vvEPA
March 2023
EPA Document No.
822P23005
PUBLIC COMMENT DRAFT
Toxicity Assessment and Proposed Maximum
Contaminant Level Goal for Perfluorooctanoic Acid
(PFOA) in Drinking Water
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PUBLIC COMMENT DRAFT
Toxicity Assessment and Proposed Maximum Contaminant Level Goal for
Perfluorooctanoic Acid (PFOA) in Drinking Water
Prepared by:
U.S. Environmental Protection Agency
Office of Water (4304T)
Health and Ecological Criteria Division
Washington, DC 20460
EPA Document Number: EPA 822P23005
March 2023
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Disclaimer
This document is a public comment draft for review purposes only. This information is
distributed solely for the purpose of public comment. It has not been formally disseminated by
the U.S. Environmental Protection Agency. It does not represent and should not be construed to
represent any agency determination or policy. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Acknowledgments
This document was prepared by the Health and Ecological Criteria Division, Office of Science
and Technology, Office of Water (OW) of the U.S. Environmental Protection Agency (EPA).
The agency gratefully acknowledges the valuable contributions of EPA scientists from the OW,
Office of Research and Development (ORD), the Office of Children's Health Protection
(OCHP), and the Office of Land and Emergency Management (OLEM). OW authors of the
document include Brittany Jacobs; Casey Lindberg; Carlye Austin; Kelly Cunningham; Barbara
Soares; Ruth Etzel; and Colleen Flaherty. ORD authors of the document include J. Michael
Wright; Elizabeth Radke; Michael Dzierlenga; Todd Zurlinden; Jacqueline Weinberger; Thomas
Bateson; Hongyu Ru; and Kelly Garcia. OCHP authors of the document include Chris
Brinkerhoff; and Greg Miller (formerly OW). EPA scientists who provided valuable
contributions to the development of the document from OW include Adrienne Keel; Joyce
Donohue (now retired); Amanda Jarvis; James R. Justice; from ORD include Timothy Buckley;
Allen Davis; Peter Egeghy; Elaine Cohen Hubal; Pamela Noyes; Kathleen Newhouse; Ingrid
Druwe; Michelle Angrish; Christopher Lau; Catherine Gibbons; and Paul Schlosser; and from
OLEM includes Stiven Foster. Additional contributions to draft document review from managers
and other scientific experts, including the ORD Toxicity Pathways Workgroup and experts from
the Office of Chemical Safety and Pollution Prevention (OSCPP), are greatly appreciated. The
agency gratefully acknowledges the valuable management oversight and review provided by
Elizabeth Behl (OW); Jamie Strong (formerly OW; currently ORD); Susan Euling (OW);
Kristina Thayer (ORD); Andrew Kraft (ORD); Viktor Morozov (ORD); Vicki Soto (ORD); and
Garland Waleko (ORD).
The systematic review work included in this assessment was prepared in collaboration with ICF
under the U.S. EPA Contracts EP-C-16-011 (Work Assignment Nos. 4-16 and 5-16) and PR-
OW-21-00612 (TO-0060). ICF authors serving as the toxicology and epidemiology technical
leads were Samantha Snow and Sorina Eftim. ICF and subcontractor authors of the assessment
include Kezia Addo; Barrett Allen; Robyn Blain; Lauren Browning; Grace Chappell; Meredith
demons; Jonathan Cohen; Grace Cooney; Ryan Cronk; Katherine Duke; Hannah Eglinton;
Zhenyu Gan; Sagi Enicole Gillera; Rebecca Gray; Joanna Greig; Samantha Goodman; Anthony
Hannani; Samantha Hall; Jessica Jimenez; Anna Kolanowski; Madison Lee; Cynthia Lin;
Alexander Lindahl; Nathan Lothrop; Melissa Miller; Rachel O'Neal; Ashley Peppriell; Mia
Peng; Lisa Prince; Johanna Rochester; Courtney Rosenthal; Amanda Ross; Karen Setty; Sheerin
Shirajan; Raquel Silva; Jenna Sprowles; Wren Tracy; Joanne Trgovcich; Janielle Vidal;
Maricruz Zarco; and Pradeep Raj an (subcontractor).
ICF contributors to this assessment include Sarah Abosede Alii; Tonia Aminone; Caelen
Caspers; Laura Charney; Kathleen Clark; Sarah Colley; Kaylyn Dinh; Julia Finver; Lauren
Fitzharris; Caroline Foster; Jeremy Frye; Angelina Guiducci; Pamela Hartman; Cara Henning;
Audrey Ichida; Caroline Jasperse; Kaedra Jones; Michele Justice; Afroditi Katsigiannakis;
Gillian Laidlaw; Yi Lu; Denyse Marquez Sanchez; Alicia Murphy; Emily Pak; Joei Robertson;
Lucas Rocha Melogno; Andrea Santa-Rios; Alessandria Schumacher; Nkoli Ukpabi; and
Wanchen Xiong.
ii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Contents
Disclaimer i
Acknowledgments ii
Contents iii
Figures vi
Tables xi
Acronyms and Abbreviations xiii
1 Background 1-1
1.1 National Primary Drinking Water Regulation for Per- and Polyfluoroalkyl
Substances under the Safe Drinking Water Act 1-1
1.2 Background on PFAS 1-1
1.3 Evaluation of PFOA Under SDWA 1-2
1.4 Purpose of this Document 1-3
1.5 Chemical Identity 1-5
1.6 Occurrence Summary 1-6
1.6.1 Biomonitoring 1-6
1.6.2 Ambient Water 1-7
1.6.3 Drinking Water 1-8
2 Summary of Assessment Methods 2-10
2.1 Introduction to the Systematic Review Assessment Methods 2-10
2.1.1 Literature Search 2-11
2.1.2 Literature Screening 2-12
2.1.3 Study Quality Evaluation for Epidemiological Studies and Animal
Toxicological Studies 2-13
2.1.4 Data Extraction 2-13
2.1.5 Evidence Synthesis and Integration 2-14
2.2 Dose-Response Assessment 2-15
2.2.1 Approach to POD and RfD Derivation for Non-Cancer Health Outcomes. 2-15
2.2.2 Cancer Assessment 2-17
2.3 MCLG Derivation 2-19
3 Results of the Health Effects Systematic Review and Toxicokinetics Methods 3-1
3.1 Literature Search and Screening Results 3-1
3.1.1 Results for Epidemiology Studies of PFOA by Health Outcome 3-4
3.1.2 Results for Animal Toxicological Studies of PFOA by Health Outcome 3-4
3.2 Data Extraction Results 3-5
3.3 Toxicokinetic Synthesis 3-5
3.3.1 ADME 3-5
3.3.2 Pharmacokinetic Models 3-17
3.4 Non-Cancer Health Effects Evidence Synthesis and Integration 3-25
iii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.1 Hepatic 3-25
3.4.2 Immune 3-93
3.4.3 Cardiovascular 3-143
3.4.4 Developmental 3-190
3.4.5 Evidence for Other Health Outcomes 3-261
3.5 Cancer Evidence Study Quality Evaluation, Synthesis, Mode of Action Analysis
and Weight of Evidence 3-261
3.5.1 Human Evidence Study Quality Evaluation and Synthesis 3-261
3.5.2 Animal Evidence Study Quality Evaluation and Synthesis 3-268
3.5.3 Mechanistic Evidence 3-272
3.5.4 Weight Of Evidence for Carcinogenicity 3-289
3.5.5 Cancer Classification 3-306
4 Dose-Response Assessment 4-1
4.1 Non-Cancer 4-1
4.1.1 Study andEndpoint Selection 4-1
4.1.2 Estimation or Selection of Points of Departure (PODs) for RfD
Derivation 4-14
4.1.3 Pharmacokinetic Modeling Approaches to Convert Administered Dose to
Internal Dose in Animals and Humans 4-18
4.1.4 Application of Pharmacokinetic Modeling for Animal-Human
Extrapolation of PFOA Toxicological Endpoints and Dosimetric
Interpretation of Epidemiological Endpoints 4-28
4.1.5 Derivation of Candidate Chronic Oral Reference Doses (RfDs) 4-41
4.1.6 RfD Selection 4-49
4.2 Cancer 4-54
4.2.1 Animal Toxicological Studies 4-54
4.2.2 Epidemiological Studies 4-56
4.2.3 CSF Selection 4-59
4.2.4 Application of Age-Dependent Adjustment Factors 4-59
5 MCLG Derivation 5-1
6 Effects Characterization 6-1
6.1 Addressing Uncertainties in the Use of Epidemiological Studies for Quantitative
Dose-Response Analyses 6-1
6.2 Comparisons Between Toxicity Values Derived from Animal Toxicological
Studies and Epidemiological studies 6-4
6.3 Updated Approach to Animal Toxicological RfD Derivation Compared to the
2016 PFOA HESD 6-5
6.4 Consideration of Alternative Conclusions Regarding the Weight of Evidence of
PFOA Carcinogenicity 6-7
6.5 Health Outcomes with Evidence Integration Judgments of Evidence Suggests
Bordering on Evidence Indicates 6-10
6.6 Challenges and Uncertainty in Modeling 6-12
iv
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6.6.1 Modeling of Animal Internal Dosimetry 6-12
6.6.2 Modeling of Human Dosimetry 6-13
6.6.3 Approach of Estimating a Benchmark Dose from a Regression
Coefficient 6-15
6.7 Human Dosimetry Models: Consideration of Alternate Modeling Approaches 6-16
6.8 Sensitive Populations 6-19
6.8.1 Fetuses, Infants, Children 6-20
6.8.2 Sex Differences 6-20
6.8.3 Other Susceptible Populations 6-21
7 References 7-1
v
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figures
Figure 1-1. Distribution of PFOA Concentrations in Surface Waters by State/Waterbody
(excluding Great Lakes) in the U.S 1-8
Figure 3-1. Summary of Literature Search and Screening Process for PFOA 3-3
Figure 3-2. Summary of Epidemiology Studies of PFOA Exposure by Health System and
Study Designa 3-4
Figure 3-3. Summary of Animal Toxicological Studies of PFOA Exposure by Health
System, Study Design, and Speciesa'b 3-5
Figure 3-4. Schematic for a Physiologically Motivated Renal Resorption PK Model for
PFOA 3-22
Figure 3-5. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects Published Before 2016 (References in the
2016 HESD) 3-26
Figure 3-6. Overall ALT Levels from Pre-2016 HESD Epidemiology Studies Following
Exposure to PFOA 3-27
Figure 3-7. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effectsa 3-30
Figure 3-8. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects (Continued)51 3-31
Figure 3-9. Odds of Elevated ALT Levels from Epidemiology Studies Following Exposure
to PFOA 3-33
Figure 3-10. ALT Levels from Medium Confidence Epidemiology Studies Following
Exposure to PFOA 3-34
Figure 3-11. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects 3-36
Figure 3-12. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects (Continued) 3-37
Figure 3-13. Relative Liver Weight in Rodents Following Exposure to PFOA (logarithmic
scale) 3-40
Figure 3-14. Percent Change in Serum Enzyme Levels Relative to Controls in Male Mice
Following Exposure to PFOAa'b 3-42
Figure 3-15. Percent Change in Enzyme Levels Relative to Controls in Male Rats
Following Exposure to PFOAa 3-43
Figure 3-16. Percent Change in Enzyme Levels Relative to Controls in Female Rodents
Following Exposure to PFOAa 3-45
vi
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figure 3-17. Summary of Mechanistic Studies of PFOA and Hepatic Effects 3-51
Figure 3-18. Summary of Study Quality Evaluation Results Epidemiology Studies of PFOA
and Immune Effects Published Before 2016 (References in 2016 HESD) 3-94
Figure 3-19. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immunosuppression Effects 3-96
Figure 3-20. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immunosuppression Effects (Continued) 3-97
Figure 3-21. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA 3-99
Figure 3-22. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-100
Figure 3-23. Overall Diphtheria Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA 3-101
Figure 3-24. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immune Hypersensitivity Effects 3-109
Figure 3-25. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Autoimmune Effects 3-114
Figure 3-26. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Immune Effects 3-116
Figure 3-27. Globulin Levels in Rodents Following Exposure to PFOA (logarithmic scale). 3-122
Figure 3-28. Summary of Mechanistic Studies of PFOA and Immune Effects 3-122
Figure 3-29. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects Published Before 2016 (References from
2016 HESD) 3-144
Figure 3-30. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects 3-147
Figure 3-31. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued) 3-148
Figure 3-32. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued) 3-149
Figure 3-33. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids Published Before 2016 (References from 2016 PFOA
HESD) 3-156
Figure 3-34. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids 3-159
vii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figure 3-35. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued) 3-160
Figure 3-36. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued) 3-161
Figure 3-37. Odds of High Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA 3-167
Figure 3-38. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA 3-168
Figure 3-39. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-169
Figure 3-40. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-170
Figure 3-41. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-171
Figure 3-42. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Cardiovascular Effects 3-175
Figure 3-43. Serum Lipid Levels in Rodents Following Exposure to PFOA (logarithmic
scale) 3-178
Figure 3-44. Summary of Mechanistic Studies of PFOA and Cardiovascular Effects 3-179
Figure 3-45. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Developmental Effects Published before 2016 (References from
2016 PFOA HESD) 3-191
Figure 3-46. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects 3-198
Figure 3-47. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued) 3-199
Figure 3-48. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued) 3-200
Figure 3-49. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA 3-202
Figure 3-50. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-203
Figure 3-51. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-204
Figure 3-52. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-205
viii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figure 3-53. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Small for Gestational Age and Low Birth Weight Effectsa 3-208
Figure 3-54. Odds of Small-for-gestational-age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA 3-209
Figure 3-55. Odds of Small-for-gestational-age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA (Continued) 3-210
Figure 3-56. Odds of Small-for-gestational-age in Children from Medium Confidence
Epidemiology Studies Following Exposure to PFOA 3-211
Figure 3-57. Odds of Low Birthweight in Children from Epidemiology Studies Following
Exposure to PFOA 3-212
Figure 3-58. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Length Effects 3-214
Figure 3-59. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Length Effects (Continued) 3-215
Figure 3-60. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Head Circumference Effects 3-218
Figure 3-61. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Postnatal Growth 3-224
Figure 3-62. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Preterm Birth Effects 3-228
Figure 3-63. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Fetal Loss 3-231
Figure 3-64. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Defects 3-232
Figure 3-65. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Developmental Effects 3-233
Figure 3-66. Maternal Body Weight in Rodents Following Exposure to PFOA (logarithmic
scale) 3-236
Figure 3-67. Placental Weights in Mice Following Exposure to PFOA 3-237
Figure 3-68. Offspring Mortality in Rodents Following Exposure to PFOAa 3-240
Figure 3-69. Offspring Body Weight in Rodents Following Exposure to PFOA (logarithmic
scale)a 3-243
Figure 3-70. Summary of Mechanistic Studies of PFOA and Developmental Effects 3-247
Figure 3-71. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cancer Effects Published Before 2016 (References from 2016
PFOAHESD) 3-263
ix
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figure 3-72. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cancer Effects 3-265
Figure 3-73. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Cancer Effects 3-269
Figure 3-74. Summary of Mechanistic Studies of PFOA and Cancer Effects 3-272
Figure 4-1. Model Structure for Life Stage Modeling 4-21
Figure 4-2. Gestation/Lactation Predictions of PFOA in the Rat 4-23
Figure 4-3. Gestation/Lactation Predictions of PFOA in the Mouse in a Cross-Fostering
Study 4-24
Figure 4-4. Comparison of Candidate RfDs Resulting from the Application of Uncertainty
Factors to PODheds Derived from Epidemiological and Animal Toxicological
Studies 4-50
Figure 4-5. Schematic depicting selection of the overall RfD for PFOA 4-53
x
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Tables
Table 1-1. Chemical and Physical Properties of PFOA 1-6
Table 3-1. Database Literature Search Results 3-1
Table 3-2. Associations Between PFOA Exposure and Cell Death or Necrosis in Rodents 3-48
Table 3-3. Evidence Profile Table for PFOA Hepatic Effects 3-87
Table 3-4. Associations between PFOA Exposure and Vaccine Response in Faroe Islands
Studies 3-102
Table 3-5. Effects of PFOA Exposure on Cytokines Impacting Adaptive Immune
Responses 3-128
Table 3-6. Effects of PFOA Exposure on Pro-Inflammatory Cytokines and Markers of
Inflammation 3-132
Table 3-7. Evidence Profile Table for PFOA Immune Effects 3-136
Table 3-8. Evidence Profile Table for PFOA Cardiovascular Effects 3-184
Table 3-9. Study Design for Perinatal and Postweaning Exposure Levels for Fi Male and
Female Rats for the NTP (2020, 7330145) Study 3-235
Table 3-10. Evidence Profile Table for PFOA Developmental Effects 3-255
Table 3-11. Incidences of Liver Adenomas in Male Sprague-Dawley Rats as Reported by
NTP (2020, 7330145) 3-270
Table 3-12. Incidences of Pancreatic Acinar Cell Adenomas in Male Sprague-Dawley Rats
as Reported by NTP (2020, 7330145) 3-271
Table 3-13. Incidences of Uterine Adenocarcinomas in Female Sprague-Dawley Rats from
the Standard and Extended Evaluations (Combined) as Reported by NTP (2020,
7330145) 3-272
Table 3-14. Mutagenicity Data from In Vitro Studies 3-276
Table 3-15. DNA Damage Data from In Vivo Studies 3-276
Table 3-16. DNA Damage Data from In Vitro Studies 3-277
Table 3-17. Comparison of the PFOA Carcinogenicity Database with the Likely Cancer
Descriptor as Described in the Guidelines for Carcinogen Risk Assessment
{U.S. EPA, 2005, 6324329} 3-307
Table 4-1. Summary of Endpoints and Studies Considered for Dose-Response Modeling
and Derivation of Points of Departure for All Effects in Humans and Rodents 4-10
Table 4-2. Benchmark Response Levels Selected for BMD Modeling of Health Outcomes.... 4-16
xi
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 4-3. PK Parameters from Wambaugh et al. (2013, 2850932) Meta-Analysis of
Literature Data for PFOA 4-19
Table 4-4. Model Predicted and Literature PK Parameter Comparisons for PFOA 4-20
Table 4-5. Additional PK Parameters for Gestation/Lactation for PFOA 4-22
Table 4-6. Updated and Original Chemical-Specific Parameters for PFOA in Humans 4-26
Table 4-7. Summary of Studies Reporting the Ratio of PFOA Levels in Breastmilk and
Maternal Serum or Plasma 4-28
Table 4-8. PODheds Considered for the Derivation of Candidate RfD Values 4-31
Table 4-9. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
from Epidemiological Studies {U.S. EPA, 2002, 88824} 4-43
Table 4-10. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
from Animal Toxicological Studies {U.S. EPA, 2002, 88824} 4-45
Table 4-11. Candidate Reference Doses (RfDs) 4-47
Table 4-12. Cancer Slope Factors based on Animal Toxicological Data 4-56
Table 4-13. Cancer Slope Factors based on Epidemiological Data 4-58
Table 6-1. Comparison of Candidate RfDs Derived from Animal Toxicological Studies for
Priority Health Outcomes51 6-7
Table 6-2. Comparison of the PFOA Carcinogenicity Database with Cancer Descriptors as
Described in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329} 6-9
xii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Acronyms and Abbreviations
3D
Three-dimensional
BBB
Blood brain barrier
8-N02Gua
8-nitroguanine
Bel-2
B-cell lymphoma 2
8-OHdG
8-hydroxydeoxy-
BCRP
Breast cancer resistance
guanosine
protein
AASLD
American Association
BK
Bradykinin
for the Study of Liver
BMD
Benchmark dose
Diseases
BMDio
Dose corresponding to a
ABC
ATP Binding Casette
10% change in response
ACG
American College of
BMDL
Benchmark dose lower
Gastroenterology
limit
AChE
Acetylcholinesterase
BMDLio
Dose level
Acot
Acyl-CoA thioesterase
corresponding to the
ACOX
Acyl-CoA oxidase
95% lower confidence
Acsll
Acyl-CoA synthetase
limit of a 10% change
ADME
Absorption, distribution,
BMDS
Benchmark Dose
metabolism, excretion
Software
AFFF
Aqueous film forming
BMI
Body mass index
foam
BMR
Benchmark response
AL
Human-hamster hybrid
BTB
Blood testes barrier
cells
BWT
Birth weight
ALP
Alkaline phosphatase
C3a
Complement 3
ALSPAC
Avon Longitudinal
Clast7
Average concentration
Study of Parents and
over final week of study
Children
CAD
Coronary artery disease
ALT
Alanine
CalEPA
California
aminotransferase
Environmental
Aplsl
Adaptor related protein
Protection Agency
complex 1 subunit
CAR
Constitutive androstane
sigma 1
receptor
APC
Antigen presenting cell
CASRN
Chemical Abstracts
APFO
Ammonium
Service Registry Number
perfluorooctanoate
CAT
Catalase
APOA4
Apolipoprotein A4
Cavg
Average blood
apoB
Apolipoprotein B
concentration
ApoC-III
Apolipoprotein C-III
Cavg,pup,gest
area under the curve
AST
Aspartate
normalized per day
aminotransferase
during gestation
AT SDR
Agency for Toxic
Cavg, pup, gest,lact
area under the curve
Substances and Disease
normalized dose per day
Registry
during gestation/lactation
AUC
Area under the curve
BAFF
B cell activating factor
xiii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Cavg,pup,lact
area under the curve
normalized per day
during lactation
Cavg,pup,total
area under the curve in
gestation/lactation added
to the area under the
curve from diet (post-
weaning) divided by two
years
CCL
Contaminant Candidate
List
CCK
Cholecystokinin
CCK-8
Cell Counting Kit-8
CD
Circular dichronism
CDC
Centers for Disease
Control and Prevention
cDNA
complementary DNA
Ces
Carboxylesterases
CETP
Cholesteryl ester transfer
protein
C-F
Carbon-fluorine
c-fos
Transcription factor
complex
CHD
Coronary heart disease
CHF
Congestive heart failure
CHO
Chinese hamster ovary
CHOP
C/EBP homologous
protein
CI
Confidence interval
CIMT
Carotid intima-media
thickness test
CLr
Renal clearance
Cmax
Maximum blood
concentration
Cmax pup gest
Maximum fetal
concentration during
gestation
Cmax pup lact
Maximum fetal
concentration during
lactation
CNS
Central nervous system
Cptla
Carnitine
palmitoyltransferase la
cs
collagen sandwich
CSF
cancer slope factor
CSM
Cholestyramine
CVD
Cardiovascular disease
DBP
Diastolic blood pressure
DCF
27 '-dichlorofluorescein
DCF-DA
Dichlorodihydro-
fluorescein diacetate
DDE
Dichlorodiphenyl
dichloroethane
DMP
3,5-dimethyl pyrazole
DMSO
Dimethyl sulfoxide
DNA
Deoxyribonucleic acid
DNBC
Danish National Birth
Cohort
DNMT
Deoxyribonucleic acid
methyltransferases
DNP
Dinitrophenyl
dpf
Days post fertilization
DPP
Diabetes Prevention
Program
DPPOS
Diabetes Prevention
Program and Outcomes
Study
DWI-BW
Body weight-based
drinking water intake
E2
Estradiol
eGFR
Estimated glomerular
filtration rate
EPA
Environmental
Protection Agency
ER
Endoplasmic reticulum
ER-
Estrogen receptor
negative
ETC
Electron transport chain
Fi
First generation
f2
Second generation
Fabp
Fatty acid binding
protein
FACS
Fluorescence activated
cell sorting
FeNO
Fractional exhaled nitric
oxide
FFA
Free fatty acids
FT4
Free thyroxine
FXR
Farnesoid X receptor
xiv
-------
DRAFT FOR PUBLIC COMMENT
GBCA
Genetic and Biomarker
Study for Childhood
Asthma
GCL
Glutamate-cysteine
ligase
GD
Gestation day
GFR
Glomerular filtration rate
GGT
y-glutamyltransferase
GM
Geometric mean
GO
Gene Ontology
GSH
Glutathione
GSPE
Grape seed
proanthocyandidn extract
GSSG
Glutathione disulfide
GST
Glutathione S-
transferases
HAT
Histone acetylase
HAWC
Health Assessment
Workplace Collaborative
HDAC
Histone deacetylase
HDL
High-density-lipoprotein
HED
Human equivalent dose
HEK-293
Human embryonic
kidney
HERO
Health and
Environmental Research
Online
HESD
Health Effects Support
Document
HFC
7-hy droxytrifluoro-
methylcoumarin
HFD
High-fat diet
HFMD
Hand, foot, and mouth
disease
HFPO
Hexafluoropropylene
oxide
Hib
Haemophilus influenza
type b
HK
High-molecular-weight
kininogen
hL-FABP
Human liver fatty acid
binding protein
HMOX
Heme oxygenase
HNF
Hepatocyte nuclear
factor
March 2023
HOME
Health Outcomes and
Measures of the
Environment
HPA
Hypothalamic-pituitary-
adrenal
HR
Hazard ratio
HRL
Health reference level
HSA
Human serum albumin
IARC
International Agency for
Research on Cancer
IDL
Intermediate-density
lipoprotein
IFN
Interferon
Ig
Immunoglobulin
IGF-1
Insulin-like growth factor 1
IHD
Ischemic heart diseases
IHIC
Hepatic immune cell
IL
Inflammatory cytokine
INMA
Spanish Environment
and Childhood (Infancia
y Medio Ambiente)
IP
Intraperitoneal
IPCS
International Programme
on Chemical Safety
IQR
Interquartile range
IRIS
Integrated Risk
Information System
IV
Intravenous
ki2
Intercompartment
transfer rate
ka
Absorption rate
Kd
Disassociation constant
Kh
Henry's Law Constant
KK
Kallikrein-kinn system
KLH
Keyhole limpet
hemocyanin
Kmem/w
Membrane/water
partition coefficients
Koc
Organic carbon-water
partitioning coefficient
Kow
Octanol-water partition
coefficient
LBW
Low birth weight
xv
-------
DRAFT FOR PUBLIC COMMENT
March 2023
LCM
Liver capsular
NAFLD
Non-alcoholic fatty liver
macrophage
disease
LCT
Leydig cell tumors
NCI
National Cancer Institute
LD
Lactation day
NF-kB
Nuclear factor kappa B
LDL
Low-density lipoprotein
NHANES
National Health and
L-FABP
Liver fatty acid binding
Nutrition Examination
protein
Survey
LH
Luteinizing hormone
NK
Natural killer
LOAEL
Lowest-observed-
NO
Nitric oxide
adverse-effect level
NOAEL
No-observed-adverse-
LOD
Limit of detection
effect level
Lpl
Lipoprotein lipase
NOD
Nucleotide-binding and
LTRI
Lower respiratory tract
oligomerization domain
infection
NOS
Nitric oxide synthase
LXR
Liver X receptor
NP
Niemann-Pick disease
LYZ
Lysozyme
NPDWR
National Primary
M/P
Milk/plasma
Drinking Water
MAIT
Mucosal associated
Regulation
invariant T
Nrf2
Nuclear factor erythroid
MCLG
Maximum Contaminant
2-related factor 2
Level Goal
NTCP
S odium-taurochol ate
Me-PFOSA-AcOH
cotransporting
polypeptide
or MeFOSAA
2-(N-Methyl-
NTP
National Toxicology
perfluorooctane
Program
sulfonamido) acetic acid
OATPs
Organic anion
MDA
Malondialdehyde
transporting polypeptides
MFC
7-methoxy-4-
OATs
Organic anion
trifluoromethylcoumarin
transporters
miRNA or miRs
Microribonucleic acids
OCM
Organotypic culture
MMP
Mitochondrial membrane
models
potential
OECD
Organisation for
MMR
Measles, mumps, and
Economic Co-operation
rubella
and Development
MOA
Mode of action
OPR
Opioid Receptor
MOBA
Norwegian Mother,
OR
Odds Ratio
Father, and Child Cohort
ORD
Office of Research and
Study
Development
MRL
Minimum reporting level
OST
Office of Science and
mRNA
Messenger ribonucleic
Technology
acid
Po
Parental generation
MRPs
Multidrug resistance-
pOAL
Mitochondrial deficient
associated proteins
cell line
MS
Multiple sclerosis
PACT
Pancreatic acinar cell
MyD
Myeloid differentiation
tumors
xvi
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PAD
PanIN
PBMC
PBPK
PC
PDCD
PECO
PERK
PFAA
PFAS
PFBA
PFCAs
PFDA
PFDoDA
PFHpA
PFHxA
PFHxS
PFNA
PFOA
PFOS
PG
Pion
PK
pKa
PLCO
Pmilk
Peripheral artery disease
Pancreatic intraepithelial
neoplasia
Peripheral blood
mononuclear cells
Physiologically-based
pharmacokinetic
Partition coefficient
Programmed cell death
protein
Populations, Exposures,
Comparator, and
Outcome
Protein kinase-like
endoplasmic reticulum
kinase
Perfluoroalkyl acids
Per- and polyfluoroalkyl
Substances
Perfluorobutanoic acid
Perfluoroalkyl
carboxylic acids
Perfluorodecanoic acid
Perfluorododecanoic
acid
Perfluoroheptanoic acid
Perfluorohexanoic acid
Perfluorohexane-
sulfonate
Perfluorononanoic acid
Perfluorooctanoic acid
Perfluorooctane sulfonic
acid
Prostaglandin
Passive anionic
permeability
Pharmacokinetic
Negative base-10
logarithm of acid
dissociation constant
Prostate, Lung,
Colorectal, and Ovarian
Screening Trial
Maternal milk: blood
partition coefficient
PND
PNW
POD
PODhed
POUNDS-Lost
PP2A
PPAR
PPK
ppm
PR-
PSA
PTB
PWS
PXR
Qi
Q2
Q3
Q4
QA
Ro
r°milk
r^iik
r2miik
r3milk
RARa
RASA3
RCC
RD
RfD
Rfm
r'milk
Postnatal day
Postnatal week
Point of departure
Point of departure human
equivalent dose
Prevention of Obesity
Using Novel Dietary
Strategies-Lost
Protein phosphatase 2A
Peroxisome proliferator
activated receptor
Plasma prekallikrein
Parts per million
Progesterone receptor
negative
Prostate-specific antigen
Preterm birth
Public water system
Pregnane X receptor
Quartile one
Quartile two
Quartile three
Quartile four
Quality assurance
Baseline risk
Starting milk
consumption rate
Week 1 milk
consumption rate
Week 2 milk
consumption rate
Week 3 milk
consumption rate
Retinoic acid receptor a
RAS P21 protein
Activator 3
Renal cell carcinoma
Regular diet
Reference dose
Fetus: mother
concentration ratio
Milk consumption rate
for the ith week of
lactation
XVII
-------
DRAFT FOR PUBLIC COMMENT
RNA
Ribonucleic acid
RNS
Reaction nitrogen
species
ROS
Reactive oxygen species
RR
Rate ratio
RRBS
Reduced representation
bisulfite sequencing
RSC
Relative source
contribution
SAB
Science Advisory Board
SBP
Systolic blood pressure
SDWA
Safe Drinking Water Act
SES
Socioeconomic status
SGA
Small for gestational age
SIRT
Sirtuin
slcold
Solute carrier organic
anion transporter
SMR
Standardized mortality
ratios
SOD
Superoxide dismutase
SRBC
Sheep red blood cells
SREBP
Sterol regulatory
element-binding protein
T1D
Type 1 diabetes
T4
Thyroxine
TC
Total cholesterol
TET
Methylcytosine
dioxygenases
tfc
Transcription factor
tgf
Transforming growth
factor
TLDA
Taqman low density
arrays
TLR
Toll-like receptor
Tmax
Time to Cmax
TNF
Tumor necrosis factor
TNP
Trinitrophenyl
TReg
Regulatory T cell
March 2023
TSCATS
Toxic Substance Control
Act Test Submissions
TTEs
Transplacental
efficiencies
TTR
Transthyretin
TXB
Thromboxane
UCMR3
Third Unregulated
Contaminant Monitoring
Rule
UF
Uncertainty factors
UFa
Interspecies UF
UFd
Database UF
UFh
Intraspecies UF
UFl
LOAEL-to-NOAEL
extrapolation UF
UFs
UF for extrapolation
from a sub chronic to a
chronic exposure
duration
UFc
Composite uncertainty
factor
|iM
Micromolar
UPR
Unfolded protein
response
UV-vis
Ultravi ol et-vi sibl e
Vd
Volume of distribution
vtgl
Vitellogenin 1
VLDL
Very low-density
lipoproteins
Vldlr
Very low-density
lipoproteins receptor
WHO
World Health
Organization
WoS
Web of Science
WTC
World Trade Center
XBP1
Spliced X box-binding
protein 1
ZFL
Zebrafish liver cell line
xviii
-------
DRAFT FOR PUBLIC COMMENT
March 2023
1 Background
1.1 National Primary Drinking Water Regulation for Per- and
Polyfluoroalkyl Substances under the Safe Drinking Water Act
The U.S. Environmental Protection Agency (EPA) has initiated the process to develop a
Maximum Contaminant Level Goal (MCLG) and National Primary Drinking Water Regulation
(NPDWR) for per- and polyfluoroalkyl substances (PFAS), including perfluorooctanoic acid
(PFOA), under the Safe Drinking Water Act (SDWA). As part of the proposed rulemaking, EPA
prepared Proposed Approaches to the Derivation of a Draft Maximum Contaminant Level Goal
for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking Water that described the
derivation of candidate oral cancer toxicity values and noncancer toxicity values, a relative
source contribution (RSC), and cancer classification, which could be subsequently used to derive
an MCLG for PFOA. The agency sought peer review from the EPA Science Advisory Board
(SAB) on key scientific issues related to the development of the MCLG, including the systematic
review approach, oral toxicity values, RSC, and cancer classification.
The SAB provided draft recommendations on June 3, 2022 and final recommendations on
August 23, 2022 {U.S. EPA, 2022, 10476098}, and EPA addressed those recommendations in
the development of this updated assessment, Toxicity Assessment and Proposed Maximum
Contaminant Level Goal (MCLG) for Perfluorooctanoic Acid (PFOA) in Drinking Water, which
derives toxicity values and an MCLG for PFOA To be responsive to the SAB recommendations,
EPA has, for example:
• updated and expanded the scope of the studies included in the assessment;
• expanded the systematic review steps beyond study quality evaluation to include evidence
integration to ensure consistent hazard decisions;
• separated hazard identification and dose-response assessment;
• added protocols for all steps of the systematic review and more transparently described the
protocols;
• evaluated alternative pharmacokinetic models and further validated the selected model;
• conducted additional dose-response analyses using additional studies and endpoints;
• evaluated and integrated mechanistic information;
• strengthened the weight of evidence for cancer and rationale for the cancer classification;
• strengthened the rationales for selection of points of departure for the noncancer health
outcomes; and
• clarified language related to the relative source contribution determination including the
relevance of drinking water exposures and the relationship between the reference dose
(RfD) and the relative source contribution.
1.2 Background on PFAS
PFAS are a large group of anthropogenic chemicals that share a common structure of a chain of
linked carbon and fluorine atoms. The PFAS group includes PFOA, perfluorooctane sulfonic
acid (PFOS), and thousands of other chemicals. While the number of PFAS used globally in
commercial products in 2021 was approximately 250 substances {Buck, 2021, 9640864}, the
1-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
universe of PFAS, including parent chemicals, metabolites, and degradants, is greater than
12,000 compounds (https://comptox.epa.eov/dashboard/chemical4ists/PFA.SMA.STER). The
2018 Organisation for Economic Co-operation and Development (OECD) New Comprehensive
Global Database of Per- and Polyfluoroalkyl Substances (PFASs) includes over 4,700 PFAS
{OECD, 2018, 5099062}.
PFAS have been manufactured and used in a wide variety of industries around the world,
including in the United States, since the 1950s. PFAS have strong, stable carbon-fluorine (C-F)
bonds, making them resistant to hydrolysis, photolysis, microbial degradation, and metabolism
{Ahrens, 2011, 2657780; Beach, 2006, 1290843; Buck, 2011, 4771046}. The chemical
structures of PFAS enable 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 make many PFAS extremely persistent in the human body and the environment
{Calafat, 2007, 1290899; Calafat, 2019, 5381304; Kwiatkowski, 2020, 7404231}. Due to their
widespread use, physicochemical properties, persistence, and bioaccumulation potential, many
different PFAS co-occur in environmental media (e.g., air, water, ice, sediment) and in tissues
and blood of aquatic and terrestrial organisms, including humans.
Based on structure, there are many families or classes of PFAS, each containing many individual
structural homologues that can exist as either branched-chain or straight-chain isomers {Buck,
2011, 4771046}. These PFAS families can be divided into two primary categories: non-polymers
and polymers. The non-polymer PFAS include perfluoroalkyl acids (PFAAs), fluorotelomer-
based substances, and per- and polyfluoroalkyl ethers. PFOA and PFOS belong to the PFAA
family of the non-polymer PFAS category and are among the most researched PFAS in terms of
human health toxicity and biomonitoring studies (for review, see Podder et al. (2021, 9640865)).
1.3 Evaluation of PFOA Under SDWA
SDWA, as amended in 1996, requires EPA to publish a list every 5 years of unregulated
contaminants that are not subject to any current proposed or promulgated NPDWRs, are known
or anticipated to occur in public water systems (PWSs), and might require regulation under
SDWA. This list is known as the Contaminant Candidate List (CCL). PFOA is included on the
third CCL (CCL 3) {U.S. EPA, 2009, 1508321} and on the fourth CCL (CCL 4) {U.S. EPA,
2016, 6115068}.
After PFOA and PFOS were listed on the CCL 3 in 2009, EPA initiated development of health
effects support documents (HESDs) for PFOA and PFOS that provided information to federal,
state, tribal, and local officials and managers of drinking water systems charged with protecting
public health when these chemicals are present in drinking water {U.S. EPA, 2016, 3603365;
U.S. EPA, 2016, 3603279}. The two HESDs were peer-reviewed in 2014 and revised based on
consideration of peer reviewers' comments, public comments, and additional studies published
through December 2015. The resulting 2016 Health Effects Support Document for
Perfluorooctanoic Acid (PFOA) {U.S. EPA, 2016, 3603279} described the assessment of cancer
and noncancer health effects and the derivation of a noncancer RfD that served as the basis for
the non-regulatory 2016 Drinking Water Health Advisory for Perfluorooctanoic Acid (PFOA)
{U.S. EPA, 2016,3982042}.
1-2
-------
DRAFT FOR PUBLIC COMMENT
March 2023
SDWA requires EPA to make regulatory determinations for at least five CCL contaminants
every 5 years. EPA must begin developing an NPDWR when the agency makes a determination
to regulate based on a finding that a contaminant meets all three of the following criteria:
• The contaminant may have an adverse effect on the health of persons.
• The contaminant is known to occur or there is substantial likelihood the contaminant will
occur in PWSs with a frequency and at levels of public health concern.
• In the sole judgment of the Administrator, regulating the contaminant presents a
meaningful opportunity for health risk reductions.
To make these determinations, the agency considers a range of information, including data to
analyze occurrence of these compounds in finished drinking water and data on health effects that
represent the latest science.
In the Final Regulatory Determinations for Contaminants on the Fourth Drinking Water
Contaminant Candidate List {U.S. EPA, 2021, 9640861}, the agency made a determination to
regulate PFOA and PFOS with an NPDWR. The agency concluded that all three criteria were
met—PFOA and PFOS may have adverse health effects; they occur in PWSs with a frequency
and at levels of public health concern; and, in the sole judgment of the Administrator, regulation
of PFOA and PFOS presents a meaningful opportunity for health risk reduction for persons
served by PWSs {U.S. EPA, 2021, 7487276}. As noted above in Section 1.1, EPA prepared
Proposed Approaches to the Derivation of a Draft Maximum Contaminant Level Goal for
Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking Water as part of this
rulemaking.
In June 2022, EPA published an interim Drinking Water Health Advisory for PFOA {U.S. EPA,
2022, 10671184} to supersede the 2016 Drinking Water Health Advisory based on analyses of
more recent data described in the Proposed Approaches to the Derivation of a Draft Maximum
Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking
Water, which showed that PFOA can impact human health at exposure levels much lower than
reflected by the 2016 Drinking Water Health Advisory {U.S. EPA, 2016, 3982042; U.S. EPA,
2022, 10671184}.
1.4 Purpose of this Document
Consistent with SDWA Section 1412(b)(3)(A) and (B), the primary purpose of this draft
document is to obtain public comment on EPA's toxicity assessment and proposed MCLG for
PFOA by describing the best available science on health effects in order to derive an MCLG. To
derive an MCLG, the latest science is identified, described, and evaluated, and then a cancer
classification, toxicity values (i.e., a noncancer RfD and cancer slope factor (CSF)), and RSC are
developed (Section 2.3). The draft cancer and noncancer toxicity values, cancer classification,
and RSC derived in this assessment build upon the work described in the Proposed Approaches
to the Derivation of a Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid
(PFOA) (CASRN 335-67-1) in Drinking Water, the 2016 PFOA HESD {U.S. EPA, 2016,
3603279}, and the previous 2016 PFOA Drinking Water Health Advisory {U.S. EPA, 2016,
3982042}.
1-3
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In addition to documenting EPA's basis for the proposed MCLG, this document serves the
following purposes:
• Transparently describe and document the literature searches conducted and systematic
review methods used to identify health effects information (epidemiological and animal
toxicological studies and physiologically-based pharmacokinetic (PBPK) models) in the
literature.
• Describe and document literature screening methods, including use of the Populations,
Exposures, Comparators, and Outcomes (PECO) criteria and the process for tracking
studies throughout the literature screening.
• Identify epidemiological and animal toxicological literature that report health effects after
exposure to PFOA (and its associated salts) as outlined in the PECO criteria.
• Evaluate and document the available mechanistic information (including toxicokinetic
understanding) associated with PFOA exposure to inform interpretation of findings related
to potential health effects in studies of humans and animals, with focus on five main
health outcomes (developmental, hepatic, immune, and cardiovascular effects, and
cancer).
• Describe and document the study quality evaluations conducted on epidemiological and
animal toxicological studies considered potentially useful for point-of-departure (POD)
derivation.
• Describe and document the data from high and medium confidence epidemiological and
animal toxicological studies (as determined by study quality evaluations) that were
considered for POD derivation; in cases of health effects with few available studies, data
may be extracted from low confidence studies and used in the evidence syntheses. For
dose-response assessment, only high and medium confidence studies were used to
quantify health effects.
• Synthesize and document the adverse health effects evidence across studies, assessing
health outcomes using a narrative approach. The assessment focuses on synthesizing the
available evidence for five main health outcomes—developmental, hepatic, immune, and
cardiovascular effects, and cancer—but also provides secondary syntheses of evidence for
dermal, endocrine, gastrointestinal, hematologic, metabolic, musculoskeletal, nervous,
ocular, renal, and respiratory effects; reproductive effects in males or females; and general
toxicity.
• Develop and document strength of evidence judgments across studies (or subsets of
studies) separately for epidemiological and for animal toxicological lines of evidence and
integrate mechanistic analyses into judgments for the five main health outcomes.
• Develop and document integrated expert judgments across lines of evidence (i.e.,
epidemiological and animal toxicological lines of evidence) as to whether and to what
extent the evidence supports that exposure to PFOA has the potential to be hazardous to
humans. The judgments will be directly informed by the evidence syntheses and based on
structured review of an adapted set of considerations for causality first introduced by
Austin Bradford Hill {Hill, 1965, 71664}.
1-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
• Describe and document the dose-response analyses conducted on the studies identified for
POD derivation.
• Derive candidate RfDs and/or CSFs and select the RfD and/or CSF for PFOA and
describe the rationale.
• Determine PFOA's cancer classification using a weight of evidence approach.
• Characterize hazards (e.g., uncertainties, data gaps).
1.5 Chemical Identity
PFOA is a perfluorinated aliphatic carboxylic acid. It is a fully fluorinated organic synthetic acid
that was used in the United States primarily as an aqueous dispersion agent and emulsifier in the
manufacture of fluoropolymers and in a variety of water-, oil-, and stain-repellent products (e.g.,
adhesives, cosmetics, fire-fighting foams, greases and lubricants, paints, polishes) {NLM, 2022,
10369700}. It can exist in linear- or branched-chain isomeric form. PFOA is a strong acid that is
generally present in solution as the perfluorooctanoate anion. Therefore, this assessment applies
to all isomers of PFOA, as well as nonmetal salts of PFOA that would be expected to dissociate
in aqueous solutions of pH ranging from 4 to 9 (e.g., in the human body).
PFOA is water soluble and mobile in water, with an estimated log organic carbon-water partition
coefficient (log Koc) of 2.06 {Zareitalabad, 2013, 5080561}. PFOA is stable in environmental
media because it is resistant to environmental degradation processes, such as biodegradation,
photolysis, and hydrolysis. In water, no natural degradation has been demonstrated, and it
dissipates by advection, dispersion, and sorption to particulate matter. PFOA has low volatility in
its ionized form but can adsorb to particles and be deposited on the ground and into water bodies.
Because of its persistence, it can be transported long distances in air or water, as evidenced by
detections of PFOA in arctic media and biota, including polar bears, ocean-going birds, and fish
found in remote areas {Lindstrom, 2011, 1290802; Smithwick, 2006, 1424802}.
Physical and chemical properties and other reference information for PFOA are provided in
Table 1-1. There is uncertainty in the estimation, measurement, and/or applicability of certain
physical/chemical properties of PFOA in drinking water, including the Koc {Li, 2018, 4238331;
Nguyen, 2020, 7014622}, octanol-water partition coefficient (Kow), and Henry's Law Constant
(Kh) {ATSDR, 2021, 9642134; NCBI, 2022, 10411459}. For example, for Kow, the Agency for
Toxic Substances and Disease Registry (ATSDR) (2021, 9642134) and Lange et al. (2006,
10411376) reported that a value could not be measured because PFOA is expected to form
multiple layers in octanol-water mixtures.
For a more detailed discussion of the chemical and physical properties and environmental fate of
PFOA, please see the PFAS Occurrence & Contaminant Background Technical Support
Document {U.S. EPA, 2023, 10692764}, the 2016 PFOA Drinking Water Health Advisory
{U.S. EPA, 2016, 3982042}, and th q Draft Aquatic Life Ambient Water Quality Criteria for
Perfluorooctanoic Acid (PFOA) {U.S. EPA, 2022, 10671186}.
1-5
-------
DRAFT FOR PUBLIC COMMENT March 2023
Table 1-1. Chemical and Physical Properties of PFOA
Property
Perfluorooctanoic Acid;
Experimental Average
Source
Chemical Abstracts Service Registry
335-67-1
NLM, 2022, 10369702
Number (CASRN)3
Chemical Abstracts Index Name
2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-
pentadecafluorooctanoic acid
Synonyms
PFOA; pentadecafluoro-l-octanoic
EPA CompTox Chemicals
acid; pentadecafluoro-n-octanoic acid;
Dashboard
octanoic acid, pentadecafluoro-;
perfluorocaprylic acid;
pentadecafluorooctanoic acid;
perfluoroheptanecarboxylic acid
Chemical Formula
C8HFi502
NLM, 2022, 10369702
Molecular Weight
414.069 g/mol
NLM, 2022, 10369700
Color/Physical State
White to off-white powder (ammonium
NLM, 2022, 10369700
salt)
Boiling Point
192°C
NLM, 2022, 10369700
Melting Point
54.3°C
NLM, 2022, 10369700
Vapor Pressure
0.0316 mm Hg at 19°C
NLM, 2022, 10369700
0.017 mm Hg at 20°C
ATSDR, 2021, 9642134
(extrapolated)
Henry's Law Constant (KH)
0.362 Pa-m3/mol (converts to
ATSDR, 2021, 9642134
3.57E-06 atm-m3/mol)
pKa
1.30,2.80, -0.5-4.2,0.5,0.5
NLM, 2022, 10369700
ATSDR, 2021, 9642134
Koc
631 ± 7.9 L/kg (mean ± 1 standard
Zareitalabad et al., 2013, 5080561
deviation of selected values)
(converted from log Koc to Koc)
Solubility in Water
2,290 mg/L at 24°C (estimated);
NLM, 2022, 10369700
3,300 mg/L at 25°C; 4,340 mg/L at
24.1°C
ATSDR, 2021, 9642134
9,500 mg/L at 25°C; 3,300 mg/L at
25°C
Notes'. Koc = organic carbon-water partitioning coefficient; K0w = octanol-water partition coefficient; pKa: negative base-10
logarithm of acid dissociation constant.
a The CASRN given is for linear PFOA, but the toxicity studies are based on both linear and branched, thus, this assessment
applies to all isomers of PFOA.
1.6 Occurrence Summary
1.6.1 Biomonitoring
The U.S. Centers for Disease Control and Prevention (CDC) National Health and Nutrition
Examination Survey (NHANES) has measured blood serum concentrations of several PFAS in
the general U.S. population since 1999. PFOA and PFOS have been detected in up to 98% of
serum samples taken in biomonitoring studies that are representative of the U.S. general
population. Blood levels of PFOA and PFOS dropped 60% to 80% between 1999 and 2014,
presumably due to restrictions on their commercial usage in the United States {CDC, 2017,
4296146}. In 2006, EPA secured a commitment from the eight major companies in the PFAS
industry to reduce PFOA from facility emissions and product content by 95% no later than 2010,
1-6
-------
DRAFT FOR PUBLIC COMMENT
March 2023
and to work toward eliminating PFOA from emissions and product content by 2015
(https://www.epa.gov/assessing~and~managing~chemicals~under~tsca/fact~sheet~2QlQ2Q15~pfoa~
stewardship-program) {U.S. EPA 2006, 3005012}. Manufacturers have since shifted to
alternative short-chain PFAS, such as hexafluoropropylene oxide (HFPO) dimer acid and its
ammonium salt (two "GenX chemicals"). Additionally, other PFAS were found in human blood
samples from recent (2011-2016) NHANES surveys (e.g., perfluorodecanoic acid (PFDA),
perfluorododecanoic acid (PFDoDA), perfluoroheptanoic acid (PFHpA),
perfluorohexanesulfonate (PFHxS), perfluorononanoic acid (PFNA), 2-(N-methyl-
perfluorooctane sulfonamido) acetic acid (Me-PFOSA-AcOH or MeFOSAA)). There is less
publicly available information on the occurrence and health effects of these replacement PFAS
than for PFOA, PFOS, and other members of the carboxylic acid and sulfonate PFAS categories.
1.6.2 Ambient Water
Among the PFAS with established analytical methods for detection, PFOA (along with PFOS) is
one of the dominant PFAS compounds detected in ambient water both in the U.S. and worldwide
{Ahrens, 2011, 2657780; Benskin, 2012, 1274133; Dinglasan-Panlilio, 2014, 2545254;
Nakayama, 2007, 2901973; Remucal, 2019, 5413103; Zareitalabad, 2013, 5080561}. Most of the
current, published PFOA occurrence studies have focused on a handful of broad geographic
regions in the U.S., often targeting sites with known manufacturing or industrial uses of PFAS
such as the Great Lakes, the Cape Fear River, and waterbodies near Decatur, Alabama
{Boulanger, 2004, 1289983; Cochran, 2015, 9416545; Hansen, 2002, 1424808; Konwick, 2008,
1291088; Nakayama, 2007, 2901973; 3M Company, 2000, 9419083}. PFOA concentrations in
global surface waters range over seven orders of magnitude, generally in pg/L to ng/L
concentrations, but sometimes reaching |ig/L levels {Jarvis, 2021, 9416544; Zareitalabad, 2013,
5080561}. Figure 1-1 (adapted from {Jarvis, 2021, 9416544}) shows the distribution of PFOA
concentrations (ng/L) measured in surface waters for each U.S. state or waterbody (excluding the
Great Lakes) with reported data in the publicly available literature.
1-7
-------
DRAFT FOR PUBLIC COMMENT
March 2023
10000
<
o
Uh
Ph
CO
-------
DRAFT FOR PUBLIC COMMENT
March 2023
from 117 PWSs had detections of PFOA (i.e., greater than or equal to the MRL). PFOA
concentrations for these detections ranged from 0.02 |ig/L (the MRL) to 0.349 |ig/L (median
concentration of 0.03 |ig/L; 90th percentile concentration of 0.07 |ig/L).
Because PFOS and PFOA cause similar types of adverse health effects and their 2016 lifetime
Health Advisory values were the same, EPA recommended an additive approach when PFOA
and PFOS co-occur at the same time and location in drinking water sources {U.S. EPA, 2016,
3603365; U.S. EPA, 2016, 3603279}. This approach was used in the analysis for Regulatory
Determination for Contaminants on the Fourth Drinking Water Contaminant Candidate List
{U.S. EPA, 2021, 7487276; U.S. EPA, 2021, 9640861} and the reported maximum summed
concentration of PFOA and PFOS was 7.22 |ig/L2 and the median summed value was 0.05 |ig/L.
Summed PFOA and PFOS concentrations reported in UCMR 3 exceeded one-half the health
reference level (HRL)3 (0.035 |ig/L) at a minimum of 2.4% of PWSs (115 PWSs) and exceeded
the HRL (0.07 |ig/L) at a minimum of 1.3% of PWSs (63 PWSs). Since the time of UCMR 3
monitoring, some sites where elevated levels of PFOA and PFOS were previously detected may
have installed treatment for PFOA and PFOS, may have chosen to blend water from multiple
sources, or may have otherwise remediated known sources of contamination. However, the
extent of these changes is unknown. The identified 63 PWSs serve a total population of
approximately 5.6 million people and are located across 25 states, tribes, or U.S. territories {U.S.
EPA, 2017, 9419085}.
Data from more recent state monitoring efforts demonstrate occurrence in multiple geographic
locations consistent with UCMR 3 monitoring {U.S. EPA, 2021, 7487276}. In 2021, at the time
of publication of the final regulatory determinations for PFOA and PFOS, the finished water data
available from fifteen states collected since UCMR 3 identified at least 29 PWSs where the
summed concentrations of PFOA and PFOS exceeded the EPA HRL {U.S. EPA, 2021,
7487276}. The agency notes that some of these data are from targeted sampling efforts and thus
may not be representative of levels found in all PWSs within the state or represent occurrence in
other states. The state data demonstrate occurrence in multiple geographic locations and support
EPA's finding that PFOA and PFOS occur with a frequency and at levels of public health
concern in drinking water systems across the United States.
Likewise, Glassmeyer et al. (2017, 3454569) sampled source and treated drinking water from 29
drinking water treatment plants for a suite of emerging chemical and microbial contaminants,
including 11 PFAS. In this study, PFOA was reported in source water at 76% of systems, at a
median concentration of 6.32 ng/L and maximum concentration of 112 ng/L. Similarly, in treated
drinking water, PFOA was detected in 76% of systems, with a median concentration of 4.15 ng/L
and maximum concentration of 104 ng/L.
2 Sum of PFOA + PFOS results rounded to 2 decimal places in those cases where a laboratory reported more digits.
3 An HRL is a health-based concentration against which the agency evaluates occurrence data when making decisions about
regulatory determinations. The HRL for PFOA that was used to evaluate UCMR 3 results was 0.070 (ig/L (equal to the 2016
Drinking Water Health Advisory value).
1-9
-------
DRAFT FOR PUBLIC COMMENT
March 2023
2 Summary of Assessment Methods
This section summarizes the methods used for the systematic review of the health literature for
all isomers of PFOA and PFOS, as well as nonmetal salts of PFOA and PFOS that would be
expected to dissociate in aqueous solutions of pH ranging from 4 to 9 (e.g., in the human body).
The purposes of the systematic review were to identify the best available and most relevant
health effects literature, to evaluate studies for quality, and to subsequently identify and consider
studies that can be used for dose-response assessment. A detailed description of these methods is
provided as a protocol in the Appendix (see PFOA Appendix).
The information that was gathered in the systematic review was used to update EPA's 2016
HESD for PFOA {U.S. EPA, 2016, 3603279} and to derive an MCLG to support a National
Primary Drinking Water Regulation under the Safe Drinking Water Act.
2.1 Introduction to the Systematic Review Assessment
Methods
The methods used to conduct the systematic review for PFOA are consistent with the methods
described in the draft and final EPA ORD Staff Handbook for Developing IRIS Assessments
{U.S. EPA, 2020, 7006986; U.S. EPA, 2022, 10367891} (hereafter referred to as the Integrated
Risk Information System (IRIS) Handbook) and a companion publication {Thayer, 2022,
10259560}. EPA's IRIS Handbook has incorporated feedback from the National Academy of
Sciences (NAS) at workshops held in 2018 and 2019 and was well regarded by the NAS review
panel for reflecting "significant improvements made by EPA to the IRIS assessment process,
including systematic review methods for identifying chemical hazards" {NAS, 2021, 9959764}.
Furthermore, EPA's IRIS program has used the IRIS Handbook to develop toxicological reviews
for numerous chemicals, including some PFAS. Though the IRIS Handbook was finalized
concurrently with this assessment, the alterations in the final IRIS Handbook compared to the
draft version did not conflict with the methods used in this assessment. In fact, many of the NAS
recommendations incorporated into the final IRIS handbook (e.g., updated methods for evidence
synthesis and integration) were similarly incorporated into this assessment protocol {NAS, 2021,
9959764}. However, some of the study evaluation refinements recommended by NAS {2021,
9959764}, including clarifications to the procedure for evaluating studies for sensitivity and
standardizing the procedure for evaluating reporting quality between human and animal studies,
were not included in this assessment protocol, consistent with a 2011 NASEM recommendation
not to delay releasing assessments until systematic review methods are finalized {NRC, 2011,
710724}. The assessment team concluded that implementing these minor changes in study
quality evaluation would not change the assessment conclusions. Therefore, EPA considers the
methods described herein to be consistent with the final IRIS Handbook and cites this version
accordingly.
For this updated toxicity assessment, systematic review methods used were comparable to those
in the IRIS Handbook for the steps of literature search, screening, study quality evaluation, data
extraction, and the display of study quality evaluation results for all health outcomes through the
2020 literature searches {U.S. EPA, 2022, 10476098}. EPA then focused the subsequent steps of
the systematic review process (synthesis of human, experimental animal, and mechanistic data;
evidence integration; derivation of toxicity values) on health effects outcomes with the strongest
2-10
-------
DRAFT FOR PUBLIC COMMENT
March 2023
weight of evidence (developmental, hepatic, immune, cardiovascular, and cancer) based on the
conclusions presented in EPA's preliminary analysis, Proposed Approaches to the Derivation of
a Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-
1) in Drinking Water, and consistent with the recommendations of the SAB {U.S. EPA, 2022,
10476098}.
This section provides a summary of methods used to search and screen the literature identified,
evaluate the studies and characterize study quality, extract data, and identify studies that can be
used for dose-response analysis. Extracted data are available in interactive visual formats (see
Section 3) and can be downloaded in open access formats.
The systematic review protocol (see PFOA Appendix) provides a detailed description of the
systematic review methods that were used. The particular focus of the protocol is the description
of the problem formulation and key science issues guiding this assessment.
2.1.1 Literature Search
EPA assembled an inventory of epidemiological, animal toxicological, mechanistic, and
toxicokinetic studies for this updated toxicity assessment based on three data streams: 1)
literature published from 2014 through 2019 and then updated in the course of this review (i.e.,
through February 3, 2022) identified via literature searches of a variety of publicly available
scientific literature databases, 2) literature identified via other sources (e.g., searches of the gray
literature and studies shared with EPA by the SAB), and 3) literature identified in EPA's 2016
HESDs for PFOA and PFOS {U.S. EPA, 2016, 3603279; U.S. EPA, 2016, 3603365}.
The search strings for the new searches for this updated assessment focused on the chemical
name (PFOA, PFOS, and their related salts) with no limitations on lines of evidence (i.e.,
human/epidemiological, animal, in vitro, in silico) or health outcomes. EPA conducted an
updated literature search in 2019 (covering January 2013 through April 11, 2019), which was
subsequently updated by a search covering April 2019 through September 3, 2020 (2020
literature search) and another covering September 2020 through February 3, 2022 (2022
literature search) using the same search strings used in 2019.
The publicly available databases listed below were searched for literature containing the
chemical search terms outlined in the PFOA Appendix:
• Web of Science™ (WoS) (Thomson Reuters),
• PubMed® (National Library of Medicine),
• ToxLine (incorporated into PubMed post 2019), and
• TSCATS (Toxic Substances Control Act Test Submissions).
In addition to the databases above, other review efforts and searches of publicly available
sources were used to identify relevant studies, as listed below:
• studies cited in assessments published by other U.S. federal, international, and/or U.S.
state agencies (this included assessments by ATSDR and California Environmental
Protection Agency (CalEPA)),
2-11
-------
DRAFT FOR PUBLIC COMMENT
March 2023
• studies identified during mechanistic or toxicokinetic synthesis (i.e., during manual review
of reference lists of relevant mechanistic and toxicokinetic studies deemed relevant after
screening against mechanistic- and ADME-specific PECO criteria), and
• studies identified by the SAB in their final report dated August 23, 2022 {U.S. EPA,
2022, 10476098}.
The details of the studies included from the 2016 PFOA HESD as well as the search strings and
literature sources searched are described in the Appendix (see PFOA Appendix).
EPA relied on epidemiological and animal toxicological literature identified in the 2016 PFOA
HESD to identify studies for this updated assessment on five major health outcomes, as
recommended by SAB and consistent with EPA's preliminary analysis in the Proposed
Approaches to the Derivation of a Draft Maximum Contaminant Level Goal for
Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking Water. The 2016 HESD for
PFOA contained a summary of all relevant literature identified in searches conducted through
2013. EPA's 2016 HESD relied on animal toxicological studies for quantitative analyses
whereas epidemiology studies were considered qualitatively, as a supporting line of evidence.
This updated assessment includes the study quality evaluation of epidemiological studies that
were identified and included in the 2016 HESD for the five main health outcomes that had the
strongest evidence. It also includes "key" animal toxicological studies from the HESD, which
includes studies that were selected in 2016 for dose-response modeling. More details are
provided in the Appendix (see PFOA Appendix).
All studies identified in the literature searches as well as those brought forward from the 2016
PFOA HESD were uploaded into the Health and Environmental Research Online (HERO)
database (https://hero.epa.eov/hero/index.cfm/proiect/paee/proiect id/2608) and are publicly
available.
EPA has continued to monitor the literature published since February 2022 for other potentially
relevant studies published after the 2022 literature search update. Potentially relevant studies
identified after February 2022 that were not recommended by the SAB in their final report are
not included as part of the evidence base for this updated assessment but are provided in a
repository detailing the results and potential impacts of new literature on the assessment (see
PFOA Appendix A.3).
2.1.2 Literature Screening
This section summarizes the methods used to screen the identified health effects, mechanistic,
and absorption, distribution, metabolism, excretion (ADME) literature. Briefly, PECO statements
were established and detail the criteria used to screen all of the literature identified from
literature searches in this assessment, prioritize the dose-response literature for dose-response
assessment, and identify studies containing potentially important supplemental information that
may inform key science questions described in the protocol. The PECO criteria used for
screening the literature are provided in the Appendix (see PFOA Appendix).
Consistent with protocols outlined in the IRIS Handbook {U.S. EPA, 2022, 10476098}, studies
identified in the literature searches and stored in HERO were imported into the Swift-Review
software platform and the software was used to identify those studies most likely to be relevant
2-12
-------
DRAFT FOR PUBLIC COMMENT
March 2023
to human health risk assessment. Studies captured then underwent title and abstract screening by
at least two reviewers using DistillerSR or SWIFT ActiveScreener software, and studies that
passed this screening underwent full-text review. Dose-response studies that met PECO inclusion
criteria following both title and abstract screening and full-text review underwent study quality
evaluation as described below. Studies tagged as supplemental and containing potentially
relevant mechanistic or ADME (or toxicokinetic) data following title and abstract and full-text
level screening underwent further screening using mechanistic- or ADME-specific PECO
criteria, and those deemed relevant underwent light data extraction of key study elements (e.g.,
extraction of information about the tested species or population, mechanistic or ADME
endpoints evaluated, dose levels tested; see PFOA Appendix). Supplemental studies that were
identified as mechanistic or ADME via screening did not undergo study quality evaluation.
2.1.3 Study Quality Evaluation for Epidemiological Studies
and Animal Toxicological Studies
For study quality evaluation of the PECO-relevant human epidemiological and animal
toxicological studies identified in the three literature searches (all health outcomes for the 2019
and 2020 searches; the five priority health outcomes for the 2022 search), epidemiological
studies from the 2016 HESD that reported results on one or more of the five priority health
outcomes, and key animal toxicological studies from the 2016 HESD, two or more quality
assurance (QA) reviewers, working independently, assigned ratings about the reliability of study
results {good, adequate, deficient (or "not reported"), or critically deficient) for different
evaluation domains. These study quality evaluation domains are listed below and details about
the domains, including prompting questions and suggested considerations, are described in the
PFOA Appendix.
• Epidemiological study quality evaluation domains: participant selection; exposure
measurement criteria; outcome ascertainment; potential confounding; analysis; selective
reporting; and study sensitivity.
• Animal toxicological study quality evaluation domains: reporting; allocation;
observational bias/blinding; confounding/variable control; reporting and attrition bias;
chemical administration and characterization; exposure timing, frequency, and duration;
endpoint sensitivity and specificity; and results presentation.
The independent reviewers performed study quality evaluations using a structured platform
housed within EPA's Health Assessment Workplace Collaboration (HAWC;
https://hawcproi ect.ore/). Once the individual domains were rated, reviewers independently
evaluated the identified strengths and limitations of each study to reach an overall classification
on study confidence of high, medium, low, or uninformative for each PECO-relevant endpoint
evaluated in the study. A study can be given an overall mixed confidence classification if
different PECO-relevant endpoints within the study receive different confidence ratings (e.g.,
medium and low confidence classifications).
2.1.4 Data Extraction
Data extraction was conducted for all relevant human epidemiological and animal toxicological
studies determined to be of medium and high confidence after study quality evaluation. Data
2-13
-------
DRAFT FOR PUBLIC COMMENT
March 2023
were also extracted from low confidence epidemiological studies when data were limited for a
health outcome or when there was a notable effect, consistent with the IRIS Handbook {U.S.
EPA, 2022, 10476098}. Studies evaluated as being uninformative were not considered further
and therefore did not undergo data extraction. All health endpoints were considered for
extraction, regardless of the magnitude of effect or statistical significance of the response relative
to the control group. The level of detail in data extractions for different endpoints within a study
could differ based on how the data were presented for each outcome (i.e., ranging from a
narrative to a full extraction of dose-response effect size information).
Extractions were conducted using DistillerSR for epidemiological studies and HAWC for animal
toxicological studies. An initial reviewer conducted the extraction, followed by an independent
QA review by a second reviewer who confirmed accuracy and edited/corrected the extraction as
needed. Discrepancies in data extraction were resolved by discussion and confirmation within
the extraction team.
Data extracted from epidemiology studies included population, study design, year of data
collection, exposure measurement, and quantitative data from statistical models. Data extracted
from statistical models reported in the studies included the health effect category, endpoint
measured, sample size, description of effect estimate, covariates, and model comments. Data
extracted from animal toxicological studies included information on the experimental design and
exposure duration, species and number of animals tested, dosing regime, and endpoints
measured. Further information about data extraction can be found in the PFOA Appendix.
2.1.5 Evidence Synthesis and integration
For the purposes of this assessment, evidence synthesis and integration are considered distinct
but related processes. Evidence synthesis refers to the process of analyzing the results of the
available studies (including their strengths and weaknesses) for consistency and coherence, often
by evidence stream (e.g., human or animal) and health effect outcome. In evidence integration,
the evidence across streams is considered together and integrated to develop judgments (for each
health outcome) about whether the chemical in question poses a hazard to human health.
The evidence syntheses are summary discussions of the body of evidence for each evidence
stream (i.e., human and animal) for each health outcome analyzed. The available human and
animal health effects evidence were synthesized separately, with each synthesis resulting in a
summary discussion of the available evidence. For the animal toxicological evidence stream,
evidence synthesis included consideration of studies rated high and medium confidence. For the
epidemiological evidence stream, evidence synthesis was based primarily on studies of high and
medium confidence, including discussion of study quality considerations, according to the
recommendations of the SAB {U.S. EPA, 2022, 10476098}. Inferences drawn from studies
described in the 2016 PFOA HESD were considered when drawing health effects conclusions.
Epidemiological studies were excluded from the evidence synthesis narrative if they included
data that were reported in multiple studies (e.g., overlapping NHANES studies). Studies
reporting results from the same cohort and the same health outcome as another study were
considered overlapping evidence, and these additional studies were not discussed in the evidence
synthesis narrative to avoid duplication or overrepresentation of results from the same group of
participants. In cases of overlapping studies, the study with the largest number of participants
and/or the most accurate outcome measures was given preference. Consistent with the IRIS
2-14
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Handbook {U.S. EPA, 2022, 10476098}, low confidence epidemiological studies and results
were used only in a supporting role and given less weight during evidence synthesis and
integration compared to high or medium confidence studies. Low confidence epidemiological
studies were included in evidence syntheses in order to capture all of the available data for PFOA
in the weight of evidence analyses.
For evidence integration, integrated judgments that took into account mechanistic considerations
for the five priority health outcomes (i.e., cancer, hepatic, immune, cardiovascular, and
developmental) were drawn for each health outcome across human and animal lines of evidence.
The evidence integration provides a summary of the causal interpretations between PFOA
exposure and health effects based on results of the available epidemiological and animal
toxicological studies, in addition to the available mechanistic evidence. Considerations when
evaluating the available studies included risk of bias, sensitivity, consistency, strength (effect
magnitude) and precision, biological gradient/dose-response, coherence, and mechanistic
evidence related to biological plausibility.
The evidence integration was conducted according to guidance outlined in the IRIS Handbook
and the Systematic Review Protocol for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (Anionic
and Acid Forms) IRIS Assessments {U.S. EPA, 2020, 8642427}. The evidence integration
included evidence stream evaluation, in which the qualitative summaries on the strength of
evidence from studies in animals and humans were evaluated, and subsequent inference across
all evidence streams. Human relevance of animal models as well as mechanistic evidence to
inform mode of action were considered. Evidence integration produced an overall judgment
about whether sufficient or insufficient evidence of an association with PFOA exposure exists
for each human health outcome, as well as the rationale for each judgment. The potential
evidence integration judgments for characterizing human health effects are evidence
demonstrates, evidence indicates (likely), evidence suggests, evidence inadequate, and strong
evidence supports no effect.
Details about evidence synthesis and integration are summarized in the Appendix (see PFOA
Appendix).
2.2 Dose-Response Assessment
Evidence synthesis and integration enabled identification of the health outcomes with the
strongest weight of evidence supporting causal relationships between PFOA exposure and
adverse health effects, as well as the most sensitive cancer and noncancer endpoints. Studies
were evaluated for use in POD derivation on the basis of study design, study quality evaluation,
and data availability. For human evidence, all high or medium confidence studies were
considered; for animal evidence, only animal toxicological studies with at least two PFOA
exposure groups and also of high or medium confidence were considered.
2.2.1 Approach to POD and RfD Derivation for Non-Cancer
Health Outcomes
The current, recommended EPA human health risk assessment approach described in EPA's A
Review of the Reference Dose and Reference Concentration Processes, which is a multistep
approach to dose-response assessment, includes analysis of dose and response within the range
2-15
-------
DRAFT FOR PUBLIC COMMENT
March 2023
of observation, followed by extrapolation to lower exposure levels {U.S. EPA, 2002, 88824}.
For non-cancer health outcomes, EPA performed dose-response assessments to define points of
departure (PODs) and extrapolated from the PODs to RfDs.
For PFOA, EPA performed benchmark dose (BMD) modeling of all animal toxicological studies
considered for dose-response to refine the POD in deriving the RfD. The BMD modeling
approach involves dose-response modeling to obtain BMDs (i.e., dose levels corresponding to
specific response levels near the low end of the observable range of the data) and identifies the
lower limits of the BMDs (BMDLs) which serve as potential PODs for deriving quantitative
estimates below the range of observation {U.S. EPA, 2012, 1239433}. EPA used the publicly
available Benchmark Dose Software (BMDS) program developed and maintained by EPA
(https://www.epa.eov/bmds). BMDS fits mathematical models to the data and determines the
dose (benchmark dose or BMD) that corresponds to a pre-determined level of response
(benchmark response or BMR). For dichotomous data, the BMR is typically set at either 5 or
10% above the background or the response of the control group. For continuous data, a BMR of
one half or one standard deviation from the control mean is typically used when there are no
outcome-specific data to indicate what level of response is biologically significant {U.S. EPA,
2012, 1239433}. For dose-response data for which BMD modeling did not produce an adequate
model fit, a no-observed-adverse-effect level (NOAEL) or lowest-observed-adverse-effect level
(LOAEL) was used as the POD.
For the epidemiological studies considered for dose-response assessment, EPA used multiple
modeling approaches to determine PODs, depending upon the health outcome and the data
provided in the studies. For the developmental, hepatic, and serum lipid dose-response studies,
EPA used a hybrid modeling approach that involves estimating the incidence of individuals
above or below a level considered to be adverse and determining the probability of responses at
specified exposure levels above the control {U.S. EPA, 2012, 1239433} for cases in which EPA
was able to define a level considered clinically adverse for these outcomes (see PFOA Appendix
for details). EPA also performed BMD modeling and provided study LOAELs/NOAELs for the
hepatic and serum lipid dose-response studies as sensitivity analyses of the hybrid approach. For
the immune studies, where a clinically defined adverse level is not well defined, EPA used
multivariate models provided in the studies and determined a BMR according to EPA guidance
to calculate BMDs and BMDLs {U.S. EPA, 2012, 1239433}.
See the PFOA Appendix for additional details on the study-specific modeling.
The general steps for deriving an RfD for PFOA are summarized below.
Step 1: Evaluate the data to identify and characterize endpoints affected by exposure to
PFOA. This step involves selecting the relevant studies and adverse effects to be considered for
BMD modeling. Once the appropriate data are collected, evaluated for study quality, and
characterized for adverse health outcomes, the risk assessor selects health endpoints/outcomes
judged to be relevant to human health and among the most sensitive, defined as effects observed
in the lower exposure range. Considerations that might influence selection of endpoints include
whether data have dose-response information, percent change from controls, adversity of effect,
and consistency across studies.
2-16
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Step la (for dose-response data from a study in an animal model): Convert administered
dose to an internal dose. A pharmacokinetic model is used to predict the internal dose (in the
animals used in the toxicity studies or in humans) that would correspond to the administered
dose used in the study (see 4.1.3 for additional detail). A number of dose-metrics across life
stages are selected for simulation in a mouse, rat, monkey, or human. Concentrations of PFOA in
blood are considered for all the internal dose-metrics.
Step 2: Conduct dose-response modeling. See above and the PFOA Appendix for study-
specific details.
Step 3: Convert the POD to a human equivalent dose (HED) or point of departure human
equivalent dose (PODhed). The POD (a BMDL, NOAEL, or LOAEL) is converted to an HED
following the method described in Section 4.1.3. Briefly, a pharmacokinetic model for human
dosimetry is used to simulate the HED from the animal PODs from Step 2. Pharmacokinetic
modeling is also used to simulate selected epidemiological studies to obtain a chronic dose that
would result in the internal POD obtained from dose-response modeling (Section 4.1.3). Based
on the available data, a serum PFOA concentration was identified as a suitable internal dosimetry
target for the human and animal endpoints of interest.
Step 4: Select appropriate uncertainty factors (UFs) and provide rationale for UF selection.
UFs are applied in accordance with EPA guidelines considering variations in sensitivity among
humans, differences between animals and humans (if applicable), the duration of exposure in the
critical study compared to the lifetime of the species studied, and the completeness of the
epidemiological or animal toxicological database.
Step 5: Calculate the chronic RfD. The RfD is calculated by dividing PODhed by the
composite (total) UF.
PODhed = calculated from the BMDL, NOAEL, or LOAEL using the human pharmacokinetic
(PK) model presented in Section 4.1.3.2.
UFc = Composite (total) UF calculated by multiplying the selected individual UFs for variations
in sensitivity among humans, differences between animals and humans, duration of exposure in
the critical study compared to the lifetime of the species studied, and completeness of the
toxicology database, in accordance with EPA guidelines {U.S. EPA, 2002, 88824}.
2.2.2 Cancer Assessment
2.2.2.1 Approach for Cancer Classification
In accordance with EPA's 2005 Guidelines for Carcinogen Risk Assessment, a descriptive
weight of evidence expert judgment is made, based on all available animal, human, and
mechanistic data, as to the likelihood that a contaminant is a human carcinogen and the
conditions under which the carcinogenic effects may be expressed {U.S. EPA, 2005, 6324329}.
A narrative is developed to provide a complete description of the weight of evidence and
where:
2-17
-------
DRAFT FOR PUBLIC COMMENT
March 2023
conditions of carcinogenicity. The potential carcinogenicity descriptors (presented in the 2005
guidelines) are:
• Carcinogenic to humans
• Likely to be carcinogenic to humans
• Suggestive evidence of carcinogenic potential
• Inadequate information to assess carcinogenic potential
• Not likely to be carcinogenic to humans
More than one carcinogenicity descriptor can be applied if a chemical's effects differ by dose,
exposure route, or mode of action (MOA)4. For example, a chemical may be carcinogenic to
humans above but not below a specific dose level if a key event in tumor formation does not
occur below that dose. MOA information informs both the qualitative and quantitative aspects of
the assessment, including the human relevance of tumors observed in animals. MOA must be
considered separately for each target organ.
2.2.2.2 Derivation of a Cancer Slope Factor
EPA's 2005 Guidelines for Carcinogen Risk Assessment recommends a two-step process for the
quantitation of cancer risk. First, a model is used to fit a dose-response curve to the data, based
on the doses and associated tumors observed. For animal toxicological studies, EPA used the
publicly available Benchmark Dose Software (BMDS) program developed and maintained by
EPA (https://www.epa.eov/bmds). For cancer data, BMDS fits multistage models and the model
is used to identify a POD for extrapolation to the low-dose region based on the BMD associated
with a significant increase in tumor incidence above the control. According to the 2005
guidelines, the POD is the lowest dose that is adequately supported by the data. The BMDio (the
dose corresponding to a 10% increase in tumors) and the BMDLio (the 95% lower confidence
limit on that dose) are also reported and are often used as the POD.
In the second step of quantitation, the POD is extrapolated to the low-dose region of interest for
environmental exposures. The approach for extrapolation depends on the MOA for
carcinogenesis (i.e., linear or nonlinear). When evidence indicates that a chemical causes cancer
through a mutagenic MOA (i.e., mutation of deoxyribonucleic acid (DNA)) or the MOA for
carcinogenicity is not known, the linear approach is used and the extrapolation is performed by
drawing a line (on a graph of dose vs. response) from the POD to the origin (zero dose, zero
tumors). The slope of the line (Aresponse/Adose) gives rise to the CSF, which can be interpreted
as the risk per mg/kg/day. In addition, according to EPA's Supplemental Guidance for Assessing
Susceptibility from Early-Life Exposure to Carcinogens {U.S. EPA, 2005, 88823}, affirmative
determination of a mutagenic MOA (as opposed to defaulting to a mutagenic MOA based on
insufficient data or limited data indicating potential mutagenicity) indicates the potential for
higher cancer risks from a given exposure occurring early in life compared with exposure during
adulthood, and so requires that the application of age-dependent adjustment factors (ADAFs) be
considered in the quantification of risk to account for additional sensitivity of children. The
ADAFs are 10- and 3-fold adjustments that are combined with age specific exposure estimates
4MOA is defined as a sequence of key events and processes, starting with interaction of an agent with a cell, proceeding through
operational and anatomical changes, and resulting in cancer formation. It is contrasted with "mechanism of action," which
implies a more detailed understanding and description of events.
2-18
-------
DRAFT FOR PUBLIC COMMENT
March 2023
when estimating cancer risks from early life (<16 years of age) exposure to a mutagenic
chemical.
In cases for which a chemical is shown to cause cancer via an MOA that is not linear at low
doses, and the chemical does not demonstrate mutagenic or other activity consistent with
linearity at low doses, a nonlinear extrapolation is conducted. EPA's 2005 Guidelines for
Carcinogen Risk Assessment state that "where tumors arise through a nonlinear MOA, an oral
RfD or inhalation reference concentration, or both, should be developed in accordance with
EPA's established practice of developing such values, taking into consideration the factors
summarized in the characterization of the POD." In these cases, an RfD-like value is calculated
based on the key event5 for carcinogenesis or the tumor response.
Once a POD is determined, a PK model is used to calculate the HED for animal oral exposures
(PODhed). The CSF is then calculated by dividing the selected BMR by the PODhed.
For epidemiological data, EPA used linear regression between PFOA exposure and cancer
relative risk to estimate dose-response as well as the generalized least-squares for trend (gist)
modeling {Greenland, 1992, 5069} using STATA vl7.0 (StataCorp. 2021. Stata Statistical
Software: Release 17. College Station, TX: StataCorp LLC). The CSF was then calculated as the
excess cancer risk associated with each ng/mL increase in serum PFOA. The internal serum CSF
was converted to an external dose CSF, which describes the increase in cancer risk per 1 ng/kg-
day increase in dose. EPA also considered evaluating the dose-response data using the BMDS;
however, categorical data from case-control studies cannot be used with the BMDS since these
models are based on cancer risk, and the data needed to calculate risks (i.e., the denominators)
were not available.
See the PFOA Appendix for additional details on the study-specific modeling.
2.3 MCLG Derivation
As provided in SDWA Section 1412(b)(4)(A), EPA establishes the MCLG at the level at which
no known or anticipated adverse effects on the health of persons occur and which allows an
adequate margin of safety. EPA assesses the available science examining cancer and noncancer
health effects associated with oral exposure to the contaminant. Consistent with the statutory
definition of MCLG, EPA establishes MCLGs of zero for carcinogens classified as Carcinogenic
to Humans or Likely to be Carcinogenic to Humans6 for which there is insufficient information
5The key event is defined as an empirically observed precursor step that is itself a necessary element of the MOA or is a
biologically based marker for such an element.
6The MCLG is derived depending on the available noncancer and cancer evidence for a particular chemical. Establishing the
MCLG for a chemical has typically been accomplished in one of three ways depending upon a three-category classification
approach {U.S. EPA, 1985, 9207; U.S. EPA, 1991, 5499}. The categories are based on the available evidence of carcinogenicity
after exposure via ingestion. The starting point in categorizing a chemical is through assigning a cancer descriptor using EPA's
current Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}. The descriptors in the 2005 Guidelines
replaced the prior alphanumeric groupings, although the basis for the classifications is similar. In prior rulemakings, the agency
typically placed Group A, Bl, and B2 contaminants into Category I, Group C into Category II, and Group D and E into Category
III based on the agency's previous cancer classification guidelines (i.e., Guidelines for Carcinogen Risk Assessment, published in
51 FR 33992, September 24, 1986 {U.S. EPA, 1986,199530} and the 1999 interim final guidelines {U.S. EPA, 1999, 41631;
U.S. EPA, 2001,10442464}):
• Category I chemicals have "strong evidence [of carcinogenicity] considering weight of evidence, pharmacokinetics, and
exposure {U.S. EPA, 1985, 9207; U.S. EPA, 1991, 5499}." EPA's 2005 cancer descriptors associated with this category
2-19
-------
DRAFT FOR PUBLIC COMMENT
March 2023
to determine that a carcinogen has a threshold below which there are no carcinogenic effects
{U.S. EPA, 1998, 10442462; U.S. EPA, 2000, 10442463; U.S. EPA, 2001, 10442464}.
For nonlinear carcinogenic contaminants, contaminants that are suggestive carcinogens, and non-
carcinogenic contaminants, EPA establishes the MCLG based on a toxicity value, typically an
RfD, but a similar toxicity value (e.g., ATSDR Minimal Risk Level) may also be used when it
represents the best available science. A noncancer MCLG is designed to be protective of
noncancer effects over a lifetime of exposure with an adequate margin of safety, including for
sensitive populations and life stages consistent with SDWA 1412(b)(3)(C)(i)(V) and
1412(b)(4)(A). The calculation of a noncancer MCLG includes an oral toxicity reference value
such as an RfD, body weight-based drinking water intake (DWI-BW), and RSC as presented in
the equation below:
RfD = chronic reference dose—an estimate (with uncertainty spanning perhaps an order of
magnitude) of a daily oral exposure of the human population to a substance that is likely to be
without an appreciable risk of deleterious effects during a lifetime. The RfD is equal to a
PODhed divided by a composite uncertainty factor.
DWI-BW = An exposure factor in the form of the 90th percentile body weight-adjusted drinking
water intake value for the identified population or life stage, in units of liters of water consumed
per kilogram body weight per day (L/kg bw-day). The DWI-BW considers both direct and
indirect consumption of drinking water (indirect water consumption encompasses water added in
the preparation of foods or beverages, such as tea or coffee). Chapter 3 of EPA's Exposure
Factors Handbook {U.S. EPA, 2019, 7267482} provides DWI-BWs for various populations or
life stages within the general population for which there are publicly available, peer-reviewed
data such as NHANES data.
RSC = relative source contribution—the percentage of the total exposure attributed to drinking
water sources {U.S. EPA, 2000, 19428}, with the remainder of the exposure allocated to all
other routes or sources. The purpose of the RSC is to ensure that the level of a contaminant (e.g.,
MCLG value), when combined with other identified sources of exposure common to the
population of concern, will not result in exposures that exceed the RfD. The RSC is derived by
applying the Exposure Decision Tree approach published in EPA's Methodology for Deriving
are: "Carcinogenic to Humans" or "Likely to be Carcinogenic to Humans" {U.S. EPA, 2005, 6324329}. EPA's policy
under SDWA is to set MCLGs for Category I chemicals at zero, based on the principle that any exposure to known or
likely human carcinogens might represent some finite level of risk. In cases when there is sufficient evidence to
determine a nonlinear cancer mode of action, the MCLG is based on the RfD approach described below.
• Category II chemicals have "limited evidence [of carcinogenicity] considering weight of evidence, pharmacokinetics,
and exposure {U.S. EPA, 1985, 9207; U.S. EPA, 1991, 5499}." EPA's 2005 cancer descriptor associated with this
category is: "Suggestive Evidence of Carcinogenic Potential" {U.S. EPA, 2005, 6324329}. The MCLG for Category II
contaminants is based on noncancer effects {U.S. EPA, 1985, 9207; U.S. EPA, 1991,5499}.
• Category III chemicals have "inadequate or no animal evidence [of carcinogenicity] {U.S. EPA, 1985, 9207; U.S. EPA,
1991, 5499}." EPA's 2005 cancer descriptors associated with this category are: "Inadequate Information to Assess
Carcinogenic Potential" and "Not Likely to Be Carcinogenic to Humans" {U.S. EPA, 2005, 6324329}. The MCLG for
Category III contaminants is based on noncancer effects.
Where:
2-20
-------
DRAFT FOR PUBLIC COMMENT March 2023
Ambient Water Quality Criteria for the Protection of Human Health {U.S. EPA, 2000, 19428}.
Further description of the RSC for PFOA can be found in the Appendix (see PFOA Appendix).
2-21
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3 Results of the Health Effects Systematic
Review and Toxicokinetics Methods
3.1 Literature Search and Screening Results
Studies referenced in this assessment are cited as "Author Last Name, Publication Year, HERO
ID" and are available in EPA HERO: A Database of Scientific Studies and References. The
HERO ID is a unique identifier for studies available in HERO. Additional study metadata are
publicly available and can be obtained by searching for the HERO ID on the public facing
webpage available here: https://hero.epa.eov/.
The three database searches yielded 6,007 unique records prior to running SWIFT Review. Table
3-1 shows the results from database searches conducted in April 2019, September 2020, and
February 2022.
Table 3-1. Database Literature Search Results
Database
Date Run: Results
WoS
4/10/2019: 3,081 results
9/3/2020: 1,286 results
2/2/2022: 1,021 results
PubMed
4/10/2019: 2,191 results
9/3/2020: 811 results
2/2/2022: 1,728 results
TOXLINE
4/10/2019: 60 results
TSCATS
4/11/2019: 0 results
Total number of references from all databases for all
searches"
4/2019: 3,382 results
9/2020: 1,153 results
2/2022: 1,858 results
Total number of references after running SWIFT
Review3
4/2019: 1,977 results
9/2020: 867 results
2/2022: 1,370 results
Total number of unique studies moved to screeningb
3,921
a The number of studies includes duplicate references across search dates due to overlap between search years.
b Duplicates across search dates removed.
The additional sources of literature outlined in Section 2.1.1 (i.e., assessments published by other
agencies, studies identified during mechanistic or toxicokinetic syntheses, and studies identified
by the SAB) yielded 200 unique records.
The 3,921 studies captured with the SWIFT Review evidence streams filters and the 200 records
identified from additional sources yielded a total of 4,121 unique studies. These 4,121 studies
were moved to the next stage of screening—title and abstract screening (using either DistillerSR
or SWIFT ActiveScreener). Of the 4,121 unique studies, 918 moved on to full-text level review,
1,589 were excluded during title and abstract screening, and 1,614 were tagged as containing
potentially relevant supplemental material. Of the 918 screened at the full-text level, 618 were
considered to meet PECO eligibility criteria (see PFOA Appendix) and included relevant
information on PFOA. The 618 studies that were determined to meet PECO criteria after full-text
3-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
level screening included 443 epidemiological (human) studies, 37 animal toxicological studies, 9
PBPK studies (2 of which were also relevant epidemiological studies), and 131 studies that were
not extracted (e.g., low confidence studies, meta-analyses, studies that did not evaluate effects on
one of the priority health outcomes). An additional 20 PBPK studies were identified during the
toxicokinetic screening for a total of 29 PBPK studies. Details of the literature search and
screening process are shown in Figure 3-1.
The 443 epidemiological studies and 37 animal toxicological studies underwent study quality
evaluation and were subsequently considered for data extraction as outlined in Sections 2.1.3 and
2.1.4 (see PFOA Appendix for more details). The results of the health outcome-specific study
quality evaluations and data extractions are described in Sections 3.4 and 3.5.
Additionally, the 29 studies tagged as containing relevant PBPK models were reviewed by PK
subject matter experts for inclusion consideration. The included studies are summarized in
Section 3.3.2 and parameters described in these studies were considered for incorporation into
the animal and human PK models, which are summarized in Section 4.1.3.
Finally, the 113 toxicokinetic and 270 mechanistic studies identified as relevant for PFOA
moved on to a limited data extraction as described in the Appendix (see PFOA Appendix). The
toxicokinetic studies pertaining to ADME are synthesized in Section 3.3.1. The mechanistic
studies relevant to the 5 prioritized health outcomes are synthesized in Sections 3.4 and 3.5 and
were considered as part of the evidence integration.
