) Response Units Dose
Design (mg/kg-day)
PFPA Liver Weights
CERI 2002, 8728368 28-Day Oral Liver Weight, Absolute Rat, Crj:CD(SD)IGS (
5
I
20
• 1
80
1 • 1
320
1 • 1
Rat, Crj:CD(SD)IGS (?, N=6) g 0
5
1 1
1—•
> 1
20
I 1
80
320
1 • 1
1 • 1
Liver Weight, Relative Rat, Crj:CD(SD)IGS (<*, N=6) g/100g 0
1 1
1 1
5
1
20
1 • 1
80
1 • 1
320
Rat, Crj:CD(SD)IGS (?, N=6) g/100g 0
1—(
1 • 1
1 1
5
1 • 1
20
I •—
an
^ Percent control response
O Statistically significant
M 95% CI
320
•
0
5 -10 -5
5 10 15 20 25 30 35 40 45 5
Percent Control Response
Figure 4. Liver Weight Changes in Rats Following Oral Expose to PFPrA
Interactive figure available on HAWC.
The liver weight increases in males were accompanied by noticeable enlargement of the
liver at necropsy in the 320 mg/kg-day dose group and increased incidence of centrilobular
hypertrophy (slight to moderate severity) at the two highest doses (2/6 and 6/6 animals at 80 and
320 mg/kg-day, respectively) (see Figure 5). Slight, focal necrosis was observed in 1/6 males in
both the 20 and 80 mg/kg-day dose groups. These liver lesions were not observed in the controls
or in females (0, 20, 80 and 320 mg/kg-day dose groups were evaluated in males and 0 and
320 mg/kg-day dose groups were evaluated in females for liver histopathology). Livers were no
longer visibly enlarged nor were any microscopic lesions noted in males following the 14-day
recovery period.
Study Name
Study
Design
Endpoint Name
Animal Description
Incidence
Dose
(mg/kg-day)
CERI 2002, 8728368
28-Day Oral
Hepatocellular Hypertrophy
Rat, Crj:CD(SD)IGS (cf)
0/6 (0.0%)
0
20
2/6 (33.3%) 80
6/6(100.0%)
320
Rat, Crj:CD(SD)lGS (9)
0/6 (0.0%)
0
320
Hepatocyte Necrosis. Focal
Rat, Crj:CD(SD)IGS (d1)
0/6 (0.0%)
0
1/6(16.7%)
20
80
0/6 (0.0%)
320
| Statistically significant
I INo significant change
Rat, Crj:CD(SD)IGS (9) 0/6 (0.0%) 0
320
PFPrA Liver Histopathology
i 6
incidence
Figure 5. Liver Histopathology in Rats Following Oral Expose to PFPrA
Interactive figure available on HAWC.
12
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Serum enzyme levels were also examined in the 28-day rat study (CERI, 2002c) (see
H.A.WC figure for more details). Alanine aminotransferase (ALT) and aspartate aminotransferase
(AST) are markers of hepatocellular damage, while alkaline phosphatase (ALP) and gamma-
glutamyl transferase (GGT) are markers of hepatobiliary damage (Hall et at.. 2012; EMEA.
2008; Boone et at.. 2005). In males, ALT and ALP levels showed an increasing trend (p < 0.05)
across dose groups, reaching statistical significance at the highest dose (40% at 320 mg/kg-day)
for ALT and at 80 mg/kg-day (30%) for ALP. Levels of AST and GGT were not significantly
elevated in either male or female rats. Effects in other clinical markers relevant to liver function
were also reported. Blood proteins such as albumin and globulin are routinely evaluated in
clinical chemistry and changes in the levels of these blood proteins can be indicators of kidney or
liver damage (Whalan. ). The albumin/globulin (A/G) ratio was significantly increased in all
dose groups in males (15%—25%), and albumin levels were slightly elevated in the high-dose
male group (4%); the increases in albumin and A/G ratio displayed a significant trend.
Significant decreases in total bilirubin occurred in males (33% in all dose groups) and in females
(44% at doses >80 mg/kg-day), but the biological significance of this decrease is unclear.
Although results were not always coherent across endpoints, changes in some serum markers
provide support for potential liver damage in PFPrA-exposed animals (i.e., increased ALT and
ALP, and possibly A/G ratio).
Other health effects were observed in rats after 28 days of exposure, but they generally
occurred only at the highest dose, were sporadic, or were not supported by corroborative
evidence of toxicity. For example, kidney weight changes were observed in male and female rats
but were not accompanied by significant histological lesions or biochemical indicators of kidney
toxicity in the blood or urine. Indeed, female absolute kidney weights were significantly
increased (p < 0.05) in the 80 mg/kg-day group only (16%), with a 10% increase also observed
in the 20 and 320 mg/kg-day groups. Similarly, the relative kidney weights in females were
significantly increased at 80 mg/kg-day (9%,p < 0.05). Absolute kidney weights in males were
increased by 10% or more in the 80 and 320 mg/kg-day groups, but the changes were not
statistically significant. Relative kidney weights were significantly increased in males exposed to
>20 mg/kg-day (15%—18%,p < 0.05).
Sporadic, statistically significant changes in clinical chemistry also occurred, mainly in
male rats. An increase in cholinesterase (73%) and a decrease in calcium (5%) were reported in
males exposed to 80 mg/kg-day. Activated partial thromboplastin time was increased by 11%-
18% in females at >5 mg/kg-day but did not follow a dose-response and was not accompanied by
changes in platelets or prothrombin time. Total cholesterol was decreased (31 %—41 %) in males
in all exposure groups. The biological significance of these changes in the absence of additional
data is unclear.
In summary, the liver appears to be a primary target organ for PFPrA after short-term oral
exposure, with male rats more sensitive than female rats. Coherent liver effects in males were
reported at >20 mg/kg-day across organ weights, histopathology, and clinical serum markers,
including dose-related increases in relative liver weights, hepatocyte lesions (mainly
centrilobular hypertrophy but also some evidence of degenerative changes [slight focal
necrosis]) and changes in serum markers indicative of hepatocellular/hepatobiliary injury
(i.e., increased ALT and ALP).
13
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Table 3. Available Experimental Animal Oral Toxicity Data for PFPrA
Species (Strain),
Study Details
Dose
(mg/kg-d)
Critical Effects
Other Effects
Reference,
Confidence
\cnie
No data available.
Shori-ierni
kal <( ij CI) LSI)J l(iSi
28-d
Oral gavage
ii. 5. 2d. So.
320
Increased relalne li\erweiulil in males al
>20 mg/kg-d, accompanied by hepatocyte
lesions (primarily hypertrophy with some
evidence of slight focal necrosis) and serum
markers of hepatocellular/hepatobiliary injury
(i.e., increased ALT, ALP) at >80 mg/kg-d
Decreased mm emenl and increased
salivation after treatment; increased
activated partial thromboplastin time,
albumin, albumin/globulin ratio, kidney
weight; decreased total cholesterol, total
bilirubin; increased incidence of
forestomach lesions
High
Snhcliroiiic
\n dala a\ ailahle
( limine ( arciikiuemcils
\n dala a\ ailahle
Reproduce e 1 )e\ ekipnieiiial
No data available.
ALT = alanine aminotransferase; ALP = alkaline phosphatase.
14
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Other Data
Other human health relevant studies conducted on PFPrA include an Ames test to
evaluate genotoxicity and a chromosomal aberration study (see Table 4). For the Ames test, a
dose range finding study and a main study were conducted (CI 32a). The main study
included five concentrations ranging from 313 to 5,000 |ig/plate with the highest dose of
5,000 |ig/plate diluted for the remaining four doses with a geometric progression of two. The
compound was found not to have potential for mutagenicity. In the chromosomal aberration test,
Chinese hamster lung fibroblasts were exposed to concentrations of 0, 410, 820, or 1,640 |ag/m L
of PFPrA, and no increase was found in the frequencies of cells with total aberrations (chromatid
breaks, chromatid exchanges, chromosome breaks, chromosome exchanges) using the short-term
or the continuous treatment methods (CERI, 2002b). Growth inhibition tests, including
concentrations as low as 6.41 |ig/mL, demonstrated some inhibition at the 1,640 |ag/m L
concentration using the short-term method and at concentrations >205 |ag/m L in the continuous
treatment method, but no increases in structural or numerical aberrations.
Table 4. Summary of PFPrA Genotoxicity Studies
Endpoint
Test System
Concentrations
Tested
Results
Without
Activation3'b
Results With
External
Activation3'b
References
Ames Assay
(revertant
colonies)
Salmonella
typhimurium
(TA100, 1535, 98,
1537), Escherichia
coli (WP2 uvrA) in
the presence and
absence of
metabolic
activation system
(S9 mix)
0,313,625,
1,250, 2,500,
5,000 ng/plate
CERI (2002a)
Chromosomal
aberration
Chinese hamster
lung fibroblasts
with or without S9
mix
0, 410, 820,
1,640 ng/mL
CERI (2002b)
aResults reported as - = negative.
15
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
DERIVATION OF REFERENCE VALUES
The hazard and dose-response database for PFPrA is limited to studies via the oral route
of exposure. There are no known inhalation or dermal studies for PFPrA. Further, no known
studies have evaluated potential cancer effects of PFPrA, and studies relevant to potential cancer
mechanism(s) are sparse and inconclusive. The purpose of this assessment is to inform human
health hazard(s) associated with chronic duration/lifetime exposures to PFPrA. Therefore, only a
noncancer chronic reference dose (RfD) is derived in this assessment for the oral route of
exposure. The RfD derived in this assessment is an estimate of an oral exposure to the human
population (including susceptible subgroups and lifestages) likely to be without an appreciable
risk of adverse health effects over a lifetime.
Derivation of Oral Reference Dose
The hazard and dose-response database for PFPrA is limited to one medium confidence
(Duam et at.. 2020) and two low confidence (Sons et at.. 2018; Li et at.. 2017) epidemiological
studies that evaluated potential associations between health effects and PFPrA blood serum
concentrations in humans, and one high confidence repeat-dose (28-day) oral gavage study in
rats [conducted by the Chemicals Evaluation and Research Institute, Japan; (CERI. 2002c)l. Two
of the three human studies have multiple limitations discussed previously (see the Human
Studies summaries for more details) that diminish confidence in reported associations and
decrease their ability to inform conclusions. In addition, studies that used measurements of
biomarkers in tissues or bodily fluids as the metric for exposure were considered suitable only
for dose-response analysis if data or physiologically based pharmacokinetic (PBPK) models are
available to extrapolate between the reported biomarker measurement and the route-specific
level of exposure. As such, the human studies were not considered further for toxicity value
derivation.