3-2
-------
DRAFT FOR PUBLIC COMMENT
March 2023
References identified through
database search
2013 to 2019
2019 to 2020
2020 to 2022
3,382
1,153
1,858
References after
SWIFT fitters applied
2013 to 2019:
2019 to 2020
2020 to 2022
1,977
867
1,370
References screened for
PFOA and PFOS3
4,121
References assessed for eligibility
for PFOA and PFOS
References identified from
2016 PFOA & PFOS HESDs
References identified through
other sources
References excluded 1,589
Not PECO Relevant 1,117
Excluded by SWIFT-Active 472
Supplemental Tag
Other Supplemental
Mechanistic
Toxicokinetic
References excluded
1,614
1,309
525
232
Deduplication
61
Not PECO Relevant
14
Supplemental Tag
163
Other Supplemental
132
Mechanistic
70
—
Toxicokinetic
19
Relevant references for PFOA
Toxicokinetic/Mechanistic
; 829
references assessed for eligibility
Mechanistic 652
Toxicokinetic 280
Did Not Extract
131
Human Animal
443 37
PBPK6
29
Toxicokinetic
113
Mechanistic
270
Figure 3-1. Summary of Literature Search and Screening Process for PFOA
Interactive figure and additional study details available on Tableau.
Interactive figure based on work by Magnuson et al. (2022, 10442900).
"Other sources" include assessments published by other agencies, studies identified during mechanistic or toxicokinetic
syntheses, and studies identified by the SAB.
a Includes number of unique references after deduplication of studies captured with the SWIFT Review evidence streams filters
and records identified from additional sources.
b Includes number of unique references considered to meet PECO eligibility criteria at the full text level and include relevant
information on PFOA.
c Includes number of unique references identified during title/abstract screening, full text screening, and data extraction assessed
for toxicokinetic and/or mechanistic eligibility.
d Only includes studies with relevant information on PFOA.
e Includes 9 PBPK studies (2 of which were also relevant epidemiological studies) determined to meet PECO criteria plus an
additional 20 PBPK studies identified during the toxicokinetic screening.
3-3
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.1.1 Results for Epidemiology Studies of PFOA by Health
Outcome
Of the 443 epidemiological studies that met the inclusion criteria, 189 had a cohort study design,
175 had a cross-sectional design, 40 had a case-control design, and 39 had other study designs
(e.g., nested case-control). Epidemiological studies were categorized into 18 health outcomes.
Most studies reported on the cardiovascular (n = 93), developmental (n = 92), metabolic
(n = 78), or immune systems (n = 64). Studies that reported outcomes spanning multiple health
outcomes were not counted more than once in the grand totals shown in Figure 3-2.
Study Design
Health System
Case-control
Cohort
Cross-sectional
Other
Grand Total
Cancer
6
6
3
5
20
Cardiovascular
5
21
60
7
93
Dermal
0
1
0
0
1
Developmental
4
61
20
7
92
Endocrine
1
8
18
B
35
Gastrointestinal
1
6
0
0
7
Hematologic
0
0
7
1
e
Hepatic
1
6
20
4
31
Immune
5
33
17
9
64
Metabolic
7
36
30
5
78
Musculoskeletal
0
1
6
2
9
Nervous
3
26
5
3
37
Ocular
0
0
1
0
1
Renal
0
6
18
2
26
Reproductive, Male
0
7
14
1
22
Reproductive, Female
9
24
22
4
59
Respiratory
1
4
1
0
6
Other
0
3
3
0
6
Grand Total
40
189
175
39
443
Figure 3-2. Summary of Epidemiology Studies of PFOA Exposure by Health System and
Study Design3
Interactive figure and additional study details available on Tableau.
a A study can report on more than one health system. Column grand totals represent the number of unique studies and are not a
sum of health system tags.
3.1.2 Results for Animal Toxicological Studies of PFOA by
Health Outcome
Of the 37 animal toxicological studies that met the inclusion criteria, most studies had either
short-term (n = 16) or developmental (n = 13) study designs and most were conducted in mice (n
= 30). The mouse studies had short-term (n = 15), developmental (n = 13), and subchronic
(n = 2) study designs. The remaining studies reported results for rats (n = 7) using chronic
(n = 3), short-term (n = 2), subchronic (n = 1), or reproductive (n = 1) study designs, or monkeys
(n = 1) using a chronic study design. Animal toxicological studies were categorized into 15
3-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
health outcomes. Most studies reported results for the hepatic (n = 27), whole body (n = 23; i.e.,
systemic effects such as bodyweight), reproductive (n = 18), or developmental (n = 14) systems.
Studies that reported outcomes spanning multiple health outcomes, study designs, or species
were not counted more than once in the grand totals shown in Figure 3-3.
Study Design & Species
Health System
Short-term
Mouse
Rat
Subchronic
Mouse
Rat
Chronic
Monkey Rat
Developmental
Mouse
Reproductive
Rat
Grand Total
Cancer
0
0
0
0
0
3
0
0
3
Cardiovascular
2
2
0
0
0
2
2
0
7
Developmental
0
0
0
0
0
1
12
1
14
Endocrine
3
2
0
0
0
3
2
1
10
Gastrointestinal
0
0
0
0
1
2
0
3
Hematologic
1 1
0
0
0
1
0
3
Hepatic
11
2
2
1
0
3
8
1
27
Immune
5
2
2
0
0
2
2
1
13
Metabolic
0 1
0
0
0
2
2
5
Musculoskeletal
1
0
0
0
0
0
0
1
Nervous
2
0
0
0
0
1
2
1
6
Renal
1
1
1
0
0
2
1
1
7
Reproductive
3
1
1 1
0
3
8
1
18
Respiratory
0
1
0
0
0
1
0
2
Whole Body
¦
2
2
1
0
3
5
1
23
Grand Total
15
2
2
1
1 3
13
1
37
Figure 3-3. Summary of Animal Toxicological Studies of PFOA Exposure by Health
System, Study Design, and Speciesa'b
Interactive figure and additional study details available on Tableau.
a A study can report on more than one study design and species. Row grand totals represent the number of unique studies and are
not a sum of study design and species tags.
b A study can report on more than one health system. Column grand totals represent the number of unique studies and are not a
sum of health system tags.
3.2 Data Extraction Results
Data extracted from the 443 epidemiological studies are available via Tableau Public and data
extracted from the 37 animal toxicological studies are available in the public HAWC site,
displayed as exposure-response arrays, forest plots, and trees. See Sections 3.4 and 3.5 for health
outcome-specific data extracted for synthesis development. Additionally, the limited data
extractions from the ADME and mechanistic studies can be found via Tableau Public here and
here, respectively.
3.3 Toxicokinetic Synthesis
As described in Section 3.1, EPA identified 113 and 29 studies containing information relevant
to the toxicokinetics and PBPK modeling of PFOA, respectively. The results of these studies are
described in the subsections below and additional information related to toxicokinetic
characteristics of PFOA can be found in Appendix B.
3.3.1 ADME
PFOA is resistant to metabolic and environmental degradation due to its strong carbon-fluorine
bonds. It also is resistant to metabolic biotransformation. Thus, the toxicity and
pharmacodynamics of the parent compound (the anion when dissociated in water or the body)
are the concern. Because of its impacts on cellular receptors and proteins, PFOA can influence
the biotransformation of dietary constituents, intermediate metabolites, and other xenobiotic
chemicals by altering enzyme activities and transport kinetics. PFOA is known to activate
3-5
-------
DRAFT FOR PUBLIC COMMENT
March 2023
peroxisome proliferator activated receptor (PPAR) pathways by increasing transcription of
mitochondrial and peroxisomal lipid metabolism, sterol, and bile acid biosynthesis and retinol
metabolism genes. Findings of transcriptional activation of many genes in peroxisome
proliferator activated receptor alpha (PPARa)-null mice after PFOA exposure, however, indicate
that the effects of PFOA are mediated by other MO As in addition to PPAR activation and
consequent peroxisome proliferation {Wen; 2019, 5080582; Oshida, 2015, 2850125; Oshida,
2015, 5386121; Rosen, 2017, 3859803; U.S. EPA, 2016, 3603279}. The available data indicate
that PFOA exposure can also activate the constitutive androstane receptor (CAR), farnesoid X
receptor (FXR), and pregnane X receptor (PXR), and can affect metabolic activities linked to
these nuclear receptors {Oshida, 2015, 2850125; Oshida, 2015, 5386121; Rosen, 2017, 3859803;
U.S. EPA, 2016, 3603279}. Activation of these receptors resulting from PFOA exposure could
in turn impact the toxicokinetics of PFOA itself {Andersen, 2008, 3749214}.
PFOA is not readily eliminated from humans and other primates. Toxicokinetic profiles and the
underlying mechanism for half-life differences between species and sexes are not completely
understood, although many of the differences appear to be related to elimination kinetics and
factors that control membrane transport. Thus far, three transport families appear to play a role in
PFOA absorption, distribution, and excretion: organic anion transporters (OATs), organic anion
transporting polypeptides (OATPs), and multidrug resistance-associated proteins (MRPs)
{Klaassen, 2010, 9641804; Launay-Vacher, 2006, 9641802}. These transporters are critical for
gastrointestinal absorption, uptake by the tissues, and excretion via bile and the kidney. These
transport systems are located at the membrane surfaces of the kidney tubules, intestines, liver,
lungs, heart, blood brain barrier (BBB), blood placental barrier, blood testes barrier (BTB), and
mammary glands where they function to protect the organs, tissues, and fetus through active
removal of foreign compounds {Ito, 2003, 9641803; Klaassen, 2010, 9641804, Zair, 2008,
9641805}. However, luminal transporters in the kidney may cause reuptake of PFOA from the
proximal tubule resulting in decreased excretion from the body {Weaver, 2009, 2010072}. This
reuptake would lead to PFOA persisting in the body over time. Transporters involved in
enterohepatic circulation have also been identified that may facilitate uptake and reuptake of
PFOA from the gut {Ruggiero, 2021, 9641806}.
There are differences in transporters across species, sexes, and individuals. In addition, more
PFOA-specific information is available for the OAT and OATP families than for the MRPs.
These data limitations have hindered the development of PK models for use in predicting effects
in humans based on the data from animal toxicological studies.
3.3.1.1 Absorption
PFOA absorption data are available in laboratory animals for oral, inhalation, and dermal
exposures, and extensive data are available from humans demonstrating the presence of PFOA in
serum (descriptions of available studies are provided in the PFOA Appendix). In vitro absorption
data indicate that uptake is influenced by pH, temperature, and concentration as well as OATP
activity (see PFOA Appendix).
3.3.1.1.1 Cellular Uptake
The available information indicates that the absorption process requires transport from the
external environment across the interface of the gut, lung, or skin. Uptake in cells cultured in
vitro is fast and saturable, consistent with a role of transporters. Cellular transfection of cells
3-6
-------
DRAFT FOR PUBLIC COMMENT
March 2023
with vectors coding for organic ion transporters have confirmed their role in uptake of PFOA
{Kimura, 2017, 3981330; Nakagawa, 2007, 2919370; Nakamura, 2009, 2919342; Yang, 2009,
2919328; Yang, 2010, 2919288}. Several studies suggest involvement of OATs, OATPs, and
MRPs in enterocytes in the uptake of PFOA {Klaassen, 2010, 9641804; Zair, 2008, 9641805}.
Few studies have been conducted on the intestinal transporters for PFOA in humans or
laboratory animals, although one study supports a role for OATPs in PFOA uptake by
immortalized intestinal cells {Kimura, 2017, 3981330}. Most of the research has focused on
transporters in the kidney that are relevant to excretion and were carried out using cultured cells
transfected with the transporter proteins.
In addition to facilitated transport, there is evidence supporting passive diffusion in cells cultured
in vitro {Yang, 2009, 2919328} and in placenta in vivo {Zhang, 2013, 3859792}. Since PFOA is
moderately soluble in aqueous solutions and oleophobic (i.e., minimally soluble in body lipids),
movement across interface membranes was thought to be dominated by transporters or
mechanisms other than simple diffusion across the lipid bilayer. Recent mechanistic studies,
however, support transporter-independent uptake through passive diffusion processes. Ebert et
al. (2020, 6505873) determined membrane/water partition coefficients (Kmem/w) for PFOA and
examined passible permeation into cells by measuring the passive anionic permeability (Pion)
through planar lipid bilayers. In this system, the partition coefficients (PCs) were considered
high enough to explain observed cellular uptake by passive diffusion in the absence of active
uptake processes.
Uptake by cells may be influenced by interactions with lipids and serum proteins. PFOA
exhibited lower levels of binding to lipids and phospholipids relative to PFOS, which correlated
with uptake into lung epithelial cells {Sanchez Garcia, 2018, 4234856}. Phospholipophilicity
correlated to cellular accumulation better than other lipophilicity measures. The extent to which
PFOA phospholipophilicity influences absorption through the gastrointestinal tract, lungs, or
skin is unknown.
3.3.1.1.2 Absorption and Bioavailability in Humans and Animals
In vivo, PFOA is well-absorbed following oral exposure, as evidenced by the presence of PFOA
in serum of humans following exposure to contaminated drinking water {Xu, 2020, 6781357;
Worley, 2017, 3859800}. Studies on male rats administered PFOA by gavage using a single or
multiple dose regimen estimated dose absorption of at least 92.3% {Gibson, 1979, 9641813; Cui,
2010, 2919335}. In rats, the time to reach the maximum PFOA plasma concentration (Tmax)
following oral exposure is very fast and varies by sex {Kim, 2016, 3749289; Dzierlenga, 2019,
5916078}. For example, the study by Kim and colleagues estimated Tmax after a single oral dose
of 1 mg/kg to be 1.44 hours in female rats vs. 2.07 days in males.
Recent studies confirm that bioavailability of PFOA after oral exposure is very high in rats.
Serum concentrations after oral dosing ranged from 82-140% of levels measured after
intravenous (IV) dosing, which may reflect increased reabsorption by intestinal transporters by
the oral route relative to the IV route of exposure {Kim, 2016, 3749289; Dzierlenga, 2019,
5916078}. Bioavailability of PFOA appears to be modified by diet. Using in vitro and in vivo
(BALB/c mice) systems, Li et al. (2015, 2851033) found that PFOA bioavailability is strongly
influenced by diet, with high fat diets associated with reduced absorption. The authors suggest
3-7
-------
DRAFT FOR PUBLIC COMMENT
March 2023
that colloidal stability in intestinal solutions may be an important factor influencing PFOA
bioaccessibility.
The available data, although limited, also support PFOA absorption through both inhalation
1 Hinderliter, 2006, 135732} and dermal routes 1 Fasano, 2005, 3749187; O'Malley, 1981,
4471529; Kennedy, 1985, 3797585}.
33.1.2 Distribution
3.3.1.2.1 PFOA Binding to Blood Fractions and Serum Proteins
Detailed study descriptions of literature regarding the distribution of PFOA in humans and
animals are provided in the Appendix (see PFOA Appendix). Distribution of absorbed material
requires vascular transport from the portal of entry to receiving tissues. Distribution of PFAS to
plasma has been reported to be chain length-dependent {Jin, 2016, 3859825}. Increasing chain
length (from C6 to CI 1) correlated with an increased mass fraction in human plasma. Within the
blood cell constituents, PFOA preferentially accumulates in platelets over red blood cells and
leukocytes {De Toni, 2020, 6316907}. Among different kinds of human blood samples, PFOA
accumulates to highest levels in plasma, followed by whole blood and serum {Forsthuber, 2020,
6311640; Jin, 2016, 3859825; Poothong, 2017, 4239163}. Poothong etal. (2017, 4239163)
found that median PFOA concentrations in plasma, serum, and whole blood were 1.90, 1.60, and
0.93 ng/mL, respectively. These findings suggest that the common practice of multiplying by a
factor of 2 to convert the concentrations in whole blood to serum {Ehresman, 2007, 1429928}
will not provide accurate estimates for PFOA.
PFOA is distributed within the body by noncovalently binding to plasma proteins. Many studies
have investigated PFOA interactions with human serum albumin (HSA) {Wu, 2009, 536376;
MacManus-Spencer, 2010, 2850334; Qin, 2010, 3858631; Salvalaglio, 2010, 2919252; Weiss,
2009, 534503; Luebker, 2002, 1291067; Zhang, 2013, 5081488; Cheng, 2018, 5024207; Gao,
2019, 5387135; Yue, 2016, 3479514}. In vitro analyses found that plasma proteins can bind
97%-100% of the PFOA in plasma from humans, cynomolgus monkeys, and rats {Kerstner-
Wood, 2003,4771364}.
HSA is the primary PFOA binding protein in plasma {Han, 2003, 5081471} and intermolecular
interactions are mediated through van der Waals forces and hydrogen bonds {MacManus-
Spencer, 2010, 2850334; Chen, 2020, 6324256}. Beesoon and Martin (2015, 2850292)
determined that linear PFOA molecules bound more strongly to calf serum albumin than the
branched chain isomers in the order of 4m < 3m < 5m < 6m (iso) < linear. PFOA-mediated
conformational changes may interfere with albumin's ability to transport its natural ligands and
pharmaceuticals {Wu, 2009, 536376} such as fatty acids, thyroxine (T4), warfarin, indole, and
benzodiazepine.
Binding to albumin and other serum proteins may affect transfer of PFOA from maternal blood
to the fetus {Gao, 2019, 5387135}. Since there is effectively a competition between PFOA
binding in maternal serum vs. cord blood, lower cord blood albumin levels compared to maternal
blood albumin levels are likely to reduce transfer from maternal serum across the placenta.
Consistent with this hypothesis, Pan et al. (2017, 3981900) found that high concentration of cord
serum albumin was associated with higher PFOA transfer efficiencies, whereas high maternal
serum albumin concentration was associated with reduced transfer efficiency.
3-8
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Other plasma proteins that bind PFOA, albeit with lower affinity than HSA, include low-density
lipoproteins (LDLs), alpha-globulins (alpha-2-macroglobulin), gamma-globulins, transferrin, and
fibrinogen {Kerstner-Wood, 2003, 4771364}. PFOA also binds the serum thyroid hormone
transport protein, transthyretin (TTR), causing up to a 50% inhibition of T4 binding to TTR
{Weiss, 2009, 534503}. In contrast to serum proteins, little is known regarding PFOA binding to
proteins in the gut. One study found that PFOA can bind to and cause a conformational change in
pepsin {Yue, 2016, 3479514}, though it is unclear whether PFOA-pepsin interactions impact
absorption from the gut or distribution to other compartments in the body.
3.3.1.2.2 PFOA Binding to Subcellular Fractions, Intracellular Proteins, and
Transporters
Han et al. (2005, 5081570) observed a sex-dependent subcellular distribution of PFOA in the
liver and kidney of male and female adult rats necropsied 2 hours after oral gavage dosing. The
proportion of PFOA in the liver cytosol of female rats was almost twice that of the male rats.
They hypothesized that females might have a greater amount than males of an unknown liver
cytosolic binding protein with an affinity for perfluorinated acids. In the kidney, the subcellular
distribution did not show a sex difference comparable to the one seen for liver; however, the
protein-bound fraction in males (42%) was about twice that of females (17%), which differs
from the sex differences found for the liver.
In a study of human cells {Zhang, 2020, 6316915}, PFOA preferentially distributed to cytosol
followed by nuclei and mitochondria in human colorectal cancer cells, human lung epithelial
cells, and human normal liver cells. In liver cells, PFOA binds to the liver fatty acid binding
protein (L-FABP) through polar and hydrophobic interactions {Luebker, 2002, 1291067; Zhang,
2013, 5081488; Yang, 2020, 6356370}. L-FABP is an intracellular lipid carrier protein that
reversibly binds long-chain fatty acids, phospholipids, and an assortment of peroxisome
proliferators {Erol, 2004, 5212239} and constitutes 2%-5% of the cytosolic protein in
hepatocytes.
PFOA interactions with various protein transporters play a role in the tissue uptake of orally
ingested PFOA. The transporters are located at the interface between serum and a variety of
tissues (e.g., liver, kidneys, lungs, heart, brain, testes, ovaries, placenta, uterus) {Klaassen, 2010,
9641804}. The liver is an important uptake site for PFOA. OATPs and MRPs, at least one OAT,
and the sodium-taurocholate cotransporting polypeptide (NTCP)—a hepatic bile uptake
transporter—have been identified at the boundary of the liver at the portal blood and/or the
canalicular membranes within the liver {Kim, 2003, 9641809; Kusuhara, 2009, 9641810; Zair,
2008, 9641805}. Transporters responsible for PFOA transport across the placenta are not well
understood, though preliminary studies examining transporter expression identified OAT4 as a
candidate receptor {Kummu, 2015, 3789332}. The expression of 9 transporter genes was found
to vary at different stages of gestation {Li, 2020, 6505874}, though direct experimental evidence
for these transporters in mediating transfer of PFOA to the fetus is lacking.
3.3.1.2.3 Tissue Distribution in Humans and Animals
Evidence from human autopsy and surgical tissues demonstrates that PFOA distributes to a wide
range of tissues, organs, and matrices throughout the body. Although blood and liver are major
sites of PFOA accumulation {Olsen, 2001, 9641811}, recent findings confirm PFOA
accumulation in other tissues and fluids including brain and cerebral spinal fluid {Fujii, 2015,
3-9
-------
DRAFT FOR PUBLIC COMMENT
March 2023
2816710; Wang, 2018, 5080654}, thyroid gland {Pirali, 2009, 757881}, and follicular fluid
{Kang, 2020, 6356899}. Perez et al. (2013, 2325349) measured PFOA levels in autopsy tissue
samples (liver, kidney, brain, lung, and bone) collected within 24 hours of death and found
PFOA in bone (60.2 ng/g), lung (29.2 ng/g), liver (13.6 ng/g), and kidney (2.0 ng/g), with levels
below the limit of detection (LOD) in the brain. It should be noted, however, that autopsy and
surgical tissues may not necessarily accurately reflect PFAS tissue distribution in the living body
{Cao, 2021,9959613}.
Most whole animal toxicological studies that measured PFOA distribution were conducted in rats
and mice by oral dosing. Studies in primates measured PFOA in blood and liver following oral
administration {Butenhoff 2002, 1276161; Butenhoff, 2004, 3749227}. PFOA primarily
distributes to serum, liver, lungs, and kidney across a range of dosing regimens and durations
{Ylinen, 1990, 5085631; Kemper, 2003, 6302380; NTP, 2020, 7330145; NTP, 2019, 5400977}
in rats and in mice {Lau, 2006, 1276159; Lou, 2009, 2919359; Burkemper, 2017, 3858622; Li,
2017, 4238518; Guo, 2019, 5080372}. Sex-specific differences in PFOA levels were observed in
several rat studies. For example, in a 28-day study {NTP, 2019, 5400977}, PFOA plasma
concentrations were higher in males than in females across all dose groups even though females
were administered a 10-fold higher dose of PFOA, suggesting that female rats excrete PFOA
more efficiently than males. Sex-specific differences were less striking in studies conducted in
mice compared to rats {Lau, 2006, 1276159; Lou, 2009, 2919359}.
Liver PFOA levels are regulated in part by PPARa. In human and rodent hepatocytes, PPARa
activation induces expression of genes involved in lipid metabolism and cholesterol homeostasis.
PFOS and PFOA structurally resemble fatty acids and are well-established ligands of PPARa in
the rat and mouse liver. As PPARa agonists, PFOS and PFOA can induce B-oxidation of fatty
acids, induce fatty acid transport across the mitochondrial membrane, decrease hepatic very low-
density lipoprotein (VLDL)-triglyceride and apolipoprotein B (apoB) production, and promote
lipolysis of triglyceride-rich plasma lipoproteins {Fragki, 2020, 8442211}. The liver can
transport PFOA from hepatocytes to bile ducts, which is mediated at least partly by PPARa
{Minata, 2010, 1937251}. PFOA levels were significantly lower in PPARa-null mice than in
wild-type mice exposed to doses of 25 and 50 |imol/kg, supporting a role for PPARa in PFOA
clearance in the liver {Minata, 2010, 1937251} but not excluding other factors regulating PFOA
levels. It is unclear what role PPARa plays in PFOA clearance in the liver of humans.
Studies administering radiolabeled PFOA to whole animals demonstrate the range of tissue
distribution in rats {Kemper, 2003, 6302380} and mice {Burkemper, 2017, 3858622;
Bogdanska, 2020, 6315801} that includes the central nervous system (CNS), cardiovascular,
gastrointestinal, renal, immune, reproductive, endocrine, and musculoskeletal systems. PFOA
crossed the BBB in males an order of magnitude more efficiently than in females {Ylinen, 1990,
5085631}. Fujii and colleagues (2015, 2816710) found that PFOA can cross the BBB in mice,
although a relatively small amount of administered PFOA was measured in the brains (0.1%).
Also in mice, Burkemper et al. (2017, 3858622) observed the highest PFOA levels in bone, liver,
and lungs. Bogdanska et al. (2020, 6315801) also observed PFOA in testes of C57BL/6 mice at
levels similar to those observed in epididymal fat and in intestines. In BALB/c mice exposed to
PFOA for 28 days, PFOA levels in the testes increased with increasing dose {Zhang, 2014,
2850230}, and PFOA accumulated in the epididymis of BALB/c mice in a dose-dependent
manner {Lu, 2016, 3981459}.
3-10
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Fujii and colleagues (2015, 2816710) observed that perfluoroalkyl carboxylic acids (PFCAs) (C6
and C7) were excreted rapidly through urine in mice, whereas longer-chained PFCAs (> C8)
accumulated in the liver. Moreover, PFAS with longer chain lengths were found to exhibit
increasing affinity for serum and L-FABPs. The authors suggest that differential lipophilicity
driven by chain length may account for the distribution patterns of PFAS, which is consistent
with the findings of high levels of PFOA accumulation in serum and liver. These large
sequestration volumes of PFOA observed in the liver seem to be attributable to the liver's large
binding capacity in mice.
3.3.1.2.4 Distribution During Reproduction and Development
Several studies have confirmed PFOA distribution from rat and mouse dams to fetuses and pups,
as well as variable PFOA levels across many fetal tissues {Han, 2003, 5081471; Hinderliter
2006, 3749132; Butenhoff, 2004, 1291063; Mylchreest, 2003, 9642031; Fenton, 2009, 194799;
Macon, 2011, 1276151; White, 2011, 1276150; Blake, 2020, 6305864}. Interestingly, Fujii et al.
(2020, 6512379) found that the milk/plasma (M/P) concentration ratio for PFOA also exhibited a
U-shaped curve with increasing chain length but it did not correlate to lipophilicity of PFAS in
FVB/NJcl mice. These findings suggest that the amount transferred from mother to pup during
lactation may also relate to chain length-dependent clearance.
Many recent human studies have quantified the distribution of PFOA from pregnant mothers to
their fetuses and from mothers to their infants. Distribution from pregnant mother to fetus has
been confirmed by measuring PFOA levels in placenta, cord blood, and amniotic fluid during
gestation and at birth. The ratio of PFOA in placenta relative to maternal serum during
pregnancy (Rpm) ranged from 0.326 to 0.460 {Zhang, 2013, 3859792; Chen, 2017, 3859806}.
Gestational age and PFOA branching characteristics influence transport across the placenta.
PFOA concentrations within the placenta increase during gestation from the first to third
trimester {Mamsen, 2019, 5080595}. Linear PFOA is detected at a higher frequency and at
higher concentrations in maternal serum than branched PFOA isomers. However, branched
PFOA is more efficiently transported into the placenta than linear PFOA {Cai, 2020, 6318671;
Chen, 2017, 3859806}.
Several studies reported a strong positive correlation between maternal and cord serum PFOA
levels in humans {Kato, 2014, 2851230; Porpora, 2013, 2150057}. The ratio of PFOA in cord
serum relative to maternal serum ranged from 0.55 to 1.33 (see PFOA Appendix) and generally
increased with gestational age {Li, 2020, 6505874}. Factors such as exposure sources, parity,
and other maternal demographics are postulated to influence variations in maternal serum PFAS
concentrations and cord:maternal serum ratios {Kato, 2014, 2851230; Brochot, 2019, 5381552}.
Cord:maternal serum ratios represent transplacental efficiencies (TTEs), which exhibit a U-
shaped curve with PFAS chain length {Zhang, 2013, 3859792} and generally increase as the
PFAS branching point moves closer to the carboxyl or sulfonate moiety {Zhao, 2017, 5085130}.
Lower levels of PFOA were measured in amniotic fluid compared to the placenta and cord blood
(all collected at delivery) {Zhang, 2013, 3859792}. The mean concentration ratio between
amniotic fluid and maternal blood (collected no more than one hour before delivery) was higher
for PFOA (0.13) than for PFOS (0.0014). The mean concentration ratio between amniotic fluid
and cord blood was higher for PFOA (0.023) than for PFOS (0.0065). Authors attributed the
3-11
-------
DRAFT FOR PUBLIC COMMENT
March 2023
differences in ratios between the two compartments to the solubilities of PFOS and PFOA and
their respective protein binding capacities in the two matrices.
PFOA also distributes widely in human fetal tissues. Mamsen et al. (2017, 3858487) measured
the concentrations of five PFAS in fetuses, placentas, and maternal plasma from a cohort of 39
pregnant women in Denmark. PFOA was detected in placenta and fetal liver, extremities, heart,
intestines, lungs, connective tissues, spinal cord, and ribs, and concentrations were highest in the
placenta and lung. Different patterns of PFOA distribution were observed in fetal tissues
depending on fetal age {Mamsen, 2019, 5080595}. Fetal tissue: maternal serum ratios of PFAS
were calculated by dividing the fetal tissue concentration by the maternal serum concentration. In
general, fetal tissue:maternal serum ratios of PFOA increased from the first trimester to the third
trimester, except for the liver and heart, which showed the highest fetal tissue:maternal serum
ratios in the second trimester compared with the third trimester.
New studies in humans also confirm that the distribution of PFOA from nursing mothers to their
infants via breastmilk correlates with duration of breastfeeding {Mondal, 2014, 2850916; Cariou,
2015, 3859840, Mogensen, 2015, 3859839, Gyllenhammar, 2018, 4778766}. Distribution is
influenced by the chemical properties of PFAS including length, lipophilicity, and branching. In
the Mondal study {Mondal, 2014, 2850916}, the mean maternal serum PFOA concentrations
were lower in breastfeeding mothers vs. non-breastfeeding mothers. Conversely, breastfed
infants had higher mean serum PFOA than infants who were never breastfed. Maternal serum
concentrations decreased with each month of breastfeeding {Mondal, 2014, 2850916; Mogensen,
2015, 3859839}. Cariou et al. (2015, 3859840) reported that PFOA levels in breastmilk were
approximately 30-fold lower relative to maternal serum and the ratio between breastmilk and
maternal serum PFOA was 0.038 ±0.013. The authors noted that the transfer rates of PFAS from
serum to breastmilk were lower compared to other lipophilic persistent organic pollutants such as
poly chlorinated biphenyls.
3.3.1.2.5 Volume of Distribution in Humans and Animals
In humans, the volume of distribution (Vd) for PFOA has been assigned values between 170 and
200 mL/kg (see PFOA Appendix). Vd values may be influenced by differences in distribution
between males and females, between pregnant and non-pregnant females, and across serum,
plasma, and whole blood.
Vd estimates derived in mice and rats vary by species, age, sex, and dosing regimen. For
example, Dzierlenga et al. (2019, 5916078) calculated the apparent volume of central and
peripheral distribution in male and female adult rats after oral dosing. A one-compartment model
for males and a two-compartment model for females was used to characterize PFOA levels.
Peripheral Vd values were dramatically lower than central Vd values at all doses after oral
administration and, interestingly, also after IV administration. While peak tissue levels were
reached readily in both males and females, tissue levels in males were steady over the course of
several days whereas tissue levels in females dropped quickly, in the span of hours. Further
discussion on the Vd for PFOA can be found in Section 6.6.2.
3-12
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.3.1.3 Metabolism
Consistent with other peer-reviewed, published reports and reviews {U.S. EPA, 2016, 3603279;
ATSDR, 2021, 9642134; Pizzurro, 2019, 5387175}, the available evidence demonstrates that
PFOA is not metabolized in humans, primates, or rodents.
3.3.1.4 Excretion
Excretion data are available for oral exposure in humans and laboratory animals. Most studies
have investigated the elimination of PFOA in humans, cynomolgus monkeys, and rats. Fewer
studies measured elimination in mice, hamsters, and rabbits. Available evidence supports urine
as the primary route of excretion in most species, though fecal elimination is prominent in rats.
In rats, hair is another route of elimination in both males and females. In female humans and
animals, elimination pathways include menstruation, pregnancy (cord blood, placenta, amniotic
fluid, and fetal tissues) and lactation (breast milk) (see PFOA Appendix). Results of elimination
half-life determination studies in mammals suggest that PFOA elimination time is longest in
humans (years), intermediate in monkeys (days to weeks), and shortest in rodents (hours to
days).
3.3.1.4.1 Urinary and Fecal Excretion
Studies in laboratory animals provide evidence that urine is typically the primary route of
excretion but that sex impacts excretion by both routes, and these sex differences appear to be
species-specific. Limited evidence supports excretion via the fecal route in laboratory animals
and humans and via hair in animals. Most studies in all species indicate that excretion by the
fecal route is substantially lower than that observed by the urinary route. Excretion through the
fecal route appears to be more prominent in males compared to females and in rodents compared
to humans. Nevertheless, a comprehensive set of principles governing resorption by renal,
hepatic, and enteric routes and how these impact excretion and retention of PFOA has not been
established in either humans or animals.
Human studies examined PFOA excretion after oral exposure, primarily through the urinary
route. The urinary excretion of PFOA in humans is impacted by the isomeric composition of
the mixture present in blood and the sex and age of the individual. The half-lives of the
branched-chain PFOA isomers are shorter than those for the linear molecule, indicating that
renal resorption is less prevalent for the branched-chain isomers {Zhang, 2014, 2851103;
Fu, 2016, 3859819}.
Fujii et al. (2015, 2816710) measured PFOA clearance in mice and humans. Male and
female FVB/NJcl mice were administered PFOA by IV (0.31 |imol/kg) or gavage
(3.13 |iinol/kg) and serum concentration data were analyzed using a two-compartment
model. Mouse urinary clearance was analyzed by dividing the total amount excreted in the
urine during a 24-hour period with the area under the curve (AUC) of the serum
concentration. Human data were analyzed from paired (bile-serum) archived samples from
patients undergoing nasobiliary drainage, percutaneous transhepatic biliary drainage, or
percutaneous transhepatic gallbladder drainage for 24 hours. Urine-serum pairs were
collected from healthy donors. Urinary and biliary clearance was determined by dividing the
cumulative urine or bile excretion in a 24-h period with the serum concentration. Fecal
clearance was calculated using the estimated biliary resorption rate.
3-13
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The authors estimated that the total human clearance for PFOA was 0.096 mL/kg/day;
PFOA clearance rates via urinary, biliary, and fecal routes were estimated to be 0.044, 2.62,
and 0.052 mL/kg/day, respectively. The reabsorption rate of bile excreting PFOA was
estimated to be 0.98 (derived by assigning a Vd of 200 mL/kg, a serum half-life of 3.8 years,
and the presumption that PFOA could only be excreted into the urine and feces via the bile).
The authors also noted that estimated total human clearance was 50-100 times lower than
the estimated PFOA clearances in mice after oral gavage dosing
In rats, urine PFOA concentrations differed with age, dose, and sex {Hinderliter, 2006,
3749132}. For all rats dosed between 3 and 8 weeks of age, urinary excretion of PFOA was
substantially higher in females than in males, and this difference increased with age. Several
additional studies in rats found that females excreted much higher levels in urine compared to
males and compared to feces {Kim, 2016, 3749289; Benskin, 2009, 1617974; Cui, 2010,
2919335}.
3.3.1.4.2 Renal and Enterohepatic Resorption
Several studies have been conducted to elucidate the cause of the sex difference in the
elimination of PFOA by rats {Kudo, 2002, 2990271; Cheng, 2006, 6551310; Hinderliter, 2006,
3749132}. Many of the studies have focused on the role of transporters in the kidney tubules,
especially the OATs and OATPs located in the proximal portion of the descending tubule
{Nakagawa, 2007, 2919370; Nakagawa, 2009, 2919342; Yang, 2009, 2919328; Yang, 2010,
2919288}.
The results of in vitro studies were consistent with an in vivo analysis of OATPs gene and
protein expression in rat kidneys {Yang, 2009, 2919328}. Organic anion transporters
polypeptide lal (OATPlal) is located on the apical side of proximal tubule cells and is a
potential mechanism for renal reabsorption of PFOA in rats. The level of messenger ribonucleic
acid (mRNA) of OATPlal in male rat kidney is 5-20-fold higher than in female rat kidney and
is regulated by sex hormones. Thus, higher expression of OATPlal in male rats would favor
resorption of PFOA in the glomerular filtrate which is consistent with reduced excretion in
males.
Fewer studies have investigated enterohepatic resorption of PFOA. Gastrointestinal elimination
of PFOA was reported in a case report of a single human male with high serum levels of
perfluorinated chemicals who was treated with a bile acid sequestrant (cholestyramine (CSM))
{Genuis, 2010, 2583643}. Before treatment, PFOA was detected in urine (3.72 ng/mL) but not in
stool (LOD = 0.5 ng/g) or sweat samples. After treatment with CSM for 1 week, the serum
PFOA concentration decreased from 5.9 ng/g to 4.1 ng/g, and stool PFOA levels increased to
0.96 ng/g. This observation suggests that PFOA is excreted in bile and that enterohepatic
resorption via intestinal transporters limits the loss of PFOA via feces.
Studies in mice {Maher, 2008, 2919367; Cheng, 2008, 758807} suggest that increased expression
of MRP3 and MRP4, coupled with decreased OATP levels, leads to increased biliary excretion
of bile acids, bilirubin, and potentially toxic exogenous substances, including PFOA. Based on
the greater observed downregulation of OATP-encoding genes in wild-type mice exposed to
PFDA compared to PPARa-null mice exposed to PFDA, the authors concluded that the changes
in receptor proteins were primarily linked to activation of PPARa {Cheng, 2008, 758807}.
3-14
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Zhao et al. (2017, 3856461) demonstrated that PFOA was a substrate for human OATP1B1,
OATP1B3, and OATP2B1 transporters expressed in liver using in vitro studies of Chinese
hamster ovary (CHO) and human embryonic kidney (HEK-293) cells transfected with
transporter complementary DNA (cDNA). Under these conditions, the three OATPs expressed in
human hepatocytes can transport the longer chain PFOA (C8) and perfluorononanoate (C9), but
not the shorter chain perfluoroheptanoate {CI). Preliminary evidence suggests that enterohepatic
resorption could limit elimination of PFOA by the fecal route, including the recent observation
that PFOA binds to NTCP, a transporter that mediates the uptake of conjugated bile acids
{Ruggiero, 2021, 9641806}. The extent to which this pathway operates in vivo and whether
enterohepatic resorption plays a substantial role in the retention of PFOA in humans and animals
is still unknown.
3.3.1.4.3 Maternal Elimination through Lactation and Fetal Partitioning
In humans, PFOA can readily pass from mothers to their fetuses during gestation and through
breast milk during lactation. In conjunction with elimination through menstruation, discussed in
Section 3.3.1.4.4, human females clearly eliminate PFOA through routes not available to males.
The total daily elimination of PFOA in pregnant human females was estimated to be 11.4 ng/day,
lower than the 30.1 ng/day estimated forPFOS {Zhang, 2014, 2850251}. Mamsen et al. (2019,
5080595) estimated a placenta PFOA accumulation rate of 0.11% increase per day during
gestation and observed that the magnitude of elimination may be influenced by the sex of the
fetus. A human study by Zhang et al. (2013, 3859792) observed that the mean levels in the cord
blood, placenta, and amniotic fluid were 58%, 47%, and 1.3%, respectively, of those in the
mother's blood, demonstrating that cord blood, placenta, and amniotic fluid are additional routes
of elimination in pregnant females. Blood loss during childbirth could be another source of
excretion. Underscoring the importance of pregnancy as a life-stage when excretion is altered,
Zhang et al. (2015, 2851103) observed that the partitioning ratio of PFOA concentrations
between urine and whole blood in pregnant women (0.0011) was lower than the ratios found in
non-pregnant women (0.0028). The rate and extent of elimination through these routes are
affected by parity {Lee, 2017, 3983576; Jusko, 2016, 3981718} and may be affected by the
increase in blood volume during pregnancy {Pritchard, 1965, 9641812}.
Women can also eliminate PFOA via lactation {Tao, 2008, 1290895; Thomsen, 2010, 759807;
Kang, 2016, 3859603}. Cariou et al. (2015, 3859840) measured PFOA in maternal serum, cord
serum, and breast milk from females with planned Cesarean births. The observed mean ratio of
cord serum to maternal serum PFOA was 0.78 in this study. However, the mean ratio between
breast milk and maternal serum was 0.038, suggesting transfer from maternal blood to breast
milk is lower than transfer from maternal blood to cord blood.
Studies in laboratory animals support elimination through pregnancy and lactation similar to
what has been observed in humans. Fujii et al. (2020, 6512379) used the M/P concentration ratio
as a measure of chemical transferability in FVB/NJcl mice. Maternal plasma PFOA
concentrations were significantly higher than in milk (M/P ratio was 0.32). The M/P ratios were
similar for C8, C9, C12, and C13, arguing against a direct relationship with lipophilicity.
Potential roles for binding proteins in breast milk or transporters in breast tissue have not been
investigated.
3-15
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In summary, partitioning to the placenta, amniotic fluid, fetus, and breast milk represent
important routes of elimination in humans, and may account for some of the sex differences
observed for blood and urinary levels of PFOA by sex and life stage.
3.3.1.4.4 Other Routes of Elimination
Menstruation may be an important factor in the sex-specific differences observed in PFOA
elimination. Zhang et al. (2013, 3859849) estimated a menstrual serum PFOA clearance rate of
0.029 mL/day/kg. The link between menstruation and PFOA elimination is based on several
observations. First, postmenopausal females and adult males have longer PFOA elimination half-
lives than premenopausal adult females {Zhang, 2013, 3859849}. Challenging the assumption
that this is due to menstruation, Singer et al. (2018, 5079732) failed to find evidence of
associations between menstrual cycle length and PFAS concentrations. Second, several studies
reported on an association between increased serum concentrations of PFOA and PFOS and
early menopause {Knox, 2011, 1402395; Taylor, 2014, 2850915}. However, areanalysis of
these data {Ruark, 2017, 3981395} suggested that the association between increased serum
PFAS and early menopause could be explained by reversed causality, and more specifically, that
pharmacokinetic bias could account for the observed association with epidemiological data.
Ruark et al. (2017, 3981395) thus highlight the importance of considering menstrual blood loss
as a PFAS elimination pathway. Additional studies may be needed to clarify the significance of
the menstruation in PFOA elimination.
One study, Gao et al. (2015, 2851191), found that hair is a potential route of PFAS elimination in
rats. A dose-dependent increase in hair PFOA concentration was observed in all exposed
animals. Interestingly, hair PFOA concentrations for all treatment doses were significantly
higher in males than in females. The sexually dimorphic difference in hair concentrations may be
attributed to the sex differences observed in PFOA elimination rate and the transfer from serum
to hair.
3.3.1.4.5 Half-Life Data
Because there is no evidence that PFOA is metabolized in mammals, half-life determinations are
governed by excretion. There have been several studies of half-lives in humans all supporting a
long residence time for serum PFOA with estimates measured in years rather than months or
weeks (see PFOA Appendix). The calculated PFOA half-lives reported in the literature vary
considerably, which poses challenges in predicting both the routes and rates of excretion. Half-
life estimates vary considerably by species, being most rapid in rodents (measured in hours to
days), followed by primates (measured in days to weeks) and humans (measured in years). Half-
life estimates were shorter in human and rodent females relative to males. In the single primate
study discussed below, half-lives were shorter in males compared to females.
PFOA half-life values in humans ranged from 0.53 years for a branched PFOA in young adult
females {Zhang, 2013, 3859849} to 22 years in a study of primiparous women in Sweden
{Glynn, 2012, 1578498} and varied by geographical region {Gomis, 2017, 3981280} (see PFOA
Appendix). Age, lifestage, and sex differences in PFOA half-lives have not been rigorously
evaluated, though estimates in males are generally longer than those in females {Fu, 2016,
3859819; Gomis, 2017, 3981280; Li, 2017, 4238434} and exhibit an age-related increase in
adults {Genuis, 2014, 2851045, Zhang, 2013, 3859849}. While most studies were conducted in
adults and/or adolescents, one study in newborns {Spliethoff, 2008, 2919368} calculated a half-
3-16
-------
DRAFT FOR PUBLIC COMMENT
March 2023
life for PFOA of 4.4 years. Linear isomers exhibit longer half-lives than branched isomers
{Zhang, 2013, 3859849}.
Half-life estimates in humans rely on measured serum and/or urine concentrations. However,
relatively few studies calculated PFOA half-lives along with measured intake and serum and
urine PFOA concentrations {Xu, 2020, 6781357; Worley, 2017, 3859800; Fu, 2016, 3859819;
Zhang, 2013, 2639569} (see PFOA Appendix). PFOA half-life values among these 4 studies
varied from 1.7 years in Xu et al. (2020, 6781357) to 4.7 years in Fu et al. (2016, 3859819).
These comparisons support principles suggested by the broader literature. First, sex related
differences with males exhibiting somewhat longer half-lives compared to females (especially
females of reproductive age) may relate, at least in part, to menstruation as a route of elimination
{Zhang, 2013, 3859849}. Second, blood and urine concentrations varied by several orders of
magnitude across these four studies. While blood and urine PFOA concentrations varied by two
orders of magnitude across these studies, half-life estimates were similar, ranging from 1.77 to
4.70 years. This variability in serum and urine concentrations may reflect the role of non-urinary
routes of PFOA excretion; the variability in concentrations may also reflect the difficulty in
measuring renal resorption. Finally, only two studies estimated PFOA intake in subjects {Xu,
2020, 6781357; Worley, 2017, 3859800}. Altogether, there is insufficient data to correlate PFOA
intake measurements to serum/plasma and urine concentrations. These factors, as well as age and
health status of subjects, likely contribute to the reported variability in PFOA half-life estimates
in humans.
In experimental animals, half-life values are reported in days rather than in years. Values in
cynomolgus monkeys ranged from 13.6 to 41.7 days {Butenhoff, 2004, 3749227} and were
generally longer than those observed in rodents, but much shorter than values observed in
humans. Depending on the experimental conditions, half-lives in rats ranged from 0.03 days in
females exposed to a high dose of 40 mg/kg {Dzierlenga, 2019, 5916078} to 13.4 days in males
exposed to a relatively low dose of 0.4 mg/kg {Benskin, 2009, 1617974}. Rats exposed by the
IV route exhibited shorter half-lives than rats administered the same dose by the oral gavage
route {Kim, 2016, 3749289; Dzierlenga, 2019, 5916078}. Similar to humans and mice, half-life
estimates were shorter in adult female rats compared to male rats. In contrast, female half-life
values exceeded male values in cynomolgus monkeys, suggesting that species-specific factors
impact elimination across sexes. Similar to findings in humans, PFOA branched isomers
exhibited shorter half-lives compared to linear forms.
33.2 Pharmacokinetic Models
Pharmacokinetic (PK) models are tools for quantifying the relationship between external
measures of exposure and internal measures of dose. For this assessment, PK models were
evaluated for their ability to allow for 1) cross-species PK extrapolation of animal studies of both
cancer and noncancer effects and 2) the estimation of the external dose associated with an
internal dose metric that represents the POD calculated from either animal toxicological or
epidemiological studies. The following sections first describe and evaluate published PK
modeling efforts and then present conclusions from analyses that assessed the utility of the
models to predict internal doses for use in dose-response assessment.
Numerous PK models for PFOA have been developed and published over the years to
characterize the unique ADME described in Section 3.3.1. These approaches can be classified
3-17
-------
DRAFT FOR PUBLIC COMMENT
March 2023
into three categories: classical compartmental models, modified compartmental models, and
PBPK models. With classical compartmental modeling, the body is defined as either a one- or
two-compartment system with volumes and intercompartmental transfer explicitly fit to the
available PFAS PK dataset. Modified compartmental models are more physiologically based in
that they attempt to characterize unique aspects of in vivo ADME through protein binding,
cardiac output, and known renal elimination from the published literature. However, these
models still rely on explicit fitting of data to the non-physiological parameters. Finally, PBPK
models describe the tissues and organs of the body as discrete, physiologically-based
compartments with transport between compartments informed by the available data on
physiologically relevant quantifications of blood flow and tissue perfusion. Determining
additional, non-physiological parameters typically requires explicitly fitting the PBPK model to
time-course concentration data. However, the number of parameters estimated through data
fitting is generally fewer than for classical PK or modified compartmental models. A review of
the available PK models regarding their ability to predict PFOA ADME is provided below.
3.3.2.1 Classical Compartmental Analysis
The most common approach for the prediction of serum levels of PFOA is to apply a relatively
simple one-compartment model. This type of model describes the toxicokinetics of the substance
with a single differential equation that describes the rate of change in the amount or
concentration of the substance over time as a function of the exposure rate and the clearance rate.
This type of model describes the relationship between exposure, serum concentration, and
clearance and can be used to predict one of these values when the other two values are set.
Additionally, because the model can produce predictions of changes in exposure and serum
concentration over time, these models can be applied to fill the temporal gaps around or between
measured serum concentrations or exposures.
The most common use for these models in human populations is to predict serum concentrations
from estimated exposures. Some examples of this include the work by Shin et al. (2011,
2572313) who evaluated the exposure predictions from an environmental fate and transport
model by comparing the predicted serum PFOA concentrations to observed values and by
Avanasi et al. (2016, 3981510) who extended the work of Shin et al. (2011, 5082426) by
applying a population model to investigate how variability and uncertainty in model parameters
affect the prediction of serum concentrations.
Some examples of one-compartment models used to predict human exposure from serum
concentrations include the work of Dassuncao et al. (2018, 4563862) who used a model to
describe historical changes in exposure in seafood and consumer products, Hu et al. (2019,
5381562) who used paired tap water and serum concentration to estimate the proportion of total
exposure that originates from drinking water, and Balk et al. (2019, 5918617) who used
measured concentrations in drinking water, dust and air samples, and serum concentrations in
developing children (measured at several time points) to assess the relative proportion of
exposure that originates from dietary exposure. Zhang et al. (2019, 5080526) performed a similar
study using community tap water measurements and serum concentrations to estimate the
proportion of PFOA exposure in humans that originates from drinking water.
Other applications are used to better understand the toxicokinetics of PFOA in humans by
combining estimated exposure values and serum values to estimate clearance and half-life in a
3-18
-------
DRAFT FOR PUBLIC COMMENT
March 2023
population of interest. One example of this type of model application was presented by Gomis et
al. (2016, 3749264) who used measurements of serum and exposure, in the form of air
concentrations during occupational exposure, to estimate an elimination half-life for PFOA.
Those authors were also able to identify the relative contributions of direct occupational
exposure to PFOA, indirect occupational exposure to PFOA precursors, and background, non-
occupational PFOA exposure. Another example was presented by Worley et al. (2017, 3859800)
who estimated the half-life of PFOA using exposure predicted from drinking water PFAS
concentration in a community with contaminated drinking water. Fu et al. (2016, 3859819) used
paired serum and urine samples from an occupational cohort to estimate the half-life separately
from renal clearance (CLr) (in urine) and in the whole body (in serum). One challenge in the
estimation of half-life is the problem of estimating exposure to PFOA. Russell et al. (2015,
2851185) addressed this problem by estimating the amount of bias in elimination half-life that is
introduced when the ongoing background exposure is not taken into account, with application to
PFOA as an example.
One common modification of the one-compartment model is to perform a "steady-state
approximation" (i.e., to assume that the rate of change of the serum concentration is zero). This
scenario occurs when an individual experiences constant exposure, constant body habitus, and
constant clearance over a timespan of several half-lives. Due to the long half-life of PFOA,
steady state is a reasonable assumption for adults starting from the age of 25 and above.
However, the steady state approximation cannot be applied for ages younger than 21 years of age
(EPA defines childhood as < 21 years of age; {U.S. EPA, 2021, 9641727}) due to ongoing
development during childhood and adolescence. This growth dilutes the concentration of the
chemical in the body and results in lower levels than would be seen in its absence. Even though
pubertal development including skeletal growth typically ends several years prior to the age of
25, there is a period after growth ceases during which PFOA levels increase until the adult
steady-state level is reached. The general acceptability of the steady-state assumption in adults
has the caveat that pregnancy or breastfeeding will result in changes in serum concentration and
will not be accounted for in the steady-state approximation.
When adopting a steady-state assumption, the rate of change in serum levels over time is zero. It
follows that the ratio between exposure to the substance and clearance determines the serum
concentration. This is the approach used in the 2016 PFOA HESD to determine the constant
exposure associated with a serum concentration {U.S. EPA, 2016, 3603279}. A similar approach
was used in the recent risk assessment performed by CalEPA {CalEPA, 2021, 9416932}.
Publications reporting applications of similar models include the work of Zhang et al. (2015,
2851103) who used paired human urine and serum data to estimate the total intake of PFOA and
compared it to the rate of urinary elimination, and Lorber et al. (2015, 2851157) who examined
the effects of regular blood loss due to phlebotomy on PFOA levels and extrapolated that finding
to clearance via menstruation.
In animals, three classical PK models for PFOA have been published since the 2016 HESD for
PFOA. In Dzierlenga et al. (2020, 5916078), male Sprague-Dawley rats were dosed with PFOA
via oral gavage at 6, 12, and 48 mg/kg, or intravenously at 6 mg/kg. Female Sprague-Dawley rats
were dosed with PFOA via oral gavage at 40, 80, 320 mg/kg or intravenously at 40 mg/kg.
Following the administration of PFOA, rats were sacrificed from five minutes to 50 days post-
dosing for males and from five minutes to 12 days post-dosing in females. Differences in length
3-19
-------
DRAFT FOR PUBLIC COMMENT
March 2023
of study for each sex represent the sex-dependent difference in half-lives for which adult female
rats eliminate PFOA more rapidly than adult males. For both sexes, measured plasma
concentrations characterized the biphasic PK curve. From these exposure scenarios, Dzierlenga
et al. (2020, 5916078) developed a two-compartment model to characterize PK parameters of
interest such as the alpha- and beta-phase half-life, central and peripheral compartment volumes,
and total PFOA clearance. For each dosing scenario, a single set of PK parameters were fit,
making extrapolation to other dosing scenarios difficult. However, the authors demonstrate a
significant difference between males and females in beta-phase half-life and overall clearance.
This difference in half-life is critical when considering internal dosimetry for a pregnant dam
during developmental PK studies.
Fujii et al. (2015, 2816710) conducted a PK analysis in mice by dosing male and female mice
either intravenously with 0.313 [j,mol/kg or through oval gavage with 3.13 [j,mol/kg. Following
administration of PFOA, blood concentrations were collected through tail veins beginning
immediately following dosing up to 24 hours post-dosing. Fujii et al. (2015, 2816710) used these
data to develop a two-compartment model to describe sex-dependent PK in mice. Unfortunately,
the follow-up time of 24 hours post-dosing is not long enough to accurately characterize the
beta-phase elimination of PFOA, which the authors predicted was 627 days. The small amount of
change in PFOA levels within a 24-hour timespan will make the estimated terminal half-life
from a two-compartment model unreliable because PFOA will still be in the distribution phase.
In addition, the functional form fit for the oral gavage data in Fujii et al. (2015, 2816710) reflects
a one-compartment model with gavage dosing making it not possible to compare the predicted
half-lives between the two routes of exposure. While the reported data could be used for
characterizing absorption and distribution of PFOA, it cannot be used for characterizing the
elimination phase. Additionally, a study with a much longer follow-up time of 80 days post-
dosing reported a half-life of 15.6 - 21.7 days {Lou, 2009, 2919359}.
Finally, Gomis et al. (2016, 3749264) utilized the functional form of a two-compartment model
with oral gavage to predict internal dosimetry of PFOA in rats using PK data from Perkins et al.
(2004, 1291118). However, because the scope of the Gomis et al. (2017, 3981280) study
involved predicting internal dose points-of-departure, PK parameters are not presented.
3.3.2.2 Modified Compartmental Models
In addition to the common one-compartment models described above, several models for
humans have been developed to extend the simple one-compartment model to describe the PK
during pregnancy and lactation. The key factors that must be introduced into the model are the
changes in body habitus that occur during pregnancy (e.g., increases in blood plasma volume and
body weight), the distribution and transfer of the substance between the maternal and fetal
tissues, the transfer from the mother to the infant during nursing, and postnatal development,
including growth of the infant during the early period of life. The mathematical formulation of
this type of model requires two differential equations, one describing the rate of change in
amount or concentration in the mother and one describing the rate of change in infants. One such
developmental model with a lactational component was used to predict the maternal serum
concentrations and exposure from measurements of PFOA concentrations in breast milk
{Abdallah, 2020, 6316215}. Verner et al. (2016, 3299692) presented another developmental
model to predict PFOA serum concentrations in the mother and child and predict previous
exposure using mother/child paired serum measurements at different times. This model included
3-20
-------
DRAFT FOR PUBLIC COMMENT
March 2023
all the key aspects previously mentioned for developmental PK models. Another developmental
model was developed by Goeden et al. (2019, 5080506) to evaluate the relationship between
drinking water concentrations and infant serum levels during breastfeeding resulting from
gestational and lactational transfer of PFOA that had accumulated within the mother. A
distinguishing feature of the Goeden et al. (2019, 5080506) model is that it incorporates an
adjustment for the increased intracellular water in infants and young children compared to adults,
under the assumption that PFAS distribution into tissues, quantified by the Vd, will increase
proportionally to intracellular water content. This life stage difference in intracellular water
content may explain why the ratio of PFOA in cord blood vs. maternal blood at childbirth tends
to be less than one. Monroy et al. (2008, 2349575) reported median cord blood PFOA
concentration to be 87% of maternal serum, while the median ratio of fetal tissue to placenta
PFOA concentration was found to be generally greater than one {Mamsen, 2019, 5080595}. One
oversight of this model is that the rate equations take concentration into account, but they do not
account for decreases in concentration due to increasing body weight (growth dilution). This is a
significant factor for infants who grow quickly.
Other unique analyses that extended the one-compartment framework were publications by Shan
et al. (2016, 3360127), who estimated the exposure to specific isomers of PFOA using
measurements in food, tap water, and dust to estimate the isomeric profiles of the substances in
human serum, and Convertino et al. (2018, 5080342) who used a two-compartment PK-
pharmacodynamic model to describe changes in serum concentration during a dose-escalation,
phase one clinical trial with PFOA and describe how those serum changes are correlated with
changes in serum total cholesterol (TC) and free thyroxine (FT4).
Some other models have added features to accommodate longer half-life values and allow for
dose-dependent changes in excretion rate compared to the classic 1- or 2- compartment
approaches {Andersen, 2006, 818501; Wambaugh, 2013, 2850932; Loccisano, 2011, 787186;
Loccisano, 2012, 1289830; Loccisano, 2012, 1289833; Loccisano, 2013, 1326665}. The
underlying assumption for all the models is saturable resorption from the kidney filtrate, which
consistently returns a portion of the excreted dose to the systemic circulation and prolongs both
clearance from the body (e.g., extends half-life) and the time needed to reach steady state. These
more complex models have been developed for humans, monkeys, mice, and rats.
One of the earliest PK models {Andersen, 2006, 818501} was created using the post-dosing
plasma data from the Butenhoff et al. (2004, 3749227) study in cynomolgus monkeys. In this
study, groups of six monkeys (three per sex per group) were dosed for 26 weeks with 0, 3, 10, or
20 mg/kg PFOA (and also a high dose of 30 mg/kg PFOA for only the first 12 days) and
followed for more than 160 days after dosing. Metabolism cages were used for overnight urine
collection. Since urine specimens could only account for overnight PFOA excretion, total
volume and total PFOA were extrapolated to 24-hour values based on the excretion rate (volume
per hour) for the volume collected and the hours of collection.
The Andersen et al. (2006, 818501) model was based on the hypothesis that saturable resorption
capacity in the kidney would possibly account for the unique half-life properties of PFOA across
species and sexes. The model structure was derived from a published model for glucose
resorption from the glomerular filtrate via transporters on the apical surface of renal tubule
epithelial cells {Andersen, 2006, 818501}.
3-21
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The renal-resorption model includes a central compartment that receives the chemical from the
oral dose and a filtrate compartment for the glomerular filtrate from which resorption with
transfer to the central compartment can occur. Transfer from the filtrate compartment to the
central compartment decreases the rate of excretion. The resorption in the model was saturable,
meaning that there was proportionally less resorption and greater excretion at high serum PFOA
concentrations than at low concentrations. In addition to decreased renal excretion due to the
renal resorption, excretion is also reduced in the model by implementing a constant proportion of
PFOA that is bound to protein in plasma and is not available for renal filtration.
The model was parameterized using the body weight and urine output of cynomolgus monkeys
{Butenhoff, 2004, 3749227} and a cardiac output of 15 L/h-kg from the literature {Corley, 1990,
10123}. A 20% blood flow rate to the kidney was assumed based on data from humans and dogs.
Other parameters were optimized to fit the data for plasma and urine at lower concentrations and
then applied for the 20 mg/kg/day dose, which was assumed to represent a concentration at
which renal resorption was saturated. Based on the data for the dose of 20 mg/kg/day, the model
was able to predict the decline in plasma levels after the cessation of dosing. The predictions
were adequate for one of the three modeled monkeys; for the other two monkeys, the model
predicted higher serum concentrations than were observed. That discrepancy between model
prediction and observations could have occurred because the model did not allow for efflux of
PFOA into the glomerular filtrate through transporters on the basolateral surface of the tubular
cells. The authors also observed that three of the monkeys had faster CLR of PFOA than the other
three monkeys, indicating interindividual variability in clearance.
Second
Compartment
(V«, Ctissue)
Oral Dose
k.
Central
r
k 9U" Compartment
(Vc Ccentral' ^plasma)
IV Dose
Filtrate q((
Compartment in
(V„|, Cfli)
Figure 3-4. Schematic for a Physiologically Motivated Renal Resorption PK Model for
PFOA
Adapted from Wambaugh et al. (2013,2850932).
Building on the work of other researchers, Wambaugh et al. (2013, 2850932) developed and
published a PK model to support the development of an EPA RfD for PFOA {U.S. EPA, 2016,
3603279}. The model was applied to data from studies conducted in monkeys, rats, or mice that
demonstrated an assortment of systemic, developmental, reproductive, and immunological
effects. A saturable renal resorption term was used. This concept has played a fundamental role
3-22
-------
DRAFT FOR PUBLIC COMMENT
March 2023
in the design of all of the published PFOA models summarized in this section. The model
structure is depicted in Figure 3-4 (adapted from Wambaugh et al. (2013, 2850392)).
Wambaugh et al. (2013, 2850932) placed bounds on the estimated values for some parameters of
the Andersen et al. (2006, 818501) model to support the assumption that serum carries a
significant portion of the total PFOA body load. The Andersen et al. (2006, 818501) model is a
modified two-compartment model in which a primary compartment describes the serum and a
secondary deep tissue compartment acts as a specified tissue reservoir. Wambaugh et al. (2013,
2850932) constrained the total Vd such that the amount in the tissue compartment was not greater
than 100 times that in the serum. As a result, the ratio of the two volumes (serum vs. total) was
estimated in place of establishing a rate of transfer from the tissue to serum, but the rate of
transfer from serum to tissue was also estimated from the data. A nonhierarchical model for
parameter values was also assumed. Under this assumption, a single numeric value represents all
individuals of the same species, sex, and strain. This sex assumption was applied to male and
female rats to determine sex-specific parameters because of established sex-specific
toxicokinetic differences. Conversely, monkeys and mice were only grouped by species and
strain with only female parameters available for mice and male/female monkey data pooled
together for a single set of parameters. Body weight, the number of doses, and magnitude of the
doses were the only parameters varied for different studies. Measurement errors were assumed to
be log-normally distributed. Table 4-3 in Section 4.1.3.1.1 provides the estimated and assumed
PK parameters applied in the Wambaugh et al. (2013, 2850932) model for each of the species
evaluated.
The PK data that supported the Wambaugh et al. (2013, 2850932) analysis were derived from
two in vivo PFOA PK studies. The monkey PK data were derived from Butenhoff et al. (2004,
3749227), and the data for the rats (M/F) were from Kemper et al. (2003, 6302380). Two strains
of female mice were analyzed separately, with CD1 information derived from Lou et al. (2009,
2919359) and C57BL/6 information derived from DeWitt et al. (2008, 1290826). The data were
analyzed within a Bayesian framework using Markov Chain Monte Carlo sampler implemented
as an R package developed by EPA to allow predictions across species, strains, and sexes and to
identify serum levels associated with the NOAEL and LOAEL external doses. Prior distributions
for the parameters were chosen to be broad, log-normal distributions, allowing the fitted
parameters to be positive and for the posterior distribution to be primarily informed by the data
likelihood rather than by the priors.
3.3.2.3 PBPK Models
An alternative approach to the use of a classical or modified compartmental model is a PBPK
model, which describes the changes in substance amount or concentration in a number of
discrete tissues. One of the main advantages of a PBPK model is the ability to define many
parameters based on physiological data, rather than having to estimate them from chemical-
specific data. Such physiological parameters include, for example, organ volumes and the blood
flow to different organs; they can be measured relatively easily and are chemical independent.
Another advantage is that amount and concentration of the substance can be predicted in specific
tissues, in addition to blood. This can be valuable for certain endpoints where it is expected that a
tissue concentration would better reflect the relevant dosimetry compared to blood concentration.
3-23
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The first PBPK model developed for this chemical was reported in a series of publications by
Loccisano et al. which together describe the PK of PFOA in rats, monkeys, and humans, in both
adult and developmental (for rat and human) scenarios {Loccisano, 2011, 787186; Loccisano,
2012, 1289830; Loccisano, 2012, 1289833; Loccisano, 2013, 1326665}. These models were
developed based on an earlier "biologically motivated" model that served as a bridge between a
one-compartment model and PBPK by implementing a tissue compartment (similar to a 2-
compartment model), an absorption compartment, and a renal filtrate compartment with
saturable renal resorption {Tan, 2008, 2919374}. The work of Tan et al. (2008, 2919374) was a
development of the earlier work of Andersen et al. (2006, 818501) previously discussed. The
PBPK model of Loccisano and colleagues then extended this "biologically motivated" model by
the addition of discrete tissue compartments, rather than a single compartment representing all
tissues.
A series of follow-up studies applied the Loccisano and coauthors' model structure, with
extensions, to address how PK variation in human populations could bias the result of the study.
This consisted of the work of Wu et al. (2015, 3223290) who developed a detailed model of
adolescent female development during puberty and menstrual clearance of PFOA to investigate
the interaction between chemical levels and the timing of menarche, Ruark et al. (2017,
3981395) who added a detailed description of menopause to evaluate how that affects serum
levels and the epidemiological association between early menopause and PFOA levels, Ngueta et
al. (2017, 3860773) who implemented a reduction in menstrual clearance in individuals using
oral contraceptives and the interaction between oral contraceptive use, endometriosis, and serum
PFOA levels, and Dzierlenga et al. (2020, 6315786; 2020, 6833691) who applied a model of
thyroid disease {Dzierlenga, 2019, 7947729} to describe changes in PFOA and PFOS urinary
clearance due to disease state.
In addition to this set of studies, Fabrega et al. (2014, 2850904) updated the model of Loccisano
et al. (2013, 1326665) for humans by modeling a human population using regional food and
drinking water measurements and human tissue data collected from cadavers in a region of
Spain. The use of human tissue data is relatively rare due to the challenges in sourcing human
tissue but may prove preferable to the assumption that human distribution is similar to
distribution in an animal model. However, Fabrega et al. (2014, 2850904) estimated their tissue
to blood partition coefficients from the ratio of tissue concentrations in the cadavers to the
average serum concentrations in live volunteers who lived in the same region but were sampled
several years earlier {Ericson, 2007, 3858652} and they provided no details on how their renal
resorption parameters were estimated from the human blood concentrations. This model was
further applied to a population in Norway and extended to other PFAS {Fabrega, 2015,
3223669}.
Brochot et al. (2019, 5381552) presented the application of a PBPK model for PFOA with
gestation and lactation life stages to describe development and predicted maternal, infant, and
breastmilk concentrations over a variety of scenarios including the prediction of maternal levels
across multiple pregnancies.
One of the major challenges in the parameterization of PBPK models for PFOA is the estimation
of the chemical-dependent parameters such as those involved in protein binding and renal
clearance. One way to investigate this issue is to perform in vitro experiments to help inform the
parameters. Worley et al. (2015, 3981311) used in vitro measurements of renal transporter
3-24
-------
DRAFT FOR PUBLIC COMMENT
March 2023
activity to describe in detail the various steps involved in the renal filtration, resorption, and
excretion of PFOA. Cheng et al. (2017, 3981304) went farther in their use of in vitro data and
used measurements of PFOA interactions with binding proteins, as well the measured rates of
several transporters, to parameterize a rat PBPK model.
No new animal PBPK models for PFOA have been published since the 2016 PFOA HESD {U.S.
EPA, 2016, 3603279}. See the 2016 HESD {U.S. EPA, 2016, 3603279} for a more in-depth
review of PFOA PBPK models.
3.4 Non-Cancer Health Effects Evidence Synthesis and
Integration
3.4.1 Hepatic
EPA identified 32 epidemiological studies (reported in 38 publications)7'8 and 28 animal
toxicological studies that investigated the association between PFOA and hepatic effects. Of the
epidemiological studies, 21 were classified as medium confidence, 8 as low confidence, 1 as
mixed {medium/low) confidence, and 8 were considered uninformative (Section 3.4.1.1). Of the
28 animal toxicological studies, 5 were classified as high confidence, 19 as medium confidence,
2 as low confidence, and 2 were considered mixed (medium/uninformative and
medium/low /uninformative) (Section 3.4.1.2). Studies have mixed confidence ratings if different
endpoints evaluated within the study were assigned different confidence ratings. Though low
confidence epidemiology and animal toxicological studies are considered qualitatively in this
section (e.g., to inform the weight of the evidence for hazard assessment), they were not
considered quantitatively for the dose-response assessment (Section 4).
3.4.1.1 Human Evidence Study Quality Evaluation and Synthesis
3.4.1.1.1 Introduction and Summary of Evidence from the 2016 PFOA HESD
Serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are
considered reliable markers of hepatocellular function/injury, with ALT considered more
specific and sensitive {Boone, 2005, 782862}. Bilirubin and y-glutamyltransferase (GGT) are
also routinely used to evaluate potential hepatobiliary toxicity {Boone, 2005, 782862; EMEA,
2008, 3056793; Hall, 2012, 2718645}. Elevated liver serum biomarkers are frequently an
indication of liver injury, though not as specific as structural or functional analyses such as
histology findings and liver disease.
There are 12 epidemiological studies (13 publications)8 from the 2016 PFOA HESD {U.S. EPA,
2016, 3603279} that investigated the association between PFOA exposure and hepatic effects,
and study quality evaluations are shown in Figure 3-5.Error! Reference source not found.
Emmett et al. {2006, 1290905} was rated as uninformative and will not be further discussed.
Nine out of the twelve remaining studies were rated as medium quality and all investigated
changes in serum liver enzymes.
7 Multiple publications of the same data: Jain and Ducatman (2019, 5381566); Jain and Ducatman (2019, 5080621); Jain (2019,
5381541); Jain (2020, 6833623); Omoike et al. (2020, 6988477); Liu et al. (2018, 4238514); Gleason et al. (2015,2966740) all
used NHANES data from overlapping years.
8 Olsen (2003, 1290020) is the peer-review paper of Olsen (2001,10228462); however, data for PFOA and hepatic outcomes is
reported in Olsen (2001,10228462).
3-25
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Costa et al., 2009, 1429922-
Emmettetal., 2006, 1290905-
Gallo et al.. 2012. 1276142-
+
+
+
•
•
+
-
L.
Legend
9 Good (metric) or High confidence (overall)
+ Adequate (metric) or Medium confidence (overall)
I - I Deficient (metric) or Low confidence (overall)
9 Critically deficient (metric) or Uninformativ© (overall)
4c Multiple judgments exist
¦
-
+
D -
¦
+
B
+
+
++
+
+
+
+
+
Lin et al., 2010, 1291111 -
+
+
+
+
+
Olsen and Zobel. 2007, 1290836-
Olsen et al., 2000, 1424954-
Olsen et al., 2001, 10228462 -
Olsen et al., 2003, 1290020-
Sakret al., 2007, 1291103-
Sakr et al., 2007, 1430761 -
St©enland and Woski©. 2012. 2919168-
St©enland ©t al.. 2015. 2851015-
Yamaguchi etal., 2013, 2850970-
+
+
+
¦
+
-
+
-
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
¦
+
+
¦
+
+
+
+
f*
-
+
¦
+
-
+
¦
+
+¦
++
+
-
-
-
++
-
-
-
+
-
Figure 3-5. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects Published Before 2016 (References in the 2016 HESD)
Interactive figure and additional study details available on HAWC.
Lin et al. (2010, 1291111) is a medium confidence study that examined 2,216 adults in the
NHANES study (1999-2000, and 2003-2004) and observed that higher serum concentrations of
PFOA were associated with abnormal liver enzymes increases in the U.S. general population.
For each increase in log-PFOA, the serum ALT and GGT concentrations (U/L) increased by
1.86 units (95% CI: 1.24, 2.48), and 0.08 units (95% CI: 0.05, 0.11), respectively {Lin, 2010,
1291111}. Importantly, when PFOS, PFHxS, and PFNA were simultaneously added in the fully
adjusted regression models, the associations remained and were slightly larger; one unit increase
in serum log-PFOA concentration was associated with a 2.19 unit (95% CI: 1.4, 2.98) increase in
serum ALT concentration (U/L), and a 0.15 unit (95% CI: 0.11, 0.19) increase in serum log-GGT
concentration (U/L). Another medium confidence cross-sectional study {Yamaguchi, 2013,
2850970} conducted in Japan reported a positive correlation between PFOA and ALT.
A medium confidence study in a highly exposed community provides further support for the
positive association between PFOA exposure and ALT findings in the U.S. general population.
One of the largest studies of PFOA exposure and ALT in adults, Gallo et al. (2012, 1276142),
evaluated 47,092 adults from the C8 Health Project living in communities in Ohio and West
Virginia impacted by a manufacturing-related PFOA-contaminated drinking water supply.
Natural log transformed serum PFOA concentrations were associated with ln-ALT in linear
3-26
-------
DRAFT FOR PUBLIC COMMENT
March 2023
regression models (regression coefficient: 0.022; 95% CI: 0.018, 0.025) and with elevated ALT
in logistic regression models across deciles of PFOA (OR = 1.10; 95% CI: 1.07, 1.13). The
evidence of an association between PFOA and GGT or bilirubin was less consistent. The level of
bilirubin increased with increasing PFOA at low PFOA concentrations and decreased with
increasing PFOA levels at higher PFOA concentrations, producing an inverse roughly U-shaped
curve of the relationship between PFOA and bilirubin.
Several medium confidence cross-sectional occupational studies reported that higher
concentrations of PFOA were associated with higher liver enzyme levels, such as ALT, AST,
GGT, and total bilirubin {Sakr, 2007, 1291103; Sakr, 2007, 1430761; Costa, 2009, 1429922}.
However, other medium confidence cross-sectional occupational studies in PFOA production
workers reported mostly null findings, with some positive associations with ALT in specific
locations or specific years {Olsen, 2000, 1424954; Olsen, 2001, 10228462; Olsen, 2003,
1290020; Olsen, 2007, 1290836).
Confidence
Rating
Reference
Exposure
Matrix
Study Design Exposure Levels
Sub-population
Comparison
EE
0
1
Effect Estimate
2
3
4
Medium
confidence
Gallo et
aL, 2012
Serum
Cross -
sectional
Median=28.0 ng/mL (25th -
75th percentile: 13.5-70.8
ng/mL)
-
Regression coefficient
(per 1-ln ng/mL
increase in PFOA)
0.02
1
1
1
1
Gleason et
al.t 2015
Serum
Cross -
sectional
Median=3.7 ng/mL (25th-75th
percentile: 2.5 - 5.2 ng/mL)
-
Regression coefficient
(per 1-ln ngfmL
increase in PFOA)
0.04
l
1
»
1
i
Lin et a!.,
2010
Serum
Cross -
sectional
Mean: 4.06 ng/mL (standard
error 1.04 ng/mL)
Women
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
1.87
I
1
1
1
1
•
Mean: 4.22 ng/mL (standard
error 1.04 ng/mL)
Ages >=60
years
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
1.93
t
1
1
1
1
•
Mean: 4.48 ng/mL (standard
error 1.03 ng/mL)
Ages 18-39
years
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
1.02
1
1
1
1
1
•
Mean: 4.71 ng/mL (standard
error 1.04 ng/mL)
Ages 40-59
years
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
1.83
T
1
1
1
1
•
Mean: 5.05 ng/mL (standard
error. 1.03 ng/mL)
Men
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
1.55
"I
1
1
1
1
•
Median: 4.20 ng/mL
(25th-75th percentile:
2.90-5.95 ng/mL)
-
Regression coefficient
(per 1-log ng/mL
increase in PFOA)
4.05
t
1
1
1
1
•
Olsen et
at., 2000
Serum
Cross -
sectional
1993 Median (min-max): 1.1
ppm (0.0-80.0 ppm): 1995:
1.2 ppm ( 0.0-114.1 ppm);
1997: 1.3 ppm (0.1-81.3 ppm)
-
Regression coefficient
(per 1 ppm increase
in serum PFOA)
4.47
1
1
1
1
1
•
Sakr et aL
Serum
Cohort
Mean: 1.13 ppm (standard
Regression coefficient
(per ppm increase in
PFOA)
0.54
w
1
2007a
deviation: 2.1 ppm)
1
1
Sakretal..
2007b
Serum
Cross -
sectional
Mean (SD): 0.428 ppm
(0.86): Min-max: 0.005-9.550
ppm
-
Regression coefficient
(per ppm increase
PFOA)
0.02
T
1
1
1
Workers not on
lipid-lowering
medications
Regression coefficient
(per ppm increase
PFOA)
0.03
l
1
fc
1
1
0
1
2
3
4
Figure 3-6. Overall ALT Levels from Pre-2016 HESD Epidemiology Studies Following
Exposure to PFOA
Interactive figure and additional study details available on Tableau.
The associations with ALT indicate the potential for PFOA to affect liver function; however,
studies of functional hepatic endpoints were limited to two studies in an occupational cohort. The
first study was a low confidence study that observed no association between PFOA and hepatitis
3-27
-------
DRAFT FOR PUBLIC COMMENT
March 2023
or fatty liver disease; however, there was a positive association with non-hepatitis liver disease
with a 10-year lag time {Steenland, 2015, 2851015}. A medium confidence cohort mortality
study of workers exposed to PFOA at a DuPont chemical plant in West Virginia observed no
association between PFOA exposure levels and non-malignant chronic liver disease deaths
{Steenland and Woskie, 2012, 2919168}.
In conclusion, the majority of the medium confidence studies support an association between
PFOA exposure and increases in serum ALT in multiple populations, including occupational and
highly exposed communities as well as the general population (see Figure 3-6). Multiple studies
demonstrated statistically significant increases in ALT {Gallo, 2012, 1276142; Lin, 2010,
1291111; Yamaguchi, 2013, 2850970; Olsen, 2000, 1424954 for 1997 data) or elevated ALT
{Gallo, 2012, 1276142} after PFOA exposure. Increases were also observed for AST and GGT,
though less consistently across the available studies.
3.4.1.1.2 Study Quality Evaluation Results for the Relevant Epidemiology
Studies Identified from the Updated Literature Review
There are 20 epidemiological studies (25 publications)9 that were identified from recent
systematic literature search and review efforts conducted after publication of the 2016 PFOA
HESD {U.S. EPA, 2016, 3603279} that investigated the association between PFOA and hepatic
effects. Study quality evaluations for these 25 publications are shown in Figure 3-7 and Figure
3-8. Of these 25 publications, 12 were classified as medium confidence, 6 as low confidence, and
7 were considered uninformative.
The following informative studies examined liver enzymes in adults: two cross-sectional studies
{Wang, 2012, 2919184; Nian, 2019, 5080307}; multiple publications of data from NHANES
{Jain, 2019, 5381541; Liu, 2018, 4238514; Omoike, 2020, 6988477; Jain, 2019, 5080621; Jain,
2019, 5381566; Gleason, 2015; 2966740}; one cohort with retrospective exposure assessment
{Darrow, 2016, 3749173}; one prospective cohort {Salihovic, 2018, 5083555}; one open-label
controlled trial {Convertino, 2018, 5080342}; and one occupational cohort {Olsen, 2012,
2919185}. Most of these studies were in general population adults, but some assessed specific
populations such as the elderly {Salihovic, 2018, 5083555} and fluorochemical plant workers
{Wang, 2012, 2919184; Olsen, 2012, 2919185}. In addition, one occupational cohort {Girardi,
2019, 6315730} and three cross-sectional studies {Darrow, 2016, 3749173; Rantakokko, 2015,
3351439; Liu, 2018, 4238396} examined functional liver endpoints in adults (histology, liver
disease, hepatic fat mass). In children and adolescents, four studies were available, including one
cohort {Mora, 2018, 4239224} and three cross-sectional studies {Khalil, 2018, 4238547; Jin,
2020, 6315720; Attanasio, 2019, 5412069}, with one examining histology endpoints {Jin, 2020,
6315720}.
All of the studies of adults and children in the general population, except for Darrow et al. (2016,
3749173), and one of the two occupational cohorts {Olsen, 2012, 2919185} measured exposure
to PFOA using biomarkers in blood. Darrow et al. (2016, 3749173) modeled exposure based on
residential history, drinking water sources, and water consumption rates. The other occupational
cohort study estimated PFOA exposure based on job duties {Girardi, 2019, 6315730}. The
9 Multiple publications of the same data: Jain and Ducatman (2019, 5381566); Jain and Ducatman (2019, 5080621); Jain (2019,
5381541); Jain (2020, 6833623); Omoike et al. (2020, 6988477); Liu et al. (2018, 4238514); and Gleason et al. (2015,2966740)
all used NHANES data from overlapping years.
3-28
-------
DRAFT FOR PUBLIC COMMENT
March 2023
uninformative studies were excluded due to potential confounding {Jiang, 2014, 2850910;
Predieri, 2015, 3889874; Abraham, 2020, 6506041; Sinisalu, 2021, 7211554}, lack of
information on participant selection {Sinisalu, 2020, 9959547}, or use of PFAS as the dependent
variable (in a publication with a more suitable analysis {Jain, 2020, 6833623}, or in cases where
the independent variable is a genetic variant and thus not affected by PFAS exposure {Fan, 2014,
2967086}).
High and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies (e.g., AST serum levels and functional assays), the evidence synthesis below included
details on any low confidence studies available. Studies considered uninformative were not
considered further in the evidence synthesis.
3-29
-------
DRAFT FOR PUBLIC COMMENT
March 2023
9^
4is^ 0<$^0
>°^0^00^VS6Ve^i^ C
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Abraham et al., 2020, 6506041 -
I
+
l
+
l
+
D
i
I
+
i
+
~
Attanasio, 2019, 5412069-
+
+
+
+
+
+
+
Convertino et al., 2018, 5080342 -
+
++
+
-
+
+
+
-
Darrow et al., 2016, 3749173-
+
+
+
++ ++
+
+
+
Fan et al., 2014, 2967086 -
+
+
-
+
+
+
+
-
Girardi et al., 2019, 6315730 -
-
+
-
-
¦
+
-
-
Gleason et al., 2015, 2966740 -
+
+
++
+
+
+
+
+
Jain and Ducatman, 2019, 5381566 -
+
+
-
+
-
+
+
-
Jain et al., 2019, 5080621 -
++
++
++
+
+
+
+
+
Jain, 2020, 6833623 -
+
+
+
+
+
+
+
~
Jiang et al., 2014, 2850910-
-
++
+
Bl-
+
-
~
Jinet al., 2020, 6315720-
+
+
++
-
+
+
-
~
Figure 3-7. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects3
Interactive figure and additional study details available on HAWC.
a Multiple publications of the same data: Jain and Ducatman (2019, 5381566); Jain and Ducatman (2019, 5080621); Jain (2019,
5381541); Jain (2020, 6833623); Omoike et al. (2020, 6988477); Liu et al. (2018, 4238514); Gleason etal. (2015,2966740) all
use NHANES data from overlapping years.
3-30
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Ne^0°
8 °°^
*0®
Khalilet al., 2018, 4238547-
-
+
+
-
+
+
"
-
Liu et al., 2018, 4238396-
-
+
++
+
+
+
+
+
Liu et al., 2018, 4238514-
+
+
+
+
+
+
+
+
Moraet al., 2018, 4239224-
+
+
++
+
++
+
+
+
Nian et al., 2019, 5080307-
+
++
+
+
++
+
+
+
Olsen et al., 2012, 2919185-
+
+
+
-
+
+
-
Omoike et al., 2020, 6988477 -
++
++
+
+
+
+
+
+
Predieri et al., 2015, 3889874 -
+
+
-
¦
+
+
- ¦
Rantakokko et al., 2015, 3351439 -
+
+
+
+
+
+
"
+
Salihovic et al., 2018, 5083555 -
+
+
++
+
+
+
+
+
Sinisalu et al., 2020, 7211554-
+
+
+
B
-
+
- B
Sinisalu et al., 2021, 9959547 -
--
+
+
-
+
- H
Wang et al., 2012, 2919184-
-
+
+
-
+
+
-
~
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Figure 3-8. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects (Continued)3
Interactive figure and additional study details available on HAWC.
a Multiple publications of the same data: Jain andDucatman (2019, 5381566); Jain andDucatman (2019, 5080621); Jain (2019,
5381541); Jain (2020, 6833623); Omoike et al. (2020,6988477); Liu et al (2018, 4238514); Gleason et al. (2015, 2966740) all
use NHANES data from overlapping years.
3-31
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.1.1.3 Synthesis of Hepatic Injury from the Updated Literature Review
Results for the studies that examined ALT are presented in the Appendix (see PFOA Appendix).
As shown in Figure 3-9 and Figure 3-10, of the available informative studies that measured ALT
in adults, statistically significant positive associations between ALT and PFOA (i.e., increased
ALT as a continuous measure with higher PFOA exposure levels) were observed in all of the
medium confidence studies, which consisted of one cross-sectional study {Nian, 2019,
5080307}, two cohort studies {Darrow, 2016, 3749173; Salihovic, 2018, 5083555}, and two
NHANES publications {Gleason, 2015, 2966740; Jain, 2019, 5381541}.
In addition, an exposure-response gradient was observed in the single study that examined
quintiles of exposure {Darrow, 2016, 3749173}. This study additionally examined elevated ALT
as a dichotomous outcome and reported an OR of 1.16 (95% CI 1.02, 1.33) in the highest vs.
lowest quintiles of exposure (Figure 3-9). The positive associations in Jain (2019, 5381541)
were observed only in certain sub-groups (e.g., by renal function (i.e., glomerular filtration
stage), obesity status) and according to no clear pattern across sub-groups (NHANES 2003-
2014), but in Gleason et al. (2015, 2966740), the positive association was observed in the entire
study population (NHANES 2007-2010). Results of the low confidence studies of ALT in adults
are further described in the PFOA Appendix and not described further in this section because
there are numerous medium confidence studies describing ALT measures in adults that were
included in the 2016 PFOA HESD (see Section 1.13.1.1.1) or identified in the updated literature
search.
3-32
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Confidence Literature Exposure Study Exposure
Rating Search Tag Reference Matrix Design Levels Sub-population Comparison
Medium Pre-2016 Galloetal.. Serum Cross- Median=28.0 ¦
confidence Literature 2012 sectional ng/mL (25th-
Search 75th
percentile:
13.5-70.8
ng/mL)
GSeason et
al., 2015
Updated
Literature
Search
i Cross - Median=3.7
sectional ng/mL
(25th-75th
percentile:
2.5 - 5.2
ng/mL)
Darrowetal., Modeled Cohort and 20th-80th
2016 (serum) cross- percentile:
sectional 191.2 -
3997.6
y-ng/mL
Median =
16.5 ng/mL,
(range: 2.6 -
3559 ng/mL)
EE
OR (for decile 2 vs. decile 1 of PFOA) 1.09
OR (for decile 3 vs. decile 1 of PFOA) 1.19
OR (for decile 4 vs. decile 1 of PFOA) 1.26
OR (for decile 5 vs. decile 1 of PFOA) 1.4
OR (for decile 6 vs. decile 1 of PFOA) 1.39
OR (for decile 7 vs. decile 1 of PFOA) 1.31
OR (for decile 8 vs. decile 1 of PFOA) 1.42
OR (for decile 9 vs. decile 1 of PFOA) 1.4
OR (for decile 10 vs. decile 1 of PFOA) 1.54
OR (per 1-In ng/mL increase in PFOA) 1.1
OR (for Q2 vs. Q1) 1.43
OR (for Q3 vs. Q1) 1.56
OR (for Q4 vs. Q1) 1.52
OR (per Hn y-ng/mL increase in modeled cumulative - n.
PFOS)
OR [for quintile 2 (191.2 - <311.3 y-ng/mL) vs. quintile 1 1 1?
(50.3 - <191.2 y-ng/mL) modeled cumulative PFOA)
OR [for quintile 3 (311.3 - < 794.1 y-ng/mL) vs. quintile 1 . ..
(50.3 - <191.2 y-ng/mL) modeled cumulative PFOA)
OR [for quintile 4 (794.1 - <3997.6 y-ng/mL) vs. quintile 1 1 -
(50.3 - <191.2 y-ng/mL) modeled cumulative PFOA)
OR [for quintile 5 (3997.6 - 205667.3 y-ng/mL) vs. quintile . 1fi
1 (50.3 - <191.2 y-ng/mL) modeled cumulative PFOA)
OR (per 1-ln y-ng/mL increase in 2005/2006 estimated 1 n.
serum PFOA) u
OR [for quintile 2 (5.8- <11.4 ng/mL) vs. quintile 1 (2.6 - Q
<5.8 ng/mL) 2005/2006 estimated serum PFOS]
OR [for quintile 3 (11.4-< 26.7 ng/mL) vs. quintile 1 (2.6 - 1 nfi
<5.8 ng/mL) 2005/2006 estimated serum PFOS]
OR [for quintile 4 (26.7-< 81.5 ng/mL) vs quintile 1 (2.6 - .
<5.8 ng/mL) 2005/2006 estimated serum PFOS]
OR [for quintile 5 (81 5 - 3558.8 ng/mL) vs quintile 1 (2.6 - 1 .
<5.8 ng/mL) 2005/2006 estimated serum PFOS]
Effect Estimate +
1.5 2.0
Figure 3-9. Odds of Elevated ALT Levels from Epidemiology Studies Following Exposure
to PFOA
Interactive figure and additional study details available on Tableau.
3-33
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Confidence
Rating
Reference
Exposure Study
Matrix Design
Exposure Levels
Sub-population
Comparison
EE
0.00 0.02
Effect Estimate +
0.04 0.06 0.08
Medium
confidence
Darrow el
al., 2016
Modeled
(serum)
Cohort and
cross -
sectional
20th - 80th percentile: 191.2 - 3997.6
y-ng/mL
-
Regression coefficient (per 1-ln
y-ng/mL increase in modeled
cumulative PFOA)
0.01
1
1
1
< 50 years old
Regression coefficient (per 1-ln
y-ng/mL increase in modeled
cumulative PFOA)
0.01
I
1
1
>= 50 years old
Regression coefficient (per 1-ln
y-ng/mL increase in modeled
cumulative PFOA)
0.01
1
1
Female
Regression coefficient (per 1-ln
y-ng/mL increase in modeled
cumulative PFOA)
0.01
i
i
i
Male
Regression coefficient (per 1-ln
y-ng/mL increase in modeled
cumulative PFOA)
0.01
t
i
i
Median = 16.5 ng/mL, (range: 2.6 - 3559
ng/mL)
-
Regression coefficient (per 1-ln
ng/mL increase in 2005/2006
estimated serum PFOA)
0.01
j
i
i
< 50 years old
Regression coefficient (per 1-ln
ng/mL increase in 2005/2006
estimated serum PFOA)
0.01
1
i
i
>= 50 years old
Regression coefficient (per 1-ln
ng/mL increase in 2005/2006
estimated serum PFOA)
0.01
1
i
i
Female
Regression coefficient (per 1-ln
ng/mL increase in 2005/2006
estimated PFOA)
0.01
1
i
i
Male
Regression coefficient (per 1-ln
ng/mL increase in 2005/2006
estimated PFOA)
0.01
i
i
i
Jain et al..
2019
Serum
Cohort
Geometric mean (95% CI) = 2.0 ng7mL (1.8
-2.1)
Obese
Regression coefficient (per
1-log10 ng/mL increase in PFOA)
0.07
1
i
i
i
•
Geometric mean (95% CI) = 2.2 ng/mL (2.0
-2.3)
Non-obese
Regression coefficient (per
1-log10 ng/mL increase in PFOA)
0.01
t
i
i •
i
Nian et al..
Serum
Cross -
Median=6.19 ng/mL (25th-75lh percentile:
Excluding
Regression coefficient (per 1-ln
0.05
I
i
2019
sectional
4.08-9.31 ng/mL)
medicine takers
ng/mL increase in PFOA)
i
i
Salihovic et
Plasma
Cohort
Median (25th-75th percentile): Age 70: 3.31
ng/mL (2.52-4.39): Age 75: 3.81 ng/mL
(2.71-5.41); Age 80: 2.53 ng/mL (1.82-3.61)
Regression coefficient (per 1-ln
0.04
1
i
al.. 2018
ng/mL increase in PFOA)
i
i
0.00 0.02
0.04
0.06 0.08
Figure 3-10. ALT Levels from Medium Confidence Epidemiology Studies Following
Exposure to PFOA
Interactive figure and additional study details available on Tableau.
In children and adolescents, positive associations were observed in girls (with exposure-response
gradient across quartiles) in the medium confidence study by Attanasio et al. (2019, 5412069)
and in the low confidence study of obese children {Khalil, 2018, 4238547}. However, inverse
associations were observed in boys in Attanasio et al. (2019, 5412069) and Mora et al. (2018,
4239224), which may indicate that the associations in children are less consistent than in adults
or that there are sex differences in children. Insufficient data were available to assess the
potential for effect modification by sex.
The studies that examined AST are presented in the Appendix (see PFOA Appendix). In adults
in the general population, positive associations were observed in the two medium confidence
studies {Jain, 2019, 5381541; Nian, 2019, 5080307}. In the two low confidence studies of
fluorochemical plant workers {Olsen, 2012, 2919185; Wang, 2012, 2919184}, no associations
were observed. In children including adolescents, the medium confidence study {Attanasio, 2019,
5412069} reported a positive association in girls but an inverse association in boys. In the low
confidence study {Khalil, 2018, 4238547}, the direction of association was inverse, but the result
was extremely imprecise. For the other liver enzymes (bilirubin, GGT), results were generally
consistent with those of ALT and AST, with the exception that inverse associ ations for bilirubin
were observed in some studies {Salihovic, 2018, 5083555; Darrow, 2016, 3749173}.
3-34
-------
DRAFT FOR PUBLIC COMMENT
March 2023
For functional measures of liver injury, two medium confidence studies (one in adults and one in
children including adolescents) examined histology endpoints. Both studies examined lobular
inflammation. Rantakokko et al. (2015, 3351439) reported that higher PFOA exposure levels
were associated with extremely reduced odds of lobular inflammation (OR = 0.02, p < 0.05),
whereas Jin et al. (2020, 6315720) reported the opposite direction of association, though the
results in the latter study were non-monotonic and not statistically significant. Jin et al. (2020,
6315720) additionally reported lower odds of ballooning and portal inflammation, but higher
odds of steatosis (association non-monotonic) and nonalcoholic steatohepatitis. Three additional
studies examined some form of liver disease. In a medium confidence study, Darrow et al. (2016,
3749173) reported no increases in any liver disease or specifically enlarged liver, fatty liver, or
cirrhosis. In contrast, in a low confidence study, Girardi and Merler (2019, 6315730) reported
that workers at a PFAS production plant had higher mortality from liver cancer or cirrhosis when
compared to regional mortality statistics and a control group of non-chemical workers (p < 0.05
for some comparisons). Lastly, a second low confidence study by Liu et al. (2018, 4238396)
examined hepatic fat mass and found no correlation with PFOA exposure.
3.4.1.2 Animal Evidence Study Quality Evaluation and Synthesis
There are 9 animal toxicological studies from the 2016 PFOA HESD {U.S. EPA, 2016,
3603279} and 19 studies identified from recent systematic literature searches and review efforts
conducted after publication of the 2016 PFOA HESD that investigated the association between
PFOA and hepatic effects. Study quality evaluations for these 28 studies are shown in Figure
3-11 and Figure 3-12.
3-35
-------
DRAFT FOR PUBLIC COMMENT
March 2023
s^s o^°V^
Abbott et al„ 2007, 1335452 -
l
+
I
+
I
+
I
+
I
+
++
++
++
++
B
Biegel et al., 2001, 673581 -
++
++
NR
++
++
B
++
++
++
++
Blake et al., 2020, 6305864 -
+
+
++
+
+
++
B
++
B
Butenhoff et al., 2004, 1291063 -
++
NR
NR
++ ++
0
++
++
++
++
Butenhoff et al., 2012, 2919192-
+
++
NR
-
+
++
++
B
++
B
Cope et al., 2021, 10176465-
+
+
++
++
+
D
++
0
B
Crebelli et al., 2019, 5381564 -
+
+
~
++
+
+
+
+
+
~
De Guise et al., 2021, 9959746 -
+
+
NR
+
+
++
++
++
B
Guoetal., 2019, 5080372-
+
+
NR
++
+
++
B
B
B
Guoetal., 2021, 7542749-
£
+
NR
+
+
D
++
++*
B
Guoetal., 2021, 9960713-
+
'A
+
-
++
++
ft
B
Guo etal.,2021, 9963377-
+
+
NR
+
-*
++
++
++~
++
a
Huetal., 2010, 1332421 -
++
NR
NR
++
-
B
++
D
++
B
Lauetal., 2006, 1276159-
+
+
NR
+
+
++
++
++
B
B
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
B Critically deficient (metric) or Uninformative (overall)
Not reported
* Multiple judgments exist
Figure 3-11. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects
Interactive figure and additional study details available on HAWC.
3-36
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Li et al., 2017, 4238518
Li et al., 2018, 5084746-
Loveless et al., 2008, 988599 -
Macon etal., 2011, 1276151
NTP, 2019, 5400977
NTP, 2020, 7330145-
Perkins et al,, 2004, 1291118 -
Shi et al., 2020, 7161650
Wolf etal., 2007, 1332672
Yan et al., 2014, 2850901 -
Yan etal., 2017, 3981501 -
Yu etal., 2016, 3981487-
Zhang et al., 2020, 6505878
Zhang etal., 2021, 10176453
NR
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Not reported
* Multiple judgments exist
Figure 3-12. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-37
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Hepatic effects (e.g., increased absolute and relative liver weight, altered clinical parameters
indicating potential liver injury, and histopathological alterations of liver tissue) were observed
in male and female mice, rats, and monkeys after oral PFOA doses of different durations. Data
from numerous studies provide evidence confirming that the liver is a target of PFOA toxicity.
3.4.1.2.1 Liver Weight
Generally, increases in absolute and/or relative liver weight were observed in all available PFOA
animal toxicological studies, regardless of species, sex, life stage, and exposure paradigm (Figure
3-13). Significant increases in absolute and relative liver weight were reported at doses as low as
0.05 mg/kg/day and 0.31 mg/kg/day, respectively {Li, 2017, 4238518; Yan, 2014, 2850901},
and were often observed at the lowest dose administered in each study. In male mice, significant
increases in absolute and/or relative liver weights were observed at doses ranging from 0.31-
30 mg/kg/day after 4-5 weeks of exposure {Loveless, 2008, 988599; Minata, 2010, 1937251;
Yan, 2014, 2850901; Yu, 2016, 3981487; Li, 2017, 4238518; Crebelli, 2019, 5381564; Guo,
2019, 5080372; Guo, 2021, 9963377; Shi, 2020, 7161650}. Similarly, significant increases in
absolute and relative liver weights were reported in male rat short-term/sub chronic studies at
doses of 0.625-30 mg/kg/day {Perkins, 2004, 1291118; Loveless, 2008, 988599; Cui, 2009,
757868; NTP, 2019, 5400977}. Two subchronic dietary studies in adult male rats with exposures
lasting 13-16 weeks reported significantly increased absolute and relative liver weights at doses
as low as 1 mg/kg/day {Perkins, 2004, 1291118; NTP, 2020, 7330145}. In one chronic study in
male Crl:CD BR (CD) rats, relative liver weight was significantly increased after 15 months of
exposure to 13.6 mg/kg/day via the diet {Biegel, 2001, 673581}. Similar results were observed
at the 1-year interim sacrifice of a 2-year dietary study in male Sprague-Dawley rats exposed to
14.2 mg/kg/day PFOA, but the effect was not statistically significant at the 2-year timepoint
{Butenhoff, 2012, 2919192}. Male cynomolgus monkeys orally administered PFOA capsules
daily for 26 weeks also had significantly increased absolute liver weights at doses
> 3 mg/kg/day, though the increase in relative liver weight was only statistically significant in
the highest dose group (30/20 mg/kg/day) {Butenhoff, 2002, 1276161}.
Several systemic toxicity studies evaluating liver weight in female mice and rats after short-term,
subchronic, or chronic PFOA exposures are also available {Butenhoff, 2012, 2919192; De
Guise, 2021, 9959746; Li, 2017, 4238518; NTP, 2019, 5400977; NTP, 2020, 7330145; Zhang,
2020, 6505878}. Two 28-day studies in female mice reported significant increases in absolute
liver weight at doses ranging from 0.05-5 mg/kg/day (relative liver weight not reported) {Li,
2017, 4238518; Zhang, 2020, 6505878}. A third 28-day study in female B6C3F1 mice reported
significant increases in absolute and relative liver weights at both doses tested (1.88 and
7.5 mg/kg/day) {De Guise, 2021, 9959746}. NTP (2019, 5400977) conducted a 28-day gavage
study in female Sprague-Dawley rats and reported significant increases in both absolute and
relative liver weights at doses > 25 mg/kg/day. In a chronic feeding study (see study design
details in Section 3.4.4.2.1.2), NTP (2020, 7330145) reported significant increases in absolute
and relative liver weight in female Sprague-Dawley rats after 16 weeks of exposure to 63.4 but
not 18.2 mg/kg/day PFOA. A 2-year feeding study in female Sprague-Dawley rats similarly
found no significant difference in absolute or relative liver weight at doses of 1.6 or
16.1 mg/kg/day PFOA {Butenhoff, 2012, 2919192}.
There are also multiple reproductive and developmental toxicity studies that report maternal
and/or offspring liver weight in rodents after gestational PFOA exposures. Blake et al. (2020,
3-38
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6305864) reported significant increases in absolute and relative liver weights in CD-I mouse
dams exposed to PFOA at doses of 1 or 5 mg/kg/day from GD 1.5-11.5 or GD 1.5-17.5. Yahia
et al. (2010, 1332451) similarly reported significant increases in maternal ICR mouse absolute
liver weights at doses > 5 mg/kg/day and relative liver weights at doses > 1 mg/kg/day. In a 2-
generation reproductive toxicity study in Sprague-Dawley rats {Butenhoff, 2004, 1291063}, Po
dams dosed with 1, 3, 10, or 30 mg/kg/day PFOA at least 70 days prior to mating through
lactation did not show consistent alterations in absolute or relative liver weights at the time of
sacrifice on PND 22. However, significantly increased absolute and relative liver weights were
observed in Po males and male Fi offspring starting at the lowest dose of 1 mg/kg/day, whereas
no statistically significant differences in absolute or relative liver weights were reported for
female Fi offspring.
Several other developmental toxicity studies reported significantly increased maternal, fetal,
and/or pup liver weights associated with gestational PFOA exposure, but the authors did not
further examine tissue or serum samples for hepatic effects {Lau, 2006, 1276159; Wolf, 2007,
1332672; Abbott, 2007, 1335452; White, 2009, 194811; Macon, 2011, 1276151; White, 2011,
1276150; Tucker, 2015, 2851046; Li, 2018, 5084746; Cope, 2021, 10176465}. For example,
White et al. (2011, 1276150) orally dosed pregnant CD-I mice with 0, 1, or 5 mg/kg/day PFOA
from GD 1 to GD 17. Fi offspring liver-to-body weight ratios were significantly increased at
1 mg/kg/day on PND 22 and at 5 mg/kg/day on PND 22 and PND 42. Macon et al. (2011,
1276151) exposed pregnant CD-I mice to PFOA from GD 1 to GD 17 (full gestation) or GD 10
to GD 17 (late gestation). At PND 7, significantly increased absolute and relative liver weights in
offspring were observed as low as 0.3 mg/kg/day after full-gestation exposure; significantly
increased absolute and relative liver weights were also observed at the high dose of 1 mg/kg/day
PFOA after late-gestation exposure (PND 4 and PND 7; relative liver weights were also
significantly increased at PND 14). Wolf et al. (2007, 1332672) reported that offspring of
pregnant CD-I mice orally dosed with 0 and 5 mg/kg/day on GD 7-GD 17, GD 10-GD 17, GD
13-GD 17, and GD 15-GD 17 or with 20 mg/kg/day on GD 15-GD 17 had significantly
increased liver-to-body weight ratios at PND 22. White et al. (2009, 194811) reported that
offspring of CD-I mice exposed to 5 mg/kg/day PFOA during gestation or during gestation plus
lactation had significantly increased liver-to-body weight ratios on PND 1. Inconsistent results
were observed on PND 22 and PND 128 in male and female CD-I mice gestationally exposed to
0.1 and 1 mg/kg/day PFOA from GD 1.5-17.5 and then given either a high- or low-fat diet
starting on PND 22 {Cope, 2021, 10176465}. Specifically, increased relative liver weights were
observed at PND 22 for both males and females exposed to 1 mg/kg/day (statistically significant
in males only), but not at PND 128 {Cope, 2021, 10176465}. One study reported no significant
change in relative liver weights, which were only measured on PND 48 in the female offspring
of C57BL/6N mouse dams exposed to 0.5 or 1 mg/kg/day PFOA in drinking water from GD 6-
17 {Hu, 2010, 1332421}.
3-39
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA Hepatic Effects - Relative Liver Weight
Study Name Study Design Observation Time Animal Description # No significant change A Significant increase T Significant decrease |
Blake etal., 2020.6305864 developmental (GD1.5-11.5) GD11.5 P0 Mouse, CD-1 ( + , N=11) < — ——
developmental (GD1.5-17.5) GD17.5 P0 Mouse, CD-1 { ], N=11) < i A A
Li et al., 2018, 5084746 developmental (GD1-17) GD18 P0 Mouse, Kunming N=10) ' *^~~~ ——Ai—
Wolf etal., 2007, 1332672 developmental (GD1-17) PND22 P0 Mouse, CD-1 (*, N=25-39) < — AA
F1 Mouse, CD-1 (;, N=11-14) < — A-A
F1 Mouse, CD-1 (y, N=11-14) < - AA
developmental (GD15-17) PND22 P0 Mouse, CD-1 (J, N=3-10) » i —^A
F1 Mouse, CD-1 (V, N=3-10) < A ~
developmental (GD1-PND22) PND22 F1 Mouse, CD-1 ( ?, N=12-14) < A A
F1 Mouse, CD-1 ($, N=12-14) < — A-A
developmental (PND1-22) PND22 F1 Mouse, CD-1 ( ', N=11-14) < - AA
F1 Mouse, CD-1 ($, N=11-14) < - A-A
Abbott et al., 2007, 1335452 developmental (GD0-GD17) PND22 P0 Mouse, 129S1/SvlmJ (£, N=11-36) « I • • A A I
F1 Mouse, 129S1/SvlmJ N=4-9) < A A AA
Hu etal., 2010, 1332421 developmental (GD6-17) PND48 F1 Mouse, C57BL/6n (?, N=8) < ' ¦ >
Macon et al., 2011, 1276151 developmental (GD10-17) PND1 F1 Mouse, CD-1 ( , N=3-5) < ^^
PND7 F1 Mouse, CD-1 (V. N=3-5) < • A
PND14 F1 Mouse, CD-1 ft, N=2-5) i I A
PND21 F1 Mouse, CD-1 ($, N=2-5) < i • #
developmental (GD1-17) PND7 F1 Mouse, CD-1 ( ', N=3-6) < ^
F1 Mouse, CD-1 ($, N=4-5) < < AAA
PND14 F1 Mouse, CD-1 (o, N=4-6) < i • A
F1 Mouse, CD-1 (y, N=4-6) < ' • A A
PND21 F1 Mouse, CD-1 (', N=4) < m 9 A
F1 Mouse, CD-1 ( f, N=3-6) < i I • A
PND84 F1 Mouse. CD-1 (:', N=3-5) < > I t
F1 Mouse, CD-1 (¥, N=2-5) < • I ~
Cope etal.: 2021, 10176465 developmental (GD1.5-GD 17.5) PND22 F1 Mouse, CD-1 (;, N=8) ¦ • A
F1 Mouse, CD-1 (f, N=9) ' • ~
developmental (GD1.5-17.5) PND128 F1 Mouse, CD-1 (y, N=7) < ' I ~
F1 Mouse, CD-1 (¦•?, N=8) < • ~
Yan etal., 2014. 2850901 short-term (28d) 28d Mouse, BALB/c(o, N=16) < ^^^A^^^~A
Yu etal., 2016,3981487 short-term (28d) 28d Mouse, BALB/c 0,N=5) ' • A
Guo et al., 2021, 9963377 short-term (28d) 28d Mouse, BALB/c (.T. N= 12) " AAA
De Guise et al., 2021, 9959746 short-term (4wk) 28d Mouse, B6C3F1 (L. N=12-16) < ' —-A
Guo etal., 2019, 5080372 short-term (4wk) 4wk Mouse, BALB/c(o, N= 12) < AAA
Loveless et al., 2008,988599 short-term (29d) 29d Mouse, Crl:CD-1(ICR)BR (d, N=20) ' • A A A
Shi etal., 2020,7161650 subchronic (5wk) 5wk Mouse, C57BL/6J 0, N=8) < i AAA
Butenhoff et al., 2004, 1291063 reproductive (84d) LD22 P0 Rat. Crl:CD(SD)IGS BR (y. N=26-29) < • ~ W •
reproductive (64d) 106d P0 Rat, Crl:CD(SD)IGS BR ( ¦, N=29-30) < A A A A
reproductive (GD0-PND106) LD22 F1 Rat, Crl:CD(SD)IGS BR (£, N=28-29) < ' • • • >
reproductive (GD0-PND120) PND120 F1 Rat, Crl:CD(SD)IGS BR (->, N=29-30) < A A A A
Loveless et al., 2008.988599 short-term <29d) 29d Rat, Sprague-Dawley Cri:Cd(Sd)(Br) ( f, N=10) n • • ———A~~tA
NTP, 2019.5400977 short-term (28d) 29d Rat, Sprague-Dawley (;?, N=10) < i AAA A—A
Rat, Sprague-Dawley {'+, N=9-10) < t t A A li i
Perkins et al., 2004, 1291118 subchronic (13wk) 13wk Rat, Sprague-Dawley Crl:Cd Br (f, N=15) < i ^ > A~^tA
NTP, 2020,7330145 chronic (GD6-PNW21) 16wk F1 Rat, Sprague-Dawley (-?, N=10) ' A A
chronic (GD6-PNW107) 16wk F1 Rat, Sprague-Dawley (;?, N=10) < AAA
F1 Rat, Sprague-Dawley ($., N=10) <' • —
chronic (PND21-PNW21) 16wk F1 Rat, Sprague-Dawley (;t, N=10) ''
chronic (PND21-PNW107) 16wk F1 Rat, Sprague-Dawley (
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA concentration is presented in logarithmic scale to optimize the spatial presentation of data. Interactive figure and additional
study details available on HAWC.
GD = gestation day; PND = postnatal day; PNW = postnatal week; LD = lactational day; Po = parental generation; Fi = first
generation; d = day; wk = week; y = year.
3.4.1.2.2 Clinical Chemistry Measures
Albumin, a blood protein that plays a major role in PFOA toxicokinetics (Section 3.3), is
synthesized by the liver. Increases in serum albumin were reported in several short-term and
chronic studies in male rodents, with increases observed at doses as low as 0.4 and
1.3 mg/kg/day in mice and rats, respectively {Butenhoff, 2012, 2919192; Yan, 2014, 2850901;
Guo, 2019, 5080372; NTP, 2020, 7330145}. Females appeared to be less sensitive, with
increased albumin at doses > 25 mg/kg/day in rats after short-term or chronic exposures and no
significant differences or inconsistent decreases in pregnant mice after gestational exposures
{Yahia, 2010, 1332451; Butenhoff, 2012, 2919192; NTP, 2019, 5400977; Blake, 2020,
6305864; NTP, 2020, 7330145}. The albumin/globulin ratio was significantly increased in both
adult males and females after PFOA exposure for 28 days or 16 weeks {Guo, 2019, 5080372;
NTP, 2019, 5400977; NTP, 2020, 7330145}.
Similar to albumin, inconsistent results were observed for total protein, with statistically
significant decreases observed in some studies in male rats {NTP, 2019, 5400977; NTP, 2020,
7330145} and pregnant female mice in one study {Blake, 2020, 6305864}, and increases or no
significant changes observed in several other studies in adult male rats or mice {Guo, 2019,
5080372; Butenhoff, 2012, 2919192} and in female rats {Butenhoff, 2012, 2919192; NTP, 2019,
5400977; NTP, 2020, 7330145}.
Increases in enzymes including ALT, ALP, and AST following PFOA exposures were observed
across multiple species, sexes, and exposure paradigms (Figure 3-14 (male mice), Figure 3-15
(male rats), Figure 3-16 (female rodents)). These enzymes are often useful indicators of hepatic
enzyme induction, hepatocellular damage, or hepatobiliary damage as increased serum levels are
thought to be due to hepatocyte damage resulting in release into the blood {EPA, 2002, 625713}.
Alterations in serum enzymes are generally considered to reach biological significance and
indicate potential adversity at levels > 2-fold compared to controls (i.e., > 100% change relative
to controls) {U.S. EPA, 2002, 625713; Hall, 2012, 2718645}.
In adult male mice dosed with PFOA for 4-5 weeks, statistically significant increases in ALT
and/or AST were observed at PFOA exposure levels ranging from 2-21.6 mg/kg/day {Minata,
2010, 1937251; Yan, 2014, 2850901; Crebelli, 2019, 5381564; Guo, 2019, 5080372}. Increases
in ALT were > 100% above control values at doses as low as 1.25 mg/kg/day {Yan, 2014,
2850901}. Biologically significant increases in AST were only observed in two of these studies
at doses > 20 mg/kg/day {Minata, 2010, 1937251; Yan, 2014, 2850901}. In the only short-term
study examining ALP in male mice, ALP was significantly increased at concentrations of 5 and
20 mg/kg/day after 28-day exposure {Yan, 2014, 2850901}; serum ALP levels were > 100%)
change at doses of 1.25 mg/kg/day and higher.
In male CD-I mice gestationally exposed to 0.1 and 1 mg/kg/day from GD 1.5-17.5 and then fed
either a high- or low-fat diet starting on PND 22, no significant changes were observed in ALT,
AST, or ALP on PND 128 {Cope, 2021, 10176465}.
3-41
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA Hepatic Effects - Serum Enzymes in Male Mice
Endpoint Study Name Study Design Observation Time Animal Description Dose (mg/kg/day) | Q Statistically significant % Not statistically significant |—\ 95% CI |
Alanine Aminotransferase (ALT) Cope etal.. 2021,10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse, CD-1 {•', N=8) 0
0.1
1
Yan etal.. 2014, 2850901 short-term (28d)
Guo et al.. 2019, 5080372 short-term (4wk)
Crebelli etal., 2019, 5381564 subchronic {5wk)
Alkaline Phosphatase (ALP) Cope etal.. 2021,10176465 developmental (GD1.5-17.5) PNW18
Yan etal., 2014, 2850901 short-term (28d)
Aspartate Aminotransferase (AST) Cope et al.. 2021,10176465 developmental (GD1.5-17.5) PNW18
Yan etal., 2014, 2850901 short-term (28d)
Guo et al., 2019, 5080372 short-term (4wk)
Crebelli etal., 2019. 5381564 subchronic (5wk)
Mouse. BALB/c (-,:, N=6) 0
0.08
0.31
1.25
Mouse, BALB/c (.¦$, N=10) 0
Mouse. CS7BI/6 ( ; , N=6-6) 0
F1 Mouse. CD-1 (¦', N=8) 0
Mouse, BALB/c (o. N=6) 0
0.08
0.31
1.25
F1 Mouse, CD-1 { N=8) 0
Mouse. BALB/c (, '. N=6) 0
0.31
1.25
Mouse, BALB/c (..>, N=10) 0
Mouse. C57BI/6 (-?, N=6-8) 0
'*'
'M
it*—i
i*i
'<
»
i
4
i
i
!
i
i
l—
• 1
i
i
(
1
i
»
i
i
i
i
i*i
'*'
'*'
i
p
i*i
i*i
'*'
:~!
i*i
i*i
i
•
i
i
•
i
D
i*i
i*i
i*i
<*<
!~!
,*,
i*i
i
i
•
i
>:
,*,
i ©i
i
0
!~!
i
5i
500 1,000 1,500 2,000 2,500 3,000
Percent control response (%)
Figure 3-14. Percent Change in Serum Enzyme Levels Relative to Controls in Male Mice
Following Exposure to PFOAa'b
Interactive figure and additional study details available on HAWC and Tableau.
ALT = alanine aminotransferase; ALP = alkaline phosphatase; AST = aspartate aminotransferase; d = day; wk = week;
CI = confidence interval.
a The red dashed lines indicate a 100% increase or 100% decrease from the control response.
bResults for Yan et al. (2014, 2850901) are presented for 6 doses (0, 0.08, 0.31, 1.25, 5, and 20 mg/kg/day), and a statistically
significant response of 7,000% occurred at the highest dose for the ALT endpoint. The axis has been truncated at 3,000% to
allow results at lower doses for other studies and endpoints to be legible.
NTP (2019, 5400977; 2020, 7330145) reported significantly increased ALT and ALP at all doses
tested in the 28-day and 16-week exposures of male Sprague-Dawley rats to PFOA (dose range
of 0.625-32.1 mg/kg/day). However, increases in ALT did not exceed 100% change in either
study. Similarly, increases in ALP did not exceed 100% change in the 28-day gavage study
{NTP, 2019, 5400977} and only exceeded 100% change with doses > 15.6 mg/kg/day at the 16-
week interim time point of the chronic dietary study {NTP, 2020, 7330145}. In another chronic
dietary study, Butenhoff et al. (2012, 2919192) generally observed increased ALT and ALP in
male Sprague-Dawley rats dosed with 1.3 and 14.2 mg/kg/day PFOA at time points ranging from
3-42
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3 months to 2 years of administration. Increases in ALT were above or approximately 100%
change in both dose groups at 6, 12, and 18 months of exposure. ALP levels were elevated at all
time points with 14.2 mg/kg/day PFOA but were only above 100% change at the 18-month time
point. AST was also less sensitive than ALT or ALP in male rats. NTP (2019, 5400977)
observed statistically significant but not biologically significant increases in AST at doses of
2.5 mg/kg/day and higher (up to 10 mg/kg/day) after 4 weeks. Butenhoff et al. (2012, 2919192)
did not observe biologically significant increases in AST at any time of assessment during the 2-
year feeding study.
PFOA Hepatic Effects - Serum Enzymes in Male Rats
Endpoint Study Name Study Design Observation Time Animal Description
Alanine Aminotransferase (ALT) NTP, 2019. 5400977 short-term (28d) 29d Ral. Sprague-Dawley (fj, N=10)
Dose (mg/kg/day)
NTP, 2020, 7330145
; (GD6-PNW21) 16wk
: (GD6-PNW107) 16wk
chronic (PND21-PNW21) 16wk
chronic (PND21-PNW107) 16wk
Butenhoff et al., 2012, 2919192 chronic (2y)
Alkaline Phosphatase (ALP) NTP, 2019,5400977 short-term (28d)
NTP, 2020. 7330145
: (GD6-PNW21) 16wk
: (GD6-PNW107) 16wk
: (FND21-PNW21) 16wk
chronic (PND21-PNW107) 16wk
Butenhoff etal., 2012. 2919192 chronic (2y)
Aspartate Aminotransferase (AST) NTP, 2019, 5400977 short-term (28d) 29d
Butenhoff etal., 2012. 2919192 chronic (2y)
F1 Rat, Sprague-Dawley (,C, N=10)
F1 Rat, Spraguc-Dawlcy (J, N=10)
F1 Rat. Sprague-Dawley ( N=10)
F1 Rat, Sprague-Dawley ( ¦', N=10)
Ral. Sprague-Dawley Crl:Cd(Sd)(Br) N=14-15)
Rat, Sprague-Dawley (-.*. N=10)
F1 Rat. Sprague-Dawley (', N=10)
F1 Rat. Sprague-Dawley ( >, N=10)
F1 Rat. Sprague-Dawley N=10)
F1 Ral, Sprague-Dawley (rf, N=10)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) ( i'. N=14-15)
Rat, Spraguo-Dawloy N=10)
Rat. Sprague-Dawley Cri:Cd(Sd)(Br) N=14-15)
I—#—
I • I
I •—
-40 -20 0 20
100 120 140 160 180 200
Figure 3-15. Percent Change in Enzyme Levels Relative to Controls in Male Rats Following
Exposure to PFOA3
Interactive figure and additional study details available on HAWC and Tableau.
3-43
-------
DRAFT FOR PUBLIC COMMENT
March 2023
ALT = alanine aminotransferase; ALP = alkaline phosphatase; AST = aspartate aminotransferase; GD = gestation day;
PND = postnatal day; PNW = postnatal week; Fi = first generation; d = day; wk = week; CI = confidence interval.
a The red dashed line indicates a 100% increase from the control response.
In addition to the findings in rodents, no consistent responses of serum enzymes were observed
in the one available study in male cynomolgus monkeys dosed with PFOA for 26 weeks
{Butenhoff, 2002, 1276161}.
The only available studies measuring ALT, AST, or ALP in female mice were after gestational
PFOA exposures. Blake et al. (2020, 6305864) reported no statistically significant effects on
ALT or ALP levels in CD-I dams after gestational PFOA exposure, and significantly increased
AST (113% increase over control) only after exposure to the high dose of 5 mg/kg/day from GD
1.5-17.5. In contrast, Yahia et al. (2010, 1332451) reported biologically significant increases in
ALT and AST in dams after gestational exposure to 5 or 10 mg/kg/day PFOA (150% and 372%)
increase from control ALT levels, respectively; 312%> and 813%) increase from control AST
levels, respectively). Biologically significant increases in ALT, ALP, and AST were only
observed at the highest dose of 10 mg/kg/day. In a study in which female CD-I mice were
gestationally exposed to 0.1 or 1 mg/kg/day from GD 1.5-17.5 and then given a low-fat diet
starting on PND 22, no significant changes were observed in ALT, AST, or ALP on PND 128
{Cope, 2021, 10176465}. However, in the group of females exposed to 1 mg/kg/day and then
given a high-fat diet, statistically significant increases were observed in ALT (130%> control),
AST (23% control), and ALP (43%> control).
Short-term and chronic studies reported statistically but not biologically significant increases in
ALT in female rats after 4- or 16-week PFOA exposures between 50-100 mg/kg/day {NTP,
2019, 5400977; NTP, 2020, 7330145}. The 4- and 16-week studies also reported no biologically
significant changes in ALP with any PFOA dose, though PFOA exposures resulted in
statistically significant ALP increases at gavage doses as low as 6.25 mg/kg/day after 4 weeks
{NTP, 2019, 5400977; NTP, 2020, 7330145}. NTP (2019, 5400977) and found no statistically or
biologically significant differences in AST in adult female Sprague-Dawley rats following 4-
week PFOA gavage dosing. Butenhoff et al. (2012, 2919192) also did not observe statistically
significant changes in ALT, AST, or ALP in adult female Sprague-Dawley rats exposed to 1.6 or
16.1 mg/kg/day PFOA for up to 2 years.
3-44
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA Hepatic Effects — Serum Enzymes in Female Rodents
Endpolnt Study Name Study Design Observation Time Animal Description Dose (mg/kg/day) [ © Statistically significant # Not statistically significant |—195% Cl |
le Aminotransferase (ALT) Blake et al , 2020, 6305864 developmental (GD1.5-11.5) GD11.5 PO Mouse, CD-1 N=5)
developmental (GD1.5-17,5) GD17.5 PO Mouse, CD-1 (i, N=4-6) 0
Cope el al.. 2021. 10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 (',. N=7)
NTP, 2019, 5400977
irt-term (28d) 29d
NTP 2020,7330145 chronic (GD6-PNW107) 16wk
chronic (PND21-PNW107) 16wk
Alkaline Phosphatase (ALP) Blake et al., 2020, 6305864 developmental (GD1.5-11.5) GD11.5
developmental (GD1.5-17.5) GD17.5
NTP, 2019, 5400977 short-term (28d)
NTP, 2020,7330145 chronic (GD6-PNW107) 16wk
chronic (PND21-PNW107) 16wk
Aspartate Aminotransferase (AST) Blake el al.. 2020. 6305864 developmental (GD1.5-11.5) GD11.5
NTP, 2019, 5400977
i (28d) 29d
Rat, Sprague-Dawley (-, N=9-10) 0
6.25
F1 Rat, Sprague-Dawley (-, N=10) 0
18.4
63.5
F1 Rat, Sprague-Dawley (-, N=10) 0
18.2
63.4
P0 Mouse, CD-1 (". N=5) 0
1
5
PO Mouse. CD-1 (k, N=4-6) 0
Cope el al., 2021, 10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 N=7)
Rat, Spraguo-Dawley (-, N=9-10) 0
6.25
12.5
25
50
100
F1 Rat, Sprague-Dawley (", N=10) 0
18.4
63.5
F1 Rat, Sprague-Dawley (-, N=10) 0
18.2
63.4
P0 Mouse, CD-1 (", N=5) 0
developmental (GD1.5-17.5) GD17.5 P0 Mouse. CD-1 (-, N=4-6) 0
Cope et al.. 2021,10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 {'„• N=7)
Rat, Sprague-Dawley (", N=9-10) 0
hin
&
h*~l
jH
m
~
-I 1 I—
-150 -100 -50 0 50 100 150 200 250 300
Percent control response (%)
Figure 3-16. Percent Change in Enzyme Levels Relative to Controls in Female Rodents
Following Exposure to PFOA1'
Interactive figure and additional study details available on HAWC and Tableau.
ALT = alanine aminotransferase; ALP = alkaline phosphatase; AST = aspartate aminotransferase; GD = gestation day;
PND = postnatal day; PNW = postnatal week; Po = parental generation; Fi = first generation; d = day; wk = week;
CI = confidence interval.
a The red dashed lines indicate a 100% increase or 100% decrease from the control response.
3-45
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.1.2.3 Histopathology
The available animal toxicology literature provides evidence of alterations in liver
histopathology were observed after PFOA exposure. Increased cell proliferation/division, bile
duct hyperplasia, and hepatocellular hypertrophy were common responses across multiple
studies. Loveless et al. (2008, 988599) reported increased incidence and severity of
hepatocellular hypertrophy with increasing doses of PFOA (0.3-30 mg/kg/day) in male CD-I
mice dosed for 29 days (incidences of 0/19, 20/20, 20/20, 20/20, and 19/19 (all severity grades
combined) in the 0, 0.3, 1, 10, and 30 mg/kg/day groups, respectively). Several other 28-day
studies in adult male mice provided qualitative descriptions and images as evidence of increased
hypertrophy, though results were not quantitatively reported {Minata, 2010, 1937251; Yan,
2017, 3981501; Li, 2017, 4238518; Guo, 2019, 5080372}.
Doses as low as 0.3 mg/kg/day PFOA resulted in increased incidence and severity of
hypertrophy in male rats dosed for 28 or 29 days {Perkins, 2004, 1291118; Loveless, 2008,
988599; NTP, 2019, 5400977}; female rats dosed for 28 days showed slight increases at
50 mg/kg/day (20%) and a 100% hypertrophy incidence rate at 100 mg/kg/day compared to 0%
incidence at all lower doses (6.25, 12.5, or 25 mg/kg/day) and in controls (n = 10) {NTP, 2019,
5400977}. Butenhoff et al. (2012, 2919192) reported significant increases in the incidence of
hypertrophy in male and female adult Sprague-Dawley rats administered PFOA for 1 or 2 years
at the highest dose tested for each sex (14.2 and 16.1 mg/kg/day for males and females,
respectively). NTP (2020, 7330145) also reported increased incidence of hepatocellular
hypertrophy in male and female adult rats dosed with PFOA for 16 or 107 weeks (see study
design details in Section 3.4.4.2.1.2). At the 16-week interim necropsy, males had significantly
increased incidences of hypertrophy at all doses tested (1-32.1 mg/kg/day); significantly
increased incidences of hypertrophy were only observed in females at the highest doses tested
(63.4/63.5 mg/kg/day) at 16 weeks. At 107-weeks, significantly increased incidences of
hypertrophy were observed in males and females at doses >1.1 mg/kg/day and
> 18.2 mg/kg/day, respectively.
In a developmental toxicity study, Blake et al. (2020, 6305864) observed 100% incidence of
hepatocellular hypertrophy with decreased glycogen and intensely eosinophilic granular
cytoplasm at both the GD 11.5 and GD 17.5 time points with doses of 1 and 5 mg/kg/day
compared to 0% incidence in controls (all n = 5-6); however, control CD-I mouse dams at the
GD 17.5 time point also exhibited what the authors characterized as hepatocellular hypertrophy
consistent with pregnancy at that stage of gestation. Quist et al. (2015, 6570066) similarly
reported increased severity of hepatocellular hypertrophy with increasing PFOA doses (0.01-
1 mg/kg/day) in PND 91 female CD-I mouse offspring exposed from GD 1-17. In a standard 2-
generation reproductive toxicity study, significant increases in the incidence of diffuse
hepatocellular hypertrophy were reported for male Fi Sprague-Dawley rat offspring at doses of
3 mg/kg/day and higher {Butenhoff, 2004, 1291063}.
In addition to hepatocellular hypertrophy, significantly increased incidences of mitotic figures
and bile duct hyperplasia were observed in adult male CD-I mice exposed to 10 or 30 mg/kg/day
PFOA for 29 days {Loveless, 2008, 988599}. NTP (2020, 7330145) reported significantly
increased incidences of mitoses and bile duct hyperplasia in female Sprague-Dawley rats dosed
with 63.5 mg/kg/day PFOA for 2 years, but not in males. In contrast, Filgo et al. (2015,
2851085) reported the incidence and severity of bile duct hyperplasia in two strains of 18-month-
3-46
-------
DRAFT FOR PUBLIC COMMENT
March 2023
old wild-type female mice exposed to PFOA during gestation and found no alterations in CD-I
mice and a significant decrease in the severity of bile duct hyperplasia in 129/Sv mice. However,
increased mitoses were observed (data not provided) in ICR mouse dams exposed to 1-
10 mg/kg/day PFOA during gestation {Yahia, 2010, 1332451}.
Several studies reported cytoplasmic alterations including cytoplasmic vacuolization resulting
from PFOA exposures. Male mice dosed with PFOA for 28 days were reported to have increased
vacuolation at doses between 5.4-21.6 mg/kg/day (incidence data not provided) and significantly
decreased numbers of nuclei per unit area with 28-day exposures to > 0.4 mg/kg/day {Minata,
2010, 1937251; Guo, 2019, 5080372}. Male rats were particularly susceptible to cytoplasmic
alterations; NTP (2019, 5400977; 2020, 7330145) reported incidences of 90-100% in animals
receiving doses > 1 mg/kg/day for 4 or 16 weeks compared to 0% incidences in controls (all
n = 10). In the 2-year study, males receiving >2.1 mg/kg/day showed a 58% or greater incidence
rate compared to 0% incidence rates in controls (all n = 50) {NTP, 2020, 7330145}.
Female rats receiving doses > 25 mg/kg/day for 4, 16, or 107 weeks had 98%—100% incidence
rates of cytoplasmic alterations compared to 0% incidence rates in controls {NTP, 2019,
5400977; NTP, 2020, 7330145}. In CD-I mouse dams, 100%) incidence rates of cytoplasmic
vacuolization were observed only at the highest dose of 5 mg/kg/day but at both gestational time
points (GD 11.5 and GD 17.5) compared to 0% incidence rates in controls (n = 5-6) {Blake,
2020, 6305864}. In this study, the vacuoles frequently contained remnant membrane material as
myelin figures.
Cell and tissue death10 and degeneration was the final category of hepatic histological effects
observed across multiple studies, species, and sexes (Table 3-2). Incidence rates of individual
cell necrosis in male CD-I mice dosed with PFOA for 29 days were above 50% at doses
> 1 mg/kg/day {Loveless, 2008, 988599}. There was similarly a significantly increased
percentage of necrotic liver cells, analyzed by flow cytometry, in male C57BL/6 mice
administered 5 mg/kg/day PFOA in drinking water for 5 weeks {Crebelli, 2019, 5381564}.
Significantly increased incidences of single cell death were observed in male Sprague-Dawley
rats after 16 weeks of exposure to doses as low as 1 mg/kg/day but were not increased in females
at this time point {NTP, 2020, 7330145}. Incidence rates of single cell death in male and female
rats after 2-year exposures as reported in NTP (2020, 7330145) are provided in Table 3-2 (see
further study design details in Section 3.4.4.2.1.2). Apoptosis and single-cell necrosis were also
observed in livers of pregnant CD-I mice after gestational exposures of 1 and 5 mg/kg/day, with
increasing length of exposure resulting in increased incidence rates {Blake, 2020, 6305864}. In
male and female CD-I mice gestationally exposed to 0.1 and 1 mg/kg/day from GD 1.5-17.5 and
then given a low-fat diet on PND 22, incidences of single cell necrosis were higher in the
exposed groups but not significantly increased at PNW 18 (Table 3-2) {Cope, 2021, 10176465}.
However, in females exposed to 1 mg/kg/day and then to a high-fat diet, incidences of single cell
necrosis were significantly increased at PNW 18.
In male CD-I mice exposed to PFOA for 29 days, the incidence of hepatic focal necrosis
increased with increasing PFOA doses between 1-30 mg/kg/day {Loveless, 2008, 988599}. In
10 In this document, EPA used the cell death nomenclature as reported in the individual studies to describe the observed effects.
Cell "necrosis" is a type of cell death, the term for which is generally used when a specific method to distinguish necrotic cells
from other dying cells (e.g., apoptotic cells) has been employed {Elmore, 2016, 10671182}. EPA did not evaluate the methods of
individual studies to ensure that the nomenclature used by the authors accurately reflected the type of cell death reported.
3-47
-------
DRAFT FOR PUBLIC COMMENT
March 2023
the same study, increased incidences of necrosis were reported in male Sprague-Dawley rats only
with the highest dose tested (30 mg/kg/day) {Loveless, 2008, 988599}. Inconsistent incidences
of hepatic necrosis were observed in male and female Sprague-Dawley rats administered PFOA
in feed for 16 weeks, though there were increases reported after 2 years {NTP, 2020, 7330145}.
Table 3-2 depicts the 2-year data for males and females. In a separate 2-year study, there were no
significant differences in the incidences of hepatic necrosis in male or female Sprague-Dawley
rats {Butenhoff, 2012, 2919192}. Blake et al. (2020, 6305864) did not observe consistent
increases in the incidence of focal necrosis in mouse CD-I dams dosed with PFOA during
gestation. However, Butenhoff et al. (2004, 1291063) reported significant increases in focal and
multifocal necrosis in Fi generation male Sprague-Dawley rats in a 2-generation reproductive
toxicity study (data not provided).
Table 3-2. Associations Between PFOA Exposure and Cell Death or Necrosis in Rodents
Reference
Study Design
Endpoint Name
Incidence
Males
NTP (2019,
5400977)
Loveless (2008,
988599)
Perkins (2004,
1291118)a
Butenhoff (2012,
2919192)
Cope (2021,
10176465)b
NTP (2020,
7330145)
28-day Sprague Dawley rat
oral gavage dosing; 0, 0.625,
1.25, 2.5, 5, 10 mg/kg/day
29-day Crl:CD(SD)IGS BR
rat oral gavage dosing; 0,
0.3, 1, 10, 30 mg/kg/day
29-day Crl:CD-l(ICR)BR
mouse oral gavage dosing; 0,
0.3, 1, 10, 30 mg/kg/day
29-day Crl:CD-l(ICR)BR
mouse oral gavage dosing; 0,
0.3, 1, 10, 30 mg/kg/day
4-week Crl:CD®BR rat
feeding study; 0, 0.06, 0.64,
1.94, 6.5 mg/kg/day
7-week Crl:CD®BR rat
feeding study; 0, 0.06, 0.64,
1.94, 6.5 mg/kg/day
13-week Crl:CD®BR rat
feeding study; 0, 0.06, 0.64,
1.94, 6.5 mg/kg/day
2-year Crl:COBS®
CD(SD)BR rat feeding
study; 0, 1.3,
14.2 mg/kg/day
Gestational CD-I mouse
gavage dosing from GD 1.5-
GD 17.5 (offspring); 0, 0.1,
1 mg/kg/day
16-week Hsd: Sprague
Dawley SD rat feeding
study, with and without
perinatal exposure; 0/0,
0/150, 0/300, 150/150, and
300/300 ppm
Focal Hepatocellular Necrosis
Focal Necrosis
Individual Cell Necrosis
Focal Necrosis
Coagulative Necrosis
Coagulative Necrosis
Coagulative Necrosis
Focal Hepatocellular Necrosis
Hepatocyte Single Cell Necrosis
Hepatocellular Single Cell Death
Necrosis
0/10, 0/10, 0/10, 0/10, 1/10,
0/10
0/10, 0/10, 0/10, 1/10, 4/10
0/19, 0/20, 11/20, 20/20,
19/19
0/19, 1/20, 3/20, 4/20, 7/19
0/15,0/15,0/15, 1/15,2/14
0/15,0/15,0/15,0/15, 1/15
0/15, 1/15,0/15, 1/15,0/15
3/50, 5/50, 5/50
2/8, 5/9, 6/9
0/10, 10/10, 10/10, 9/10,
10/10
0/10, 6/10, 2/10, 2/10, 4/10
3-48
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Reference
Study Design
Endpoint Name
Incidence
16-week Hsd:Sprague
Dawley SD rat feeding
study, with and without
perinatal exposure; 0/0, 0/20,
0/40, 0/80, 300/0, 300/20,
300/40, 300/80 ppm
2-year Hsd:Sprague Dawley
SD rat feeding study, with
and without perinatal
exposure; 0/0, 0/20, 0/40,
0/80, 300/0, 300/20, 300/40,
300/80 ppm
Hepatocellular Single Cell Death
Necrosis
Hepatocellular Single Cell Death
Necrosis
0/10, 7/10, 9/10, 10/10,
0/10, 5/10, 8/10, 10/10
1/10, 1/10, 6/10, 4/10, 0/10,
2/10, 3/10, 1/10
1/50, 1/50, 11/50, 24/50,
1/50, 3/50, 5/50, 29/50
2/50, 17/50, 23/50, 20/50,
1/50, 11/50, 14/50,21/50
Females
NTP (2019,
5400977)°
Butenhoff (2012,
2919192)
Blake (2020,
6305864)
Cope (2021,
10176465)b
NTP (2020,
7330145)
28-day Hsd:Sprague Dawley
SD rat oral gavage dosing; 0,
6.25, 12.5, 25, 50,
100 mg/kg/day
2-year Crl:COBS@
CD(SD)BR rat feeding
study; 0, 1.6,
16.1 mg/kg/day
Gestational CD-I mouse
gavage dosing from GD 1.5-
GD 11.5 (dams); 0, 1,
5 mg/kg/day
Gestational CD-I mouse
gavage dosing from GD 1.5-
GD 17.5 (dams); 0, 1,
5 mg/kg/day
Gestational CD-I mouse
gavage dosing from GD 1.5-
GD 17.5 (offspring); 0, 0.1,
1 mg/kg/day
16-week Hsd:Sprague
Dawley SD rat feeding
study, with and without
perinatal exposure; 0/0,
0/300, 0/1,000, 150/300, and
300/1,000 ppm
2-year Hsd:Sprague Dawley
SD rat feeding study, with
and without perinatal
exposure; 0/0, 0/300,
0/1,000, 150/300, and
300/1,000 ppm
Focal Hepatocellular Necrosis
Focal Hepatocellular Necrosis
Focal Necrosis
Cell Death (including apoptosis and
single-cell necrosis of individual
hepatocytes)
Focal Necrosis
Cell Death (including apoptosis and
single-cell necrosis of individual
hepatocytes)
Hepatocyte Single Cell Necrosis
Hepatocellular Single Cell Death
Necrosis
Hepatocellular Single Cell Death
Necrosis
0/10, 0/10, 0/10, 0/10, 0/10,
0/10
5/50, 6/50, 2/50
1/5, 0/5, 2/5
0/5, 1/5, 3/5
0/5, 0/5, 0/6
0/5, 5/5, 6/6
1/9, 3/9, 4/10
0/10, 0/10, 1/10, 0/10, 0/10
0/10, 0/10, 2/10, 0/10, 0/10
0/50, 4/50, 29/50, 5/50,
32/50
0/50, 1/50, 8/50, 4/50, 5/50
Notes: GD = gestation day.
incidence data as reported by Perkins et al. (2004,1291118) were split into severity categories within the original study. For the
purposes of this table, all non-grade 0 severities were considered an incidence (results for severity grades 1-3 were combined).
bData are summarized for low-fat diet only from Cope et al. (2021, 10176465).
c Incidence data not explicitly reported by NTP (2019, 5400977).
3-49
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Cystic degeneration was also observed across two chronic feeding studies in male rats. Butenhoff
et al. (2012, 2919192) reported incidences of cystic degeneration characterized as areas of
multilocular microcysts in the liver parenchyma in 4/50 (8%), 7/50 (14%), and 28/50 (56%) male
rats dosed for 2 years with 0, 1.3, or 14.2 mg/kg/day, respectively. NTP (2020, 7330145)
similarly reported increases in the incidence of cystic degeneration in the liver of male rats
administered 4.6 mg/kg/day PFOA for 107 weeks.
3.4.1.2.4 Additional Hepatic Endpoints
A suite of other liver effects was observed but were either not included as endpoints of interest
across multiple studies or had inconsistent results between studies, sexes, and/or species. These
included serum measures of gamma-glutamyl transpeptidase (only measured in one short-term
study of male BALB/C mice that showed increases at 2 and 10 mg/kg/day exposures) {Guo,
2021, 9963377}, bile acids (study results generally showed no response or increases at high
doses) {Butenhoff, 2002, 1276161; Yan, 2014, 2850901; NTP, 2019, 5400977; Blake, 2020,
6305864; NTP, 2020, 7330145; Guo, 2021, 9963377}, bilirubin (study results showed no change
or minimal increases at high doses) {Butenhoff, 2002, 1276161; Butenhoff, 2012, 2919192;
Yahia, 2010, 1332451; NTP, 2019, 5400977; Guo, 2021, 7542749}, and histopathological
findings such as hepatic inflammation (study results showed increased incidence/severity,
decreased incidence, or no response) {Filgo, 2015, 2851085; Quist, 2015, 6570066; NTP, 2020,
7330145}, increased incidence of cellular infiltration {Cope, 2021, 10176465; Butenhoff, 2012,
2919192}, and increased incidence of hepatocytomegaly {Zhang, 2020, 6505878}. NTP (2020,
7330145) also reported a variety of other histopathological outcomes including eosinophilic or
mixed-cell foci (significant increases in male Sprague-Dawley rats) and pigmentation
(significant increases in males and females). Butenhoff et al. (2004, 1291063) similarly reported
increased discoloration of the liver in male Fi Sprague-Dawley rats analyzed during a standard 2-
generation reproductive toxicity study.
3.4.1.3 Mechanistic Evidence
Mechanistic evidence linking PFOA exposure to adverse hepatic outcomes is discussed in
Sections 3.2.1, 3.2.2, 3.2.3, 3.2.7, 3.2.8, 3.2.9, 3.3.2, 3.3.3, 3.3.4, 3.4.1, 3.4.2, 3.4.3, 3.4.4, and
4.2 of the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}. There are 81 studies from recent
systematic literature search and review efforts conducted after publication of the 2016 PFOA
HESD that investigated the mechanisms of action of PFOA that lead to hepatic effects. A
summary of these studies is shown in Figure 3-17.
3-50
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Mechanistic Pathway
Animal
Human
In Vitro
Grand Total
Atherogenesis And Clot Formation
0
0
1
1
Big Data, Non-Targeted Analysis
8
0
10
17
Cell Growth, Differentiation, Proliferation, Or Viability
17
1
36
50
Cell Signaling Or Signal Transduction
14
1
17
30
Extracellular Matrix Or Molecules
1
0
1
2
Fatty Acid Synthesis, Metabolism, Storage, Transport, Binding, B-Oxidation
20
0
19
36
Hormone Function
6
1
1
8
Inflammation And Immune Response
5
1
3
9
Oxidative Stress
8
0
14
21
Xenobiotic Metabolism
7
1
12
19
Other
0
0
3
3
Grand Total
41
2
46
81
Figure 3-17. Summary of Mechanistic Studies of PFOA and Hepatic Effects
Interactive figure and additional study details available on Tableau.
3.4.1.3.1 Nuclear Receptor Activation
3.4.1.3.1.1 Introduction
The ability of PFOA to mediate hepatotoxicity via nuclear receptor activation has been
investigated for several receptor-signaling pathways, including that of the peroxisome
proliferator-activated receptors (PPARa, PPARS, PPARy), the pregnane X receptor (PXR), and
the constitutive androstane receptor (CAR). PPARa is a major target for PFOA. A primary
mechanism of hepatic injury associated with PFOA-mediated activation of PPARa relates to
impacts on hepatic lipid metabolism caused by altered expression of genes and proteins within
the PPARa signaling pathway {U.S. EPA, 2016, 3603279; Das, 2017, 3859817; Hui, 2017,
3981345; Li, 2019, 5387402; Pouwer, 2019, 5080587; Rebholz, 2016, 3981499; van Esterik,
2015, 2850288; Wang , 2013, 2850952; Wen, 2019, 5080582; Yan, 2015, 2851199; Yang , 2014,
2850321}. Activation of PPARa has been cited as a mechanism of action for PFAS, including
PFOA {U.S. EPA, 2016, 3603279}, because of the association between hepatic lesions and/or
increased liver weight and peroxisome proliferation downstream of PPARa activation in rats.
However, increased hepatic lipid content in the absence of a strong PPARa response (i.e.,
activation of downstream target genes) is a characteristic of exposure to PFOA. Additionally,
many of the genes activated by PFOA are regulated by transcription factors other than PPARa,
including CAR, PPARy, PXR, Era, and HNF4a {U.S. EPA, 2016, 3603279}. PPARs, CAR, and
PXR are nuclear receptors that can form heterodimers with one another to induce transcription of
linked genes. Other factors impacting nuclear receptor activation in hepatocytes include dose and
duration of PFOA exposure and the genetic background, diet, and sex of exposed animals. Sex-
specific hepatic effects varied by strain, and long-term PFOA oral exposure in mice with pre-
existing steatosis had protective effects against hepatic injury {NTP, 2019, 5400977; Li, 2017,
4238518; Li, 2019, 5080362}. Thus, the underlying mechanism(s) of PFOA-induced
hepatotoxicity may involve multiple nuclear receptors. Additionally, hepatic effects observed
3-51
-------
DRAFT FOR PUBLIC COMMENT
March 2023
with PFAS exposure, including inflammation and necrosis, cannot be fully explained by PPARa
activation (Section 3.4.1.2.3). This updated assessment includes a summary of studies that have
examined PPARs, CAR, PXR, Era, and HNF4a activation as potential mechanisms underlying
the health effects induced by PFOA.
3.4.1.3.1.2 PPARa Receptor Binding and Activation
Receptor binding and activation assays have been performed to examine the association between
activation of PPARs, CAR, and/or PXR, and PFOA-mediated hepatotoxicity. PPARs modulate
gene expression in response to exogenous or endogenous ligands and play essential roles in lipid
metabolism, energy homeostasis, development, and cell differentiation {U.S. EPA, 2016,
3603279}.
Several studies used luciferase reporter assays to examine the activation of PPARa by PFOA in
vitro using human and animal cell lines transfected with mouse and human PPARa {Wolf, 2014,
2850908; Rosenmai, 2018, 4220319; Behr, 2020, 6305866; Buhrke, 2013, 2325346}. In African
green monkey kidney COS-1 cells transfected with mouse PPARa, PFOA was the most potent
activator of PPARa among the 5 PFAS tested, with PPARa activation observed at less than 1 [xM
after a 24 h exposure {Wolf, 2014, 2850908}. A study in human HEK293T cells found that
human PPARa was activated at a concentration of 50 |iM PFOA after a 24 h exposure {Behr,
2020, 6305866}. Whether PFOA activates other nuclear receptors is less clear from studies
conducted in HEK293 cells and may be cell type- and dose-dependent. PFOA had no activity in
HEK293 cells transfected with constructs encoding other nuclear receptors, including PPARS,
CAR, PXR, the farnesoid X receptor (FXR), the liver X receptor a (LXRa), the retinoid X
receptor a (RXRa) and retinoic acid receptor a (RARa), at concentrations up to 100 |iM for 24
hours {Behr, 2020, 6305866}. In a second study using a human PPARa construct in HEK293
cells, PFOA induced PPARa activation at concentrations of 25 |iM and higher, whereas PFOA
concentrations of at least 100 |iM were necessary to activate PPARy and PPARS {Buhrke, 2013,
2325346}. Results from the single study conducted in a human hepatic cell line (HepG2) were
consistent with results in other cell lines {Rosenmai, 2018, 4220319}. Of the 14 PFAS
substances tested, PFOA was the most potent PPARa activator, showing significant elevation of
luciferase activity after a 24 hour exposure to 30 and 100 |iM PFOA. While luciferase levels
were elevated at 10 |iM of PFOA, the increase did not reach significance. These in vitro studies
support PPARa activation by PFOA.
Another study measured the expression of hepatic carboxylesterases (Ces) that function in the
metabolism of drugs, chemical toxicants, and endogenous lipids {Wen, 2019, 5080582}. PFOA
upregulated expression of the PPARa target gene, Cyp4al4, in the livers of male C57BL/6 NCrl
mice after exposure to 3 mg/kg/day by gavage for 7 days. PFOA exposure also led to alterations
to the expression of Ces genes: Cesld, le, If, 1 g, 2c, and 2e mRNA levels were increased
between 1.5- and 2.5-fold, while Ceslc and 2b transcripts were decreased. In a second study
within Wen et al. (2019, 5080582), Ces genes were measured in the livers of C57BL/6NTac
mice and PPARa-null mice also exposed to 3 mg/kg/day PFOA by gavage for 7 days. Cesle and
7/mRNA and protein levels were PPARa dependent, whereas Ceslc, Id, 1 g, 2a, 2b, and 2e
mRNA and CES2 protein levels were induced by PFOA in PPARa-null mice, implicating a
CAR-mediated pathway for differential expression of these genes.
3-52
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The mechanism by which PFOA activates PPARa is likely dependent on interactions with liver
fatty acid binding protein (L-FABP). L-FABP facilitates the nucleo-cytoplasmic shuttling of
activator ligands, such as fatty acids, for nuclear receptors, including PPAR activators, PXRs,
and LXRs. PFOA is structurally similar to fatty acids, and both exhibit a strong binding affinity
with L-FABP (Section 3.3.1.2). Thus, L-FABP is responsible for delivering PFOA to the nuclei
of hepatic cells for access to nuclear receptors. Sheng et al. (2018, 4199441) used circular
dichroism (CD) spectroscopy, fluorescence displacement assays, and molecular docking
approaches to evaluate the binding mode and capacity of PFOA as well as PFOS and PFAS
replacement chemicals to purified human L-FABP (hL-FABP). The purified recombinant hL-
FABP was calculated to consist of 15.7% a-helix and 54.4% P-sheet. In the presence of PFOA,
a-helix content of the protein increased slightly, whereas the P-sheet content decreased. The
dissociation constant (Kd) of PFOA to hL-FABP was 8.03 ± 2.10 [xM, which was higher than
PFOS and lower than some (but not all) replacement PFAS substances. By molecular docking,
PFOA bonded with hL-FABP in a "head-out" mode, such that the carboxyl head of PFOA will
interacted with R122 amino acid residue through hydrogen bonding and N111 amino acids
residue through hydrophobic interactions. Introduction of oxygen molecules into the backbone
could flip the binding prediction to a "head-in" mode characterized by interactions with amino
acid residue N61. By comparing PFOA to PFOS and replacement PFAS chemicals, the authors
demonstrated that these three parameters correlated both with cytotoxicity in human liver HL-
7702 cells and binding affinity for hL-FABP. Notably, expression of select PPARa-regulated
genes showed no significant change across the chemicals tested, with one exception, the Cd36
gene. Expression of other genes, including cell cycle genes, did correlate with these binding
parameters. These findings suggest that binding of PFAS to hL-FABP can mediate toxicity in a
manner that is not exclusively dependent on PPARa-mediated changes in gene expression in
liver cells, but possibly through effects on other FABP-related events such as binding to the
CD36 protein or effects on cell proliferation.
3.4.1.3.1.3 Receptor Binding and Activation of Other Nuclear Receptors
PFOA can activate PPARa in the liver of rodents and humans. However, the extent by which
activation of PPARa mediates hepatoxicity may be species-specific, and activation of other
receptors may also contribute to toxicity {U.S. EPA, 2016, 3603279}. Indeed, studies in mice
and rats indicate that PFOA may activate PPARa, CAR, and PXR in the liver {NTP, 2019,
5400977; Wen, 2019, 5080582; Li, 2019, 5080362; Rose, 2016, 9959775}.
Several studies observed perturbations in lipid transport, fatty acid metabolism, triglyceride
synthesis, and cholesterol synthesis in PFOA-exposed mice {Das, 2017, 3859817; Rosen, 2017,
3859803; Li, 2019, 5387402}. A few of these studies, Das et al. (2017, 3859817), Rosen et al.
(2008, 1290832), and Rosen et al. (2017, 3859803), investigated the effects of PFOA on lipid
metabolism and homeostasis in the absence of PPARa by using knockout mouse models. After
exposure to 10 mg/kg/day PFOA for 7 days, Das et al. (2017, 3859817) observed that a smaller
subset of genes related to lipid homeostasis was activated in PPARa null mice compared to wild-
type (WT) mice. Increased expression of genes regulating fatty acid and triglyceride synthesis
and transport into hepatocytes was attenuated but not entirely abolished in PFOA-exposed
PPARa null mice compared to WT mice. Gene expression changes in PPARa null mice
implicate a role for PPARp/S and/or PPARy in the absence of PPARa {Rosen, 2008, 1290832}.
Mechanistically, these changes correlated with the development of steatosis in PFOA-exposed
WT mice consistent with increased triglyceride accumulation. In contrast, elevated triglyceride
3-53
-------
DRAFT FOR PUBLIC COMMENT
March 2023
levels and steatosis develop in PPARa null mice even in the absence of PFOA exposure. The
authors propose that PFOA exposure alters lipid metabolism to favor biosynthesis and
accumulation over P-oxidation, leading to hepatic steatosis. PFOA increased the expression of
genes related to fatty acid P-oxidation, lipid catabolism, lipid synthesis, and lipid transport in
both strains; however, gene induction was lower in PPARa null mice {Rosen, 2008, 1290832;
Rosen, 2017, 3859803}. In fact, the authors suggest that the transcriptome of the mice resembled
that of mice treated with PPARy agonists, thus indicating a role for other PPAR isoforms in the
dysregulation of lipid synthesis {Rosen, 2017, 3859803}. Furthermore, Rosen and colleagues
{2017, 3859803} demonstrated that PFOA significantly downregulated the Signal Transducer
and Activator of Transcription 5B gene (STAT5B), a transcription factor and member of the
STAT family, in a PPARa-dependent manner. STAT5B has been demonstrated in regulation of
sexually-dimorphic gene expression in the liver between males and females, raising the
possibility that that PFOA exposure may promote feminization of the liver in male mice
{Oshida, 2016, 6781228; Rosen, 2017, 3859803}.
Increasing evidence links CAR activation as a mechanism of PFOA-induced liver toxicity {NTP,
2019, 5400977; Wen, 2019, 5080582; Li, 2019, 5080362}. The use of genetically modified mice
and gene expression analyses has demonstrated that PFOA exposure activates both PPARa and
CAR receptors {NTP, 2019, 5400977; Abe, 2017, 3981405; Li, 2019, 5080362; Rosen, 2017,
3859803; Wen, 2019, 5080582; Li, 2019, 5080362; Oshida, 2015, 2850125; Oshida, 2015,
5386121}.
Five recent studies also examined PFOA activation of CAR-specific genes {Abe, 2017,
3981405; NTP, 2019, 5400977; Wen, 2019, 5080582; Rosen, 2017, 3859803; Rose, 2016,
9959775}. Additionally, one study used both a cell-based reporter assay and in silico approaches
to examine PFOA activation of PXR {Zhang, 2020, 6324307}, and one study examined other
PFOA effects on other nuclear receptors in vitro {Buhrke, 2015, 2850235}. In support of PFOA
as a CAR receptor activator, PFOA induced expression of the CAR target genes CYP2B6 in a
human hepatocyte cell line in vitro (HepaRG), and Cyp2bl0 in wild type mice but not CAR-null
mice in vivo {Abe, 2017, 3981405}. Evidence of CAR-specific gene expression was also noted
in male and female rats administered PFOA. Exposed animals exhibited significant increases in
expression of PPARa-stimulated genes (Acoxl, Cyp4al) and CAR-specific genes (Cyp2bl,
Cyp2b2) in livers compared to controls, suggesting increases in PPARa and CAR activity {NTP,
2019, 5400977}. Males were exposed to a range of doses between 0 and 10 mg/kg/day and
females to between 0 and 100 mg/kg/day PFOA for 28 days. Gene expression in liver tissue was
analyzed using qRT-PCR. Female rats displayed the greatest fold increase for the CAR-related
genes Cyp2bl whereas males exhibited the greatest fold increase for Cyp4al and Cyp2bl
compared to controls.
Rosen et al. (2008, 1290832) postulated that gene expression changes in the liver should overlap
between PFOA and phenobarbital, a known CAR activator. To test this, differentially expressed
genes in wild-type or CAR-null mice treated with PFOA by gavage (3 mg/kg/day) for 7 days
were compared to differentially expressed genes in the livers of mice exposed to 100 mg/kg/day
phenobarbital for three days {Rosen, 2017, 3859803}. Similarity in differentially expressed
genes between the two studies (i.e., overlap) was analyzed using a Running Fisher Test for
pairwise comparisons. As expected, there was significant similarity between the lists of
differentially expressed genes for PFOA and phenobarbital in WT mice, but not in CAR-null
3-54
-------
DRAFT FOR PUBLIC COMMENT
March 2023
mice. In fact, close to 15% of genes differentially expressed upon PFOA exposure in liver were
considered PPARa-independent. Two gene expression compendium studies further analyzed
these data using gene expression biomarker signatures built using microarray profiles from livers
ofWT mice, CAR-null mice {Oshida, 2015, 2850125}, and PPARa-null mice {Oshida, 2015,
5386121}. These analyses found that both CAR and PPARa were activated by PFOA, and that
CAR activation was generally more significant in PPARa-null mice. The authors concluded that
CAR likely plays a subordinate role to PPARa in mediating the adverse hepatic effects of PFOA
{Oshida, 2015,2850125}.
Activation of CAR may occur via direct activation or indirect activation. Indirect activation of
CAR by phenobarbital involves blockade of the downstream phosphorylation pathway of EGFR
protein phosphatase 2A (PP2A), which dephosphorylates CAR to enable nuclear translocation.
Using a COS-1 fibroblast cell-based reporter gene assay that is capable of detecting CAR ligands
but not indirect activators, Abe et al. (2017, 3981405) observed that PFOA failed to activate
reporter gene expression. In a second study using primary mouse hepatocytes, PFOA exposure
led to CAR-mediated expression of Cyp2bl0 even in the presence of okadaic acid, a PP2A drug
inhibitor. Together these findings suggest the mechanism of PFOA-mediated CAR activation
indirect and distinct from that of phenobarbital. Moreover, an analysis of historical and new data
of gene expression in PPARa- and CAR-null mice indicate the pathway of PFOA-mediated CAR
activation is PPARa-independent {Rosen, 2017, 3859803}. Thus, the precise mechanism of
CAR activation by PFOA remains to be determined.
Several studies evaluated PFOA activation of other nuclear receptors. Rosen et al. {2017,
3859803} noted that PFOA activated PPARy and ERa in trans-activation assays from the
ToxCast screening program. Zhang et al. {2020, 6324307} used a cell-based reporter assay and
an in silico approach to estimate PFOA-mediated activation of the PXR receptor. The PFOA log
EC50 was 5.04 M in the luciferase-based PXR reporter assay, a higher concentration (i.e., less
potent) than observed for PPARa. These authors also developed classical QSAR and 3D-QSAR
models that predicted very similar values of log EC50 of 4.92 M and 4.94 M, respectively. Both
models suggested that molecular structural factors including molecular polarizability, charge,
and atomic mass are key parameters dictating hPXR agonistic activity of PFOA and other
perfluoroalkyl chemicals.
In addition to the key role of PPARa and other nuclear receptors discussed above, other
transcription factors and epigenetic mechanisms influence PFOA-mediated changes in lipid
metabolism and storage. Beggs et al. (2016, 3981474) observed a decrease in hepatocyte nuclear
factor alpha (HNF4a) protein, a master regulator or hepatic differentiation, in the livers of ten-
week-old CD-I mice exposed to 3 mg/kg/day PFOA once daily by oral gavage for 7 days.
HNF4a regulates liver development (hepatocyte quiescence and differentiation), transcriptional
regulation of liver-specific genes, and regulation of lipid metabolism. In this study, PFOA
exposure correlated with downregulation of HNF4a target genes involved in differentiation
(Cyp7al) and induced pro-mitogenic genes including CCND1. Other genes altered by PFOA
exposure mapped to pathways involved in lipid metabolism, liver cholestasis, and hepatic
steatosis. PFOA also led to diminished accumulation of HNFa protein. This decrease in HNF4a
was not accompanied by a change in expression of the gene, suggesting that the decrease in
HNF4a occurs post-translationally. The decreased HNFa correlated with upregulation of genes
that are negative targets of HNF4a. HNF4a is considered an orphan receptor, with various fatty
3-55
-------
DRAFT FOR PUBLIC COMMENT
March 2023
acids as its endogenous ligands. These fatty acids maintain the structure of the receptor
homodimer. PFOA and PFOS are analogous in structure to fatty acids and may also provide
stabilization of the homodimer. The authors investigated the role of PFOA and PFOS interaction
with this protein via in silico docking models, which showed a displacement of fatty acids by
PFOA/PFOS, possibly tagging HNF4a for degradation. The authors hypothesize that steatosis,
hepatomegaly, and carcinoma in rodents may be a consequence of the loss of this protein and
also presents a mechanism for PFOA-induced hepatic effects in humans.
In primary human hepatocytes exposed to 1, 25, or 100 |iM PFOA for 24 hours, the number of
differentially regulated genes was 43, 109, and 215, respectively, as measured using a human
genome gene chip {Buhrke, 2015, 2850235}. Based on known activators of the differentially
expressed genes, the authors suggest that in addition to PPARa, PPARy and HNF4a may
contribute to changes in expression of genes involved in carnitine metabolism. PFOA-mediated
induction of ERa signaling was also predicted based on pathway analysis.
3.4.1.3.1.4 Host Factors Impacting PPARa Signaling
The effects of PFOA on PPARa activation depend on diet and pre-existing conditions {Li, 2019,
5080362}. Mice were subjected to control diet or high-fat diet (HFD) for 16 weeks to induce
non-alcoholic fatty liver disease (NAFLD), after which they were exposed to vehicle or 1
mg/kg/day PFOA by oral gavage for 2, 8, or 16 weeks; control diet and HFD were continued
throughout this exposure period. Preexisting NAFLD in mice fed a HFD enhanced the induction
of PPARa activation by PFOA early in the exposure but reduced the severity of macrovesicular
steatosis and sinusoidal fibrosis induced by a HFD, and reversed HFD-induced increase in body
weight and serum alanine aminotransferase (ALT). The authors hypothesized that PFOA
exposure in animals with a lipid burden in the liver leads to PFOA-mediated inhibition of fatty
acid biosynthesis pathways by the metabolic end-product feedback effect. The authors also
observed reduced Tgf-P gene expression in PFOA-treated HFD-fed mice compared to vehicle-
treated HFD-fed mice, which could account for the diminished level of hepatic stellate cell
activation and collagen production associated with fibrosis. Furthermore, the duration of PFOA
exposure impacted gene expression and hepatic injury. For example, PFOA induced Srebfl and
Srebf2 genes in the fatty acid biosynthesis pathway following 2 weeks of treatment, but this
effect was not seen following 8 or 16 weeks of PFOA treatment. Notably, this increase in Srebfl
expression following 2 weeks of PFOA exposure was only observed with the co-treatment of
PFOA and HFD; the Srebfl effect was not observed in the PFOA-treated mice fed the control
diet.
PFOA-driven changes in PPARa-mediated gene expression may also be modified be age, strain,
or species. Pregnant Kunming mice were exposed to PFOA at doses of 1, 2.5, 5 and 10
mg/kg/day from gestational days 1-17, and female offspring were analyzed on postnatal day 21
{Li, 2019, 5387402}. Genes involved in fatty acid P-oxidation including acyl-CoA synthetase
(Acsll), carnitine palmitoyl transferase I, Palmitoyl-CoA oxidase (Acoxl), acyl-CoA
thioesterase 1 (Acotl), and carnitine palmitoyltransferase la (Cptla) were significantly
downregulated at the two highest doses, as was the PPARa gene. In this strain of mouse,
perinatal PFOA disrupts the gene expression of enzymes involved in fatty acid oxidation induced
by PPARa, possibly through an epigenetic mechanism. In contrast, several studies have shown
PFOA to upregulate expression of PPAR signaling pathway genes, including Acox in rats and
mice {Li, 2019, 5080362; NTP, 2019, 5400977; Cavallini, 2017, 3981367}. One such study
3-56
-------
DRAFT FOR PUBLIC COMMENT
March 2023
proposed that the PFOA-mediated gene expression changes are due to changes in the activity of
histone acetyltransferase (HAT) and HDAC (histone deacetylase) {Li, 2019, 5387402}. In
female offspring of pregnant Kunming mice treated with PFOA by oral gavage at doses between
0 and 10 mg/kg/day on GD 1-17, the overall levels of histone H3 and H4 acetylation were
decreased in a dose-dependent manner in liver tissues in the pups at post-natal day 21. Histone
acetylase (HAT) activity was reduced in pups at all doses except for the highest dose (10
mg/kg/day), in which there was no significant difference in HAT activity compared to controls.
HDAC activity was increased in all dose groups. The changes in HAT and HDAC activity did
not follow a dose-responsive pattern. Notably, gene-specific alterations in histone acetylation
activity were not measured; thus, follow-up studies are needed to clarify the relationship between
the global histone modifications and the gene expression changes.
Additional support for species-specificity derives from studies demonstrating that PFOA-
mediated gene expression changes were distinctly different in primary human hepatocytes
compared to primary mouse hepatocytes {Rosen, 2013, 2919147}. Custom Taqman PCR arrays
were generated to include transcripts regulated by PPARa as well as transcripts regulated
independently of this nuclear receptor. Mouse and human hepatocytes were exposed to PFOA at
doses ranging from 0-100 and 0-200 [xM, respectively, or the PPARa activator Wyl4,643. In
mouse cells, many fewer genes were altered by PFOA treatment compared to whole livers from
mice exposed in vivo. Also, genes typically regulated by PPARa agonists were not altered by
PFOA in mouse cells, including Acoxl, Mel, Acaala, Hmgcsl, and Slc27al. The CAR target
gene Cyp2bl0 was also unchanged in cultured mouse hepatocytes. In contrast, a larger group of
genes were differentially expressed in primary human hepatocytes, including PPARa-
independent genes (CYP2B6, CYP3 A4, and PPARy). These findings underscore some of the
difficulty in extrapolating in vitro results from rodents to humans after PFOA exposure and
suggest PPARa may elicit species-specific changes in gene expression.
3.4.1.3.1.5 Conclusions
Although activation of PPARa is a widely cited mechanism of liver toxicity induced by PFAS
exposure, PFOA has been shown to activate a number of other nuclear receptors, including
PPARy, CAR/PXR, Era, and HNF4a. Many of these nuclear receptors, including CAR and
PPARy, are also known to play an important role in liver homeostasis and have been implicated
in liver dysfunction, including steatosis {Armstrong, 2019, 6956799}. Therefore, there is
accumulating evidence that PFOA exposure may lead to liver toxicity through the activation of
multiple nuclear receptors in both rodents and humans. However, the contribution of gene
expression changes induced and associated toxicity by these other receptors is not clear. Also, it
is possible that other receptors may play compensatory roles in PPARa null mice. In addition,
PFOA-mediated changes in hepatic gene expression and toxicity exhibit strain, sex, and species
specificity. Thus, the interplay between nuclear receptor activation and host factors may dictate
the nature and severity of liver toxicity in response to PFOA exposure.
3.4.1.3.2 Lipid Metabolism, Transport, and Storage
3.4.1.3.2.1 Introduction
The liver is the prime driver of lipid metabolism, transport, and storage within an organism. It is
responsible for the absorption, packaging, and secretion of lipids and lipoproteins. Lipids are
absorbed from digestion through biliary synthesis and secretion, where they are converted to
3-57
-------
DRAFT FOR PUBLIC COMMENT
March 2023
fatty acids {Trefts, 2017, 10284972}. These fatty acids are then transported into hepatocytes,
cells that make up roughly 80% of the liver mass, via a variety of transport proteins such as
CD36, FATP2, and FATP5 {Lehner, 2016, 10284974}. Fatty acids can be converted to
triglycerides, which can be packaged with high or very-low-density lipoproteins (HDL or
VLDL) for secretion. Lipid handling for the liver is important for energy metabolism (e.g., fatty
acid P-oxidation) in other organs and for the absorption of lipid-soluble vitamins {Huang, 2011,
10284973}. De novo cholesterol synthesis is another vital function of the liver. Cholesterol is
important for the assembly and maintenance of plasma membranes. Dysregulation of any of
these functions of the liver can have implications for metabolic and homeostatic processes within
the liver itself and other organs, and can contribute to the development of diseases such as non-
alcoholic fatty liver disease, steatosis, hepatomegaly, and obesity.
PFOA accumulates in liver tissue, and as such, not only influences lipid levels but can also alter
gene expression for a variety of pathways involved in biological processes {U.S. EPA, 2016,
3603279}. PFAS have been shown to induce steatosis and increase hepatic triglyceride levels in
rodents via inducing changes in genes directly involved with fatty acid and triglyceride synthesis
that may have variable effects on serum triglyceride levels depending on species, sex, and
exposure conditions {Das, 2017, 3859817; Rosen, 2013, 2919147; Rosen, 2017, 3859803; Li,
2019, 5387402; Beggs, 2016, 3981474; Liang, 2019, 5412467}. These include genes such as
fatty acid binding protein 1 (Fabpl), sterol regulatory element binding protein 1 (Srebpl), VLDL
receptor (Vldlr), and lipoprotein lipase (Lpll) {Armstrong, 2019, 6956799}. Various studies
have also shown that PFOA alters expression of genes directly involved in cholesterol
biosynthesis {Pouwer, 2019, 5080587; Das, 2017, 3859817; Rosen, 2017, 3859803; Li, 2019,
5387402} and in P-oxidation of fatty acids (e.g., Acoxl and/or carnitine palmitoyltransferase 1A
(Cptla)) {Lee, 2020, 6323794; NTP, 2019, 5400977; Cavallini, 2017, 3981367; Li, 2019,
5387402; Rosen, 2013, 2919147; Schlezinger, 2020, 6833593}. Genes involved in lipid
metabolism and homeostasis can be altered through PPARa, PPARy, CAR, and HNF4a
induction pathways and are dose-, life stage-, species-, and sometimes sex-dependent.
3.4.1.3.2.2 In Vivo Models
3.4.1.3.2.2.1 Rats
Two studies conducted in Sprague Dawley rats reported marked effects on lipid metabolism,
including sex-dependent effects, of PFOA on hepatic outcomes {NTP, 2019, 5400977; Cavallini,
2017, 3981367}.
The study conducted by NTP in 2019 {NTP, 2019, 5400977} used an oral dosing paradigm of 0,
0.625, 1.25, 2.5, 5, or 10 mg/kg (males) or 0, 6.25, 12.5, 25, 50, or 100 mg/kg/day (females) for
28 days. Males exhibited higher plasma levels of PFOA despite receiving a 10-fold lower dose
across the dose groups.
Serum cholesterol levels were decreased in PFOA exposed males and females, whereas serum
triglyceride levels were decreased in males but increased in females. In liver, PPARa- and CAR-
induced genes including Acoxl, Cyp4al, Cyp2bl, and Cyp2b2 were upregulated in both males
and females compared to controls. In females, the CAR-induced Cyp2bl and Cyp2b2 exhibited a
greater increase than that of Acoxl and Cyp4al, whereas Cyp4al and Cyp2bl exhibited the
greatest fold increase in males. Acoxl was more strongly upregulated in males than females.
This gene expression profile indicates a stronger PPARa signal in males relative to females, and
3-58
-------
DRAFT FOR PUBLIC COMMENT
March 2023
stronger CAR activation signal in females. Bile acid concentrations were increased at the two
highest dose groups (5 and 10 mg/kg/day) in males, but were not measured in females.
PFOA is known to activate PPAR receptors and proliferation of peroxisomes, and increase
expression of Acyl-CoA oxidase (ACOX) activity, the first enzyme in the fatty acid beta
oxidation pathway. In one study, a single dose of PFOA (150 mg/kg) in male Sprague-Dawley 2-
month-old rats caused increased liver weight associated with an eight-fold and a fifteen-fold
increase in ACOX after 2 and 4 days, respectively {Cavallini, 2017, 3981367}. PFOA exposure
was associated with generation of new, ACOX rich peroxisomes. Autophagy was induced in
fasted rats by an injection of an antilipolytic agent (3,5-dimethyl pyrazole (DMP)). In PFOA-
treated rats, DMP-induced autophagy delayed the decrease in ACOX activity relative to controls.
The authors hypothesized that autophagy may preferentially target older peroxisomes for
degradation. However, another possibility not considered by the authors is that PFOA could
disrupt drug-induced autophagy, which may represent an interesting area for further research.
3.4.1.3.2.2.2 Mice
Several studies were conducted to investigate the effects of PFOA on lipid accumulation in
hepatocytes by histopathological and metabolomic methods using mice of different genetic
backgrounds and life stages, and mice genetically modified to mimic human lipid metabolism
{Wang, 2013, 2850952; Pouwer, 2019, 5080587; Hui, 2017, 3981345; Rebholz, 2016, 3981499;
van Esterik, 2015, 2850288}. Other studies focused on the transcription and translation of genes
involved in lipid metabolism and biliary pathways. The focus of these studies was to identify key
genes, gene products, and transcriptional regulators affected by PFOA exposure and to examine
how PFOA alters metabolism of lipids {Zhang, 2020, 6833704; Das, 2017, 3859817; Rosen,
2017, 3859803; Li, 2019 5387402; Beggs, 2016, 3981474; Yan, 2015, 2851199; Yu, 2016,
3981487; Song, 2016, 9959776; Wu, 2018, 4238318}.
3.4.1.3.2.2.2.1 Changes in hepatic lipid homeostasis
Many biochemical changes occurred with lipids and bile within the liver as well as lipid
transport out of the liver (serum/plasma values). In several mouse studies, PFOA increased
hepatic lipid levels including triglycerides, total cholesterol, and LDL, which correlated with
histopathological changes that are often consistent with steatosis.
In Das et al. (2017, 3859817), WT male SV129 mice administered 10 mg/kg/day PFOA for 7
days had increased lipid accumulation in liver, as seen by Oil Red O staining, as well as
increased liver triglyceride levels. These effects were mainly attributed to activation of PPARa,
as they were attenuated in PFOA-exposed PPARa null mice (Section 3.4.1.2). In contrast, in
male BALB/c mice administered 0.08, 0.31, 1.25, 5, or 20 mg/kg/day PFOA for 28 days, liver
cholesterol was significantly decreased at 0.31 mg/kg/day and above, while triglycerides were
significantly decreased at 0.08 and 20 mg/kg/day and significantly increased at 1.25 mg/kg/day
(no changes were seen at other concentrations) {Yan, 2015, 2851199}. An increase in the
transcriptional activity of PPARa and sterol regulatory element-binding proteins (SREBPs) was
also observed. The authors hypothesize that altered lipid metabolism is induced by PPARa
activation, with increased SREBP activity as a mediator in this pathway.
One study evaluated PFOA effects on storage in hepatic lipid droplets (LDs) in BALB/c mice
{Wang, 2013, 2850952}. LDs are storage structures for neutral lipids that form in the
3-59
-------
DRAFT FOR PUBLIC COMMENT
March 2023
endoplasmic reticulum and release into the cytoplasm. In addition to lipid storage, they influence
lipid metabolism, signal transduction, intracellular lipid trafficking, and protein degradation.
Four-week-old BALB/c mice fed either regular or HFD were dosed with 5, 10, or 20 mg/kg/day
PFOA by gavage for 14 days. Cytoplasmic LDs were apparent in both regular- and HFD-fed
mice, though more were observed in HFD-fed mice. However, in PFOA-exposed mice, LDs
transferred from the cytoplasm to the nucleus, forming hepatocyte intranuclear inclusions in a
dose-dependent manner. The authors suggest that this translocation of LDs to the nucleus is a
critical factor in PFOA-mediated liver toxicity. As discussed below (Section 3.4.1.3.2.2.2.2), at
least two genes involved in lipid droplet formation, PLIN2 and PLIN4, were increased in PFOA-
exposed HepaRG cells in vitro, supporting a role for PFOA in altering lipid droplets in
hepatocytes {Louisse, 2020, 6833626}.
A targeted metabolomics approach was used to directly identify alterations in 278 metabolites in
livers of BALB/c mice exposed to either 0.5 or 2.5 mg/kg/day PFOA for 28 days by gavage {Yu,
2016, 3981487}. A total of 274 of these metabolites were identified in liver and were mapped to
KEGG metabolic pathways including amino acid, lipid, carbohydrate, and energy metabolism. In
liver, nine metabolites mapped to lipid metabolism as evidenced by alterations in the relative
concentrations of acylcarnitines, sphingomyelins, phosphatidylcholines, and oxidized
polyunsaturated fatty acids. Among the 18 liver metabolites that were significantly different
between exposed and control mice were six acylcarnitines, one phosphatidylcholine, and two
polyunsaturated fatty acids, which could serve as potential biomarkers of PFOA exposure. The
altered lipid profiles are consistent with the finding that PFOA upregulates hepatic nuclear
receptors and their target genes directly involved in lipid metabolism and the P-oxidation of fatty
acids {Lee, 2020, 6323794}. The profile of both phosphatidylcholine and fatty acid metabolites
indicated a PFOA-mediated shift to phosphatidylcholines with more carbons and more double
bonds. Because a change to fatty acids with more carbon atoms and double bonds is due to
biosynthesis reactions of saturated and unsaturated fatty acids, these findings suggest PFOA
exposure may stimulate fatty acid biosynthesis, which may account for the altered profile of both
phosphatidylcholines and fatty acids in liver. Thus, PFOA may regulate both catabolic and
anabolic lipid metabolism in liver.
3.4.1.3.2.2.2.2 Gene expression and metabolite accumulation impacting lipid homeostasis
Several studies probed the genes and pathways by which PFOA alters hepatic lipid homeostasis.
Hui et al. {2017, 3981345} demonstrated that the expression of genes and proteins associated
with lipid storage in was altered in the liver of PFOA-exposed BALB/c mice. Male mice were
exposed to 1 or 5 mg/kg/day for 7 days and the expression of lipid metabolism genes was
analyzed. Triglyceride and free fatty acid contents in serum were reduced, while hepatic
triglyceride levels were increased in the PFOA-exposed mice compared to controls. In liver,
transcript levels of hepatic lipoprotein lipase (Lpl) and fatty acid translocase (Cd36) were
elevated, while apolipoprotein-BlOO (ApoB) expression was diminished. LPL and CD36
regulate lipid intake through lipid hydrolysis and transport of lipids from blood to liver, whereas
APOB is required for lipid export from liver. Protein levels aligned with the changes in transcript
levels for these genes. The authors suggest that dysregulation of lipid metabolism and,
specifically, fatty acid trafficking, leads to decreased body weights and lipid malnutrition and
deposition of lipids in liver. These findings are consistent with observations in male Kunming
mice exposed to 5 mg/kg/day PFOA for 21 days {Wu, 2018,4238318}. In these mice, PFOA
exposure led to reduced APOB and elevated CD36 protein levels as measured
3-60
-------
DRAFT FOR PUBLIC COMMENT
March 2023
immunohistochemically and correlated to increased liver triglyceride levels.In addition to genes
directly involved in regulating lipid metabolism and storage, Eldasher et al. (2013, 2850979)
demonstrated that Bcrp mRNA and protein are increased in the livers, but not the kidneys of
male C57BL/6 mice exposed to 1 or 3 mg/kg/day PFOA by gavage for 7 days. BCRP is an ATP-
binding cassette efflux transporter protein involved in active transport of various nutrients and
drugs and implicated in transport of xenobiotics. In addition, BCRP can function sterol transport
and its ATPase activity can be stimulated with cholesterol {Neumann, 2017, 10365731}. Further
studies are needed to elucidate the role of BCRP or other transport proteins in PFOA-mediated
disruption of lipid metabolism.