The 28-day oral rat study CERI (2002c) was conducted consistent with OECD guideline
protocol and under GLP conditions and had an overall high confidence rating based on a study
quality evaluation (see ik for more details). Male and female control and PFPrA-
treated rats were evaluated across a comprehensive panel of general toxicity, clinical chemistry,
organ weight, and histopathological parameters. The liver of rats was identified as a primary
target of PFPrA toxicity following 28 days of oral exposure; male rats were more sensitive than
females across all parameters evaluated and thus were prioritized for dose-response analyses.
Alterations included increases in relative liver weights at >20 mg/kg-day and increased absolute
liver weights in males at >80 mg/kg-day; gross enlargement of the liver and histopathological
indicators of altered tissue architecture or injury (e.g., centrilobular hypertrophy; some evidence
of focal necrosis) at >80 mg/kg-day; and increased serum enzymes indicative of hepatic injury
(i.e., ALT and ALP) at >80 mg/kg-day. According to Hall et al. (2012). this constellation of
effects is consistent with criteria supporting a determination of liver injury. Specifically, the
PFPrA-induced liver effects are indicative of an interrelated pattern of toxicity to parenchymal
(i.e., hepatocyte hypertrophy, necrosis, and ALT release into systemic circulation) and
nonparenchymal (e.g., hepatic biliary epithelial release of ALP into systemic circulation) cell
populations. Despite the lack of additional oral repeat-dose studies examining liver effects of
PFPrA by which to evaluate similarity of results, this profile of PFPrA-induced liver effects is
consistent with liver toxicity observed in experimental rodents following oral exposure to
perfluorobutanoic acid, a closely related linear short-chain (4-carbon) perfluorocarboxylic acid
16
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
( 202 le). These effects observed in animals are considered relevant for humans in the
absence of experimental data that provide direct information to the contrary. In total, evidence in
animals indicates that PFPrA exposure may cause liver effects in humans, but few studies were
available to contribute to the evaluation. The main study that this conclusion is based on assessed
dose levels of 5-320 mg/kg-day and was conducted according to well-established experimental
animal guidelines (CERI. 2002c). Despite limitations in the availability of repeat-dose toxicity
studies in the database (including in species other than rat), the 28-day rat study by CERI
(2002c) was considered for the derivation of a chronic RfD for PFPrA. The RfD for PFPrA may
be useful for certain decision contexts, such as providing a sense of the magnitude of potential
human health risks, ranking potential hazards, or informing PFAS mixtures assessment in which
PFPrA is a component ( €5, 2000).
The PFPrA-induced liver effects observed in male rats from the CERI (2002c) 28-day
study were evaluated for amenability to benchmark dose (BMD) modeling (see Table 5).
Consistent with EPA's Benchmark Dose Technical Guidance (U.S. EPA. 2020a. 2012. 2002).
the BMDs and 95% lower confidence limits on the BMDs (BMDLs) for increased relative liver
weight, serum ALT, serum ALP, and incidence of hepatocyte hypertrophy were estimated using
a benchmark response (BMR) representative of a biologically or statistically significant level of
change for continuous (e.g., relative liver weight; serum ALT and ALP) or dichotomous
(e.g., incidence of hepatocyte hypertrophy) endpoints. For liver weight changes, a 10% increase
over control is considered biologically significant for this assessment. For serum ALT and ALP,
a 1-standard deviation (SD) change over control was used. For hepatocyte hypertrophy, a 10%
increased incidence over control was used (U.S. EPA. 2020a. 2012. 2002). The full results of the
BMD modeling are provided in Appendix B.
Table 5. Data for Liver Effects in Adult Male Crj:CD (SD) IGS (SPF) Rats Exposed to
PFPrA for 28 Days via Gavage (CERI. 2002c)
Endpoint
Dose, mg/kg-db
0
5
20
80
320
Relative liver weight -
% of BWa
3.15 ± 0.19
3.24 ±0.21
(+3%)
3.58 ±0.34*
(+14%)
3.92 ±0.29**
(+24%)
4.27 ±0.43**
(+36%)
Serum ALT-IU/La
25 ±5
27 ±3
(+8%)
27 ±2
(+8%)
29 ±4
(+16%)
35 ±4**
(+40%)
Serum ALP-IU/La
420 ± 48
428 ± 74
(+2%)
242 ± 83
(+1%)
545 ±79**
(+30%)
518±107
(+23%)
Hepatocyte hypertrophy
- incidence
0
ND
0
2
6
Animals (n)
6
6
6
6
6
ALT = alanine aminotransferase; ALP = alkaline phosphatase; BW = body weight; ND = not determined.
aValues expressed as mean ± SD. Parentheses show % change relative to control = ([treatment mean - control mean] control
mean) x 100.
bDosimetry: Oral rat exposures are expressed in mg/kg-day as reported by the study authors.
*Biologically significant change from control. **Statistically (p < 0.05) and biologically significant change from control.
Following dose-response modeling of the liver effect data in male rats, BMDLs were
converted to corresponding human equivalent doses (HEDs) (see Table 6). In Recommended Use
17
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
of Body Weight4 as the Default Method in Derivation of the Oral Reference Dose (
2011\ EPA endorses a hierarchy of approaches to derive human equivalent oral exposures from
data on laboratory animal species, with the preferred approach being physiologically based
toxicokinetic modeling. Without a complete physiologically based toxicokinetic model, other
approaches might include using available chemical-specific information (e.g., clearance or
plasma half-life values). In the absence of chemical-specific models or data to inform the
derivation of human equivalent oral exposures, EPA recommends doses be scaled allometrically
using body weight (BW)3/4 as a default method to extrapolate toxicologically equivalent doses of
orally administered agents from laboratory animals to humans to derive an RfD, under certain
exposure conditions. For PFPrA, no toxicokinetic data were identified, so no chemical-specific
data are available to inform cross-species kinetics between rats and humans4. As such, the male
rat BMDLs were converted to the human equivalent points of departure (PODheds) using default
BW3/4 scaling. Table 6 provides the candidate points of departure (i.e., PODheds) obtained from
the BMD-modeled liver effects data from male rats of the 28-day study (CERI. 2002c).
Table 6. Candidate PODs for Derivation of the Chronic RfD for PFPrA
Endpoint
BMDL
mg/kg-d
POD type
PODhed3
mg/kg-d
Reference
Increased relative liver weight
in adult males
6.3
BMDLio
1.6
CERI (2002c)
Increased hepatocyte
hypertrophy in adult males
7.9
BMDLio
2.0
CERI (2002c)
Increased serum ALP in adult
males
20
BMDLisd
5.0
CERI (2002c)
Increased serum ALT in adult
males
28
BMDLisd
7.0
CERI (2002c)
ALP = alkaline phosphatase; ALT = alanine aminotransferase; BMDL = 95% lower confidence limit on the BMD
(subscripts denote BMR: i.e., 10 = dose associated with a 10% extra risk in parameter, 1SD = dose associated with 1
standard deviation relative risk from the control); BMR = benchmark response; POD = point of departure;
PODhed = human equivalent point of departure.
aHEDs were calculated using species-specific application of a dosimetric adjustment factor (DAF), as recommended
by U.S. EPA (20.1.1). The DAFs are calculated as follows: DAF = (BWa1/4 BWh1'4), where BWa = animal body
weight, and BWh = human body weight. Default body weight for male SD-derived rats (0.267 kg [for subchronic
duration]) and a reference body weight of 80 kg for humans, as recommended in U.S. EPA (.1.988). were used to
calculate the DAFs.
Considering that confidence among the candidate points of departure (PODs) is
approximately equivalent (i.e., the same study population/species, exposure paradigm, and
quality of exposure and outcome measurement), the PODhed of 1.6 mg/kg-day for increased
relative liver weight represented the most sensitive effect in rats and was identified as the POD
4 To inform cross-species extrapolation for PFPrA, toxicokinetic (TK) data for PFBA, a closely related linear short-
chain (4-carbon) perfluorocarboxylic acid, was considered. For PFBA, TK data exist in relevant animals and
humans, leading to a data-informed extrapolation approach (i.e., ratio of the clearance (CL) in humans to animals,
CLh:CLa) for estimating the DAF in U.S. EPA (2021e). For comparison, the DAF for PFPrA. based on the default
(BW)3'4 approach in male rats, is 0.25 which is similar to the data-informed DAF for male rats for PFBA of 0.229.
18
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
for deriving a chronic RfD for oral PFPrA exposure. Under EPA's A Review of the Reference
Dose and Reference Concentration Processes ( E002) and Methods for Derivation of
Inhalation Reference Concentrations and Application of Inhalation Dosimetry ( 4),
five possible areas of uncertainty and variability were considered in deriving the chronic RfD for
PFPrA. The chronic RfD is derived by applying a composite uncertainty factor (UFc) of 3,000
(reflecting an interspecies uncertainty factor [UFa] of 3, an intraspecies uncertainty factor [UFh]
of 10, a duration uncertainty factor [UFs] of 10, and a database uncertainty factor [UFd] of 10) to
the PODhed of 1.6 mg/kg-day. Table 7 summarizes the uncertainty factors for the chronic RfD
for PFPrA. Confidence in the chronic RfD for PFPrA is low, as described in Table 8. The low
confidence in the chronic RfD, resulting primarily from the limited available hazard and dose-
response relevant evidence in the database, indicate a high level of uncertainty in the derived
RfD. Nevertheless, this RfD may be useful for some decision purposes ( ^005).
Chronic RfD = PODhed ^ UFc
= 1.6 mg/kg-day3,000
= 0.0005 or 5 x 10"4 mg/kg-day
19
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Table 7. Uncertainty Factors for the Chronic RfD for PFPrA (CASRN 422-64-0)
UF
Value
Justification
UFa
3
A UFa of 3 (10°5) is applied to account for uncertainty in characterizing the toxicokinetic and
toxicodynamic differences between animals and humans following oral PFPrA exposure. Cross-
species dosimetric adjustment (HED calculation) was performed using default allometric BW3'4
scaling between rats and humans. This scaling is applied to account for some aspects of the cross-
species toxicokinetic processes. Further, cross-species toxicokinetic (TK) data for PFB A, a closely
related linear short-chain (4-carbon) perfluorocarboxylic acid, was considered. For PFBA, TK data
exist in relevant animals and humans, leading to a data-informed extrapolation approach (i.e., ratio
of the clearance (CL) in humans to animals, CLh:CLa) for estimating the DAF as suggested in U.S.
EPA (2022a). For comparison purposes, for PFPrA, the DAF of 0.25 for male rats based on the
application of the default (B W)3/4 approach is similar to the data-informed DAF for male rats for
PFBA of 0.229. This suggests that although a default allometric BW scaling approach is used for
the 3-carbon structure PFPrA, the resulting DAF is similar to a data-informed DAF for the 4-
carbon PFBA. As such, a factor of 3 is applied to account for residual toxicokinetic uncertainty and
potential toxicodynamic differences across species.
UFd
10
A UFd of 10 is applied to account for deficiencies and uncertainties in the database. The database
for oral exposure to PFPrA is limited to three human epidemiological studies (one medium
confidence and two low confidence) and a single high confidence, 28-day repeat-dose oral rat
study. No longer-duration repeat-dose studies, examining potential systemic, reproductive,
developmental or immunotoxicity effects are available following exposure via any route.
UFh
10
A UFh of 10 is applied to account for interindividual variability in the susceptibility of the human
population because of both intrinsic and extrinsic factors that can influence the response to dose, in
the absence of chemical-specific information to assess toxicokinetic and toxicodynamic variability
of PFPrA in humans.
UFl
1
A UFl of 1 is applied because the POD is a BMDL.
UFS
10
A UFS of 10 is applied to account for uncertainty in how a significantly longer exposure duration
might impact the incidence and or severity of liver injury. The POD was derived from a 28-day rat
study; studies of PFPrA exposures for longer than 28 days were not available to evaluate and
characterize the potential for increasing magnitude or incidence of injury in the liver with
increasing exposure duration.
UFC
3,000
Composite UF = UFA x UFD x UFH x UFL x UFS.
BMDL = benchmark dose lower confidence limit; BW = body weight; HED = human equivalent dose; POD = point
of departure; RfD = reference dose; UF = uncertainty factor; UFa = interspecies uncertainty factor; UFC = composite
uncertainty factor; UFd = database uncertainty factor; UFH = intraspecies uncertainty factor; UFL = LOAEL-to-
NOAEL uncertainty factor; UFS = less-than-chronic duration uncertainty factor.
20
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Table 8. Confidence Descriptors for the Chronic RfD for PFPrA (CASRN 422-64-0)
Confidence Categories
Designation
Discussion
Confidence in study
H
Confidence in the t>rincit>al studv CERI (2002c) is high. The
study was performed by an industry/contract lab using an
established OECD protocol for 28-day oral exposures in rodents
and under GLP conditions. All but one of the toxicity study rating
criteria were of "Good" or "High" confidence (see Figure 3 and
information available on HAWC).
Confidence in database
L
Confidence in the database for PFPrA is low. The relevant human
health assessment database consists of one medium and two low
confidence human epidemiological studies, and a single 28-day
repeat-dose oral rat study. No longer-duration repeat-dose studies,
examining potential systemic, reproductive, developmental or
immunotoxicity effects are available following exposure via any
route.
Confidence in quantification of
the PODhed
M
Confidence in the quantification of the POD and RfD is medium.
The POD was based on BMD modeling within the range of the
observed data. Dosimetric adjustment of the POD was based on
default BW3'4 scaling due to the lack of chemical specific
toxicokinetic data (e.g., clearance, half-life).
Confidence in the chronic RfD
L
The overall confidence in the chronic RfD is low and is primarily
driven by low confidence in the available database for PFPrA.
BMD = benchmark dose; BW = body weight; GLP = Good Laboratory Practice; HED = human equivalent dose;
POD = point of departure; RfD = reference dose.
Derivation of Inhalation Reference Concentrations
No studies have been identified that examine noncancer effects of PFPrA via the
inhalation exposure route.
Summary of Noncancer Reference Values
Noncancer reference values are summarized in Table 9.
Table 9. Summary of the Noncancer Reference Values for PFPrA (CASRN 422-64-0)
Toxicity Type
(units)
Species/Sex
Critical Effect
Reference
Value
mg/kg-d
POD
Method
PODhed
mg/kg-d
UFc
Principal
Study
Chronic RfD
(mg/kg-d)
Rat/M
Increased
relative liver
weight
0.0005
BMDLio
1.6
3,000
CERI
(2002c)
Chronic RfC
(mg/m3)
NDr
BMDL = benchmark dose lower confidence limit (subscripts denote benchmark response: i.e., 10 = dose associated
with a 10% extra risk in parameter); M = male(s); NDr = not derived; PODhed = human equivalent point of
departure; RfC = reference concentration; RfD = reference dose; UFC = composite uncertainty factor.
21
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
CARCINOGENICITY ASSESSMENT
No studies have been identified that examine potential carcinogenicity of PFPrA via any
route of exposure.
22
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
REFERENCES
Ahrens. L. (2011). Polyfluoroalkyl compounds in the aquatic environment: a review of their occurrence
and fate | Review|. J Environ Monit 13: 20-31. http://dx.doi.org/10.1039/c0em00373e.
Bailey. SA; Zidelt RH; Perry. RW. (2004). Relationships between organ weight and body/brain weight in
the rat: What is the best analytical endpoint? Toxicol Pathol 32: 448-466.
http://dx.doi.org/10.1080/01926230490465874.
Bannach-Brown. A; Przvbvfa. P; Thomas. J; Rice. ASC; Ananiadou. S; Liao. J; Macleod. MR. (2018).
Machine learning algorithms for systematic review: reducing workload in a preclinical review of
animal studies and reducing human screening error (pp. 1-26). bioRxiv.
http://dx.doij ).
Beach. SA; Newsted. XL; Coadv. K; Giet (2006). Ecotoxicological evaluation of
perfluorooctanesulfonate (PFOS) [Review]. Rev Environ Contam Toxicol 186: 133-174.
http://dx.doi.org/10.1007/0-387-32883-1 5.
Boone. L; Mever. D; Cusick. P; Ennui at. D; Bolliger. AP; Everds. N; Meador. V; Elliott. G; Honor. I);
Bounous. I); Jordan. H. (2005). Selection and interpretation of clinical pathology indicators of
hepatic injury in preclinical studies [Review]. Vet Clin Pathol 34: 182-188.
http://dx.doij •165X.2005.tb00041.x.
Buck. RC; Franklin. J; Berge roder. JM; Cousins. IT; de Voogt. P; Jensen. A A; Kannan. K;
Maburv. SA; van Leeuwen. SP. (2011). Perfluoroalkyl and polyfluoroalkyl substances in the
environment: terminology, classification, and origins [Review]. Integr Environ Assess Manag 7:
513-541. http://dx.doi.or! 32/ieam.258.
Calafat. AM; Kato. K; Hubbard. K; Ji \ * * otelho. JC; Won 1 1 \ (2019). Legacy and alternative per-
and polyfluoroalkyl substances in the U.S. general population: Paired serum-urine data from the
2013-2014 National Health and Nutrition Examination Survey. Environ Int 131: 105048.
http://dx.doi.Org/10.10.l.6/i.envint.2019.105048.
Calafat. AM; Woa 1 * \ K^klenvik. Z; Reidv. JA; Needham. LL. (2007). Polyfluoroalkyl chemicals in
the US population: Data from the National Health and Nutrition Examination Survey (NHANES)
2003-2004 and comparisons with NHANES 1999-2000. Environ Health Perspect 115: 1596-
1602. http://dx.doi.org/10.1289/ehp..1.0598.
Carlson. LA; Angrish. M; Shirke. AY; Radke. EG; Schulz. B; Kraft. A; Judson. R; Patlewicz. G; Blain.
R; Lin. C; Vetter >. i > .neris. C; Hartman. P; Hubbaui i I \t iiaga. X; Davis. A; Dishaw. LV;
Druwe. IL; Hollinger. H; Jones. R; Kaiser. JP; Lizarraga, L; Noves. PD; Taylor. M; Shapiro. AJ;
Williams. AJ; Thayer. KA. (2022). Systematic evidence map for over one hundred and fifty per-
and polyfluoroalkyl substances (PFAS) [Review]. Environ Health Perspect 130: 56001.
http://dx.doi.org/10.1289/EH.P10343.
CERI (Chemicals Evaluation and Research Institute. Japan). (2002a). Bacterial reverse mutation test of T-
7701. (3M-MPCA-00215756; K01-2703). Japan: Sumitomo 3M Limited.
CERI (Chemicals Evaluation and Research Institute. Japan). (2002b). Chromosomal aberration test of T-
7701 using cultured mammalian cells. (3M-MPCA-00215776; K06-0937). Japan: Sumitomo 3M
Limited.
CERI (Chemicals Evaluation and Research Institute. Japan). (2002c). Twenty-cight-day repeated-dose
oral toxicity study ofT-7701 in rats. (3M-MPCA-00215660; B11-0691). Japan: Sumitomo 3M
Limited.
Cohen. AM; Hersh. WR; Peterson. K; Yen. PY. (2006). Reducing workload in systematic review
preparation using automated citation classification. J Am Med Inform Assoc 13: 206-219.
http://dx.doij 7iamia.Ml.929.
DeWitt. JC. (2015). Toxicological effects of perfluoroalkyl and polyfluoroalkyl substances. New York.
NY: Springer International, http://dx.doi 7/978-3-3.1.9-155.1.8-0.
23
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Duan i. H; Yao. Y; Meng. Y; Li. Y. (2020). Distribution of novel and legacy pe r-/po 1 y fl u o roal k y 1
substances in serum and its associations with two glycemic biomarkers among Chinese adult men
and women with normal blood glucose levels. Environ Int 134: 105295.
http://dx.doi.' >/i .envint.20.1.9. .1.05295.
ECHA (European Chemicals Agency). (2020). Information on Chemicals | Database|. Helsinki. Finland:
European Chemicals Agency, European Union. Retrieved from
https://echa.europa.eu/infomiation-on-chemicals
EMEA (European Medicines Agency). (2008). Non-clinical guideline on drug induced hepatotoxicity.
(Doc. Ref. EMEA/CHMP/SWP/150115/2006). London, UK.
http://www.ema.europa.eu/docs/en GB/document library/Scientific guideline/2009/09/WC5000
03355.pdf.
Evidence Partners. (2022). DistillerSR. Retrieved from
https://www.evidencepartners.com/products/distillersr-svstematic-review-software/
Fenton. SE; Ducatman. A; Boobis. A; DeWitt. JC; Lau. C; Ng. C; Smith. JS; Roberts. SM. (2021). Per-
and polyfluoroalkyl substance toxicity and human health review: Current state of knowledge and
strategies for informing future research [Review]. Environ Toxicol Chem 40: 606-630.
http://dx.doi.org/10.1002/etc.4890.
Hall. AP; Elcombe. €R; Foster. JR; Harada. T; Kaufmann. W; Knippel. A; Kuttler. K; Malarkev. DE;
Maronpot. RR: Nishikawa. A; Nolte. T; Schulte. A; Strauss. k. MJ. (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/ 70.1.926233.1.2448935.