MicroRNAs (miRNAs or miRs) are also altered after exposure to PFOA in mice in a dose-
dependent manner. In serum of male BALB/c mice, 24 and 73 circulating miRNAs were altered
in mice exposed to 1.25 and 5 mg/kg/day PFOA, respectively, for 28 days {Yan, 2014,
2850901}. Changes in expression of six miRNAs (miR-28-5p, miR-32-5p, miR-34a-5p, miR-
200c-3p, miR-122-5p, miR-192-5p) were confirmed in liver, including two (miR-122-5p and
miR-192-5p) considered to be biomarkers for drug-induced liver injury. MiRNAs may play a
specific role in regulating expression of genes involved in lipid metabolism and storage.
Cui et al. (2019, 5080384) observed that PFOA exposure (5 mg/kg/day PFOA for 28 d) led to a
significant increase of miR-34a, but not miR-34b or miR-34c, in the livers of male BALB/c
mice, consistent with the findings of Yan et al. {Yan, 2014, 2850901}.
Liver toxicity was evaluated by Cui et al. (2019, 5080384) by measuring liver weight, elevated
liver enzymes, and hepatic cell swelling manifested in both WT mice and in miR-34a-null mice
generated on a C57BL/6J background. RNASeq analysis of hepatic tissue showed that
expression of lipid metabolism genes was significantly altered in both WT mice and in miR-34a-
null mice after PFOA exposure; however, fewer genes were altered in livers of miR-34a-null
mice. Metabolism genes dominated those changed by miR-34a, including Fabp3, Cyp7al, and
Apoa4. Based on the transcriptome analysis, the authors found that miR-34a mainly exerts a
metabolic regulation role, rather than the pro-apoptosis and cell cycle arrest role reported
previously in vitro.
In addition to perturbed expression of genes as a consequence of activating PPARa and other
nuclear receptors, PFOA may directly target enzymes involved in fatty acid metabolism. Shao et
al. (2018, 5079651) postulated that based on the electrophilic properties of PFOA, it may
preferentially bind to proteins harboring reactive cysteine residues. To test this hypothesis,
proteomic and metabolomic approaches were applied. Two cysteine-targeting probes were used
to enrich putative target proteins in mouse liver extracts in the absence or presence of PFOA,
resulting in the identification of AC AC A and ACACB as novel target proteins of PFOA. Parallel
reaction monitoring (PRM)-based targeted proteomics combined with thermal shift assay-based
chemical proteomics was used to verify AC AC A and ACACB as PFOA binding targets. Next,
the authors used a metabolomic approach to analyze liver extracts from female C57BL/6 mice
four hours after IP injection with a very high dose (300 mg/kg) of PFOA to confirm abnormal
fatty acid metabolism, including significantly elevated levels of carnitine and acyl-carnitines.
ACACA and ACACB are acetyl-CoA carboxylases that can regulate fatty acid biosynthesis. The
authors suggest PFOA interactions with these carboxylases leads to a downregulation of
malonyl-CoA, required for the rate-limiting step of fatty acid biosynthesis and an inhibitor of
carnitine palmitoyl transferase 1 (Cptl). Despite the correlation to altered fatty acid profiles,
3-61
-------
DRAFT FOR PUBLIC COMMENT
March 2023
additional studies are required to confirm PFOA binding to these lipid enzyme targets and
changes in hepatic fatty acid metabolism.
3.4.1.3.2.2.2.3 Host factors influencing lipid metabolism and storage
Rebholz et al. (2016, 3981499) underscored the relevance of genetic background, sex, and diet in
PFOA-mediated alterations of hepatic gene expression and highlighted the role of genes involved
in sterol metabolism and bile acid production Young, sexually immature male and female
C57BL/6 and BALB/c mice were placed on diets to target a dose of approximately 0.56
mg/kg/day of PFOA and supplemented with 0.25% cholesterol and 32% fat.
Hypercholesterolemia developed in male and female C57BL/6 mice exposed to PFOA.
Hypercholesterolemia was also observed in male BALB/c mice but to a lesser degree than
C57BL/6, and did not manifest in female BALB/c mice. The PFOA-induced
hypercholesterolemia appeared to be the result of increased liver masses and altered expression
of genes associated with hepatic sterol output, specifically bile acid production. These data
support genetic background and dietary levels of fat and cholesterol as important variables
influencing PFOA-mediated changes in cholesterol. However, an important caveat in this study
is that female mice in the control groups for both strains had higher than expected blood PFOA
levels.
PFOA-mediated changes in lipid levels may be programmed during early life exposure.
C57BL/6JxFVB hybrid mice were exposed during gestation and lactation via maternal feed {van
Esterik, 2015, 2850288} to seven doses of PFOA targeting 0.003-3 mg/kg/day. The dose range
was chosen to be at or below the NOAEL used for current risk assessment. Liver morphology
and serum lipids were analyzed at in the pups at 26 weeks (males) and 28 weeks (females) of
age. Histopathological changes, including microvesicular steatosis and nuclear dysmorphology,
were more frequent in PFOA-exposed mice compared to controls, though the incidence did not
reach statistical significance over the dose range. However, perinatal exposure induced a sex-
dependent change in lipid levels. In females only, serum cholesterol and triglycerides showed a
dose-dependent decrease with a maximum change of -20% for cholesterol and -27% for
triglycerides (BMDLs of 0.402 and 0.0062 mg/kg/day, respectively). The authors suggest that
perinatal exposure to PFOA in mice alters metabolic programming in adulthood. Based on the
sexually dimorphic lipid levels, as well as extrahepatic changes, females appear more sensitive to
PFOA-mediated alterations in metabolic programming.
The potential developmental effects of PFOA in liver are also of interest considering recent
findings that PFOA regulates expression of homeobox genes involved in both development and
carcinogenesis {Zhang, 2020, 6833704}. Adult male C57BL/6 mice, PPARa-null mice, or CAR-
null mice were given a single IP administration of 41.4 mg/kg and livers were collected on Day
5. PFOA induced mRNA expression of Hoxa5, b7, c5, dlO, Pdxl and Zeb2 in wild-type mice in
a manner dependent on PPARa and CAR. Whether exposure to PFOA alters homeobox genes
during perinatal exposure, and the potential for homeobox proteins to alter PFOA susceptibility
in different life stages remains to be determined.
One difference between human and rodent lipid metabolism relates to transfer of cholesterol
ester from HDL to the APOB-containing lipoproteins in exchange for triglycerides. Mice lack
cholesteryl ester transfer protein (CETP) and rapidly clear APOB-containing lipoproteins. In
contrast, a higher proportion of HDL relative to LDL is observed in humans and primates due to
the function of CETP. APOE*3-Leiden.CETP transgenic mice, a strain that expresses human
3-62
-------
DRAFT FOR PUBLIC COMMENT
March 2023
CETP, exhibit a more human-like lipoprotein metabolism with transfer of cholesterol ester from
HDL to the APOB-containing lipoproteins in exchange for triglycerides resulting in delayed
APOB clearance. Pouwer et al. (2019, 5080587) utilized these transgenic mice to evaluate the
effect of PFOA on plasma cholesterol and the mechanism for the hypolipidemic responses
observed with PFOA exposures. APOE*3-Leiden.CETP mice were fed a Western-type diet
(0.25% cholesterol (wt/wt), 1% corn oil (wt/wt), and 14% bovine fat (wt/wt)) with PFOA (0.01,
0.3, or 30 mg/kg/day) for 4-6 weeks. The doses were chosen to parallel environmental and
occupational exposures in humans. PFOA exposure did not alter plasma lipids at lower doses,
but did decrease plasma triglycerides, total cholesterol, and non-HDL levels, and increased HDL
levels. Overall, these findings mirrored a clinical trial in humans demonstrating PFOA-induced
decreases in cholesterol levels. This lipid profile could be attributed to decreased very low-
density lipoprotein (VLDL) production and increased VLDL clearance by the liver through
increased lipoprotein lipase activity. The concomitant increase in HDL was attributed to
decreased CETP activity subsequent to PPARa activation and the downregulation of hepatic
genes involved in lipid metabolism, including Apoal, Scarbl, and Lipc (genes involved in HDL
formation, HDL clearance, and HDL remodeling, respectively). Based on the lipid profiles, gene
expression analysis, and pathway analysis, the authors propose a mechanistic model in which
high PFOA exposure increases VLDL clearance by the liver through increased LPL-mediated
lipolytic activity. These changes lead to lower VLDL serum levels consistent with reduced
VLDL particle formation and secretion from the liver due to reduced ApoB transcript levels and
de novo synthesis.
To further explore mechanistic differences in PFOA-induced changes in lipid metabolism
between humans and mice, Schlezinger et al. (2020, 6833593) investigated PFOA-mediated lipid
dysregulation in mice expressing human PPARa (hPPARa) and compared results to PPARa-null
mice. Male and female mice were fed an American style diet (51.8% carbohydrate, 33.5% fat,
and 14.7%) protein, based on an analysis of what 2-to-19-year-old children and adolescents eat
using NHANES datal) and exposed to PFOA (8 |iM) in drinking water for 6 weeks that led to
serum PFOA levels of 48 [j,g/mL. Both hPPARa-null and PPARa-null mice developed
hepatosteatosis after PFOA exposure. Changes in gene expression and increased serum
cholesterol that was more pronounced in males than females correlated with changes in
expression of genes that regulate cholesterol homeostasis. PFOA decreased expression of Hmgcr
in a PPARa-dependent manner. Ldlr and Cyp7al were also decreased but in a PPARa-
independent manner. Apob expression was not changed. While many of the target genes
analyzed were similarly regulated in both sexes, some sex-specific changes were observed.
PFOA induced PPARa target genes in livers of both sexes including Acoxl (involved in fatty
acid P-oxidation), Adrp (involved in coating lipid droplets), and Mogatl (involved in
diacylglyerol biosynthesis). PPARy target genes were also upregulated in both sexes and
included Fabp4 and Cd36 that contribute to lipid storage and transport as was the CAR target
gene Cyp2bl0. PFOA exposure decreased expression of Cyp7al required for conversion of
cholesterol to bile acids and efflux, but more so in females than in males.
Sex-specific changes in hepatic gene expression in response to PFOA exposure was also
observed in zebrafish {Hagenaars, 2013, 2850980}. Adult zebrafish were exposed to 0.1, 0.5, or
1 mg/L PFOA for 28 days. Livers were harvested and subjected to transcriptomic analysis.
Similar to observations in mice, expression of genes regulating fatty acid metabolism and
cholesterol metabolism and transport were generally upregulated in males and suppressed in
3-63
-------
DRAFT FOR PUBLIC COMMENT
March 2023
females. Thus, sex-specific effects of PFOA on fatty acid and cholesterol metabolism is observed
across different vertebrate species, but also exhibits species specificity. For example, genes in the
cytochrome P450 family involved in cholesterol metabolism and transport were suppressed in
female zebrafish but upregulated in male zebrafish {Hagenaars, 2013, 2850980}. However,
Cyp2b genes downstream of CAR (e.g., Cyp2bl and Cyb2bl0) were more strongly upregulated
in females compared to males in both rats and mice {Schlezinger, 2020, 6833593; NTP, 2019,
5400977}. Differences in expression of Cyp450 genes may in part relate to species-specific
activity of nuclear receptors, and the fact that no CAR orthologues have been identified in
zebrafish nor any other fish species {Schaaf, 2017, 10365760}.
3.4.1.3.2.2.3 In Vitro Studies
In vitro studies reported genetic profiles and pathway analyses in mouse and human hepatocytes
to determine the effect of PFOA treatment on lipid homeostasis and bile synthesis. Six studies
investigated the effect of PFOA on lipid homeostasis using primary hepatocytes and human cell
lines such as HepG2, HepaRG, and HL-7702 cells. Various endpoints were also investigated in
these cell lines such as mRNA expression through microarray and qRT-PCR assays; lipid,
triglyceride, cholesterol, and choline content; and protein levels via ELISA or western blot. In
addition, two studies evaluated PFOA-mediated changes to lipids using metabolomic
approaches.
Franco et al. (2020, 6507465) exposed HepaRG cells to PFOA and PFOS and evaluated
metabolomics at a dose range of 100 pM to 1 [xM. The highest PFOA exposure levels (10-100
|iM) were associated with significant increases in total lipid concentrations, especially at the
three highest concentrations tested (10, 100, and 1,000 nM). Interestingly, hepatocyte lipids were
decreased in response to increasing PFOS exposure in this system. The affected classes of lipids
also diverged, with PFOA associated with increased diglycerides, triglycerides, and
phosphatidylcholines, whereas PFOS was associated with decreased diglycerides, ceramides, and
lysophosphatidylcholines. Staining of neutral lipids was also prominent in PFOA-treated
hepatocytes, suggesting an obesogenic role PFOA that may directly impact hepatic steatosis. The
authors further hypothesized that the concentration-dependent decrease in lipid accumulation
associated with PFOS may be related to differential ability of these compounds to interact with
PPARs, including PPARy.
Peng et al. (2013, 2850948) evaluated disturbances of lipids in the human liver cell line L-02
using metabolomic and transcriptomic approaches. Specifically, PFOA exposure was associated
with altered mitochondrial metabolism of carnitine to acylcarnitines. The effect was dose-
dependent and correlated with altered expression levels of key genes involved in this pathway.
Downstream of this pathway, cholesterol biosynthesis was upregulated as measured by both
increased cholesterol content and elevated expression levels of key genes. The profile of PFOA-
associated disturbance in lipid metabolism was consistent with initial changes in fatty acid
catabolism in cytosol that altered mitochondrial carnitine metabolism, ultimately impacting
cholesterol biosynthesis.
In contrast to the findings of Peng et al. (2013, 2850948) in L-02 cells, Das et al. (2017,
3859817) reported that PFOA did not inhibit palmitate-supported respiration (mitochondrial
metabolism) in HepaRG cells. There was no effect on oxidation or translocation of
palmitoylcarnitine, an ester involved the in metabolism of fatty acids, as part of the tricarboxylic
3-64
-------
DRAFT FOR PUBLIC COMMENT
March 2023
acid (TCA) cycle in the mitochondrial fraction. This may indicate less of a perturbation to fatty
acid metabolism in this cell line. This suggests that intermediary steps in fatty acid activation,
transport, and/or oxidation are affected. The authors suggest that PFOA effects on mitochondrial
synthesis of fatty acid and other lipids are secondary and possibly compensatory to any
mitochondrial-induced toxicity, rather than as the result of activation of peroxisomes, which are
mediated by PPARs.
Rosen et al. (2013, 2919147) exposed mouse and human primary hepatocytes to 0-100 or 0-200
|iM PFOA, respectively. Gene expression was evaluated using microarrays and qRT-PCR. For
PFOA-exposed murine hepatocytes, a much smaller group of genes was found to be altered
compared to the whole liver. These genes included those associated with P-oxidation and fatty
acid synthesis such as Ehhadh and Fabpl, which are upregulated by PFOA. In contrast to the
transcriptome of primary mouse hepatocytes, a large group of genes related to lipid metabolism
was differentially expressed in primary human hepatocytes including perilipin 2 (PLIN2) and
CYPTA1, which were upregulated at 100 |iM PFOA. The authors attribute some of these
differences between mouse and human hepatocytes to a less robust activation of PPARa in
humans. Further, many of the genes investigated were chosen to explore effects of PFOS
exposure that are independent of PPARa activation but may include other nuclear receptors such
as CAR, LXR, PXR, and AhR (Section 3.4.1.3.1). Beggs et al. (2016, 3981474) exposed human
primary hepatocytes to 0.01-10 |iM PFOA for 48 or 96 hours to determine pathways affected by
PFOA exposure. PFOA treatment altered 40 genes (20 upregulated and 20 downregulated).
Upregulated genes were primarily associated with lipid metabolism, hepatic steatosis and
cholestasis, and liver hyperplasia. Among the top 10 upregulated genes were PLIN2, CYP4A22,
and apolipoprotein A4 (APOA4).
Differential regulation of lipid metabolism and storage genes was also observed in HepG2 cells
exposed to PFOA (dose range of 20-200 pM) for 48 hours {Wen, 2020, 6302274}. Some
specific metabolic pathway genes were not altered, including genes encoding the acyl-CoA
dehydrogenase enzyme. FABP1, which encodes for a key protein responsible for fatty acid
uptake, transport, and metabolism, exhibited decreased expression. Acyl-CoA oxidase 2
(ACOX2), which is involved in the peroxisome-mediated degradation of fatty acids, was also
decreased. In contrast, a number of genes involved in fatty acid anabolism were upregulated. The
authors linked PFOA-mediated gene expression changes to diminished global methylation,
implicating epigenetic factors in PFOA-mediate changes in gene expression.
In human hepatic cell lines such as HepaRG, PFOA treatment led to downregulation of genes
involved in cholesterol homeostasis. Louisse et al. (2020, 6833626) noted a concentration-
dependent increase in triglycerides, a decrease of cholesterol at a high dose, and a
downregulation of cholesterogenic genes especially after 24 hours of exposure to the high dose
of 200 |iM PFOA in HepaRG cells. Cellular cholesterol biosynthesis genes are regulated by
SREBPs, which were also downregulated with PFOA exposure. In contrast, PPARa-responsive
genes were upregulated with PFOA exposure, particularly at higher doses. Behr et al. (2020,
6505973) also exposed HepaRG cells to 0-500 |iM PFOA for 24 or 48 hours. Similar to the
results from Louisse et al. (2020, 6833626), at 24 hours, genes related to cholesterol synthesis
and transport were downregulated at the highest dose except for several genes that were
upregulated, including bile and cholesterol efflux transporters (SLC51B and ABCG1), and genes
involved in bile acid and bilirubin detoxification (CYP3A4, UGT1 Al). The gene profiles after
3-65
-------
DRAFT FOR PUBLIC COMMENT
March 2023
48 hours of exposure were similar, except at the high dose, at which there was an attenuation of
the response in cholesterol synthesis and transport. Cholesterol content was significantly higher
in the supernatant at the highest dose of 500 |iM but there was no significant difference after 48
hours between treated cells and controls, which aligns with the attenuation of gene expression
changes. Both studies also observed a PFOA-associated decrease in CYP7A1, a key enzyme
involved in the initial step of cholesterol catabolism and bile acid synthesis.
3.4.1.3.2.2.4 Conclusions
Despite some inconsistencies in the literature, an emerging picture of PFOA-related dyslipidemia
is largely initiated by activation of nuclear receptors targeted by PFOA, primarily PPARa,
PPARy, and CAR. A primary consequence of this interaction is altered expression of genes
regulating hepatic lipid homeostasis. Gene expression profiles of lipid metabolism genes were
observed both in vivo and in vitro, and in a diverse set of study designs. While changes in gene
expression were consistently observed, the magnitude of the changes varied according to dose,
dose duration, and model system. PPARa appears to be the primary driver regulating gene
expression. However, studies in PPARa-null mice and analysis of nuclear receptor-specific
genes implicate PPARy, CAR, and possibly PPAR8 as important contributors to the changes in
PFOA-mediated gene expression. It should be noted, however, that a thorough analysis of
potential compensatory changes in gene knock-out mice was not discussed in the literature
reviewed here.
Two of the primary pathways targeted by PFOA-induced changes in gene expression include
metabolism of fatty acids leading to triglyceride synthesis and metabolism of cholesterol and bile
acids. In both mice and rats, gene expression changes generally correlated with increased
triglyceride levels in liver, and decreased levels of circulating serum triglycerides. For
cholesterol, in vitro studies were conflicting but suggest hepatic cholesterol content generally
increases in PFOA-exposed animals. However, serum cholesterol levels were reduced in rats but
were generally elevated in mice. Hepatic changes in lipid-regulating gene expression appear to
influence circulating levels of lipids in serum in a manner that varies by sex, species, and life
stage. For example, adult male rats exhibited decreases in serum triglycerides, whereas adult
female rats exhibited increases {NTP, 2019, 5400977}. However, in mice exposed perinatally
and then examined in adulthood, females, but not males, exhibited decreased serum levels of
triglycerides, a treatment effect that was not observed in males {van Esterik, 2015, 2850288}.
Male Kunming mice also exhibited a dose-dependent decrease in serum triglycerides and an
increase in liver triglycerides {Wu, 2018, 4238318}. For cholesterol, serum levels were
decreased in PFOA exposed male rats and increased in female rats {NTP, 2019, 5400977}. In
contrast, young male and female C57BL/6 mice exhibited hypercholesterolemia after PFOA
exposure, though this was less striking male among BALB/c mice and did not manifest in female
BALB/c mice {Rebholz, 2016, 3981499}. Elevated serum cholesterol was also more pronounced
in males than females in mice expressing human PPARa {Schlezinger, 2020, 6833593}.
Importantly, changes in gene expression and lipid content in liver ultimately manifest in altered
hepatocyte morphology. Most strikingly and consistently, steatosis manifests in PFOA-exposed
animals. Other pathogenetic changes associated with PFOA included hepatomegaly, cholestasis,
hyperplasia, and carcinoma. The finding of steatosis is interesting in light of observation that
PFOA exposure downregulates expression of HNF4a in liver with concomitant changes in
3-66
-------
DRAFT FOR PUBLIC COMMENT
March 2023
HNF4a target genes because HNF4a-deficient mice develop steatosis in the absence of exposure
to toxicants.
While the precise events that lead to steatosis have yet to be elucidated, the current studies
conducted in animals and in vitro studies supports the following key molecular and cellular
events related to PFOA-mediated hepatoxicity specific to changes in lipid metabolism: (1) PFOA
accumulation in liver activates nuclear receptors; (2) nuclear receptors, including PPARa, then
alter expression of genes involved in lipid homeostasis and metabolism; (3) the products of the
genes altered by activated nuclear receptors modify the lipid content of liver to favor triglyceride
accumulation, and possibly also cholesterol accumulation; (4) altered lipid content in liver leads
to accumulation of lipid droplets promoting development of steatosis and other changes leading
to liver dysfunction; and (5) alterations in lipid metabolism leads to alterations in serum levels of
triglycerides and cholesterol. An intriguing possibility that may be concurrent to these events is
direct binding of PFOA to ACACA and ACACB enzymes in a manner that interferes with fatty
acid biosynthesis. Although this series of events is plausible, significant gaps remain in
understanding this process, including how these events interface with other cellular processes
such as cell growth and survival, oxidative stress, and others in understanding the mechanisms of
PFOA-mediated hepatoxicity.
There are challenges in the extrapolation of results from research related to PFOA-mediated
changes to lipid metabolism in animals to humans. As presented in the 2016 PFOA HESD {U.S.
EPA, 2016, 3603279}, serum lipid levels were variably altered in humans exposed to PFOA in
their environments. In occupationally exposed humans and humans exposed to high levels of
PFOA, there was a general association with increased serum total cholesterol and LDL, but not
HDL. At least one obstacle to extrapolating from rodent to humans is that the cholesteryl ester
transfer protein encoded by the CETP gene in humans is absent in rodents. Mice lack CETP and
rapidly clear apoB-containing lipoproteins. In contrast, a higher proportion of HDL relative to
LDL is observed in humans and primates due to the function of CETP. New models designed to
develop mice that are "humanized" for lipid metabolism, including APOE*3-Leiden.CETP
{Pouwer, 2019, 5080587}, and mice expressing human nuclear receptors {Schlezinger, 2020,
6833593}, are likely to accelerate the extrapolation of mechanistic information from animals to
humans.
3.4.1.3.3 Hormone Function and Response
While much of the literature relevant to hormone function and response is focused on
reproductive or endocrine outcomes (see PFOA Appendix), recent literature has also shown a
relationship between hepatic hormonal effects and PFOA exposure. PFOA has been found to
affect thyroid mechanisms in hepatic cells. Huang et al. (2013, 2850934) studied the effect of 5,
10, 25, or 50 mg/L PFOA in a human non-tumor hepatic cell line (L-02 cells) and found that
PFOA exposure downregulated thyroid hormone binding protein precursor.
While there are a small number of studies regarding hormone function and response specifically
within the liver, there is evidence that PFOA has the potential to perturb hormonal balance in
hepatic cells, particularly in regard to thyroid function. This could have implications for hormone
function and responses in other organ systems and may also be important for MOA
considerations for hepatotoxicity.
3-67
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.1.3.4 Xenobiotic Metabolism
Xenobiotic metabolism is the detoxification and elimination of endogenous and exogenous
chemicals via enzymes (i.e., cytochrome P450 (CYP) enzymes) and transporters (i.e., organic
anion transporting peptides [OATPs]) {Lee, 2011, 3114850}. As described in Section 3.3.1.3,
the available evidence demonstrates that PFOA is not metabolized in humans or other species.
However, several studies have investigated how PFOA could alter xenobiotic metabolism in the
liver by downregulating or upregulating the gene expression of enzymes and transporters.
Li et al. (2017, 3981403) summarized the literature on molecular mechanisms of PFOA-induced
toxicity in animals and humans. The authors noted how Elcombe et al. (2007, 5085376) and
Guruge et al. (2006, 1937270) reported PFOA activation of PXR/CAR and subsequent
manipulation of the expression of genes responsible for xenobiotic metabolism {Li, 2017,
3981403}. For instance, Cheng and Klaassen {2008, 2850410} concluded that PFOA induced
the gene expression of CYP2B10 in mice.
Overall, results from both in vivo and in vitro model systems suggest that genes responsible for
xenobiotic metabolism are upregulated as a result of PFOA exposure.
3.4.1.3.4.1 In Vivo Models
Three studies investigated xenobiotic metabolism endpoints in in vivo models with two using
mice {Li, 2019, 5080362; Wen, 2019, 5080582} and one using zebrafish {Jantzen, 2016,
3860109}.
Li et al. (2019, 5080362) examined 5-6-week-old male C57BL/6 mice administered PFOA (1
mg/kg/day) via oral gavage for 2, 8, or 16 weeks. CYP2B and CYP3A activity were assessed via
PROD and BQ assays as an indicator of CAR/PXR activity in the liver. As discussed in Section
3.4.1.3.1, the authors reported upregulation of Cyp2b and Cyp3a gene expression with
downstream effects to CAR/PXR activation and xenobiotic metabolism. Similarly, Wen et al.
(2019, 5080582) investigated CYP gene expression (including Cyplal, Cyp2bl0, and Cyp3all)
with a focus on the activation of the nuclear receptor PPARa and downstream alteration of
metabolism and excretion of xenobiotics. Adult, male wild-type C57BL/6NTac and PPARa-null
mice were administered PFOA (3 mg/kg/day) for 7 days {Wen, 2019, 5080582}. Expression of a
targeted list of genes, including Cyplal, Cyp2bl0, and Cyp3al 1, was quantified by qRT-PCR.
In PFOA-treated wild-type mice, gene expression of Cyplal and Cyp3al 1 were not significantly
changed. Conversely, in PFOA-treated PPARa-null mice, gene expression of Cyp2bl0 and
Cyp3al 1 were significantly altered compared to the wild-type mice (11-fold increase for
Cyp2bl0 and 1.7-fold increase for Cyp3al 1). Authors noted the differences between wild-type
and PPARa mice were consistent with a previous study {Corton, 2014, 2215399}.
One study examined the expression of four genes related to xenobiotic metabolism in zebrafish
{Jantzen, 2016, 3860109}. Zebrafish embryos (AB strain) were exposed to 2.0 |iM PFOA
dissolved in water from 3 to 120 hours post-fertilization (hpf) and evaluated 180 days post-
fertilization (dpf) at adult life stage for gene expression. Females and males both had significant
reductions in slcoldl expression; however, only males had significant reductions in slco2bl
expression {Jantzen, 2016, 3860109}. Jantzen et al. (2016, 3860109) noted that in their previous
study {Jantzen, 2016, 3860114}, PFOA exposure from 5 to 14 dpf resulted in significantly
increased slco2bl expression. Given the fluctuation in gene expression from short-term to long-
3-68
-------
DRAFT FOR PUBLIC COMMENT
March 2023
term, further studies with additional timepoints are needed to elucidate the effect of PFOA
exposure on OATPs expression.
3.4.1.3.4.2 In Vitro Models
CYP2B6 is expressed in the liver and is predominately responsible for xenobiotic metabolism;
similar to previous studies, Behr et al. (2020, 6305866) investigated activation of nuclear
receptors by PFAS. Authors exposed HEK293T cells and HepG2 cells to varying concentrations
of PFOA (0, 50, 100, or 250 |iM) for 24 hours. As discussed further in Section 3.4.1.3.1, the
authors reported the downstream effects of PFOA-mediated PPARa activation. At the highest
concentration of 250 |iM, Behr et al. (2020, 6305866) reported that PFOA significantly induced
gene expression of CYP2B6 by 11.2-fold. CYP2B6 gene expression was assessed in an
additional study that used primary human and mouse hepatocytes {Rosen, 2013, 2919147}. In
primary human hepatocytes, PFOA concentrations ranged between 0 and 200 |iM; in mouse
hepatocytes, concentrations ranged between 0 and 100 |iM. Results varied between human and
mouse hepatocytes, with CYP2B6 upregulated in human hepatocytes but not in mouse
hepatocytes. The authors noted that the differences between gene expression of the human and
mouse hepatocytes were unclear; however, cell density, collection methods, and time in culture
were possible factors.
Franco et al. (2020, 6315712) assessed the expression of genes encoding several phase I and II
biotransformation enzymes following exposure to PFOA concentrations (10-10, 10-9, 10-8, 10-7,
10-6 M) for 24 or 48 hours. Gene expression of phase I enzymes (CYP1A2, CYP2C19, and
CYP3A4) varied across concentrations and between the 24- and 48-hour exposures. For
CYP1A2, after 24 hours, expression was significantly upregulated at concentrations >10-9 M;
however, after 48 hours, expression was significantly downregulated at concentrations >10-8 M.
CYP2C19 was downregulated across all concentrations after both 24- and 48-hour exposures;
downregulation was significant for concentrations after both 24- and 48-hour exposures with the
exception of 10-8 M after 24-hours. The authors concluded that PFOA exposure can significantly
reduce expression of phase I biotransformation enzymes.
Evidence varied across studies for the effect of PFOA on the expression of CYP3A4, a phase I
enzyme involved in bile acid metabolism and detoxification by hydroxylation and xenobiotic
metabolism, depending on the model and duration of exposure, as well as whether gene
expression or enzyme activity was assessed {Behr, 2020, 6505973; Franco, 2020, 6315712;
Louisse, 2020, 6833626; Rosen, 2013, 2919147; Shan, 2013, 2850950}. Franco et al. (2020,
6315712) reported that after 24-hours, there were not significant changes in CYP3A4 expression.
However, after 48 hours, there was a five-fold reduction in the expression. Conversely, Behr et
al. (2020, 6505973) and Louisse et al. (2020, 6833626) reported upregulation of CYP3A4
enzyme activity following 24- or 48-hour PFOA exposure in HepaRG cells; specifically, Behr et
al. (2020, 6505973) reported significant upregulation at 50 and 100 |iM after both 24- and 48-
hour PFOA exposure.
Rosen et al. (2013, 2919147) also reported upregulation of CYP3A4 expression following PFOA
exposure (0-100 |iM) in human hepatocytes; however, significant changes were not reported for
mouse hepatocytes. Lastly, Shan et al. (2013, 2850950) reported no significant changes in
CYP3A4 enzyme activity following PFOA exposure (0, 100, 200, 300, or 400 |iM) in HepG2
cells.
3-69
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Franco et al. (2020, 6315712) also assessed gene expression of phase II enzymes, glutathione-s-
transferase mul (GST-MI) and UDP glucuronosyltransferase-1 Al (UGT-1 Al), which were not
significantly affected by exposure to PFOA after 24 or 48 hours. The authors noted that it was
unclear where and how PFOA alters gene expression of phase I enzymes and not phase II
enzymes. Further research is needed to determine whether altered gene expression occurs by
interference with cytoplasm receptors, inhibition of nuclear translocation, and/or inhibition of the
interaction of nuclear translocator complexes with DNA sequences {Franco, 2020, 6315712}.
Orbach et al. (2018, 5079788) focused on the gene expression of the CYP2E1 enzyme. PFOA
was added to primary human hepatocytes and primary rat hepatocytes at either V2 LC50 or LC50
(500 |iM for both humans and rats) for 24 hours. CYP2E1 enzymatic activity was estimated by
the conversion of 7-methoxy-4-trifluoromethylcoumarin (MFC) to 7-
hydroxytrifluoromethylcoumarin (HFC). However, in both human and rat hepatocytes, there
were no significant changes in CYP2E1 activity.
Song et al. (2016, 9959776) analyzed the expression of over 1,000 genes by expression
microarray analysis following exposure of HepG2 cells with increasing concentrations (0-1,000
|iM) of PFOA for 48h. As a result, 1,973 genes expressed >1.5-fold changes in the exposed
groups compared to the control group, including 20 genes responsible for metabolism of
xenobiotics by cytochrome P450.
3.4.1.3.4.3 Conclusions
Several studies are available that assessed xenobiotic metabolism endpoints as a response to
PFOA exposure, including studies in mice {Li, 2019, 5080362; Wen, 2019, 5080582}, zebrafish
{Jantzen, 2016, 3860109}, primary hepatocytes {Orbach, 2018, 5079788; Rosen, 2013,
2919147}, or hepatic cell lines {Behr, 2020, 6305866; Franco, 2020, 6315712; Louisse, 2020,
6833626; Shan, 2013, 2850950; Song, 2016, 9959776}. Jantzen et al. (2016, 3860109) reported
significant reductions in the expression of OATPs (slcoldl and slco2bl). While the majority of
studies reported altered gene expression of CYP enzymes, the direction and magnitude of change
varied across doses and exposure durations. Jantzen et al. (2016, 3860109) and Franco et al.
(2020, 6315712) both noted the need for further research to elucidate any potential relationships
between PFOA exposure and xenobiotic metabolism.
3.4.1.3.5 Cell Viability, Growth and Fate
3.4.1.3.5.1 Cytotoxicity
Several in vitro studies have examined the cytotoxic effect of PFOA on cell viability assays in
both primary hepatic cell cultures {Beggs, 2016, 3981474; Xu, 2019, 5381556} and in hepatic
cell lines {Wen, 2020, 6302274; Hu, 2014, 2325340; Rosenmai, 2018, 4220319; Shan, 2013,
2850950; Lv, 2019, 5080368; Yan, 2015, 2851199; Zhang, 2020, 6316915; Sheng, 2018,
4199441; Wiels0e, 2015, 2533367; Florentin, 2011, 2919235; Franco, 2020, 6315712; Ojo,
2020, 6333436; Franco, 2020, 6507465; Huang, 2014, 2851292; Cui, 2015, 3981517; Behr,
2020, 6505973; Song, 2016, 9959776}, with varying results depending on the exposure
concentration and duration, cell line, and culturing methods.
In mouse primary hepatocytes, cell viability as determined by cell counting Kit-8 (CCK-8) assay
did not significantly change at concentrations of PFOA in the range of 10-500 |iM; however, a
41% decrease in viability was observed after 24 hours of exposure to 1000 |iM PFOA {Xu,
3-70
-------
DRAFT FOR PUBLIC COMMENT
March 2023
2019, 5381556}. In primary rat hepatocytes exposed to PFOA for 24 hours showed no changes
in cell viability at concentrations <25 |iM, but cell viability was increased by approximately 16%
in the 100 |iM concentration {Liu, 2017, 3981337}.
PFOA exposure duration and concentration affect cytotoxicity. In HepG2 cells, 100 |iM PFOA
did not affect cell viability after 1-3 hours of exposure {Florentin, 2011, 2919235; Shan, 2013,
2850950}. However, after 72 hours, cell viability as determined by neutral red assay was reduced
by nearly 80% in the same cell line {Buhrke, 2013, 2325346}, suggesting that PFOA
cytotoxicity is increased with long-term exposure. Additionally, in human HEPG2 cells treated at
different concentrations of PFOA for 24 hours, viability as determined by MTT assay did not
change with 100 |iM PFOA, but was significantly reduced by 14% at 200 |iM, 22% at 400 |iM,
47%) at 600 |iM, and 69% at 800 |iM, suggesting a concentration-dependent reduction in cell
viability {Florentin, 2011, 6333436}. In contrast, cell viability dropped below 80% in HepaRG
cells exposed to 100 |iM PFOA at 24 hours {Franco 2020, 6315712}. Another study in HepaRG
cells {Louisse, 2020, 6833626} showed no effect on cell viability up to concentrations of 400
|iM for 24 hours. Although some results are conflicting, overall, these studies suggest that
exposure duration and concentration, type of cell lines, species, and viability assessment methods
are determinants of PFOA-induced cytotoxicity.
IC50 values in hepatic cell lines ranged from approximately 42 |iM PFOA after 72 hours
{Buhrke et al., 2013, 2325346}, 102-145 |iM after 24 hours {Ojo, 2020, 6333436; Franco, 2020,
6315712}, to 305 |iM after 48 hours of exposure in HepG2 cells {Song, 2016, 9959776}. In a
fetal liver cell line (HL-7702), IC50 values were 647 |iM after 24 hours exposure and 777 |iM
after 48 hours exposure {Hu, 2014, 2325340; Sheng, 2018, 4199441}. One study in zebrafish
liver cells reported IC50 values of 84.76 |ig/mL after 48 hours exposure {Cui, 2015, 3981517}.
3.4.1.3.5.2 Apoptosis
To determine the mechanism underlying PFOA-induced cytotoxicity, several studies have
interrogated the apoptosis pathway as a potential mechanism {Li, 2017, 4238518; Buhrke, 2013,
2325346; Cui, 2015, 3981517}. Apoptosis is characterized by biochemical and morphological
changes in cells. Flow cytometry has been used to quantify the percentage of apoptotic cells and
their phase in cells exposed to PFOA. The percentage of apoptotic cells in the early and late
phases of apoptosis nearly doubled in isolated C57BL/6J mice hepatocytes exposed to 500 |iM
and 1,000 |iM PFOA for 24 hours {Xu, 2019, 5381556}. In zebrafish liver cells exposed to the
IC50 (84.76 |ig/mL) and IC80 (150.97 |ig/mL) for 48 hours, the percentage of dead cells in the
late phase of apoptosis did not change in cells exposed to the IC50 compared to control, while a
significant increase in the percentage of apoptotic cells in the late phase of apoptosis was
observed in the cells exposed to the IC80 {Cui, 2015, 3981517}.
Activation of cysteine aspartic acid-specific protease (caspase) family is essential for initiation
and execution of apoptosis. PFOA-induced apoptosis via caspase activities have been examined
in primary mouse hepatocytes, mouse cell lines, and human cell lines after exposure to various
PFOA concentrations {Sun, 2019, 5024252; Cui, 2015, 3981517; Buhrke, 2013, 2325346;
Huang, 2013, 2850934; Li, 2017, 4238518; Xu, 2020, 6316207}. In mouse hepatocytes, PFOA
induced caspase activity in a dose-dependent manner {Li, 2017, 4238518}. In male C57BL/6J
mouse hepatocytes treated with PFOA for 24 hours, caspase 3 activity did not change at doses
below 1,000 |iM but increased by more than 1,000% at 1,000 |iM {Xu, 2020, 6316207}. In a
3-71
-------
DRAFT FOR PUBLIC COMMENT
March 2023
spheroid model of mouse liver cells (AML12), increased activity of caspase 3/7 was detected
from 14 to 28 days of >100 |iM PFOA exposure {Sun, 2019, 5024252}. In contrast, 100 |iM
PFOA did not change caspase 3/7 activity in HepG2 cells exposed for 48 hours {Buhrke, 2013,
2325346}.
Another key feature of cells undergoing apoptosis is the release of lactate dehydrogenase (LDH).
Many studies have reported intracellular release of LDH in hepatocytes treated with PFOA {Yan
2015, 3981567; Shan, 2013, 2850950; Wiels0e, 2015, 2533367; Sun, 2019, 5024252}. In male
C57BL/6J mouse primary hepatocytes treated with PFOA for 24 hours, 35% increase in LDH
was observed at the 10 mM dose compared to control. However, for all concentrations below 10
mM, the difference was not significant {Xu, 2020, 6316207}.
Changes in mRNA and protein expression of apoptotic genes is a hallmark of apoptosis.
Increased expression of p53, Bcl-2, Bcl-2 associated X-protein (Bax), caspase-3, nuclear factor
kappa B (NF-kB) mRNA and protein was observed in zebrafish liver {Cui, 2015, 3981517}. In
human hepatoma SMM-721 cells treated with 10 or 100 |ig/mL PFOA for 3 hours, BAX mRNA
was significantly increased while B-cell lymphoma 2 (Bcl-2) decreased compared to control {Lv,
2019, 5080368}. Proteomic analysis of 28 proteins differentially expressed in PFOA-exposed
human non-tumor hepatic cells (L-02) led the authors to conclude that PFOA induces apoptosis
by activating the p53 mitochondria pathway {Huang, 2013, 2850934}. This result is consistent
with several studies showing that PFOA-induced liver apoptosis is in part mediated through p53
activation {Li, 2017, 4238518; Sun, 2019, 5024252}. In a third study that examined miRNA
expression in the mouse liver, an increase in the expression of miR-34a-5p, which has been
shown to be involved in p53-mediated apoptosis, was observed {Yan, 2014, 2850901}.
PFOA has been shown to induce apoptosis through morphological changes to the mitochondrial
membrane {Xu, 2020, 6316207; Li, 2017, 4238518}. One study in Balb/c male mice gavaged
with PFOA (0.08-20 mg/kg/day) for 28 days suggested that hepatocyte apoptosis following
exposure to PFOA may be caused by endoplasmic reticulum stress, mediated by the induction of
ER stress markers including phosphorylated eukaryotic initiation factor 2a (p-elf2a), spliced X
box-binding protein 1 (XBP1), and C/EBP homologous protein (CHOP) {Yan, 2015, 3981567}.
An RNA-sequencing study in primary human hepatocytes found that PFOA exposure was
associated with changes in gene expression that aligned with cell death and hepatic system
disease, including necrosis, cholestasis, liver failure, and cancer {Beggs, 2016, 3981474}.
Another RNA-sequencing study showed that PFOA induced intracellular oxidative stress in
Sprague Dawley rats leading to apoptosis {Liu, 2017, 3981337}. Other mechanisms underlying
PFOA-induced apoptosis include DNA damage {Wielsoe, 2015, 2533367}, autophagosome
accumulation {Yan, 2015, 3981567; Yan, 2017, 3981501}, induction of ER stress biomarkers
and oxidative stress {Li, 2017, 4238518; Huang, 2013, 2850934; Panaretakis, 2001, 5081525;
Wielsoe, 2015, 2533367}, and reduction of mitochondrial ATP {Mashayekhi, 2015, 2851019;
Sun, 2019, 5024252}. Although many studies have reported oxidative stress as a potential
mechanism underlying PFOA-induced apoptosis, Florentin et al. (2011, 2919235) did not
observe an increase in DNA damage or ROS at doses that proved cytotoxic to HEPG2 cells,
leading the authors to conclude that PFOA-induced apoptosis is not related to DNA damage nor
oxidative stress.
3-72
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA-induced apoptosis has been shown to differ between males and females. In male and
female Balb/c mice gavaged with PFOA at doses ranging from 0.01-2.5 mg/kg/day for 28 days,
caspase-9 activity and dissipation of the mitochondrial membrane potential were higher in
females than males. Specifically, mitochondrial membrane dissipation was 25% in males and
39% in females for mice in the 2.5 mg/kg/day groups. In the 0.05 mg/kg/day group, caspase-9
activity was elevated by 72% in females compared to 40% in males. The sexual dimorphic
changes in caspase-9 and mitochondrial membrane dissipation were accompanied by
morphological changes in the mitochondria characterized by increased mitochondrial vesicle
formation and swelling in female than male hepatocytes, suggesting that female livers are more
susceptible to PFOA-induced apoptosis than males {Li, 2017, 4238518}.
3.4.1.3.5.3 Cell Cycle and Proliferation
Alterations in cell proliferation and cell cycle were also seen in many in vivo and in vitro studies
{Zhang, 2020, 6316915; Zhang, 2016, 3748826; 9959776; Beggs, 2016, 3981474; Buhrke, 2013,
2325346; Buhrke, 2015, 2850235; Lv, 2019, 5080368; Wen, 2020, 6302274}. In mice exposed
to 3 mg/kg/day PFOA for 7 days by oral gavage, proliferation in the liver, as seen through
proliferation cell nuclear antigen (PCNA) staining, was increased relative to control {Beggs,
2016, 3981474}. HL-7702 cells were treated with PFOA at concentrations of 50-400 |iM for 48
or 96 hours {Zhang, 2016, 3748826}. All except the highest dose (400 |iM) group showed an
increase in cell proliferation compared to control at 48 hours. Other studies have reported a
similar pattern where proliferation is significantly increased at low doses and decreased at high
doses of PFOA in human primary hepatocytes {Buhrke, 2015, 2850235}, HepG2 {Buhrke, 2013,
2325346}, and HepaRG cells {Behr, 2020, 6505973}. Together these studies suggest that higher
concentration of PFOA may interfere with cell cycle progression by reducing cell proliferation
rather than severely inducing apoptosis.
In contrast, a study in primary hepatocytes of Sprague Dawley rats found increased proliferation
at the highest dose and no proliferative effect at low doses. Approximately 16% increase in
proliferation was observed with PFOA exposures of 100 |iM for 24 hours compared to controls
{Liu, 2017, 3981337}. However, no changes in cell number as measured by MTT assay was
observed at the PFOA concentration range of 0.4-25 |iM at the same duration, adding to the
evidence that PFOA-induced proliferation is dose-dependent and may vary by cell type.
PFOA has also been shown to disrupt cell cycle progression. Using flow cytometry, Zhang et al.
2016, 3748826) found that in HL-7702 cells, the proportion of cells in the G0/G1 phase (non-
dividing) significantly decreased while cells in the S phase increased after 48 hours of exposure
to 50 and 100 |iM PFOA. However, at the 200 |iM and 400 |iM exposure for 48 hours,
percentage of cells in the G0/G1 phase increased while cells in the G2/M/S phase (interphase
growth/mitosis) decreased significantly compared to control. Interestingly, the same trend was
observed in cells incubated at the same dose for 96 hours {Zhang, 2016, 3748826}. A second
study in immortalized non tumor cells derived from human normal liver tissue (L-02 cells) also
used flow cytometry to examine changes in the cell cycle after 72 hours at 25 and 50 mg/L and
found that PFOA increased the percentage of cells in G2/M phases but decreased the number of
cells in G0/G1 and S phases {Huang, 2013, 2850934}. Additionally, the percentage of cells in
apoptotic sub-Gl (G1-) phase increased significantly from 19% to 33% compared to 10% of
cells in the Gl- phase in the control group, leading the authors to conclude that PFOA treatment
disrupt cell cycle in L-02 cells by arresting cells in G2/M phase while inducing apoptosis. A
3-73
-------
DRAFT FOR PUBLIC COMMENT
March 2023
third study in a zebrafish liver cell line also used flow cytometry to identify changes in the cell
cycle after 85 and 151 |ag/m L PFOA exposure for 48 hours. In corroboration with the study in L-
02 cells, PFOA concentration of 151 |ag/m L showed an increase in the percentage of cells in the
G2/M/S stage and a decrease in the percentage of cells in the G1/G0 phase {Cui, 2015,
3981517}. Together, these studies suggest that PFOA interferes with the balance between
apoptosis and proliferation by disrupting cell cycle progression.
PFOA-induced changes in cell proliferation and cell cycle progression are often accompanied
with changes in mRNA and protein expression of genes implicated in cell cycle progression.
Pathway analysis of protein expression in human HL-7702 normal liver cells exposed to 50 |iM
PFOA for 48 and 96 hours identified 68 differentially expressed proteins that are related to cell
proliferation and apoptosis {Zhang, 2016, 3748826}. Western blot analysis from the same study
showed differential protein expression of positive cell cycle-regulators, including cyclins and
cyclin-dependent kinases (Cyclin/CDKs) that are known to control G1/G2/S/M cell cycle
progression, as well as negative regulators (p53, p21, MYTI, and WEE1). Interestingly,
expression of cell cycle regulations was dose-dependent. Significant induction of cyclin Dl,
CDK6, cyclin E2, cyclin A2, CDK2, p-CDKl, p53, p21, p-WEEl and myelin transcription factor
1 (MYTI) was observed at low dose (50 or 100 |iM). However, cyclin A2, cyclin B1 and p21
proteins were significantly inhibited at high dose (400 |iM) at the same duration (48 hours)
{Zhang, 2016, 3748826}. In primary human hepatocytes treated with 10 |iM PFOA, CCND1 and
Aldo-keto reductase family 1 member B10 (AKR1B10) mRNA were significantly induced after
96 hours {Beggs, 2016, 3981474}. AKR1B10 is a promitogenic gene that has been associated
with the progression of hepatocellular carcinoma {Matkowskyj, 2014, 10365736}. In addition,
two microarray studies in hepatic cell lines found that PFOA exposures ranging from 100-305
|iM for up to 48 hours were associated with pathways involved in the regulation of cellular
proliferation or the cell cycle {Song, 2016, 9959776; Louisse, 2020, 6833626}.
PFOA has been shown to decrease the expression of hepatocyte nuclear factor 4-alpha (HNF4a),
a regulator of hepatic differentiation and quiescence, in multiple studies and is thought to
mediate steatosis following PFOA exposure {Behr, 2020, 6505973; Beggs, 2016, 3981474}. One
study suggested that PFOA-induced proliferation may be mediated by the degradation of HNF4a
{Beggs, 2016, 3981474}. This study, using wild type CD-I and HNF4a knockout mice, reported
that 11 out of 40 genes altered by PFOA exposure were regulated by HNF4a. PFOA exposure
decreased the expression of HNF4a in both male mice and primary human hepatocytes and
increased the expression of Nanog, a stem cell marker, suggesting that PFOA may be de-
differentiating hepatocytes. Increased relative liver weight in PFOA-exposed mice was observed
in this study and the authors concluded that hepatomegaly, along with other liver effects such as
steatosis, may be mediated by PFOA-induced dysregulation of HNF4a.
3.4.1.3.5.4 Conclusions
Hepatotoxicity is widely cited as a type of toxicity induced by PFOA exposure. PFOA has been
shown to trigger apoptosis at high doses and induce cell proliferation at low doses. PFOA-
induced apoptosis is activated through a cascade of mechanisms including activation of caspase
activity, intracellular release of LDH, induction of apoptotic genes, morphological changes to the
mitochondria membrane, and activation of p53 mitochondria pathway. Additionally, PFOA
induced hepatocyte proliferation both in vivo and in vitro by disrupting cell cycle progression
3-74
-------
DRAFT FOR PUBLIC COMMENT
March 2023
leading to liver dysfunction, including steatosis and hepatomegaly. Therefore, PFOA exposure
may lead to liver cytotoxicity through a myriad of intracellular events.
3.4.1.3.6 Inflammation and Immune Response
The liver is an important buffer between the digestive system and systemic circulation and is
thus exposed to compounds that are potentially immunogenic, resulting in protective immune
and inflammatory responses. Kupffer cells constitute the majority of the liver-resident
macrophages and make up one third of the non-parenchymal cells in the liver. Kupffer cells
phagocytose particles, dead erythrocytes, and other cells from the liver sinusoids and play a key
role in preventing immunoreactive substances from portal circulation from entering systemic
circulation {Dixon, 2013, 10365841}. While Kupffer cells can be protective in drug- and toxin-
induced liver toxicity, dysregulation of Kupffer cell-mediated inflammatory responses is
associated with a range of liver diseases, including steatosis. Other liver-resident immune cells
include natural killer (NK) cells, invariant NKT cells, mucosal associated invariant T (MAIT)
cells, yST cells, and memory CD8+T cells {Wang et al., 2019, 10365737}. The non-immune
cells of the liver, liver sinusoidal endothelial cells (LSECs), hepatocytes, and stellate cells, also
participate in immunity. They can express pattern recognition receptors and present antigens to T
cells {Robinson, 2016, 10284350}. However, the impact of PFOA on the immune function of
these cell types has not been thoroughly investigated.
3.4.1.3.6.1 In Vivo Studies
Investigations into the liver immune response have been conducted in a single human study in
the C8 Health Project cohort {Bassler, 2019, 5080624}, and in several rodent studies {Botelho,
2015, 2851194; Li, 2019, 5080362; Liu, 2016, 3981762; Yu, 2016, 3981487; Hui, 2017,
3981345; Wu, 2018, 4238318}. Bassler et al. (2019, 5080624) collected 200 serum samples from
participants of the C8 Health Project to analyze mechanistic biomarkers of non-alcoholic fatty
liver disease (NAFLD) and test the hypothesis that PFAS exposures are associated with
increased hepatocyte apoptosis and decreased proinflammatory cytokines. PFOA levels were
significantly correlated with decreases in serum levels of the proinflammatory cytokine tumor
necrosis factor a (TNFa). In contrast, both interferon y (IFNy) and cleaved complement 3 (C3a)
were positively associated with PFOA levels. The authors state that these results are consistent
with other findings that PFAS are immunotoxic and downregulate some aspects of the immune
responses, but paradoxically result in increased apoptosis, which may subsequently result in
progression of liver diseases (including NAFLD).
A study in mice acutely exposed to PFOA also linked hepatic injury to activation of the
complement system. In contrast to the human study {Bassler, 2019, 5080624}, a decrease in
serum C3a was observed in mice {Botelho 2015, 2851194}. C57BL/6 mice exposed to a 10-day
dietary treatment with PFOA (0.002-0.02%, w/w) exhibited hepatomegaly, elevated serum
triglycerides, elevated alanine aminotransferase (ALAT), hepatocyte hypertrophy, and
hepatocellular necrosis at all doses. At the highest dose only, PFOA-induced hepatic injury
coincided with deposition of the complement factor C3a fragment in the hepatic parenchyma.
The findings support activation of the classical, but not alternative complement cascade in liver,
and correlated with diminished C3 levels in serum. In serum, commercial hemolytic assays
indicated attenuation of both the classical and alternative complement pathways. These authors
proposed that that PFOA-mediated induction of hepatic parenchymal necrosis is the initiation
event that leads to activation of the complement cascade and pro-inflammatory responses.
3-75
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In another study in mice, the effects of PFOA exposure on inflammatory changes in liver varied
depending on the presence of pre-existing NAFLD {Li, 2019, 5080362}. Mice were subjected to
control diet or HFD for 16 weeks to induce NAFLD, after which they were exposed to vehicle or
1 mg/kg/day PFOA by oral gavage for 2, 8, or 16 weeks; the control diet and HFD were
continued throughout the exposure period until necropsy. In mice on the control diet,
inflammatory changes were not observed in the first 8 weeks of PFOA treatment. However, after
16 weeks of PFOA treatment, mild hepatic lobular inflammation was observed in 3 of 5 animals,
suggesting that chronic exposure to PFOA induces inflammatory changes in liver. In HFD-fed
mice, focal inflammation was seen as early as 2 weeks after initiating PFOA treatment and
inflammatory foci were observed in 2 of 5 mice after 16 weeks of PFOA exposure. Gene
expression of Tnfa measured by qRT-PCR was elevated in the HFD group exposed to PFOA for
all three treatment durations (2, 8, or 16 weeks of PFOA). Similarly, Liu et al. (2016, 3981762)
observed an induction of TNFa in liver homogenates, measured by ELISA, in male Kunming
mice fed a regular diet {Liu, 2016, 3981762} and exposed to a higher dose of PFOA (10
mg/kg/day for 2 weeks). This study observed significantly elevated levels of both TNFa and IL-6
in liver homogenates.
Li et al. (2019, 5080362) also confirmed increased expression of inflammatory genes using an
RNA-Seq transcriptomic approach. Compared to mice on the control diet, the HFD group
exposed to PFOA resulted into 537 differentially expressed genes. The inflammatory response
was among the top enriched Gene Ontology (GO) terms for the gene set specific to the PFOA-
exposed HFD. Analysis using Ingenuity Pathway Analysis showed significant upregulation of
chemokines and chemokine-related genes and toll-like receptor (TLR) related genes in the
PFOA-exposed HFD group compared to mice fed the control diet. Taken together with the
histopathological findings, these gene expression changes suggest that that preexisting fatty liver
may enhance PFOA-mediated inflammatory changes in liver.
Another potential nexus between changes in hepatic lipid metabolism and inflammation comes
from a high-throughput metabolomics study in male BALB/c mice {Yu, 2016, 3981487}. After a
28-day exposure to 0, 2.5 or 5 mg/kg/day PFOA, livers were subjected to metabolomic analysis.
Metabolite analysis indicated PFOA altered polyunsaturated fatty acid metabolism including the
arachidonic acid pathway. Arachidonic acid is a precursor in production of inflammatory
mediators including prostaglandins, thrombaxanes, and leukotrienes. Prostaglandins (PGD2,
PGE2, and PGF2a) were slightly elevated but increases did not reach statistical significance.
However, the ratio of the thromboxane A2 (TXBA2) metabolite thromboxane X2 (TXB2) to
prostaglandin 12 (PGI2) was significantly decreased in PFOA-exposed mice. Given the
prothrombotic role of TXBA2 and the vasodilatory role of PGI2, the authors suggest these
changes are consistent with ischemic liver injury that is characterized by vasodilation of
microvasculature, lessened adherent leukocytes, and improved flow velocity in liver. Two
leukotrienes, LTD4 and LTB4 were significantly lower in the high dose group. Both leukotrienes
can also regulate vascular permeability and the authors suggest these changes are consistent with
PFOA-induced inflammation in liver. PFOA also upregulates CD36 gene expression in
hepatocytes {Hui, 2017, 3981345; Wu, 2018, 4238318}, which is a negative regulator of
angiogenesis {Silverstein, 2009, 10365842}. Together with the PFOA-mediated changes in
abundance of prostaglandins and thrombaxanes, these findings raise the possibility that PFOA-
mediated alterations of the hepatic microvasculature are key events in the development or
persistence of liver inflammation.
3-76
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.1.3.6.2 In Vitro Studies
In a study investigating the hepatic effects of PFOA in vitro, Song et al. (2016, 9959776)
evaluated gene expression changes in human liver hepatocellular carcinoma HepG2 cells using a
whole genome expression microarray. After exposing these cells to 306 |iM PFOA (the IC20
dose for cell viability inhibition) for 48 hours, gene expression changes were evaluated. PFOA
exposure led to differential regulation of 1,973 genes. Through KEGG pathway analyses, the
authors reported that genes related to immune response were among the most differentially
expressed biological process out of the 189 processes with altered genetic profiles. The authors
identified 17 immune-associated genes that were differentially expressed. These genes mapped
to the TNF signaling pathway, nucleotide-binding and oligomerization domain (NOD)-like
receptor signaling, cytokine-cytokine receptor interactions, and the complement and coagulation
cascade system. These findings support a role for PFOA in dysregulating innate immune
mechanisms.
Alterations in cytokines associated with regulation of adaptive immunity were also observed
using multi-cellular hepatic organotypic culture models composed of primary human or rat cells
{Orbach, 2018, 5079788}. This system involved seeding primary liver sinusoidal epithelial cells
and Kupffer cells encapsulated in extracellular matrix proteins above the hepatocytes. This
culture system forms a stratified three-dimensional (3D) structure designed to more accurately
mimic liver tissue. Organotypic cultures were exposed to 500 [xM PFOA for 24 hours (the LC50
in human cultures). PFOA exposure led to a 62% decrease in IL-10 levels. In addition to being a
key cytokine in development of T helper lymphocytes, IL-10 has anti-inflammatory properties.
Thus, the decrease in IL-10 observed in organotypic culture is consistent with the
proinflammatory changes in liver associated with PFOA exposure. Using a proteomic approach,
another cytokine, IL-22, has also been shown to be downregulated in PFOA-exposed human
hepatic L-02 cells {Huang, 2013, 2850934}. IL-22, a member of the IL-10 cytokine family,
exerts protective effects in liver during acute inflammation and alcoholic liver injury {Ki, 2010,
10365730; Zenewicz, 2007, 10365732}. T helper (Th22) cells are a T-cell subset responsive to
IL-22. Th22 cells function in maintaining the integrity of the epithelial barriers {Hossein-
Khannazer, 2021, 10365738}. As such, diminished levels of IL-22 in the liver suggest that
PFOA could interfere with the protective effects of IL-22 and Th22 cells.
3.4.1.3.6.3 Conclusions
The limited number of studies reviewed support a role PFOA in inducing hepatic inflammation
through dysregulation of innate immune responses. This includes elevated levels of TNFa as
well as changes in prostaglandin and thromboxane levels. Gene expression studies also suggest a
role for chemokines in elaborating inflammation in liver. Expression of genes coding for
products involved in innate immune defense systems were altered, including TLRs, molecules
involved in NOD signaling, and C3a, a key indicator of complement cascade activation. Far less
is known regarding PFOA effects on adaptive immunity in liver. PFOA exposure caused a
reduction in IL-10 levels in organotypic culture of liver. IL-10 has anti-inflammatory properties
in addition to promoting differentiation of Th2 CD4+ T cells. Intriguingly, IL-22 levels were
diminished in PFOA-exposed hepatic cells. This cytokine may impact the function of Th22 T
lymphocytes and impact the epithelial barriers in liver. Moreover, IL-22 reduction may reduce
the protective effects of this cytokine during inflammation. Altogether, induction of
inflammation appears to be an important mechanism that impacts liver pathogenesis in response
to PFOA exposure, though the contribution of specific populations of resident or infiltrating liver
3-77
-------
DRAFT FOR PUBLIC COMMENT
March 2023
immune cells and the series of events that produce inflammation have yet to be elucidated.
Adaptive immune responses are disrupted in PFOA-exposed animals (Section 3.4.2.2). However,
whether alterations in adaptive immunity impact pathogenetic mechanisms in liver remain
unknown.
3.4.1.3.7 Oxidative Stress and Antioxidant Activity
3.4.1.3.7.1 Introduction
Oxidative stress, caused by an imbalance of reactive oxygen species (ROS) production and
detoxification processes, is a key part of several pathways, including inflammation, apoptosis,
mitochondrial function, and other cellular functions and responses. In the liver, oxidative stress
contributes to the progression and damage associated with chronic diseases, such as alcoholic
liver disease, non-alcoholic fatty liver disease, hepatic encephalopathy, and Hepatitis C viral
infection {Cichoz-Lach, 2014, 2996796}. Indicators of oxidative stress include but are not
limited to increased oxidative damage (e.g., malondialdehyde (MDA) formation); increased
reactive oxygen species (ROS) production (e.g., hydrogen peroxide and superoxide anion);
altered antioxidant enzyme levels or activity (e.g., superoxide dismutase (SOD) and catalase
(CAT) activity); changes in total antioxidant capacity (T-AOC); changes in antioxidant levels
(e.g., glutathione (GSH) and glutathione disulfide (GSSG) ratios); and changes in gene or protein
expression (e.g., nuclear factor-erythroid factor 2-related factor 2 (Nrf2) protein levels). PFOA
has been implicated as a chemical that can induce these indicators of oxidative stress,
inflammation, and cell damage.
3.4.1.3.7.2 In Vivo Models
3.4.1.3.7.2.1 Mouse
Yan et al. (2015, 3981567) examined livers from male Balb/c mouse following PFOA exposure
of 0.08, 0.31, 1.25, 5, or 20 mg/kg/day for evidence of oxidative stress, including changes in
expression of oxidative stress-related genes. While no change was observed in Cat expression
levels, increases in Sesnl, Sodl, and Sod2 were observed in livers from mice exposed to 1.25, 5,
and 20 mg/kg/day PFOA, respectively. PFOA exposure led to increased CAT activity and
decreased SOD activity in mouse livers. MDA contents were decreased at all dose levels, and
levels of the antioxidant GSH increased at 5 and 20 mg/kg/day PFOA. Authors concluded that
the changes in SOD, CAT, GSH, and MDA reflect PFOA-induced disruptions to the antioxidant
defense system in the livers of exposed mice. However, no significant oxidative damage was
observed.
Li et al. (2017, 4238518) explored the role of ROS accumulation in apoptosis in male and female
Balb/c mice dosed with 0.05, 0.5, or 2.5 mg/kg/day PFOA for 28 days. The authors explored
how activation of PPARa and suppression of the electron transport chain (ETC) sub-unit
Complex I influenced ROS generation. Excluding the lowest male dose group, PFOA exposure
significantly increased 8-OHdG levels in the liver, a key indicator of oxidative DNA damage. 8-
OHdG levels were higher among dosed females compared to males, which authors suggest
signals stronger genotoxicity in females. Authors explored the connection between the oxidative
stress and apoptosis through the p53 signal pathway. Increases in p53 levels occurred in the same
dose groups with elevated 8-OHdG, which authors suggest indirectly links oxidative stress to
apoptosis. Authors posited that ROS hypergeneration led to increased 8-OHdG levels, and DNA
damage then leads to increases in programmed cell death protein 5 (PDCD5), which activates
3-78
-------
DRAFT FOR PUBLIC COMMENT
March 2023
p53 to induce apoptosis. At 0.5 and 2.5 mg/kg/day, PFOA exposure decreased expression of
electron transport chain (ETC) proteins, which corresponds to an increase in ROS generation and
accumulation. For two ETC subunits, ACP and NDUV2, expression was increased, which also
indicates an accumulation of ROS and an increase in antioxidant activity to counter ROS
generation. At 0.05 mg/kg/day, female mice showed more oxidative stress than males. In these
females, Complex I suppression drove ultimate apoptosis, while PPARa activation drove
apoptosis among males.
Two studies examined changes in oxidative stress endpoints in male Kunming mice exposed to
PFOA {Yang, 2014, 2850321; Liu, 2016, 3981762}, and an additional two studies evaluated
oxidative stress endpoints in pregnant female Kunming mice and their pups {Li, 2019, 537402;
Song, 2019, 5079965}. In the livers of male Kunming mice exposed to 2.5, 5, or 10 mg/kg/day
PFOA for 14 days, MDA at all doses and H2O2 at 5 and 10 mg/kg/day levels were significantly
increased compared to controls {Yang, 2014, 2850321}. Liu et al. (2016, 3981762) explored
grape seed proanthocyanidn extract (GSPE) as a protective agent against PFOA damage in the
liver. The authors reported significantly increased MDA and H2O2, significantly decreased Nrf2
protein levels, and significantly decreased SOD and CAT activity in the liver following PFOA
exposure. Additionally, expression of SOD and CAT, measured via qRT-PCR, were significantly
decreased in the livers of exposed mice. Li et al. (2019, 5387402) found that serum levels of
SOD and 8-OHdG were significantly increased in pups of females dosed at 2.5, 5, and 10
mg/kg/day PFOA. Serum levels of CAT were increased at 5 and 10 mg/kg/day PFOA. PFOA-
induced changes in SOD, CAT, and 8-OHdG reflect increased antioxidant activity in response to
increased oxidative stress and increased DNA damage. In their study examining the protective
effects of lycopene against PFOA-induced damage, Song et al. (2019, 5079965) exposed
pregnant mice to 20 mg/kg/day PFOA via oral gavage from gestational days (GD) 1-7. After
sacrifice on GD 9, levels of MDA were significantly increased in livers of pregnant mice treated
with 20 mg/kg/day PFOA, while SOD and GSH-Ps levels were significantly decreased compared
to controls, providing evidence of oxidative damage in the liver following PFOA exposure.
Three studies dosed C57B1/6 mice with PFOA to study impacts on oxidative stress endpoints
{Wen, 2019, 5080582; Crebelli, 2019, 5381564; Kamendulis, 2014, 5080475}. In male C57B1/6
mice dosed with 28 mg/L PFOA, Crebelli et al. (2019, 5381564) found slightly decreased T-
AOC, but the results were not statistically significant. MDA levels were below detection limits in
all collected samples. Additionally, there was no statistically significant change in the levels of
liver TBARS that would indication lipid peroxidation. Kamendulis et al. (2014, 5080475)
exposed male C57B1/6 mice to 5 mg/kg/day and found that PFOA exposure led to a 1.5-fold
increase in 8-iso-PGF2a levels, a measure of lipid peroxidation that indicates oxidative damage.
Additionally, PFOA led to a nearly 2-fold increase in mRNA levels of Sodl in liver cells
extracted from mice dosed at 2.5 and 5 mg/kg/day PFOA. mRNA levels of Sod2 and Cat were
increased 3-fold and 1.3-fold, respectively. The same doses of PFOA also led to a nearly 2-fold
increase in Nqol mRNA levels. The induction of genes for detoxifying enzymes following
PFOA exposure suggests PFOA causes increased oxidative stress activity. In a different study
{Wen, 2019, 5080582}, 1 and 3 mg/kg/day PFOA exposure in wild-type C57BL/6 NCrl male
mice increased gene expression of Nrf2 and Nqol, measured via qRT-PCR assays, by 50-300%.
One gene expression compendium study aimed to examine the relationship between activation of
xenobiotic receptors, Nrf2, and oxidative stress by comparing the microarray profiles in mouse
3-79
-------
DRAFT FOR PUBLIC COMMENT
March 2023
livers (strain and species not specified) {Rooney, 2019, 6988236}. The study authors compiled
gene expression data from 163 chemical exposures found within Illumina's BaseSpace
Correlation Engine. Gene expression data for PFOA exposure was obtained from a previously
published paper by Rosen et al. (2008, 1290832). In WT (129Sl/SvlmJ) and Ppara-null male
mice, Nrf2 activation was observed (as seen by increases in gene expression biomarkers) after a
7-day exposure to 3 mg/kg/day PFOA via gavage. Similar to Nrf2, CAR was also activated in
both mouse strains after PFOA exposure. The authors proposed that CAR activation by chemical
exposure (PFOA or otherwise) leads to Nrf2 activation, and that oxidative stress may be a
mediator.
3.4.1.3.7.3 In Vitro Models
Rosen et al. (2013, 2919147) assessed oxidative stress-related gene expression changes using
Taqman low density arrays (TLDA) in both mouse and human primary hepatocytes exposed to
levels of PFOA ranging from 0-200 |iM. PFOA exposure led to a decrease in the expression of
the heme oxygenase 1 (Hmoxl) gene in human primary hepatocytes. There were no changes
observed in the nitric oxide synthase 2 (Nos2) gene nor in either gene in primary mouse
hepatocytes.
Orbach et al. (2018, 5079788) examined the impacts of 500 |iM PFOA exposure in multi-cellular
organotypic culture models (OCM) of primary human and rat hepatocytes and in collagen
sandwich (CS) models via high-throughput screening. In exposed rat and human cells, PFOA
decreased GSH levels by <10%. The authors suggest that PFOA did not bind to or oxidize GSH.
In human OCMs, mitochondrial integrity decreased 37% following PFOA exposure. In human
CS models, the decrease was 39%. In rat OCMs, exposure decreased mitochondrial integrity by
47%), and by 45%> in rat CS models.
In primary rat hepatocytes incubated with 100 |iM PFOA for 24-hours, Liu et al. (2017,
3981337) found that intracellular oxidant intensity increased to more than 120% of control levels
as measured by mean fluorescence intensity of 2',7'-dichlorofluorescein (DCF). In addition, cells
incubated with 6.25, 25, or 100 |iM PFOA displayed significantly increased levels of
mitochondrial superoxide, measured by MitoSOX fluorescence. In cells exposed to 100 |iM
PFOA, mitochondrial superoxide levels were elevated to 130%> of those of controls. Authors
suggest that these results indicate that mitochondrial superoxide is a more sensitive marker of
oxidative stress than intracellular ROS levels.
Two studies examined oxidative stress endpoints following PFOA exposure in mitochondria
isolated from Sprague Dawley rats {Mashayekhi, 2015, 2851019; Das, 2017, 3859817}.
Mashayekhi et al. (2015, 2851019) examined oxidative damage in the mitochondria, an
important organelle in the oxidative stress pathway, associated with PFOA exposure. In
mitochondria isolated from the livers of male Sprague Dawley rats, significant increases in the
percent ROS formation were observed following exposure to 0.75, 1, or 1.5 mM PFOA for up to
20 minutes. At 30 minutes and longer, significant increases were observed at the two highest
concentrations only. Mashayekhi et al. (2015, 2851019) also observed significantly increased
levels of ROS formation in complexes I and III of the mitochondrial respiratory chain, key
sources of ROS production. Disruption to the chain can lead to accumulation of ROS and,
ultimately, oxidative stress. In complex II, activity levels were significantly decreased at 0.75
and 1.5 mM PFOA exposure. There was no significant difference in MDA of GSH content in
3-80
-------
DRAFT FOR PUBLIC COMMENT
March 2023
liver mitochondria following PFOA exposure. PFOA exposure from 0.5-1.5 mM significantly
decreased mitochondrial membrane potential and ATP levels and significantly increased
mitochondrial swelling, suggesting a decrease in mitochondrial function following exposure to
PFOA.
Xu et al. (2019, 5381556) exposed mouse hepatic primary cells from C57B1/6J male mice to
0.01, 0.1, 0.5, or 1 mM PFOA for 24 hours. ROS levels, measured by a CM-H2DCFA
fluorescent probe, were significantly increased in cells exposed to 0.5 and 1 mM PFOA.
Interestingly, SOD activity was significantly increased in cells exposed to 0.5 and 1 mM PFOA,
up to 123% with 1 mM, while CAT activity was reduced to 7.7% in cells at the highest
concentration. Increasing PFOA exposure also led to alterations in the structure of SOD,
resulting in a significantly decreased percentage of a-helix structures (20%) and an increased
percentage of P-sheet structures (29%), providing evidence of polypeptide chain unfolding and
decreased helical stability. These structural changes suggest that PFOA interacts directly with
SOD, resulting in polypeptide chain extension and, ultimately, diminished antioxidant
capacity. Additionally, GSH content was increased by 177% and 405% in cells exposed to 0.5
mM and 1 mM PFOA, respectively. The authors suggest that increases in GSH may reflect
cellular adaptations to oxidative stress and can lead to detoxification of oxidized GSSG to GSH.
Xu et al. (2020, 6316207) exposed cultured primary mouse hepatocytes to 0.01, 0.1, 0.5, or 1
mM of PFOA for 24 hours to examine oxidative stress-related apoptosis. The authors examined
the impact of PFOA exposure on endogenous levels of lysozyme (LYZ), an enzyme that inhibits
oxidative stress-induced damage, and demonstrated that PFOA exposure impacted LYZ
molecular structure, subsequently decreasing activity levels, leading to oxidative stress-induced
apoptosis. Decreases in peak intensity at 206 nm during ultraviolet-visible (UV-vis) absorption
spectrometry represented an unfolding of the LYZ molecule following exposure to PFOA, which
inhibited enzyme activity. At concentrations of 100 |iM and above, LYZ enzyme activity
decreased to 91% of control levels. Such an impact on LYZ activity was deemed to be related to
the high affinity of PFOA for key central binding sites on the LYZ molecule.
In human HL-7702 liver cells, 24 hours of PFOA exposure at 1, 2.5, or 7.5 |ig/mL led to a dose-
dependent increase in 8-OHdG levels in cells exposed to the two highest concentrations {Li,
2017, 4238518}. The authors noted that DNA damage, which frequently accompanies increases
in 8-OHdG, was observed in their in vivo models following PFOA exposure, suggesting
increased oxidative stress following exposure. In human non-tumor hepatic cells (L-02) exposed
to 25 or 50 mg/L PFOA for 72 hours, Huang et al. (2013, 2850934) observed concentration-
dependent increases in ROS levels measured via DCFH-DA fluorescent probe, evidence of the
role of PFOA in inducing oxidative stress.
Six additional studies examined oxidative stress endpoints following PFOA exposure in HepG2
cell lines {Wan, 2016, 3981504; Wiels0e, 2015, 2533367; Shan, 2013, 2850950; Florentin, 2011,
2919235; Panaretakis, 2001, 5081525; Yan, 2015, 3981567}. Four studies reported increases in
ROS levels following PFOA exposure {Wan, 2016, 3981504; Wiels0e, 2015, 2533367;
Panaretakis, 2001, 5081525; Yan, 2015, 3981567}, while two studies did not observe statistical
differences in ROS levels following 1- or 24-hour PFOA exposures up to 400 |iM {Florentin,
2011, 2919235} or following 3-hour PFOA exposures up to 400 |iM {Shan, 2013, 2850950}.
3-81
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Wiels0e et al. (2015, 2533367) incubated HepG2 cells with up to 2 x 10-4 M PFOA to detect
changes in ROS, T-AOC, and DNA damage. PFOA exposure significantly increased ROS
production, as measured with the carboxy-H2DCFDA, and significantly decreased T-AOC at all
concentrations by 0.70-0.82-fold compared to controls. Additionally, PFOA induced DNA
damage, specifically, increased mean percent tail intensity, an indicator of strand breaks,
measured via comet assay. In cells exposed up to 400 |iM PFOA for up to 24 hours, Panaretakis
et al. (2001, 5081525) observed increased ROS levels, measured via DCFH-DA and
dihydroethidium fluorescent probes, following 3 hours PFOA exposure. H2O2 levels were
detectable in 91% and 98% of the cell population at 200 and 400 |iM PFOA, respectively.
Additionally, superoxide anion levels were detectable in 43% and 71% of cells exposed to 200
and 400 |iM PFOA, respectively. Authors reported evidence of depolarized mitochondrial
membranes in cells exposed up to 24 hours. Yan et al. (2015, 3981567) observed significantly
increased ROS levels in cells incubated with 100 and 200 |iM PFOA for 24 hours, but no
changes were observed in superoxide anion levels. After 72 hours of exposure, however, ROS
levels decreased at those concentrations, with statistically significant results observed at 200 |iM
PFOA. Activity levels of SOD and CAT were not altered in exposed cells compared to controls,
nor were MDA or GSH contents. Similarly, in HepG2 cells treated with PFOA for 24 hours, Yan
et al. (2015, 3981567) found ROS levels significantly increased, but no significant changes were
observed in SOD and CAT activity or MDA and GSH levels. Yarahalli Jayaram et al. (2018,
5080662) examined the impacts of PFOA exposure on oxidative stress endpoints and small
ubiquitin-like modifiers (SUMO), which play a key role in posttranslational protein
modifications. SUMOylation of a protein has been identified as a key part of the oxidative stress
pathway. In cells incubated with 250 |iM PFOA, ROS levels were significantly increased. Cells
incubated with PFOA also showed increased levels of nitric oxide (NO). Additionally,
expression levels of genes related to SUMOylation were measured. PFOA treatment significantly
increased levels of SUM02 in HepG2 cells, but did not impact SUMOl, SUM03, or UBC9
mRNA levels.
In cells exposed to 10 and 200 |iM PFOA for 24 hours, Florentin et al. (2011, 2919235) observed
significant increases in the percentage of DNA tails, an indicator of DNA damage measured via
comet assay. However, no such changes were observed at the 1-hour time point or at other
concentrations (5, 50, 100, or 400 |iM) after 24 hours. Additionally, no significant changes in
ROS generation were observed. Shan et al. (2013, 2850950) exposed HepG2 cells to 100 [xM
PFOA for 3 hours and found an increase in ROS generation, though the effect was not
statistically significant. Additionally, no changes were observed in the GSH/GSSG ratio.
In two cell lines derived from Hepalc-lc7 mouse cells, CR17 and HepaV cells, Melnikov et al.
(2018, 5031105) found that Hmoxl gene expression was significantly decreased in cells exposed
to PFOA for 24 hours compared to controls. Additionally, exposed HepaV cells showed
significantly decreased expression of Gclc and Gclm. There were no significant changes in GSH
levels after exposure to 100 |iM PFOA for 24 hours. CR17 cells have increased glutamate-
cysteine ligase (GCL) activity, leading to increased GSH content. Authors anticipated that the
elevated GSH levels in the CR17 cell line would better resist PFOA toxicity. They concluded
that the observed changes in gene expression in PFOA exposed HepaV cell lines, but not in
CR17 cell lines, supported this hypothesis.
3-82
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Sun et al. (2019, 5024252) examined the impacts of PFOA exposure on both a monolayer and a
scaffold-free three-dimensional spheroid model of mouse liver cells (AML12). Monolayer cells
were exposed to 6.25-2,000 |iM PFOA for 24 and 72 hours. The spheroid cell model was
exposed to 50, 100, and 200 |iM PFOA for up to 28 days. In monolayer cells exposed to 200 |iM
PFOA for 72 hours, ROS levels, measured via an ROS-Glo assay kit, increased 1.6-fold
compared to controls. In the spheroid cell models, however, ROS levels decreased in cells
exposed to 100 and 200 |iM PFOA for 24 and 72 hours, which authors report suggests that
monolayer cells demonstrate higher PFOA toxicity due to the absence of an endogenous
extracellular matrix with the potential to inhibit PFOA diffusion. After 14 days of exposure, ROS
levels in spheroid cells significantly increased at all concentrations. Gene expression of
glutathione S-transferases alpha 2 (Gsta2), Nqol, and Ho-1 increased with increasing PFOA
concentration and duration of exposure, which provides additional evidence of PFOA's effect on
oxidative stress.
3.4.1.3.7.4 Conclusions
Results from new studies published since the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}
further support the 2016 conclusions that PFOA can cause oxidative stress and related cellular
damage. Evidence of increased oxidative stress in the liver, including increased ROS levels,
changes in GSH and GSSG levels, and decreases in T-AOC, was observed following both in
vivo and in vitro exposures to PFOA. PFOA exposure was also associated with increased levels
of markers of oxidative damage and decreased activity or levels of protective antioxidants that
play a role in the reduction of oxidative damage. There was also evidence that PFOA can disrupt
the structure and subsequent function of crucial enzymes that mitigate ROS production and
oxidative damage, SOD and LYZ. While further research is needed to understand the underlying
mechanisms of PFOA-induced oxidative stress responses, it is clear that PFOA induces oxidative
stress in hepatic tissues.
3.4.1.4 Evidence Integration
There is moderate evidence for an association between PFOA exposure and hepatic effects in
humans based on associations with liver biomarkers, especially ALT, in several medium
confidence studies. Across the studies in the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and
this updated systematic review, there is consistent evidence of a positive association between
exposure to PFOA and ALT in adults. An exposure-response gradient observed in one medium
quality study that examined categorical exposure in adults {Darrow, 2016, 3749173} increases
certainty in the association. These associations were observed in studies of the general
population, in communities with high exposure in water due to contamination events, and in
occupational studies. Consistency in the direction of association across these different population
sources increases certainty in the results and reduces the likelihood that they can be explained by
confounding across PFAS. For example, studies in communities with high exposure in water and
occupational participants are less susceptible to potential confounding from other PFAS due to
PFOA exposure predominating over other PFAS. In addition, the single general population that
performed multi-pollutant modeling {Lin, 2010, 1291111} found no attenuation of the
association, further increasing confidence in the association between PFOA exposure and
increased ALT. The positive associations with ALT are also supported by the recent meta-
analysis of 25 studies in adolescents and adults {Costello, 2022, 10285082}. Associations for
3-83
-------
DRAFT FOR PUBLIC COMMENT
March 2023
other hepatic outcomes were less consistent, including for functional outcomes such as liver
disease. This may be due to a relative lack of high confidence studies of these outcomes.
The animal evidence for an association between PFOA exposure and hepatic toxicity is robust
based on 25 high or medium confidence animal toxicological studies. However, it is important to
distinguish between alterations that may be non-adverse (e.g., hepatocellular hypertrophy alone)
and those that indicate functional impairment or lesions {U.S. EPA, 2002, 625713; FDA, 2009,
6987952; EMEA, 2010, 3056796; Hall, 2012, 2718645}. EPA considers responses such as
increased relative liver weight and hepatocellular hypertrophy adverse when accompanied by
hepatotoxic effects such as necrosis, inflammation, or biologically significant increases in
enzymes indicative of liver toxicity {U.S. EPA, 2002, 625713}. Many of the studies discussed in
this section reported dose-dependent increases in liver weight and hepatocellular hypertrophy in
rodents of both sexes. However, a limited number of these studies additionally examined
functional or histopathological hepatic impairment to provide evidence that the enlargement of
hepatic tissue was an adverse, and not adaptive, response {Minata, 2010, 1937251; Yan, 2014,
2850901; Crebelli, 2019, 5381564; Guo, 2019, 5080372; Blake, 2020, 6305864; Loveless, 2008,
7330145; NTP, 2020, 7330145}.
EPA identified the following studies as providing the most comprehensive evidence of dose-
dependent hepatoxicity resulting from oral PFOA exposure: a chronic dietary study in male and
female Sprague Dawley rats {NTP, 2020, 7330145} (see study design details in Section
3.4.4.2.1.2); a developmental study in male and female CD-I mice {Cope, 2021, 10176465}; and
a 29-day oral gavage study in male rats and mice {Loveless, 2008, 988599}. NTP (2020,
7330145) conducted histopathological examinations of liver tissue in male and female rats and
reported dose-dependent increases in the incidence of hepatocellular hypertrophy and
hepatocellular cytoplasmic vacuolation, as well as increases in the incidence of hepatocellular
single cell death and hepatocellular necrosis at the same dose levels. Cope et al. (2021,
10176465) also provides evidence of hepatic lesions in adult male and female CD-I mice
offspring exposed gestationally from GD 1.5-17.5. When the offspring were weaned, they were
placed on a low- or high-fat diet. At 18 weeks there were increases in the incidence and severity
of hepatocellular single cell death in females on either the low- or high-fat diets and males on the
low-fat diet. Loveless et al. (2008, 988599) similarly provides concurrent evidence of liver
enlargement and hepatic lesions in male mice gavaged with PFOA for 29 days. Increases in the
incidence and severity of hepatocellular hypertrophy and individual cell or focal cell necrosis
were dose-dependent. Similar to the NTP (2020, 7330145) study, Loveless et al. (2008, 988599)
provides a comprehensive report of hepatotoxicity, with a low dose range resulting in dose-
dependent increases in histopathological outcomes indicating adversity.
An important element of understanding the underlying mechanism(s) of toxicity is species-
specificity and relevance of data collected from laboratory models in relation to observed human
effects as well as in consideration of human hazard. There are several studies that have proposed
potential underlying mechanisms of the hepatotoxicity observed in rodents exposed to PFOA,
such as induction of hepatocytic proliferation leading to hypertrophy or nuclear receptor
activation leading to lipid droplet accumulation and steatosis. Generally, mechanistic evidence
supports the ability of PFOA to induce hepatotoxicity which may explain elevated serum ALT
levels in humans (and animals). However, mechanistic studies did not specifically relate (or,
"anchor") mechanistic data with serum ALT levels in animals, and challenges exist in the
3-84
-------
DRAFT FOR PUBLIC COMMENT
March 2023
extrapolation of evidence for PFOA-mediated changes in rodents to humans. For example, there
is substantial evidence that PFOA-induced liver toxicity, specifically alterations to lipid
metabolism and accumulation, occurs via the activation of multiple nuclear receptors, including
PPARa. Activation of PPARa by PFOA has been demonstrated in multiple studies across
various model systems, both in vivo and in vitro. Several studies examined the activation of
PPARa in vitro in both human and animal cell lines transfected with mouse and human PPARa
using luciferase reporter assays, the results of which demonstrate that PFOA can activate human
PPARa in vitro. In addition to PPARa, evidence also exists indicating that PFOA can activate
CAR, PXR, PPARy, ERa, and HNFa, as evidenced by receptor activation assays as well as
changes in target genes of these receptors. PFOA showed the highest potency for PPARa in
comparison to PPARy and PPAR8, although PFOA did activate these receptors at concentrations
of 100 |iM (compared to 25 [xM for PPARa). Like PPARa, PPARy and CAR are known to play
important roles in liver homeostasis, and dysregulation of these nuclear receptors can lead to
steatosis and liver dysfunction, potentially presenting an important mechanism for the liver
effects observed in rodent studies. Beyond receptor activation assays, individual target genes that
represent reliable markers of CAR and PPARa activation (e.g., Cyp2bl and Cyp4al,
respectively) have been clearly demonstrated to be altered by PFOA, and changes to these
nuclear receptors have important implications in regard to hepatotoxicity, specifically steatosis.
PPARa has vastly different expression in rodents compared to humans, and this species
difference is known to play a major role in differences in liver effects between the two species.
PPARa is the most demonstrated nuclear receptor to be activated by PFOA, and it should be
noted that using PPARa-null mice to study PPARa-independent effects of PFOA may lead to
compensatory mechanisms involving other nuclear receptors.
Another example of species-specificity for an effect of PFOA is the presence or absence of a
transfer protein that is important in cholesterol accumulation, CETP, which is expressed in
humans but not in rodents. Transgenic mice that express human CETP exhibit a more human-
like lipoprotein metabolism. Laboratory models that are designed to better predict human-
relevant mechanisms, such as mice expressing human CETP or PPARa, will continue to aid in
accuracy of the extrapolation of mechanistic findings in rodents to humans. Despite these
challenges, the evidence that PFOA leads to hepatotoxicity via activation of hepatic nuclear
receptors and dysregulation of lipid metabolism and accumulation is clear.
When considering the evidence from both in vivo and in vitro studies, PFOA-mediated
hepatoxicity specific to changes in lipid metabolism leading to steatosis, the most commonly
reported hepatocytic morphological alteration in PFOA-exposed animals, likely occurs through
the following molecular and cellular events: (1) PFOA accumulation in liver activates nuclear
receptors, including PPARa; (2) expression of genes involved in lipid homeostasis and
metabolism is altered by nuclear receptor activation; (3) gene products (translated proteins)
modify the lipid content of liver to favor triglyceride accumulation and potentially cholesterol
accumulation; (4) altered lipid content in the liver leads to accumulation of lipid droplets, which
can lead to the development of steatosis and liver dysfunction; and (5) alterations in lipid
metabolism lead to alterations in serum levels of triglycerides and cholesterol. Although
individual studies have not demonstrated every step of this proposed process, each event has
been demonstrated for PFOA, including steatosis in PFOA-exposed animals. It has also been
suggested that PFOA could interfere with fatty acid biosynthesis by binding to the Acetyl-CoA
carboxylase 1 and Acetyl-CoA carboxylase 2 enzymes; however, only a single study has
3-85
-------
DRAFT FOR PUBLIC COMMENT
March 2023
demonstrated such a binding event and further research is needed to understand the plausibility
of this binding occurring across species and exposure scenarios.
In addition (and potentially related) to the abundance of evidence related to hepatic nuclear
receptors, PFOA also alters apoptosis and cell proliferation in the liver. Specifically, PFOA
exposure at high doses causes apoptosis through a cascade of mechanisms including activation of
caspase activity, intracellular release of LDH, induction of apoptotic genes, morphological
changes to the mitochondria membrane, autophagy, and activation of the p53 mitochondria
pathway. PFOA has been shown to induce hepatocytic proliferation at low doses by disrupting
cell cycle progression, leading to steatosis, hepatomegaly, and liver dysfunction in general.