Hekster. FM; Laane. RWP. M; de Voogt. P. (2003). Environmental and toxicity effects of
perfluoroalkylated substances [Review]. In GW Ware (Ed.), Reviews of environmental
contamination and toxicology: Continuation of residue reviews (pp. 99-121). New York, NY:
Springer-Verlag. http://dx.doi.or:> (0 100 ' 0-38? J I 1 J
Howard. BE; Phillips. J: Miller. K; Tandon. A: Mav. I); Shah. MR; Holmgren. S; Pelch. KE: Walker. V;
Roonev. AA; Macleod. M; Shah. RR; Thaver. K (2016). SWIFT-Review: A text-mining
workbench for systematic review. Syst Rev 5: 87. http://dx.doix i1 10 II ^ ^s.1.3643-01 c 0.63-z.
Howard. BE; Phillips don. A; Maharana. A; Elmore. R; Mav. I); Sedvkh. A; Thayer. K; Merrick.
BA; Walker. V; Roonev. A; Shah. RR. (2020). SWIFT-Active Screener: Accelerated document
screening through active learning and integrated recall estimation. Environ Int 138: 105623.
http://dx.doi.' >/i .envint.2020. .1.05623.
L eng. Y; Xie. Z; Zeng. F. (2017). 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/srep43380.
Magnuson. K; Cawlev. M; Reillv. D; Varghese. A. (2018). Improving efficiency of systematic reviews
through machine learning for automated record deduplication and text analytics for iterative
keyword streamlining. Poster presented at Society of Toxicology 57th Annual Meeting and
ToxExpo, March 11-15, San Antonio, Texas.
NASEM (National Academies of Sciences. Engineering, and Medicine). (2018). Progress toward
transforming the Integrated Risk Information System (IRIS) program: A 2018 evaluation.
Washington, DC: National Academies Press, http://dx.doi.oi '226/25086.
NASEM (National Academies of Sciences. Engineering, and Medicine). (2021). Review of U.S. EPA's
ORD staff handbook for developing IRIS assessments: 2020 version. Washington, DC: National
Academies Press, http://dx.doi.org/10.17226/26289.
Pelch. KE; Reade. A; Wolffe. TAM; Kwiatkowski. CF. (2019). PFAS health effects database: Protocol
for a systematic evidence map [Review]. Environ Int 130: 104851.
http://dx.doi.' >/i .envint.20.1.9.05.045.
PFASToxDatabase, T, he Endocrine Disruption Exchange. PFAS systematic evidence map.
PFASToxDatabase, The Endocrine Disruption Exchange, https://pfastoxdatabase.org/.
24
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Radke. E; Wright. Ml; Christensen. K; Lin. CJ; Goldstot Glenn. 6; Thayer. K. (2022).
Epidemiology evidence for health effects of 150 per- and polyfluoroalkyl substances: A
systematic evidence map [Review]. Environ Health Perspect 130: 96003.
http://dx.doi.org/10.1289 I H!M I IS>,
Radke. EG; Braun. JM; Meeker. J I): Cooper. GS. (2018). Phthalate exposure and male reproductive
outcomes: A systematic review of the human epidemiological evidence [Review]. Environ Int
121: 764-793. http://dx.doto i.envint.20.18.07.029.
Sciome (Sciome. LLC ). (2021). SWIFT-Active Screener. Available online at
https://www.sciome.com/swift-activescreener/ (accessed
Song. N ( nig. S: Zlm > hen. 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.tH ' 10 IPIo/i.envint.20IS 01 010.
Sunderland. EM; Hu. XC; Dassuncao. C; Tokranov. AK; Wagner. CC; Allen. JG. (2019). A review of the
pathways of human exposure to poly- and perfluoroalkyl substances (PFASs) and present
understanding of health effects [Review]. J Expo Sci Environ Epidemiol 29: 131-147.
http://dx.doi.org/10.103 8/s41370-018-0094- .1..
Tableau Software. LLC. . (2023). Tableau [Computer Program], Retrieved from
https ://www. tableau .com/
U.S. EPA (U.S. Environmental Protection Agency). Health & Environmental Research Online (HERO).
https://lieronet.epa.gov/lieronet/index.cftn/content/liome.
U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation of
biological values for use in risk assessment [EPA Report]. (EPA600687008). Cincinnati, OH.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm ?deid=34855.
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/recordisplav.cfm7debi Mnn3&( iM* • \ \ ' i^-"-)&CFTOKEN=25
0063.1.7.
U.S. EPA (U.S. Environmental Protection Agency). (2000). Supplementary guidance for conducting
health risk assessment of chemical mixtures (pp. 1-209). (EPA/630/R-00/002). Washington, DC:
U.S. Environmental Protection Agency, Risk Assessment Forum.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm ?deid=20533.
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/ftles/2014-12/documents/rfd-ftnal.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/ftles/2013-09/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/ftles/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/bench.mark-dose-technical-guidance.
U.S. EPA (U.S. Environmental Protection Agency). (2018). Chemistry Dashboard. Washington. DC.
Retrieved from https://comptox.epa.gov/dashboard
U.S. EPA (U.S. Environmental Protection Agency). (2019a). Nation.
https://lieronet.epa.gov/lieronet/index.cftn/proiect/page/proiect id/28.1. .1..
25
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
U.S. EPA (U.S. Environmental Protection Agency). (2019b). PFAS 150.
https://lieronet.epa.gov/lieronet/index.cfin/proiect/page/proiect id/2826.
(U.S. Environmental Protection Agency). (2020a). Benchmark Dose Software (BMDS).
Version 3.2: User guide [EPA Report]. (EPA/600/R-20/216). Washington, DC: U.S.
Environmental Protection Agency, Office of Research and Development.
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey= ST2.txt.
U.S. EPA (U.S. Environmental Protection Agency). (2020b). ORD staff handbook for developing IRIS
assessments (public comment draft) [EPA Report]. (EPA/600/R-20/137). Washington, DC: U.S.
Environmental Protection Agency, Office of Research and Development, Center for Public
Health and Environmental Assessment.
https://cfpub.epa.gov/ncea/iris drafts/recordisplav.cfm?deid=350086.
(U.S. Environmental Protection Agency). (2020c). PFAS 150 (2020).
https://liawc.epa.gOv/assessment/.l.00500085/.
(U.S. Environmental Protection Agency). (2020d). PFAS 150 (2020) downloads.
https://liawc.epa.gov/assessment/100500085/downloads/.
(U.S. Environmental Protection Agency). (2021a). CompTox Chemicals Dashboard:
Perfluoropropanoic acid 422-64-0 | DTXSID8059970.
https://eomptox.epa.gov/d3shboard/dsstoxdb/results7seardFDT.XSID8059970.
U.S. EPA (U.S. Environmental Protection Agency). (202 lb). Health Assessment Workspace
Collaborative (HAWC). https://liawc.epa.gov/portal/.
U.S. EPA (U.S. Environmental Protection Agency). (202 lc). 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). (202 Id). 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/recordisplav.cfm ?deid=350888.
U.S. EPA (U.S. Environmental Protection Agency). (202 le). Toxicological review of perfluorobutanoic
acid (PFBA) and related compound ammonium perfluorobutanoic acid (public comment and
external review draft, Aug 2021) [EPA Report]. (EPA/635/R-20/424a). Washington, DC: U.S.
Environmental Protection Agency, Integrated Risk Information System.
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey= BF.txt.
U.S. EPA (U.S. Environmental Protection Agency). (2022a). IRIS toxicological review of
perfluorobutanoic acid (PFBA) and related salts. (EPA/635/R-22/277F).
U.S. EPA (U.S. Environmental Protection Agency). (2022b). ORD staff handbook for developing IRIS
assessments [EPA Report]. (EPA 600/R-22/268). Washington, DC: U.S. Environmental
Protection Agency, Office of Research and Development, Center for Public Health and
Environmental Assessment.
https://cfpub.epa.gov/ncea/iris drafts/recordisplav.cfin?deid=356370.
Whalan. IE. (2015). A toxicologist's guide to clinical pathology in animals: Hematology, clinical
chemistry, urinalysis. Switzerland: Springer International Publishing.
http://dx.doi.org/10.1007/978-3-319-15853-2.
Yost EE; Euling. eaver. JA; Beverly. BEJ; Keshava. N; Mudipalli. A; Arzuaga. X; Blessinger. T;
Dishaw. L; Hotchkiss. A; Makris. SL. (2019). Hazards of diisobutyl phthalate (DIBP) exposure:
A systematic review of animal toxicology studies [Review]. Environ Int 125: 579-594.
http://dx.doij >/i.envint.20.1.8.09.038.
26
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
APPENDIX A. SYSTEMATIC LITERATURE SEARCH METHODS AND RESULTS
Methods
The following describes the systematic review methods used to collect epidemiological
and toxicological evidence for -150 PFAS as part of the larger PFAS systematic review effort
described in Carlson et al. (2022); Radke et al. (2022)3. The methods outlined below are taken
from (Carlson et al.. 2022) and further details can be found directly in the published manuscript.
Perfluoropropanoic acid (PFPrA) was part of the list of 150 PFAS, and for the purposes of this
summary, we isolated the PFPrA-specific results found as a result of the processes outlined
below.
Populations, Exposures, Comparators, and Outcomes (PECO) Criteria and Supplemental
Material Tagging
PECO criteria are used to focus the scope of an evidence map or systematic review by
defining the research question(s), search terms, and inclusion/exclusion criteria. The PECO
criteria for PFPrA are presented in Table A-l. In addition to PECO-relevant studies, studies that
did not meet PECO criteria but contained "potentially relevant" supplemental material were
tracked during the literature screening process. Supplemental material was tagged by category, as
outlined in Table A-2. Note that "supplemental" material does not refer to findings contained in
the supplement of papers identified.
Literature Search and Screening Strategies
Database Search Term Development
Chemical search terms were used to search for relevant literature in the databases listed
below. The detailed search strategy for each database, including specific search stings are
presented in the supplemental materials of Carlson et al. (2022).
• PubMed (National Library of Medicine)
• Web of Science (Thomson Reuters)
• ToxLine via TOXNET (included in the 2019 search; no longer operational in the
2020 or 2021 search updates)5
The literature search for the -150 PFAS consisted only of the chemical name, synonyms,
and trade names and no additional limits, with the exception of the Web of Science (WoS) search
strategy. Due to the specifics of searching WoS, a chemical name-based search can retrieve a
very large number of off-topic references. Given the number of PFAS included in the 150 PFAS
screening effort, a more targeted WoS search strategy was used to identify the records most
likely applicable to human health (see supplemental materials of Carlson et al. (2022)). Chemical
synonyms for PFAS were identified by using synonyms in the Dashboard ( 32la)
indicated as "valid" or "good." The preferred chemical name (as presented in the Dashboard),
CASRN, and synonyms were then shared with EPA information specialists who used these
5 As part of a broader National Library of Medicine (NLM) reorganization, TOXNET has moved and most of NLM's
toxicology information services have been integrated into other NLM products and services.
27
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
inputs to develop search strategies tailored for PubMed, Web of Science, and ToxLine (see
supplemental materials of Carlson et al. (2022)).
Database Searches
The database searches were conducted by an EPA information specialist in August 2019
for the 150 PFAS, and searches were updated in December 2020 and again in December 2021.
All records were stored in EPA's Health and Environmental Research Online (HERO) database
v ii \ 2019a. b). The HERO database
(https://heronet.epa.gov/heronet/index.cfm/content/home) (U.S. EPA.) is used to provide access
to the references used in the EPA's scientific assessments, including this effort. After
deduplication in HERO using unique identifiers (e.g., PMID, WoSID, or DOI) and citations, the
references went through an additional round of deduplication using ICF's Deduper tool
(described in detail in the supplemental materials of Carlson et al. (2022). "DeDuper"), which
uses a two-phase approach to identify duplicates by a) locating duplicates using automated logic
and b) employing machine learning built from Python's Dedupe package to predict likely
duplicates which are then verified manually (Maenuson et al.. 2018). Following deduplication,
SWIFT-Review software (Sciome. 2021; Howard et al.. 2016) was used to identify which of the
unique references were most relevant for human health risk assessment. In brief, SWIFT-Review
was used to filter the unique references based on the software's preset literature search strategies
(titled "evidence stream"). These evidence streams were developed by information specialists
and can be used to separate the references most relevant to human health from those that are not
(e.g., environmental fate studies). References are tagged to a specific evidence stream if the
search terms from that evidence stream appear in the title, abstract, keyword, and/or medical
subject headings (MeSH) fields of that reference. For the PFAS 150 SEM, the following SWIFT-
Review evidence stream were applied: human, animal models for human health, and in vitro
studies. Specific details on the evidence stream search strategies are available through Sciome's
SWIFT-Review documentation at https://www.sciome.com/wp-
content/uploads/2019/08/SWIFT-Review-Search-Strategies-Evidence-Stream.docx. Studies not
retrieved using the search strategies were not considered further.
Other Resources Consulted
The literature search strategies described above are intentionally broad; however, it is still
possible that some studies were not captured (e.g., cases where the specific chemical is not
mentioned in title, abstract, or keyword content; "gray" literature that is not indexed in the
databases listed above). Additionally, if incomplete citation information was provided (e.g., if
reference lists searched did not include titles) no additional searching was conducted. Thus, in
addition to the databases identified above, the sources below were used to identify studies that
could have been missed during the database searches. Additional descriptions of these sources
can be found in Table 4 of Carlson et al. (2022).
• Reference list from the PFAS-Tox Database, a 2019 evidence map of 29 PFAS
(Pelch et al.. 2019). available at
https://public.tableau.eom/profile/the.endocrine.disruption.exchange#l/vizhome/P
FASToxDatabase/PFASDatabase-BETA and https://pfastoxdatabase.org/.
( oxDatabase)
28
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
• Reference lists from all PECO-relevant animal and epidemiological studies
identified in the database searches meeting PECO criteria (see supplemental
materials of Carlson et al. (2022))
• National Toxicology Program (NTP) database of study results and research
projects. The was accomplished by personal communication with NTP rather than
manual search of the NTP database for all the PFAS included in the evidence
map.
• References from EPA's CompTox Chemicals Dashboard ToxValDB (Toxicity
Values Database) (U.S. EPA. 2018) to identify studies or assessments that present
POD information. ToxValDB collates publicly available toxicity dose-effect
related summary values typically used in risk assessments. Many of the PODs
presented in ToxValDB are based on gray literature studies or assessments not
available in databases such as PubMed and WoS, etc. It is important to note that
ToxValDB entries have not undergone quality control to ensure accuracy or
completeness and may not include recent studies.
o ToxValDB includes POD data collected from data sources within ACToR
(Aggregated Computational Toxicology Resource) and ToxRefDB
(Toxicity Reference Database) and no-observed and lowest-observed
(adverse) effect level (NOEL, NOAEL, LOEL, LOAEL) data extracted
from repeated dose toxicity studies submitted under REACH
(Registration, Evaluation, Authorisation and Restriction of Chemicals).
Also included are reference dose and concentration values (RfDs and
RfCs) from EPA's Integrated Risk Information System (IRIS) and dose
descriptors from EPA's Provisional Peer-Reviewed Toxicity Values
(PPRTV) documents. Acute toxicity information in ToxValDB comes
from several sources, including Organization for Economic Cooperation
and Development (OECD) eChemPortal, National Library of Medicine
(NLM), Hazardous Substances Data Bank (HDSB), ChemlDplus via EPA
Toxicity Estimation Software Tool (TEST), and the European Union (EU)
Joint Research Centre (JRC) AcutoxBase and the EU COSMOS project
and the European Chemicals Agency (ECHA) registration dossiers to
identify data submitted by registrants, available at
http://echa.europa.eu/information-on-chemicals/information-from-
existing-substances-regulation. (ECHA, 2020)
Records from these other sources were uploaded into DistillerSR (Evidence Partners.
2022) and annotated with respect to source of the record. The specific methods and results for
searching each source are described below. Results of searches of these sources is summarized to
include the source type or name, the search string (when applicable), the number of results
present within the resource, and the URL (when available and applicable).
ECHA
A search of the ECHA registered substances database was conducted using the CASRN.
The registration dossier associated with the CASRN was retrieved by navigating to and clicking
29
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
the eye-shaped view icon displayed in the chemical summary panel. The General Information tab
and all subpages under the Toxicological Information tab were downloaded in PDF format,
including all nested reports that had unique URLs. In addition, the data were extracted from each
dossier page and used to populate an Excel tracking sheet with this data. Extracted fields
included data from the general information page regarding the registration type and publication
dates, and on a typical ECHA dossier page the primary fields reported in the administrative data,
data source, and effect levels sections. Each study summary resulted in more than one row in the
tracking sheet if more than one data source or effect level was reported.
At this stage, each reference was reviewed for inclusion based on PECO criteria. ECHA
dossiers without information under the ToxCategory column were excluded from review because
these refer to data extracted from the General Information tab. Toxicological and end point
summary pages, study protocols, and dossiers with data waiving were also excluded from
review. When a reference that was considered relevant reported data from a named study or lab
report, a citation for the full study was either retrieved or generated in HERO and verified that it
was not already identified from the peer-reviewed literature search prior to moving forward to
screening in DistillerSR. If citation information was not available and a full text could not be
retrieved, ECHA and ToxValDB references were compared using information on the chemical,
points of departure, study type, species, strain, sex, exposure route and method, and critical effect
to determine whether any of these references were previously accounted for in ToxValDB. When
there were no overlaps between references, a citation was created in HERO using the
information provided in the ECHA dossier. The generated PDF for the dossier was used as the
full text for screening, and these citations were annotated accordingly for Tableau and HAWC
visualizations by adding "(ECHA)" to the citation.
EPA CompTox chemical dashboard (ToxValDB)
ToxValDB data was retrieved for the PFAS chemicals from the EPA CompTox
Dashboard (U.S. EPA. 2018). Data available from the Hazard tab for each chemical was
exported from the Dashboard by U.S. EPA staff and provided as an Excel file output. Using this
ToxValDB POD summary file, citations were identified for references that apply to human
health PODs. A citation for each reference, except those indicated as "ECHA" or "ECHA
IUCLID," was either retrieved or generated in HERO and verified that it was not already
identified from the database search prior to moving forward to screening in DistillerSR.
References in ToxValDB described as from an ECHA or ECHA IUCLID source were
confirmed to be accounted for in the ECHA results retrieved above. A comparison was
performed between 25% of the ECHA references from ToxValDB and the full ECHA results
retrieved above, and although the comparison noted discrepancies (5 out of 34), these were found
to be inaccuracies in ToxValDB, most likely because the data was removed or modified during
an update to ECHA since the last time ToxValDB imported ECHA data. That is, the ECHA
dossiers retrieved above were determined to be more accurate and up to date than the ToxValDB
ECHA entries and could supersede the ECHA data from ToxValDB.
Screening and Tagging Process
After selection of evidence steams and chemicals in SWIFT-Review as described in the
"Database Searches" section, the filtered studies were imported into SWIFT-Active Screener
(version 1.061; Sciome LLC) for title or abstract (TIAB) screening. SWIFT-Active Screener is a
30
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
web-based collaborative software application that uses active machine-learning approaches to
reduce the screening effort (Howard et at.. 2020). The screening process was designed to
prioritize records that appeared to meet PECO criteria or that included supplemental material
content based on TIAB content (i.e., both types of records were screened as "include" for active-
learning purposes). Studies were screened in SWIFT-Active Screener until the software indicated
a likelihood of 95% that all relevant studies had been captured. This threshold is comparable to
human error rates (Baimach-Brown et at., 2018; Howard et al., 2016; Cohen et al., 2006) and is
used as a metric to evaluate machine-learning performance. Any studies in "partially screened"
status at the time of reaching the 95% threshold were fully screened.
Studies that met these criteria from TIAB screening were then imported into DistillerSR
(Evidence Partners. 2022) for more specific TIAB tagging (i.e., to separate studies meeting
PECO criteria versus supplemental content and to tag the specific type of supplemental content
and, if necessary, the chemical). Supplemental content tags are described in Table A-2. For
studies meeting PECO criteria at the DistillerSR TIAB level, full text articles were retrieved
through EPA's HERO database. References that could be retrieved within 45 days were
identified to be unavailable.
Studies identified via the gray literature searches were imported directly into DistillerSR
at the TIAB phase. References identified in the gray searches that had previously been screened
as not relevant to PECO criteria at either the SWIFT-Review or SWIFT-Active Screener stage
were rescreened in Distiller.
Two independent reviewers conducted each level of screening (TIAB and full text). At all
levels (SWIFT-Active Screener TIAB, DistillerSR TIAB, and DistillerSR full-text review), any
conflicts in screening were resolved by discussion between the two independent reviewers; a
third reviewer was consulted if any conflicts remained thereafter. Conflicts between screeners in
applying the supplemental tags were resolved by discussion at both the TIAB and full-text levels,
erring on the side of over tagging at the TIAB level. At the TIAB level, articles without an
abstract were screened based on title (title should indicate clear relevance), and number of pages
(articles two pages or fewer in length were assumed to be records with no original data) For
additional information, please see Table A-2 for supplemental categorization information. All
studies identified as supplemental material at TIAB and full-text levels were tagged to their
respective chemical(s) using the preferred chemical names. All studies identified as PECO were
tagged to the preferred chemical name after the full-text screening stage. A caveat to tagging at
the TIAB level was that tagging was based only on information provided in the abstract and
could therefore miss additional details that may have been provided in the full text of the
manuscript. Additionally, sources that did not list a specific PFAS in the TIAB (i.e., included
terms like "PFAS") were tagged to "chemical not specified." However, if any PFAS were
specified, they were tagged and the "chemical not specified" tag was not selected, even though it
was possible that additional PFAS chemicals were reported in the full text. All chemical tagging
was reviewed by an expert in chemistry (with a doctoral or similar degree). Where chemical
identity presented in the manuscript was unclear, the original authors were contacted to resolve
the chemical species.