There are other mechanisms that may be involved in PFOA-induced hepatotoxicity, but the
evidence for such is limited and the relevance to liver outcomes is less clear. These include
hormone perturbation, inflammatory response, and oxidative stress. There are very limited data
demonstrating the potential of PFOA to perturb hormone balance, particularly related to thyroid
function. There are also a limited number of studies that reported inflammation in the liver,
including changes in cytokine levels and the expression of genes involved in innate immunity.
PFOA can cause oxidative stress in the liver, as demonstrated by standard indicators of oxidative
stress including increased ROS levels, changes in GSH and GSSG levels, and decreased total
antioxidant capacity in both in vivo and in vitro exposures to PFOA. The direct relevance of
oxidative stress to liver pathology induced by PFOA requires further study, but it is clear that
PFOA can cause oxidative stress. These other mechanisms that have a limited evidence base may
also occur in relation to the more well-characterized mechanisms of PFOA-induced
hepatotoxicity. For example, while the role of alterations in adaptive immunity in PFOA-induced
liver pathology is not clear, it is plausible that the inflammatory response is related to fatty liver
and associated liver dysfunction, such as the liver outcomes observed in humans and rodents,
that can occur via nuclear receptor-mediated pathways.
3.4.1.4.1 Evidence Integration Judgment
Overall, considering the available evidence from human, animal, and mechanistic studies, the
evidence indicates that PFOA exposure is likely to cause hepatotoxicity in humans under
relevant exposure circumstances (Table 3-3). This conclusion is based primarily on coherent
liver effects in animal models following exposure to doses as low as 0.3 mg/kg/day PFOA. In
human studies, there is consistent evidence of a positive association with ALT in adults, at
median PFOA levels as low as 1.3 ng/mL. The available mechanistic information provides
support for the biological plausibility of the phenotypic effects observed in exposed animals as
well as the activation of relevant molecular and cellular pathways across human and animal
models in support of the human relevance of the animal findings.
3-86
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 3-3. Evidence Profile Table for PFOA Hepatic Effects
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase Factors that Decrease
Certainty Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
Evidence from Studies of Exposed Humans (Section 3.4.1.1)
Serum biomarkers of
hepatic injury
9 Medium confidence
studies
4 Low confidence studies
Studies in adults
consistently reported
significant increases in
ALT (5/7). Findings for
AST and GGT in adults
were generally positive
(3/6). Some studies
reported conflicting or
non-significant
associations, however,
these were mostly of low
confidence. Findings for
liver enzymes in children
were mixed and different
by sex at times.
• Medium confidence
studies
• Consistent direction
of effect for ALT in
both the 2016
epidemiological
studies and the
updated literature
• Coherence of findings
between liver enzyme
increases
• Low confidence studies
• Inconsistent direction of
effect in children
Liver disease or injury
4 Medium confidence
studies
2 Low confidence studies
A limited number of
studies examined liver
disease or injury in general
population adults. One
study reported increased
risk of liver disease (1/2)
but was of low confidence,
whereas the only medium
confidence study reported
no significant association.
The only occupational
study reported
significantly higher
mortality from liver cancer
or cirrhosis compared to
the general population.
Other measures of
• No factors noted
»Association only
observed in Low
confidence studies
»Incoherence of findings
among measures of liver
inflammation
®©o
Moderate
0©O
'Evidence Indicates (likely)
Primary basis and cross-
stream coherence:
Human data indicated
consistent evidence of
Evidence for hepatic
effects is based on
increases in ALT in adults.1
Supporting evidence hepatoxicity as noted by
includes increases in other increased serum
liver enzymes such as ASTbiomarkers of hepatic
and GGT and increased ^ (Pnmanl-v ALT^
incidence of liver disease
mortality in occupational
settings. Minor
uncertainties remain
_regarding mixed liver
enzyme findings in
children and coherence of
with coherent results for
increased incidence of
hepatic nonneoplastic
lesions, increased liver
weight, and elevated serum
biomarkers of hepatic
injury in animal models.
liver enzyme and albumin Although a few
associations between other
findings.
serum biomarkers of
hepatic injury and PFOA
exposure were identified in
medium confidence
epidemiological studies,
there is considerable
uncertainty in the results
due to inconsistency across
studies.
Human relevance and
other inferences:
3-87
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
inflammation in the liver
were mixed and lacked
coherence.
Serum protein Significant increases in
3 Medium confidence albumin were consistently
studies observed in adults (4/5).
2 Low confidence studies Findings for total serum
protein and fibrinogen
were mixed or imprecise.
• Medium confidence
studies
• Consistent direction
of effect for albumin
• Low confidence studies
• Lmprecision of findings
Serum iron
1 Medium confidence
study
Only one large cross-
sectional study examined
serum iron concentrations
and reported a significant
positive association.
»Medium confidence
study
> Limited number of
studies examining
outcome
Evidence Integration
Summary Judgment
The available mechanistic
information overall provide
support for the biological
plausibility of the
phenotypic effects
observed in exposed
animals as well as the
activation of relevant
molecular and cellular
pathways across human
and animal models in
support of the human
relevance of the animal
findings.
3-88
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key Factors that Increase Factors that Decrease
Findings Certainty Certainty
Evidence Stream
Judgment
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.1.2)
Evidence Integration
Summary Judgment
Histopathology
3 High confidence
studies
9 Medium confidence
studies
Histopathological
alterations in liver were
observed in male and
female rodents exposed to
PFOA for various
durations (12/12).
Increased hepatocellular
hypertrophy (8/12) and
necrosis (5/12) were the
most common lesions.
Other lesions included
inflammation or cellular
infiltration (4/12),
cytoplasmic alteration or
vacuolation (3/12), mitosis
or mitotic figures (3/12),
bile duct hyperplasia
(2/12), cystic/cystoid
degeneration (2/12), fatty
change (2/12), and/or
pigment (1/12).
• High and medium
confidence studies
• Consistent direction
of effects across study
design, sex, and
species
• Dose-dependent
response
• Coherence of findings
in other endpoints
indicating liver
damage (i.e.,
increased serum
biomarkers and liver
weight)
• Large magnitude of
effect, with some
responses reaching
100% incidence in
some dose groups
(i.e., hypertrophy,
vacuolation, single
cell death) or are
considered severe
(i.e., cell or tissue
death/necrosis and
cystoid degeneration)
> No factors noted
Liver weight
5 High confidence
studies
19 Medium confidence
studies
Absolute (17/19) and
relative (17/20) liver
weights were increased in
male and female rodents
exposed to PFOA for
various durations. Several
• High and medium
confidence studies
• Consistent direction
of effects across study
> Confounding variables
such as decreases in
body weights
©0©
Robust
Evidence is based on 26
high or medium confidence
animal toxicological
studies indicating
increased incidence of
hepatic nonneoplastic
lesions, increased liver
weight, and elevated serum
biomarkers of hepatic
injury. However, it is
important to distinguish
between alterations that
may be non-adverse (e.g.,
hepatocellular hypertrophy
alone) and those that
indicate functional
impairment or lesions.
EPA considers responses
such as increased relative
liver weight and
hepatocellular hypertrophy
adverse when
accompanied by
hepatotoxic effects such as
necrosis, inflammation, or
.biologically significant
(i.e., greater than 100%
change) increases in
enzymes indicative of
hepatobiliary damage.
Many of the studies
3-89
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
studies that included both
males and females
suggested that males may
be more sensitive than
females (4/7).
design, sex, and
species
• Dose-dependent
response
• Coherence of effects
with other responses
indicating increased
liver size (e.g.,
hepatocellular
hypertrophy)
Serum biomarkers of
hepatic injury
3 High confidence
studies
7 Medium confidence
studies
Increases were observed in
ALT (6/9), AST (6/7),
ALP in (4/6), and GGT
(1/1). Biologically
significant changes
(>100%) in an enzyme
level were observed in 6/9
studies. Albumin (5/6) and
albumin/globulin ratio
(3/3) were increased. Bile
acids were increased in
males (4/4) and unchanged
in females (3/3).
Inconsistent changes in
bilirubin were observed
with direct bilirubin
increased in males (2/2) or
females (0/1), increased
indirect bilirubin in males
(1/1), and mixed effects on
total bilirubin in males (2)
and transient effects in
females (1). Total protein
was decreased in males
(3/5) and females (1/4).
• High and medium
confidence studies
• Consistent direction
of effects across study
design, sex, and
species
• Dose-dependent
response
• Coherence of findings
with other responses
indicating
hepatobiliary damage
(i. e.,
histopathological
lesions)
• Large magnitude of
effect, with evidence
of biologically
significant increases
(i.e., >100% control
responses) in serum
liver enzymes
indicating adversity
• Limited number of
studies examining
specific outcomes
discussed in this section
reported dose-dependent
increases in liver weight
and hepatocellular
hypertrophy in rodents of
both sexes. Although a
limited number of these
studies additionally
examined functional or
histopathological hepatic
Jmpairment, several
provide evidence of
adverse hepatic
responses.
3-90
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key Factors that Increase Factors that Decrease
Findings Certainty Certainty
Evidence Stream
Judgment
Mechanistic Evidence and Supplemental Information (Section 3.4.1.3)
Biological Events or
Pathways
Summary of Key Findings, Interpretation, and Limitations
Evidence Stream
Judgement
Molecular initiating
events - PPARa
Key findings and interpretation:
• Activation of human PPARa in vitro.
• Increased expression of PPARa-target genes in vitro in rat and human
hepatocytes, and cells transfected with rat or human PPARa.
• Altered expression of genes involved in lipid metabolism and lipid
homeostasis.
Limitations:
• Increased hepatic lipid content has also been reported for PFOA in the
absence of a strong PPARa response.
Overall, studies in rodent
and human in vitro and in
vivo models suggest that
PFOA induces hepatic
effects, at least in part,
through PPARa. The
evidence also suggests a
role for PPARa-
independent pathways in
the MOA for noncancer
liver effects of PFOA.
Molecular or cellular Key findings and interpretation:
initiating events - other . Increased apoptosis is a high dose effect demonstrated in vivo, as well as in
pathways
vitro, occurring through a cascade of mechanisms:
o activation of caspase activity, intracellular release of LDH, induction of
apoptotic genes, morphological changes to the mitochondria membrane,
autophagy, and activation of p53 mitochondria pathway.
• Inflammation of the liver (e.g., changes in cytokine levels and the expression
of genes involved in innate immunity) has been reported in a limited number
of studies.
• Induction of oxidative stress in vivo and in vitro, including increased ROS
levels, changes in GSH and GSSG levels, and decreased total antioxidant
capacity.
• Indirect evidence of activation of alternative pathways, including activation
of other nuclear receptors, primarily CAR and PPARy, following
observations in knockout or humanized PPARa mice.
Limitations:
• The direct relevance of oxidative stress to liver pathology induced by PFOA
requires further study.
Evidence Integration
Summary Judgment
3-91
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation Evidence Integration
~ ~ " ~ ~~ ~ " 7 ~ " ~~ ~7T! I Summary Judgment
Studies and Summary and Key Factors that Increase Factors that Decrease Evidence Stream
Interpretation Findings Certainty Certainty Judgment
• Very limited database for other pathways, with the exception of apoptosis
and cell cycle changes.
Notes: ALP = alkaline phosphatase; ALT = alanine transaminase; AST = aspartate transaminase; CAR = constitutive androstane receptor; EPA = Environmental Protection
Agency; GGT = gamma-glutamyl transferase; GSH = glutathione; GSSG = glutathione disulfide; LDH = lactate dehydrogenase; MOA = mode of action; PPARy = peroxisome
proliferator-activated receptor gamma; PPARa = peroxisome proliferator-activated receptor alpha; ROS = reactive oxygen species.
3-92
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2 immune
EPA identified 49 epidemiological and 13 animal toxicological studies that investigated the
association between PFOA and immune effects. Of the epidemiological studies, 1 was classified
as high confidence, 28 as medium confidence, 12 as low confidence, 6 as mixed (6 medium/low)
confidence, and 2 were considered uninformative (Section 3.4.2.1). Of the animal toxicological
studies, 3 were classified as high confidence, 9 as medium confidence, and 1 was considered
mixed {medium/low) confidence (Section 3.4.2.2). Studies have mixed confidence ratings if
different endpoints evaluated within the study were assigned different confidence ratings.
Though low confidence studies are considered qualitatively in this section, they were not
considered quantitatively for the dose-response assessment (Section 4).
3.4.2.1 Human Evidence Study Quality Evaluation and Synthesis
3.4.2.1.1 Immunosuppression
Immune function—specifically immune system suppression—can affect numerous health
outcomes, including risk of common infectious diseases (e.g., colds, influenza, otitis media) and
some types of cancer. The WHO guidelines for immunotoxicity risk assessment recommend
measures of vaccine response as a measure of immune effects, with potentially important public
health implications {WHO, 2012, 9522548}.
There are 11 epidemiological studies from the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}
that investigated the association between PFOA and immune effects. Study quality evaluations
for these 11 studies are shown in Figure 3-18.
3-93
-------
DRAFT FOR PUBLIC COMMENT
March 2023
,c®
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Costa et al„ 2009, 1429922-
i
+
I
+
I
+
I
+
I—
+
I—
+
I
~
Dong etal., 2013, 1937230-
+
++
+
+
++
+
+
n
Emmettet al., 2006, 1290905-
-
+
"
-
Fei etal., 2010, 1290805-
+
++
+
+
++
+
+
+
Grandjean et al., 2012, 1248827 -
+
+
++
+
+
+
+
+
Granum et al., 2013, 1937228-
+ *
+
+
-
+*
Humblet et al., 2014, 2851240 -
+
+
+
++
+
+
+
Looker et al., 2014, 2850913 -
+
+
++
+
+
+
+
+
Okada etal., 2012, 1332477-
+
+
+
+
+
+
+
+
Steenland et al., 2015, 2851015-
-
+
+
++
+
+
-
Wang etal., 2011, 1424977-
-
+
+
+
+
+
+
+
Figure 3-18. Summary of Study Quality Evaluation Results Epidemiology Studies of PFOA
and Immune Effects Published Before 2016 (References in 2016 HESD)
Interactive figure and additional study details available on HAWC.
Three studies reported decreases in response to one or more vaccines in relation to higher PFOA
exposure in children {Grandjean, 2012, 1248827; Granum, 2013, 1937228} and adults {Looker,
2014, 2850913}. Antibody responses for diphtheria and tetanus in children (n = 587) were
examined at multiple timepoints in a study on a Faroese birth cohort {Grandjean, 2012,
1248827}. Prenatal and age five serum PFOA concentrations were inversely associated with
3-94
-------
DRAFT FOR PUBLIC COMMENT
March 2023
childhood anti-diphtheria antibody response at all measured timepoints, and the association was
significant for anti-diphtheria antibody response at age seven in separate models for prenatal and
age five serum PFOA concentrations. The association was less pronounced when examining
anti-tetanus antibody responses in relation to prenatal PFOA measurements, but the anti-tetanus
antibody response (age seven) was significantly decreased in relation to PFOA measured in child
serum at five years of age. Prenatal PFOA exposure was associated with diminished vaccine
response in a different birth cohort study {Granum, 2013, 1937228}. Decreases in the anti-
rubella antibody response were significantly associated with elevated prenatal PFOA
concentrations among three-year-old children. A C8 Health Project study examining influenza
vaccine responses in highly exposed adults {Looker, 2014, 2850913} observed that pre-
vaccination PFOA concentrations were inversely associated with GM A/H3N2 antibody titer
rise, but no association was found with antibody titers for A/H1N1 and influenza type B. In the
studies of children, there was concern that the associations were also seen with other correlated
PFAS, but this was not considered a limitation in the study in adults, which was conducted in a
population with known high PFOA exposure (the C8 Health Project study).
Associations between prenatal PFOA exposure and risk of infectious diseases (as a marker of
immune suppression) were not seen in one study, although there was some indication of effect
modification by gender (i.e., associations seen in females but not in males). Fei et al. (2010,
1290805) examined hospitalizations for infectious diseases in early childhood in a Danish birth
cohort with mean maternal PFOA concentration of 0.0056 [j,g/mL. A slightly higher risk for
hospitalizations was observed in females whose mothers had higher PFOA concentrations
(incidence rate ratio [IRR] = -1.20, 1.63, 1.74 for quartile 2 [Q2], quartile 3 [Q3], and quartile 4
[Q4], respectively compared with quartile 1 [Ql]; see PFOA Appendix), and the risk for males
was below 1.0 for each quartile. Overall, there was no association between hospitalizations due
to infectious diseases and maternal PFOA exposure.
Overall, the 2016 PFOAHESD {U.S. EPA, 2016, 3603279} found consistent evidence of an
association between PFOA exposure and immunosuppression.
3.4.2.1.2 Immunosuppression Study Quality Evaluation and Synthesis from the
Updated Literature Review
There are 26 epidemiological studies identified from recent systematic literature search and
review efforts conducted after publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}
that investigated associations between prenatal, childhood, or adult PFOA exposure and
immunosuppression since publication of the 2016 PFOA HESD. Study quality evaluations for
these 26 studies are shown in Figure 3-19 and Figure 3-20.
One study from the 2016 assessment {Grandjean, 2012, 1248827} was updated during this
period, and the update was included in the systematic review {Grandjean, 2017, 3858518}.
3-95
-------
DRAFT FOR PUBLIC COMMENT
March 2023
&
,C.®
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Abraham et al., 2020, 6506041 -
+
+
%
-
-
+
+
-
Ait Bamai et al., 2020, 6833636 -
+
+
+
+
++
+
+
+
Bulka et al., 2021, 7410156-
++
+
+
+
+
+
+
+
Dalsager et al., 2016, 3858505 -
-
++
-
+
+
+
-
-
Dalsager et al., 2021, 7405343 -
+
+
+
+
+
+
+
+
Goudarzi et al., 2017, 3859808 -
++
+
+
+
+
+
+
+
Grandjean et al., 2017, 3858518 -
+
++
++
+
+
+
+
++
Grandjean et al., 2017, 4239492-
+
++
++
-
+
+
+
+
Grandjean et al., 2020, 7403067 -
-
++
+
+
+
+
+
+
Huang et al., 2020, 6988475 -
+
+
+
+
+
+
+
+
Impinen et al., 2018, 4238440-
+
+
-
+
+
+
-
-
Impinen et al., 2019, 5080609 -
++
++
-
+
++
+
-
-
Ji et al., 2021, 7491706-
-
+
+
+
+
+
-
+
Figure 3-19. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immunosuppression Effects
Interactive figure and additional study details available on HAWC.
3-96
-------
DRAFT FOR PUBLIC COMMENT
March 2023
A\o® .
&
Kielsen et al., 2016, 4241223-
-
-
+
+
+
~
Kvalem etal., 2020, 6316210-
+
+
-
+
++
+
+
"
Lopez-Espinosa et al., 2021, 7751049-
+
++
+
++
+
+
+
Manzano-Salgado et al,, 2019, 5412076 -
+
+ +
+
+
++
+
+
+
Mogensen et al., 2015, 3981889 -
+
++
+
+
++
+
+
+
Pilkerton et al., 2018, 5080265-
++
+ *
+
+ *
+
+
+
+ *
Shih etal,, 2021, 9959487-
+
++
++
+
++
+
+
+
Stein et al., 2016, 3860111 -
-
++
++
++
+
•
Timmermann et al., 2020, 6833710 -
+
+
+ *
+
++
+
"
+
Timmermann et al., 2021, 9416315-
+
+
+
+
++
+
+
+
Wang etal., 2022, 10176501 -
+
++
+
+
++
+
+
+
Zeng etal., 2019, 5081554-
-
+
++
+
+
-
Zeng et al., 2020, 6315718 -
+
+
+
++
+
+
-
Legend
B
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
B
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-20. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immunosuppression Effects (Continued)
Interactive figure and additional study details available on HAWC.
High and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies, the evidence synthesis below included details on any low confidence studies available.
Studies considered iminformative were not considered further in the evidence synthesis.
3-97
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.1.2.1 Vaccine Response
Nine studies (ten publications)11 studied the relationship between antibody response to
vaccination and PFOA exposure. Five of these studies (six publications) investigated antibody
response to vaccination in children {Timmermann, 2020, 6833710; Abraham, 2020, 6506041;
Grandjean, 2017, 3858518; Mogensen, 2015, 3981889; Grandjean, 2017, 4239492;
Timmermann, 2021, 9416315}. In adults, two studies investigated antibody response to
diphtheria and tetanus {Kielsen, 2016, 4241223; Shih, 2021, 9959487}, one study investigated
hepatitis vaccine response {Shih, 2021, 9959487}, one study investigated adult flu vaccine
response {Stein, 2016, 3860111}, and one study measured rubella antibodies in both adolescents
(aged 12 and older) and adults {Pilkerton, 2018, 5080265}. In addition to these studies on
vaccine response, one study {Zeng, 2019, 5081554} measured natural antibody response to
hand, foot, and mouth disease (HFMD), and one study {Zeng, 2020, 6315718} measured
antibody response to hepatitis B infection in adults. Overall, seven studies were medium
confidence {Grandjean, 2017, 3858518; Grandjean, 2017, 4239492; Timmermann, 2020,
6833710; Mogensen, 2015, 3981889; Pilkerton, 2018, 5080265; Shih, 2021, 9959487;
Timmermann, 2021, 9416315}, four were low confidence {Stein, 2016, 3860111, Zeng, 2019,
5081554; Zeng, 2020, 6315718; Abraham, 2020, 6506041}, and one study {Kielsen, 2016,
4241223} was uninformative.
Of the studies that measured antibody response to vaccination in children, four studies were
cohorts {Timmermann, 2020, 6833710; Grandjean, 2017, 3858518; Grandjean, 2017; 4239492;
Mogensen, 2015, 3981889}, and two were cross-sectional {Abraham, 2020, 6506041;
Timmermann, 2021, 9416315} (maternal serum was also available for a subset of participants in
Timmermann et al. (2021, 9416315)). These included multiple prospective birth cohorts in the
Faroe Islands, one with enrollment in 1997-2000 and subsequent follow-up to age 13
{Grandjean, 2017, 3858518} and one with enrollment in 2007-2009 and follow-up to age five
{Grandjean, 2017, 4239492}. One additional cohort in the Faroe Islands examined outcomes in
adults with enrollment in 1986-1987 and follow-up to age 28 {Shih, 2021, 9959487}. Five of
these studies measured antibody response to tetanus vaccination {Abraham, 2020, 6506041;
Grandjean, 2017; 3858518; Grandjean, 2017; 4239492; Mogensen, 2015, 3981889;
Timmermann, 2021, 9416315}; the same studies also measured antibody response to diphtheria
vaccination; one study measured antibody response to measles vaccination {Timmermann, 2020,
6833710}, and one study to Haemophilus influenza type b (Hib) antibodies {Abraham, 2020,
6506041}.
The results for this set of studies in children are shown in Table 3-4 and the Appendix (see
PFOA Appendix). The Faroe Islands studies {Grandjean, 2017, 3858518; Grandjean, 2017;
4239492; Mogensen, 2015, 3981889} observed associations between higher levels of PFOA and
lower antibody levels against tetanus and diphtheria in children at birth, 18 months, age 5 years
(pre-and post-booster), and at age 7 years, with some being statistically significant. These studies
measured PFOA exposure levels in maternal blood during the perinatal period and at later time
periods from children (at ages 5, 7, and 13 years). There are a few results in the opposite
direction for sub-analyses of the Faroe Island cohorts {Grandjean, 2017, 3858518; Grandjean,
2017; 4239492}, such as maternal PFOA exposure and anti-tetanus antibodies at 7 years (Table
11 Multiple publications of the same study: the study populations are the same in Grandjean et al. (2017, 3858518) and Mogensen
et al. (2015,3981889).
3-98
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3-4). No biological rationale has been identified as to whether one particular time period or
duration of exposure or outcome measurement is more sensitive to an overall immune response
to PFOA exposure.
It is plausible that the observed associations between decreased antibody concentration and
PFOA exposure observed in the Faroe Islands cohort could be partially explained by
confounding across the PFAS (e.g., exposure levels to PFOS were higher than PFOA (PFOS
17 ng/mL, PFOA 4 ng/mL); there was a moderately high correlation between PFOA and PFOS,
PFHxS, and PFNA (0.50, 0.53, 0.54, respectively) {Grandjean, 2017, 3858518; Grandjean, 2017,
4239492}). To investigate this, the authors assessed the possibility of confounding in a follow-up
paper {Budtz-Jorgensen, 2018, 5083631}. In these analyses, estimates were adjusted for PFOS
and there was no notable attenuation of the observed effects. The other available studies did not
perform multipollutant modeling, so it is difficult to determine the potential for highly correlated
PFAS to confound the effect estimates. However, as described above, one study {Looker, 2014,
2850913} observed an association with PFOA in a population where PFOA exposure
predominated (the C8 Health Project population), and this is not likely to be confounded by other
PFAS. Overall, the available evidence suggests that confounding is unlikely to explain the
observed effects.
Figure 3-21. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on Tableau.
3-99
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Effect Estimate
60 -50
Medium Grandjean el Ago S PFOA. Geometric Percent deference {per daubing Hui
confidence al. 2012 mean=4.Q6 nglinL (25th 75lh tn age 5 FTOA)
pereerilite=3J33-4-96 n^mL)
Adj for Age 5 Ab Age 5 Age 7 28.2
Age S Age 5 -13.3
Maternal PFOA: Geometric Percent difference {per daubing JJuf
mean-320 ligfai. (2Sth-75th in maternal PFOA i
DereenUle-2.5fl-4.01 ngTnL)
nl charge {per daubing af Cohort 3
Coticrt 3 and 5
Adj for Age 5 Ab Prenatal Age?
Post-booBer Prenatal Age 5
Pre bixreter Prenatal Age 5
Age 1.5 Age5
Age 5 Age 5
Cord Ago5
Age 1 Ji Age 5
Median = 2.2 nqimL (25th 75th Petcenl charge {per doubing of - . ,
percentile. 1.3 - 2.8 ngl'mLl PFOA I Cohort5
Median - 2.8 ngihiL (25th 75th Percent change ,'por Ooutoirig af _
percentile: 2.0 4.5 ngfmL) PFOA) umcrt s
Age 5 Age 5 25.26
Age 13 Age 5 16.31
Age 7 Age 7 20.5
Prenatal Age 7-12 -7
Age 7 12 Age 7 12 -8
Figure 3-22. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on Tableau.
Grandjean et al., 2012 was reviewed as a part of the 2016 HESD
3-100
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Exposure Outcome
Medium Grandjean et al. Age 5 PFOA. Geometric mean-4.Q8 Percent difference (per
confidence 2012 ngYnL t25lh-7S(h percentile^3.33-4.96 doubling ri age 5 PFOAl
ng*nL>
Adj for Age 5 Ab Age 5
Age 5 Age 5 6.8
Maternal PFOA. Geometric itkm*i=3.20 Percent difference jper Nul Prenatal Age 7
rtgtad. I2SSH-751H pcfcentiles2-5S-4.01 doubling iri msKmi PFOA I
nptnL)
Adj far Age 5 Ab Prenatal Age 7 16.8
Prenatal Age 5 16.2
Age 5 Age 5 -6.84
Cord blood Age S -35.17
Median ~ 2.2 ngftriL (25th • 75lh Percent change (per
percentile: 1.6 2.8 ngimLJ doubling t/ PFOSi
Median - 2.8 ngiVriL (25lh - 75fli Percent charge (per
percentile: 2.0 4 j ngftnL) doubling nf PFOA)
Cohort 5 AgeS Age5 18.31
Cotnrt 5 Age 1.5 Age 5 4.19
doubling of PFOA
je 7-12 Age 7-12 22
Figure 3-23. Overall Diphtheria Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on Tableau.
Grandjean et al., 2012 was reviewed as a part of the 2016 HESD
3-101
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 3-4. Associations between PFOA Exposure and Vaccine Response in Faroe Islands Studies
Exposure
measurement
timing, PFOA
levels (ng/mL)a
Diphtheria Antibody Associations with PFOA by Age at
Assessment
Tetanus Antibody Associations with PFOA by Age at
Assessment
5 years
(Pre-Booster)
(C3 and/or C5)
7 years
(C3 only)
13 years
(C3 only)
5 years
(Pre-Booster)
(C3 and/or C5)
7 years
(C3 only)
13 years
(C3 only)
Maternal
| (C3; age, sex)b
| (C3; age, sex,
| (C3; age, sex)b
| (C3; age, sex,
C3: GM: 3.20
BMD/BMDL (C3 &
booster type, and the
BMD/BMDL
booster type, and the
(2.56-4.01)
5; sex, birth cohort,
child's specific
(C3&5; sex, birth
child's specific
log-PFOA)°
antibody
cohort, log-PFOA)°
antibody
concentration at age
concentration at age
5 years)b
5 years)b
Birth
| (C3; age, sex)d
-
| (C3; age, sex)d
-
(modeled)
|| (C3 & 5; age,
|| (C3 & 5; age,
sex)d
sex)d
|| (C5; age, sex)d
|| (C5; age, sex)d
18 months
| (C3; age, sex)d
-
| (C3; age, sex)d
-
C3:NR
| (C3 & 5; age, sex)d
|| (C3 & 5; age,
C5: 2.8 (2.0-
| (C5; age, sex)d
sex)d
4.5)
|| (C5; age, sex)d
5 years
| (C3; age, sex)b
|| (C3; age, sex,
| (C3; age, sex)b
|| (C3; age, sex,
C3: GM: 4.06
| (C3; age, sex)d
booster type, and the
| (C3; age, sex)d
booster type, and the
(3.33-4.96)
| (C3 & 5; age, sex)d
child's specific
| (C3 & 5; age, sex)d
child's specific
C5: 2.2 (1.8-
| (C5; age, sex)d
antibody
|| (C5; age, sex)d
antibody
2.8)
concentration at age
concentration at age
5 years)b
5 years)b
BMD/BMDL (C3;
BMD/BMDL (C3;
sex, age, and booster
sex, age, and booster
type at age 5)e
type at age 5)e
BMD/BMDL (C3;
BMD/BMDL (C3;
sex, booster type at
sex, booster type at
age 5, log-PFOA)°
age 5, log-PFOA)°
3-102
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Exposure
measurement
timing, PFOA
levels (ng/mL)a
Diphtheria Antibody Associations with PFOA by Age at
Assessment
Tetanus Antibody Associations with PFOA by Age at
Assessment
5 years
(Pre-Booster)
(C3 and/or C5)
7 years
(C3 only)
13 years
(C3 only)
5 years
(Pre-Booster)
(C3 and/or C5)
7 years
(C3 only)
13 years
(C3 only)
7 years
C3: 4.4 (3.5
5.7)
IJ, (C3; age, sex, J, (C3; sex, age at
booster type)f antibody assessment,
I (C3; sex, age at booster type at age
antibody assessment, 5)g
booster type at age
51
I (C3; age, sex, | (C3; sex, age at
booster type)f antibody assessment,
| (C3; sex, age at booster type at age
antibody assessment, 5)g
booster type at age
51
13 years
C3: 2.0 (1.6
2.5)
I (C3; sex, age at
antibody assessment,
booster type at age
51
| (C3; sex, age at
antibody assessment,
booster type at age
51
Notes'. C3 = cohort 3, born 1997-2000; C5 = cohort 5, born 2007-2009; GM = geometric mean; NR = not reported.
Arrows indicate direction of association with PFOA levels; double arrows indicate statistical significance (p < 0.05) where reported. Arrows are followed by parenthetical
information denoting the cohort(s) studied and confounders (factors the models presented adjusted for).
a Exposure levels reported from serum as median (25th-75th percentile) unless otherwise noted.
bGrandjean et al. (2012,1248827); medium confidence
c Budtz-Jergensen and Grandjean (2018, 5083631); medium confidence
dGrandjean et al. (2017,4239492); medium confidence
e Grandjean and Budtz-Jergensen (2013,1937222); medium confidence
f Mogensen et al. (2015, 3981889); medium confidence
g Grandjean et al. (2017, 3858518); high confidence
3-103
-------
DRAFT FOR PUBLIC COMMENT
March 2023
A cross-sectional study of these antibody levels in Greenlandic children {Timmermann, 2021,
9416315} reported results that differed in direction of association based on the covariate set
selected. The exposure measurement in these analyses may not have represented an etiologically
relevant window; cross-sectional analyses in the Faroe Islands studies at similar ages also found
weaker associations than analyses for some other exposure windows. A subset of the study
population did have maternal samples available and those results were also inconsistent by
vaccine. However, this study was the only one to examine the OR for not being protected against
diphtheria (antibody concentrations, which has clear clinical significance, and they reported
elevated odds of not being protected (based on antibody concentrations <0.1 IU/mL, OR (95%
CI) per unit increase in exposure: 1.41 (0.91, 2.19)).
In children from Guinea-Bissau, West Africa, Timmermann et al. (2020, 6833710) observed
non-significant associations between elevated levels of PFOA and decreased adjusted anti-
measles antibody levels across time in the group with no measles vaccination at age 9 months.
This association was not seen in the group with one measles vaccination. The same pattern was
observed at the 2-year follow-up.
Lastly, the low confidence cross-sectional study of one-year-old children in Germany, Abraham
et al. (2020, 6506041), reported statistically significant correlations between PFOA
concentrations and adjusted levels of antibodies against tetanus, Hib, and diphtheria.
Of the three studies that measured vaccine response in adults or adolescents, two were cohorts
{Stein, 2016, 3860111; Shih, 2021, 9959487} and one was a cross-sectional analysis {Pilkerton,
2018, 5080265}. The medium confidence study by Shih et al. (2021, 9959487) measured PFOA
in cord blood and at multiple points through childhood to early adulthood in people in the Faroe
Islands, with outcome measurement at age 28 years. The study by Stein et al. (2016, 3860111)
was rated low confidence because it utilized convenience sampling to recruit participants, had
low seroconversion rates, and was at high risk of residual confounding. The study of the adult
population in Pilkerton et al. (2018, 5080265) suffered from potential exposure misclassification
due to concurrent exposure and outcome measurements and was also rated low confidence, but
this was less of a concern for the adolescent participants, so the study of this sub-population was
rated as medium confidence.
In adults and adolescents, results were less consistent than in children. Shih et al. (2021,
9959487) reported inverse associations for all exposure windows in the total cohort (not
statistically significant) for hepatitis B antibodies but for other vaccines (diphtheria, tetanus, and
hepatitis A), the direction of association was inconsistent across exposure windows. Results also
differed by sex for all vaccines, but without a consistent direction (i.e., stronger associations
were sometimes observed in women and sometimes in men). Similar to the results in 13-year-old
children in the other Faroe Islands cohorts, this may indicate that by age 28, the effect of
developmental exposure is less relevant. Pilkerton et al. (2018, 5080265) observed statistically
significant associations between high-quartile PFOA levels and decreased rubella IgA levels
compared with low-quartile PFOA levels in adult men but found no association between PFOA
exposure and anti-rubella antibody levels in adolescents. Stein et al. (2016, 3860111) reported no
immunosuppression based on seroconversion following FluMist vaccination.
Despite the imprecision (i.e., wide CIs) of some of the exposure-outcome analysis pairs, the
findings are generally consistent with respect to an association between PFOA exposure and
3-104
-------
DRAFT FOR PUBLIC COMMENT
March 2023
immunosuppression in children. Changes in antibody levels of 10%-20% per doubling of
exposure were observed in the Faroe Islands cohorts {Grandjean, 2017, 3858518; Grandjean,
2017, 4239492}. The variability in some of the results could be related to differences in
etiological relevance of exposure measurement timing, vaccine type, and timing of the boosters,
as well as differences in timing of antibody measurements in relation to the last booster.
However, these factors cannot be explored further with currently available evidence. Overall, the
evidence indicates an association between increased serum PFOA levels and decreased antibody
production following routine vaccinations, particularly in children.
In addition to these studies of antibody response to vaccination, there are two studies that
examined antibody response to HFMD {Zeng, 2019, 5081554} and hepatitis B infection {Zeng,
2020, 6315718}. This birth cohort study in China {Zeng, 2019, 5081554} measured antibody
levels in infants at birth and age 3 months, which represent passive immunity from maternal
antibodies. This study {Zeng, 2019, 5081554} was rated low confidence because the clinical
significance of the outcome is difficult to interpret in infants and there are concerns for
confounding by timing of HFMD infection as well as other limitations. Statistically significant
increased odds of HFMD antibody concentration below clinically protective levels per doubling
of PFOA were observed. This is coherent with the vaccine antibody results, but there is
uncertainty due to study deficiencies. Zeng et al. (2020, 6315718) observed negative associations
(p > 0.05) between serum PFOA concentration and hepatitis B surface antibody; however, there
are study limitations due to concurrent measurement of exposure and outcome and potential for
reverse causality, and this study was rated low confidence.
In a C8 Health Project study, Lopez-Espinoza et al. (2021, 7751049) measured serum PFAS and
white blood cell types in 42,782 adults in 2005-2006 and 526 adults in 2010 from an area with
PFOA drinking water contamination in the Mid-Ohio Valley (USA). Generally positive
monotonic associations between total lymphocytes and PFOA were found in both surveys
(difference range: 1.12%—5.50% for count and 0.36-1.24 for percentage, per PFOA IQR
increment). Findings were inconsistent for lymphocyte subtypes. However, the magnitude of the
differences was small.
3.4.2.1.2.2 Infectious Disease
Overall, ten studies (eleven publications)12 measured associations between PFOA exposure and
infectious diseases (or disease symptoms) in children with follow-ups between 1 and 16 years.
Infectious diseases measured included common cold, respiratory tract infections, respiratory
syncytial virus, otitis media, pneumonia, chickenpox (varicella), bronchitis, bronchiolitis, ear
infections, gastric flu, urinary tract infections, and streptococcus. Of the studies measuring
associations between infectious disease and PFOA exposure, eight (nine publications) were
cohorts {AitBamai, 2020, 6833636; Dalsager, 2016, 3858505; Dalsager, 2021, 7405343;
Kvalem, 2020, 6316210; Manzano-Salgado, 2019, 5412076; Gourdazi, 2017, 3859808; Impinen,
2019, 5080609; Wang, 2022, 10176501; Huang, 2020, 6988475}, one was a case control study
nested in a cohort {Impinen, 2018, 4238440}, and one was a cross-sectional study {Abraham,
2020, 6506041}. Six studies measured PFOA concentrations from mothers during pregnancy
{AitBamai, 2020, 6833636; Dalsager, 2016, 3858505; Manzano-Salgado, 2019, 5412076;
12 Multiple publications of the same study: both Dalsager et al. (2016, 3858505) and Dalsager et al. (2021, 7405343) use data
from the Odense cohort in Denmark and thus have overlapping, though not identical populations. They received different ratings
due to outcome ascertainment methods.
3-105
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Gourdazi, 2017, 3859808; Impinen, 2019, 5080609; Wang, 2022, 10176501}. Two studies
{Impinen, 2018, 4238440; Huang, 2020, 6988475} measured PFOA concentrations from cord
blood at delivery. Two studies measured PFOA concentrations in children's serum at age
one year {Abraham, 2020, 6506041} and at age 10 years {Kvalem, 2020, 6316210}.
Several of the studies measured infectious disease incidences as parental self-report, which may
have led to outcome misclassification {Kvalem, 2020, 6316210; Abraham, 2020, 6506041;
Impinen, 2018, 4238440; Impinen, 2019, 5080609}. Four studies measured infections as the
doctor-diagnosed incidence of disease over a particular period {Gourdazi, 2017, 3859808;
Manzano-Salgado, 2019, 5412076; AitBamai, 2020, 6833636; Huang, 2020, 6988475}, and
Wang et al. (2022, 10176501) used a combination of parental report and medical records. One
study used hospitalizations as an outcome, with events identified based on medical records
{Dalsager, 2021, 7405343}. Overall, six studies were medium confidence {Ait Bamai, 2020,
6833636; Goudarzi, 2017, 3859808; Manzano-Salgado, 2019, 5412076; Dalsager, 2021,
7405343; Wang, 2022, 10176501; Huang, 2020, 6988475} and five were low confidence
{Abraham, 2020, 6506041; Dalsager, 2016, 3858505; Impinen, 2018, 4238440; Impinen, 2019,
5080609; Kvalem, 2020, 6316210}.
Increased incidence of some infectious diseases in relation to PFOA exposure was observed,
although results were not consistent across studies (see PFOA Appendix). The most commonly
examined types of infections were respiratory, including pneumonia/bronchitis, upper and lower
respiratory tract, throat infections, and common colds. Dalsager et al. (2021, 7405343) reported
higher rates of hospitalization for upper and lower respiratory tract infections with higher PFOA
exposure (statistically significant only for lower respiratory tract). Among studies that examined
incidence, two studies (one medium and one low confidence) examining pneumonia/bronchitis
observed statistically significant associations between elevated PFOA concentrations and
increased risk of developing pneumonia in 0- to 3-year-old children {Impinen, 2019, 5080609}
and 7-year-old children {Ait Bamai, 2020, 6833636}; one other low and one other medium
confidence study did not report an increase in infections {Abraham, 2020, 6506041; Wang,
2022, 10176501}. Huang et al. (2020, 6988475), a medium confidence study, examined recurrent
respiratory infections and found no association. Two low confidence studies and one medium
confidence study found positive associations with lower respiratory tract infection {Kvalem,
2020, 6316210; Impinen, 2018, 4238440; Dalsager, 2021, 7405343}, while another medium
confidence study reported no association {Manzano-Salgado, 2019, 5412076}. In addition, non-
statistically significant positive associations were reported with upper respiratory tract infection
{Dalsager, 2021, 7405343} and throat infection {Impinen, 2019, 5080609}. There were also
statistically significant associations seen for PFOA in relation to respiratory syncytial virus,
rhinitis, throat infection, and pseudocroup {AitBamai, 2020, 6833636; Kvalem, 2020, 6316210;
Impinen, 2019, 5080609}, but findings were inconsistent across studies. No positive associations
were reported with common cold {Impinen, 2019, 5080609; Kvalem, 2020, 6316210}. Outside
of respiratory tract infections, two medium confidence studies examined total infectious diseases.
Dalsager et al. (2021, 7405343) reported higher rates of hospitalization for any infections with
higher PFOA exposure (not statistically significant), while Goudarzi et al. (2017, 3859808)
reported higher odds of total infectious disease incidence in girls (p > 0.05) but not boys. Results
for other infection types, including gastrointestinal, generally did not indicate a positive
association. Lastly, one study {Dalsager, 2016, 3858505} measured common infectious disease
symptoms in children aged 1 to 4 years and found a positive association with fever and nasal
3-106
-------
DRAFT FOR PUBLIC COMMENT
March 2023
discharge, but not cough, diarrhea, or vomiting. Overall, the observed associations provide some
coherence with the associations observed with vaccine response, but inconsistency across studies
reduces confidence in the evidence.
In addition to the studies in children, three studies examined infectious disease in adults, {Ji,
2021, 7491706; Grandjean, 2020, 7403067; Bulka, 2021, 7410156} (see PFOA Appendix). All
three studies were medium confidence. Ji et al. (2021, 7491706) was a case-control study of
COVID-19 infection. They reported higher odds of infection with higher PFOA exposure (OR
(95% CI) per log-2 SD increase in PFOA: 2.73 (1.71, 4.55)). In contrast, a cross-sectional study
examining severity of COVID-19 illness in Denmark using biobank samples and national
registry data {Grandjean, 2020, 7403067} reported no association between PFOA exposure and
increased COVID-19 severity. Bulka et al. (2021, 7410156) used NHANES data from 1999-
2016 in adolescents and adults and examined immunoglobulin G (IgG) antibody levels to several
persistent infections, including cytomegalovirus, Epstein Barr virus, hepatitis C and E, herpes
simplex 1 and 2, HIV, Toxoplasma gondii and Toxocara species. High levels of these antibodies
were interpreted as presence of a persistent infection. They found higher prevalence of herpes
simplex viruses 1 and 2 and total pathogen burden with higher PFOA exposure in adults but no
association with other individual pathogens.
3.4.2.1.3 Immune Hypersensitivity Study Quality Evaluation and Synthesis
from the Updated Literature Review
Another major category of immune response is the evaluation of sensitization-related or allergic
responses resulting from exaggerated immune reactions (e.g., allergies or allergic asthma) to
foreign agents {IPCS, 2012, 1249755}. A chemical may be either a direct sensitizer (i.e.,
promote a specific immunoglobulin E (IgE)-mediated immune response to the chemical itself) or
may promote or exacerbate a hypersensitivity-related outcome without evoking a direct response.
For example, chemical exposure could promote a physiological response resulting in a
propensity for sensitization to other allergens (e.g., pet fur, dust, pollen). Hypersensitivity
responses occur in two phases. The first phase, sensitization, is without symptoms, and it is
during this step that a specific interaction is developed with the sensitizing agent so that the
immune system is prepared to react to the next exposure. Once an individual or animal has been
sensitized, contact with that same or in some cases, a similar agent leads to the second phase,
elicitation, and symptoms of allergic disease. While these responses are mediated by circulating
factors such as T cells, IgE, and inflammatory cytokines, there are many health effects associated
with hypersensitivity and allergic response. Functional measures of sensitivity and allergic
response consist of health effects such as allergies or asthma and skin prick tests.
In the 2016 HESD for PFOA, two epidemiological studies reported higher odds of asthma with
higher PFOA exposure in children {Dong, 2013, 1937230; Humblet, 2014, 2851240}. A case-
control study {Dong, 2013, 1937230} of children in Taiwan reported increased odds of asthma
with increasing childhood PFOA exposure. The magnitude of association was particularly large
comparing each of the highest quartiles of exposure to the lowest. In cross-sectional analyses of
asthmatic children, the study authors reported monotonic increases for IgE in serum, absolute
eosinophil counts, eosinophilic cationic protein, and asthma severity score. A study on NHANES
(1999-2000, 2003-2008) adolescents also reported significantly increased odds of'ever asthma'
per doubling of concurrent PFOA measurements, where 'ever asthma' was defined as ever
having received an asthma diagnosis from a healthcare professional {Humblet, 2014 2851240}.
3-107
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Results were less consistent for measures of hypersensitivity (e.g., food allergy, eczema);
however, among female infants, decreased cord blood IgE {Okada, 2012, 1332477} was
significantly associated with prenatal PFOA exposure.
There are 24 epidemiological studies from recent systematic literature search and review efforts
conducted after publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} that
investigated the association between PFOA and hypersensitivity (i.e., asthma, allergy, and
eczema) effects. Study quality evaluations for these 24 studies are shown in Figure 3-24. High
and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies, the evidence synthesis below included details on any low confidence studies available.
Studies considered uninformative were not considered further in the evidence synthesis.
3-108
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Ait Bamai et al., 2020, 6833636 -
"
+
i
•
+
•
+
++
+
——
-
Averina et al., 2019, 5080647 -
+
-
-
+
+
-
-
Beck et al., 2019, 5922599-
+
++
-
+
+
+
+
+
Buser et al., 2016, 3859834 -
++
+
++
+
++
+
+
+
Chen et al., 2018, 4238372-
+
+
++
+
+
+
+
+
Gaylord et al., 2019, 5080201 -
+
+
B
+
+
+
+
Goudarzi et al., 2016, 3859523-
+
+
++
+
+
+
+
++
Impinen et al., 2018, 4238440 -
+
+
SS
+
+
+
-
Impinen et al., 2019, 5080609 -
Q
~
1
+
H
+
-
Jackson-Browne et al., 2020, 6833598-
S
+
++
+
+
+
Kvalem et al., 2020, 6316210 -
+
+
+*
+
-
-
+*
Manzano-Salgado et al., 2019, 5412076-
+
+
+
+
+
Shen et al., 2022, 10176753-
+
+
+
+
+
-
+
Smit et al., 2015, 2823268 -
+
++
++
+
++
+
+
+
Timmermann et al., 2017, 3858497-
+
++
2*
+
+
+
+
+*
Wen et al., 2019, 5081172-
+
+
+
+
+
+
+
+
Wen et al., 2019, 5387152-
+
+
+
++
+
+
+
Workman et al., 2019, 5387046 -
++
+
++
-
-
Xu et al., 2020, 6988472-
++
+
++
+
++
+
+
+
Zeng et al., 2019, 5081554-
+
-
++
+
+
-
Zeng et al., 2019, 5412431 -
+
++
+
+
+
+
+
Zhou et al., 2017, 3858488 -
+
+
+
+
+
-
Zhou et al., 2017, 3981296-
+
+
+
+
+
+
-
Zhu et al., 2016, 3360105-
+
+
+
+
-
Legend
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-24. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Immune Hypersensitivity Effects
Interactive figure and additional study details available on HAWC.
3-109
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Thirteen studies (fifteen publications)13 examined asthma (or asthma symptoms) and PFOA
exposure. Nine of these studies were cohorts {Averina, 2019, 5080647; Beck, 2019, 5922599;
Kvalem, 2020, 6316210; Manzano-Salgado, 2019, 5412076; Zeng, 2019, 5412431; Impinen,
2019, 5080609; Smit, 2015, 2823268; Timmermann, 2017, 3858497; Workman, 2019,
5387046}; three studies (five publications) were case-control investigations {Zhou, 2017,
3981296; Zhou, 2017, 3858488; Zhu, 2016, 3360105}, including one nested case-control,
{Gaylord, 2019, 5080201; Impinen, 2018, 4238440}; and one was a cross-sectional analysis
{Jackson-Browne, 2020, 6833598}. Seven studies measured the prevalence of "current" asthma
for at least one time point {Averina, 2019, 5080647; Beck, 2019, 5922599; Manzano-Salgado,
2019, 5412076; Kvalem, 2020, 6316210; Impinen, 2018, 4238440; Impinen, 2019, 5080609;
Zeng, 2019, 5412431}. Nine studies measured 'ever asthma' for at least one time point
{Averina, 2019, 5080647; Kvalem, 2020, 6316210; Manzano-Salgado, 2019, 5412076; Jackson-
Browne, 2020, 6833598; Gaylord, 2019, 5080201; Impinen, 2018, 4238440; Impinen, 2019,
5080609; Smit, 2015, 2823268; Timmermann, 2017, 3858497}. Incident or recurrent wheeze
was examined in one study {Workman, 2019, 5387046}. For asthma, ten publications were rated
medium confidence and five publications were rated low confidence (Figure 3-24). Timmermann
et al. (2017, 3858497) was low confidence for asthma because the questionnaire used to ascertain
status was not validated. Averina et al. (2019, 5080647) was considered low confidence because
results were not provided quantitatively. Two studies from the Genetic and Biomarker Study for
Childhood Asthma (GBCA) {Zhou, 2017, 3858488; Zhu, 2016, 3360105} were considered low
confidence based on participant selection. Cases and controls were recruited from different
catchment areas, and the resulting differences between cases and controls indicated potential for
residual confounding by age. Additionally, the timing of exposure assessment in relation to
outcome assessment was unclear, and it was not reported whether outcome status was confirmed
in controls.
Results across these studies were inconsistent (see PFOA Appendix), and few statistically
significant results were observed. Several studies observed positive associations with ORs
greater than 1.2 between PFOA concentration levels and increased "current" or "ever" asthma
{Beck, 2019, 5922599; Timmermann, 2017, 3858497; Jackson-Browne, 2020, 6833598;
Kvalem, 2020, 6316210; Zeng, 2019, 5412431; Averina, 2019, 5080647}, but often only within
population subgroups. Averina et al. (2019, 5080647) observed statistically significant increased
odds of self-reported doctor diagnosed asthma among adolescents in their first year of high
school. Beck et al. (2019, 5922599) observed statistically significant increased odds of self-
reported asthma per PFOA increase in boys, but this was not observed in girls. For doctor
diagnosed asthma in the same study, an inverse association (p > 0.05) was observed in boys and
a positive association (p > 0.05) was observed in girls. Kvalem et al. (2020, 6316210) reported
increased odds of asthma in girls at age 10 (p < 0.05) and between 10-16 years of age, but null
associations at 16 years, while the opposite was true for boys. Zeng et al. (2019, 5412431)
observed a positive association in girls and an inverse association in boys (both p > 0.05).
Jackson-Browne et al. (2020, 6833598) also observed statistically significant increased odds of
"ever" asthma from increased PFOA concentrations in children aged 3-5. However, these
associations were null in other age groups and in sex and race categories. Gaylord et al. (2019,
5080201) reported non-significant positive associations in youths of 13-22 years in age. The low
13 Three publications {Zhou, 2017, 3981296; Zhou, 2017, 3858488; Zhu, 2016, 3360105} reported on the same cohort (Genetic
and Biomarker study for Childhood Asthma) and outcome and are considered one study.
3-110
-------
DRAFT FOR PUBLIC COMMENT
March 2023
confidence study by Timmermann et al. (2017, 3858497) observed positive associations
(p < 0.05) between increased asthma odds and elevated PFOA concentrations in a small subset of
children aged 5 and 13 who did not receive their measles, mumps, and rubella (MMR)
vaccination before age 5. However, in children of the same ages who had received their MMR
vaccination before age 5, an inverse association was observed (p > 0.05). Low confidence studies
from the GBCA study {Zhou, 2017, 3858488; Zhu, 2016, 3360105} observed elevated PFOA
levels (p < 0.001) in children with asthma compared to those without {Zhou, 2017, 3981296},
and the odds of current asthma were also found to be elevated among boys and girls with
increasing PFOA exposure {Zhu, 2016, 3360105}. Two other studies {Impinen, 2018, 4238440;
Impinen, 2019, 5080609} observed small positive associations (OR: 1.1); in Impinen et al.
(2019, 5080609), this was only observed for current asthma in boys. Two studies reported non-
significant inverse associations with asthma {Manzano-Salgado, 2019, 5412076; Smit, 2015,
2823268}, and one low confidence study did not observe a significant effect for recurrent wheeze
{Workman, 2019, 5387046}.
In addition to the studies of asthma in children, one medium confidence study using data from
NHANES examined fractional exhaled nitric oxide (FeNO), a measure of airway inflammation,
in adults ({Xu, 2020, 6988472}; see PFOA Appendix). Among participants without current
asthma, this study found higher FeNO levels with higher PFOA exposure, indicating greater
inflammation (percent change (95% CI) for tertiles vs. Tl, T2: 5.29 (1.88, 8.81); T3: 6.34 (2.81,
10.01)).
Overall, there is some evidence of an association between PFOA exposure and asthma, but there
is considerable uncertainty due to inconsistency across studies and sub-populations. Sex-specific
differences were reported in multiple studies, but there was inconsistency in the direction of
association within each sex. There is not an obvious pattern of results by analysis of "ever" vs.
"current" asthma, and no studies beyond the Dong et al. (2013, 1937230) study described in the
2016 PFOA HESD examined asthma incidence.
Seven studies observed associations between PFOA exposure and allergies, specifically allergic
rhinitis or rhinoconjunctivitis, skin prick test, and food or inhaled allergies. Five of these studies
were cohorts {Goudarzi, 2016, 3859523; AitBamai, 2020, 6833636; Kvalem, 2020, 6316210;
Impinen, 2019, 5080609; Timmermann, 2017, 3858497}, one study was a case-control analysis
{Impinen, 2018, 4238440}, and one study was a cross-sectional study using data from NHANES
2005-2010 {Buser, 2016, 3859834}. One study was considered high confidence {Goudarzi,
2016, 3859523} and the rest were considered medium confidence for allergy outcomes. PFOA
concentrations were measured at a variety of time points: three studies measured PFOA during
pregnancy {Goudarzi, 2016, 3859523; AitBamai, 2020, 6833636; Impinen, 2019, 5080609};
three studies measured PFOA concentrations in children at age 5 years {Timmermann, 2017,
3858497}, age 10 years {Kvalem, 2020, 6316210}, age 13 years {Timmermann, 2017,
3858497} and ages 12-19 years {Buser, 2016, 3859834); and one study measured PFOA in cord
blood at delivery {Impinen, 2018, 4238440} (see PFOA Appendix).
Results were generally inconsistent across studies. Three studies conducted skin prick tests on
participants to determine allergy sensitization at age 10 years {Kvalem, 2020, 6316210; Impinen,
2018, 4238440}, at age 13 years {Timmermann, 2017, 3858497}, and at age 16 years {Kvalem,
2020, 6316210}. Skin prick tests were conducted to test sensitization to dust mites, pets, grass,
trees and mugwort pollens and molds, cow's milk, wheat, peanuts, and cod. Kvalem et al. (2020,
3-111
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6316210) reported a statistically significant but small association (OR: 1.1) with a positive skin
prick test at ages 10 and 16 years. Timmermann et al. (2017, 3858497) also reported a positive
association (p > 0.05) in children who had received an MMR before age 5 years (but an inverse
association in those who had not received an MMR) and results in Impinen et al. (2018,
4238440) were null. Five studies measured symptoms of "current" or "ever" allergic rhinitis or
rhinoconjunctivitis {Goudarzi, 2016, 3859523; AitBamai, 2020, 6833636; Impinen, 2018,
4238440; Kvalem, 2020, 6316210; Timmermann, 2017, 3858497}. Rhinitis was defined as at
least one symptom of runny or blocked nose or sneezing. Rhinoconjunctivitis was defined as
having symptoms of rhinitis, in addition to itchy and watery eyes. Rhinitis was increased with
exposure at age 16 years (p < 0.05) but decreased at age 10 years in Kvalem et al. (2020,
6316210). Non-significant increases in rhinitis were also reported in Impinen et al. (2018,
4238440) and Timmermann et al. (2017, 3858497), but results were null in Ait Bamai et al.
(2020, 6833636) and Goudarzi et al. (2016, 3859523) for rhinoconjunctivitis. Impinen et al.
(2019, 5080609) measured parent-reported, doctor-diagnosed "current" or "ever" allergy
symptoms at age 7 years in addition to known food and inhaled allergies and reported higher
odds of current food allergies and ever inhaled allergies (both p > 0.05), but not ever food
allergies or current inhaled allergies. Buser et al. (2016, 3859834) measured food sensitization
(defined as having at least one food-specific serum IgE > 0.35 kU/L) and self-reported food
allergies and reported statistically significant positive associations with self-reported food
allergies in NHANES 2007-2010 but not in in NHANES 2005-2006.
Seven studies measured the association between PFOA concentration and eczema (described by
some authors as atopic dermatitis). Six of these studies were cohorts {Goudarzi, 2016, 3859523;
Wen, 2019, 5387152; Wen, 2019, 5081172; Manzano-Salgado, 2019, 5412076; Chen, 2018,
4238372; Timmermann, 2017, 3858497}, and one was a case-control analysis {Impinen, 2018,
4238440}. Four studies measured PFOA concentrations in cord blood at delivery {Wen, 2019,
5387152; Wen, 2019, 5081172; Chen, 2018, 4238372; Impinen, 2018, 4238440}, three studies
measured maternal PFOA concentrations during pregnancy {Goudarzi, 2016, 3859523;
Manzano-Salgado, 2019, 5412076; Timmermann, 2017, 3858497}, and one study measured
PFOA concentrations in children at age 5 and 13 years {Timmermann, 2017, 3858497}. All of
the studies were considered medium confidence for eczema (see PFOA Appendix).
Two studies (three publications) observed statistically significant associations between increased
odds of eczema within the highest quantiles of PFOA exposure {Wen, 2019, 5387152; Wen,
2019, 5081172; Chen, 2018, 4238372}; however, the associations were non-monotonic across
categories of exposure. Impinen et al. (2018, 4238440) also observed a non-significant
association between higher PFOA concentrations and "ever" eczema at age 2 years; however,
results were null for "current" eczema at age 10 years. Results from Goudarzi et al. (2016,
3859523), Manzano-Salgado et al. (2019, 5412076) and Timmermann et al. (2017, 3858497)
were null.
One medium confidence nested case-control study examined chronic spontaneous urticaria
{Shen, 2022, 10176753}. They found no association between PFOA exposure and case status.
3-112
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.1.4 Autoimmune Disease Study Quality Evaluation and Synthesis from
the Updated Literature Review
Autoimmunity and autoimmune disease arise from immune responses against endogenously
produced molecules. The mechanisms of autoimmune response rely on the same innate and
adaptive immune functions that respond to foreign antigens: inflammatory mediators, activation
of T lymphocytes, or the production of antibodies for self-antigens {IPCS, 2012, 1249755}.
Chemical exposures that induce immune response or immunosuppression may initiate or
exacerbate autoimmune conditions through the same functions. Autoimmune conditions can
affect specific systems in the body, such as the nervous system (e.g., multiple sclerosis (MS)), or
the effects can be diffuse, resulting in inflammatory responses throughout the body (e.g., lupus).
The 2016 HESD for PFOA {U.S. EPA, 2016, 3603279} identified one occupational study in
workers highly exposed to PFOA (part of the C8 Health Project) {Steenland, 2015, 2851015}
that reported significant positive trends for rheumatoid arthritis and ulcerative colitis with
increasing cumulative PFOA exposure. The C8 Science Panel concluded there was a probable
link between PFOA and ulcerative colitis {C8 Science Panel, 2012, 1430770}.
There are 6 epidemiological studies from recent systematic literature search and review efforts
conducted after publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} that
investigated the association between PFOA and autoimmune disease. Study quality evaluations
for these 6 studies are shown in Figure 3-25. High and medium confidence studies were the focus
of the evidence synthesis for endpoints with numerous studies, though low confidence studies
were still considered for consistency in the direction of association (and details are provided in
PFOA Appendix). For endpoints with fewer studies, the evidence synthesis below included
details on any low confidence studies available. Studies considered uninformative were not
considered further in the evidence synthesis.
3-113
-------
DRAFT FOR PUBLIC COMMENT
March 2023
r\N0
-------
DRAFT FOR PUBLIC COMMENT
March 2023
disease, resulting in some concern for reverse causation. Additionally, there was potential for
residual confounding by SES which was not considered in the analysis. These factors together
contributed to a low confidence rating. Information on participant selection, particularly control
selection, was not reported in Ammitzb0ll (2019, 5080379). Additionally, PFOA was evaluated
as a dependent rather than independent variable, making no informative determinations about
associations between PFOA exposure and risk of MS.
In a C8 Health Project study {Steenland, 2013, 1937218}, associations for rheumatoid arthritis
were generally consistent and positive across untagged and 10-year lagged PFOA quartiles. The
risk of rheumatoid arthritis was significantly elevated comparing those in the third quartile of 10-
year lagged exposure to participants in the first quartile, but this was the only significant
association. The risk of MS was non-significantly elevated in untagged and 10-year lagged
models {Steenland, 2013, 1937218}. Significantly increased risk of ulcerative colitis among
adults across increasing quartiles of PFOA exposure was also observed (p-trend < 0.0001).
Associations with lupus and Crohn's disease were non-significant and inconsistent in the
direction of effect {Steenland, 2013, 1937218}.
Evidence from a case-control study suggested lower PFOA concentrations among healthy
controls compared to those with MS {Ammitzb0ll, 2019, 5080379}. Serum PFOA
concentrations were 12% lower (95% CI: -24%, 2%; p = 0.099) in healthy controls compared to
cases of relapsing remitting MS and clinically isolated MS. Restricting the analysis to men,
serum PFOA levels were 28% lower (95% CI: -42%, -9%; p = 0.006) in healthy controls
compared to cases, but this effect was not seen in women. Steenland et al. (2018, 5079806)
detected significantly increased levels of PFOA in ulcerative colitis cases vs. those with Crohn's
disease or controls and observed statistically significantly increased odds of ulcerative colitis
with increased PFOA exposure among combined children and adults; however, the trend was not
consistent across increasing quintiles of PFOA exposure, with a peak in the third quintile. Xu et
al. (2020, 6315709) observed significant decreases in risk of Crohn's disease in an early
exposure period, but not in later exposure periods, or for UC in children and adults from a high-
exposure community in Sweden (Ronneby cohort).
The risk of celiac disease was elevated among children and young adults (<21 years old) in a
case-control study {Gaylord, 2020, 6833754}, particularly in females (p < 0.05), but the
association did not reach significance among the whole population.
In the prospective observational Finnish Diabetes Prediction and Prevention (DIPP) study in
which children genetically at risk to develop type 1 diabetes (T1D) and celiac disease were
followed from birth, with blood samples taken at birth and 3 months of age {Sinisalu, 2020,
7211554}, there was no significant difference in the levels of PFOA exposure in those children
that later developed celiac disease, which may be due to the small sample size, but age at
diagnosis of celiac disease was strongly associated with the PFOA exposure.
3.4.2.2 Animal Evidence Study Quality Evaluation and Synthesis
There are 4 studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and 9 studies from
recent systematic literature search and review efforts conducted after publication of the 2016
PFOA HESD that investigated the association between PFOA and immune effects in animal
models. Study quality evaluations for these 13 studies are shown in Figure 3-26.
3-115
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Butenhoff et al., 2004, 1291063-
++
NR
NR
++
++
++
++
++
Butenhoff et al., 2012, 2919192 -
+
++
NR
-
+
++
++
B
Crebelli et al., 2019, 5381564 -
+
+
NR
++
+
+
+
+
+
+
De Guise et al., 2021, 9959746 -
+
+
NR
+
+
+
++
++
++
+
Dewitt et al., 2008, 1290826 -
+
+
NR
+
+
+
+
++
++
+
Guo et al., 2019, 5080372-
+
+
NR
++
++
B
a
B
Guo et al., 2021, 7542749-
+
+
NR
D
++
Huet al., 2010, 1332421 -
++
NR
NR
++
¦
++
¦
++
Hu et al., 2012, 1937235-
+
+
NR
++
++
++
++
++
Loveless et al., 2008, 988599 -
+
+
NR
++
++
++
++
B
NTP, 2019, 5400977-
++
++
NR
++
++
++
a
++
++
++
NTP, 2020, 7330145-
++
++
NR
++
++
++
++
++
++
++
Shi et al., 2020,7161650-
+
+
+
+
IL
D
++
n
++
B
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Not reported
Multiple judgments exist
Figure 3-26. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Immune Effects
Interactive figure and additional study details available on HAWC.
The data available on immunological responses of animals following oral exposure to PFOA are
extensive, especially as they apply to mice. A number of studies reported effects on spleen and
3-116
-------
DRAFT FOR PUBLIC COMMENT
March 2023
thymus weights, immune system cellular composition, and the ability to generate an immune
response following PFOA doses ranging from approximately 1-40 mg/kg/day.
3.4.2.2.1 Organ Weight/Histopathology
Short-term exposure studies by Yang et al. (2000, 699394), Yang et al. (2001, 1014748), Qazi et
al. (2009, 1937259), and Yang et al. (2002, 1332453) using male C57BL/6 mice, by DeWitt et
al. (2008, 1290826) using female C57BL/6 mice, and by DeWitt et al. (2016, 2851016) using
female C57BL/6Tac mice were conducted using relatively high PFOA doses (up to
approximately 40 mg/kg/day). In each study, the PFOA-treated C57BL/6 mice exhibited
significant reductions in spleen and thymus weights after 5-16 days of exposure. Yang et al.
(2000, 699394) and DeWitt et al. (2008, 1290826) observed up to an approximately 80%
reduction in absolute and relative thymus weight and up to a 30%-48% reduction in absolute and
relative spleen weight. Similar reductions in absolute thymus and spleen weights were observed
in Yang et al. (2002, 1332453); relative weights were not reported. In DeWitt et al. (2016,
2851016), relative spleen weights were significantly reduced by 30% after exposure to
30 mg/kg/day, and relative thymus weights were significantly reduced by 55.4% after exposure
to 7.5 mg/kg/day (but not after exposure to 30 mg/kg/day). Absolute weights were not reported
in this study. In male CD-I mice exposed for 29 days via gavage to 1, 10, or 30 mg/kg/day
PFOA, absolute and relative spleen weights were reduced to approximately 90%, 60%, and 50%
of controls, respectively {Loveless, 2008, 988599}. Absolute and relative thymus weights were
decreased to approximately 50% of controls in the 10 and 30 mg/kg/day groups. Spleen and
thymus weights were only reduced by up to 9% (not statistically significant) in male ICR mice
administered 47.21 mg/kg/day PFOA in drinking water for 21 days {Son, 2009, 1290821}. In
male BALB/c mice dosed with 0.4, 2, or 10 mg/kg/day PFOA via gavage for 28 days, absolute
spleen weights were significantly reduced to 88% and 50% of the control in the 2 and
10 mg/kg/day groups, respectively {Guo, 2021, 7542749}. Relative spleen weights in these
groups were similarly reduced to 84% and 56% of the control. In the same study, however, no
significant changes in spleen or thymus weights were observed in male Sprague Dawley rats. In
a separate 28-day study, male Sprague Dawley rats administered 2.5-10 mg/kg/day displayed
significantly lower absolute spleen weights that reached 76% of control at the highest dose
{NTP, 2019, 5400977}. Absolute thymus weight was decreased to 74% of control in males
administered 10 mg/kg/day compared to those of the vehicle group. Female spleen and thymus
weights were not altered.
In one developmental study, pregnant C57BL/6N mice were exposed to 0.5 or 1 mg/kg/day
PFOA from GD 6-17; the relative spleen and thymus weights of the female offspring were
unchanged at PND 48 {Hu, 2010, 1332421}. The male offspring were not assessed in this study.
However, a reduction in spleen and thymus weights has been reported in male rats following
developmental PFOA exposure. NTP (2020, 7330145) exposed pregnant rats to PFOA beginning
on GD 6, and exposure was continued in offspring postweaning for a total of 107 weeks. Dose
groups for this report are referred to as "[perinatal exposure level (ppm)]/[postweaning exposure
level (ppm)]" (see further study design details in Section 3.4.4.2.1.2). Following perinatal and
postweaning PFOA exposure (150/150 and 300/300 ppm), significant reductions in absolute and
relative spleen weight and absolute thymus weight were observed at 16 weeks in male rats.
Reduced absolute and relative spleen weights were also observed in rats following 300/20,
300/40, and 300/80 ppm PFOA exposure. Postweaning exposure alone (0/20, 0/40, 0/150, and
0/300 ppm) significantly reduced absolute and relative spleen weights. Absolute thymus weight
3-117
-------
DRAFT FOR PUBLIC COMMENT
March 2023
was reduced following 0/150 and 0/300 ppm {NTP, 2020, 7330145}. No changes in spleen or
thymus weights were reported in females.
Two studies describing effects of subchronic PFOA exposure in adult male mice {Crebelli, 2019,
5381564; Shi, 2020, 7161650} and one chronic study in adult male rats {Butenhoff, 2012,
2919192} did not report reduced spleen weight, and thymus weights were not examined. No
changes to spleen weights were observed in C57BL/6 male mice administered <5 mg/kg/day for
5 weeks {Crebelli, 2019, 5381564; Shi, 2020, 7161650}. Although the changes were not
statistically significant, Shi et al. (2020, 7161650) observed 21%, 32%, and 32% reductions in
relative spleen weight (compared to controls) in mice exposed to 0.5, 1, or 3 mg/kg/day,
respectively. Body weight gain was also significantly reduced in these groups, and absolute
spleen weight was not reported. Similarly, spleen weight was not affected in male Sprague-
Dawley rats chronically exposed to 30 or 300 ppm (1.3 or 14.2 mg/kg/day) for 1 or 2 years
{Butenhoff, 2012, 2919192}. An increase in absolute and relative spleen weight (40% and 30%
increase, respectively) was observed only in female rats exposed to 30 ppm (1.6 mg/kg/day) for
2 years.
3.4.2.2.2 Histopathology
Several studies reported on histological evaluations of the spleen and thymus from rodents orally
administered PFOA at varying doses and durations. In male Crl:CD-l (ICR)BR mice
administered PFOA for 29 days, decreased spleen weights at 10 and 30 mg/kg/day correlated
with the gross observation of small spleens {Loveless, 2008, 988599}. An increased incidence of
spleen atrophy was also observed in the 30 mg/kg/day group. The decreased thymus weights at
these doses correlated with the microscopic finding of lymphoid depletion and with the gross
observation of small thymuses {Loveless, 2008, 988599}. Loveless et al. (2008, 988599) also
reported increased incidences of granulocytic hyperplasia of the bone marrow in mice in the 10
and 30 mg/kg/day groups.
Other microscopic findings were reported in Son et al. (2009, 1290821) in the histological
evaluation of male ICR mice administered PFOA (0.49-47.21 mg/kg/day) for 21 days. The
thymus of mice exposed to 47.21 mg/kg/day PFOA revealed atrophy with decreased thickness of
the cortex and medulla compared to control, but increased cellular density of lymphoid cells in
the cortex was observed {Son, 2009; 1290821}. The authors also reported an enlargement of the
spleen with marked hyperplasia of the white pulp in the 47.21 mg/kg/day PFOA-treated group,
and an increased area of the lymphoid follicles in the spleen with increased cellular density {Son,
2009, 1290821}. In contrast, in a study in male BALB/c mice administered 0.4-10 mg/kg/day
PFOA via gavage, the authors noted decreased white pulp content, with the white pulp content in
the highest dose group being reduced to nearly in half of that of the control group (quantitative
results were not provided) {Guo, 2021, 7542749}.
After 5-6 days of recovery, Loveless et al. (2008, 988599) observed increased extramedullary
hematopoiesis in the spleens of male Crl:CD(SD)IGS BR rats and Crl:CD-l (ICR)BR mice
exposed to 30 mg/kg/day PFOA for 23-24 days. However, these changes were not observed in
rats and mice after a continuous 29-day exposure {Loveless, 2008, 988599}. Likewise, splenic
hematopoiesis was not affected in male or female Sprague-Dawley rats administered 0.625-10 or
6.25-50 mg/kg/day PFOA, respectively {NTP, 2019, 5400977}.
3-118
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Two studies in male Sprague-Dawley rats exposed to up to 30 mg/kg/day PFOA for 28-29 days
reported no histopathological changes in the spleen, thymus, and/or lymph nodes {Loveless,
2008, 988599; NTP, 2019, 5400977}. However, a significant increase in bone marrow
hypocellularity of minimal severity was reported in male rats exposed to 10 mg/kg/day (6/10
compared to 1/10 in controls) but not in female rats {NTP, 2019, 5400977}.
Histological evaluation of the spleen following chronic PFOA exposure was only reported in one
study, which administered 30 or 300 ppm PFOA to male and female Sprague-Dawley rats for
2 years. Hemosiderin, an iron-rich pigment, was found in greater amounts in the spleens of males
dosed with 300 ppm (approximately 15 mg/kg/day), though this change was not significant, but
was significantly reduced in the 30 ppm groups (approximately 1.5 mg/kg/day) and in the 300
ppm females {Butenhoff, 2012, 2919192}. However, no histopathological changes in the
thymus, spleen, bone marrow, or lymph nodes were reported in a study that exposed Sprague-
Dawley rats to up to 300 ppm PFOA for 16 weeks (males and females) or up to 80 ppm PFOA
(males) or 300 ppm (females) for 2 years {NTP, 2020, 7330145}.
Histological evaluation of the spleen and thymus following reproductive PFOA exposure was
only reported in one study {Butenhoff, 2004, 1291063}. Po males and females were administered
1-30 mg/kg/day PFOA from premating until the end of lactation and the Fi generation was
exposed throughout their life. The authors note that no histopathological changes were reported,
though qualitative results were not provided.
3.4.2.2.3 Immune Cellularity
3.4.2.2.3.1 White Blood Cells and Differentials
Evidence supporting an effect of PFOA exposure on immune system-associated cellularity has
been reported. A decrease in total serum white blood cells to 28% of control was observed in
male C57BL/6 (H-2b) mice fed 40 mg/kg/day for 10 days {Qazi, 2009, 1937259}. Total number
of circulating neutrophils and lymphocytes (T and B cells) were decreased to 50% and 27% of
control, respectively. The numbers of circulating monocytes, eosinophils, and basophils were too
small to be determined reliably, according to the study {Qazi, 2009, 1937259}.
In a similar study, male Crl:CD-l(ICR)BR mice were exposed to PFOA (10 or 30 mg/kg/day) by
oral gavage for 29 days. At both doses tested, increases in total serum neutrophils and monocytes
(reaching 296% and 254% of control, respectively, at the highest dose), and a decrease in total
number of eosinophils (approximately 60% of control, data not statistically significant) were
observed {Loveless, 2008, 988599}. Loveless et al. (2008, 988599) also reported a decrease in
lymphocytes in male mice dosed with 30 mg/kg/day, but these data were not provided in the
study. In a second short-term study, white blood cell count was significantly decreased to 71%
and 36%) of the control in male BALB/c mice exposed to 2 and 10 mg/kg/day PFOA,
respectively, for 28 days {Guo, 2021, 7542749}. White blood cell differentials were not
measured in this study.
In a short-term study in male and female Sprague-Dawley exposed to 0.625-10 or 6.25-100
mg/kg/day PFOA, respectively, no changes in white blood cell counts or differentials were
reported {NTP, 2019, 5400977}.
3-119
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In male and female Sprague-Dawley rats chronically exposed to 30 or 300 ppm PFOA
(approximately 1.5 or 15 mg/kg/day) for 2 years, PFOA did not affect total white blood cell
count, blood lymphocytes, or neutrophils {Butenhoff, 2012, 2919192}. However, white blood
cell counts were increased in males through the first year of the study. The authors suggest that
these changes were due to increases in absolute counts of lymphocytes at 3 and 6 months and in
neutrophils at 12 months {Butenhoff, 2012, 2919192}.
3.4.2.2.3.2 Spleen, Thymus, Lymph Nodes, and Bone Marrow Cellularity
Short-term PFOA exposure (10-40 mg/kg/day) significantly decreased splenocyte and
thymocyte cell populations by up to approximately 30% and 15% of control, respectively, in
male Crl:CD-l (ICR)BR mice {Loveless, 2008, 988599} and male C57BL/6 mice {Yang, 2001,
1014748}. Similarly, in male C57BL/6 mice administered 40 mg/kg/day PFOA for 7 days, the
number of thymocytes was decreased to 14% of control; immature thymocyte populations
(CD4+CD8+) were the most affected {Yang, 2000, 699394}. In the spleen, both B and T cells
were significantly reduced in these mice, and the number of total splenocytes was decreased to
20% of control {Yang, 2000, 699394}. Reduced splenocyte and thymocyte CD4+CD8+ cells
were also observed in male ICR mice administered PFOA (0, 0.49, 2.64, 17.63, and
47.21 mg/kg/day) in drinking water for 21 days, reflecting an impairment in cell maturation
{Son, 2009, 1290821}.
No changes in splenocyte and thymocyte cell populations were observed in one study of male
Sprague-Dawley rats exposed to 0.3-30 mg/kg/day PFOA for 29 days {Loveless, 2008,
988599}.
Developmental PFOA exposure may also impact cellularity of the spleen. In one study by Hu et
al. (2012, 1937235), an approximate 22% reduction in splenic regulatory T cells
(CD4+CD25+Foxp3+) was observed in PND 42 male and female offspring from C57BL/6N dams
exposed to 2 mg/kg/day PFOA from gestation through lactation. Thymic cellularity was not
examined in this study {Hu, 2012, 1937235}.
3.4.2.2.4 Ability to Generate an Immune Response
The ability to generate an immune response following PFOA has been investigated in rodent
models. Male Crl:CD-l (ICR)BR mice were exposed to PFOA (0, 0.3, 1, 10, or 30 mg/kg/day)
by oral gavage for 29 days and received an injection of serum sheep red blood cells (SRBC) on
day 24 {Loveless, 2008; 988599}. The induced immunoglobulin M (IgM) response was
significantly reduced to 80% and 12% of controls in mice exposed to 10 and 30 mg/kg/day,
respectively. The same study found no changes in IgM in rats. After an injection with keyhole
limpet hemocyanin (KLH), a similar reduction in anti-KLH IgM response was observed in
female B6C3F1 mice administered 1.88 and 7.5 mg/kg/day PFOA in drinking water for 28 days
{De Guise, 2021, 9959746}. The IgM response in the mice exposed to 1.88 or 7.5 mg/kg/day
was significantly reduced to 29% and 8% of the control's response, respectively. The ability to
respond to an immunological challenge was also reduced in female C57BL/6N mice exposed to
3.75 to 30 mg/kg/day PFOA in drinking water for 15 days {DeWitt, 2008, 1290826}. The mice
showed a dose-dependent reduction in IgM levels (between 11% and 30% decrease) after
injection with SRBC to induce an immune response. The IgG response to SRBC significantly
increased by approximately 15% following 3.75 and 7.5 mg/kg/day PFOA exposure, but no
change was observed at higher doses {DeWitt, 2008, 1290826}. In a separate study, female
3-120
-------
DRAFT FOR PUBLIC COMMENT
March 2023
C57BL/6Tac mice were exposed to 0, 7.5, or 30 mg/kg/day PFOA in drinking water for 15 days
and injected with SRBC on day 11 {DeWitt, 2016; 2851016}. Exposure to 30 mg/kg/day PFOA
reduced SRBC-specific IgM antibody responses by 16%. Similarly, male C57BL/6 mice were
fed approximately 40 mg/kg/day PFOA for 10 days and then evaluated for their immune
response to horse red blood cells {Yang, 2002, 1332454}. PFOA-exposed mice were unable to
produce an increase in plaque-forming cells in response to the immune challenge, compared to
control mice, suggesting a suppression of the humoral immune response.
One developmental study assessed the ability to generate an immune response following
gestational exposure to PFOA {Hu, 2010, 1332421}. In this study, pregnant C57BL/6N mice
were exposed to 0.5 or 1 mg/kg/day PFOA from GD 6-17. The adult female offspring were
immunized with SRBC on PND 44. No change in the immune response was observed, as
measured through IgM titers (PND 48) and IgG titers 2 weeks later (PND 63) following an
SRBC booster.
Alterations in the serum levels of globulin can be associated with decreases in antibody
production {FDA, 2002, 88170}. PFOA exposure at 12.5 mg/kg/day and up to 100 mg/kg/day
for 28 days decreased globulin concentrations in female Sprague-Dawley rats by up to 79% of
control. In males, a decrease in globulin concentrations was observed at 0.625 mg/kg/day (74%
of control) and up to 10 mg/kg/day (61% of control), highlighting greater PFOA tolerance in
females compared to males (Figure 3-27) {NTP, 2019, 5400977}. In contrast, an increase in
globulin concentrations, by approximately 7%, was observed in male BALB/c mice exposed to
0.4 or 2 mg/kg/day PFOA (but not 10 mg/kg/day) for 4 weeks (Figure 3-27) {Guo, 2019,
5080372}. In a similar study by the same group, immunoglobulins were measured, and IgA
concentrations were found to be significantly increased by 12%, 16%, and 33% in male BALB/c
mice exposed to 0.4, 2, or 10 mg/kg/day, respectively, PFOA for 4 weeks {Guo, 2021,
7542749}. IgM was increased by 3% and 6% in mice exposed to 2 or 10 mg/kg/day,
respectively, and IgG was increased by 6% in mice exposed to 10 mg/kg/day.
Globulin levels were also decreased in pregnant ICR dams on GD 18 following 5 or
10 mg/kg/day PFOA from GD 0-18 {Yahia, 2010, 1332451}. Globulin levels were decreased to
78 and 68% of control, respectively. Globulin levels in offspring were not measured. In a
developmental study conducted by NTP (2020, 7330145), Sprague-Dawley rats were exposed
perinatally and/or postweaning for a total of 107 weeks to varying doses of PFOA ((perinatal
exposure level (ppm))/(postweaning exposure level (ppm)); see further study design details in
Section 3.4.4.2.1.2). In male Sprague-Dawley rats at the 16-week interim timepoint, perinatal
exposure to 300 ppm (300/0) and/or postweaning exposure to doses ranging from 20-300 ppm
(0/150, 0/300, 150/150, 300/300, 0/20, 0/40, 0/80, 300/20, 300/40, or 300/80 ppm) significantly
decreased globulin levels. Female rats displayed decreased globulin levels following exposure to
0/300, 0/1,000, 150/300, or 300/1,000 ppm PFOA {NTP, 2020, 7330145} (Figure 3-27).
3-121
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint Study Name Study Design Observation Time Animal Description
Globulin (G) Guo etal., 2019, 5080372 short-term (4wk) 4wk Mouse, BALB/c (r?, N=10)
NTP, 2019, 5400977 short-term (28d) 29d Rat, Sprague-Dawley (,?, N=10)
Rat, Sprague-Dawley (V, N=9-10)
NTP, 2020,7330145 chronic (GD6-PNW21) 16wk F1 Rat, Sprague-Dawley (;\ N=10)
chronic (GD6-PNW107) 16wk F1 Rat, Sprague-Dawley (?, N=10)
F1 Rat, Sprague-Dawley Q, N=10)
chronic (PND21-PNW21) 16wk F1 Rat, Sprague-Dawley N= 10)
chronic (PND21-PNW107) 16wk F1 Rat, Sprague-Dawley (.?, N=10)
F1 Rat, Sprague-Dawley (i, N=10)
0.01 0,1 1 10 100
Concentration (mg/kg/day)
PFOA Immune Effects - Globulin
# No significant change A Significant increase"V~significant decrease
v v v v v
~ v v ^
-v-v
—^2—3^
-V—¥
-V—5—7
* ~
Figure 3-27. Globulin Levels in Rodents Following Exposure to PFOA (logarithmic scale)
PFOA concentration is presented in logarithmic scale to optimize the spatial presentation of data. Interactive figure and additional
study details available on HAWC.
GD = gestation day; PND = postnatal day; PNW = postnatal week; Fi = first generation; d = day; wk = week.
3.4.2.3 Mechanistic Evidence
Mechanistic evidence linking PFOA exposure to adverse immune outcomes is discussed in
Sections 3.3.2 and 3.4.1 of the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}. There are 22
studies from recent systematic literature search and review efforts conducted after publication of
the 2016 PFOA HESD that investigated the mechanisms of action of PFOA that lead to immune
effects. A summary of these studies is shown in Figure 3-28.
Mechanistic Pathway Animal Human In Vitro Grand Total
Cell Growth, Differentiation, Proliferation, Or Viability
3
0
3
6
Cell Signaling Or Signal Transduction
3
0
1
4
Fatty Acid Synthesis, Metabolism, Storage. Transport, Binding, B-Oxidation
1
0
1
2
Inflammation And Immune Response
11
6 5
20
Oxidative Stress
1
0
2
3
Not Applicable/Not Specified/Review Article
1
0
0
1
Grand Total
12
6
7
22
Figure 3-28. Summary of Mechanistic Studies of PFOA and Immune Effects
Interactive figure and additional study details available on Tableau.
A consistent pattern of findings from human (Section 3.4.2.1) and animal (Section
3.4.2.2) studies support that higher serum concentrations of PFOA are associated with
immunosuppression. Additional findings included reduced spleen and thymus weights, reduced
cellularity of white blood cells and differentials in circulation, reduced immune cellularity in
primary and secondary lymphoid organs, and altered globulin levels. Mechanistic data available
from in vitro, in vivo, and epidemiological studies were used to evaluate the mode of action of
PFOA-associated immunosuppression and other effects on the immune system.
3-122
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.3.1 Mechanistic Evidence for PFOA-mediated Effects on Immune System
Development and Physiology
Reductions in lymphocyte numbers have been consistently reported in animal toxicological
studies (Section 3.4.2.2), with parallel observations of reduced antibody responses in human
studies (Section 3.4.2.1). PFOA can alter the number of various B and T cell subsets in primary
and secondary lymphoid organs, which may reflect effects on immune system development
including effects on proliferation, differentiation, and/or apoptosis of immune cells.
Two in vivo studies were identified that evaluated PFOA-mediated effects on immune system
development, reflected in numbers of B and T cell populations. In female BALB/c mice dermally
exposed to PFOA for 14 days, the total numbers of splenic CD4+ T cells were reduced, as were
the total numbers and percent of CD4+ T cells in the lymph nodes. The percent of splenic CD4+
T cells was increased {Shane, 2020, 6316911}. The authors also observed that the absolute
number and percent of splenic B cells were reduced, an observation which could be explained by
increased apoptosis of B cells in the spleen or diminished proliferation in the bone marrow,
where B cells develop. Effects on B cell differentiation may also reflect reduced cellularity of
bone marrow, thymus, and spleen. Qazi et al. (2012, 1937236) reported reduced percentages of
the relatively undifferentiated pro/pre-B cells (CD19+/CD138+/IgM-) in the bone marrow of
male C57BL/6 mice fed diets containing 0.02% PFOA for 10 days. Morphological assessment of
the bone marrow was consistent with the reduced cell populations; mice treated with 0.02%
PFOA displayed hypocellularity in the bone marrow. The authors note that food consumption of
the mice exposed to 0.02% PFOA can be reduced up to 35%. Moreover, although experimentally
restricting food consumption by 35% in the absence of PFOA exposure affects pro/pre-B cell
populations in a similar manner to PFOA, the effect is not identical, which may support that
PFOA exposure is associated with decreased pro/pre-B cells in the bone marrow independent of
reduced food consumption. The study also demonstrated that the number of myeloid cells
(Grl+/CD1 lb+) is reduced by 0.02% PFOA but to a lesser magnitude than that of B lymphoid
cells (CD 19+), suggesting that the B-lymphoid cell lineage is more sensitive than the myeloid
cell lineage.
Several in vitro studies have reported reductions in immune cell viability or increases in
cytotoxicity following exposure to PFOA {S0rli, 2020, 5918817; Rainieri, 2017, 3860104},
which could also contribute to reduced lymphocyte cellularity or reduced immune organ weight
observed in the animal literature (Section 3.4.2.2).
Reductions in immune cellularity of B and T cell populations in the thymus and spleen (Section
3.4.2.2) as well as the bone marrow may reflect perturbations in cellular and/or molecular events
including cell proliferation, apoptosis, and oxidative stress. An in vitro study by Rainieri et al.
(2017, 3860104) evaluated the effects of PFOA on cell proliferation by quantifying the
distribution of cells in different stages of the cell cycle in a human macrophage cell line (TLT
cells). Significantly more cells were in G2/M phase of mitosis following exposure to PFOA in
parallel with a lower proportion of cells in the G0/G1 phase, suggesting increased cell
proliferation. However, increased cell proliferation is inconsistent with the immune organ
atrophy reported in animal toxicological studies (Section 3.4.2.2) and findings of other
mechanistic studies in immune organs. Yang et al. (2002, 1332453) reported significant
reductions in the proportion of thymocytes in the S and G2/M phases and significant increases in
the G0/G1 phases of mice treated with PFOA, which were attenuated in PPARa-null mice. These
3-123
-------
DRAFT FOR PUBLIC COMMENT
March 2023
results imply that reductions in cell numbers in the S and G2/M phases of the cell cycle are
partially mediated by PPARa.