31
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Data Extraction of Study Methods and Findings
Animal Toxicology Studies
Studies that met PECO criteria after full-text review were summarized using custom
forms (a standard operating procedure for populating the forms is included in the supplemental
material of Carlson et al. (2022)) in DistillerSR. For animal studies, the following study
summary information was captured in a literature inventory: PFAS assessed, study type [acute
(<24 hours), short term (1-30 days), subchronic (30-90 days), chronic (>90 days),
developmental, peripubertal, multigenerational], route of exposure, species, sex, and health
system(s) assessed. For epidemiological studies, the following study summary information was
captured in a literature inventory: PFAS assessed, sex, population, study design, exposure
measurement (e.g., blood, feces), and health system(s) assessed. Summaries were then extracted
into DistillerSR by one team member, and the extracted data were checked for quality by at least
one other team member. The data from these summary literature inventories were exported from
DistillerSR to an Excel format and then modified and transformed using Excel's 'Get and
Transform' features for import into Tableau visualization software (https://www.tableau.com/)
(version 2019.4; Tableau Software LLC) (Tableau Software. 2023). These data transformations
include pivoting multiple columns of data to single columns, appending data from multiple
literature inventories, and merging detailed reference information and chemical ID information
into the dataset.
The literature inventory was used to prioritize animal toxicological studies with exposure
to the 150 PFAS for repeat dose studies of 21-day and longer durations, or with study designs
focusing on exposure windows targeting reproduction or development. Studies meeting these
exposure timing and duration parameters were moved forward for study evaluation (described in
the next section) and endpoint-level data extraction. Animal toxicology studies not meeting these
criteria did not move forward and were summarized at the literature inventory level only.
Data extraction was conducted for prioritized animal toxicology studies by two members
of the evaluation team using EPA's version of Health Assessment Workspace Collaborative
(HAWC) (\ v < < \ _0C I h), a free and open source web-based software application designed to
manage and facilitate the process of conducting literature assessments. Data extracted included
basic study information (e.g., full citation, funding, author-reported conflicts of interest);
experiment details (e.g., study type, chemical name, chemical source, and purity); animal group
specifics (species, strain, sex, age at exposure and assessment, husbandry); dosing regimen;
endpoints evaluated; and results (qualitative or quantitative) by endpoint. Authors were not
contacted for information that was not reported in a study. Data extraction was performed by one
member of the evaluation team (primary extractor) and checked by a second member for
completeness and accuracy (secondary extractor). Data extraction results were used to create
HAWC visualizations (e.g., exposure-response arrays) by health system and effect for each
PFAS chemical. The detailed HAWC extractions for animal studies are available for download
from EPA HAWC in Excel format at https://hawcprd.epa.eov/assessment/100500085
/downloads/ ( )20d). The data extraction output will also be available as an excel file
from the Dashboard ToxValDB database in a future release.
32
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Subsequent to HAWC data extraction, an EPA toxicologist reviewed each study to
identify study-level, no-observed-adverse-effect levels [NOAELs] and lowest-observed-adverse-
effect levels [LOAELs], These judgments were made at the individual study level.
Epidemiology Studies
Epidemiological studies did not undergo full endpoint-level data extraction. A more
detailed analysis of these studies, however, is being pursued as part of a separate activity (Radke
et al., 2022).
Study Evaluation
Study evaluation was conducted for prioritized animal toxicological studies (>21 -day
exposure durations or exposure occurring during reproduction or development) and human
epidemiological studies by two reviewers using EPA's version of HAWC (\ v < < \ 1 ^)-
Reviews were made by toxicologists or epidemiologists with multiple years of experience in
developing chemical human health assessments. During study evaluation, in each evaluation
domain, at least two reviewers reached a consensus rating of Good, Adequate, Deficient, Not
Reported or Critically Deficient as defined in HAWC. Key study evaluation considerations were
potential sources of bias (factors affecting the magnitude or direction of an effect in systematic
way) or insensitivity (factors limiting detection of a true effect). Core and prompting questions
used to guide the judgment for each domain are described in more detail in the IRIS Handbook
( 2020b). After a consensus rating was reached, the reviewers considered the identified
strengths and limitations to reach an overall study confidence rating of High, Medium, Low, or
Uninformative for each health outcome. The definitions below follow the standard template
language that is used in systematic evidence maps developed by the EPA and have only been
adjusted, where appropriate, for the specific needs of the PFAS 150 SEM.
• High: A well-conducted study with no notable deficiencies or concerns identified
for the outcome(s) of interest; the potential for bias is unlikely or minimal, and the
study used sensitive methodology. "High" confidence studies generally reflect
judgments of good across all or most evaluation domains.
• Medium: A study where some deficiencies or concerns were noted for the
outcome(s) of interest, but the limitations are unlikely to be of a notable degree.
Generally, "medium" confidence studies will include adequate or good judgments
across most domains, with the impact of any identified limitation not being
judged as severe.
• Low: A study where one or more deficiencies or concerns were noted for the
outcome(s) of interest, and the potential for bias or inadequate sensitivity could
have a significant impact on the study results or their interpretation. Typically,
"low" confidence studies would have a deficient evaluation for one or more
domains, although some "medium" confidence studies may have a deficient rating
in domain(s) considered to have less influence on the magnitude or direction of
the results. Generally, in an assessment context (or a full systematic review), low
confidence results are given less weight in comparison with high or medium
confidence results during evidence synthesis and integration and are generally not
33
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
used as the primary sources of information for hazard identification or derivation
of toxicity values unless they are the only studies available. Studies rated as low
confidence only because of sensitivity concerns about biases toward the null
would require additional consideration during evidence synthesis.
• Uninformative: A study where serious flaw(s) make the results unusable for
informing hazard identification for the outcome(s) of interest. Studies with
critically deficient judgments in any evaluation domain will almost always be
classified as "uninformative" (see explanation above). Studies with multiple
deficient judgments across domains may also be considered uninformative. As
mentioned above, although outside the scope of this SEM, in an assessment or full
systematic review, uninformative studies would not be considered during the
synthesis and integration of evidence for hazard identification or for dose
response but might be used to highlight possible research gaps. Thus, data from
studies deemed uninformative are not depicted in the results displays included in
this SEM.
Rationales for each study evaluation classification, including a description of how
domain ratings impacted the overall study confidence rating, are available in the supplemental
materials of Carlson et al. (2022) and are documented and retrievable in HAWC
(https://hawcprd.epa. gov/
summary/visual/assessment/100500085/animal-studv-evaluation-heatmap) ( Z020c).
Table A-l. Populations, Exposures, Comparators, and Outcomes (PECO) Criteria
Pi:( Oelemenl
Description
Populations
Human: Any population and lifestage (occupational or general population, including children
and other potentially sensitive populations).
Animal: Nonhuman mammalian animal species (whole organism) of any lifestage (including
preconception, in utero, lactation, peripubertal, and adult stages).
Exposures
Relevant forms: PFPrA-specific results isolated from search for ~150 PFAS chemicals +
Nafion represented by -170 PFAS structures and substances previously identified.
Human: Any exposure to PFAS via the oral and inhalation routes because these are the most
relevant routes of human exposure and typically the most useful for developing human health
toxicity values. Studies are also included if biomarkers of PFAS exposure are evaluated
(e.g., measured PFAS or metabolite in tissues or bodily fluids) but the exposure route is unclear
or reflects multiple routes. Other exposure routes, including dermal, and mixture-only studies
(i.e., without effect estimates for individual PFAS of interest) are tracked during title and
abstract screening and are tagged as supplemental material.
Animal: Any exposure to PFAS via oral or inhalation routes. Studies involving exposures to
mixtures are included only if a treatment group consists of exposure to a PFAS alone. Other
exposure routes, including dermal or injection, and mixture-only studies are tagged as
supplemental material.
Comparators
Human: A comparison or referent population exposed to lower levels (or no
exposure/exposure below detection limits) or exposed for shorter periods of time. Worker
surveillance studies are considered to meet PECO criteria, however, even if no referent group is
presented. Case reports describing findings in 1-3 people in nonoccupational or occupational
settings are tracked as supplemental material.
34
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Pi:( Odenum
Dcsci'ipliou
Animal: A concurrent control group exposed to vehicle-only treatment or untreated control
(control could be a baseline measurement).
Outcomes
All health outcomes (cancer and noncancer).
35
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Table A-2. Major categories of "potentially relevant supplemental material"
Csili'Sion
Ik'MTiplion
In vitro, ex vivo, or in silico
"mechanistic" studies
In vitro, ex vivo, or in silico studies reporting measurements related to a health
outcome that inform the biological or chemical events associated with phenotypic
effects, in both mammalian and nonmammalian model systems.
Absorption, distribution,
metabolism, and excretion
(ADME)
ADME studies are primarily controlled experiments where defined exposures
usually occur by intravenous, oral, inhalation, or dermal routes, and the
concentration of particles, a chemical, or its metabolites in blood or serum, other
body tissues, or excreta are then measured. These data are used to estimate the
amount absorbed (A), distributed to different organs (D), metabolized (M), and/or
excreted/eliminated (E) through urine, breath, feces, etc. ADME data can also be
collected from human subjects who have had environmental or workplace
exposures that are not quantified or fully defined. However, to be useful, such data
must involve either repeated measurements over a time period when exposure is
known (e.g., is zero because previous exposure ended) or time- and subject-
matched tissue or excreta concentrations (e.g., plasma and urine, or maternal and
cord blood). ADME data, especially metabolism and tissue partition coefficient
information, can be generated using in vitro model systems. Although in vitro data
may not be as definitive as in vivo data, these studies should also be tracked as
ADME. For large evidence bases it may be appropriate to separately track the in
vitro ADME studies. Note: Studies describing environmental fate and transport or
metabolism in bacteria are not tagged as ADME.
Classical pharmacokinetic
(PK) model Studies, or
physiologically based
pharmacokinetic (PBPK)
modeling studies
Classical PK or dosimetry modelins studies: Classical PK or dosimetry modelins
usually divides the body into just one or two compartments, which are not
specified by physiology, where movement of a chemical into, between, and out of
the compartments is quantified empirically by fitting model parameters to ADME
data.
PBPK or mechanistic dosimetry modelins studies: PBPK models represent the
body as various compartments (e.g., liver, lung, slowly perfused tissue, richly
perfused tissue) to quantify the movement of chemicals or particles into and out of
the body (compartments) by defined routes of exposure, metabolism, and
elimination, and thereby estimate concentrations in blood or target tissues.