Two studies {Wang, 2014, 3860153; Rainieri, 2017, 3860104} have reported increased apoptosis
in immune cells following PFOA exposure in vivo and in vitro. Increased apoptosis may
contribute to the reductions in immune organ weight observed in the animal literature and/or
reduced populations of immune cells (Section 3.4.2.2). Wang et al. (2014, 3860153) exposed
BALB/c mice to 0, 5, 10, or 20 mg/kg/day PFOA via gavage for 14 days and reported that the
percent of apoptotic cells increased in the spleen at 10 and 20 mg/kg/day and increased in the
thymus at 20 mg/kg/day. Increased apoptosis was associated with atrophy of these immune
system organs, suggesting that PFOA-induced apoptosis may contribute to organ atrophy. In
parallel, the authors explored the association between lipid metabolism and immunotoxicity of
PFOA by including a high-fat diet (HFD) group in addition to the regular diet (RD) group; there
was a higher percentage of apoptosis in the HFD vehicle control group than the RD vehicle
control group, indicating that HFD could cause or exacerbate apoptosis. Based on these diet-
related results along with gene expression data showing that PPARa and PPARy were also up-
regulated in the thymus and the spleen, the authors concluded that immunomodulation by PFOA
occurs via the PPAR pathway and the induction of mitochondrial damage and lymphocyte
apoptosis pathway. Rainieri et al. (2017, 3860104) evaluated apoptosis in TLT cells exposed to
0, 50, 250, or 500 mg/L PFOA for 12 hours. The percentage of apoptotic cells was significantly
elevated only at the highest concentration.
Generation of oxidative stress is a potential underlying mechanism linking PFOA to the
aforementioned effects on proliferation, differentiation, and/or apoptosis of immune cells.
Oxidative stress has been implicated in PFOA immunotoxicity by one in vivo study and several
in vitro studies {Wang, 2014, 3860153; Yahia, 2014, 2851192; Rainieri, 2017, 3860104}. Wang
et al. (2014, 3860153) observed that the spleens of mice treated with PFOA had mitochondrial
swelling and cavitation as well as swollen and ruptured cristae, which suggests impaired
oxidative processes. However, there were no significant changes in H2O2 concentrations or
superoxide dismutase (SOD) activity in spleens of mice exposed to PFOA versus controls. There
were no differences in mitochondrial ultrastructure between the HFD group and the RD group,
implying that although PFOA-related mitochondrial damage may contribute to apoptosis in
lymphocytes, the mechanism may not involve perturbed lipid metabolism. Rainieri et al. (2017,
3860104) reported increased lipid peroxidation in zebrafish embryos that coincided with a dose-
dependent increase in gene expression of glutathione S-transferase pi 1.2 (gstpl) and heat shock
cognate 70-kd protein, like (hsp701), which is typically observed in response to oxidative stress.
However, it is important to note that lipid peroxidation and gene expression analyses were
evaluated in whole zebrafish embryos and therefore may not necessarily be specific to effects in
immune organs. Oxidative DNA damage was reported by Yahia et al. (2014, 2851192) in a
human lymphoblast cell line (TK6 cells) exposed to PFOA at concentrations of 0, 125, 250, and
500 ppm, including a dose-dependent increase in 8-OHdG levels that coincided with increases in
tail moment, Olive Tail moment, and tail length in the comet assay at 250 and 500 ppm, which is
indicative of DNA damage. Altogether, the evidence suggests that PFOA can induce oxidative
stress in immune cells, including oxidation of lipids and DNA, potentially leading to DNA
damage.
3-124
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.3.2 Mechanistic Evidence for PFOA-mediated Effects on Adaptive
Immune Responses
3.4.2.3.2.1 Mechanistic data informing suppression of immune responses to vaccines
and infectious diseases
PFOA-associated immunosuppressive effects are described in Section 3.4.2.2.1. Adaptive
immune responses include B and T cell-mediated responses to infection and vaccination, as well
as allergic responses related to allergens or autoimmune responses. Mechanistic studies suggest
that chemicals, such as PFOA, can perturb the function of mature B or T lymphocytes by acting
at several stages of leukocyte function, including antigen recognition, antigen signaling through
the antigen receptor, activation, proliferation, and differentiation {Klaassen, 2013, 2993368}. In
mice, PFOA has been shown to diminish the immune response to sheep red blood cells (SRBC),
a T cell-dependent antibody response (Section 3.4.2.2), indicating that B and/or T cells can be
impacted by PFOA. A review of antigen-specific IgM antibody responses by NTP (2016,
4613766) indicated that both T cell-independent responses (e.g., immunized with dinitrophenyl
(DNP) or trinitrophenyl (TNP)) and T cell-dependent responses were reduced by PFOA.
One study provided evidence that antibody glycosylation patterns could be perturbed by PFOA:
Liu et al. (2020, 6833599) reported that children with higher levels of serum PFOA had altered
levels of N-glycosylation of IgG antibodies, which could perturb normal cell-cell interactions
through protein receptors involved in antigen recognition and presentation.
Activation of T cells can be demonstrated by transcriptional changes in the genes that encode
cytokines (e.g., IL-2) and cell surface proteins (e.g., IL-2 receptor); however, none of the
transcriptomic studies reported significant associations with IL-2 levels and PFOA. Although not
significant, one study by Zhu et al. (2016, 3360105) reported trending reductions in the levels of
IL-2 and increased serum PFOA concentrations in male and female asthmatic children.
The effect of PFOA on immunoglobulin classes was evaluated in a study by Zhang et al. (2014,
2851150), in which zebrafish were exposed to 0, 0.05, 0.1, 0.5, or 1 mg/L PFOA and
immunoglobulin gene expression was quantified in spleens. In contrast to mammals, which have
five different classes of immunoglobulin (i.e., IgM, IgA, IgD, IgE, and IgG), zebrafish have three
(IgM, IgD, and IgZ). The authors reported a dose-dependent reduction in IgM and non-
monotonic dose responses in IgD and IgZ, where the greatest increases in expression were
observed at the middle doses. Another zebrafish study by Zhong et al. (2020, 6315790) reported
a similar inverse U-shaped dose-response curve for IgD after 7 or 14 days of exposure to 0, 0.05,
0.1, 0.5, or 1 mg/L PFOA, but reported that IgZ and IgM were elevated in groups exposed to 0.1
or 0.5 mg/L PFOA. Additionally, the effect of PFOA on gene expression of B cell activating
factor (baff) paralleled that of IgD, suggesting that PFOA disrupts immunoglobulin levels by
interfering with baff mRNA expression.
Differentiation of B and T cells into mature effector cells can also be affected by PFOA
exposure. The cytokine milieu surrounding the T cell and antigen presenting cell (APC)
influences the fate of the T cell. In addition to the cytokines mentioned above, fluctuations have
been reported in IL-10, IL-5, and IL-4 levels. Associations between PFOA exposure and IL-4 or
IL-5 are discussed in relation to allergic and asthmatic responses below. The data on IL-10 is
limited to a single developmental study by Hu et al. (2012, 1937235), which exposed pregnant
3-125
-------
DRAFT FOR PUBLIC COMMENT
March 2023
C57BL/6N mice to 0, 0.02, 0.2, or 2 mg/kg PFOA via gavage and examined cytokine levels in
the spleens of male and female PND 21 offspring. In males, IL-10 was reduced by approximately
70% relative to IL-10 released from control animals at every exposure level. In contrast, IL-10
was unaffected in females at every exposure level except for an elevation at 0.02%. IL-10 is
released by regulatory T (TReg) cells and function to inhibit macrophage responses, therefore the
aforementioned impacts of PFOA on macrophages may be downstream of an effect on TRegs.
The impacts of PFOA on the adaptive immune system may reflect dysregulation of cell-signaling
pathways involved in adaptive immune responses. The predominant cell-signaling pathways
implicated in PFOA-mediated immunotoxicity include the PPAR and NF-kB signaling
pathways, which are both involved in the generation of adaptive immune responses. PPARy
activation is involved in the differentiation and development of TH1, TH2, and NK cells, and
inhibits the production of inflammatory cytokines in monocytes {Liang, 2021, 9959458}.
Multiple in vitro and in vivo studies have investigated the involvement of the PPAR pathway in
PFOA-immunotoxicity {Wang, 2014, 3860153; Yang, 2002, 1332453; Dewitt, 2016, 2851016}.
Wang et al. evaluated the effects of PFOA in thymocytes of mice exposed to PFOA (0, 5, 10, or
20 mg/kg/day) via gavage and fed RD or HFD. PFOA upregulated gene expression of PPARa
and PPARy in the thymus of RD animals at the highest dose and elicited a dose-dependent
elevation in PPARy in the thymus for HFD animals that reached significance at 10 mg/kg group.
An additional study using PPARa knock-out mice suggested the immunosuppressive effects of
PFOA are independent of PPARa {DeWitt, 2016, 2851016}. In this study, female C57BL/6Tac
PPARa knock-out mice and C57BL/6Tac wild-type mice were exposed to 0, 7.5, or 30
mg/kg/day PFOA in drinking water for 14 days and then injected with SRBC on day 11
{DeWitt, 2016; 2851016}. Exposure to 30 mg/kg/day PFOA for 15 days reduced SRBC-specific
IgM antibody responses in both wild-type and PPARa knock-out mice by 16% and 14%,
respectively. There was no significant difference between genotypes, suggesting that PPARa
may not be responsible for the suppression of the immune system induced by PFOA exposure.
Interestingly, this study also reported reductions in relative spleen weights (30% reduction after
exposure to 30 mg/kg/day PFOA) and thymus weights (55.4% after exposure to 7.5 mg/kg/day
PFOA) in the wild-type mice, but not in the knockout mice. Similarly, absolute spleen weights of
male Sv/129 PPARa-null mice fed approximately 40 mg/kg/day for 7 days were unaffected by
PFOA exposure, whereas in male C57BL/6 wild-type mice, absolute spleen weights were
significantly reduced by 39% {Yang, 2002, 1332453}. A significant decrease in absolute thymus
weight was observed in PFOA-exposed PPARa-null mice, to a lesser degree compared to the
reduction observed in PFOA-exposed wild-type mice (39% reduction in PPARa-null mice and
79%) reduction in wild-type mice).
One transcriptomics study in humans reported significant associations between maternal blood
levels of PFAS (including PFOA), enrichment of genes in neonatal cord blood samples, and
episodes of the common cold and antibody titers against the rubella vaccine in children
{Pennings, 2016, 3352001}. Enrichment of PPARD in neonatal cord blood samples was
correlated with maternal PFAS exposure and later common cold episodes in the children. The
NF-kB pathway was proposed to be involved in this phenomenon; a comparison of the
transcriptomics to the number of common cold episodes revealed that several genes in the NF-kB
pathway were altered.
3-126
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The NF-kB signaling pathway is essential for many parts and functions of the immune system,
including a pro-survival role during lymphopoiesis and regulation of T cell differentiation. Wang
et al. (2014, 3860153) provided indirect evidence that NF-kB pathway stimulation may be
involved in PFOA immunotoxicity. Gene expression of the glucocorticoid receptor (GR), which
stimulates the NF-kB pathway, was increased in the thymus of PFOA-treated animals at the
highest exposure level (20 mg/kg), suggesting mechanisms involving NF-kB pathway
stimulation may be involved in PFOA immunotoxicity. Additionally, the authors observed that
IL-1B gene expression was elevated in the thymus, suggesting that the NF-kB pathway is not
suppressed.
3.4.2.3.2.2 Mechanistic data informing allergic or asthmatic responses
Several studies evaluated potential associations between PFOA exposure and allergic responses
or asthma. An epidemiological study by Zhu et al. (2016, 3360105) explored the associations
between PFOA exposure and TH1/ TH2 polarization in asthmatic children. Male asthmatic
children with higher serum levels of PFOA tended to have higher serum IL-4 and IL-5, evident
of a TH2 skew. This association was not observed in females, suggesting that the exacerbation of
asthma by PFOA involving TH2 cytokines may be male-specific (Table 3-5).
More detailed mechanistic evidence on the relationship between PFOA and allergic responses is
available from animal toxicological studies. A dermal exposure study by Shane et al. (2020,
6316911) applied 0.5-2 % (w/v; equivalent to 12.5-50 mg/kg) PFOA to the skin of BALB/C
mice and evaluated allergic sensitization and IgM response. PFOA did not elicit an irritancy
response, suggesting that PFOA is not an allergic sensitizer or dermal irritant. However, the
splenic IgM response to SRBC was suppressed after 4 days of exposure to 2% PFOA, implying
that T cell-dependent immune responses to dermal allergens may be affected by PFOA.
Moreover, mice exposed to PFOA had increased expression of Tslp, which is associated with a
polarization towards a TH2 response {Shane, 2020, 6316911}. In adult zebrafish, the effect of
PFOA exposure on mRNA expression of IL-4 was mixed: it was elevated at most doses tested,
but reduced at the highest dose {Zhang, 2014, 2851150}. More data from mammalian models on
the associations between IL-4 or IL-10 and PFOA are needed to better understand the potential
impacts of PFOA on adaptive immune responses involving T cell subsets.
An in vitro study conducted by Lee et al. (2017, 3981419) demonstrated that PFOA increased
IL-ip gene and protein expression in a dose-related manner in IgE-stimulated RBL-2H3 cells (a
rat basophil cell line). Elevated IL-ip was also observed in a study of human bronchial epithelial
cells (HBEC3-KT cells) stimulated with a pro-inflammatory agent, Poly I:C, and then treated
with 0.13, 0.4, 1.1, 3.3, or 10 ^MPFOA {S0rli, 2020, 5918817}.
Several studies have evaluated molecular signaling pathways to better understand the
mechanistic underpinnings of allergic or asthmatic responses related to exposure to PFOA. At
least four mechanistic studies have evaluated the involvement of the NF-kB signaling pathway,
which plays an important role in the regulation of inflammation and immune responses,
including expression of pro-inflammatory cytokines {Lee, 2017, 3981419; Shane, 2020,
6316911; Zhong, 2020, 6315790; Zhang, 2014, 2851150}. Histamine release and mast cell
degradation were increased in parallel with increased nuclear localization of NF-kB and
concomitant reduction in IkB in IgE-stimulated mast cells, suggesting that allergic immune
responses and inflammation are exacerbated by PFOA through a mechanism involving the NF-
3-127
-------
DRAFT FOR PUBLIC COMMENT
March 2023
kB pathway {Lee, 2017, 3981419}. Zhang et al. (2014, 2851150) reported thatPFOA exposure
for 21 days can disrupt the NF-kB pathway to mediate inflammatory cytokines in zebrafish. The
authors reported a non-monotonic dose-response in gene expression of the p65 transcription
factor in RNA isolated from zebrafish splenocytes. In a more recent study, zebrafish were
exposed to PFOA for a shorter period (7 or 14 days) and the authors reported that splenic p65
gene expression was increased in all exposed groups {Zhong, 2020, 6315790}. Shane et al.
(2020, 6316911) showed that gene expression of NF-kB (Nfkbl) was reduced in the skin of
female BALB/c mice dermally exposed to 1 or 2% PFOA after 14 days. However, the study
design did not quantify nuclear NF-kB, so it is difficult to discern whether the NF-kB pathway
was activated. The authors also reported that gene expression of PPARa was reduced by more
than 50% in female mice dermally exposed to 1% or 2% PFOA for 14 days. Mechanistically,
PPARa is known to block the NF-kB pathway and thereby modulate immune responses. These
data suggest that the NF-kB pathway activity can be reduced independent of action by PPARa in
PFOA-mediated immunotoxicity with respect to allergic responses in the skin.
Table 3-5. Effects of PFOA Exposure on Cytokines Impacting Adaptive Immune Responses
Study
Species or Cell
Type
Study
Type
Cytokine
Measurement
Significant
Change in
Cytokine
Relevant
Immune
response
{Zhu, 2016,
3360105}
Human males
and females,
GBCA study
Epi
IL-2
IL-4
serum protein
(ELISA)
serum protein
(ELISA)
None
Ta
Allergy
Allergy
IL-5
serum protein
(ELISA)
Allergy
{Hu, 2012,
1937235}
C57BL/6N
mice
Ex vivo
IL-10
IL-10 production
assay in
CD4+CD25+ T
cells'1
TReg responses
Notes'. ELISA = enzyme-linked immunosorbent assay; GBCA = Genetic and Biomarkers study for Childhood Asthma; IL-
2 = Interleukin 2; IL-4 = Interleukin 4; IL-5 = Interleukin 5; IL-10 = Interleukin 10; TReg = regulatory T cells.
a Males only
b Purity of CD4+CD25+ T cells derived by cell estimate to be 84-95% based on manufacturer specification for the cell isolation
kit.
3.4.2.3.2.3 Mechanistic data informing autoimmune diseases
Select data on PFOA and autoimmune diseases in humans have been summarized by NTP (2016,
4613766). NTP's conclusion that PFOA was presumed to be an immune hazard to humans was
partially based on the positive associations that exist between PFOA exposure and rheumatoid
arthritis, ulcerative colitis, and auto-antibodies specific to neural and non-neural antigens.
However, the association was considered low confidence by the NTP. No animal or in vitro
studies have been identified to inform the potential associations between PFOA and
autoimmunity.
3-128
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.3.3 Mechanistic Evidence for PFOA-mediated Effects on Innate Immune
Responses
Neutrophils are important cells of the innate immune system that contribute to inflammation and
are the first cells to arrive at the site of injury or infection. Reductions in neutrophil migration to
the site of injury have been noted in zebrafish exposed to PFOA {Pecquet, 2020, 6833701},
suggesting diminished innate immune responses.
Neutrophil migration occurs in response to inflammation and in response to effector cytokines
such as IL-8 released from macrophages, which may also be sensitive to PFOA. Qazi et al.
(2010, 1276154) evaluated liver homogenates from male C57BL/6 mice and found that ex vivo
production of TNF-a was significantly decreased in animals treated with 0.002% or 0.005%
PFOA. Because macrophages are the major producers of TNF-a, the authors propose that PFOA
may directly or indirectly affect specialized hepatic macrophages (e.g., Kupffer cells). The
decrease in TNF-a release from macrophages could also be related to PFOA effects on the
adaptive immune system, given that macrophage responses are inhibited by IL-10 released by
TReg cells. Indeed, Hu et al. (2012, 1937235) demonstrated that ex vivo release of IL-10 from
splenocytes was reduced in male mice. Furthermore, cells of the monocyte/macrophage lineage
express PPARa and PPARy {Zhu, 2016, 3360105; Braissant, 1998, 729555}, which supports a
mechanism for immunosuppression involving macrophages and PPAR pathways.
Rainieri et al. (2017, 3860104) also conducted an in vitro assessment using TLT cells and found
that PFOA led to an increase in relative reactive oxygen species (ROS) production measured via
the dichlorodihydrofluorescein diacetate (DCF-DA) assay, indicating that PFOA can induce
ROS in macrophages.
Although the innate immune system also includes natural killer (NK) cells, no mechanistic
studies were identified that evaluated associations with PFOA. One study by Qazi et al. (2010,
1276154) reported that there were no significant differences in number or percent of NK cells in
isolated hepatic immune cells (IHICs) of mice exposed to 0.002% (w/w) PFOA in the diet for 10
days.
3.4.2.3.4 Mechanistic Evidence for PFOA-mediated Effects on Intrinsic Cellular
Defense Pathways
Zhang et al. (2014, 2851150) exposed zebrafish to PFOA (0.05, 0.1, 0.5, and 1 mg/L) for 21
days. After exposure, spleens were analyzed for expression patterns of myeloid differentiation 88
(MyD88) and toll-like receptor 2 (TLR2) as well as several cytokines. In addition to the above-
mentioned effects on gene expression of IL-4, PFOA exerted dose-dependent effects on IL-ip
and IL-21 that were stimulated at a low exposure concentration (0.05 mg/L) and inhibited at
higher exposure concentrations (>0.1 mg/L). The Myd88/NF-KB pathway was found to mediate
inflammatory cytokine (IL-1 and IL-21) gene expression in zebrafish spleen. Interestingly,
exposure of zebrafish to 1 mg/L PFOA reduced TLR2 mRNA expression in spleen by 56%
compared to controls. These findings suggest that exposure to PFOA in zebrafish can activate the
NF-kB pathway and interfere with TLR2 expression in a dose-dependent manner to enhance pro-
inflammatory cytokine gene expression.
3-129
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.2.3.4.1 Mechanistic Evidence for PFOA-mediated Effects on Inflammation
The observed increases in circulating leukocytes (neutrophils and monocytes) of experimental
animals (Section 3.4.2.2) are consistent with an inflammatory response. Inflammation is a
physiological response to tissue damage or infection that can induce components of the innate
and adaptive immune system {Klaassen, 2013, 2993368}. Processes that contribute to
inflammation and are affected by PFOA include the complement cascade, release and/or
upregulation of pro-inflammatory cytokines, and neutrophil migration.
3.4.2.3.4.1.1 Pro-inflammatory Responses Including Cytokines
The available mechanistic data support that pro-inflammatory cytokines such as IL-ip, TNF-a,
and possibly IL-6 are elevated by PFOA exposure (Table 3-6). However, the effect of PFOA (or
lack thereof) for some cytokines varies between model organisms and exposure levels. Altered
production and/or release of these cytokines may represent an underlying mechanism of the
reductions in innate and/or adaptive immune function that has been reported in the human
(Section 3.4.2.1) and animal (Section 3.4.2.2) literature.
Elevation of IL-ip is consistent across study designs in mammalian models in vivo and in vitro.
Wang et al. (2014, 3860153) exposed 4-5-week-old male BALB/C mice to 0, 5, 10, or 20
mg/kg/day PFOA via gavage for 14 days in combination with HFD or RD and measured gene
expression of cytokines in the thymus and spleen. In the thymus, IL-ip was elevated in mice
exposed to 20 mg/kg/day and fed RD. There were no significant effects in the spleen for mice
fed RD at any PFOA concentration. In HFD-fed mice, there was an increase in IL-ip in the
spleen for the 10 mg/kg/day PFOA group, but no significant changes at any exposure level in the
thymus. Likewise, Lee et al. (2017, 3981419) and S0rli et al. (2020, 5918817) have demonstrated
that PFOA elevates IL-ip gene and/or protein expression in various cell lines. In contrast to the
consistent increases in IL-ip reported in mammalian models, one study in adult zebrafish
reported decreased IL-ip mRNA in the spleen following exposure to 0.1, 0.5, or 1 mg/L PFOA
for 21 days {Zhang, 2014, 2851150}. More research is needed to determine whether inter-
species differences exist in immunomodulation by PFOA. Elevated production of IL-ip is
triggered by activation of the inflammasome, which is an innate immune response known to be
activated by xenobiotics, and this mechanism may deserve further investigation {Mills, 2013,
2556647}.
Several studies have reported elevated levels of TNF-a during immune responses following
exposure to PFOA. Qazi et al. (2010, 1276154) reported decreased levels of TNF-a in liver
homogenates of male C57BL/6 mice orally exposed to 0.002% PFOA for 10 days. Lee et al.
(2017, 3981419) quantified TNF-a levels in blood from male ICR mice following an active
systemic anaphylaxis experiment. Mice were sensitized to ovalbumin on day 0 and day 7 via
intraperitoneal (i.p.) injection, and PFOA was orally administered on day 9, 11, and 13.
Following ovalbumin challenge (i.p.) on day 14, a dose-dependent increase in TNF-a levels in
blood was observed, suggesting PFOA aggravates allergic inflammation. In the same study, in
vitro experiments using three independent methods (Western blot, RT-PCR, and ELISA)
demonstrated a dose-dependent elevation in TNF-a in RBL-2H3 cells sensitized with anti-DNP
IgE, then treated with PFOA for 24 hours. Likewise, an in vitro study by Brieger et al. (2011,
1937244) observed a slight increase in TNF-a released from peripheral blood mononuclear cells
(PBMCs) obtained from the blood of 11 human donors. Not all studies reported positive
associations of PFOA and TNF-a. Although Bassler et al. (2019, 5080624) reported positive
3-130
-------
DRAFT FOR PUBLIC COMMENT
March 2023
associations between serum PFOA levels and IFN-y, the authors found inverse associations with
TNF-a.
A few of the studies that observed increases in IL-ip and TNF-a also evaluated other pro-
inflammatory cytokines such as IL-8 and IL-6. The in vitro studies by Lee et al. (2017, 3981419)
did not find significant effects of PFOA on IL-8 expression. This finding was consistent with
those of S0rli et al. (2020, 5918817) and Bassler et al. (2019, 5080624). IL-6 gene and protein
expression were elevated in the study by Lee et al. (2017, 3981419), which was consistent with
results of Brieger et al. (2011, 1937244) in human PBMCs stimulated with LPS. Most other
studies reported either no effect or inverse associations with IL-6 {Mitro, 2020, 6833625; Shane,
2020, 6316911}. Gimenez-Bastida et al. (2015, 3981569) reported that PFOA attenuated the
elevation in IL-6 levels that normally follows IL-ip-induction in a human colon cell line (CCD-
18Co).
IFN-y is released from activated T cells and NK cells and induces macrophages to produce a
variety of inflammatory mediators and reactive oxygen and nitrogen intermediates that
contribute to inflammation {Klaassen, 2013, 2993368}. In general, studies did not find
associations between PFOA and changes in IFN-y. The sole exception by Zhong et al. (2020,
6315790) reported elevations in IFN gene expression in splenocytes of adult zebrafish exposed
to 0.05, 0.1, 0.5, or 1 mg/L PFOA for 7 days. Zhu et al. (2016, 3360105) reported that children
with asthma generally had higher serum PFOA concentrations and lower levels of IFN-y than
non-asthmatic children, but there was not a significant association between IFN-y and PFOA.
Qazi et al. (2010, 1276154) measured IFN-y levels secreted from IHICs of 6-8-week-old male
C57BL/6 (H-2b) mice that were exposed to 0 or 0.002% (w/w) PFOA in feed for ten days. A
subgroup of IHIC were stimulated with Concanavalin A, which activates T cells to produce IFN-
y. No PFOA-related differences in IFN-y production were observed in any group in IHICs. The
authors also reported a 37% reduction in hepatic levels of IFN-y, in parallel with reductions in
hepatic levels of IL-4 and TNF-a.
Inflammatory responses can be accompanied by increased levels of the activated pro-
inflammatory transcription factor, NF-kB. Sirtuins (SIRTs) have been shown to deacetylate NF-
kB, which suppresses its transcriptional activation, thereby inhibiting the production of pro-
inflammatory cytokines. Park et al. (2019, 5412425) exposed a macrophage cell line (RAW
264.7 cells) to 0, 0.5, 5 or 50 [xM PFOA and observed significant increases in expression for
SIRT3 and SIRT6 at 5 |iM exposure, which is inconsistent with a model where PFOA induces
inflammation. Interestingly, SIRT4 and SIRT7 expression was more sensitive to PFOA and
exhibited non-linear dose-response curves; SIRT4 was significantly reduced at 0.5 |iM and
significantly elevated at 5 [xM, whereas SIRT7 was significantly elevated at 0.5 |iM and
significantly reduced at 5 and 50 |iM, Altogether, the results support that a pro-inflammatory
response of PFOA may not follow a linear dose-response.
3.4.2.3.4.1.2 Complement Pathways
PFOA can affect both the innate and adaptive immune system to perturb activation of one of the
three main pathways of the complement cascade. A study conducted in the C8 Health Project
cohort found that serum biomarkers of PFOA were positively associated with serum C3a levels
in men, but negatively associated in women, supporting sex-specific perturbations in immune
function {Bassler, 2019, 5080624}. Also using data from the C8 Health Project, another group of
3-131
-------
DRAFT FOR PUBLIC COMMENT
March 2023
researchers, Genser et al. (2015, 3271854) found evidence that PFOA blood levels were
negatively associated with blood levels of C-reactive Protein (CRP), which is essential for the
classical pathway of complement activation {Klaassen, 2013, 2993368}. However, another
human study, that measured CRP as one among several blood biomarkers of cardiometabolic
disruption reported that serum PFOA was "generally weakly" (i.e., not significantly) associated
with CRP and other biomarkers in women 3 years postpartum {Mitro, 2020, 6833625}. In
contrast to the human evidence, serum C3 levels were reduced in male C57BL/6 (H-2b) mice
exposed to 0.02% w/w PFOA in feed for 10 consecutive days {Botelho, 2015, 2851194}. Female
mice were not studied. Reduced activities of the classical and alternative complement pathways
(reflected by CH50 and AH50 response, respectively) were also reported, supporting that PFOA
can disrupt the classical (IgM/IgG dependent) and alternative pathways of complement
activation, which both require C3.
Table 3-6. Effects of PFOA Exposure on Pro-Inflammatory Cytokines and Markers of
Inflammation
Study
Species or Cell
Type
Study
Type
Cytokine or
Inflammatory
Marker
Measurement
Direction of
Change
Following PFOA
Exposure
Mitro et al.
Human females,
In vivo
IL-6
blood protein (ELISA)
T
(2020,
3 years post-
6833625)
partum
CRP
blood protein
1
(immunoturbidimetric high-
sensitivity assay)
Bassler et al.
Human males
In vivo
IL-6
serum protein (Multispot
None
(2019,
and females, C8
Immunoassay)
5080624)
Health Project
TNF-a
serum protein (Multispot
1
Immunoassay)
IL-8
serum protein (Multispot
None
Immunoassay)
IFNy
serum protein (Multispot
T
Immunoassay)
C3a
serum protein (ELISA)
None
Sorli et al.
Human
In vitro
11-6
culture supernatant protein
None
(2020,
bronchial
(ELISA)
5918817)
epithelial cell
IL-la
culture supernatant protein
None
line
(ELISA)
IL-1|3
culture supernatant protein
T
(ELISA)
CXCL8
culture supernatant protein
None
(ELISA)
Wang et al.
BALB/c mice
In vivo
IL-1|3
Gene expression
T
(2014,
3860153)
Shane et al.
BALB/c mice
In vivo
IL-1|3
Gene expression
T
(2020,
IL-6
Gene expression
None
6316911)
3-132
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Study
Species or Cell
Type
Study
Type
Cytokine or
Inflammatory
Marker
Measurement
Direction of
Change
Following PFOA
Exposure
Qazi et al.
(2010,
1276154)
C57BL/6 mice
Ex vivo
IFN-y
culture supernatant protein
(ELISA)
None
Notes: IL-6 = Interleukin 6; CRP = C-Reactive Protein; TNF-a = Tumor Necrosis Factor a; IL-8 = Interleukin 8;
IFNy = Interferon y; C3a = cleavage product of Complement 3
3.4.2.3.5 Conclusions
Overall, the available evidence supports that PFOA affects the innate and adaptive immune
system as well as immune organ physiology at multiple levels including immune system
development, survival, proliferation, and differentiation of B and T cells, inflammatory
responses, neutrophil migration, and complement activation. One study provided evidence that
antibody glycosylation patterns could be perturbed. Mechanistic data available from in vitro, in
vivo, and epidemiological studies were used to evaluate the etiology and mode of action of
PFOA-associated immunosuppression and other effects on the immune system. The pleotropic
immunomodulatory effects of PFOA, including impaired vaccine responses, may reflect
perturbed function of B and/or T cells. At the molecular level, dysregulation of the NF-kB
pathway may contribute to the immunosuppressive effects of PFOA. The NF-kB pathway
facilitates initial T cell responses by supporting proliferation and regulating apoptosis,
participates in the regulation of CD4+ T cell differentiation, and is involved in mediating
inflammatory responses. Dysregulation of the NF-kB pathway by PFOA, potentially consequent
to the induction of oxidative stress, may be a key component of the mechanism underlying
PFOA-mediated immunosuppression. Reduced NF-kB activation and consequent elevation of
apoptosis is consistent with increased apoptosis in multiple cell types, the reduction of pre/pro-B
cell numbers, and dysregulation of pro-inflammatory cytokines and mediators of inflammation.
NF-kB activation also facilitates the induction of apoptosis during negative selection of T cells in
the thymus, which is essential for the deletion of T cells that recognize self. In contrast, NF-kB
acts as a pro-survival factor during the negative selection of B cells. In human studies, PFOA
exposure has been associated with autoimmune diseases including ulcerative colitis. Further
mechanistic evidence is needed to determine the directionality of the effect of PFOA on NF-kB,
which will inform the cell types that predominantly contribute to the etiology of autoimmune
diseases associated with PFOA exposure.
3.4.2.4 Evidence Integration
There is moderate evidence for an association between PFOA exposure and immunosuppressive
effects in human studies based on largely consistent decreases in antibody response following
vaccinations (against two different infectious agents: tetanus and diphtheria) in multiple medium
confidence studies in children. Reduced antibody response is an indication of
immunosuppression and may result in increased susceptibility to infectious disease. The antibody
response results present a consistent pattern of findings that higher prenatal, childhood, and adult
serum concentrations of PFOA were associated with suppression of at least one measure of the
anti-vaccine antibody response to common vaccines in two well-conducted (though overlapping)
birth cohorts in the Faroe Islands, supported by a low confidence study in adults.
3-133
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The results in human epidemiological studies measuring PFOA concentrations and
hypersensitivity were mixed. Significant associations between PFOA exposure and "ever" or
"current" asthma were seen primarily in sex- or age-specific subgroups but were null or
insignificant in whole study analyses. For allergy and eczema outcomes, results were
inconsistent across studies.
The associations between PFOA exposure and human autoimmune disease were also mixed.
Two studies {Steenland, 2013, 1937218; Steenland, 2018, 5079806} found significant
associations indicating increased risk of autoimmune disease. Also, PFOA levels were found to
be lower in healthy controls compared to cases with MS {Ammitzb0ll, 2019, 5080379}. Results
were most consistent for ulcerative colitis, with significant associations indicating increased risk
with increasing PFOA exposure in one medium confidence study {Steenland, 2013, 1937218}
and one low confidence study {Steenland, 2018, 5079806}.
The animal evidence for an association between PFOA exposure and immunosuppressive
responses is moderate based on 13 high or medium confidence animal toxicological studies.
Short-term and developmental PFOA exposure in rodents resulted in reduced spleen and thymus
weights, altered immune cell populations, and decreased splenic and thymic cellularity. In
functional assessment of the immune response, PFOA exposure was associated with reduced
globulin and immunoglobulin levels {Dewitt, 2008, 1290826; Loveless, 2008, 988599}.
Suppression of the immunoglobulin response in these animals is consistent with decreased
antibody response seen in human subpopulations.
Mechanistic data related to the human immunomodulatory effects were similarly inconsistent
compared to the human epidemiological data. The available mechanistic data indicate that pro-
inflammatory cytokines such as IL-ip, TNF-a, and possibly IL-6 are elevated by PFOA
exposure. However, the specific effects vary across model organisms and exposure levels.
Altered production and/or release of these cytokines may reflect reductions in innate and/or
adaptive immune function that has been reported in the human and animal literature.
While evidence exists for reduced antibody response, such as diminished immune response to
sheep red blood cells in mice treated with PFOA (a T cell-dependent antibody response), data are
limited. Both T cell-dependent and T cell-independent responses are reduced by PFOA,
according to a systematic review conducted by the NTP {NTP, 2016, 4613766}. Alterations to
these responses could explain the decreased antibody response in humans. Although the evidence
is not consistent across studies or between sexes and/or model systems, several studies have
reported that PFOA appears to exacerbate allergic immune and inflammatory response, likely
through disruption to the NF-kB pathway, increased TNFa, and/or TH2 response.
One proposed mechanism of immunotoxicity involves apoptosis of immune cells, which appears
to be a high-dose phenomenon, as evidenced by in vivo and in vitro studies in which the effects
were only seen at > 10 mg/kg/day in mice or 500 mg/L in the human macrophage TLT cell line.
Relatedly, NF-kB activation also facilitates the induction of apoptosis during negative selection
of T cells in the thymus, which is essential for the deletion of T cells that recognize host cells
(i.e., "self'). In contrast, NF-kB acts as a pro-survival factor during the negative selection of B
cells. PFOA has been shown to disrupt the NF-kB pathway. At the molecular level,
dysregulation of the NF-kB pathway may contribute to the immunosuppressive effects of PFOA.
The NF-kB pathway facilitates initial T cell responses by supporting proliferation and regulating
3-134
-------
DRAFT FOR PUBLIC COMMENT
March 2023
apoptosis, participating in the regulation of CD4+ T cell differentiation, and participating in
mediating inflammatory responses. Dysregulation of the NF-kB pathway by PFOA, potentially
consequent to the induction of oxidative stress, may be a key component of the mechanism
underlying PFOA-mediated immunosuppression. Reduced NF-kB activation and consequent
elevation of apoptosis is consistent with increased apoptosis in multiple cell types, the reduction
of pre/pro B cell numbers, and dysregulation of pro-inflammatory cytokines and mediators of
inflammation.
There is conflicting evidence regarding the involvement of PPAR signaling in immunotoxic
effects of PFOA: there is evidence of PPAR-independent alterations to adaptive immunity, while
suppressive effects of innate immunity appear to involve macrophages and PPAR signaling.
3.4.2.4.1 Evidence Integration Judgment
Overall, considering the available evidence from human, animal, and mechanistic studies, the
evidence indicates that PFOA exposure is likely to cause adverse immune effects, specifically
immunosuppression, in humans under relevant exposure circumstances (Table 3-7). The hazard
judgment is driven primarily by consistent evidence of reduced antibody response from
epidemiological studies at median levels as low as 1.1 ng/mL PFOA. The evidence in animals
showed coherent immunomodulatory responses at doses as low as 1 mg/kg/day PFOA that are
consistent with potential immunosuppression and supportive of the human studies, although
issues with overt organ/systemic toxicity raise concerns about the biological significance of some
of these effects. While there is some evidence that PFOA exposure might also have the potential
to affect sensitization and allergic responses in humans given relevant exposure circumstances,
the human evidence underlying this possibility is uncertain and with limited support from animal
or mechanistic studies. Based on the antibody response data in humans, children and young
individuals exposed during critical developmental windows may represent a potential susceptible
population for the immunosuppressive effects of PFOA. The absence of additional
epidemiological studies or any long-term/chronic exposure studies in animals examining
alterations in immune function or immune-related disease outcomes during different
developmental life stages represents a source of uncertainty in the immunotoxicity database of
PFOA.
3-135
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 3-7. Evidence Profile Table for PFOA Immune Effects
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
Evidence from Studies of Exposed Humans (Section 3.4.2.1)
Immunosuppression
1 High confidence study
15 Medium confidence
studies
8 Low confidence studies
1 Mixed1 confidence
study
Studies conducted in the •
Faroe Islands examined
antibody levels among •
children at various
timepoints compared to •
exposure measured
prenatally and throughout
childhood. Lower
antibody levels against
tetanus and diphtheria
were observed in children
at birth, 18 months, age 5
years (pre-and post-
booster), and at age 7
years, with some being
statistically significant.
Findings in the three
studies examining adults
and adolescents were less
consistent than in children.
One study reported an
inverse association for
hepatitis B antibodies, but
other antibody responses
were inconsistent across
all exposure windows.
Infectious disease was
examined in 11 studies of
children. Studies
examining infections of
the respiratory system
observed some positive
High and medium •
confidence studies •
Consistent direction
of effect
Coherence of
findings between
antibody response
and increased
infectious disease
Low confidence studies
Imprecision of findings
®©o
Moderate
Evidence for immune
effects is based on
decreases in childhood
antibody responses to
pathogens such as
diphtheria and tetanus.
Reductions in antibody
response were observed at
multiple timepoints in
childhood, using both
prenatal and childhood
exposure levels. An
increased risk of upper
and lower respiratory tract
infections was observed
among children, coherent
with findings of reduced
antibody response. There
was also supporting
evidence of increased risk
of asthma, eczema, and
autoimmune disease,
however, the number of
studies examining the
same type of autoimmune
disease was limited.
0©O
" Evidence Indicates (likely)
Primary basis and cross-
stream coherence:
Human data indicated
consistent evidence of
reduced antibody response.
Evidence in animals
showed coherent
immunomodulatory
responses that are
consistent with potential
immunosuppression and
supportive of the human
studies, although issues
with overt organ/systemic
toxicity raise concerns
about the biological
significance of some of
these effects. While there is
some evidence that PFOA
exposure might also have
the potential to affect
sensitization and allergic
responses in humans given
relevant exposure
circumstances, the human
evidence underlying this
possibility is uncertain and
has only limited support
from animal or mechanistic
studies.
3-136
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
associations (4/11),
although many findings
from other studies were
not precise. Findings for
infectious disease in adults
were mixed, with two
studies reporting
inconsistent results for
COVID-19 infections.
Immune
hypersensitivity
1 High confidence study
16 Medium confidence
studies
5 Low confidence studies
2 Mixed* confidence
studies
Examination of immune
hypersensitivity includes
outcomes such as asthma,
allergies, and eczema.
Increased odds of asthma
were reported in most
medium confidence studies
(6/9), although
associations were often
inconsistent by subgroups.
Low confidence studies
supported the findings of
increased odds of asthma
or higher exposure levels
among asthmatics,
although results were not
always consistent or
precise. Seven studies
examined allergies,
rhinitis, or
rhinoconjunctivitis. Some
positive associations (3/7)
were observed, although
this varied by outcome
timing and were at times
inconsistent. Significantly
High and medium
confidence studies
Consistent direction
of effect for asthma
across medium
confidence studies
Low confidence studies
Lnconsistent direction
of effect between
subpopulations
Evidence Integration
Summary Judgment
Human relevance and other
inferences:
Based on the antibody
response data in humans,
children and young
individuals exposed during
critical developmental
windows may represent a
potential susceptible
population for the
immunosuppressive effects
of PFOA. The absence of
additional epidemiological
studies or any long-
term/chronic exposure
studies in animals
examining alterations in
immune function or
immune-related disease
outcomes during different
developmental life stages
represents a source of
uncertainty in the
immunotoxicity database of
PFOA.
3-137
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
increased odds of eczema
were observed for those in
the highest exposure group
(3/8), but the associations
were non-monotonic
across exposure groups.
Autoimmune disease
2 Medium confidence
studies
4 Low confidence studies
Increased risk of <
autoimmune disease was
reported in several studies
(4/6). One study reported a
significantly increased risk
of rheumatoid arthritis,
and two studies reported a
significantly increased risk
of ulcerative colitis. Two
studies reported positive
associations for multiple
sclerosis, with one
reaching significance. One
study observed increased
risk of celiac disease
among children and young
adults. Findings for
Crohn's disease and type 1
diabetes were less
consistent.
Medium confidence
studies
Low confidence studies
Limited number of
studies examining
outcome
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.2.2)
Organ weights
Decreases in absolute
• High and medium •
Lnconsistent direction
®©o
3 High confidence
(6/8) and relative (4/8)
confidence studies
of effects across sex
Moderate
studies
spleen weights and in
• Dose-response •
Confounding variables
7 Medium confidence
absolute (5/5) and relative
relationship seen
such as decreases in
Evidence is based on 13
studies
(3/5) thymus weights were
within multiple
body weights
high or moderate
observed across studies
studies
confidence animal
regardless of study design.
• Coherence of
toxicological studies.
Overall, decreases in
findings of other
Short-term and
3-138
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
spleen and thymus weights
were more frequently
observed in males than
females and tended to
coincide with reductions
in body weight.
immunological
endpoints
Immune cellularity
1 High confidence study
4 Medium confidence
studies
Globulins and
immunoglobulins
2 High confidence
studies
Of the studies that •
measured circulating
WBCs and differentials, •
one short-term study in
male mice found decreases
in WBC counts, while a
chronic rat study observed •
transient increases in
males that were attributed
to increased counts of
lymphocytes and
neutrophils. One short-
term study in male rats
and mice reported
increased neutrophils and
monocytes, decreased
eosinophils, as well as
reduced splenocytes and
thymocytes in mice but no
changes in rats. One
developmental study in
mice observed decreases
in splenic regulatory T
cells in males and females.
High and medium
confidence studies
Dose-response
relationship seen
within multiple
studies
Coherence of
findings
Inconsistent direction
of effects across
species, sex, and study
design
Limited number of
studies examining
specific outcomes
Mixed results were
reported for concentrations
of globulins and
immunoglobulins.
Decreased globulin levels
High and medium
confidence studies
Dose-response
relationship
Inconsistent direction
of effects between
species
developmental PFOA
exposure in rodents
resulted in reduced spleen
and thymus weights,
altered immune cell
populations, and decreased
splenic and thymic
cellularity. In functional
assessments of the
immune response, PFOA
exposure was associated
with reduced globulin and
immunoglobulin levels.
Suppression of the
immunoglobulin response
in these animals is
consistent with decreased
antibody response seen in
human subpopulations.
Evidence Integration
Summary Judgment
3-139
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
2 Medium confidence
studies
(2/3) were observed in
male and female rats, in a
dose-dependent manner
(1/3), following short-term
and chronic exposure to
PFOA. One short-term
study reported increased
globulins (1/3) in male
mice. Additional findings,
including increases in IgA,
IgG, and IgM, were found
in male mice.
Limited number of
studies examining
specific outcomes
Immune response
4 Medium confidence
studies
Dose-dependent decreases
in IgM following a SRBC
or KLH challenge was
seen in three short-term
studies in mice (3/4).
No changes in IgM were
observed in chronically
exposed male rats nor
developmentally exposed
female mice (2/4). In a
short-term study that
assessed female mice,
increased IgG levels were
observed after a SRBC
challenge (1/2), but a
developmental study in
female mice found no
changes in IgG levels
(1/2).
Medium confidence
studies
Dose-response
relationship seen
within multiple
studies
Inconsistent direction
of effects across study
design and species
Limited number of
studies examining
specific outcomes
Histopathology
3 High confidence
studies
A short-term study in male •
mice and rats reported
increased incidence of •
granulocytic hyperplasia
High and medium
confidence studies
Coherence of
findings
Limited number of
studies examining
specific outcomes
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
3-140
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Evidence Integration
Studies and Summary and Key Factors that Increase Factors that Decrease Evidence Stream Summary Judgment
Interpretation Findings Certainty Certainty Judgment
2 Medium confidence of the bone marrow and
studies increased incidence of
splenic and thymic
atrophy in mice but not
rats. One high confidence
short-term study in male
and female rats observed
no changes in the spleen,
thymus, or lymph nodes
but found increased bone
marrow hypocellularity in
male rats. One chronic
study found decreased
incidence of splenic
hemosiderosis in male and
female rats. One chronic
and one developmental
study observed
histopathological changes
in the spleen, thymus,
bone marrow, and/or
lymph nodes of male and
female rats.
Mechanistic Evidence and Supplemental Information (Section 3.4.2.3.4)
Summary of Key Findings, Interpretation, and Limitations
Evidence Stream
Judgement
Key findings and interpretation:
• Apoptosis of immune cells is a high dose immunotoxic phenomenon that has been observed in both in
vivo and in vitro studies of PFOA.
• Disruption of the NF-kB signaling pathway, which is involved in T-cell responses, regulation of
apoptosis, and inflammatory response, has been demonstrated both directly and indirectly in in vivo
human and animal data, as well as in vitro.
• Inconsistent evidence of exacerbation of allergic immune and inflammatory responses via NF-kB
pathway, increased TNFa, and/or TH2 response.
Findings support
plausibility that PFOA
exposure can lead to
dysregulation of signaling
pathways related to
immune response;
however, data have
inconsistencies.
3-141
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
Limitations:
• Inconsistent findings between sexes, model systems, and studies regarding allergic immune
response.
• Limited database for immune response data.
• While PPARa is mechanistically linked to immune signaling (blocking the NF-kB pathway), it is
not clear if PFOA-induced alterations to PPARa are involved in immunomodulatory effects: some
PPARa-knockout mouse studies have suggested that immunomodulation occurs independent of
Notes: COVID-19 = coronavirus disease 2019; WBC = white blood cells; IgA = immunoglobulin A; IgG = immunoglobulin G; IgM = immunoglobulin M; SRBC = sheep red
blood cells; KLH = keyhole limpet hemocyanin; NF-kB = nuclear factor kappa B; TNFa = tumor necrosis factor alpha; Th2 = T helper 2; PPARa = peroxisome proliferator
activated receptor alpha.
aStudies may be of mixed confidence due to differences in how individual outcomes within the same study were assessed (e.g., clinical test vs self-reported data).
PPARa.
3-142
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.3 Cardiovascular
EPA identified 112 epidemiological and 9 animal toxicological studies that investigated the
association between PFOA and cardiovascular effects. Of the 54 epidemiological studies
addressing cardiovascular endpoints, 3 were classified as high confidence, 28 as medium
confidence, 14 as low confidence, 5 as mixed (1 high/medium and 4 medium/low) confidence,
and 4 were considered uninformative (Section 3.4.3.1). Of the 87 epidemiological studies
addressing serum lipid endpoints, 1 was classified as high confidence, 29 as medium confidence,
32 as low confidence, 19 as mixed (1 high/medium and 18 medium/low) confidence, and 8 were
considered uninformative (Section 3.4.3.1). Of the animal toxicological studies, 3 were classified
as high confidence, 4 as medium confidence, and 2 were considered low confidence (Section
3.4.3.2). Studies have mixed confidence ratings if different endpoints evaluated within the study
were assigned different confidence ratings. Though low confidence studies are considered
qualitatively in this section, they were not considered quantitatively for the dose-response
assessment (Section 4).
3.4.3.1 Human Evidence Study Quality Evaluation and Synthesis
3.4.3.1.1 Cardiovascular Endpoints
3.4.3.1.1.1 Introduction
Cardiovascular disease (CVD) is the primary cause of death in the United States with
approximately 12% of adults reporting a diagnosis of heart disease {Schiller, 2012, 1798736}.
Studied health effects include ischemic heart diseases (IHD), coronary artery disease (CAD),
coronary heart disease (CHD), hypertension, cerebrovascular disease, atherosclerosis (plaque
build-up inside arteries and hardening and narrowing of their walls), microvascular disease,
markers of inflammation (e.g., C-reactive protein), and mortality. These health outcomes are
interrelated—IHD is caused by decreased blood flow through coronary arteries due to
atherosclerosis resulting in myocardial ischemia.
There are 6 epidemiological studies from the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}
that investigated the association between PFOA and cardiovascular effects. Study quality
evaluations for these 6 studies are shown in Figure 3-29.
The 2016 PFOA Health Advisory {U.S. EPA, 2016, 3982042} and HESD {U.S. EPA, 2016,
3603279} did not identify strong evidence for an association between CVD and PFOA, based on
five occupational studies. Several occupational studies examined cardiovascular-related cause of
death among PFOA-exposed workers at the West Virginia Washington Works plant {Leonard,
2008, 1291100; Sakr, 2009, 2593135; Steenland, 2012, 2919168} and the 3M Cottage Grove
plant in Minnesota {Lundin, 2009, 1291108; Gilliland, 1993, 1290858; Raleigh, 2014,
2850270}. This type of mortality is of interest because of the relation between lipid profiles (e.g.,
LDL) and the risk of CVD. A study in West Virginia did not find an association between
cumulative PFOA levels and IHD mortality across four quartiles of cumulative exposure
{Steenland, 2012, 2919168}. Based on these data from the worker cohorts (part of the C8 Health
Project), the C8 Science Panel (2012, 1430770) concluded that there is no probable link between
PFOA and stroke and CAD. The analysis of the workers at the Minnesota plant also did not
observe an association between cumulative PFOA exposure and IHD risk, but an increased risk
of cerebrovascular disease mortality was seen in the highest exposure category {Lundin, 2009,
3-143
-------
DRAFT FOR PUBLIC COMMENT
March 2023
1291108}. These studies are limited by the reliance on mortality (rather than incidence) data,
which can result in a substantial degree of under ascertainment and misclassification. Evidence
was limited in studies on the general population, with only one high-exposure community study
and two NHANES studies examining the association between PFOA and hypertension risk.
Increased risk of hypertension was observed in a C8 community study {Winquist, 2014,
2851142}; however, the association was imprecise for estimates comparing the highest two
quintiles to the lowest quintile of exposure. One NHANES study identified in the 2021 ATSDR
Toxicological Profile for Perflaorocilkyls {ATSDR, 2021, 9642134} observed a large increased
risk of hypertension for adults not using hypertensive medication in the highest exposure quartile
{Min, 2012, 2919181}. The other NHANES study reported a decreased risk of hypertension in
children {Geiger, 2014, 2851286}.
,0®
Geiger etal., 2014, 2851286-
+
++ ++
+
+
+
+
+
Min etal., 2012, 2919181 -
-
+
+
+
+
+
+
+
Raleigh et al„ 2014, 2850270 -
+
+
+
+
Steenland and Woskie, 2012, 2919168-
+ *
+
+
+
+
+
+ *
Steenland etal., 2015, 2851015-
-
+
+ *
+
++
+
+
"
Winquist and Steenland, 2014, 2851142 -
+
+
+ *
+
++
+
+
+ *
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-29. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects Published Before 2016 (References from 2016 HESD)
Interactive figure and additional study details available on HAWC.
Since publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}, 49 new
epidemiological studies report on the association between PFOA and CVD, including outcomes
such as hypertension, CAD, congestive heart failure (CHF), microvascular diseases, and
mortality. Of these, 21 examined blood pressure or hypertension in adults. Pregnancy-related
hypertension is discussed in the Appendix (See PFOA Appendix). Two of the publications
3-144
-------
DRAFT FOR PUBLIC COMMENT
March 2023
{Girardi, 2019, 6315730; Steenland, 2015, 2851015} were occupational studies and the
remainder were conducted on the general population. Six general population studies {Honda-
Kohmo, 2019, 5080551; Hutcheson, 2020, 6320195; Bao, 2017, 3860099; Mi, 2020, 6833736;
Yu, 2021, 8453076; Ye, 2021, 6988486} were conducted in a high-exposure community in
China (i.e., C8 Health Project and "Isomers of C8 Health Project" populations), and three studies
{Canova, 2021, 10176518; Pitter, 2020, 6988479; Zare Jeddi, 2021, 7404065} were conducted
in a high-exposure community in Italy (i.e., Vento Region). Different study designs were also
used including three controlled trial studies {Cardenas, 2019, 5381549; Liu, 2018, 4238396;
Osorio-Yanez, 2021, 7542684}, 11 cohort studies {Fry, 2017, 4181820; Donat-Vargas, 2019,
5080588; Girardi, 2019, 6315730; Li, 2021, 7404102; Lin, 2020, 6311641; Manzano-Salgado,
2017, 4238509; Matilla-Santander, 2017, 4238432; Mitro, 2020, 6833625; Papadopoulou, 2021,
9960593; Steenland, 2015, 2851015; Warembourg, 2019, 5881345}, one case-control study
{Mattsson, 2015, 3859607}, and 35 cross-sectional studies {Averina, 2021, 7410155; Bao, 2017,
3860099; Canova, 2021, 10176518; Chen, 2019, 5387400; Christensen, 2016, 3858533;
Christensen, 2019, 5080398; Graber, 2019, 5080653; He, 2018, 4238388; Honda-Kohmo, 2019,
5080551; Huang, 2018, 5024212; Hutcheson, 2020, 6320195; Jain, 2020, 6311650; Jain, 2020,
6833623; Jain, 2020, 6988488; Zare Jeddi, 2021, 7404065; Khalil, 2018, 4238547; Khalil, 2020,
7021479; Koshy, 2017, 4238478; Koskela, 2022, 10176386; Leary, 2020, 7240043; Lin, 2020,
6988476; Liao, 2020, 6356903; Lin, 2013, 2850967; Lin, 2016, 3981457; Lind, 2017, 3858504;
Liu, 2018, 4238514; Ma, 2019, 5413104; Mi, 2020, 6833736; Mobacke, 2018, 4354163; Pitter,
2020, 6988479; Shankar, 2012, 2919176; Yang, 2018, 4238462; Yu, 2021, 8453076;Ye, 2021,
6988486}. The three controlled trial studies {Cardenas, 2019, 5381549; Liu, 2018, 4238396;
Osorio-Yanez, 2021, 7542684} were not controlled trials of PFAS exposures, but rather health
interventions: prevention of type 2 diabetes in the Diabetes Prevention Program (DPP) and
Outcomes Study (DPPOS) {Cardenas, 2019, 5381549; Osorio-Yanez, 2021, 7542684} and
weight loss in Prevention of Obesity Using Novel Dietary Strategies Lost (POUNDS-Lost) Study
{Liu, 2018, 4238396}. Thus, these studies can be interpreted as cohort studies for evaluating
cardiovascular risk purposes.
The studies were conducted in different study populations with the majority of studies conducted
in the United States {Cardenas, 2019, 5381549; Christensen, 2016, 3858533; Christensen, 2019,
5080398; Fry, 2017, 4181820; Graber, 2019, 5080653; He, 2018, 4238388; Honda-Kohmo,
2019, 5080551; Huang, 2018, 5024212; Hutcheson, 2020, 6320195; Jain, 2020, 6311650; Jain,
2020, 6833623; Jain, 2020, 6988488; Khalil, 2018, 4238547; Khalil, 2020, 7021479; Koskela,
2022, 10176386; Leary, 2020, 7240043; Koshy, 2017, 4238478; Li, 2021, 7404102; Liao, 2020,
6356903; Lin, 2020, 6311641; Liu, 2018, 4238514; Liu, 2018, 4238396; Ma, 2019, 5413104;
Mi, 2020, 6833736; Mitro, 2020, 6833625; Osorio-Yanez, 2021, 7542684; Shankar, 2012,
2919176; Steenland, 2015, 2851015}. The remaining studies were conducted in China {Bao,
2017, 3860099; Yang, 2018, 4238462; Yu, 2021, 8453076; Ye, 2021, 6988486}, Taiwan {Lin,
2013, 2850967; Lin, 2016, 3981457; Lin, 2020, 6988476}, Spain {Manzano-Salgado, 2017,
4238509; Matilla-Santander, 2017, 4238432}, Croatia {Chen, 2019, 5387400}, Sweden {Donat-
Vargas, 2019, 5080588;Lind, 2017, 3858504;Mattsson, 2015, 3859607;Mobacke, 2018,
4354163}, Italy {Canova, 2021, 10176518; Girardi, 2019, 6315730; Zare Jeddi, 2021, 7404065;
Pitter, 2020, 6988479}, Norway {Averina, 2021, 7410155}, and two studies conducted in several
European countries {Papadopoulou, 2021, 9960593; Warembourg, 2019, 5881345}. All the
studies measured PFOA in blood components (i.e., serum or plasma) with three studies
measuring levels in maternal serum {Papadopoulou, 2021, 9960593; Li, 2021, 7404102;
3-145
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Warembourg, 2019, 5881345}, and four studies measuring levels in maternal plasma
{Papadopoulou, 2021, 9960593; Warembourg, 2019, 5881345; Manzano-Salgado, 2017,
4238509; Mitro, 2020, 6833625}.
3.4.3.1.1.2 Study Quality
There are 48 epidemiological studies from recent systematic literature search and review efforts
conducted after publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} that
investigated the association between PFOA and cardiovascular effects. Study quality evaluations
for these 48 studies are shown in Figure 3-30, Figure 3-31, and Figure 3-32.
Of the 48 studies identified since the 2016 assessment, 3 studies were high confidence, 26 were
medium confidence, 12 were considered low confidence, 3 were considered mixed confidence,
and 4 studies were considered uninformative {Jain, 2020, 6833623; Jain, 2020, 6311650; Leary,
2020, 7240043; Seo, 2018, 4238334}. The main concerns with the low confidence studies
included the possibility of outcome misclassification (e.g., reliance on self-reporting) in addition
to potential for residual confounding or selection bias (e.g., unequal recruitment and participation
among subjects with outcome of interest, lack of consideration and potential exclusion due to
medication usage). Residual confounding was possible due to SES, which can be associated with
both exposure and the cardiovascular outcome. Although PFOA has a long half-life in the blood,
concurrent measurements may not be appropriate for cardiovascular effects with long latencies.
Further, temporality of PFOA exposure could not be established for several low confidence
studies due to their cross-sectional design. Several of the low confidence studies also had
sensitivity issues due to limited sample sizes {Christensen, 2016, 3858533; Girardi, 2019,
6315730; Graber, 2019, 5080653; Khalil, 2018, 4238547}. Two studies were rated adequate for
all domains, indicating lower risk-of-bias; however, both studies treated PFOA as the dependent
variable, resulting in both studies being considered uninformative {Jain, 2020, 6833623; Jain,
2020, 6311650}. Analyses treating PFOA as a dependent variable support inferences for
characteristics (e.g., kidney function, disease status, race/ethnicity) that affect PFOA levels in the
body, but it does not inform the association between exposure to PFOA and incidence of
cardiovascular disease. Small sample size (n = 45) and missing details on exposure
measurements were the primary concerns about the remaining uninformative study {Leary, 2020,
7240043}.
High and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies, the evidence synthesis below included details on any low confidence studies available.
Studies considered uninformative were not considered further in the evidence synthesis.
3-146
-------
DRAFT FOR PUBLIC COMMENT
March 2023
vO®
Averina et al., 2021, 7410155-
I
+
I
+
l
+
I
+
I
+
I
+
i
+
+
Bao et al., 2017, 3860099-
+
+
+
+
++
+
+
+
Canova etal., 2021, 10176518-
+
+
+
++
+
+
+
+
Cardenas et al., 2019, 5381549-
+
++
+
+
+
+
+
+
Chen et al., 2019, 5387400-
-
+
++
+
+
+
+
+*
Christensen et al., 2016, 3858533-
-
+
-
-
+
+
-
-
Christensen et al., 2019, 5080398-
+
+
+
+
+
+
+
+
Donat-Vargas et al., 2019, 5080588-
+
+
++
+
++
+
+
+
Fry et al., 2017, 4181820-
++
++
+
+
+
+
+
+
Girardi et al., 2019, 6315730-
-
+
-
-
+
-
-
Graber et al., 2019, 5080653-
-
+
-
-
+
+
-
-
He et al.,2018, 4238388-
-
+
++
+
-
+
+
-
Honda-Kohmo et al., 2019, 5080551 -
+
+
-
-
+
+
+
-
Huang etal., 2018, 5024212-
++ ++
+
+
+
+
+
+
Hutcheson et al., 2020, 6320195-
+
-
¦
+
+
+
+
+
Jain et al., 2020, 6988488-
+
+
+
+
+
+
+
+
Jain, 2020, 6311650-
+
+
+
+
+
+
+
~
Legend
D
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
b
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-30. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects
Interactive figure and additional study details available on HAWC.
3-147
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Jain, 2020, 6833623-
+
i
+
'
+
—1—
+
1
+
1
+
+
~
Khalil et al., 2018, 4238547-
-
+
+
-
+
+
-
Khalil et al., 2020, 7021479-
-
+
+
-
+
+
-
-
Koshy et al., 2017, 4238478 -
+
+
+*
-
+
+
+
-
Koskela etal.,2022, 10176386-
-
+
++
+
+
+
+
-
Leary et al., 2020, 7240043-
-
+
+
-
+
+
-
Li et al., 2021, 7404102-
B
++
+
++
+
+
VA
Liao etal., 2020, 6356903-
++
++
++
+
++
+
+
++
Linetal., 2016, 3981457-
B
B
fs*
+
+
+
+
+
Lin etal., 2020, 6311641 -
+
+
+
+
++
+
++
+
Lin et al., 2020, 6988476-
-
+
+
+
+
+
+
-*
Lind et al., 2017, 3858504-
+
++ ++
+
++
+
+
+
Liu etal., 2018, 4238396-
-
+
+
+
+
+
+
+
Liu et al., 2018, 4238514-
+
+
+
+
+
+
+
+
Ma et al., 2019, 5413104-
++
++
+
+
+
+
+
Manzano-Salgado et al., 2017, 4238509-
++
+
++
+
++
+
Legend
B
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
b
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-31. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-148
-------
DRAFT FOR PUBLIC COMMENT
March 2023
i\e° ^
9S>
^ ,,C
&
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Matilla-Santander et al., 2017, 4238432 -
I
+
I
+
I
+*
++ ++
i
+
i
+
n
Mattsson et al., 2015, 3859607 -
++
+
++
+
+
+
+
++
Mi et al., 2020, 6833736-
+
+
+
+
++
+
+
*
Mobacke et al., 2018, 4354163-
+
+
++
+
+
+
+
++
Osorio-Yanez et al., 2021, 7542684 -
-
+
++
+
+
+
+
+
Papadopoulou et al., 2021, 9960593-
+
+
+
+
+
+
+
+
Pitter et al., 2020, 6988479 -
+
+
+
+
++
+
+
+
Seo et al., 2018, 4238334-
-
+
-
D
-
-
~
Shankaret al., 2012, 2919176-
+
+
+*
+
++
+
+
+
Varshavsky et al., 2021, 7410195-
-
+
+
+
+
+
+
+
Warembourg et al., 2019, 5881345-
+
+
+
+
++
+
+
+
Yang et al., 2018, 4238462-
-
+
-
-
+
+
+
-
Ye et al., 2020, 6988486-
-
+
+
+
++
+
+
-
Yu et al., 2021, 8453076-
-
+
+
+
++
+
+
-
Zare Jeddi et al., 2021, 7404065 -
+
+
-*
+
+
+
+
+
Figure 3-32. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-149
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.3.1.1.3 Findings from Children and Adolescents
One high confidence study {Li, 2021, 7404102} and six medium confidence studies {Averina,
2021, 7410155; Canova, 2021, 10176518; Ma, 2019, 5413104; Manzano-Salgado, 2017,
4238509; Papadopoulou, 2021, 9960593; Warembourg, 2019, 5881345} examined blood
pressure in children and adolescents and reported no associations (see PFOA Appendix). No
association was observed in a high confidence study in infants from the Health Outcomes and
Measures of the Environment (HOME) Study {Li, 2021, 7404102} between PFOA in maternal
serum and child blood pressure measured at 12 years of age. In a cross-sectional analysis, Ma et
al. (2019, 5413104) did not observe an association between serum PFOA and blood pressure
among 2,251 NHANES (2003-2012) participants (mean age 15.5 years). Similarly, Manzano-
Salgado et al. (2017, 4238509) did not observe an association between maternal PFOA and child
blood pressure in combined or in gender-stratified analyses at age 4 and 7 years.
In a cohort of 1,277 children (age 6-11 years), PFOA measured both in maternal blood during
the pre-natal period and in plasma during the postnatal period were not associated with blood
pressure in single-pollutant models {Warembourg, 2019, 5881345}. However, the association
was significantly positive for systolic blood pressure (SBP) after co-adjustment for
organochlorine compounds (i.e., dichlorodiphenyldichloroethane (DDE) and hexachlorobenzene
(0.9; 95% CI: 0.1, 1.6; p = 0.021)). An overlapping study {Papadopoulou, 2021, 9960593}
examined the association for z-scores of blood pressure in children in a model mutually adjusted
for other PFAS and did not find an association. In a cross-sectional study of children and
adolescents in a high-exposure community {Canova, 2021, 10176518}, blood pressure was
lower among adolescents with increasing serum PFOA, but none of the associations reached
significance. An increased risk of hypertension (SBP > 130 mmHg and/or diastolic blood
pressure > 80 mmHg) was observed in a medium confidence cross-sectional study {Averina,
2021, 7410155} on Norwegian adolescents taking part in the Fit Futures. The magnitude of the
association was larger among increasing quartiles of PFOA exposure, reaching significance for
those in the fourth quartile of exposure (OR: 2.08; 95% CI: 1.17, 3.69, p = 0.013). Two low
confidence studies did not observe associations between serum PFOA and blood pressure
{Khalil, 2018, 4238547; Lin, 2013, 2850967}.
Other cardiovascular conditions reported in children and adolescents include carotid intima-
media thickness test (CIMT) and brachial artery distensibility. Two medium confidence studies
that examined CIMT among adolescents and young adults from the Young Taiwanese Cohort
Study {Lin, 2013, 2850967; Lin, 2016, 3981457} reported no associations. A low confidence
study of children and adolescents from the World Trade Center (WTC) Health Registry reported
PFOA was significantly associated with increased brachial artery distensibility (0.45; 95% CI:
0.04, 0.87; p = 0.03), but was not associated with pulse wave velocity {Koshy, 2017, 4238478}.
However, concerns for residual confounding by age and SES contributed to the low confidence.
3.4.3.1.1.4 Findings from the General Adult Population
Most of the studies identified since the last assessment were conducted among general
population adults (see PFOA Appendix). A total of 15 studies examined PFOA in association
with SBP, diastolic blood pressure (DBP), hypertension, and elevated blood pressure {Bao,
2017, 3860099; Chen, 2019, 5387400; Christensen, 2016, 3858533; Christensen, 2019, 5080398;
Donat-Vargas, 2019, 5080588; He, 2018, 4238388; Zare Jeddi, 2021, 7404065; Mitro, 2020,
3-150
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6833625; Liao, 2020, 6356903; Lin, 2020, 6311641; Liu, 2018, 4238514; Liu, 2018, 4238396;
Mi, 2020, 6833736; Pitter, 2020, 6988479; Yang, 2018, 4238462}.
Of the ten studies that examined blood pressure as a continuous measure, six reported
statistically significant positive associations {Liao, 2020, 6356903; Mi, 2020, 6833736; Bao,
2017, 3860099; Lin, 2020, 6311641; Liu, 2018, 4238396; Pitter, 2020, 6988479; Yang, 2018,
4238462}. However, the results were not always consistent between SBP and DBP.
A high confidence study in 6,967 NHANES (2003-2012) participants 20 years and older
reported a statistically significant positive association with SBP (per 10-fold change in PFOA:
1.83; 95% CI: 0.40, 3.25) in the fully adjusted model {Liao, 2020, 6356903}. No association was
observed for DBP.
A high confidence study {Mitro, 2020, 6833625} conducted among 761 women that examined
associations between PFOA concentrations measured during pregnancy and blood pressure
assessed at 3 years post-partum reported a positive but non-significant association with SBP
(beta per doubling of PFOA: 0.8; 95% CI: -0.3, 1.8). No association was observed with DBP.
Two medium confidence cross-sectional studies with overlapping data from the "Isomers of C8
Health Project," a highly-exposed population of Shenyang, China {Mi, 2020, 6833736; Bao,
2017, 3860099}, also reported positive associations for blood pressure. In 1,612 participants with
elevated PFOA levels (median 6.19 ng/mL), Bao et al. (2017, 3860099) reported large increases
in DBP (2.18; 95% CI: 1.38, 2.98) and SBP (1.69; 95% CI: 0.25, 3.13). After stratification by
sex, a positive association was observed in men only for DBP (1.48; 95% CI: 0.58, 2.37) and in
women only for SBP (6.65; 95% CI: 4.32, 8.99). In participants with high PFOA levels (median
4.8 ng/mL), Mi et al. (2020, 6833736) observed statistically significant increases in DBP (1.49;
95% CI: 0.34, 2.64). No association was observed for SBP.
Similar findings were observed in another medium confidence study in a high-exposure
community in Italy {Pitter, 2020, 6988479}. Adults (20-39 years old) included in a regional
(i.e., Vento Region) surveillance program were included in a cross-sectional analysis of blood
pressure and PFOA exposure. Significant positive associations were reported for DBP (0.34;
95% CI: 0.21, 0.47) and SBP (0.37; 95% CI: 0.19, 0.54) in the overall (n = 15,380) population.
Results were generally consistent after stratification by sex. Minor sex differences were
observed, such as slightly larger increases in SBP among men (0.46; 95% CI: 0.19, 0.73) and
larger increases in DBP among women (0.39; 95% CI: 0.21, 0.57). Monotonic trends were
observed in all quartile analyses, although significance was not reported.
Lin et al. (2020, 6311641), a medium confidence study using data from the Diabetes Prevention
Program, a randomized controlled health intervention trial, reported that an increase in baseline
PFOA concentration was significantly associated with higher SBP (1.49; 95% CI: 0.29, 2.70); no
association was observed with DBP or pulse pressure. In a medium confidence weight loss-
controlled trial population (the POUNDS Lost Study), Liu et al. (2018, 4238396) observed that
baseline PFOA was positively correlated with DBP (p < 0.05), but at 6- and 24-month follow-up
assessments, no associations were observed with SBP or DBP {Liu, 2018, 4238396}.
The findings from three low confidence studies {Chen, 2019, 5387400; He, 2018, 4238388;
Yang, 2018, 4238462} of PFOA and blood pressure were mixed. Yang et al. (2018, 4238462)
3-151
-------
DRAFT FOR PUBLIC COMMENT
March 2023
reported a statistically significant positive increased risk of high SBP (> 140 mmHg) for n-PFOA
(linear isomers), but no association for SBP as a continuous measure. Two additional studies
reported no associations for SBP {Chen, 2019, 5387400; He, 2018, 4238388}, and three studies
reported no associations for DBP {Chen, 2019, 5387400; He, 2018, 4238388; Yang, 2018,
4238462}.
Of the eleven studies that examined risk of elevated blood pressure (hypertension), six reported
statistically significant associations {Liao, 2020, 6356903; Mi, 2020, 6833736; Bao, 2017,
3860099; Lin, 2020, 6311641; Pitter, 2020, 6988479; Ye, 2021, 6988486}. Hypertension was
defined as average SBP >140 mmHg and average DBP > 90 mmHg, or self-reported use of
prescribed anti-hypertensive medication. Using a generalized additive model and restricted cubic
splines, Liao et al. (2020, 6356903) reported a non-linear (J-shaped) relationship with
hypertension, with the inflection point of PFOA at 1.80 ng/mL. Each 10-fold increase in PFOA
was associated with a 44% decrease (OR: 0.56; 95% CI: 0.32, 0.99) in the risk of hypertension
on the left side of the inflection point, and an 85% increase (OR: 1.85; 95% CI: 1.34, 2.54) on
the right side of the inflection point. A significant association with hypertension was observed
for the highest (> 4.4 ng/mL) vs. lowest (< 2.5 ng/mL) tertile (OR: 1.32; 95% CI: 1.13, 1.54),
and the test for trend was significant (p < 0.001). Additionally, positive associations were
observed among women (OR: 1.42; 95% CI: 1.12, 1.79) and in participants 60 years and older
(OR: 1.32; 95% CI: 1.03, 1.68). The studies {Mi, 2020, 6833736; Bao, 2017, 3860099; Ye,
2021, 6988486} with overlapping data on highly-exposed Isomers of C8 Health Project
participants reported significant associations. An overlapping low confidence study {Ye, 2021,
6988486} on metabolic syndrome observed a moderate increase (OR: 1.31; 95% CI: 1.11, 1.56)
in the risk of elevated blood pressure (SBP > 130 and/or DBP > 85; or medication use). Mi et al.
(2020, 6833736) reported higher risk of hypertension overall (OR: 1.72; 95% CI: 1.27, 2.31) and
among women (OR: 2.32; 95% CI: 1.38, 3.91), but not in men. Bao et al. (2017, 3860099) did
not observe an association between total-PFOA and hypertension. However, in isomer-specific
analysis, a natural-log unit (ng/mL) increase of 6-m-PFOA was significantly associated with
higher risk of hypertension among all participants (OR: 1.24; 95% CI: 1.05, 1.47) and among
women (OR: 1.86; 95% CI: 1.25, 2.78). These results suggest branched PFOA isomers have a
stronger association with increased risk of hypertension compared to linear isomers (n-PFOA).
Increased risk of hypertension was observed in a pair of overlapping studies on another high
exposure community located in Italy {Zare Jeddi, 2021, 7404065; Pitter, 2020, 6988479}. Pitter
et al. (2020, 6988479), a medium confidence study, observed a significant association (OR: 1.06;
95%) CI: 1.01, 1.12) between PFOA exposure and hypertension in a large cross-sectional sample
of adults (n = 15,786). The association remained significant in men (OR: 1.08; 95% CI: 1.02,
1.15), but was not significant in women (OR: 1.06; 95% CI: 0.97, 1.15). A similar increased risk
of hypertension was observed among all participants in the overlapping study {Zare Jeddi, 2021,
7404065}.
A medium confidence study, Lin et al. (2020, 6311641), reported in a cross-sectional analysis
that the association with hypertension was not statistically significant but was modified by sex.
Among males, a doubling of baseline plasma PFOA was associated with a significantly higher
risk of hypertension (RR: 1.27; 95% CI: 1.06, 1.53); no association with hypertension was
observed among females. In a prospective analysis, among participants who did not have
hypertension at baseline, there was no association with hypertension at the approximately
3-152
-------
DRAFT FOR PUBLIC COMMENT
March 2023
15 years of follow-up {Lin, 2020, 6311641}. In addition, three medium confidence studies
{Donat-Vargas, 2019, 5080588; Christensen, 2019, 5080398; Liu, 2018, 4238514} and a low
confidence study {Christensen, 2016, 3858533} did not observe associations with hypertension.
Ten studies examined other CVD-related outcomes including CHD, stroke, carotid artery
atherosclerosis, angina pectoris, C-reactive protein, CHF, peripheral artery disease (PAD),
microvascular disease, CIMT, and mortality.
Among the four studies that examined CHD, the findings were mixed. A high confidence study
{Mattsson, 2015, 3859607}, a medium confidence study of 10,850 NHANES participants from
1999-2014 {Huang, 2018, 5024212}, and a low confidence study {Christensen, 2016, 3858533}
all reported no associations with CHD. A low confidence study from the C8 Health Project
{Honda-Kohmo, 2019, 5080551} reported a significant inverse association between PFOA and
CHD among adults with and without diabetes. However, study limitations that may have
influenced these findings include the reliance on self-reporting of a clinician-based diagnosis for
CHD outcome classification and residual confounding by SES.
Among the two NHANES-based studies that examined CVD {Shankar, 2012, 2919176; Huang,
2018, 5024212}, the findings were mixed. Using data from NHANES 1999-2000 and 2003-
2004 cycles, Shankar et al. (2012, 2919176) reported significant associations with CVD. The
analysis by PFOA quartiles reported significantly higher odds for the presence of CVD in the
third (OR: 1.77; 95% CI: 1.04, 3.02) and the highest (OR: 2.01; 95% CI: 1.12, 3.60) quartiles
compared to the lowest quartile, with a significant trend (p = 0.01). In contrast, using a larger
dataset from NHANES 1999-2014 cycles, Huang et al. (2018, 5024212) did not observe an
association with total CVD by quartiles of exposure, nor a positive trend.
Shankar et al. (2012, 2919176) also observed a significant association with PAD. The analysis
by PFOA quartiles reported significantly higher odds for the presence of PAD (OR: 1.78; 95%
CI: 1.03, 3.08) in the highest compared to the lowest quartile, with a significant trend (p = 0.04).
Among the two studies that examined stroke, the findings also were mixed. A borderline positive
association (p = 0.045) was observed by Huang et al. (2018, 5024212). In contrast, Hutcheson et
al. (2020, 6320195) observed a significant inverse association with history of stroke in adults
with and without diabetes participating in the C8 Health Project (OR: 0.90; 95% CI: 0.82, 0.98,
p = 0.02). However, a borderline-significant inverse association was observed among non-
diabetics (OR: 0.94; 95% CI: 0.88, 1.00; p = 0.04) but not among those with diabetes, although
the interaction was not significant.
In addition, a low confidence study of adults and children did not observe an association between
serum PFOA and self-reported cardiovascular conditions, including high blood pressure, CAD,
and stroke {Graber, 2019, 5080653}. However, potential selection bias is a major concern for
this study owing to the recruitment of volunteers who already knew their PFAS exposure levels
and were motivated to participate in a lawsuit.
Huang et al. (2018, 5024212) also reported significantly higher odds of heart attack for the third
quartile (OR: 1.62; 95% CI: 1.04, 2.53) and second quartile (OR: 1.57; 95% CI: 1.06, 2.34),
compared to the first quartile. No associations were observed with CHF and angina pectoris.
3-153
-------
DRAFT FOR PUBLIC COMMENT
March 2023
No associations with microvascular diseases (defined as the presence of nephropathy,
retinopathy, or neuropathy) were observed {Cardenas, 2019, 5381549}.
One medium confidence study {Osorio-Yanez, 2021, 7542684} examined changes in
atherosclerotic plaque in a sample of participants enrolled in the Diabetes Prevention Program. A
non-significant positive association (OR: 1.17; 95% CI: 0.91, 1.50) was observed for the odds of
having a mild to moderate coronary artery calcium Agatston score (11- 400). Two studies
examined changes in heart structure {Mobacke, 2018, 4354163} and carotid atherosclerosis
{Lind, 2017, 3858504} in participants 70 years and older. Mobacke et al. (2018, 4354163)
examined alterations of left ventricular geometry, a risk factor for CVD, and reported that serum
PFOA was significantly associated with a decrease in relative wall thickness (-0.12; 95% CI:
-0.22, -0.001; p = 0.03), but PFOA was not observed to be associated with left ventricular mass
or left ventricular end diastolic diameter. Lind et al. (2017, 3858504) examined markers of
carotid artery atherosclerosis including atherosclerotic plaque, the intima-media complex, and
the CIMT (a measure used to diagnose the extent of carotid atherosclerotic vascular disease) and
observed no associations.
The association between exposure to PFOA and apolipoprotein B, a protein associated with LDL
and increased risk of arthrosclerosis, was examined in a medium confidence study {Jain, 2020,
6311650} onNHANES participants (2007-2014). Serum apolipoprotein B was significantly
increased (beta per loglO-unit increase PFOA: 0.03878; p < 0.01) with increasing PFOA
exposure in non-diabetic participants who did not take lipid-lowering medication. No significant
associations were observed among lipid-lowering medication users and participants with
diabetes. No association between PFOA and C-reactive protein levels (a risk factor for CVD)
were observed in two studies, one in women from Project Viva {Mitro, 2020, 6833625} and the
other in pregnant women from the Spanish Environment and Childhood (Infancia y Medio
Ambiente, INMA) study {Matilla-Santander, 2017, 4238432}. One medium confidence study
examined mortality due to heart/cerebrovascular diseases in 1,043 NHANES (2003-2006)
participants 60 years and older and observed no associations {Fry, 2017, 4181820}.
Overall, the findings from one high confidence study and several medium confidence studies
conducted among the general population did not provide consistent evidence for an association
between PFOA and SBP and DBP. The evidence for an association between PFOA and
increased risk of hypertension/elevated blood pressure, overall and in gender-stratified analyses
was inconsistent. Evidence for other CVD-related outcomes was more limited, and similarly
inconsistent.
3.4.3.1.1.5 Findings from Occupational Studies
Two low confidence studies examined occupational PFOA exposure and cardiovascular effects
(see PFOA Appendix). Steenland et al. (2015, 2851015) examined 1,881 workers with high
serum PFOA levels (median 113 ng/mL) from a subset of two prior studies conducted by the C8
Science Panel. No trend was observed in the exposure-response gradient for stroke, CHD, and
hypertension and. In analysis of PFOA levels by quartiles, a significantly higher risk of stroke
(no lag) was observed for the 2nd quartile vs. the 1st quartile (Rate Ratio (RR): 2.63; 95% CI:
1.06, 6.56). No association was observed with 10-year lag stroke, CHD, and hypertension,
respectively. For the assessment of stroke, this study had low confidence because of concerns for
selection bias, specifically survival bias. For other chronic diseases examined, this study is of
3-154
-------
DRAFT FOR PUBLIC COMMENT
March 2023
low confidence due to concerns about outcome misclassification, particularly for hypertension
due to lack of medical record validation. In another occupational study of 120 male workers with
very high PFOA serum levels (GM: 4,048 ng/mL), Girardi et al. (2019, 6315730) reported no
association with increased risk of mortality due to cardiovascular causes, including hypertensive
disease, ischemic heart disease, stroke, and circulatory diseases. However, the potential for
selection bias, outcome misclassification, and limited control for confounding may have
influenced the reported results.
Overall, the limited evidence available from occupational studies was inconsistent for an
association with risk of stroke and indicated PFOA is not associated with an increased risk of
CHD, hypertension, and mortality due to cardiovascular causes. However, the findings based on
two low confidence studies should be interpreted with caution due to potential biases arising
from the selection of participants and outcome misclassification.
3.4.3.1.2 Serum Lipids
3.4.3.1.2.1 Introduction
Serum cholesterol and triglycerides are well-established risk factors for CVDs. Major cholesterol
species in serum include LDL and HDL. Elevated levels of TC, LDL, and triglycerides are
associated with increased cardiovascular risks, while higher levels of HDL are associated with
reduced risks. There are 21 epidemiological studies (22 publications)14 from the 2016 PFOA
HESD {U.S. EPA, 2016, 3603279} that investigated the association between PFOA and serum
lipid effects. Study quality evaluations for these 22 studies are shown in Figure 3-33.
In the 2016 Health Assessment {U.S. EPA, 2016, 3603279} for PFOA, there was relatively
consistent and robust evidence of positive associations between PFOA and TC and LDL in
occupational {Sakr, 2007, 1291103; Sakr, 2007, 1430761; Olsen, 2003, 1290020; Costa, 2009,
1429922} and high-exposure community settings {Frisbee, 2010, 1430763; Steenland, 2009,
1291109; Fitz-Simon, 2013, 2850962; Winquist, 2014, 2851142}. Two of the studies were cross-
sectional, however, Fitz-Simon (2013, 2850962) reported positive associations for LDL and TC
in a longitudinal analysis of the change in lipids seen in relation to a change in serum PFOA.
General population studies {Lin, 2009, 1290820; Geiger, 2014, 2850925; Nelson, 2010,
1291110} in children and adults using NHANES reported positive associations for TC and
increased risk of elevated TC. Other general population studies were generally consistent,
reporting positive associations for TC in adults {Fisher, 2013, 2919156; Eriksen, 2013,
2919150} and pregnant women {Starling, 2014, 2850928}. Positive associations between PFOA
and HDL were also observed in most studies in the general population {Lin, 2009, 1290820;
Frisbee, 2010, 1430763; Steenland, 2009, 1291109; Fisher, 2013, 2919156}. Positive
associations were observed for triglycerides and LDL in high-exposure community studies
{Frisbee, 2010, 1430763; Steenland, 2009, 1291109}, but associations for triglycerides and LDL
were less consistent in other general population studies {Fisher, 2013, 2919156; Lin, 2009,
1290820; Geiger, 2014, 2850925}.
14 Olsen (2003,1290020) is the peer-review paper of Olsen (2001, 10228462).
3-155
-------
DRAFT FOR PUBLIC COMMENT
March 2023
*\v|© cSP w 0°
®c S^Cr^
Legend
I Good (metric) or High confidence (overall)
| Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Costa et al., 2009, 1429922 -
+
+*
+
+
+
~
a
Emmett et al., 2006, 1290905 -
-
-
D
-
¦
Eriksen et al., 2013, 2919150 -
+
+
+
+
++
+
+
+
Fisher et al., 2013, 2919156-
£
+
-
+
++
+
+
+
Fitz-Simon et al., 2013, 2850962 -
+
++
+*
+
++
+
+
+#
Frisbee et al., 2010, 1430763 -
+
+
+*
+
+
+
+
+*
Fu et al., 2014, 3749193-
-
+
+*
+
+
-
Geiger et al., 2014, 2850925 -
+
++ ++
+
+
+
+
+
Lin et al., 2009, 1290820-
+
+
+
+
+
+
+
Maisonet et al., 2015, 3981585 -
+
+
+#
+
+
+
+
+*
Nelson et al., 2010, 1291110 -
++
+
+
+
++
+
+
+
Olsen and Zobel, 2007, 1290836 -
+
+
+*
+
-
+
-
Olsen et al., 2000, 1424954 -
-
-
+
+
+
+
+
¦
Olsen et al., 2001, 10228462 -
-
+
-*
+
+
+
+
-*
Olsen et al., 2003, 1290020 -
-
+
-
+
+
+
+
-
Sakr et al., 2007, 1291103-
+
+
+
+
+
+
+
Sakretal., 2007, 1430761 -
-
+
+*
+
+
+*
Starling et al., 2014, 2850928 -
+
+
+*
+
++
+
+
+*
Steenland et al., 2009, 1291109 -
+
+
+*
+
+
+
+
+~
Steenland et al., 2015, 2851015 -
-
+
-
+
++
+
+
-
Timmermann et al., 2014, 2850370 -
+
+
+
+
+
+
+
Winquist and Steenland, 2014, 2851142 -
+
+
+
+
++
+
+
+
Figure 3-33. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids Published Before 2016 (References from 2016 PFOA HESD)
Interactive figure and additional study details available on HAWC.
3-156
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.3.1.2.2 Study Quality
All studies were evaluated for risk of bias, selective reporting, and sensitivity following the EPA
IRIS protocol. Three considerations were specific to evaluating the quality of studies on serum
lipids. First, because lipid-lowering medications strongly affect serum lipid levels, unless the
prevalence of medication use is assumed to be low in the study population (e.g., children),
studies that did not account for the use of lipid-lowering medications by restriction, stratification,
or adjustment were rated as deficient in the participant selection domain. Second, because
triglyceride levels are sensitive to recent food intake {Mora, 2016, 9564968}, outcome
measurement error is likely substantial when triglyceride is measured without fasting. Thus,
studies that did not measure triglycerides in fasting blood samples were rated deficient in the
outcome measures domain for triglycerides. The outcome measures domain for LDL was also
rated deficient if LDL was calculated based on triglycerides. Fasting status did not affect the
outcome measures rating for TC, directly measured LDL, and HDL because the serum levels of
these lipids change minimally after a meal {Mora, 2016, 9564968}. Third, measuring PFOA and
serum lipids concurrently was considered adequate in terms of exposure assessment timing.
Given the long half-life of PFOA (median half-life = 2.7 years) {Li, 2018, 4238434}, current
blood concentrations are expected to correlate well with past exposures. Furthermore, although
reverse causation due to hypothyroidism {Dzierlenga, 2020, 6833691} or enterohepatic cycling
of bile acids {Fragki, 2021, 8442211} has been suggested, there is not yet clear evidence to
support these reverse causal pathways.
Since publication of the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}, 66 new
epidemiological studies (65 publications)15 report on the association between PFOA exposure
and serum lipids. Except for ten studies {Olsen, 2012, 2919185; Domazet, 2016, 3981435; Lin,
2019, 5187597; Liu, 2020, 6318644; Donat-Vargas, 2019, 5080588; Liu, 2018, 4238396;
Blomberg, 2021, 8442228; Sinisalu, 2020, 7211554; Li, 2021, 7404102; Tian, 2020, 7026251},
all studies were cross-sectional. Some cohort studies provided additional cross-sectional analyses
{Blomberg, 2021, 8442228; Sinisalu, 2020, 7211554; Li, 2021, 7404102}. Most studies assessed
exposure to PFOA using biomarkers in blood, and measured serum lipids with standard clinical
biochemistry methods. Serum lipids were frequently analyzed as continuous outcomes, but a few
studies examined the prevalence or incidence of hypercholesterolemia, hypertriglyceridemia, and
low HDL based on clinical cut-points, medication use, doctor's diagnosis, or criteria for
metabolic syndrome. Study quality evaluations for these 65 studies are shown in Figure 3-34,
Figure 3-35, and Figure 3-36.