Nonmammalian model
systems
Studies in nonmammalian model systems, e.g,,Xenopus species, fish, birds,
Caenorhabditis elegans.
Transgenic mammalian
model systems
Transgenic studies in mammalian model systems.
Non-oral or non-inhalation
routes of administration
Studies in which humans or animals (whole organism) were exposed via a non-oral
or non-inhalation route (e.g., injection, dermal exposure).
Exposure characteristics (no
health outcome assessment)
Exposure characteristic studies which include data that are unrelated to health
outcomes but which provide information on exposure sources or measurement
properties of the environmental agent (e.g., demonstrate a biomarker of exposure).
Mixture studies
Mixture studies that are not considered PECO-relevant because they do not contain
an exposure or treatment group assessing only the chemical of interest. This
category is generally used for experimental studies and generally does not apply to
epidemiological studies where the exposure source may be unclear.
Case reports
Case reports describing health outcomes after exposure will be tracked as
potentially relevant supplemental information when the number of subjects is three
or fewer.
36
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Csili'Sion
Ik'MTiplion
Records u nil no original
data
Records llial do noi coiiiam original dala, such as oilier agencs assessment,
informative scientific literature reviews, editorials, or commentaries.
Conference abstracts
Records that do not contain sufficient documentation to support study evaluation
and data extraction.
European Chemicals Agency
(ECHA) read-across
Data on a nonrelevant chemical that makes inferences about a relevant PFAS
chemical.
Presumed duplicate
Duplicate studies (e.g., published vs. unpublished reports) identified during data
extraction and study quality evaluation.
37
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
APPENDIX B. BENCHMARK DOSE MODELING RESULTS
Modeling Procedure for Continuous Noncancer Data
Benchmark dose (BMD) modeling of continuous data is conducted with EPA's BMD
Software (BMDS, Version 3.2). All continuous models available within the software are fit using
a benchmark response (BMR) of 1 standard deviation (SD) relative risk or 10% extra risk when a
biologically determined BMR is available (e.g., BMR of 10% relative deviation [RD] for body
weight based on a biologically significant weight loss of 10%), as outlined in the Benchmark
Dose Technical Guidance (U.S. EPA. 2002). A BMR of 10% RD for relative liver weights is
considered a minimally biologically significant response in adult animals and was applied in this
assessment for BMD modeling purposes. The default BMR of 1 SD also was applied for
comparison. An adequate fit is judged on the basis of a %2 goodness-of-fitp-value (p> 0.1), the
magnitude of the scaled residuals near the BMR, and a visual inspection of the model fit. In
addition to these three criteria forjudging adequacy of model fit, whether the variance across
dose groups is homogeneous is determined. If a homogeneous variance model is deemed
appropriate on the basis of the statistical test provided by BMDS (i.e., Test 2), the final BMD
results are estimated from a homogeneous variance model. When the test for homogeneity of
variance is rejected (p < 0.1), the model is rerun with the variance modeled as a power function
of the mean to account for this nonhomogeneous variance. If this nonhomogeneous variance
model does not adequately fit the data (i.e., Test 3;p<0. 1), the data set is considered unsuitable
for BMD modeling. Among all models providing adequate fit, the lowest BMD lower confidence
limit (BMDL) or the benchmark concentration lower confidence limit (BMCL) is selected if the
BMDL or BMCL estimate from different models vary by greater than threefold. Otherwise, the
BMDL or BMCL from the model with the lowest Akaike's information criterion (AIC) is
selected as a potential point of departure (POD) from which to derive the oral reference dose or
inhalation reference concentration (RfD or RfC).
Modeling Procedure for Dichotomous Noncancer Data
The BMD modeling of dichotomous data is conducted with the EPA's BMDS, Version
3.2. The Gamma, Logistic, Log-Logistic, Probit, Log-Probit, Hill, Multistage, and Weibull
dichotomous models available within the software are fit using a BMR of 10% extra risk. In
general, the BMR should be near the low end of the observable range of increased risk in the
study. BMRs that are too low can result in widely disparate BMDL estimates from different
models (i.e., model dependence is high). Adequacy of model fit is judged on the basis of the %2
goodness-of-fit p-v alue (p > 0.1), magnitude of scaled residuals (absolute value <2.0), and visual
inspection of the model fit. Among all models providing adequate fit, the BMDL from the model
with the lowest AIC is selected as a potential POD, if the BMDLs are sufficiently close (less than
approximately threefold); if the BMDLs are not sufficiently close (greater than approximately
threefold), model dependence is indicated, and the model with the lowest reliable BMDL is
selected.
38
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Model Predictions for Increased Relative Liver Weight in Male Rats (CERI. 2002c)
The procedure outlined above for continuous data was applied to the data for increased
relative liver weight in adult male Crj :CD (SD) IGS (SPF) rats exposed to PFPrA for 28 days via
gavage (CERI. 2002c). The BMD modeling results are summarized in Table B-l and Figure B-l.
The constant variance model provided adequate fit to the variance data, and adequate fit to the
means was provided by some included models. The BMDLs for the models providing adequate
fit were sufficiently close (differed by -Valueb
Scaled
Residual at
Dose Nearest
BMD
AIC
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
Exponential 2
3
0.005050662
2.262925904
27.21560892
122.4887085
96.33617836
Exponential 3
3
0.005050662
2.262802575
27.2156088
122.4882507
96.3740821
Exponential 4
2
0.594805118
0.791602093
17.43812865
20.47462463
9.993505528
Exponential 5
2
0.594801226
0.79264636
17.43814174
20.5040741
9.9951759
Hill*
2
0.839332468
0.393780633
16.7493825
14.69985034
6.271785406
Polynomial 4
3
0.007285773
2.184615038
26.42770209
109.9206543
83.70900173
Polynomial 3
3
0.007285773
2.184614997
26.42770209
109.9206543
83.70900173
Polynomial 2
3
0.007285773
2.18461501
26.42770209
109.9206543
83.70900173
Power
3
0.007285773
2.184615695
26.42770209
109.9206792
83.7162667
Linear
3
0.007285773
2.184615019
26.42770209
109.9206543
83.70900173
aCERI (2002c).
bValues <0.10 failed to meet conventional goodness-of-fit criteria.
* Selected model (bold). Lowest AIC among models with adequate fit was selected (Hill).
AIC = Akaike's information criterion; BMD = maximum likelihood estimate of the dose associated with the selected
BMR; BMDL = 95% lower confidence limit on the BMD (subscripts denote BMR: i.e., 10RD = dose associated
with 10% relative deviation from the control); BMR = benchmark response; df = degree(s) of freedom.
39
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
5
4.5
4
3.5
3 ^
2.5
2
1.5
1
0.5
0
0
Frequeritist Hill Model with BMR of 0.1 Rel. Dev. for the BMD and 0.95 Lower
Confidence Limit for the BMDL
Estimated Probability
^—Response at BMD
c Data
BMD
BMDL
150
Dose
200
300
Figure B-l. Fit of Hill Model to Data for Increased Relative Liver Weight in Adult Male
Crj:CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage (CERL, 2002c)
40
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
BMD Model Output for Figure B-l:
Model Results
Benchmark Dose
BMD
14.69985034
BMDL
6.271785406
BMDU
37.95628174
AIC
16.7493825
Test 4 P-value
0.839332468
D.O.F.
2
Model Parameters
# of Parameters
5
Variable
Estimate
g
3.142501356
V
1.268335058
k
44.62975185
n
Bounded
alpha
0.078376104
Goodness of Fit
Dose
Size
Estimated
Median
Calc'd
Median
Observed
Mean
Estimated
SD
Calc'd SD
Observed
SD
Scaled
Residual
0
6
3.142501356
3.15
3.15
0.27995733
0.19
0.19
0.065609467
5
6
3.270281066
3.24
3.24
0.27995733
0.21
0.21
-0.264944526
20
6
3.534993985
3.58
3.58
0.27995733
0.34
0.34
0.393780633
80
6
3.956647282
3.92
3.92
0.27995733
0.29
0.29
-0.3206458
320
6
4.255595438
4.27
4.27
0.27995733
0.43
0.43
0.126032879
Likelihoods of Interest
# of
Model
Log Likelihood*
Parameters
AIC
A1
-4.199542866
6
20.3990857
A2
-1.541134391
10
23.0822688
A3
-4.199542866
6
20.3990857
fitted
-4.37469125
4
16.7493825
R
-21.96468014
2
47.9293603
* Includes additive constant of -27.56816. This constant was not included in the LL derivation prior to BMDS 3.0.
Tests of Interest
-2*Log(Likelihood
Test
Ratio)
Test df
p-value
1
40.8470915
8
<0.0001
2
5.31681695
4
0.25630676
3
5.31681695
4
0.25630676
4
0.350296769
2
0.83933247
41
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Model Predictions for Increased Serum Alanine Aminotransferase (ALT) in Male Rats
( II. 2002c)
The procedure outlined above for continuous data was applied to the data for increased
serum ALT in adult male Cij :CD (SD) IGS (SPF) rats exposed to PFPrA for 28 days via gavage
(CERI, 2002c). The BMD modeling results are summarized in Table B-2 and Figure B-2. The
constant variance model provided adequate fit to the variance data, and adequate fit to the means
was provided by all included models. The BMDLs for the models providing adequate fit were
not sufficiently close (differ by >threefold), so the model with the lowest BMDL (Hill) was
selected. For increased serum ALT, the BMDLisd of 28 mg/kg-day from this model was
selected.
Table B-2. BMD Modeling Results for Increased Serum ALT in Adult Male Crj:CD (SD)
IGS (SPF) Rats Exposed to PFPrA for 28 days via Gavage"
Model
df
X2 Goodness-of-
Fit />-Valueb
Scaled
Residual at
Dose Nearest
BMD
AIC
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
Exponential 2
3
0.725145
0.495322432
166.1551966
137.6285
103.6144
Exponential 3
3
0.7251452
0.495266835
166.1551956
137.6445
103.6143
Exponential 4
2
0.6611853
-0.108470434
167.6658274
86.36887
34.31788
Exponential 5
2
0.6611853
-0.108457685
167.6658274
86.36948
34.31813
Hill*
2
0.6635677
-0.118454015
167.6586338
85.62671
28.43509
Polynomial 4
3
0.7661823
0.379605768
165.9835564
124.7527
90.53213
Polynomial 3
3
0.7661823
0.379605711
165.9835564
124.7527
90.53213
Polynomial 2
3
0.7661823
0.3796057
165.9835564
124.7527
90.53213
Power
3
0.7661823
0.379605707
165.9835564
124.7526
90.53205
Linear
3
0.7661823
0.379605721
165.9835564
124.7527
90.53213
aCERI (2002c).
bValues <0.10 failed to meet conventional goodness-of-fit criteria.
* Selected model (bold). Lowest AIC among models with adequate fit was selected (Hill).
AIC = Akaike's information criterion; ALT = alanine aminotransferase; BMD = maximum likelihood estimates of
the dose associated with the selected BMR; BMDL = 95% lower confidence limit on the BMD (subscripts denote
BMR: i.e., 1SD = dose associated with 1 standard deviation relative risk from the control); BMR = benchmark
response; df = degree(s) of freedom.
42
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
Frequentist Hill Model with BMR of 1 Std. Dev. for the BMD
and 0.95 Lower Confidence Limit for the BMDL
Estimated Probability
Response at BMD
O Data
BMD
BMDL
Figure B-2. Fit of Hill Model to Data for Increased Serum ALT in Adult Male Crj:CD (SD)
IGS (SPF) Rats Exposed to PFPrA for 28 days via Gavage (CERL 2002c)
43
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
BMD Model Output for Figure B-2:
Model Results
Benchmark Dose
BMD
85.62671486
BMDL
28.43508564
BMDU
239.4949956
AIC
167.6586338
Test 4
/>-value
0.66356774
df
2
Model Parameters
# of
Parameters
5
Variable
Estimate
g
25.89898576
V
22.30697177
k
465.9927831
n
Bounded
alpha
11.99003863
Goodness of Fit
Dose
Size
Estimated
Median
Calc'd.
Median
Observed
Mean
Estimated
SD
Calc'd.
SD
Observed
SD
Scaled Residual
0
6
25.89898576
25
25
3.46266352
5
5
-0.635942931
5
6
26.13579376
27
27
3.46266352
3
3
0.611339889
20
6
26.81698177
27
27
3.46266352
2
2
0.129467182
80
6
29.16744973
29
29
3.46266352
4
4
-0.118454015
320
6
34.98078793
35
35
3.46266352
4
4
0.013590628
Likelihoods of Interest
Model
Log Likelihood*
#of
Parameters
AIC
A1
-79.41919259
6
170.838385
A2
-76.87604927
10
173.752099
A3
-79.41919259
6
170.838385
fitted
-79.82931692
4
167.658634
R
-89.92741703
2
183.854834
*Includes additive constant of-27.56816. This constant was not included in the log likelihood derivation prior to BMDS 3.0.
Tests of Interest
Test
-2 *Log(Likelihood
Ratio)
Test df
p-value
1
26.10273551
8
0.00100861
2
5.08628663
4
0.27855798
3
5.08628663
4
0.27855798
4
0.82024867
2
0.66356774
44
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Model Predictions for Increased Serum Alkaline Phosphatase (ALP) in Male Rats (CERI.
2002c)
The procedure outlined above for continuous data was applied to the data for increased
serum ALP in adult male Cij :CD (SD) IGS (SPF) rats exposed to PFPrA for 28 days via gavage
(CERI 2002c). The BMD modeling results are summarized in Table B-3 and Figure B-3. The
constant variance model provided adequate fit to the variance data, and adequate fit to the means
was provided by some included models. The BMDLs for the models providing adequate fit were
sufficiently close (differed by -Valucb
Scaled
Residual at
Dose Nearest
BMD
AIC
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
Exponential 2
3
0.0473274
-0.467410275
356.8601623
298.0884
180.9615
Exponential 3
3
0.0473274
-0.467408907
356.8601623
298.0878
180.9639
Exponential 4
2
0.2900059
-1.0137466
353.3986597
41.09192
13.77903
Exponential 5*
2
0.5081367
0.446996822
353.3608541
67.12098
20.33889
Hill
2
0.5081348
-0.001142051
353.3608582
34.13729
20.52392
Polynomial 4
3
0.0516934
-0.53163545
356.663328
284.0054
163.4765
Polynomial 3
3
0.0516934
-0.53163533
356.663328
284.0054
163.4765
Polynomial 2
3
0.0516934
-0.531635319
356.663328
284.0054
163.4765
Power
3
0.0516934
-0.531635379
356.663328
284.0054
163.479
Linear
3
0.0516934
-0.531635714
356.663328
284.0054
163.4765
aCERI (2002c).
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
*Selected model (bold). Lowest AIC among models with adequate fit was selected (Exponential 5).
AIC = Akaike's information criterion; ALP = alkaline phosphatase; BMD = maximum likelihood estimates of the
dose associated with the selected BMR; BMDL = 95% lower confidence limit on the BMD (subscripts denote BMR:
i.e., 1SD = 1 dose associated with 1 standard deviation relative risk from the control); BMR = benchmark response;
df = degree(s) of freedom.
45
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
Frequentist Exponential Degree 5 Model with BMR of 1 Std.
Dev. for the BMD and 0.95 Lower Confidence Limit for the
BMDL
700
600
500
£ 400
o
Cu
S 300
U,
200
100
0
¦O
f)
50
100
150
Dose
200
250
300
Estimated Probability
Response at BMD
O Data
BMD
BMDL
Figure B-3. Fit of Exponential Degree 5 Model to Data for Increased Serum ALP in Adult
Male Crj:CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage (CERL
2002c)
46
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
BMD Model Output for Figure B-3:
Model Results
Benchmark Dose
BMD
67.12097645
BMDL
20.33888677
BMDU
Infinity
AIC
353.3608541
Test 4 p-value
0.508136728
df
1
Model Parameters
# of Parameters
5
Variable
Estimate
a
423.9999997
b
0.015025569
c
1.253538871
d
17.99993883
log-alpha
8.607475193
Goodness of Fit
Dose
Size
Estimated
Median
Calc'd.
Median
Observed
Mean
Estimated
SD
Calc'd.
SD
Observed
SD
Scaled Residual
0
6
423.9999997
420
420
73.9757692
48
48
-0.132448211
5
6
423.9999997
428
428
73.9757692
74
74
0.132448229
20
6
423.9999998
424
424
73.9757692
83
83
7.75705E-09
80
6
531.5004808
545
545
73.9757692
79
79
0.446996822
320
6
531.5004808
518
518
73.9757692
107
107
-0.447028664
Likelihoods of Interest
Model
Log
Likelihood*
# of
Parameters
AIC
A1
-171.461476
6
354.922952
A2
-169.651633
10
359.303266
A3
-171.461476
6
354.922952
fitted
-171.6804271
5
353.360854
R
-177.8303481
2
359.660696
* Includes additive constant of-27.56816. This constant was not included in the log likelihood derivation prior to BMDS 3.0.
Tests of Interest
Test
-2*Log(Like-
lihood Ratio)
Test df
/>-value
1
16.35743003
8
0.03754072
2
3.619685881
4
0.45991465
3
3.619685881
4
0.45991465
4
0.437902174
1
0.50813673
47
-------
Human Health Toxicity Values for Perfluoropropanoic Acid
Model Predictions for Increased Hepatocyte Hypertrophy in Male Rats (CERI. 2002c)
The procedure outlined above for dichotomous data was applied to the data for increased
hepatocyte hypertrophy in adult male Crj :CD (SD) IGS (SPF) rats exposed to PFPrA for 28 days
via gavage (CERI. 2002c). The BMD modeling results are summarized in Table B-4 and
Figure B-4. All models provided adequate fit (p-value > 0.10). The BMDLs for the models
providing adequate fit were not sufficiently close (differed by >threefold), so the model with the
lowest BMDL (Multistage Degree 1) was selected. For increased hepatocyte hypertrophy, the
BMDLioer of 7.9 mg/kg-day from this model was selected.
Table B-4. BMD Modeling Results for Increased Hepatocyte Hypertrophy in Adult Male
Crj:CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage"
Model
df
X2 Goodness-
of-Fit p-
Valueb
Scaled
Residual at
Dose Nearest
BMD
AIC
BMDioer
(mg/kg-d)
BMDLo.ier
(mg/kg-d)
Dichotomous Hill
1
0.9994976
2.89723E-06
13.63817081
71.52739
22.82469
Gamma
1
0.9904814
0.000386722
13.63845452
60.62737
16.55655
Log-Logistic
3
1
-7.48247E-10
9.638170614
71.55622
22.82473
Multistage 3
3
0.9980426
0.026748569
9.713472066
51.37595
12.69572
Multistage 2
3
0.9833786
-0.384943695
9.957035722
41.5625
12.92712
Multistage 1*
2
0.3982919
-0.950683507
14.89786409
15.01799
7.890077
Weibull
3
0.9997399
0.013825013
9.657704573
57.16347
15.74647
Logistic
3
1
1.48722E-06
9.638184036
73.04749
31.07119
Log-Probit
2
0.9999999
5.62727E-08
11.63817039
69.91623
22.30673
Probit
3
0.9999977
0.000912285
9.63902246
64.88286
28.34315
aCERI (2002c).
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
* Selected model (bold). Lowest AIC among models with adequate fit was selected (Hill).
AIC = Akaike's information criterion; ALP = alkaline phosphatase; BMD = maximum likelihood estimate of the
dose associated with the selected BMR; BMDL = 95% lower confidence limit on the BMD (subscripts denote BMR:
i.e., 10ER = dose associated with 10% extra risk from the control); BMR = benchmark response; df = degree(s) of
freedom.
48
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
Frequentist Multistage Degree 1 Model with BMR of 10% Extra
Risk for the BMD and 0.95 Lower Confidence Limit for the
BMDL
Estimated Probability
Response at BMD
— — Linear Extrapolation
O Data
BMD
BMDL
Figure B-4. Fit of Multistage Degree 1 Model to Data for Increased Serum ALP in Adult
Male Crj:CD (SD) IGS (SPF) Rats Exposed to PFPrA for 28 Days via Gavage (CERL
2002c)
49
-------
Human Health Toxicity Values for Perflnoropropanoic Acid
BMD Model Output for Figure B-4:
Model Results
Benchmark Dose
BMD
15.01799345
BMDL
7.890076616
BMDU
30.73698127
AIC
14.89786409
/>-value
0.398291862
df
2
Chi2
1.841140441
Slope Factor
0.012674148
Model Parameters
# of Parameters
2
Variable
Estimate
g
1.55762E-08
bl
0.007015619
Goodness of Fit
Dose
Estimated
Probability
Expected
Observed
Size
Scaled Residual
0
1.55762E-08
9.34574E-08
0
6
-0.0003057
20
0.130913301
0.785479804
0
6
-0.9506835
80
0.429504223
2.577025336
2
6
-0.475893
320
0.894072248
5.364433486
6
6
0.8431293
Analysis of Deviance
Model
Log Likelihood
# of Parameters
Deviance
Test d.f.
p-Value
Full Model
-3.81908501
4
-
-
NA
Fitted Model
-5.448932047
2
3.25969407
2
0.1959595
Reduced Model
-15.27634004
1
22.9145101
3
<0.0001
50
------- |