Based on the considerations mentioned, one study was classified as high confidence, one study
was classified as high confidence for prospective analyses and medium confidence for cross-
sectional analyses, 21 studies were classified medium confidence for all lipid outcomes, nine
studies were rated medium confidence for TC or HDL, but low confidence for triglycerides or
LDL, 26 studies were rated low confidence for all lipid outcomes, and 7 studies were rated
uninformative for all lipid outcomes {Seo, 2018, 4238334; Abraham, 2020, 6506041; Predieri,
2015, 3889874; Huang, 2018, 5024212; Leary, 2020, 7240043; Sinisalu et al., 2020, 7211554;
Sinisalu, 2021, 9959547}. Notably, ten studies {Zeng, 2015, 2851005; Manzano-Salgado, 2017,
4238509; Canova, 2020, 7021512; Matilla-Santander, 2017, 4238432; Lin, 2020, 6988476;
Blomberg, 2021, 8442228; Tian, 2020, 7026251; Yang, 2020, 7021246; Canova, 2021,
15 Dong 2019, 5080195 counted as two studies, one in adolescents and one in adults.
3-157
-------
DRAFT FOR PUBLIC COMMENT
March 2023
10176518; DallaZuanna, 2021, 7277682} were rated low confidence specifically for
triglycerides and/or LDL because these studies measured triglycerides in non-fasting blood
samples. The low confidence studies had deficiencies in participant selection {Wang, 2012,
2919184; Khalil, 2018, 4238547; Lin, 2013, 2850967; Lin, 2020, 6315756; Fassler, 2019,
6315820; Chen, 2019, 5387400; Li, 2020, 6315681; He, 2018, 4238388; Yang, 2018, 4238462;
Christensen, 2016, 3858533; Graber, 2019, 5080653; Sun, 2018, 4241053; Rotander, 2015,
3859842; Liu, 2018, 4238396; Cong, 2021, 8442223; Khalil, 2020, 7021479; Kobayashi, 2021,
8442188; Liu, 2021, 10176563; Ye, 2021, 6988486; Yu, 2021, 8453076}, outcome measures
{Koshy, 2017, 4238478; Yang, 2018, 4238462; Christensen, 2016, 3858533; Kishi, 2015,
2850268; Graber, 2019, 5080653; Rotander, 2015, 3859842; Kobayashi, 2021, 8442188},
confounding {Wang, 2012, 2919184; Convertino, 2018, 5080342; Khalil, 2018, 4238547;
Koshy, 2017, 4238478; Olsen, 2012, 2919185; Lin, 2013, 2850967; Lin, 2020, 6315756; Fassler,
2019, 6315820; Li, 2020, 6315681; Yang, 2018, 4238462; Christensen, 2016, 3858533; Graber,
2019, 5080653; Khalil, 2020, 7021479; Liu, 2021, 10176563; Sinisalu, 2020, 7211554}, analysis
{He, 2018, 4238388; Sun, 2018, 4241053; Liu, 2018, 4238396}, sensitivity {Wang, 2012,
2919184; Khalil, 2018, 4238547; Olsen, 2012, 2919185; Christensen, 2016, 3858533; Graber,
2019, 5080653; Rotander, 2015, 3859842; Sinisalu, 2020, 7211554}, or selective reporting
{Dong, 2019, 5080195, adolescent portion}.
The most common reason for a low confidence rating was potential for selection bias, including a
lack of exclusion based on use of lipid-lowering medications {Wang, 2012, 2919184; Lin, 2020,
6315756; Chen, 2019, 5387400; Li, 2020, 6315681; He, 2018, 4238388; Yang, 2018, 4238462;
Sun, 2018, 4241053; Liu, 2018, 4238396; Cong, 2021, 8442223; Liu, 2021, 10176563; Ye,
2021, 6988486; Yu, 2021, 8453076}, potential for self-selection {Li, 2020, 6315681;
Christensen, 2016, 3858533; Graber, 2019, 5080653; Rotander, 2015, 3859842}, highly unequal
recruitment efforts in sampling frames with potentially different joint distributions of PFOA and
lipids {Lin, 2013, 2850967}, and missing key information on the recruitment process {Khalil,
2018, 4238547; Fassler, 2019, 6315820; Yang, 2018, 4238462; Khalil, 2020, 7021479}. Another
common reason for low confidence was a serious risk for residual confounding by SES {Wang,
2012, 2919184; Khalil, 2018, 4238547; Koshy, 2017, 4238478; Olsen, 2012, 2919185; Lin,
2013, 2850967; Lin, 2020, 6315756; Fassler, 2019, 6315820; Li, 2020, 6315681; Yang, 2018,
4238462; Christensen, 2016, 3858533; Graber, 2019, 5080653; Sinisalu, 2020, 7211554}.
Frequently, deficiencies in multiple domains contributed to an overall low confidence rating. The
uninformative studies had critical deficiencies in at least one domain or were deficient in several
domains. These critical deficiencies include a lack of control for confounding {Seo, 2018,
4238334; Huang, 2018, 5024212; Abraham, 2020, 6506041}, convenience sampling {Sinisalu,
2021, 9959547}, and treating PFOA as an outcome of all lipids instead of an exposure, which
limits the ability to make causal inference for the purpose of hazard determination {Predieri,
2015, 3889874}. Small sample size (n = 45) and missing details on exposure measurements were
the primary concerns of the remaining uninformative study {Leary, 2020, 7240043}.
High and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies, the evidence synthesis below included details on any low confidence studies available.
Studies considered uninformative were not considered further in the evidence synthesis.
3-158
-------
DRAFT FOR PUBLIC COMMENT
March 2023
.0^
avS^ A\^
^ M**
Sins'" QMe<
Abraham et al., 2020, 6506041 -
+
+
L+*
B
-
+
| [|
Averina et al., 2021, 7410155-
+
+
+*
+
+
+
+
+
Bjorke-Monsen et al., 2020, 7643487-
-
+
+*
+
+
+
-
Blomberg et al., 2021, 8442228 -
+
+
+*
+
+
+
+*
Canova et al., 2020, 7021512-
+
+
n*
+
+
+*
Canova et al., 2021,10176518 -
+
+
+*
+
+
+*
Chen etal., 2019, 5387400-
-
+
++
+
+
+
+
Christensen et al., 2016, 3858533-
¦
+
+
+
-
Christensen et al., 2019, 5080398-
*
+
+
+
+
+
+
+
Cong et al., 2021, 8442223-
¦
+
+
+
+
+
+
Convertino et al., 2018, 5080342-
+
++
+
-
-
+
Dalla Zuanna et al., 2021, 7277682 -
-
i
+
+*
+
-
-
+
+*
Domazet et al., 2016, 3981435-
+
+ I
+
+
+
+
+
Donat-Vargas et al., 2019, 5080588-
+
-
++
+
++
+
+
+
Dong etal., 2019, 5080195-
-
+
+
+
+
-
+
+
Fan etal., 2020, 7102734-
-
+
+
++
+
+
+
Fassler et al., 2019, 6315820-
-
+
-
+
+
+
Gardener et al., 2021, 7021199-
+
++
+
+
++
Graber et al., 2019, 5080653-
¦
+
-
+
+
-
-
Han etal., 2021, 7762348-
*
+
+
-
+
+
+
+
He et al., 2018, 4238388-
¦
+
++
+
¦
+
+
Huang etal., 2018, 5024212-
++ ++
~
-
+
+
+ | j
I
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-34. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids
Interactive figure and additional study details available on HAWC.
3-159
-------
DRAFT FOR PUBLIC COMMENT
March 2023
,o®
Jain et al., 2018, 5079656-
++
+
+
+
+
+
+
+
Jain et al.,2019, 5080642-
+
+
+
+
+
+
+
+
Jensen et al., 2020, 6833719 -
+
+
+
+
+
+*
+*
Kang et al., 2018, 4937567-
+
+
+
+
+
+
+
+
Khalil et al., 2018, 4238547-
-
+
+
-
+
+
-
Khalil et al., 2020, 7021479-
+
+
¦
+
+
-
Kishi et al.,2015, 2850268-
-
+
-
+
+
-
+
-
Kobayashi et al., 2021, 8442188 -
-
+
-
+
+
+
+
-
Kobayashi et al., 2022, 10176408 -
+
++
-
+
+
+
+
+
Koshy et al., 2017, 4238478 -
+
+
-
-
+
+
+
-
Leary et al., 2020, 7240043 -
-
+
+
-
+
+
~
Li et al., 2020, 6315681 -
+
+*
-
+
~
Li et al., 2021, 7404102-
+
+
+
+
'fr
Lin et al.,2013, 2850967-
-
+
-
+
+
+
¦
Linet al.,2019, 5187597-
++
+
++
+
++
+
+
Lin et al., 2020, 6315756-
-
+
+
+
+
+
+
+
Lin et al., 2020, 6988476 -
-
+
-
+
+
+
+
¦
Liu et al., 2018, 4238396-
-
+
++
+
+
+
+
-
Liu et al., 2018, 4238514-
+
+
+
+
+
+
+
+
Liuet al., 2020, 6318644-
+
+
+
+
++
+
+
+
Liu et al., 2021, 10176563-
-
-*
++
-
+
+
+
-
Manzano-Salgado et al., 2017, 4238509 -
-
++
+*
+
++
+
++
+*
Legend
D
Good (metric) or High confidence (overall)
r
Adequate (metric) or Medium confidence (overall)
~
Deficient (metric) or Low confidence (overall)
B
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-35. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued)
Interactive figure and additional study details available on HAWC.
3-160
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Vvsi © c,e^ \\ o°
**%# a""*
,o©
Legend
| Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Matilla-Santander et al., 2017, 4238432 -
l
+
l
+
+*
J
+
+
+*
Mora et al.,2018, 4239224-
+
+
+
+
+
+
+
Olsen et al., 2012, 2919185-
+
+
+
-
+
+
-
-
Papadopoulou et al., 2021, 9960593-
+
+
+
+
+
+
+
+
Predieri et al., 2015, 3889874-
+
+
-
-
+
+
-
~
Rotander et a I., 2015, 3859842 -
-
+
-
+
+
+
J
Seo et al.,2018, 4238334-
-
+
-
~
Sinisalu et al., 2020, 7211554-
+
+
' H
-
+
~
Sinisalu et al., 2021, 9959547 -
B
+
+
-
-
+
~
Skuladottir et al., 2015, 3749113-
+
+
+
+
+
+
+
+
Spratlen et al., 2020, 5915332 -
+
+*
+
++
+
+
+
Sun et al.,2018, 4241053-
•
++
+
+
-
+
+
-
Tian et al., 2020, 7026251 -
+
+
+*
+
++
+
+
+
Varshavsky et al., 2021, 7410195-
-
-
+
+
+
+
-
Wang etal.,2012, 2919184-
-
+
+
-
+
+
-
Yang et al.,2018, 4238462-
-
+
-
-
+
+
+
-
Yang etal.,2020, 7021246-
+
+*
+
+
+
+
+*
Ye et al., 2020, 6988486-
-
+
+
+
++
+
+
-
Yu eta I., 2021, 8453076-
-
+
+
+
++
+
+
-
Zare Jeddi et al., 2021, 7404065-
+
+
-*
+
+
+
+
+
Zeng etal.,2015, 2851005-
+
+
+*
+
+
+
+
+*
Figure 3-36. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued)
Interactive figure and additional study details available on HAWC.
3-161
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.3.1.2.3 Findings from Children
Results for the studies that examined TC in children are presented in the Appendix (see PFOA
Appendix). Eleven medium confidence and four low confidence studies examined the association
between PFOA and TC in children. Of these, five studies examined the association between
prenatal PFOA exposure and TC in childhood {Spratlen, 2020, 5915332; Jensen, 2020, 6833719;
Manzano-Salgado, 2017, 4238509; Mora, 2018, 4239224; Tian, 2020, 7026251; Averina, 2021,
7410155} and ten examined the association between childhood PFOA exposure and concurrent
TC {Mora, 2018, 4239224; Jain, 2018, 5079656; Zeng, 2015, 2851005; Kang, 2018, 4937567;
Khalil, 2018, 4238547; Koshy, 2017, 4238478; Fassler, 2019, 6315820; Dong, 2019, 5080195;
Canova, 2021, 10176518; Blomberg, 2021, 8442228}. Positive associations between PFOA and
TC were reported in seven medium confidence studies {Zeng, 2015, 2851005; Spratlen, 2020,
5915332; Jensen, 2020, 6833719; Manzano-Salgado, 2017, 4238509; Mora, 2018, 4239224;
Canova, 2021, 10176518; Blomberg, 2021, 8442228}, but the direction of association sometimes
differed by age and sex {Jensen, 2020, 6833719; Manzano-Salgado, 2017, 4238509; Blomberg,
2021, 8442228}. Of all the positive associations observed in medium confidence studies, only
three were significant, including: all children (age 12-15 years) in Zeng (2015, 2851005), among
girls in mid-childhood in Mora (2017, 4239224), and children and adolescents in the highest
quartile of exposure from Canova (2021, 10176518).
In three out of four low confidence studies, PFOA was positively associated with TC {Khalil,
2018, 4238547; Koshy, 2017, 4238478; Fassler, 2019, 6315820}. However, residual
confounding by SES may have positively biased these findings. Taken together, these studies
suggest a positive association between PFOA and TC in children. However, the true association
between PFOA and TC remains uncertain given the heterogeneity by age and sex and the
imprecise findings in most medium confidence studies.
Seven medium confidence and five low confidence studies examined the association between
PFOA and LDL in children. Of these, five examined prenatal exposure {Jensen, 2020, 6833719;
Manzano-Salgado, 2017, 4238509; Mora, 2018, 4239224; Tian, 2020, 7026251; Papadopoulou,
2021, 9960593; Mora, 2018, 4239224} and eight examined childhood exposure {Mora, 2018,
4239224; Zeng, 2015, 2851005; Kang, 2018, 4937567; Khalil, 2018, 4238547; Koshy, 2017,
4238478; Canova, 2021, 10176518; Averina, 2021, 7410155; Dong, 2019, 5080195, adolescent
portion}. The medium studies generally reported small, positive associations between PFOA and
LDL, but most of the associations were not statistically significant (see PFOA Appendix)
{Jensen, 2020, 6833719; Mora, 2018, 4239224; Kang, 2018, 4937567}. In one medium study,
the association was inverse among 3-month old infants and 18-month old boys {Jensen, 2020,
6833719}.
One low confidence study {Canova, 2021, 10176518} on children and adolescents in a high-
exposure community located in Italy observed significantly increased LDL among adolescents
(beta per ln-unit increase in PFOA: 1.03; 95% CI: 0.39, 1.66). Most low confidence studies
reported a positive association between PFOA and LDL {Khalil, 2018, 4238547; Koshy, 2017,
4238478; Zeng, 2015, 2851005; Manzano-Salgado, 2017, 4238509; Canova, 2021, 10176518},
but residual confounding by SES {Khalil, 2018, 4238547; Koshy, 2017, 4238478} and the use of
non-fasting samples {Zeng, 2015, 2851005; Manzano-Salgado, 2017, 4238509; Canova, 2021,
10176518} were concerns in these studies. Overall, increases in LDL with increasing PFOA
were observed in children, though less consistently.
3-162
-------
DRAFT FOR PUBLIC COMMENT
March 2023
One high confidence, nine medium confidence and four low confidence studies examined the
association between PFOA and HDL in children. Of these, six examined prenatal exposure
{Jensen, 2020, 6833719; Manzano-Salgado, 2017, 4238509; Mora, 2018, 4239224;
Papadopoulou, 2021, 9960593; Blomberg, 2021, 8442228; Li, 2021, 7404102} and 12 examined
childhood exposure {Mora, 2018, 4239224; Jain, 2018, 5079656; Zeng, 2015, 2851005; Khalil,
2018, 4238547; Koshy, 2017, 4238478; Fassler, 2019, 6315820; Papadopoulou, 2021, 9960593;
Blomberg, 2021, 8442228; Li, 2021, 7404102; Canova, 2021, 10176518; Averina, 2021,
7410155; Dong, 2019, 5080195, adolescent portion}. Prenatal PFOA exposure was inversely
associated with HDL, but most associations were not statistically significant {Jensen, 2020,
6833719; Manzano-Salgado, 2017, 4238509; Mora, 2018, 4239224; Papadopoulou, 2021,
9960593; Blomberg, 2021, 8442228; Li, 2021, 7404102} (see PFOA Appendix). Sex-stratified
analyses showed that the inverse association occurred mainly in boys {Manzano-Salgado, 2017,
4238509; Mora, 2018, 4239224}. Results on childhood exposure were less consistent (see PFOA
Appendix). One medium study reported a statistically significant, positive association between
PFOA and HDL in mid-childhood {Mora, 2018, 4239224}, but another medium study reported
an inverse, though statistically non-significant association {Zeng, 2015, 2851005}. One medium
confidence study {Canova, 2021, 10176518} in a high-exposure community observed a
significant increase in HDL in children, but results were less consistent in adolescents. Most low
confidence studies reported a positive association between childhood PFOA exposure and HDL
{Khalil, 2018, 4238547; Koshy, 2017, 4238478; Fassler, 2019, 6315820}. In summary, PFOA
was not consistently associated with lower HDL in children. Effect modification by exposure
window may explain this inconsistency.
One high confidence, nine medium confidence and five low confidence studies examined the
association between PFOA and triglycerides in children. Of these, seven examined prenatal
exposure {Spratlen, 2020, 5915332; Jensen, 2020, 6833719; Manzano-Salgado, 2017, 4238509;
Mora, 2018, 4239224; Papadopoulou, 2021, 9960593; Li, 2021, 7404102; Tian, 2020, 7026251}
and 11 examined childhood exposure {Domazet, 2016, 3981435; Mora, 2018, 4239224; Zeng,
2015, 2851005; Kang, 2018, 4937567; Khalil, 2018, 4238547; Koshy, 2017, 4238478; Fassler,
2019, 6315820; Papadopoulou, 2021, 9960593; Li, 2021, 7404102; Canova, 2021, 10176518;
Averina, 2021, 7410155}. No association was observed in the only high confidence study {Li,
2021, 7404102}. PFOA was significantly associated with increased triglycerides in newborns in
one medium study {Spratlen, 2020, 5915332} (see PFOA Appendix). Some medium studies also
reported positive associations, but they were not statistically significant {Jensen, 2020, 6833719;
Mora, 2018, 4239224; Kang, 2018, 4937567}. Results from other medium confidence studies
were imprecise {Papadopoulou, 2021, 9960593; Li, 2021, 7404102}. In one medium study that
examined the association between PFOA and triglycerides longitudinally, PFOA at age 9 years
was associated with lower triglycerides at age 15 years and 21 years, while PFOA at age 15 years
was associated with higher triglycerides at age 21 years {Domazet, 2016, 3981435}. None of the
associations were statistically significant. In most low confidence studies, PFOA was positively
associated with triglycerides {Manzano-Salgado, 2017, 4238509; Zeng, 2015, 2851005; Khalil,
2018, 4238547; Koshy, 2017, 4238478}, but the use of non-fasting samples and residual
confounding by SES may have biased these results upwards. Overall, increased triglycerides
with increasing PFOA were observed in children, but results were less consistent and not always
statistically significant.
3-163
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In summary, the association between PFOA and serum lipids in children remains inconclusive.
For TC, LDL, and triglycerides, positive associations were generally observed, but few were
statistically significant. Differences in the direction of association by age or sex further
contributed to inconsistency in findings; it is difficult to determine if the differences were due to
effect modification or random error. For HDL, prenatal exposure appeared to be associated with
lower HDL, especially in boys, although childhood exposure was associated with higher HDL.
Few findings were statistically significant, however, suggesting caution in interpreting these
results.
3.4.3.1.2.4 Findings from Pregnant Women
Four medium confidence studies examined the association between PFOA and TC in pregnant
women {Matilla-Santander, 2017, 4238432; Skuladottir, 2015, 3749113; DallaZuanna, 2021,
7277682; Yang, 2020, 7021246} and two reported significantly positive associations between
PFOA and TC (see PFOA Appendix) {Matilla-Santander, 2017, 4238432; Skuladottir, 2015,
3749113}. One medium confidence study in a high-exposure community {Dalla Zuanna, 2021,
7277682} considered PFOA exposure concentrations across trimesters using a generalized
additive model (GAM). Authors reported significantly decreased TC with an increasingly inverse
trend across all sampled trimesters. Results were consistent for second and third trimester
samples in sensitivity analyses, but the direction of effect was positive for first trimester samples
(see PFOA Appendix). No association between PFOA and TC was observed in a Chinese study
of pregnant women {Yang, 2020, 7021246}. No association was found in the single low
confidence study {Varshavsky, 2021, 7410195} on total serum lipids after adjustment for
race/ethnicity, insurance type, and parity. These findings suggest a consistently positive
association between PFOA and TC in pregnant women.
Two studies examined PFOA and LDL in pregnant women {Dalla Zuanna, 2021, 7277682;
Yang, 2020, 7021246} and were considered low confidence due to lack of fasting blood samples
for LDL measurement. In a high-exposure community {DallaZuanna, 2021, 7277682}, a
decrease in LDL was reported with increasing PFOA concentrations when considering exposure
concentrations sampled across trimesters. In individual trimester sensitivity analyses, results
were consistently inverse for second and third trimester samples, including a significant finding
for the third trimester. However, non-significant positive associations were observed for first
trimester samples. No associations were observed for LDL in the other low confidence study, but
a significant decrease was reported for the LDL:HDL ratio (see PFOA Appendix). Two medium
confidence studies examined PFOA and HDL and reported statistically significant positive
associations between PFOA and HDL (see PFOA Appendix) {Starling, 2017, 3858473; Dalla
Zuanna, 2021, 7277682}. Dalla Zuanna (2021, 7277682) observed significant positive
associations when considering blood samples across all trimesters of pregnancy (GAM model).
The association was consistent, but no longer significant, when trimesters were modeled
individually. Another medium confidence study {Yang, 2020, 7021246} reported no association.
One medium confidence and three low confidence studies examined the association between
PFOA and triglycerides in pregnant women. The medium confidence study reported an inverse
association between PFOA and triglycerides, but the association was small and not statistically
significant {Starling, 2017, 3858473}. The low confidence studies each reported inverse
{Matilla-Santander, 2017, 4238432; Yang, 2020, 7021246} or positive associations {Kishi,
2015, 2850268} that were not statistically significant. Each study was limited by their use of
3-164
-------
DRAFT FOR PUBLIC COMMENT
March 2023
non-fasting blood samples. Kishi et al. (2015, 2850268) additionally examined the association
between PFOA and select fatty acids in serum. PFOA was not significantly associated with any
fatty acids, but the associations were generally positive except for arachidonic acid,
docosahexaenoic acid, and omega 3. Together, these studies suggest PFOA was not associated
with triglycerides or fatty acids in pregnancy.
In summary, the available evidence supports a positive association between PFOA and HDL in
pregnancy. The available evidence does not support a consistent, positive association between
PFOA and TC or triglycerides. Finally, the available evidence is too limited to determine the
association between PFOA and LDL in pregnant women.
3.4.3.1.2.5 Findings from the General Adult Population
Ten medium confidence and 13 low confidence studies examined PFOA and TC or
hypercholesterolemia in adults (Figure 3-34, Figure 3-35, Figure 3-36). All studies examined
cross-sectional associations {Dong, 2019, 5080195; Jain, 2019, 5080642; Liu, 2018, 4238514;
Liu, 2020, 6318644; Lin, 2019, 5187597; Donat-Vargas, 2019, 5080588; Wang, 2012, 2919184;
Convertino, 2018, 5080342; Chen, 2019, 5387400; Li, 2020, 6315681; He, 2018, 4238388;
Christensen, 2016, 3858533; Graber, 2019, 5080653; Sun, 2018, 4241053; Canova, 2020,
7021512; Fan, 2020, 7102734; Liu, 2018, 4238396; Lin, 2020, 6988476; Han, 2021, 7762348;
Cong, 2021, 8442223; Bjorke-Monsen, 2020, 7643487; Khalil, 2020, 7021479; Liu, 2021,
10176563} and two studies additionally examined the association between baseline PFOA and
changes in TC or incident hypercholesterolemia {Liu, 2020, 6318644; Lin, 2019, 5187597}.
Of the ten medium confidence studies, eight reported positive associations. In a population of
young adults aged 20 to 39 years in Veneto region, Italy, an area with water contamination by
PFAS, Canova et al. (2020, 7021512) reported statistically significant, positive associations with
TC. Canova et al. (2020, 7021512) also reported a concentration-response curve when PFOA
was categorized in quartiles or deciles, with a higher slope at higher PFOA concentrations, which
tended to flatten above around 20-30 ng/mL. Results from another medium confidence study
{Lin, 2020, 6988476} on older adults in a high-exposure community in Taiwan also reported
positive associations for TC, which was consistent across quartiles of PFOA exposure.
Four of the medium confidence studies used overlapping data from NHANES 2003-2014. All
four studies reported significant positive associations between PFOA and TC in adults {Dong,
2019, 5080195; Jain, 2019, 5080642; Liu, 2018, 4238514; Fan, 2020, 7102734} (see PFOA
Appendix). Stratified analyses in Jain et al. (2019, 5080642) suggested that the positive
association occurred mainly in obese men. A significantly positive association between PFOA
and TC also was observed at baseline in the DPPOS {Lin, 2019, 5187597}. This study reported
positive associations between PFOA and prevalent, as well as incident, hypercholesterolemia.
However, the HR for incident hypercholesterolemia was relatively small and not statistically
significant (HR = 1.06, 95% CI: 0.94, 1.19). In contrast to these findings, Liu et al. (2020,
6318644) reported no association between PFOA and TC. Further, Donat-Vargas et al. (2019,
5080588) reported generally inverse associations between PFOA and TC, regardless of whether
PFOA was measured concurrently or averaged between baseline and follow-up. It is noteworthy
that all participants in Lin et al. (2019 5187597) were prediabetic, all participants in Liu et al.
(2020, 6318644) were obese and enrolled in a weight loss trial, and all participants in Donat-
3-165
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Vargas et al. (2019, 5080588) were free of diabetes for at least 10 years of follow-up. It is
unclear if differences in participants' health status explained the studies' conflicting findings.
In low confidence studies, positive associations between PFOA and TC or hypercholesterolemia
were reported in nine of thirteen studies {Chen, 2019, 5387400; Cong, 2021, 8442223; Khalil,
2020, 7021479; Li, 2020, 6315681; He, 2018, 4238388; Christensen, 2016, 3858533; Graber,
2019, 5080653; Sun, 2018, 4241053; Liu, 2018, 4238396}. However, oversampling of persons
with potentially high PFOA exposure and health problems was a concern in three of these studies
{Li, 2020, 6315681; Christensen, 2016, 3858533; Graber, 2019, 5080653}. Selection bias
concerns, including lack of consideration of lipid-lowering medication and convenience
sampling, were issues in two of the studies {Cong, 2021, 8442223; Khalil, 2020, 7021479}.
Further, He et al. (2018, 4238388) used similar data as the four medium NHANES studies and
thus added little information.
Contrary to these findings, in one low confidence study, participants treated with extremely high
levels of ammonium perfluorooctanoate (APFO) in an open-label, nonrandomized, phase 1 trial,
were found to have reduced levels of TC with increasing plasma PFOA concentrations
{Convertino, 2018, 5080342}. This study differed from the other studies in several ways. First,
all participants were solid-tumor cancer patients who failed standard therapy and may have
distinct metabolic profiles compared to the general population. Second, participants ingested
high dose levels of APFO rather than being exposed to PFOA. Third, participants' plasma PFOA
concentrations were several orders of magnitude higher than those reported in the general
population. Participant serum concentrations were of similar magnitude as serum concentrations
resulting in decreased TC serum in rodent studies (see Section 3.4.3.2). It is unclear if these
factors explained the inverse association between PFOA and TC.
Considering medium and low confidence studies together, increased TC with increasing PFOA
was observed in adults. Some inconsistencies in the direction of association across studies were
found. Further studies are needed to determine if these inconsistencies reflect effect modification
by subject characteristics or PFOA dose levels.
3-166
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Confidence Exposure Study
Rating Reference Matrix Design Exposure Lewis Sub papulation Comparison EE
Effect Estimate
0-9 1.0 1.1 1.2 U 1.4 1.5 14 1.7 1.8
Medium CanovaetaL, Serum Cross- Median-35.8 rgiVnL OR (For G2 vs. Q1] 1.2
confidence 2020 sectional <25th 75th
percentile: 136 78 8
npVnL)
OR (for 03 vs. Q1] 1.25
OR [Fir 04 vs. 01] 1.46
Females OR (fix Q2 vs. Q1] 1.12
OR (Tor 03 vs. 01] 1.19
OR (fix 04 vs. 01] 1.33
Mdes OR (fix G2 vs. 01] 1.27
OR (fix 03 vs. 01] 1.22
OR (for 04 vs. 01] 1.43
Median=2.S8 ugiL _ _ , .
Low GraberetaL, „ Cross (25th 75th ° „
confidence 2019 01 sectional percentile-1.944 69 ~
upl) PFOA|
0-9 1.0 1.1 1.2 U 1.4 1.5 1J 1.7 1.8
Figure 3-37. Odds of High Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on Tableau.
3-167
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Figure 3-38. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on Tableau.
3-168
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Cohort Bascsnc median- 2-9 (ng'mll
(25th 75U*i (Xf tee rate :224 2 ra'm')
FofouMjp median. 2.7 (ng'irtl
(25ttv75Ui paiwlrtfa: 1.9 3.6
fof lerali! 2 ws icrtile 1
Figure 3-39. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on Tableau.
3-169
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Effect Estimate
Sub population Ccnrparix-
Fisher etal. 2013 Plasma
Cases: mrxi -' 0.06 ng'mL (25 76V!
percentile: 8.75 ! 7.06 ng'mL);
Contros med-11.40 nginiL $25-758!
percentile. 9.20-17.40 ng'mL)
a: 2.3 5.3; SO 1.9
at.. 2020 Serum
Median-8,6 T19WL (2£3h-7E
iriian: 3.9 ugil (range: 0.1-37.3 20 It
Regression Coefficient
PFOA)
Regression Coefficient
toglO np'mL
sase m PFOA)
Regression Cccfficie
et 1-kig10 urtfi chn
PFOA)
Regression Coefficient
cr * lag 10 unit change t
PFOA)
Regression Coefficient
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Antvrerp Mean (SD) = 1.03
Medium elal- Serum Cohort ppm (1-09X Decalijr^l.SO
confidence ppm (1_S9>
Regression Coefficient
PFOA)
Regression Coefficient
(per ppm ncrease
PFOA)
Regression Coefficie
(per ppm ncrease
PFOA)
Steenland et Serum Cress -
Regression Coefficienl
Regression Coefficient
Regression Coefficient
' er 1-ln n^VnL ricrease 0.01
PFOA)
Regression Coefficient
far PTOA Decile 10 w 0.05
Decie 1
0.0 0-5 1.0
Effect Estimate
IjS 2J0 U) 15 4.0 4.5 SO
OA 0-5 IjO U 2jQ 2-5 U> 3l5 4.0 4^ 5.0 5l5
Figure 3-41. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on Tableau.
Six medium confidence studies examined PFOA and LDL in adults, and all reported positive
associations (Figure 3-34, Figure 3-35, Figure 3-36). Higher PFOA was significantly associated
with higher LDL at baseline in the DPPOS {Lin, 2019, 5187597} (see PFOA Appendix). This
study also reported statistically significant, positive associations between PFOA and cholesterol
in non-HDL and VLDL, which are lipoprotein fractions related to LDL and associated with
increased cardiovascular risks {Lin, 2019, 5187597}. A positive association was observed in a
cross-sectional analysis of cases and controls in a study on type 2 diabetes {Han, 2021,
7762348}. Positive associations between PFOA and LDL were also reported in the four
NHANES studies {Dong, 2019, 5080195; Jain, 2019, 5080642; Liu, 2018, 4238514; Fan, 2020,
7102734}, but statistical significance was observed in obese men only {Jain, 2019, 5080642}
and in participants from NHANES cycle 2011-2012 {Dong, 2019, 5080195; Fan, 2020,
7102734}. Liu et al. (2020, 6318644) reported that PFOA was positively associated with
cholesterol and apolipoprotein C-III (ApoC-III) in combined fractions of intermediate-density
(IDL) and LDL that contained ApoC-III; the association with ApoC-III was statistically
significant. IDL and LDL containing ApoC-III and ApoC-III itself are strongly associated with
increased cardiovascular risks. Thus, the positive associations with cholesterol and ApoC-III in
ApoC-Ill-containing fractions of IDL and LDL were consistent with the positive associations
reported for LDL.
3-171
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Consistent with these findings, nine of the thirteen low confidence studies report positive
associations between PFOA and LDL {Lin, 2020, 6315756; Chen, 2019, 5387400; Li, 2020,
6315681; He, 2018, 4238388; Canova, 2020, 7021512; Liu, 2018, 4238396; Cong, 2021,
8442223; Khalil, 2020, 7021479; Lin, 2020, 6988476; Liu, 2021, 10176563}. Altogether, the
available evidence supports a relatively consistent positive association between PFOA and LDL
in adults, especially those who are obese or prediabetic. Associations with other lipoprotein
cholesterol known to increase cardiovascular risks were also positive, which increased
confidence in the findings for LDL.
Eleven medium confidence and thirteen low confidence studies examined PFOA and HDL or
clinically defined low HDL in adults (Figure 3-34, Figure 3-35, Figure 3-36). All studies
examined cross-sectional associations {Dong, 2019, 5080195; Jain, 2019, 5080642; Christensen,
2019, 5080398; Fan, 2020, 7102734; Liu, 2018, 4238514; Liu, 2020, 6318644; Lin, 2019,
5187597; Wang, 2012, 2919184; Convertino, 2018, 5080342; Lin, 2020, 6315756; Chen, 2019,
5387400; Li, 2020, 6315681; He, 2018, 4238388; Yang, 2018, 4238462; Canova, 2020,
7021512; Liu, 2018, 4238396; Lin, 2020, 6988476; Han, 2021, 7762348; Jeddi, 2021, 7404065;
Cong, 2021, 8442223; Khalil, 2020, 7021479; Liu, 2021, 10176563; Bjorke-Monsen, 2020,
7643487; Yu, 2021, 8453076}. Two studies also examined the association between baseline
PFOA and changes in HDL {Liu, 2020, 6318644; Liu, 2018, 4238396}. In a population of young
adults aged 20 to 39 years in the Veneto region, Italy, an area with water contamination by
PFAS, Canova et al. (2020, 7021512) reported statistically significant, positive associations with
HDL. Canova et al. (2020, 7021512) also reported a concentration-response curve when PFOA
was categorized in deciles. PFOA was inversely associated with HDL at baseline in the DPPOS,
but the association was not statistically significant {Lin, 2019, 5187597} (see PFOA Appendix).
Four studies used overlapping data from NHANES 2003-2014 and reported associations with
HDL that were sometimes positive {Liu, 2018, 4238514; Christensen, 2019, 5080398; Fan,
2020, 7102734} and sometimes inverse {Dong, 2019, 5080195}. The direction of association
differed by survey cycles. Few associations in this set of NHANES analyses were statistically
significant. In an additional medium confidence study, PFOA was not associated with HDL at
baseline or changes in HDL over two years {Liu, 2020, 6318644}. Similarly, low confidence
studies also reported a mix of positive {Lin, 2020, 6315756; Li, 2020, 6315681; He, 2018,
4238388; Yang, 2018, 4238462; Liu, 2018, 4238396} associations with changes in HDL in the
6-24 months of the study), inverse {Chen 2019, 5387400; Liu 2018, 4238396} associations with
concurrent HDL or changes in HDL in the first 6 months of the study {Ye, 2020, 6988486,
positive finding for reduced HDL}, or essentially null {Wang, 2012, 2919184; Convertino, 2018,
5080342; Liu, 2021, 10176563; Khalil, 2020, 7021479; Cong, 2021, 8442223; Bjorke-Monsen,
2020, 7643487} associations, with few being statistically significant. Given the inconsistent
findings in both medium and low confidence studies, the available evidence suggests PFOA is
not associated with HDL in adults.
Nine medium confidence and sixteen low confidence studies examined the association between
PFOA and triglycerides or hypertriglyceridemia. All studies examined the cross-sectional
association {Jain, 2019, 5080642; Christensen, 2019, 5080398; Liu, 2018, 4238514; Liu, 2020,
6318644; Lin, 2019, 5187597; Donat-Vargas, 2019, 5080588; Wang, 2012, 2919184;
Convertino, 2018, 5080342; Lin, 2013, 2850967; Lin, 2020, 6315756; Chen, 2019, 5387400; Li,
2020, 6315681; He, 2018, 4238388; Yang, 2018, 4238462; Sun, 2018, 4241053; Canova, 2020,
7021512; Fan, 2020, 7102734; Liu, 2018, 4238396; Lin, 2020, 6988476; Han, 2021, 7762348;
3-172
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Zare Jeddi, 2021, 7404065; Cong, 2021, 8442223; Khalil, 2020, 7021479; Liu, 2021, 10176563;
Ye, 2021, 6988486}; three studies additionally examined the association between baseline PFOA
and changes in triglycerides or incident hypertriglyceridemia {Liu, 2020, 6318644; Lin, 2019,
5187597; Liu, 2018, 4238396}. Higher PFOA was significantly associated with higher levels of
triglycerides in the DPPOS {Lin, 2019, 5187597} (see PFOA Appendix). This study also
reported that PFOA was significantly associated with higher odds of hypertriglyceridemia at
baseline and higher incidence of hypertriglyceridemia prospectively {Lin, 2019, 5187597}.
Similarly, PFOA was associated with slightly higher levels of triglycerides in Liu et al. (2020,
6318644). The association was stronger and statistically significant for triglycerides in the apoC-
III-containing combined fractions of IDL and LDL and apoC-III-negative HDL {Liu, 2020,
6318644}. In contrast, the four medium studies using overlapping data from NHANES 2005-
2014 reported positive {Jain, 2019, 5080642; Christensen, 2019, 5080398} or inverse
associations {Jain, 2019, 5080642; Liu, 2018, 4238514; Fan, 2020, 7102734} between PFOA
and triglycerides/hypertriglyceridemia. The direction of association appeared to differ by survey
cycle, sex, and obesity status. No associations in these NHANES analyses were statistically
significant. In an additional medium confidence study, PFOA was inversely associated with
triglycerides, regardless of whether PFOA was measured concurrently or averaged between
baseline and follow-up {Donat-Vargas, 2019, 5080588}. All participants in this study were free
of diabetes for over 10 years, as opposed to the obese or prediabetic adults in Liu et al. (2020,
6318644) and Lin et al. (2019, 5187597). It is unclear if participants' different health status
explained differences in the findings across medium studies.
In low confidence studies, a mix of positive {Khalil, 2020, 7021479; Liu, 2021, 10176563; Ye,
2021, 6988486; Lin, 2020, 6315756; Chen, 2019, 5387400; He, 2018, 4238388; Yang, 2018,
4238462; Sun, 2018, 4241053; Canova, 2020, 7021512; Lin, 2020, 6988476, in women; Liu,
2018, 4238396, association with concurrent triglycerides or changes in triglycerides in the first
6 months of the study}, inverse {Lin, 2013, 2850967; Li, 2020, 6315681; Lin, 2020, 6988476, in
men; Liu, 2018, 4238396, association with changes in triglycerides in the 6-24 months of the
study}, and essentially null {Wang, 2012, 2919184; Convertino, 2018, 5080342; Cong, 2021,
8442223} associations with triglycerides or hypertriglyceridemia were reported. Some
associations were statistically significant. Overall, the available evidence suggests that PFOA
was associated with elevated triglycerides in some adults. Whether PFOA increases triglycerides
in all adults is unclear given inconsistency in reported associations.
In summary, in the general adult population, a relatively consistent, positive association was
observed between PFOA and LDL or TC. Increased triglycerides with increasing PFOA
exposure were also observed, but less consistently. HDL was not associated with PFOA.
3.4.3.1.2.6 Findings from Occupational Studies
Workers are usually exposed to higher levels of PFOA, in a more regular manner (sometimes
daily), and potentially for a longer duration than adults in the general population. At the same
time, according to the "healthy worker effect," workers tend to be healthier than non-workers,
which may lead to reduced susceptibility to toxic agents {Shah, 2009, 9570930}. Because of
these potential differences in exposure characteristics and host susceptibility, occupational
studies are summarized separately from studies among adults in the general population.
3-173
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Three low confidence studies examined the association between PFOA and TC or
hypercholesterolemia in workers. Two of these studies examined the cross-sectional association
between PFOA and TC in fluorochemical plant workers or firefighters exposed to aqueous film
forming foam (AFFF) {Wang, 2012, 2919184; Rotander, 2015, 3859842}. One investigated the
association between baseline PFOA and changes in TC over the course of a fluorochemical plant
demolition project {Olsen, 2012, 2919185}. The cross-sectional studies reported positive
{Wang, 2012, 2919184} or inverse {Rotander, 2015, 3859842} associations between PFOA and
TC; neither association was statistically significant. Olsen et al. (2012, 2919185) reported that
over the course of the demolition project, changes in PFOA were inversely associated with
changes in TC; this association was not statistically significant {Olsen, 2012, 2919185}. Taken
together, these studies suggest no association between PFOA and TC in workers.
Two studies examined PFOA and LDL in workers. One study examined PFOA and non-HDL, of
which LDL is a major component. All studies were considered low confidence. The two studies
on LDL reported positive {Wang, 2012, 2919184} or inverse {Rotander, 2015, 3859842}
association between PFOA and concurrent LDL; neither association was statistically significant.
The study examining non-HDL reported that changes in PFOA during the fluorochemical plant
demolition project were inversely associated with changes in non-HDL, but the association was
not statistically significant {Olsen, 2012, 2919185}. Overall, these studies suggest no association
between PFOA and LDL in workers.
The studies that examined LDL or non-HDL also examined the association between PFOA and
HDL {Wang, 2012, 2919184; Rotander, 2015, 3859842; Olsen, 2012, 2919185}. The two cross-
sectional studies in this set of studies reported inverse association between PFOA and HDL,
including a statistically significant finding in Wang (2012, 2919184) {Wang, 2012, 2919184;
Rotander, 2015, 3859842}. Contrary to these findings, Olsen et al. (2012, 2919185) reported that
changes in PFOA over the demolition project were positively associated with changes in HDL
{Olsen, 2012, 2919185}. This association was not statistically significant. When changes in TC
to HDL ratio were examined as an outcome, however, a statistically significant, inverse
association was observed. This suggests that increasing PFOA exposure was associated with
decreases in TC/HDL over time, potentially partly due to a positive association between changes
in PFOA and changes in HDL. Together, the occupational studies reported a consistently inverse
association between PFOA and concurrent HDL, but this cross-sectional association was not
coherent with longitudinal findings.
Two low confidence cross-sectional studies examined PFOA and triglycerides in workers and
reported inverse associations between PFOA and triglycerides {Wang, 2012, 2919184; Rotander,
2015, 3859842}. Neither association was statistically significant.
In summary, among workers, the available evidence suggests no association between PFOA and
TC or LDL. Inverse, cross-sectional associations between PFOA and HDL and triglycerides
were found, but these associations were small, often not statistically significant, and were not
coherent with longitudinal findings. Overall, the associations between PFOA and serum lipids
among workers are different than those in the general adult population. It is unclear if well-
known biases in occupational studies such as "healthy worker effect" may have attenuated the
association between PFOA and an unfavorable serum lipid profile. More higher quality
occupational studies are needed to improve hazard identification among workers.
3-174
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.3.2 Animal Evidence Study Quality Evaluation and Synthesis
There are 2 studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and 7 studies from
recent systematic literature search and review efforts conducted after publication of the 2016
PFOA HESD that investigated the association between PFOA and cardiovascular effects in
animal models. Study quality evaluations for these 9 studies are shown in Figure 3-42.
£f.V° o^C° C*
Blake etal., 2020, 6305864'
Butenhoffetal., 2012, 2919192
Cope et al„ 2021, 10176465-
Guoetal., 2021, 9960713 ¦
Loveless et al., 2008, 988599
NTP, 2019, 5400977
NTP, 2020, 7330145-
Yan etal., 2014, 2850901 -\
van Esterik et al., 2015, 2850288-
+
+
++
B
++
++
+
++
NR
¦
++
++
B
++
+
+
++
++
B
++
++
-
+
NR
B
B
++ ++
++
++
+
+
NR
++
++
B
++
++
++
++
++
NR
++
++ ++
B
++
++
++
++
++
NR
++
++
++ ++
++
++
++
++
B
NR
++
B
++ ++
++
++
++
++
NR
++
B
++
++
B
B
5
Legend
| Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NR Not reported
Figure 3-42. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Cardiovascular Effects
Interactive figure and additional study details available on HAWC.
Cardiovascular effects following exposure to PFOA were minimal according to two chronic
studies with doses between 1.1 - 14.2 mg/kg/day {Butenhoff, 2012, 2919192; NTP, 2020,
7330145} and one short-term 28-day study with doses between 0.312 - 5 mg/kg/day {NTP,
2019, 5400977}. No toxicologically-relevant changes were observed for heart weight
{Butenhoff, 2012, 2919192; NTP, 2019, 5400977; NTP, 2020, 7330145}, minimal changes were
observed for heart histopathology {Butenhoff, 2012, 2919192; NTP, 2019, 5400977; NTP, 2020,
3-175
-------
DRAFT FOR PUBLIC COMMENT
March 2023
7330145}, and no changes were observed for aorta histopathology {Butenhoff, 2012, 2919192;
NTP, 2019, 5400977} following exposure to PFOA in male and female Sprague-Dawley rats.
PFOA has been observed to cause perturbations in lipid homeostasis, which may have effects on
the cardiovascular system. Alterations in serum lipid levels have been observed in mice and rats
in subchronic, chronic, and developmental studies of oral exposure to PFOA (Figure 3-43).
Overall, studies have generally reported consistent decreases in serum lipids including TC,
triglycerides, LDL cholesterol, HDL cholesterol, and/or non-HDL cholesterol in rats {Martin,
2007, 758419; Loveless, 2008, 988599; Elcombe, 2010, 2850034; NTP, 2019, 5400977; NTP,
2020, 7330145} and mice {Loveless, 2008, 988599; De Witt, 2009, 1937261; Minata, 2010,
1937251; Yahia, 2010, 1332451; Yan, 2014, 2850901; Quist, 2015, 6570066; Blake, 2020,
6305864; Cope, 2021, 10176465}.
In a developmental study of female CD-I Po mice exposed to PFOA (0, 1, and 5 mg/kg/day) by
oral gavage from either GD 1.5-11.5 or GD 1.5-17.5, authors reported maximum decreases in
serum triglyceride levels of 58% and 66%, respectively, at the highest dose of 5 mg/kg/day. No
changes were observed for serum TC, HDL cholesterol, or LDL cholesterol {Blake, 2020,
6305864}. In a secondary developmental study of gestational PFOA exposure (0.1 and
1.0 mg/kg/day), female CD-I Pomice were exposed via gavage from GD 1.5 - 17.5 {Cope,
2021, 10176465}. Male and female Fi offspring were fed either a low-fat diet (LFD) or high fat
diet (HFD) at PND 22 and serum cholesterol markers were evaluated at PND 22 and at postnatal
week (PNW) 18. At PND 22, there was a significant reduction in serum triglycerides in males
and females and a significant reduction in LDL in males only but no effects in TC or HDL. At
PNW 18, LFD female mice exhibited non-significant decreases in TC, HDL, LDL, and
triglycerides. However, animals that were given a HFD no longer exhibited decreased levels of
TC, HDL, or triglycerides and developed significantly higher levels of LDL (1.0 mg/kg/day)
when compared to HFD control. Males fed the LFD exhibited non-significant increases in TC,
HDL, LDL, and triglycerides; however, this trend was lost when animals were fed the HFD.
Male BALB/c mice exposed to PFOA by gavage for 28 days had significant decreases in serum
TC and HDL levels at concentrations as low as 1.25 mg/kg/day {Yan, 2014, 2850901}. For
serum triglyceride levels, significant increases were observed at lower exposure concentrations
of PFOA (0.31 and 1.25 mg/kg/day) while significant decreases were seen following exposure to
higher PFOA concentrations (5 and 10 mg/kg/day); no changes were observed in serum LDL
cholesterol levels. In a study conducted by NTP, sex differences were observed in Sprague-
Dawley rats exposed to PFOA by gavage for 28 days {NTP, 2019, 5400977}. Males had
significantly decreased serum TC and triglyceride levels at exposure concentrations as low as
0.625 mg/kg/day. Female rats in the same study were exposed to 10-fold higher doses than their
male counterparts due to sex differences in PFOA excretion (see PFOA Appendix). Females had
significant increases in both serum TC and triglyceride levels at the two highest doses (50 and
100 mg/kg/day). In the available chronic study {NTP, 2020, 7330145}, Fi male and female
Sprague-Dawley rats were exposed during gestation and lactation (perinatal exposure with
postweaning exposure) or postweaning exposure only until animals were 19 weeks of age (e.g.,
16-week interim time point; see further study design details in Section 3.4.4.2.1.2). Serum TC
levels were significantly decreased only in males exposed during both the perinatal and
postweaning phases (at postweaning doses of approximately 1 and 4.6 mg/kg/day); serum
triglyceride levels were decreased in all exposure groups. Serum TC levels were significantly
3-176
-------
DRAFT FOR PUBLIC COMMENT
March 2023
decreased only in the mid-dose Fi females exposed during both perinatal and postweaning
phases; TG levels were not altered in Fi females.
Conclusions from these studies are met with limitations as the difference in serum lipid
composition between humans and commonly used rodent models may impact the relevance to
human exposures {Getz, 2012, 1065480; Oppi, 2019, 5926372}. It should noted that human-
population based PFOA exposure studies have consistently found that as PFOA exposure
increases both serum cholesterol and serum triglycerides also increase. Some rodent studies
{Yan, 2014, 2850901} exhibit a biphasic dose response where low exposure concentrations lead
to increased serum lipid levels while high exposure concentrations lead to decreased serum lipid
levels. This has called in the validity of using rodent models to predict human lipid outcomes.
The relatively high exposure and PFOA serum concentrations that produce these inverse effects
are generally beyond the scope of human relevance, though there is some evidence in humans
that similarly high serum PFOA serum concentrations result in decreased serum total cholesterol
(e.g., Convertino et al. (2018, 5080342)). This suggests that rodent models may be utilized
accurately if the tested doses are within human health relevant exposure scenarios. Additionally,
food consumption and food type may confound these results {Cope, 2021, 10176465;
Schlezinger, 2020, 6833593; Fragki, 2021, 8442211}, as diet is a major source of lipids, yet
studies do not consistently report a fasting period before serum collection and laboratory diets
contain a lower fat content compared to typical Westernized human diets. More research is
needed to understand the influence of diet on the response of serum cholesterol levels in rodents
treated with PFOA.
3-177
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint Study Name Study Design Observation Time Animal Description
High Density Lipoprotein (HDL) Blake et al.; 2020, 6305864 developmental (GD1.5-11.5) GD11.5 P0 Mouse, CD-1 (' , N=5)
developmental (GD1.5-17.5) GD17.5 P0 Mouse, CD-1 (", N=4-6)
Copeetal., 2021, 10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse, CD-1 N=8)
F1 Mouse. CD-1 (V- N=7)
Yan et al., 2014, 2850901 short-term (28d) 28d Mouse, BALB/c (o, N=6)
Loveless etal.. 2008. 988599 short-term (29d) 29d Mouse, Cr1;CD-1(lCR)BR N=20)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) (>5, N=10)
Non-HDL Cholesterol Loveless etal,. 2008.988599 short-term (29d) 29d Mouse. Cr1;CD-1(ICR)BR N=20)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
Low Density Lipoprotein (LDL) Blake etal.. 2020, 6305864 developmental (GD1.5-11.5) GD11.5 P0 Mouse, CD-1 (~, N=5)
developmental (GD1.5-17.5) GD17.5 P0 Mouse, CD-1 (• . N=4-5)
Cope et al., 2021,10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 (o, N=8)
F1 Mouse, CD-1 ($. N=7)
Yan et al., 2014, 2850901 short-term (28d) 28d Mouse. BALB/c ( '. N=6)
Total Cholesterol Blake et al., 2020,6305864 developmental (GD1.5-11.5) GD11.5 P0 Mouse, CD-1 (' , N=5)
developmental (GD1.5-17.5) GD17.5 PO Mouse, CD-1 ('\ N=4-6)
Cope et al., 2021, 10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse, CD-1 (<•'. N=8)
F1 Mouse. CD-1 (V- N=7)
Yan etal., 2014, 2850901 short-term (28d) 28d Mouse, BALB/c (o, N=6)
Loveless et al.. 2008.988599 short-term (29d) 29d Mouse. Cr1:CD-1(lCR)BR N=20)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) (o, N=10)
NTP, 2019.5400977 short-term (28d) 29d Rat, Sprague-Dawley N=10)
Rat, Sprague-Dawley (", N=9-10)
NTP, 2020,7330145 chronic (GD6-PNW21) 16wk F1 Rat: Sprague-Dawley N=10)
chronic (GD6-PNW107) 16wk F1 Rat. Sprague-Dawley {" . N=10)
F1 Rat. Sprague-Dawley { '•, N=10)
chronic (PND21-PNW21) 16wk F1 Rat. Sprague-Dawley (•. N=10)
chronic (PND21-PNW107) 16wk F1 Rat, Sprague-Dawley N=10)
F1 Rat. Sprague-Dawley {i, N=10)
Triglycerides Blake etal.. 2020,6305864 developmental (GD1.5-11.5) GD11.5 P0 Mouse, CD-1 (^. N=5)
developmental (GD1.5-17.5) GD17.5 P0 Mouse, CD-1 (^. N=4-6)
Cope et al.. 2021,10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 N=8)
F1 Mouse, CD-1 (V. N=7)
Yan et al., 2014,2850901 short-term (28d) 28d Mouse. BALB/c N=6)
Loveless etal., 2008, 988599 short-term (29d) 29d Mouse, Cri:CD-1(ICR)BR N=20)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
NTP, 2019,5400977 short-term (28d) 29d Rat, Sprague-Dawley (o, N=10)
Rat, Sprague-Dawley U. N=9-10)
NTP, 2020.7330145 chronic (GD6-PNW21) 16wk F1 Rat. Sprague-Dawley N=10)
chronic (GD6-PNW107) 16wk F1 Rat Sprague-Dawley { /, N=10)
F1 Rat, Sprague-Dawley ( ;. N=10)
chronic (PND21-PNW21) 16wk F1 Rat: Sprague-Dawley { - , N=10)
cfironic (PND21-PNW107) 16wk F1 Rat. Sprague-Dawley (N=10)
F1 Rat. Sprague-Dawley ( ;, N=10)
0.01 Oil 1 10 100
Concentration (mg/kg/day)
PFOA Cardiovascular Effects - Serum Lipids
4) No significant change Significant increase Significant decrease
-V—V
-5?
-S?
-S? S? W
-S? V-
V V V V *
• » ¦ A—
V ¦ V
V-
—A A V W
— . s? V
V
V-V
V V V
5?-V
V V .
Figure 3-43. Serum Lipid Levels in Rodents Following Exposure to PFOA (logarithmic
scale)
PFOA concentration is presented in logarithmic scale to optimize tire spatial presentation of data. Interactive figure and additional
study details available on HAWC.
GD = gestation day; Po = parental generation; PNW = postnatal week; Fi = first generation; PND = postnatal day; d = day;
wk = week.
3.4.3.3 Mechanistic Evidence
Mechanistic evidence linking PFOA exposure to adverse cardiovascular outcomes is discussed in
Sections 3.1.1.1 and 3.4.1 of the 2016 PFOA HESD (U.S. EPA, 2016, 3603279}. There are 8
studies from recent systematic literature search and review efforts conducted after publication of
the 2016 PFOA HESD that investigated the mechanism s of action of PFOA that lead to
cardiovascular effects. A summary of these studies is shown in Figure 3-44.
3-178
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Mechanistic Pathway
Animal
Human
In Vitro
Grand Total
Angiogenic, Antiangiogenic, Vascular Tissue Remodeling
0
1
0
1
Athenogenesis And Clot Formation
0
1
3
4
Big Data, Non-Targeted Analysis
1
0
0
1
Cell Growth, Differentiation, Proliferation, Or Viability
1
1
1
3
Cell Signaling Or Signal Transduction
0
0
2
2
Fatty Acid Synthesis, Metabolism, Storage, Transport, Binding, B-Oxidation
1
0
0
1
Inflammation And Immune Response
0
0
1
1
Oxidative Stress
0
2
0
2
Grand Total
2
3
3
e
Figure 3-44. Summary of Mechanistic Studies of PFOA and Cardiovascular Effects
Interactive figure and additional study details available on Tableau.
3.4.3.3.1 Lipid transport and metabolism
Blood lipid levels are associated with risk factors for cardiovascular disease. Pouwer et al. (2019,
5080587) investigated how PFOA influences plasma cholesterol and triglyceride metabolism
using a transgenic mouse model of human-like lipoprotein metabolism (APOE*3-Leiden.CETP
mice, which express the human CETP gene), human plasma samples, and in silico predictions. In
the animal toxicological study, mice were fed a semisynthetic Western-type diet (0.25%
cholesterol (wt/wt), 1% corn oil (wt/wt), and 14% bovine fat (wt/wt)) with varying levels of
PFOA added (10, 300, or 30,000 ng/g/d). At the end of 4 or 6 weeks, mice were sacrificed and
levels of triglycerides, TC, free fatty acids (FFA), ALT, glycerol, VLDL, HDL, and CETP were
measured. The authors found that administration of PFOA at the 30,000 ng/g/d levels "reduced
plasma TG and TC levels by affecting VLDL-TG production through decreased apoB synthesis
and by increasing VLDL clearance." The authors also observed that PFOA at the highest dose
decreased hepatic VLDL production rate, increased plasma VLDL clearance through enhanced
LPL activity and affected gene expression of TG and cholesterol metabolism markers. Upon
further analysis. PPARa was determined to be the major transcription factor affecting gene
expression and fatty acid oxidation that regulates triglyceride and TC levels.
One study summarized in the 2016 PFOAHESD {U.S. EPA, 2016, 3603279} evaluated a subset
of 290 individuals in the C8 Health Project for evidence that PFOA exposure can influence the
transcript expression of genes involved in cholesterol metabolism, mobilization, or transport
{Fletcher, 2013, 2850968}. Inverse associations were found between PFOA levels and
expression of genes involved in cholesterol transport including Nuclear Receptor Subfamily 1
Group H Member 2 (NR1H2), Niemann-Pick disease type C (NPC1), and ATP Binding Cassette
Subfamily G Member 1 (ABCG1). When males and females were analyzed separately, PFOA
serum concentrations were negatively associated with expression of genes involved in
cholesterol transport in both males and females, although the genes themselves differed between
3-179
-------
DRAFT FOR PUBLIC COMMENT
March 2023
sexes (males: NPC1, ABCG1, PPARa; females: Nuclear Receptor Subfamily 1, Group H,
Member 1 (NCEH1)). For additional information on the disruption of lipid metabolism,
transport, and storage in the liver following PFOA exposure, please see Section 3.4.1.3.2.
3.4.3.3.2 Apoptosis and cell cycle regulation
To elucidate the mechanisms involved in PFOA-induced vascular tissue apoptosis and CIMT,
the levels of endothelial microparticles (CD62E, CD31+/CD42a-) and platelet microparticles
(CD62P, CD31+/CD42a+) were measured in the serum of adolescents and young adults in
another epidemiological study {Lin, 2016, 3981457}. The results showed that there was no
association between PFOA serum levels and markers of apoptosis, endothelial activation, or
platelet activation. This study also measured the relationship between oxidative stress and PFOA
by measuring levels of 8-hydroxydeoxyguanosine (8-OHdG) in the urine. Similar to the markers
of apoptosis, no association was found between PFOA and 8-OHdG. Another study by the same
researchers also found that there was no association between PFOA and oxidative/nitrative stress
markers 8-OHdG and 8-nitroguanine (8-N02Gua) in Taiwanese adults {Lin, 2020, 6315756}.
One study evaluated the potential for PFOA to affect cell-cycle regulation in the heart and other
tissues {Cui, 2019, 5080384}. Male mice were orally dosed with 5 mg/kg/day PFOA for
28 days, and microRNA-34 (miR-34), a marker of tissue damage, was measured in the heart at
the end of the exposure period. To further study the role of cardiovascular miR-34a under PFOA
treatment, the authors also dosed miR-34a-knockout and wild-type mice for 28 days. In the wild-
type mice, the expression of miR-34a in the heart was not significantly different in the treatment
group compared to the control group. There were also no detectible levels miR-34b or miR-34c
in the heart for either the treatment group or the control group.
3.4.3.3.3 Mechanisms of atherogenesis and clot formation
Four groups of researchers published studies on the mechanism of atherogenesis and clot
formation. The first two studies investigated how the structure of PFOA and other PFAS leads to
activation of the plasma kallikrein-kinin system (KKS) using in vitro and ex vivo activation
assays and in silico molecular docking analysis. KKS is a key component of plasma that plays a
role in regulation of inflammation, blood pressure, coagulation, and vascular permeability.
Activation of the plasma KKS can release the inflammatory peptide bradykinin (BK), which can
lead to dysfunction of vascular permeability. The cascade activation of KKS includes the
autoactivation of Hageman factor XII (FXII), cleavage of plasma prekallikrein (PPK), and
activation of high-molecular-weight kininogen (HK) {Liu, 2018, 4238499}. Results from the ex
vivo mouse plasma study by Liu et al. (2017, 4238579) revealed that the addition of PFOA (5
mM) at the highest dose binds with FXII in a structure dependent manner and triggers the
cascade to the rest of the system. Liu et al. (2018, 4238499) observed no activation of the KKS
cascade when mouse plasma was incubated with up to 500 |iM PFOA.
Bassler et al. (2019, 5080624) focused on several disease biomarkers, including plasminogen
activator ihhibitor-1 (PAI-1), an indicator of clot formation and that may lead to atherosclerosis.
Human serum was collected from 200 patients as part of the larger C8 Health Project and
analyzed for PFOA content. The authors found that there was no statistically significant
difference in PAI-1 concentration in association with high exposure to PFOA concentrations.
3-180
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The final study among the four groups of researchers, conducted by De Toni et al. (2020,
6316907), investigated the effect of PFOA on platelet function, a key factor in atherosclerosis.
Whole blood and peripheral blood samples were taken from healthy males that lived in low
exposure areas and incubated with 400 ng/mL of PFOA. After isolating erythrocytes, leukocytes,
and platelets and quantifying the amount of PFOA present, platelets were found to be the cell
target of PFOA accumulation. The authors then used the platelets in an in vitro system and
inoculated them with 400 ng/mL of PFOA and found that substantially more PFOA accumulated
in the membrane of platelets vs. the cytoplasm. Using molecular docking analysis, they were
able to target the specific binding sites of PFOA to phosphatidylcholine, a major platelet
phospholipid, suggesting that the accumulation of PFOA in the platelet may alter the activation
process of platelets by impairing membrane stability.
3.43.4 Evidence Integration
There is moderate evidence for an association between PFOA exposure and cardiovascular
effects in humans based on consistent positive associations with serum lipids, particularly LDL,
TC, and triglycerides. Additional evidence of positive associations with blood pressure and
hypertension in adult populations supported this classification. The lack of evidence of consistent
or precise effects for CVD or atherosclerotic changes raise uncertainty related to cardiovascular
health effects following PFOA exposure. The available data for CVD and atherosclerotic
changes was limited and addressed a wider range of outcomes, resulting in some residual
uncertainty for the association between PFOA exposure and these outcomes.
Based on this systematic review of 43 epidemiologic studies, the available evidence revealed
positive associations between PFOA exposure and TC, LDL, and triglycerides effects in some
human populations. For TC, the association was consistently positive in adults from the general
population, positive but less consistently so in children and pregnant women, and generally null
in workers. For LDL, the association was generally positive among adults, positive but less
consistently so in children, and generally null in workers. Data were not available for PFOA and
LDL in pregnant women. For triglycerides, positive, often non-significant associations were
observed in some adults and children, but not pregnant women and workers. Except for workers,
these results are consistent with findings from the 2016 HESD. Differences in findings from
occupational studies between the 2016 HESD and this review may be attributable to limitations
of occupational studies in this review. Similar to the 2016 HESD, the available evidence in this
review does not support an inverse association between PFOA and HDL in any populations. The
positive associations with TC are also supported by the recent meta-analysis restricted to 14
general population studies in adults {U.S. EPA, 2022, 10369698}. Similarly, a recent meta-
analysis including data from 11 studies reported consistent associations between serum PFOA or
a combination of several PFCs including PFOA and PFOS, and increased serum TC, LDL,
triglyceride levels in children and adults {Abdullah Soheimi, 2021, 9959584}.
The epidemiological studies identified since the 2016 assessments do not provide additional
clarity on the association between PFOA and CVD. Most of the CVD evidence identified in this
review focused on blood pressure in the general adult population (13 studies). The findings from
a single high confidence study and five medium confidence studies conducted in the general
adult population did not provide consistent evidence for an association between PFOA and blood
pressure. The evidence for an association between PFOA and increased risk of hypertension
overall and in gender-stratified analysis was inconsistent. Evidence in children and adolescents
3-181
-------
DRAFT FOR PUBLIC COMMENT
March 2023
also is less consistent. Five studies in children and adolescents, and one study in pregnant women
suggest no associations with elevated blood pressure in these populations. Evidence for other
CVD-related outcomes across all study populations was more limited, and similarly inconsistent.
Consequently, the evidence for these CVD outcomes is broadly consistent with the conclusions
of the C8 Science Panel and in the 2016 PFOA assessment, which found no probable link
between PFOA exposure and multiple other conditions, including high blood pressure and CAD.
It is challenging to compare findings on CVD related mortality in the current assessment to the
prior assessment due to differences in how this outcome was defined. Findings from the prior
assessment were mixed, with one study reporting an increased risk of cerebrovascular disease
mortality observed in the highest PFOA exposure category among occupationally exposed
subjects. However, no association was reported with IHD mortality. The current evidence from a
single study indicated PFOA was not associated with an increased risk of mortality due to
cardiovascular causes, including hypertensive disease, IHD, stroke, and circulatory diseases.
Future analyses of cause-specific CVD mortality could help elucidate whether there is a
consistent association between PFOA and cerebrovascular disease mortality. No studies or
endpoints were considered for the derivation of PODs since findings for an association between
PFOA and CVD outcomes are mixed.
The animal evidence for an association between PFOA exposure and cardiovascular toxicity is
moderate based on effects on serum lipids observed in animal models in six high or medium
confidence studies. The most consistent results are for TC and triglycerides, although direction
of effect can vary by dose. The biological significance of the decrease in various serum lipid
levels observed in these animal models regardless of species, sex, or exposure paradigm is
unclear; however, these effects do indicate a disruption in lipid metabolism. No effects or
minimal alterations were noted for heart weight and histopathology in the heart and aorta.
The underlying mechanisms for the observed cardiovascular effects related to PFOA exposure
are likely related to changes in lipid metabolism, as described in detail in Section 3.4.1.3.
Specifically, alterations in lipid metabolism lead to alterations in serum levels of triglycerides
and cholesterol, as evidenced by in vivo in animal models. The events that precede and result in
the alterations in serum levels have been proposed as the following, based on experimental
evidence: (1) PFOA accumulation in liver activates nuclear receptors, including PPARa; (2)
expression of genes involved in lipid homeostasis and metabolism is altered by nuclear receptor
activation; (3) gene products (translated proteins) modify the lipid content of liver to favor
triglyceride accumulation and potentially cholesterol accumulation; (4) altered lipid content in
the liver leads to accumulation of lipid droplets, which can lead to the development of steatosis
and liver dysfunction. It should be noted that the results for PFOA-induced changes to serum
lipid levels contrast between rodents (generally decreased) and humans (generally increased).
Evidence is ultimately limited regarding a clear mechanism of alterations to serum lipid
homeostasis caused by PFOA exposure. In humans, as discussed in the 2016 PFOA HESD {U.S.
EPA, 2016, 3603279} data from the C8 Health Project indicated that PFOA exposure can
influence expression of genes involved in cholesterol metabolism, mobilization, or transport.
Specifically, an inverse association was found between PFOA levels and expression of genes
involved in cholesterol transport, with sex-specificity for some of the individual gene expression
changes. The authors of the study suggested that exposure to PFOA may promote a
hypercholesterolaemic environment. Results were inconsistent regarding effects of PFOA on
indicators or mechanisms related to atherosclerosis, including a lack of effect on an indicator of
3-182
-------
DRAFT FOR PUBLIC COMMENT
March 2023
clot formation in human serum samples, and dose-dependent effects on the plasma kallikrein-
kinin system in mouse plasma. A single study found that PFOA accumulates in platelets in
human blood samples exposed in vitro, which may alter the activation process of platelets,
although it was not directly evaluated. PFOA did not induce apoptosis or oxidative stress in
vascular tissue in humans, as evidenced in two studies that evaluated serum levels of endothelial
microparticles and platelet microparticles, and urinary 8-hydroxydeoxyguanosine (8-OHdG) in
relation to PFOA levels.
3.4.3.4.1 Evidence Integration Judgment
Overall, considering the available evidence from human, animal, and mechanistic studies, the
evidence indicates that PFOA exposure is likely to cause adverse cardiovascular effects,
specifically serum lipid effects, in humans under relevant exposure circumstances (Table 3-8).
The hazard judgment is driven primarily by consistent evidence of serum lipid responses from
epidemiological studies at median PFOA exposure levels representative of the NHANES
population (median = 3.7 ng/mL). The evidence in animals showed coherent results for
perturbations in lipid homeostasis in rodent models in developmental, subchronic, and chronic
studies following exposure to doses as low as 0.3 mg/kg/day PFOA. While there is some
evidence that PFOA exposure might also have the potential to affect blood pressure and other
cardiovascular responses in humans given relevant exposure circumstances, the human evidence
underlying this possibility is uncertain and without support from animal or mechanistic studies.
3-183
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 3-8. Evidence Profile Table for PFOA Cardiovascular Effects
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
Evidence from Studies of Exposed Humans (Section 3.4.3.1)
Serum lipids
I High confidence study
20 Medium confidence
studies
26 Low confidence
studies
II Mixed1 confidence
studies
Examination of serum
lipids included measures
of TC, LDL, HDL, and
TG. In studies of adults
from the general
population (23), there is
evidence of positive
associations with TC
(17/23), but there was
some inconsistency by sex
or health status. Positive
associations were also
observed for LDL (15/23),
and mostly positive, but
slightly mixed,
associations with TG.
Evidence from studies of
children (15) was mixed,
and observed associations
often failed to reach
significance. Findings
were mostly positive for
TC (8/15). Among studies
of pregnant women (6),
evidence indicated
positive associations with
HDL (2/6) but not other
serum lipid measures.
Evidence of inverse
associations with HDL
and TG was observed in
occupational populations,
• High and medium
confidence studies
• Consistent findings of
positive associations
with serum lipid
measures in adults
from the general
population
• Coherence of
observed associations
in adults from the
general population
with previous
evidence of serum
lipid effects
• Low confidence studies
»Inconsistent findings of
effect in children, likely
due to variations in
measured exposure
window, and
occupational populations
®©o
Moderate
Evidence for
cardiovascular effects is
based on numerous
medium confidence
studies reporting positive
associations with serum
lipids, such as TC, LDL,
and TG, in adults from the
general population.
Results from some studies
of children and pregnant
women also observed
positive effects for TC and
TG. Results from
occupational studies were
inconsistent with those
from other populations,
though this may be due to
deficiencies in the
occupational study
designs. High and medium
confidence studies of
adults reported positive
associations with blood
pressure and risk of
hypertension, though other
medium and low
confidence studies
0©O
Evidence Indicates (likely)
Primary basis and cross-
stream coherence:
Human evidence indicated
consistent evidence of
serum lipids response and
animal evidence showed
coherent results for
perturbations in lipid
homeostasis in rodent
models in developmental,
subchronic, and chronic
studies following exposure
to PFOA. While there is
some evidence that PFOA
exposure might also have
the potential to affect blood
pressure and other
cardiovascular responses in
humans given relevant
exposure circumstances, the
human evidence underlying
this possibility is uncertain
and without support from
animal or mechanistic
studies.
Human relevance and other
inferences:
No specific factors are
noted.
3-184
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
though the effects did not
reach significance, and it
is unclear if these
differences in observed
effect reflect true
differential impacts in this
subpopulation.
reported non-significant
associations. Observed
effects were inconsistent
for CVD and imprecise for
atherosclerotic changes
across all study
populations.
Blood pressure and
hypertension
3 High confidence
studies
14 Medium confidence
studies
6 Low confidence studies
Studies examining
changes in blood pressure,
including DBP and SBP,
and risk for hypertension
had mixed results with
limited statistical
significance. The majority
of studies in children (10)
did not find an association
with blood pressure and/or
hypertension (7/10),
though one study reported
positive associations with
risk of hypertension
(1/10), one reported
evidence of an increase of
SBP (1/10), another
reported evidence of
increased mean of SBP
and DBP (1/10). Studies
of adults in the general
population (16) observed
some positive associations
with continuous blood
pressure (6/16), though
results varied between
SBP and DBP, and with
risk of hypertension
• High and medium
confidence studies
• Consistent findings of
positive effects for
blood pressure
measures, including
hypertension, among
adults
• Consistent findings of
effects observed in
studies of children for
blood pressure
measures and
hypertension
• Low confidence studies
• Lmprecision of findings
• Inconsistent findings of
effects observed for SBP
and DBP across studies
in adults
3-185
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
(6/16). Other studies did
not report significant
associations. Hypertension
analyses provided
evidence of modification
by sex, with males having
higher risk in some
studies.
Cardiovascular disease
1 High confidence study
5 Medium confidence
studies
5 Low confidence studies
Measures of CVD
included CHD, PAD,
stroke, heart attack, and
MVD. Studies in adults
from the general
population (8) reported
mixed results. Positive
associations were reported
for odds of PAD and CHD
(1/8), odds of CVD (1/8),
and odds of heart attack
(1/8), while other studies
reported inverse effects for
CHD (1/8) and stroke
(1/8). Still, other studies
did not observe evidence
of CVD associations. One
study of a population of
workers (1/3) reported
significantly increased
odds of stroke in one
exposure group, but the
effect diminished when a
10-year lag was included
in analyses. Two
additional occupational
studies reported imprecise
»High and medium
confidence studies
»Low confidence studies
»Inconsistent findings for
CVD-related outcomes
> Lmprecision of findings
3-186
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
findings for other CVD or
CVD-related mortality
outcomes among workers.
Atherosclerotic changes
1 High confidence study
3 Medium confidence
studies
3 Low confidence studies
One study of children
reported increased
brachial artery
distensibility (1/3). No
significant associations
were observed for CIMT
among Taiwanese children
(2/3) or pulse wave
velocity among American
children (1/3). Studies of
adults (3) reported mixed
results for measures of
atherosclerotic changes.
Most studies did not report
associations that reached
significance, however, one
study reported decreased
left ventricular relative
wall thickness (1/3).
• High and medium
confidence studies
• Low confidence studies
• Lmprecision of findings
across children and
adult study populations
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.3.2)
Serum lipids
3 High confidence
studies
3 Medium confidence
studies
Significant decreases in
serum TC were observed
in 4/6 studies that
examined this endpoint,
regardless of species, sex,
or study design. In two
developmental studies, no
changes were observed in
mice. Similar decreases
were observed in serum
TG (6/6). In a
developmental study,
• High and medium
confidence studies
• Consistency of
findings across
species, sex, or study
design
• Dose-response
relationship observed
within multiple
studies
• Lncoherence of findings
in other cardiovascular
outcomes
• Biological significance
of the magnitude of effect
is unclear
®©o
Moderate
Evidence based on six
high or medium
confidence studies
observed that PFOA
affects serum lipids in
animal models. The most
consistent results are for
total cholesterol and
triglycerides, although
3-187
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
decreased serum TG were
observed in mice at PND
22 but not during
adulthood. In a short-term
exposure study, female
rats were given 10-fold
higher doses of PFOA
than males due to sex
differences in excretion,
and it was found that
serum TC and TG were
decreased in males but
increased in females.
Fewer studies examined
HDL and LDL, with
decreases found in HDL
(2/4). Two studies found
no changes in LDL, but
one developmental study
in mice observed
increased LDL in males at
PND 22 but no changes
during adulthood.
direction of effect can
vary by dose. The
biological significance of
the decrease in various
serum lipid levels
observed in these animal
models regardless of
species, sex, or exposure
paradigm is unclear;
however, these effects
indicate a disruption in
lipid metabolism. No
effects or minimal
alterations were noted for
heart weight and
histopathology in the heart
and aorta. However, many
of the studies identified
may not be adequate in
exposure duration to
assess potential toxicity to
the cardiovascular system.
Histopathology
2 High confidence
studies
1 Medium confidence
study
No changes in heart
histopathology were
reported in two studies.
One chronic study
reported decreased
incidence of chronic
myocarditis in female rats
in the mid-dose group
only. No changes in aorta
histopathology were noted
in two studies.
• High and medium
confidence studies
• Limited number of
studies examining
outcome
Evidence Integration
Summary Judgment
3-188
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration
Summary Judgment
Organ weight
2 High confidence
studies, 1 Medium
confidence study
No changes in absolute or
relative heart weights were
found in one short-term
study and one chronic
study in rats. One chronic
study in rats reported
decreased absolute heart
weights in males and
females, but those
reductions were found to
be related to reduced body
weights.
• High and medium
confidence studies
• Limited number of
studies examining
outcome
• Confounding variables
such as decreases in
body weights may limit
ability to interpret these
responses
Mechanistic Evidence and Supplemental Information (Section 3.4.3.3)
Summary of Key Findings, Interpretation, and Limitations
Evidence Stream
Judgment
Key findings and interpretation:
• Alterations in lipid metabolism results in alterations in serum levels ofTG and TC via:
o PFOA accumulation in liver activates nuclear receptors, including PPARa.
o Nuclear receptor activation alters the expression of genes involved in lipid homeostasis and
metabolism.
PPARa is a major transcription factor affecting expression of genes that regulate fatty acid oxidation and
triglyceride and total cholesterol levels.
Limitations:
• Only a single study demonstrating PFOA accumulation in platelets in vitro.
• Results are inconsistent and conflicting regarding effects on indicators or mechanisms related to
atherosclerosis, primarily related to clot formation.
Findings support
plausibility that
cardiovascular effects,
specifically changes to
serum TG and TC levels,
can occur through changes
in lipid metabolism related
- to PFOA exposure.
Notes: CHD = coronary heart disease; CIMT = carotid intima-media thickness; CVD = cardiovascular disease; DBP = diastolic blood pressure; HDL = high density lipoprotein;
LDL = low density lipoprotein; MVD = microvascular disease; PAD = peripheral arterial disease; PPARa = peroxisome proliferator-activated receptor alpha; SBP = systolic
blood pressure; TC = total cholesterol; TG = triglyceride.
a Mixed confidence studies had split confidence determinations for different serum lipid measures with some measures rated medium confidence and others rated low confidence.
3-189
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4 Developmental
EPA identified 100 epidemiological and 18 animal toxicological studies that investigated the
association between PFOA and developmental effects. Of the epidemiological studies, 30 were
classified as high confidence, 39 as medium confidence, 19 as low confidence, 5 as mixed (2
high/medium, 1 medium/low, 2 low/uninformative) confidence, and 7 were considered
uninformative (Section 3.4.4.1). Of the animal toxicological studies, 2 were classified as high
confidence, 11 as medium confidence, and 4 as low confidence, and 1 was considered mixed
(imedium/low) (Section 3.4.4.2). Studies have mixed confidence ratings if different endpoints
evaluated within the study were assigned different confidence ratings. Though low confidence
studies are considered qualitatively in this section, they were not considered quantitatively for
the dose-response assessment (Section 4).
3.4.4.1 Human Evidence Study Quality Evaluation and Synthesis
3.4.4.1.1 Introduction
This section describes studies of PFOA exposure and potential in utero and perinatal effects or
developmental delays, as well as effects attributable to developmental exposure. The latter
includes all studies where exposure is limited to gestation and/or early life up to 2 years of age.
Developmental endpoints can include gestational age, measures of fetal growth (e.g., birth
weight), and miscarriage, as well as infant/child development.
The 2016 PFOA HESD {U.S. EPA, 2016, 3603279} summarized epidemiological studies that
examined developmental effects in relation to PFOA exposure. There are 21 studies from the
2016 PFOA HESD {U.S. EPA, 2016, 3603279} that investigated the association between PFOA
and developmental effects. Study quality evaluations for these 21 studies are shown in Figure
3-45.
Studies included ones conducted both in the general population as well as in communities known
to have experienced high PFOA exposure (e.g., the C8 population in West Virginia and Ohio).
Results of 16 high or medium confidence epidemiological studies and five low confidence or
uninformative studies (see Section 2.1.3 for information about study quality evaluations)
discussed in the 2016 PFOA HESD are summarized below.
3-190
-------
DRAFT FOR PUBLIC COMMENT
March 2023
^>>5^s6e"
Andersen et al.,
Apelberg et al.,
Bae et al.,
Chen et al.,
Darrow et al.,
Darrow et al.,
Fei et al.,
Fei et al.,
Fei et al.,
2010, 1429893
2007,1290833
2015, 2850239-
2012, 1332466-
2013, 2850966-
2014, 2850274-
2007, 1005775-
2008, 1290822-
2008,2349574 -
Fei et al., 2010, 1430760-
Grice et al.,
Hamm et al.,
Maisonet et al.,
Monroy et al.,
Nolan et al.,
Nolan et al.,
Savitz et al.,
Savitz et al.,
Stein et al.,
Washino et al.,
2007,4930271
2010, 1290814
2012, 1332465
2008, 2349575 -\
2009,2349576
2010, 1290813
2012, 1276141
2012, 1424946
2009, 1290816-
2009, 1291133-
Whitworth et al., 2012, 2349577-
+
+
*
+
+
+
+
+
+
+
+
++
+
+
+
+
++
+
+
+
+
+
+
+
+
+
+
+
++
+
+
++
+
+
++
+
+
-
+
+
+
+
+
+
i +
++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
++
-
+
+
+
+
-
+
+ +*
-
+
+
+
¦
+
+
+
+
+
+
+
+
-
+
+
+
+
-
++
++
-
-
+
+
-
+
-
+
-
+
+
+
~
+
-
%
+
+
-
+
+
+
+
+
+
+
+
+
-
+
+
+
+
+
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
++
++
V/U
++
++
++
Legend
I Good (metric) or High confidence (overall)
+ | Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
I Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-45. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Developmental Effects Published before 2016 (References from 2016 PFOA
HESD)
Interactive figure and additional study details available on HAWC.
3-191
-------
DRAFT FOR PUBLIC COMMENT
March 2023
As noted in the 2016 HESD, several available studies measured fetal growth outcomes. Apelberg
et al. (2007, 1290833) found that birth weight was inversely associated with umbilical cord
PFOA concentration (change in birth weight per log unit increase: -104; 95% confidence
interval (CI): -213, -5; g) in a study of 293 infants born in Maryland in 2004-2005 (mean PFOA
concentration of 0.0016 |ig/mL). Maisonet et al. (2012, 1332465) evaluated fetal growth
outcomes in 395 singleton female births of participants in the Avon Longitudinal Study of
Parents and Children (ALSPAC) and found that increased maternal PFOA concentration (median
concentration of 0.0037 |ig/mL) was associated with lower birth weights (change in birth weight
per log unit increase: -34.2; 95% CI: -54.8, -13; g). A study of 252 pregnant women in Alberta,
Canada found no statistically significant association between birth weight and PFOA
concentration measured in maternal blood during the second trimester (mean concentration of
0.0021 (j,g/mL) {Hamm, 2010, 1290814}. In a prospective cohort study in Japan (2002-2005),
Washino et al. (2009, 1291133) found no association between PFOA concentration in maternal
blood during pregnancy (mean PFOA concentration of 0.0014 (j,g/mL) and birth weight. Chen et
al. (2012, 1332466) examined 429 mother-infant pairs from the Taiwan Birth Panel Study and
found no significant association between umbilical cord blood PFOA concentration (geometric
mean (GM) of 0.0018 (j,g/mL) and birth weight.
Some studies evaluated fetal growth parameters in the prospective Danish National Birth Cohort
(DNBC; 1996-2002) {Andersen, 2010, 1429893; Fei et al., 2007, 1005775; Fei, 2008,
2349574}. Maternal blood samples were taken in the first and second trimester. Fei et al. (2007,
1005775) found an inverse association between maternal PFOA concentration (blood samples
taken in the first and second trimester) and birth weight (change in birth weight per unit increase:
-8.7; 95% CI: -19.5, 2.1). Fei et al. (2008, 2349574) found an inverse association between
maternal PFOA levels and birth length and abdominal circumference in the DNBC. Change in
birth length per unit increase was 0.069 cm (95% CI: 0.024, 0.113) and change in abdominal
circumference per unit increase was 0.059 cm (95% CI: 0.012, 0.106). Andersen et al. (2010,
1429893) examined the association between maternal PFOA concentrations and birth weight,
birth length, and infant body mass index (BMI) and body weight at 5 and 12 months of age in
DNBC participants. They found a positive association between maternal PFOA concentration
and BMI measured at 5 and 12 months in boys, but not girls.
Some studies described in the 2016 PFOA HESD evaluated developmental outcomes in the C8
Health Project study population, which comprises a community known to have been subjected to
highPFAS exposure {Darrow, 2013, 2850966; Savitz, 2012, 1276141; Savitz, 2012, 1424946;
Stein, 2009, 1290816; Darrow, 2014, 2850274}. The C8 Health Project included pregnancies
within 5 years prior to exposure measurement, and many of the women may not have been
pregnant at the time of exposure measurement. As noted in the 2016 HESD, none of the studies
reported associations between PFOA and either birth weight or the risk of low birth weight.
Additionally, two studies {Nolan, 2009, 2349576; Nolan, 2010, 1290813} evaluated birth
weight, gestational age of infants, and frequencies of congenital anomalies in this community
based on whether participants were supplied with contaminated public drinking water (PFOA
concentrations were not measured in participants). The studies found no associations between
these developmental effects and water supply status. These two studies were rated low
confidence.
3-192
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4.1.2 Study Evaluation Considerations
There were multiple developmental outcome-specific considerations that informed domain-
specific ratings and overall study confidence. For the Confounding domain, downgrading of
studies occurred when key confounders of the fetal growth and PFAS relationship, such as
parity, were not considered. Some hemodynamic factors related to physiological changes during
pregnancy were also considered in this domain as potential confounders (e.g., GFR and blood
volume changes over the course of pregnancy) because these factors may be related to both
PFOA levels and the developmental effects examined here. More confidence was placed in the
epidemiologic studies that adjusted for GFR in their regression models or if they limited this
potential source of confounding by sampling PFAS levels earlier in pregnancy. An additional
source of uncertainty was the potential for confounding by other PFAS (and other co-occurring
contaminants). Although scientific consensus on how best to address PFAS co-exposures
remains elusive, this was considered in the study quality evaluations and as part of the overall
weight of evidence determination.
For the Exposure domain, all the available studies analyzed PFAS in serum or plasma using
standard methods. Given the estimated long half-life of PFOA in humans noted in Section
3.3.1.4.5, samples collected during all three trimesters, before birth or shortly after birth were
considered adequately representative of the most critical in utero exposures for fetal growth and
gestational duration measures. The postnatal anthropometric studies were evaluated with
consideration of fetal programming mechanisms (i.e., Barker hypothesis) where in utero
perturbations, such as poor nutrition, can lead to developmental effects such as fetal growth
restriction and ultimately adult-onset metabolic-related disorders and related complications (see
more on this topic in {De Boo, 2009, 6937194} and {Perng, 2016, 6814341}. There is some
evidence that birth weight (BWT) deficits can be followed by increased weight gain that may
occur especially among those with rapid growth catch-up periods during childhood {Perng,
2016, 6814341}. Therefore, the primary critical exposure window for measures of postnatal (and
early childhood) weight and height change is assumed to be in utero for study evaluation
purposes, and studies of this outcome were downgraded in the exposure domain if exposure data
were collected later during childhood or concurrently with outcome assessment (i.e., cross-
sectional analyses).
Studies were also downgraded for study sensitivity, for example, if they had limited exposure
contrasts and/or small sample sizes, since this can impact the ability of studies to detect
statistically significant associations that may be present (e.g., for sex-stratified results). In the
Outcome domain, specific considerations address validation and accuracy of specific endpoints
and adequacy of case ascertainment for some dichotomous (i.e., binary) outcomes. For example,
BWT measures have been shown to be quite accurate and precise, while other fetal and early
childhood anthropometric measures may result in more uncertainty. Mismeasurement and
incomplete case ascertainment can affect the accuracy of effect estimates by impacting both
precision and validity. For example, the spontaneous abortion studies were downgraded for
incomplete case ascertainment in the Outcome domain given that some pregnancy losses go
unrecognized early in pregnancy (e.g., before implantation). This incomplete ascertainment,
referred to as left truncation, can result in decreased study sensitivity and loss of precision.
Often, this type of error can result in bias towards the null if ascertainment of fetal loss is not
associated with PFOA exposures (i.e., non-differential). In some situations, differential loss is
possible and bias away from the null can manifest as an apparent protective effect. Fetal and
3-193
-------
DRAFT FOR PUBLIC COMMENT
March 2023
childhood growth restriction were examined using several endpoints including low BWT, small
for gestational age (SGA), ponderal index (i.e., BWT grams)/birth length (cm3) x 100),
abdominal and head circumference, as well as upper arm/thigh length, mean height/length, and
mean weight either at birth or later during childhood. The developmental effects synthesis is
largely focused on the higher quality endpoints (i.e., classified as good in the Outcome domain)
that were available in multiple studies to allow for an evaluation of consistency and other
considerations across studies. However, even when databases were more limited, such as for
spontaneous abortions, the evidence was evaluated for its ability to inform developmental
toxicity more broadly, even if available in only one study.
Overall, mean BWT and BWT-related measures are considered very accurate and were collected
predominately from medical records; therefore, more confidence was placed in these endpoints
in the Outcome domain judgments. Some of the adverse endpoints of interest examined here
included fetal growth restriction endpoints based on BWT such as mean BWT (or variations of
this endpoint such as standardized BWT z-scores), as well as binary measures such as SGA (e.g.,
lowest decile of BWT stratified by gestational age and other covariates) and low BWT (i.e.,
typically < 2500 grams; 5 pounds, 8 ounces) births. Sufficient details on the SGA percentile
definitions and stratification factors as well as sources of standardization for z-scores were
necessary to be classified as good for these endpoints in this domain. In contrast, other measures
of fetal growth that are subject to more measurement error (e.g., head circumference and body
length measures such as ponderal index) were given a rating of adequate {Shinwell, 2003,
6937192}. These sources of measurement error are expected to be non-differential with respect
to PFOA exposure status and, therefore, would not typically be a major concern for risk of bias
but could impact study sensitivity.
Gestational duration measures were presented as either continuous (i.e., per each gestational
week) or binary endpoints such as preterm birth (PTB, typically defined as gestational
age < 37 weeks). Although changes in mean gestational age may lack some sensitivity
(especially given the potential for measurement error), many of the studies were based on
ultrasound measures early in pregnancy, which should increase the accuracy of estimated
gestational age and the ability to detect associations that may be present. Any sources of error in
the classification of these endpoints would also be anticipated to be non-differential with respect
to PFOA exposure. While they could impact precision and study sensitivity, they were not
considered a major concern for risk of bias.
3.4.4.1.3 Study Inclusion
There are 79 epidemiological studies from recent systematic literature search and review efforts
conducted after publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} that
investigated the association between PFOA and developmental effects. Although every study is
included in the endpoint-specific study quality evaluation heat maps for comprehensiveness, six
developmental epidemiological studies identified in the literature search were excluded from this
synthesis due to study population overlap with other included studies (i.e., were considered
duplicative). The Li et al. (2017, 3981358) Guangzhou Birth Cohort Study overlaps with a more
recent study by Chu et al. (2020, 6315711). Four other studies {Kishi, 2015, 2850268;
Kobayashi, 2017, 3981430; Minatoya, 2017, 3981691; Kobayashi, 2022, 10176408} were also
not considered in this synthesis, because they provided overlapping data from the same
Hokkaido Study on Environment and Children's Health birth cohort as Kashino et al. (2020,
3-194
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6311632). For those studies with the same endpoints analyzed across different subsets from the
same cohort, such as mean BWT, the analysis with the largest sample size was used in forest
plots and tables (e.g., {Kashino, 2020, 6311632} for the Hokkaido birth cohort study). Although
the Kobayashi et al. (2017, 3981430) study included a unique endpoint called ponderal index,
this measure is more prone to measurement error and was not considered in any study given the
wealth of other fetal growth restriction data. Similarly, the Costa et al., (2019, 5388081) study
that examined a less accurate in utero growth estimate was not considered in lieu of their more
accurate birth outcomes measures reported in the same cohort {Manzano-Salgado, 2017,
4238465}. One study by Bae et al. {2015, 2850239} was the only study to examine sex ratio and
was not further considered here. In general, to best gauge consistency and magnitude of reported
associations, EPA largely focused on the most accurate and most prevalent measures within each
fetal growth endpoint. Three additional studies with overlapping cohorts were all included in the
synthesis, as they provided some unique data for different endpoints. For example, the Woods et
al. (2017, 4183148) publication on the Health Outcomes and Measures of the Environment
(HOME) cohort overlaps with Shoaff et al. (2018, 4619944) but the authors provided additional
mean BWT data. The mean BWT results for singleton and twin births from Bell et al. (2018,
5041287) are included in forest plots here, while the postnatal growth trajectory data in the same
UPSTATE KIDS cohort by Yeung et al. (2019, 5080619) are also included as they target
different developmental endpoints. The Bjerregaard-Olesen et al. (2019, 5083648) study from
the Aarhus birth cohort also overlaps with Bach et al. (2016, 3981534). The main effect results
are comparable for head circumference and birth length in both studies despite a smaller sample
size in the Aarhus birth cohort subset examined in Bjerregaard-Olesen et al. (2019, 5083648).
Given that additional sex-specific data are available in the Bjerregaard-Olesen et al. (2019,
5083648) study, the synthesis for head circumference and birth length are based on this subset
alone. Chen et al., (2021, 7263985) reported an implausibly large effect estimate for head
circumference. After correspondence with study authors, an error was identified, and the study
was not considered for head circumference.
Following exclusion of the seven studies above, 72 developmental epidemiological studies were
available for the synthesis. One study by Bae et al. (2015, 2850239) was the only study to
examine sex ratio and was not further considered here. Six additional studies {Alkhalawi, 2016,
3859818; Gundacker, 2021, 10176483; Jin, 2020, 6315720; Lee, 2013, 3859850; Lee, 2016,
3981528; Maekawa, 2017, 4238291} were considered uninformative due to critical deficiencies
in some risk of bias domains (e.g., confounding) or multiple domain deficiencies and are not
further examined here. Thus, 66 studies were included across various developmental endpoints
for further examination and synthesis. Forty-six of the 66 studies examined PFOA in relation to
fetal growth restriction measured by the following fetal growth restriction endpoints: SGA, low
BWT, head circumference, as well as mean and standardized BWT and birth length measures.
Twenty studies examined different measures of gestation duration, five examined fetal loss, four
examined birth defects, and 13 examined post-natal growth.
High and medium confidence studies were the focus of the evidence synthesis for endpoints with
numerous studies, though low confidence studies were still considered for consistency in the
direction of association (and details are provided in PFOA Appendix). For endpoints with fewer
studies, the evidence synthesis below included details on any low confidence studies available.
Studies considered uninformative were not considered further in the evidence synthesis.
3-195
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4.1.4 Growth Restriction: Fetal Growth
3.4.4.1.4.1 Birth Weight
Of the 43 studies examining different BWT measures in relation to PFOA exposures, 37
examined mean birth weight differences. Fifteen studies examined standardized BWT measures
(e.g., z-scores) with nine of these reporting results for mean and standardized BWT {Ashley-
Martin, 2017, 3981371; Bach, 2016, 3981534; Eick, 2020, 7102797; Gyllenhammar, 2018,
4238300; Meng, 2018, 4829851; Sagiv, 2018, 4238410; Wang, 2019, 5080598; Wikstrom, 2020,
6311677; Workman, 2019, 5387046}. Twenty-six of the 37 mean BWT were prospective birth
cohort studies, and the remaining eleven were cross-sectional analyses defined here as if
biomarker samples were collected at birth or post-partum {Bell, 2018, 5041287; Callan, 2016,
3858524; de Cock, 2016, 3045435; Gao, 2019, 5387135; Gyllenhammar, 2018, 4238300; Kwon,
2016, 3858531; Shi, 2017, 3827535; Wang, 2019, 5080598; Wu, 2012, 2919186; Xu, 2019,
5381338; Yao, 2021, 9960202}.
Eight of the 37 studies with data on the overall population relied on umbilical cord measures
{Cao, 2018, 5080197; de Cock, 2016, 3045435; Govarts, 2016, 3230364; Kwon, 2016, 3858531;
Shi, 2017, 3827535; Wang, 2019, 5080598; Workman, 2019, 5387046; Xu, 2019, 5381338}, and
one collected blood samples in infants 3 weeks following delivery {Gyllenhammar, 2018,
4238300}. Results from the Bell et al. (2018, 5041287) study were based on infant whole blood
taken from a heel stick and captured onto filter paper cards at 24 hours or more following
delivery, and one study used both maternal serum samples collected 1-2 days before delivery
and cord blood samples collected immediately after delivery {Gao, 2019, 5387135}. One of the
prospective birth cohort studies examined pre-conception maternal serum samples {Robledo,
2015, 2851197}. Twenty-four studies had maternal exposure measures that were sampled during
trimesters one {Ashley-Martin, 2017, 3981371; Bach, 2016, 3981534; Lind, 2017, 3858512;
Manzano-Salgado, 2017, 4238465; Sagiv, 2018, 4238410}, two {Buck Louis, 2018, 5016992;
Lauritzen, 2017, 3981410}, three {Callan, 2016, 3858524; Chu, 2020, 6315711; Kashino, 2020,
6311632; Luo, 2021, 9959610; Valvi, 2017, 3983872; Wang, 2016, 3858502; Wu, 2012,
2919186; Yao, 2021, 9960202}, or across multiple trimesters {Chang, 2022, 9959688; Chen,
2021, 7263985; Eick, 2020, 7102797; Hjermitslev, 2020, 5880849; Lenters, 2016, 5617416;
Marks, 2019, 5081319; Starling, 2017, 3858473; Wikstrom, 2020, 6311677; Woods , 2017,
4183148}. The study by Meng et al. (2018, 4829851) pooled exposure data from two study
populations, one which measured PFOA in umbilical cord blood and one which measured PFOA
in maternal blood samples collected in trimesters 1 and 2. For comparability with other studies of
mean BWT, only one biomarker measure was used (e.g., preferably maternal samples when
collected in conjunction with umbilical cord samples or maternal only when more than the parent
provided samples). In addition, other related publications (e.g., Gyllenhammar et al. (2017,
7323676)) or additional information or data provided by study authors were used.
Sixteen of the 37 studies reporting mean BWT changes in relation to PFOA in the overall
population were rated high in overall study confidence {Ashley-Martin, 2017, 3981371; Bach,
2016, 3981534; Bell, 2018, 5041287; Buck Louis, 2018, 5016992; Chu, 2020, 6315711; Eick,
2020, 7102797; Govarts, 2016, 3230364; Lauritzen, 2017, 3981410; Lind, 2017, 3858512; Luo,
2021, 9959610; Manzano-Salgado, 2017, 4238465; Sagiv, 2018, 4238410; Starling, 2017,
3858473; Valvi, 2017, 3983872; Wang, 2016, 3858502; Wikstrom, 2020, 6311677}, while 13
were rated medium {Chang, 2022, 9959688; Chen, 2021, 7263985; de Cock, 2016, 3045435;
3-196
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Gyllenhammar, 2018, 4238300; Hjermitslev, 2020, 5880849; Kashino, 2020, 6311632; Kwon,
2016, 3858531; Lenters, 2016, 5617416; Meng, 2018, 4829851; Robledo, 2015, 2851197; Wang,
2019, 5080598; Woods et al., 2017, 4183148; Yao, 2021, 9960202}, and eight were classified as
low {Callan, 2016, 3858524; Cao, 2018, 5080197; Gao, 2019, 5387135; Marks, 2019, 5081319;
Shi, 2017, 3827535; Workman, 2019, 5387046; Wu, 2012, 2919186; Xu, 2019, 5381338} as
shown in Figure 3-46, Figure 3-47, and Figure 3-48.
Of the 29 high or medium confidence studies highlighted in this synthesis, two had deficient
study sensitivity {Bell, 2018, 5041287; de Cock, 2016, 3045435}. Nine studies {Chen, 2021,
7263985; Lauritzen, 2017, 3981410; Lenters, 2016, 5617416; Robledo, 2015, 2851197; Starling,
2017, 3858473; Wang, 2016, 3858502; Wikstrom, 2020, 6311677; Woods, 2017, 4183148; Yao,
2021, 9960202} } were considered to have good study sensitivity, and eighteen studies {Ashley-
Martin, 2017, 3981371; Bach, 2016, 3981534; Buck Louis, 2018, 5016992; Chang, 2022,
9959688; Chu, 2020, 6315711; Eick, 2020, 7102797; Govarts, 2016, 3230364; Gyllenhammar,
2018, 4238300; Hjermitslev, 2020, 5880849; Kashino, 2020, 6311632; Kwon, 2016, 3858531;
Lind, 2017, 3858512; Luo, 2021, 9959610; Manzano-Salgado, 2017, 4238465; Meng, 2018,
4829851; Sagiv, 2018, 4238410; Valvi, 2017, 3983872; Wang, 2019, 5080598} were considered
adequate. The median exposure values across all studies ranged from 0.86 ng/mL {Callan, 2016,
3858524} to 42.8 ng/mL {Yao, 2021, 9960202}.
3-197
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(oe
Alkhalawi eta!., 2016, 3859818-
Ashley-Martin et al„ 2017, 3981371 -J
Bach et al., 2016, 3981534-
Bell eta!., 2018, 5041287-
Bjerregaard-Olesen et al., 2019, 5083648
Buck Louis et al., 2018, 5016992
Callan et al., 2016, 3858524 -
Caoetal., 2018, 5080197-
Chang et al., 2022, 9959688 -
Chen et al., 2017, 3981292
Chen et al., 2021, 7263985
Chuetal., 2020, 6315711 -
Eick et al., 2020, 7102797 A
Espindola Santos et al., 2021, 8442216 -
Gaoetal., 2019, 5387135-
Gennings et al., 2020, 7643497 - +
Govarts et al., 2016, 3230364^
Gross et al., 2020, 7014743-
Legend
| Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Figure 3-46. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects
Interactive figure and additional study details available on HAWC.
3-198
-------
DRAFT FOR PUBLIC COMMENT
March 2023
J
Gundacker etal., 2021, 10176483
Gyllenhammar et al., 2018, 4238300
Hjermitslev et al., 2020, 5880849
Jin et al., 2020, 6316202
Kashino et al., 2020, 6311632
Kishi etal., 2015, 2850268-
Kobayashi et al., 2017, 3981430
Kobayashi et al„ 2022, 10176408
Kwon et al.,2016, 3858531
Lauritzen et al., 2017, 3981410
Lee et al., 2013, 3859850
Lee et al.,2016, 3981528
Lenters et al., 2016, 5617416 -
Lind et al., 2017, 3858512
Luo et al., 2021, 9959610
Maekawa et al., 2017, 4238291
Manzano-Salgado et al., 2017, 4238465
Marks et al., 2019, 5081319
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-47. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-199
-------
DRAFT FOR PUBLIC COMMENT
March 2023
c#5 rf\0»5
scN-
>0®
Meng et al., 2018, 4829851
Minatoya et al„ 2017, 3981691
Robledo et al., 2015, 2851197 -
Sagiv et al„ 2018, 4238410
Shi etal., 2017, 3827535
Shoaffetal., 2018, 4619944
Starling et al., 2017, 3858473
Valvi et al., 2017, 3983872
Wang etal., 2016, 3858502
Wang etal., 2019, 5080598
Wikstrom et al., 2020, 6311677
Woods etal., 2017, 4183148
Workman et al., 2019, 5387046
Wu et al., 2012, 2919186
Xiao et al., 2020, 5918609
Xu et al., 2019, 5381338
Yao et al., 2021, 9960202
de Cock et al., 2016, 3045435
+
B
++
~
++
+
++
n
++
++
++
++
+
+
++
++
B
++
++
+
++
B
++
B
++
++
+
++
++
¦
++
++
B
+
+
~
++
++
++
++
+
++
++
++
++
79
++
+
B
++
++
++
++
++
+
B
++
B
++
++
++
+
++
++
B
++
79
++
+
+
-
++
++
++
++
+
+
++
B
B
++
++
+
++
+
B
++
++
++
-
-
¦
B
++
++
B
+
++
-
++
++
79.
++
+
+
++
B
++
++
B
+
-
++
++
++
+
++
++
B
++
B
+
-
~
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
^ Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-48. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-200
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4.1.4.1.1 Mean Birth Weight Study Results: Overall Population Studies
Thirty-two of the 37 included studies with mean BWT data that examined data in the overall
population {Bach, 2016, 3981534; Bell, 2018, 5041287; Buck Louis, 2018, 5016992; Callan,
2016, 3858524; Cao, 2018, 5080197; Chang, 2022, 9959688; Chen, 2021, 7263985; Chu, 2020,
6315711; de Cock, 2016, 3045435; Eick, 2020, 7102797; Gao, 2019, 5387135; Govarts, 2016,
3230364; Gyllenhammar, 2018, 4238300; Hjermitslev, 2020, 5880849; Kashino, 2020, 6311632;
Kwon, 2016, 3858531; Lauritzen, 2017, 3981410; Lenters, 2016, 5617416; Luo, 2021, 9959610;
Manzano-Salgado, 2017, 4238465; Marks, 2019, 5081319; Meng, 2018, 4829851; Robledo,
2015, 2851197; Shi, 2017, 3827535; Starling, 2017, 3858473; Valvi, 2017, 3983872; Wang,
2016, 3858502; Wikstrom, 2020, 6311677; Woods, 2017, 4183148; Wu, 2012, 2919186; Xu,
2019, 5381338; Yao, 2021, 9960202}, while five reported sex-specific data only {Ashley-
Martin, 2017, 3981371; Lind, 2017, 3858512; Marks, 2019, 5081319; Robledo, 2015, 2851197;
Wang, 2016, 3858502}. Twenty-one of the 32 PFOA studies reported some mean BWT deficits
in the overall population, albeit these were not always statistically significant (See PFOA
Appendix). Five of these mean BWT studies in the overall population reported null associations
{Bach, 2016, 3981534; Bell, 2018, 5041287; Buck Louis, 2018, 5016992; Valvi, 2017, 3983872;
Woods et al., 2017, 4183148}, while six reported increased mean BWT deficits with increasing
PFOA exposures {Chen, 2021, 7263985; de Cock, 2016, 3045435; Eick, 2020, 7102797; Gao,
2019, 5387135; Shi, 2017, 3827535; Xu, 2019, 5381338}. Seventeen of the 25 medium and high
confidence studies reported some BWT deficits in relation to PFOA exposures. Among the ten
studies presenting results based on categorical data, two studies {Meng, 2018, 4829851; Starling,
2017, 3858473} showed inverse monotonic exposure-response relationships (Figure 3-49, Figure
3-50, Figure 3-51, Figure 3-52).
Among the 21 studies showing some adverse associations in the overall population, there was a
wide distribution of deficits ranging from -14 to -267 grams across both categorical and
continuous exposure estimates with results based on a per unit (continuous measure) when
studies presented both. Among those with continuous PFOA results in the overall population, 14
of 20 studies reported deficits from -27 to -82 grams with increasing PFOA exposures. There
were no clear patterns were observed by confidence level, but there was a preponderance of
inverse associations based on studies with later biomarker sampling timing (i.e., trimester two
onward) including 15 of the overall 21 studies and 6 of the 9 high confidence studies only. The
two largest associations (one medium and one low confidence study) expressed per each PFOA
change were detected in studies with later pregnancy samples, while three of the four smallest
associations were based on earlier biomarker samples. Thus, some of these reported results may
be related to pregnancy hemodynamic influences on the PFOA biomarkers during pregnancy.
For example, 11 of the 12 largest mean BWT deficits (-48 grams or larger per unit change) in
the overall population were detected among studies with either later pregnancy samples (i.e.,
maternal samples during trimesters 2, 3, or post-partum or umbilical cord samples).
3-201
-------
DRAFT FOR PUBLIC COMMENT
March 2023
x Rasrtg Sampling Period Reference Sludy Desiyi
irience Earty pregnancy Bach eta!.. 2016 Cohcrt
sr pregnancy Bell el al.. 2016 Cress-sc
CbuetaL,2020 CcHcrt
Ddi at al.. 2020 Cohort
Govarts et al., 2016 Cohort
cl Cstrnale
Cxcosure Matrix Exposure levels
-150 -100
median?2.0 (2Sth-?Slh perrmKle im coeBider" P" ,Qfi <11 110
1.5-2.6 npfmL) nffynL>aToeaso
Regression coefficient for CO
{1.542.02 ngiVnL) vs. Ql j'-I.SC
rtffVnL)
Regression coefficient far 03
<2.03-2.64 rag/tal) vs. Q1 <<1.54 7.0
opfmL)
^efficient far Q4
<2.65 15.1 Cn(jVnL> vs. Q1 (*-1.54 4.0
nptoiL)
150 100
Figure 3-49. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA
Interactive figure and additional study details available on Tableau.
3-202
-------
DRAFT FOR PUBLIC COMMENT
March 2023
IV; too K Ij S3
Figure 3-50. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
Interactive figure and additional study details available on Tableau.
3-203
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Effect Estimate
Confidence Rating Sampling Period Reference Study Desiyi Exposure Matrix Expr
Early pregnancy Wikslrom el al.. 2020 Cohort
-100 -50
60 100
ancy Valvi el al.. 2017 Cohort
n: 42.83 ng,hiL [range: 1.16 602.79 ngftnL) ^mPFOA)
Regression coefficient (per 1 -In ng/mL change in PFOA) -68.0
Regression coefficient (for Q2 vs Q1)
Regression coefficient (for Q3 vs Q1)
Regression coefficient (ft* Q4 vs Ol I
Regression coefficient [per doubling of serum PFOA] -11.0
Regression coefficient (per 1 -
Figure 3-51. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
Interactive figure and additional study details available on Tableau.
Wikstrom (2019) has a manuscript error in the regression coefficient for Q4 vs Q1.
3-204
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Effect Eslrnate
Confidence Rafrrq Sampling Period Reference
Study Design
Exposure Mailrix Exposure levels
ristm EE 200
-100 0 100 200
Medium confidence Early pregnancy Chang et al_, 2022
Cohort
maternal serum median: 0.71 ng'mL (25th 75th
percentile: 0.45-1,07 ngiVnL)
Regression coefficient (per doubling in PFOA) -14.0
I
I
I
I
I
Regression coefficient Tfor 02 (1.44-2.19
npVnL> vs. 01 (999 ng/L)«.
T1 (<684 ngfL) 19'"J
i
i
i
1 •
i
i
i
Cohort and
i
i
i
Gylenhammar el al.. 2018
maternal serum Nul
PFOA 1270
• i
i
i
i
200 -100 0 100 2O0
Figure 3-52. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
Interactive figure and additional study details available on Tableau.
3.4.4.1.4.1.2 Mean BWT-Overall Population Summary
Overall, 21 of the 32 PFOA studies reported some mean BWT deficits in the overall population
with limited evidence of exposure-response relationships. Seventeen of the 21 studies were
medium or high confidence, but the majority of studies that showed inverse associations were
based on later biomarker sampling timing (i.e., trimester two onward). While some of the
changes were relatively large in magnitude (most were from -27 to -82 grams per each unit
PFOA change), there was also a pattern of stronger associations detected amongst studies with
later pregnancy biomarker samples. These patterns may be indicative of pregnancy
hemodynamic influences on the PFOA biomarkers during pregnancy.
3.4.4.1.4.1.3 Mean Birth Weight Study Results: Sex Specific Studies
Mean BWT findings were reported for 18 and 19 studies in female and male neonates,
respectively. Eleven of 18 epidemiological studies examining sex-specific results in female
neonates showed some BWT deficits including 10 of 16 medium and high confidence studies.
Twelve of 19 epidemiological studies examining sex-specific results in male neonates showed
some BWT deficits. The remaining 7 studies {Bach, 2016, 3981534; de Cock, 2016, 3045435;
Hjermitslev, 2020, 5880849; Lind, 2017, 3858512; Robledo, 2015, 2851197; Shi, 2017,
3-205
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3827535; Wang, 2019, 5080598} in male neonates were either null or showed larger birth
weights with increasing PFOA exposures. The low confidence study by Marks et al. (2019,
5081319) of boys only reported large deficits in the upper two PFOA tertiles (-53 and -46
grams, respectively) with no exposure-response relationship. None of the other 5 studies with
categorical data in either girls or boys showed evidence of monotonic exposure-response
relationships.
Nine of the 18 studies examining mean BWT associations in both boys and girls detected some
deficits in both sexes with one of these reporting comparable BWT deficits {Lenters, 2016,
5617416}. Five of the 9 studies showed larger deficits in girls {Ashley-Martin, 2017, 3981371;
Cao, 2018, 5080197; Hjermitslev, 2020, 5880849; Wang, 2019, 5080598; Wikstrom, 2020,
6311677} and 3 showed larger deficits among boys {Chu, 2020, 6315711; Lauritzen, 2017,
3981410; Meng, 2018, 4829851}. One study showed comparable results irrespective of sex
{Lenters, 2016, 5617416}. Three additional studies each reported mean BWT deficits either only
in boys {Kashino, 2020, 6311632; Manzano-Salgado, 2017, 4238465; Valvi, 2017, 3983872} or
girls {Hjermitslev, 2020, 5880849; Robledo, 2015, 2851197; Wang, 2016, 3858502}.
Overall, no consistent patterns in magnitude of deficits were observed with the sex-specific
studies by sample timing and other study characteristics; however, the three largest deficits in
male studies were later pregnancy sampled studies. Although other studies based on different
exposure measures were more variable, some consistency in the magnitude of deficits (range:
-80 to -90 g) was observed amongst 4 studies in girls {Ashley-Martin, 2017, 3981371; Wang,
2016, 3858502; Wang, 2019, 5080598; Wikstrom, 2020, 6311677} including three high
confidence studies based on continuous (i.e., per each In or loglO PFOA exposures increase).
The magnitude of deficits in boys across 7 studies {Ashley-Martin, 2017, 3981371; Kashino,
2020, 6311632; Lenters, 2016, 5617416; Manzano-Salgado, 2017, 4238465; Meng, 2018,
4829851; Wang, 2019, 5080598; Wikstrom, 2020, 6311677} was fairly consistent per each
continuous unit PFOA change (range: -21 to -49 g), although 3 studies {Chu, 2020, 6315711;
Lauritzen, 2017, 3981410; Valvi, 2017, 3983872} reported larger deficits in excess of-71
grams.
3.4.4.1.4.1.4 Standardized Birth Weight Measures
Fifteen studies examined standardized BWT measures including 14 studies reporting a change in
BWT z-scores on a continuous scale per each PFOA comparison. Eight of the 15 were high
confidence studies {Ashley-Martin, 2017, 3981371; Bach, 2016, 3981534; Eick, 2020, 7102797;
Gardener, 2021, 7021199; Sagiv, 2018, 4238410; Shoaff, 2018, 4619944; Wikstrom, 2020,
6311677; Xiao, 2019, 5918609}, 4 were medium {Chen, 2017, 3981292; Gyllenhammar, 2018,
438300; Meng, 2018, 4829851; Wang, 2019, 5080598} and 3 were low confidence {Espindola-
Santos, 2021, 8442216; Gross, 2020, 7014743; Workman, 2019, 5387046}.
Nine out of 15 studies with standardized BWT scores in the overall population showed some
inverse associations and 5 of these were high confidence. The high confidence study by
Gardener et al. (2021, 7021199) reported that participants in PFOA quartiles 2 (OR=0.84; 95%
CI: 0.40-1.80) and 3 (OR=0.91; 95% CI: 0.41-2.02) had a lower odds of being in the lowest
standardized birth weight category (vs. the top 3 birth weight z-score quartiles). They also
reported that there were no statistically significant interactions for their BWT-z measures by sex.
3-206
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Among the 14 studies examining continuous BWT z-score measures in the overall population, 8
showed some inverse associations of at least -0.1. The ranges of deficits were -0.1 {Ashley-
Martin, 2017, 3981371; Sagiv, 2018, 4238410; Wang, 2019, 5080598}, -0.2 {Chen, 2017,
3981292; Shoaff, 2018, 4619944; Wikstrom, 2020, 6311677}, and -0.3 {Gross, 2020, 7014743;
Xiao, 2019, 5918609}. More associations were detected among the high confidence studies
(6/8), compared to 2 of the 4 medium, and 1 of the 3 low confidence studies. None of the 5
studies {Bach, 2016, 3981534; Eick, 2020, 7102797; Sagiv, 2018, 4238410; Shoaff, 2018,
4619944; Wikstrom, 2020, 6311677} showed any evidence of exposure-response relationships.
Overall, 4 out of 6 studies in boys and 3 of 6 in girls showed lower BWT z-scores with
increasing PFOA exposures. For example, the low confidence study by Gross et al. (2020,
7014743) reported BWT z-score deficits for both males (-0.17; SE = 0.29; p-value = 0.57 and
females -0.38; SE = 0.26; p-value = 0.16) for PFOA levels greater than the mean level. Gardener
et al. (2021, 7021199) only reported that there were no statistically significant interactions for
BWT-z measures by sex in their analysis.
3.4.4.1.4.1.5 BWT z-score summary
Nine out of 15 studies with standardized BWT scores in the overall population showed some
inverse associations with PFOA exposures. Six of these 9 studies were either medium or high
confidence studies, and most of these had moderate or large exposure contrasts. Although some
studies may have been underpowered to detect associations small in magnitude relative to PFOA
exposure, there was consistent lower BWT z-scores reported across all confidence levels. There
was no apparent pattern related to magnitude of deficits across study confidence, but more
associations were evident across high confidence levels in general. Twice as many studies
showing adverse associations were based on later (6 of 9) versus early (i.e., at least some
trimester one maternal samples) pregnancy sampling (3 of 9); this might be reflective of some
impact of pregnancy hemodynamics on biomarker concentrations over time. There was no
evidence of exposure-response relationships in the 5 studies reporting categorical data. There
were also few evident patterns and minimal differences seen across sexes. Overall, 9 out of 15
overall studies in the overall population showed some suggestion of inverse associations with the
same studies showing associations in 3 out of 4 studies of male neonates and 3 of 4 studies in
females.
3.4.4.1.4.2 Small for Gestational Age/Low Birth Weight
Eleven informative and non-overlapping epidemiological studies examined associations between
PFOA exposure and different dichotomous fetal growth restriction endpoints, such as SGA (or
related intrauterine growth retardation endpoints), low birth weight (LBW), or both (i.e.,
{Manzano-Salgado, 2017, 4238465}). Five studies were high confidence {Chu, 2020, 6315711;
Lauritzen, 2017, 3981410; Manzano-Salgado, 2017, 4238465; Wang, 2016, 3858502; Wikstrom,
2020, 6311677}, three were medium confidence {Govarts, 2018, 4567442, Hjermitslev, 2020,
5880849; Meng, 2018, 4829851}, three were low confidence studies {Chang, 2022, 9959688;
Souza, 2020, 6833697; Xu, 2019, 5381338} and one as uninformative {Arbuckle, 2013,
2152344}. Four of these studies had good study sensitivity {Lauritzen, 2017, 3981410;
Manzano-Salgado, 2017, 4238465; Meng, 2018, 4829851; Wang et al. 2016, 3858502}, while
five were considered adequate {Arbuckle, 2013, 2152344; Chang, 2022, 9959688; Chu, 2020,
6315711; Hjermitslev, 2020, 5880849; Wikstrom, 2020, 6311677} and three were deficient
{Govarts, 2018, 4567442; Souza, 2020, 6833697; Xu, 2019, 5381338}.
3-207
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Arbuckle et al., 2013, 2152344 -
l
+
i
+
i
+
l
I
+
I
+
I
+
B
Chang et al., 2022, 9959688 -
+
+
+
+
+ * I
Chu et al., 2020, 6315711 -
+
++
+
+
++
Govarts et al., 2018, 4567442 -
+
+
++
B
+
-
J
Gundackeret al., 2021, 10176483-
+
+
-
-
+
-
B
Hjermitslev et al., 2020, 5880849 -
+
+
++
+
+
-
+
Lauritzen et al., 2017, 3981410-
+*-
++
B
+
++
+
++
++
Manzano-Salgado et al., 2017, 4238465 -
++
++
Z*
+
++
+
++
++
Meng etal., 2018, 4829851 -
D
++
+
++
+
++
B
Souza et al., 2020, 6833697 -
+
+
+
-
-
+
-
~
Wang et al., 2016, 3858502-
a
++
++
+
++
+
++
++
Wikstrom etal., 2020, 6311677-
++
++
++
+
++
+
B
++
Xu etal., 2019, 5381338-
¦
++
++
+
+
+
-
J
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
^ Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-53. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Small for Gestational Age and Low Birth Weight Effects3
Interactive figure and additional study details available on HAWC.
a Manzano-Salgado et al. (2017,4238465): High confidence for SGA; medium confidence for LBW.
Six of eight SGA studies {Chang, 2022, 9959688; Govarts, 2018, 4567442; Lauritzen, 2017,
3981410; Souza, 2020, 6833697; Wang, 2016, 3858502; Wikstrom, 2020, 6311677} showed
some adverse associations, while two studies were entirely null {Manzano-Salgado, 2017,
4238465; Xu, 2019, 5381338}. Although they were not always statistically significant, the
3-208
-------
DRAFT FOR PUBLIC COMMENT
March 2023
relative risks reported in the five studies examining the overall population based on either
categorical or continuous exposures (per each unit increase) were fairly consistent in magnitude
(odds ratio (OR) range: 1.21 to 2.81). The medium confidence study by Govarts et al. (2018,
4567442) reported an increased risk (OR = 1.64; 95% CI: 0.97, 2.76) per each PFOAIQR
increase. The high confidence study by Lauritzen et al. (2017, 3981410) showed a slight
increased risk in the overall population (OR = 1.21; 95% CI: 0.69, 2.11 per each ln-unit PFOA
increase), but this was driven by associations only in participants from Sweden (OR = 5.25; 95%
CI: 1.68, 16.4) including large risks detected for both girls and boys. One {Souza, 2020,
6833697} of the three studies examining exposure quartiles detected an exposure-response
relationship in the overall population (OR range: 1.26-2.81). The medium confidence study by
Chang et al. (2022, 9959688) reported non-monotonic but consistent statistically significant ORs
across the upper three quartiles (range: 2.22-2.44) in their study of African American pregnant
women. The high confidence study by Wikstrom et al. (2020, 6311677) reported comparable
ORs for the 4th quartiles (OR=1.44; 95% CI: 0.86, 2.40) as well as per each per ln-unit increase
(OR=1.43; 95%CI: 1.03, 1.99). Among females only, they reported a two-fold increased risk per
each ln-unit increase risk (OR = 1.96; 95% CI: 1.18, 3.28) and non-monotonic increased risks in
the upper two quartiles (OR range: 1.64-2.33). The high confidence study by Wang et al. (2016,
3858502) only reported sex-specific results but also showed an increased risk (OR = 1.48; 95%
CI: 0.63, 3.48 per each ln-unit increase) for SGA among girls only.
Sampling Exposure Study
Period Reference Matrix Design Exposure Levels Sub-population Comparison EE
Effect Estimate ih
012 34 5 67 89 10
OR (per doubling in
Early Manzano- Plasma, Cohort Mean (SD): 2.35 ng/mL (1.25 Boys } , . 1-18
_ . .. ' ,.\ maternal plasma PFOA)
pregnancy Salgadoet Maternal ng/mL)
al., 2017 Blood
1
1
1
1
. OR (per doubling in
Girls . . . 0.72
maternal plasma PFOA)
1
1
-•-1
1
1
OR (per doubling in QQ2
maternal plasma PFOA)
i
i
-•t-
i
i
_ . Median=l.62 ng/mL (range: OR (per In unit increase
Later Launtzen Maternal Cohort 031.7 97 /mL) Norway jn 0.66
pregnancy et al., 2017 Serum
i
i
!
i
Medien=2.33 ng/mL (range: Sweden OR (per In unit increase
0.60-6.70 ng/mL) ln PF0A>
i
i
i
i
!
„ OR (per In unit increase „ „
Sweden; Boys |n ^ 6.55
i
i
i
i
Sweden; Girls °a
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Interactive figure and additional study details available on Tableau.
Small-for-gestational-age defined as birthweight below the 10th percentile for the reference population.
Effect Estimate
Sampling Exposure Study
Period Reference Matrix Design Exposure Levels
Sub-population Comparison
Early Wikstrom Maternal Cohort Median-1.61 ng/mL Boys
pregnancy etal., 2020 Serum (25th-75th percentiles:
1.11-2.30 ng/mL)
OR (per 1-ln ng/mL
change in PFOA)
OR {for Q2 vs Ql) 0.67
OR (for Q3 vs Ql) 0.66
OR {for Q4 vs Ql) 1.04
OR (per 1-ln ng/mL
change in PFOA)
OR (for Q2 vs Ql) 1
OR{forQ3 vs Ql) 1.64
0R(forQ4vs Ql) 2.33
OR (per 1-ln ng/mL
change in PFOA)
1.43
OR (for Q2 vs Ql) 0.77
OR(forQ3 vs Ql) 0.96
0R(forQ4vs Ql) 1.44
Figure 3-55. Odds of Small-for-gestational-age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on Tableau.
Small-for-gestational-age defined as birthweight below the 10th percentile for the reference population.
3-210
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Sampling
Period
Reference
Exposure
Matrix
Study
Design
Exposure Levels Sub-population
Comparison
EE 0.0 0.5
Effect Estimate
.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Early
pregnancy
Chang etal.
2022
Maternal
Serum
Cohort
Median: 0.71 ng/mL Term Births
(25th-75th percentile:
0.45-1.07 ng/mL)
OR (per doubling
in PFOA)
12.
'b-
OR [for Q2
(0.45-0.71
ng/mL) vs. Q1
(
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Confidence Sampling Measured Exposure Study
Rating Period Reference Effect/Endpoints Matrix Design Sub-population Comparison EE
Effect Estimate
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
OR (per doubling in
High Early Manzano- Low Birth Weight Plasma, Cohort -- . . . -cnn 0.9
c , ^ . .... . maternal plasma PFOA)
confidence pregnancy Salgado etal.. Maternal
1
1
• 1
1
2017 Bl0Od oo /
_ OR (per doubling in .
Boys .. , nc_.. 1.12
maternal plasma PFOA)
1
1 _
1
1
. OR (per doubling in
Girls . . . 0.76
maternal plasma PFOA)
1
1
* 1
1
Term Low Birth Plasma, Cohort ~ °R(per doub"r,g in
Weight Maternal maternal plasma PFOA)
1
_ 1
" 1
1
_ OR (per doubling in
Boys I nc.., 1.67
maternal plasma PFOA)
1
1
1 *
1
. OR (per doubling in
Girls 0.62
maternal plasma PFOA)
1
1
1
OR (per 1 In ng/mL
Later Chu etal., 2020 Low Birth Weight Maternal Cohort - . . _J* 1.16
increase in PFOA)
pregnancy Serum
1
1
1 *
1
1.54 ng/mL PFOA) vs. Q1
(<=0.96 ng/mL PFOA) °'61
1
1
1
1
OR for Q3 (> 1.54 to
2.63 ng/mL PFOA) vs. Q1 0.27
(<=0.96 ng/mL PFOA)
1
1
• 1
1
PFOA) vs. 01 (<=0.96
ng/mL PFOA) 1
1
1
.. ,. _ Hjermitslevet , ....... . .. Maternal _ . , OR (per 1 In-ng/mL
Medium Early . Low Birth Weight _ Cohort - . 0.44
' al., 2019 M Serum change in PFOA)
confidence pregnancy
1
1
• 1
1
OR (per doubling of
Mengetal., Low Birth Weight Maternal Cohort ~ pfoaI 1
2018 Serum
1
1
OR (for Q2 vs. 01) 1.5
1
1
1 *
1
OR (for Q3 vs. Ql) 1.2
1
1 _
1
1
OR (for Q4 vs. Ql) 1.5
1
1
1
1
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Figure 3-57. Odds of Low Birthweight in Children from Epidemiology Studies Following
Exposure to PFOA
Interactive figure and additional study details available on Tableau.
Low birthweight defined as birthweight < 2,500 g.
Overall, nine of the eleven informative studies reporting main effects for either SGA or LBW or
both showed some increased risks with increasing PFOA exposures. The magnitude of the
associations was typically from 1.2 to 2.8 with limited evidence of exposure-response
relationships among the studies with categorical data. Although the number of studies was fairly
small, few discernible patterns across study characteristics or confidence ratings were evident
across the SGA or LBW findings. For example, four of the nine studies showing increased odds
of either SGA or LBW were based on early sampling biomarkers. Collectively, the majority of
SGA and LBW studies were supportive of an increased risk with increasing PFOA exposures.
3.4.4.1.4.3 Birth Length
As shown in Figure 3-58 and Figure 3-59, 34 birth length studies were considered as part of the
study evaluation. Four studies were considered iminformative {Alkhalawi, 2016, 3859818;
Gundacker, 2021, 10176483; Jin, 2020, 6315720; Lee, 2013, 3859850} and four more studies
noted above {Bach, 2016, 3981534; Kishi, 2015, 2850268, Kobayashi, 2017, 3981430;
Kobayashi, 2022, 10176408} were not further considered for multiple publications from the
same cohort studies. Among the twenty-six non-overlapping informative studies examined birth
length in relation to PFOA, including five studies with standardized birth length measures
3-212
-------
DRAFT FOR PUBLIC COMMENT
March 2023
{Chen, 2017, 3981292; Espindola-Santos, 2021, 8442216; Gyllenhammar, 2018, 4238300;
Shoaff, 2018, 4619944; Xiao, 2019, 5918609}, and one study evaluated standardized and mean
birth length changes {Workman, 2019, 5387046}. Eighteen studies examined mean birth length
differences in the overall study population. Thirteen studies examined sex-specific data with
three studies {Marks, 2019, 5081319; Robledo, 2015, 2851197; Wang, 2016, 3858502}
reporting only sex-specific results.
Nine of the 26 studies were high confidence {Bell, 2018, 5041287; Bjerregaard-Olesen, 2019,
5083648; Buck Louis, 2018, 5016992; Lauritzen, 2017, 3981410; Manzano-Salgado, 2017,
4238465; Shoaff, 2018, 4619944; Valvi, 2017, 3983872; Wang, 2016, 3858502; Xiao, 2019,
5918609}, eight were medium {Chen, 2017, 3981292; Chen, 2021, 7263985; Gyllenhammar,
2018, 4238300; Hjermitslev, 2020, 5880849; Kashino, 2020, 6311632; Luo, 2021, 9959610;
Robledo, 2015, 2851197; Wang, 2019, 5080598} and nine were low confidence {Callan, 2016,
3858524; Cao, 2018, 5080197; Espindola-Santos, 2021, 8442216; Gao, 2019, 5387135; Marks,
2019, 5081319; Shi, 2017, 3827535; Workman, 2019, 5387046; Wu, 2012, 2919186; Xu, 2019,
5381338}. Eight PFOA studies had good study sensitivity {Bjerregaard-Olesen, 2019, 5083648;
Chen, 2021, 7263985; Lauritzen, 2017, 3981410; Manzano-Salgado, 2017, 4238465; Robledo,
2015, 2851197; Shoaff, 2018, 4619944; Wang, 2016, 3858502; Wu, 2012, 2919186}, 14 had
adequate {Buck Louis, 2018, 5016992; Callan, 2016, 3858524; Cao, 2018, 5080197; Chen,
2017, 3981292; Gao, 2019, 5387135; Gyllenhammar, 2018, 4238300; Hjermitslev, 2020,
5880849; Kashino, 2020, 6311632; Luo, 2021, 9959610; Marks, 2019, 5081319; Shi, 2017,
3827535; Valvi, 2017, 3983872; Wang, 2019, 5080598; Xiao, 2019, 5918609} sensitivity and
four {Bell, 2018, 5041287; Espindola-Santos, 2021, 8442216; Workman, 2019, 5387046; Xu,
2019, 5381338} considered deficient.
3-213
-------
DRAFT FOR PUBLIC COMMENT
March 2023
,o®
Alkhalawi et al., 2016, 3859818 -
1
1
+
1
+
—1—
—1—
i
+
L
+
Bach et al„ 2016, 3981534-
+
+
+
+
++
+
++
Bellet al„ 2018, 5041287-
+
+
++
++
+
¦
++
Bjerregaard-Olesen et al., 2019, 5083648 -
+
+
++
+
++
++
Buck Louis et al., 2018, 5016992-
+
+
+
++
+
D
++
Callan etal., 2016, 3858524-
+
+
+
-
+
+
+
-
Cao etal., 2018, 5080197-
-
+
+
-
+
+
+
-
Chen et al., 2017, 3981292-
+
+
++
+
++
+
+
+
Chen et al., 2021, 7263985-
+
2
+
++
+
++
+
Espindola Santos et al., 2021, 8442216 -
+
+
+
-
+
-
-
Gao etal., 2019, 5387135-
+
+
-
-
+
+
-
Gundacker et al., 2021, 10176483 -
+
+
*
-
+
-
i
Gyllenhammar et al., 2018, 4238300 -
+
+
+
+
++
+
+
+
Hjermitslev et al., 2020, 5880849 -
+
+
+
+
+
-
+
+
Jinet al., 2020, 6316202-
"
+
D
-
+
++
--
Kashino etal., 2020, 6311632-
+
++
+
+
+
+
+
+
Kishi et al., 2015, 2850268-
+
+
+
+
+
-
+
+
Legend
1 Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-58. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Length Effects
Interactive figure and additional study details available on HAWC.
3-214
-------
DRAFT FOR PUBLIC COMMENT
March 2023
><$¦
V ^ °°^
,o0
Kobayashi et al., 2017, 3981430-
Kobayashi etal., 2022, 10176408-
Lauritzen et al., 2017, 3981410 -
Lee et al., 2013, 3859850-
Luo etal., 2021, 9959610
Manzano-Salgado et al., 2017, 4238465
Marks etal., 2019, 5081319-
Robledo et al., 2015, 2851197
Shi etal., 2017, 3827535-
Shoaffetal., 2018,4619944
Valvi etal., 2017, 3983872
Wang et al., 2016, 3858502-
Wang et al., 2019, 5080598-
Workman et al., 2019, 5387046
Wu et al., 2012, 2919186-
Xiao etal., 2020, 5918609
Xu etal., 2019, 5381338
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Figure 3-59. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Length Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-215
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Amongst the 26 birth length studies (examining mean differences or changes in standardized
scores), nine of them reported some inverse associations including three of the six studies that
reported standardized birth length data. There was limited evidence of exposure-response
relationships in the three studies that examined categorical data. The high confidence study by
Xiao et al. (2019, 5918609) reported a reduced birth length z-score (-0.14; 95%CI: -0.40, 0.13)
in the overall population per each log2 increase in PFOA that appeared to be driven by male
neonates (-0.27; 95% CI: -0.65, 0.10). The low confidence Workman et al. {2019, 5387046}
study reported a non-significant deficit similar in magnitude (-0.26; 95% CI: -1.13, 0.61). The
other study high confidence study by Shoaff et al. (2018, 4619944) of standardized birth length
measures showed a deficit only for tertile 3 (-0.32; 95% CI: -0.72, 0.07) compared to tertile 1.
In contrast, the low confidence study by Espindola-Santos et al. (2021, 8442216) reported a
larger birth length z-score per each loglO PFOA increase (0.26; 95%CI: -0.21, 0.73).
Amongst the 21 studies examining mean birth length differences, eight different studies showed
adverse associations. This included six different studies (out of 18) based on the overall
population as well two out of three studies {Robledo, 2015, 2851197; Wang, 2016, 3858502}
reporting only sex-specific results. The high confidence study by Wang et al. (2016, 3858502)
only showed deficits among females for only PFOA quartiles 1 (-0.39 cm; 95% CI: -1.80, 1.02)
and 3 (-0.60 cm; 95% CI: -1.98, 0.77). The medium confidence study by Chen et al. {2021,
7263985} reported similar birth length deficits in the overall population (-0.27 cm; 95%CI: -
0.61, 0.07), males (-0.21; 95%CI: -0.73, 0.32) and females (-0.21; 95%CI: -0.74, 0.33) per each
ln-unit PFOA increase. In the medium confidence study by Robledo et al. (2015, 2851197),
smaller deficits in birth length were detected for both male and female neonates per each 1
standard deviation (SD) PFOA increase. The high confidence study by Lauritzen et al. (2017,
3981410) showed a deficit in the overall population (-0.49 cm; 95% CI: -0.99, 0.02), but
detected the strongest association when restricted to the Swedish population (-1.2 cm; 95% CI:
-2.1, -0.3) and especially Swedish boys (-1.6 cm; 95%CI: -2.9, -0.4). Overall, four sex-
specific studies showed deficits for both boys and girls with two studies showing larger deficits
among boys. One study showed larger deficits amongst girls and the fourth study showed results
equal in magnitude.
In the overall population studies showing adverse associations, the reported magnitude of deficits
was quite variable (range: -0.16 to -1.91 cm). For example, the low confidence study by Wu et
al. (2012, 2919186) showed the largest deficit (-1.91 cm; 95% CI: -3.31,-0.52 per eachloglO
increase). The low confidence study by Cao et al. (2018, 5080197) showed consistent results
across their overall population (-0.45 cm; 95%CI: -0.79, -0.10 per each ln-unit PFOA increase),
male (-0.36 cm; 95% CI: -0.80, 0.09), and female neonates (-0.58 cm; 95% CI: -1.12, -0.04)
with evidence of exposure-response relationships in all three of these groups. Overall, 6 of 12
studies in girls and 4 of 13 studies in boys showed some birth length deficits. One of the three
studies in either or both boys and girls showed some additional evidence of exposure-response
relationships. The same study by Cao et al., {2018, 5080197} was the only study in the overall
population to show evidence of exposure-response.
Overall, nine different studies out of 26 studies examining birth length deficits in relation to
PFOA exposures. There was no apparent relationship between studies showing inverse
associations and study confidence ratings. However, six of these studies sampled PFOA
biomarkers characterized as later in pregnancy and may be more prone to potential bias from
3-216
-------
DRAFT FOR PUBLIC COMMENT
March 2023
pregnancy hemodynamic changes. Among the mean birth length studies, most showed consistent
deficits ranging from -0.21 to -0.49 cm per different PFOA comparisons. An unusually large
result (-1.91 cm per each loglO PFOA increase) was reported in an earlier study {Wu, 2012,
2919186} that reported the largest exposure range. There was a preponderance of inverse
associations amongst females (6 of 12 studies) compared to males (4 of 13); however, amongst
the four studies that reported associations in both sexes, more studies reported larger deficits in
male neonates.
3.4.4.1.4.4 Head Circumference at Birth
As shown in Figure 3-60, 21 informative studies examined head circumference at birth in
relation to PFOA exposures. Six of the 21 studies were low confidence {Callan, 2016, 3858524;
Cao, 2018, 5080197; Espindola-Santos, 2021, 8442216; Marks, 2019, 5081319; Workman, 2019,
5387046; Xu, 2019, 5381338}, while studies seven were medium {Chen, 2021, 7263985;
Gyllenhammar, 2018, 4238300; Hjermitslev, 2020, 5880849; Kashino, 2020, 6311632; Lind,
2017, 3858512; Robledo, 2015, 2851197; Wang, 2019, 5080598} and eight were high
confidence {Bell, 2018, 5041287; Bjerregaard-Olesen, 2019, 5083648; Buck Louis, 2018,
5016992; Lauritzen, 2017, 3981410; Manzano-Salgado, 2017, 4238465; Valvi, 2017, 3983872;
Wang, 2016, 3858502; Xiao, 2019, 5918609}. Four studies were deficient in study sensitivity
{Bell, 2018, 5041287; Espindola-Santos, 2021, 8442216; Workman, 2019, 5387046; Xu, 2019,
5381338}, while five were good {Chen, 2021, 7263985; Lauritzen, 2017, 3981410; Manzano-
Salgado, 2017, 4238465; Robledo, 2015, 2851197; Wang, 2016, 3858502} and twelve had
adequate study sensitivity {Bjerregaard-Olesen, 2019, 5083648; Buck Louis, 2018, 5016992;
Callan, 2016, 3858524; Cao, 2018, 5080197; Gyllenhammar, 2018, 4238300; Hjermitslev, 2020,
5880849; Kashino, 2020, 6311632; Lind, 2017, 3858512; Marks, 2019, 5081319; Valvi, 2017,
3983872; Wang, 2019, 5080598; Xiao, 2019, 5918609}.
3-217
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Bell et al.
2018, 5041287
Bjerregaard-Olesen et al.
Buck Louis et al.
Callan et al.
Cao et al.
Chen et al.
Espindola Santos et al.
Gundacker et al., 2021, 10176483
Gyllenhammar et al., 2018,4238300 -
Hjermitslev et al., 2020, 5880849
Kashino et al., 2020, 6311632
Lauritzen et al., 2017, 3981410
Lind et al., 2017, 3858512
Manzano-Salgado et al., 2017,4238465
Marks etal., 2019, 5081319
Robledo et al., 2015, 2851197 -
Valvi etal., 2017, 3983872-
Wang etal., 2016, 3858502-
Wang etal., 2019, 5080598-
Workman et al., 2019, 5387046-
Xiao et al., 2020, 5918609-
Xu etal., 2019, 5381338-
Legend
| Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-60. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Head Circumference Effects
Interactive figure and additional study details available on HAWC.
Eighteen of the 21 included studies reported PFOA in relation to mean head circumference
differences including 17 studies that provided results based on the overall population. Including
the Xiao et al. (2019, 5918609) z-score data, thirteen of these 21 studies reported sex-specific
3-218
-------
DRAFT FOR PUBLIC COMMENT
March 2023
head circumference data with four other studies {Lind, 2017, 3858512; Marks, 2019, 5081319;
Robledo, 2015, 2851197; Wang, 2016, 3858502} providing sex-specific data only.
Among the 21 studies, ten reported some inverse associations between PFOA exposures and
different head circumference measures in the overall population, in either or both male and
female neonates, across different racial strata, or different countries in the same study population.
For example, the high confidence study by Lauritzen et al. (2017, 3981410) reported a similar
deficit only in their Swedish population (-0.4 cm; 95% CI: -1.0, 0.1) per each ln-unit PFOA
change; this was largely due to an association seen in male neonates (-0.6 cm; 95% CI: -1.3,
0.1). The high confidence study by Buck Louis et al. (2018, 5016992), reported non-significant
head circumference differences (-0.14 cm; 95% CI: -0.29, 0.02) among black neonates but no
main effect association in the overall population. Six out of seventeen studies based on the
overall population reported some adverse associations between PFOA exposures and either mean
head circumference measures or standardized z-scores. The high confidence study by Xiao et al.
(2019, 5918609) reported a reduced head circumference z-score (-0.17; 95% CI: -0.48, 0.15) in
the overall population per each log2 increase in PFOA that appeared to be driven by female
neonates (-0.30; 95% CI: -0.74, 0.13) (data not shown on figures). Although it was not
statistically significant, the low confidence study by Espindola-Santos et al. (2021, 8442216)
reported a larger head circumference z-score (0.62; 95% CI: -0.06, 1.29 per each loglO PFOA).
The medium confidence study by Gyllenhammar et al. (2018, 4238300) was null based on their
standardized head circumference measure.
Among the fifteen studies that examined mean head circumference at birth in the overall
population, four of them reported adverse associations. Eight studies were largely null, and two
studies showed larger mean head circumference in the overall population with increasing PFOA
exposures. Of the eleven different studies examining sex-specific results associations including
five of ten in female neonates {Bjerregaard-Olesen, 2019, 5083648; Cao, 2018, 5080197;
Hjermitslev, 2020, 5880849; Robledo, 2015, 2851197; Wang, 2019, 5080598} and three
{Lauritzen, 2017, 3981410; Manzano-Salgado, 2017, 4238465; Wang, 2019, 5080598) of eleven
studies in male neonates. The medium confidence study by Wang et al. (2019, 5080598) reported
an association in the overall population (-0.37 cm; 95% CI: -7.00, -0.40) with larger deficits
noted in female (-0.57 cm; 95% CI: -1.07, -0.08) than in male neonates (-0.35 cm; 95% CI:
-0.79, -0.10). The medium confidence study by Hjermitslev et al. (2019, 5880849) showed a
non-significant reduction in head circumference for the overall population (-0.14 cm; 95% CI:
-0.42, 0.14 per each ng/ml PFOA increase) which seemed to be driven by results in females
(-0.25 cm; 95% CI: -0.65, 0.14). The high confidence study by Manzano-Salgado et al. (2017,
4238465) reported a non-significant decrease only in quartile 4 (-0.16 cm; 95% CI: -0.38, 0.06)
compared to quartile 1 and a deficit among male neonates only (-0.13 cm; 95% CI: -0.27, 0.0)
per each log2 PFOA increase. In the medium confidence study by Robledo et al. (2015,
2851197), opposite results were seen for male (0.18 cm; 95% CI: -0.25, 0.60) and female
neonates (-0.18 cm; 95% CI: -0.59, 0.23) per each 1 SD PFOA increase. In their low confidence
study, Cao et al. (2018, 5080197) reported an overall null association, while divergent and large
changes were seen for male (0.72 cm; 95% CI: -0.51, 1.94) and female neonates (-1.46 cm; 95%
CI: -2.96, 0.05) per each ln-unit PFOA increase. The low confidence study by Callan et al.
3-219
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(2016, 3858524) reported a -0.40 cm (95% CI: -0.96, 0.16) difference per each ln-unit PFOA
change.
Among the 21 epidemiological studies examining PFOA and head circumference, nine different
studies reported some evidence of adverse associations in the overall population or across sexes.
This included four of fifteen studies in the overall population and five of twelve sex-specific
studies in either or both sexes. Few patterns across sex were reported as deficits were found in
four or fewer studies in both male and female neonates. Apart from the Wang et al. (2019,
5080598) study, no other sex-specific studies reported reduced head circumference in both sexes.
Few patterns by study characteristics or overall confidence levels although nearly all of the high
and low confidence studies were null. Among the nine different studies reporting associations
across various populations examined there was no definitive pattern of results by biomarker
sample timing as five studies relied on early sampling periods.
3.4.4.1.4.5 Fetal Growth Restriction Summary
The majority of studies examining fetal growth restriction showed some evidence of associations
with PFOA exposures especially those that included BWT data (i.e., SGA, low BWT, as well as
mean and standardized BWT measures). The evidence for two fetal growth measures such as
head circumference and birth length were less consistent but still reported many associations. For
example, 10 (out of 21) different epidemiological studies of PFOA examining head
circumference reported some evidence of adverse associations in either the overall population or
across the sexes. Nine different studies out of 26 studies reported some birth length deficits in
relation to PFOA exposures with limited evidence of exposure-response relationships. Across the
fetal growth measures, there was not consistent evidence of sexual dimorphic differences across
the fetal growth measures; however, as noted above, many of the individual study results lacked
precision and power to detect sex-specific differences. There was minimal evidence of exposure-
response relationships reported amongst those examining categorical exposure data, but the
categorical data generally supported the linearly expressed associations that were detected.
Among the most accurate fetal growth restriction endpoints examined here, there was generally
consistent evidence for BWT deficits across different measures and types of PFOA exposure
metrics considered. For example, nearly two-thirds of studies showed BWT deficits based on
differences in means or standardized measures. There was limited evidence of exposure-response
relationships in either analyses specific to the overall population or different sexes, although the
categorical data generally supported the linearly expressed associations that were detected.
Associations were also seen for the majority of studies examining small for gestational age and
low birth weight measures. The magnitude of some fetal growth measures were at times
considered large especially when considering the per unit PFOA increases across the exposure
distributions. The range of deficits detected in the overall population across all categorical and
continuous exposure estimates ranged from -14 to -267 grams. Among those with continuous
PFOA results in the overall population. For example, 14 of the 21 studies reported deficits from -
27 to -82 grams in the overall population based on each unit increase in PFOA exposures.
The current database (since the 2016 HESD) is fairly robust given the wealth of studies included
here with most of them considered high or medium confidence (e.g., 17 out of 25 mean BWT
studies with data in the overall population) and most of them had adequate or good study
sensitivity, so the database is fairly robust. As noted earlier, one source of uncertainty is that
previous meta-analyses of PFOS by Dzierlenga et al. (2020, 7643488) and PFOA by Steenland et
3-220
-------
DRAFT FOR PUBLIC COMMENT
March 2023
al. (2018, 5079861) have shown that some measures like mean BWT may be prone to bias from
pregnancy hemodynamics especially in studies with sampling later in pregnancy. For many of
these endpoints, such as birth weight measures, there was a preponderance of associations
amongst studies with later biomarker samples (i.e., either exclusive trimester 2/3 maternal
sample or later, such as umbilical cord or post-partum maternal samples). This would seem to
comport with the PFOA meta-analysis by Steenland et al. (2018, 5079861) that suggested that
results for mean BWT may be impacted by some bias due to pregnancy hemodynamics.
Therefore, despite some consistency in evidence across these fetal growth endpoints, some
important uncertainties remain mainly around the degree that some of the results examined here
may be influenced by sample timing. This source of uncertainty and potential explanation of
different results across studies may indicate some bias due to the impact of pregnancy
hemodynamics.
3.4.4.1.5 Postnatal Growth
Thirteen studies examined PFOA exposure in relation to postnatal growth measures. The
synthesis here is focused on postnatal growth measures including body mass index
(BMI)/adiposity measures {Chen, 2017, 3981292; de Cock, 2014, 2713590; Gross, 2020,
7014743; Jensen, 2020, 6833719; Shoaff, 2018, 4619944; Starling, 2019, 5412449; Yeung,
2019, 5080619} and rapid growth during infancy {Manzano-Salgado, 2017, 4238509; Shoaff,
2018, 4619944; Starling, 2019, 5412449; Tanner, 2020, 6322293; Yeung, 2019, 5080619}, as
well as mean and standardized weight (all 13 studies except Gross et al. (2020, 7014743), Tanner
et al. (2020, 6322293), and Jensenet al. (2020, 6833719) depicted in Figure 3-61), and height
{Cao, 2018, 5080197; Chen, 2017, 3981292; de Cock, 2014, 2713590; Gyllenhammar, 2018,
4238300; Lee, 2018, 4238394; Shoaff, 2018, 4619944; Wang, 2016, 3858502; Yeung, 2019,
5080619} measures.
Six postnatal growth studies were high confidence {Jensen, 2020, 6833719; Shoaff, 2018,
4619944; Starling, 2019, 5412449; Tanner, 2020, 6322293; Wang, 2016, 3858502; Yeung, 2019,
5080619}, four were medium confidence {Chen, 2017, 3981292; de Cock, 2014, 2713590;
Gyllenhammar, 2018, 4238300; Manzano-Salgado, 2017, 4238509} and three were low
confidence {Cao, 2018, 5080197; Gross, 2020, 7014743; Lee, 2018, 4238394}. Five postnatal
growth studies had good study sensitivity {Lee, 2018, 4238394; Manzano-Salgado, 2017,
4238509; Shoaff, 2018, 4619944; Tanner, 2020, 6322293; Wang, 2016, 3858502}, six were
adequate {Cao, 2018, 5080197; Chen, 2017, 3981292; Gyllenhammar, 2018, 4238300; Jensen,
2020, 6833719; Starling, 2019, 5412449 Yeung, 2019, 5080619} and two were considered
deficient {de Cock, 2014, 2713590; Gross, 2020, 7014743}. The synthesis here is focused on
postnatal body mass index (BMI)/adiposity measures, head circumference and mean and
standardized weight and height measures. Rapid growth during infancy is also included as it was
examined in five studies {Manzano-Salgado, 2017, 4238509; Shoaff, 2018, 4619944; Starling et
al. 2019, 5412449; Tanner et al. 2020; Yeung, 2019, 5080619}. The medium confidence study by
deCock et al. (2014, 2713590) did not report effect estimates for postnatal infant height (p-
value=0.045), weight (p-value=0.35), and BMI (p-value=0.81) up to 11 months of age. But their
lack of reporting of effect estimates precluded consideration of magnitude and direction of any
associations and are not further considered below in the summaries.
The medium confidence study by Manzano-Salgado et al. (2017, 4238509) had null associations
for their overall population and female neonates measured at 6 months but reported an increased
3-221
-------
DRAFT FOR PUBLIC COMMENT
March 2023
weight gain z-score for males (0.13; 95% CI: 0.01, 0.26) per each log2 PFOA increases. The
medium confidence study by Chen et al. (2017, 3981292) did not report associations between
each per In unit PFOA exposure increase and height z-score measures up to 24 months of age.
The sex-specific data were not always consistent across time. For example, non-significant
increases small in magnitude for boys (0.11; 95% CI: -0.04, 0.27) and decreases in greater
height per each In unit PFOA increase in the 12- to 24-month window. The low confidence study
by Lee et al. (2018, 4238394) reported statistically significant associations detected for mean
height differences at age 2 years (-0.91 cm; 95% CI: -1.36, -0.47 for each PFOA In unit
increase), as well as height change from birth to 2 years (-0.86 cm; 95% CI: -1.52, -0.20).
Large differences were seen for mean weight differences at age 2 years (-210 g; 95% CI: -430,
0.20) but not for weight change from birth to 2 years. An exposure-response relationships was
detected when examined across PFOA categories with the highest exposure associated with
smaller statistically significant height increases at age 2 compared to lower exposures.
In the medium confidence study by Gyllenhammar et al. (2018, 4238300), no associations were
detected for infant height deficits among participants followed from 3 months to 60 months of
age per each IQRPFOA change. They also did not report statistically significant standardized
BWT deficits per each IQR PFOA change, but they did show slight weight deficits
(approximately -0.2) at 3 months that gradually decreased over time (to approximately -0.1) at
60 months of age. Compared to the PFOA tertile 1 referent, the low confidence study by Cao et
al. (2018, 5080197) reported slight increases (1.37 cm; 95% CI: -0.5, 3.28) in postnatal length
(i.e., height) amongst infants (median age of 19.7 months), while large postnatal weight deficits
were reported for tertile 2 (-429.2 g; 95% CI: -858.4, -0.12) and tertile 3 (-114.9 g; 95% CI:
-562.0, 332.1). These height increases were predominately due to female infants, while the
weight deficits were driven by males. Few differences were observed in the overall population
for postnatal head circumference with slight non-significant deficits seen amongst females only.
In their high confidence study, Wang et al. (2016, 3858502) reported statistically significant
childhood weight (-0.14; 95% CI: -0.39, 0.11) and height (-0.15; 95% CI: -0.38, 0.08) z-scores
for female neonates when averaged over the first 11 years and per 1-ln-unit PFOA increase.
Results were null for male neonates for childhood average weight(0.03; 95% CI: -0.11, 0.18)
and height (0.01; 95% CI: -0.24, 0.25) z-scores. However, when they examined the first 2 years
only, statistically significant deficits in both height and weight z-scores were only seen for male
neonates. These weight deficits dissipated in males later during childhood, while the height
deficits detected at age 2 years continued through age 11. In contrast, the height deficits in
female children that were detected at birth were no longer evident in older kids until later ages
(i.e., 11 years). The weight deficits in female children detected at birth did not persist.
In their high confidence study, Yeung et al. (2019, 5080619) reported statistically significant
negative growth trajectories for weight for length z-scores in relation to each log SD increase in
PFOA exposures among singletons followed for three years. In contrast, the authors showed
positive infant length (i.e., height) growth trajectory across two different measures. Some sex-
specific results were detected with larger associations seen in singleton females for weight for
length z-score (-0.13; 95% CI: -0.19, -0.06). An infant weight deficit of-12.6 g (95% CI: -
49.5, 24.3 per each 1 log SD PFOA increase) was also observed and appeared to be driven by
results in females (-30.2 g; 95% CI: -84.1, 23.6). In their high confidence study of repeated
measures at 4 weeks, 1 year and 2 years of age, Shoaff et al. (2018, 4619944) detected
3-222
-------
DRAFT FOR PUBLIC COMMENT
March 2023
statistically significant deficits for weight-for-age (-0.46; 95% CI: -0.78, -0.14) z-score, and
weight-for-length z-score (-0.34; 95% CI: -0.59, -0.08) in PFOA tertile 3 compared to tertile 1
with exposure-response relationships detected for infant weight-for-length z-score. Deficits
comparable in magnitude that were not statistically significant were observed in tertile 3 for
height measured as length for age z-score (-0.32; 95% CI: -0.72, 0.07). No associations were
found in the overall population from the high confidence study by Starling et al. (2019, 5412449)
for postnatal measures at 5 months of age, but an exposure-response relationship of increased
adiposity was seen among male neonates with increasing PFOA tertiles (2.81; 95% CI: 0.79,
4.84 for tertile 3). Similarly, no associations were found in the overall population for weight-for-
age or weight-for-length z-scores and PFOA exposures, but both measures were increased
among male neonates.
Overall, seven of nine studies with quantitative estimates (including six high and medium
confidence studies) showed some associations between PFOA exposures and different measures
of infant weight. Two of four studies with categorical data showed some evidence of inverse
monotonic exposure-response relationships. Three (two high and one low confidence) of seven
studies with quantitative estimates examining different infant height measures showed some
evidence of adverse associations with PFOA. Study quality ratings, including study sensitivity
and overall confidence, did not appear to be explanatory factors for heterogeneous results across
studies.
3-223
-------
DRAFT FOR PUBLIC COMMENT
March 2023
I I I I I I I
0#
\&e
Cao etal., 2018, 5080197-
Chen etal., 2017, 3981292-
Gross et al., 2020, 7014743
Gyllenhammar et al., 2018, 4238300
Jensen et al., 2020, 6833719 -
Lee etal., 2018, 4238394
Manzano-Salgado et al., 2017, 4238509
Shoaff et al., 2018, 4619944
Starling etal., 2019, 5412449
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Tanner et al., 2020, 6322293 -
++
+
++
+
Wang etal., 2016, 3858502-
J
+ +
+ +
+
Yeung etal., 2019, 5080619-
~
+
++
+
de Cock et al., 2014, 2713590-
+
+
+
+
Figure 3-61. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Postnatal Growth
Interactive figure and additional study details available on HAWC.
3-224
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4.1.5.1 Adiposity/BMI
The medium confidence study by Chen et al. (2017, 3981292) reported lower BMI z-scores
(-0.16; 95%CI: -0.37, 0.05) per each In unit PFOA increase in the birth to 6-months window. In
their high confidence study of repeated measures at 4 weeks, 1 year, and 2 years of age, Shoaff et
al. (2018, 4619944) detected statistically significant deficits for infant BMI z-score (-0.36; 95%
CI: -0.60, -0.12) in PFOA tertile 3 compared to tertile 1 with exposure-response relationships
detected for infant BMI z-score. The high confidence study by Yeung et al. (2019, 5080619)
reported statistically significant negative growth trajectories for BMI, BMI z-score in relation to
each log SD increase in PFOA exposures among singletons followed for three years. Some sex-
specific results were detected with larger associations seen in singleton females for BMI (-0.18
kg/m2; 95% CI: -0.27, -0.09) and BMI z-scores (-0.13; 95% CI: -0.19, -0.07). An exposure-
response relationship was evident with decreasing BMI z-scores across PFOA quartiles in the
overall population and for female neonates. An exposure-response relationship of increased
adiposity was seen among male neonates with increasing PFOA tertiles (2.81; 95% CI: 0.79,
4.84 for tertile 3) in the high confidence study by Starling et al. (2019, 5412449). The high
confidence study by Jensen et al. (2020, 6833719) reported null associations between adiposity
and per each 1-unit increase in PFOA measured at 3 and 18 months. The low confidence study
by Gross et al. (2020, 7014743) reported a null association (OR = 0.91; 95% CI: 0.36 to 2.29) of
being overweight at 18 months for PFOA levels greater than the mean level. They showed
discordant sex-specific results with higher odds of being overweight at 18 months in males (OR
= 2.62; p-value = 0.22) and lower odds among females (OR = 0.41; p-value = 0.27).
Overall, there was very limited evidence of adverse associations between PFOA exposures and
either increased BMI or adiposity measures. Only one out of seven studies in the overall
population showed evidence of increased adiposity or BMI changes in infancy in relation to
PFOA. One of these studies did report increased odds of being overweight at 18 months for
higher PFOA levels in males only. Only one of two studies showed an inverse monotonic
relationship between either BMI or adiposity with increasing PFOA exposures.
3.4.4.1.5.2 Rapid Weight Gain
Five studies {Manzano-Salgado, 2017, 4238509; Shoaff, 2018, 4619944; Starling, 2019,
5412449; Tanner, 2020; Yeung, 2019, 5080619} examined rapid infant growth, with all five
considered high confidence. Limited evidence of associations was reported with these studies, as
only one {Starling et al., 2019, 5412449} of four studies {Manzano-Salgado, 2017, 4238509;
Shoaff, 2018, 4619944; Starling, 2019, 5412449; Yeung, 2019, 5080619} showed increased
odds of rapid weight gain with increasing PFOA. For example, Starling et al. (2019, 5412449)
reported small increased ORs (range: 1.25 to 1.43) for rapid growth in the overall population
based on either weight for age z-or weight for length-based z-scores. The most detailed
evaluation by Tanner et al. (2020, 6322293) also showed some adverse associations including
higher prenatal PFOA concentrations related to a longer duration of time needed to complete
90% of the infant growth spurt (Atertile 1: 0.06; 95% CI: 0.01, 0.11). Higher prenatal PFOA
concentrations were also significantly related to delayed infant peak growth velocity (81: 0.58;
95% CI: 0.17, 0.99) and a higher post-spurt weight plateau (al: 0.81; 95% CI: 0.21, 1.41).
3.4.4.1.5.3 Postnatal Growth Summary
Seven of the nine studies reporting quantitative results for different infant weight measures
showed some evidence of adverse associations with PFOA exposures, with two of these studies
3-225
-------
DRAFT FOR PUBLIC COMMENT
March 2023
showing adverse results predominately in females and one in males only. Two other studies
showed increased weight among males only and lack of reporting of effect estimates in one study
precluded further consideration of adversity. Two {Manzano-Salgado, 2017, 4238509; Starling,
2019, 5412449} of three studies did not report adverse associations in either the overall
population and females, but did detect increased infant weight measures among males. Three of
the seven studies reporting quantitative results showed some evidence of adverse associations
between PFOA exposures and infant height. Only one out of seven studies in the overall
population showed evidence of increased adiposity or BMI changes in infancy in relation to
PFOA. One study showed increased adiposity amongst males only, while four studies each were
null or reported some inverse associations (i.e., lower adiposity/BMI with increasing PFOA).
Two of the studies showed exposure-response relationships for PFOA and decreased BMI
scores, while a third showed the opposite exposure-response for increased adiposity. Although
the data across different endpoints was not entirely consistent, the majority of infant weight
studies indicated that PFOA may be associated with post-natal growth measures up to two years
of age.
3.4.4.1.6 Gestational Duration
Twenty-two different studies examined gestational duration measures (i.e., PTB or gestational
age measures) in relation to PFOA exposures. Nine of these studies examined both PTB and
gestational age measures, while two studies only examined PTB {Liu, 2020, 6833609; Gardener,
2021, 7021199}. Two of these studies were uninformative and not considered further below
{Gundacker, 2021, 10176483; Lee, 2013, 3859850}.
3.4.4.1.6.1 Gestational Age
Eighteen different informative studies examined the relationship between PFOA and gestational
age (in weeks). Seventeen of these examined associations in the overall population and one
study reported sex-specific findings only {Lind, 2017, 3858512}. Ten of these 18 studies were
high confidence {Bach, 2016, 3981534; Bell, 2018, 5041287; Buck Louis, 2018, 5016992; Chu,
2020, 6315711; Eick, 2020, 7102797; Huo, 2020, 6505752; Lauritzen, 2017, 3981410; Lind,
2017, 3858512; Manzano-Salgado, 2017, 4238465; Sagiv, 2018, 4238410}, four were medium
{Gyllenhammar, 2018, 4238300; Hjermitslev, 2020, 5880849; Meng, 2018, 4829851; Yang,
2022, 10176806} and four were low confidence {Gao, 2019, 5387135; Workman, 2019,
5387046; Wu, 2012, 2919186; Xu, 2019, 5381338}. Six of the studies had good study sensitivity
{Huo, 2020, 6505752; Lauritzen, 2017, 3981410; Manzano-Salgado, 2017, 4238465; Meng,
2018, 4829851; Sagiv, 2018, 4238410; Wu, 2012, 2919186}, nine were adequate {Bach, 2016,
3981534; Buck Louis, 2018, 5016992; Chu, 2020, 6315711; Eick, 2020, 7102797; Gao, 2019,
5387135; Gyllenhammar, 2018, 4238300; Hjermitslev, 2020, 5880849; Lind, 2017, 3858512;
Yang, 2022, 10176806} and three {Bell, 2018, 5041287; Workman, 2019, 5387046; Xu, 2019,
5381338} were deficient.
Five (3 low confidence and 1 each medium and high confidence) of the 18 studies showed some
evidence of increased gestational age {Bach, 2016, 3981534; Gao, 2019, 5387135; Hjermitslev,
2020, 5880849; Workman, 2019, 5387046; Xu, 2019, 5381338} in relation to PFOA while six
others were largely null {Bell, 2018, 5041287; Buck Louis, 2018, 5016992; Gyllenhammar,
2018, 4238300; Huo, 2020, 6505752; Manzano-Salgado, 2017, 4238465; Sagiv, 2018,
4238410}. The remaining seven studies showed some evidence of adverse impacts on gestational
age either in the overall population or either. The high confidence study by Lind et al. (2017,
3-226
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3858512) examined only sex-specific data and reported larger deficits in female (-0.21 cm; 95%
CI: -0.61, 0.19 per each In unit PFOA increase) than male neonates (-0.10 cm; 95% CI: -0.41,
0.21). Among the other six studies with results based on the overall population, three were high
confidence, two were medium, and one was low confidence. The low confidence study by Wu et
al. (2012, 2919186) study reported an extremely large difference (-2.3 weeks; 95% CI: -4.0,
-0.6) in gestational age per each loglO unit PFOA change. The medium confidence study by
Yang et al. {2022, 10176806} reported a larger (-1.04 weeks; 95% CI: -3.72, 1.63 per each
PFOA IQR increase) difference in gestational age among preterm births than among term births
(-0.38 weeks; 95% CI: -1.33, 0.57 per each PFOA IQR increase). The medium confidence study
by Meng et al. (2018, 4829851) reported statistically significant gestational age deficits (range:
-0.17 to -0.24 weeks) across all quartiles but no evidence of an exposure-response relationship.
The high confidence study by Lauritzen et al. (2017, 3981410) reported a slight decrease in the
overall population (-0.2 weeks; 95% CI: -0.34, 0.14). They also showed larger deficits in their
Swedish population (-0.3 weeks; 95% CI: -0.9, 0.3) which was predominately driven by results
among male neonates (-0.4 weeks; 95% CI: -1.2, 0.5). The high confidence study by Chu et al.
(2020, 6315711) showed larger deficits in the overall population (-0.21 weeks; 95% CI: -0.44,
0.02) which was driven by female neonates (-0.83 weeks; 95% CI: -0.53, -0.23). The high
confidence study by Eick et al. {2020, 7102797} reported decreased gestational age only among
tertile 2 only in the overall population (-0.29 weeks; 95% CI: -0.74, 0.17), males (-0.24 weeks;
95% CI: -0.91, 0.43) and females (-0.31 weeks; 95% CI: -0.95, 0.34) relative to tertile 1.
Overall, seven of the 18 studies showed some evidence of adverse impacts on gestational age.
Six of the seven studies were either medium or high confidence studies. Few patterns emerged
based on study confidence or other study characteristics. For example, three of the null studies
were rated as having good sensitivity, along with two studies with adequate and one with
deficient ratings. There was a preponderance of associations related to sample timing possibly
related to pregnancy hemodynamic influences on the PFOA biomarkers, as five of the seven
studies reporting adverse associations were sampled later in pregnancy (i.e., exclusively trimester
two or later).
3.4.4.1.6.2 Preterm Birth
As shown in Figure 3-62, eleven studies examined the relationship between PFOA and PTB; all
of the studies were either medium {Hjermitslev, 2020, 5880849; Liu, 2020, 6833609; Meng,
2018, 4829851; Yang 2022, 10176806} or high confidence {Bach, 2016, 3981534; Chu, 2020,
6315711; Eick, 2020, 7102797; Gardener, 2021, 7021199; Huo, 2020, 6835452; Manzano-
Salgado, 2017, 4238465; Sagiv, 2018, 4238410}. Nine of the eleven studies were prospective
birth cohort studies, and the two studies by Liu et al. (2020, 6833609) and Yang et al. {2022,
10176806} were case-control studies nested with prospective birth cohorts. Four studies had
maternal exposure measures that were sampled either during trimester one {Bach, 2016,
3981534; Manzano-Salgado, 2017, 4238465; Sagiv, 2018, 4238410} or trimester three
{Gardener, 2021, 7021199}. The high confidence study by Chu et al. {2020, 6315711} sampled
during the late third trimester or within three days of delivery. Four studies collected samples
across multiple trimesters {Eick, 2020, 7102797; Hjermitslev, 2020, 5880849; Huo, 2020,
6835452; Liu, 2020, 6833609}. The medium confidence study by Meng et al. (2018, 4829851)
pooled exposure data from two study populations, one which measured PFOA in umbilical cord
blood and one which measured PFOA in maternal blood samples collected in trimesters 1 and 2.
The medium confidence study by Yang et al. (2022, 10176806) collected umbilical cord blood
3-227
-------
DRAFT FOR PUBLIC COMMENT
March 2023
samples. Four studies {Huo, 2020, 6835452; Manzano-Salgado, 2017, 4238465; Meng, 2018,
4829851; Sagiv, 2018, 4238410} were considered to have good sensitivity and one was deficient
{Liu, 2020, 6833609). The other six studies were rated adequate in this domain. The median
exposure values across all studies ranged from 0.76 ng/mL {Eick, 2020, 7102797} to 11.85
ng/rnL (Huo, 2020, 6835452}.
e\eo^e^e^ ^°e
Bachetal., 2016, 3981534-
+
+
++
+
++
+
+
++
Chu etal., 2020, 6315711 -
B
++
B
+
++
+
+
++
Eick etal., 2020, 7102797-
++
++
n
+
++
+
+
++
Gardener etal., 2021, 7021199-
++
++
++
+
++
+
+
++
Hjermitslev et al., 2020, 5880849 -
+
+
+
+
+
-
+
M
Huo etal., 2020, 6835452-
++
++
++
+
++
+
Liu etal., 2020, 6833609-
++
B
B
-
++
+
-
n
Manzano-Salgado et al., 2017, 4238465 -
++
++
++
+
++
+
++
++
Meng etal., 2018, 4829851 -
+
+
++
+
++
+
++
B
Sagiv etal., 2018, 4238410-
++
+
+
+
++
+
++
++
Yang etal., 2022, 10176806-
+
+
+
+
+
+
+
j
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Figure 3-62. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Preterm Birth Effects
Interactive figure and additional study details available on HAWC.
3-228
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Adverse associations were reported with five of the eleven studies showing increased risk of
PTB with PFOA exposures. Null or inverse associations were reported by Bach et al. (2016,
3981534), Hjermitslev et al. (2019, 5880849), Liu et al. (2020, 6833609), Manzano-Salgado et
al. (2017, 4238465) and Yang et al. (2022, 10176806). The medium confidence study by Meng et
al. (2018, 4829851) reported consistently elevated non-monotonic ORs for PTB in the upper
three PFOA quartiles (OR range: 1.7-3.2), but little evidence was seen when examined per each
doubling of PFOA exposures (OR =1.1; 95% CI: 0.8, 1.5). Although they were not statistically
significant, the high confidence study by Chu et al. (2020, 6315711) reported increased ORs of
similar magnitude per each In ng/mL increase (OR = 1.49; 95% CI: 0.94, 2.36) and when quartile
3 (OR = 1.60; 95% CI: 0.60, 4.23) and quartile 4 (OR = 1.84; 95% CI: 0.72, 4.71) exposures
were compared to the referent. ORs similar in magnitude were detected in the high confidence
study by Eick et al. (2020, 7102797) study albeit in a more monotonic fashion across all
quantiles (tertile 2: OR = 1.48; 95% CI: 0.66, 3.31); 95%CI: tertile 3: OR = 1.63; 95% CI: 0.74,
3.59). Associations between PFOA and overall PTB near or just below the null value were
consistently detected for either categorical or continuous exposures in the high confidence Huo
et al. (2020, 6835452) study. Few patterns emerged across PTB subtypes in that study, although
there was an increase in clinically indicated PTBs (OR = 1.71; 95% CI: 0.80, 3.67 per each ln-
unit PFOA increase) which seemed to be largely driven by results in female neonates (OR =
2.64; 95%CI: 0.83, 8.39). The high confidence study by Sagiv et al. (2018, 4238410) showed
increased non-significant risks (OR range: 1.1-1.2) for PTB across all PFOA quartiles. Relative
to the referent, the high confidence study by Gardener {2021, 7021199} showed higher odds of
PTB in PFOA quartiles 2 and 3 (range: 3.1-3.2) than that found in quartile 4 (OR=1.38; 95% CI:
0.32-5.97). Outside of the aforementioned Eick et al. (2020, 7102797) study, none of the other
seven studies with categorical data showed evidence of exposure-response relationships.
Limited adverse associations were reported with only three of the eight studies consistently
showing increased risk of PTB with PFOA exposures. The Meng et al. (2018, 4829851) study
reported consistently elevated non-monotonic ORs for PTB in the upper three PFOA quartiles
(OR range: 1.7-3.2), but little evidence was seen when they were examined per each doubling of
PFOA exposure (OR =1.1; 95% CI: 0.8, 1.5). Although they were not statistically significant,
Chu et al. (2020, 6315711) reported increased ORs of similar magnitude per 1 In ng/mL unit
increase (OR = 1.49; 95% CI: 0.94, 2.36) or when quartile 3 (OR = 1.60; 95% CI: 0.60, 4.23)
and quartile 4 (OR = 1.84; 95% CI: 0.72, 4.71) exposures were compared to the referent.
Associations between PFOA and (overall) PTB near or just below the null value were
consistently detected in the Huo et al. (2020, 6835452) study. Few patterns emerged across PTB
subtypes, although there was an increase in clinically indicated PTBs per each ln-unit increase in
PFOA (OR = 1.71; 95% CI: 0.80, 3.67). The high confidence study by Sagiv et al. (2018,
4238410) showed increased non-significant risks (OR range: 1.1-1.2) for PTB across all PFOA
quartiles. Null or inverse associations were reported by Bach et al. (2016, 3981534), Hjermitslev
et al. (2019, 5880849), Liu et al. (2020, 6833609), and Manzano-Salgado et al. (2017, 4238465).
None of the six studies showed strong evidence of exposure-response relationships.
Overall, five of the eleven studies showed increased risk of PTB with PFOA exposures with no
evidence of exposure-response relationships. Although small numbers limited the confidence in
many of the sub-strata comparisons, there were few apparent patterns by study evaluation ratings
or other characteristics that explained the heterogeneous results across studies. However, there
were more associations amongst studies with later sample timing data collection, as three of the
3-229
-------
DRAFT FOR PUBLIC COMMENT
March 2023
five studies with later PFOA biomarker sampling showed some increased odds of preterm birth
compared to two of six studies with earlier sampling.
3.4.4.1.6.3 Gestational Duration Summary
Overall, there was mixed evidence of adverse associations between PFOA and both gestational
age and preterm birth. Most of the associations for either gestational duration measures were
reported in medium or high confidence studies. Few other patterns were evident that explained
any between study heterogeneity.
3.4.4.1.7 Fetal Loss
Five (2 high, 2 medium and 1 low confidence) studies examined PFOA exposure and fetal loss
with limited evidence as only one study showing increased risks of miscarriage. Two studies had
good study sensitivity {Wang, 2021, 10176703; Wikstrom, 2021, 7413606}, while three had
adequate sensitivity {Buck Louis, 2016, 3858527; Jensen, 2015, 2850253; Liew, 2020,
6387285} (Figure 3-63).
The high confidence study by Wikstrom et al. {2021, 7413606} showed a statistically significant
association between PFOA and miscarriages (OR = 1.48; 95% CI: 1.09, 2.01 per doubling of
PFOA exposures. The authors also reported a monotonic exposure-response relationship across
PFOA quartiles (ORs/95%CIs: Q2: 1.69; 0.8, 3.56; Q3: 2.02; 0.95, 4.29; Q4: 2.66; 1.26, 5.65).
The medium confidence study by Liew et al. {2020, 6387285} detected a 40% increased risk of
miscarriage (OR = 1.4; 95% CI: 1.0, 1.9) per each PFOA doubling with increased risks detected
for quartiles three (OR=1.4; 95% CI: 0.8, 2.6) and four (OR = 2.2; 95% CI: 1.2, 3.9) only. No
associations were detected in the high confidence study by Wang et al. {2021, 10176703} for
preclinical spontaneous abortion (OR = 0.99; 95% CI: 0.94, 1.05) or in the medium confidence
study by Buck Louis et al. {2016, 3858527} (hazard ratio (HR) =0.81; 95% CI: 0.65, 1.00 per
each SD PFOA increase). In the low confidence study by Jensen et al. {2015, 2850253}, a
decreased risk of miscarriages was reported (OR = 0.64; 95% CI: 0.36, 1.18 per each ln-unit
PFOA increase).
Overall, there was positive evidence for fetal loss with increased relative risk estimates in two
out of five studies. In those two studies, the magnitude of associations detected ranged from 1.4
to 2.7 with an exposure-response relationship detected in one study. No patterns in the results
were detected by study confidence ratings including sensitivity.
3-230
-------
DRAFT FOR PUBLIC COMMENT
March 2023
a\0^
ft*
,c®
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Buck Louis et al., 2016, 3858527 -
+
+
+
+
+
+
+
+
Jensen et al., 2015, 2850253 -
+
+
+
+
+
+
-
Liew et al., 2020, 6387285-
B
++
+
++
+
+
+
Wang et al., 2021, 10176703-
++ ++
++
+
++
+
++
++
Wikstrom et al., 2021, 7413606 -
++ ++
++
+
++
+
++
++
Figure 3-63. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Fetal Loss
Interactive figure and additional study details available on HAWC.
3.4.4.1.8 Birth Defects
Four birth defect studies examined PFOA exposure with three of these four having adequate
study sensitivity (one was deficient) as shown in Figure 3-64. This included a medium
confidence study by Vesterholm Jensen et al. (2014, 2850926) that reported no adverse
associations for cryptorchidism (OR = 0.83; 95% CI: 0.44, 1.58 per each ln-unit PFOA increase).
A medium confidence study by Ou et al. (2021, 7493134) reported decreased risks for septal
defects (OR = 0.54; 95% CI: 0.18, 1.62), conotruncal defects (OR = 0.28; 95%CI: 0.07, 1.10),
and total congenital heart defects (OR = 0.64; 95% CI: 0.34, 1.21) among participants with
maternal serum levels over >75th PFOA percentile (relative to those <75th percentile). A low
confidence study {Cao, 2018, 5080197} of a non-specific all birth defect grouping reported
limited evidence of an association (OR = 1.24; 95%CI: 0.57, 2.61), but interpretation of an all-
birth defect grouping is challenging given that etiological heterogeneity may occur across
individual defects. Compared to the referent group of no Little Hocking Water Association
supplied water, no associations (both ORs were 1.1) were reported in a low confidence study
from Washington County, Ohio among infants born to women partially or exclusively supplied
in part by the Little Hocking Water Association {Nolan, 2010, 1290813}. The study was
3-231
-------
DRAFT FOR PUBLIC COMMENT
March 2023
considered iminformative for examination of individual defects given the lack of consideration of
confounding and other limitations in those analyses.
Overall, there was negligible evidence of associations between PFOA and birth defects based on
the four available epidemiological studies including two medium confidence studies which
reported decreased odds of birth defects relative to exposures. As noted previously, there is
considerable uncertainty in interpreting results for broad any defect groupings which are
anticipated to have decreased sensitivity to detect associations.
sf>
Cao etal.,2018, 5080197-
l
l
+
l
-
1
+
l
+
l
+
j
Nolan etal., 2010, 1290813-
+
-
+
+
-
Ou etal., 2021, 7493134-
++
+
+
+
+
+
+
Vesterholm Jensen et al., 2014, 2850926-
+
+
+
++
+
+
+
Legend
s
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
B
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-64. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Defects
Interactive figure and additional study details available on HAWC.
3.4.4.2 Animal Evidence Study Quality Evaluation and Synthesis
There are 5 studies from the 2016 PFOAHESD {U.S. EPA, 2016, 3603279} and 13 studies from
recent systematic literature search and review efforts conducted after publication of the 2016
PFOA HESD that investigated the association between PFOA and developmental effects in
animal models. Study quality evaluations for these 18 studies are shown in Figure 3-65.
3-232
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Abbott etal., 2007, 1335452 -
+
+
+
+
++
++
++
8
-*
Blake et al., 2020, 6305864 -
+
+
++
+
+
+
++
D
++
+
Butenhoff et al., 2004, 1291063-
++
NR
NR
++
++
+
++
++
++
Chen etal., 2017, 3981369-
++
NR
%
B
++
+
1
D
++
+
Cope etal., 2021, 10176465-
+
+
++
++
+
+
++
+
Hu etal., 2010, 1332421 -
++
NR
NR
++
¦
+
++
a
++
+
Hu et al., 2012, 1937235-
+
+
NR
++
++
+
++
++
++
+
Jiang etal., 2020, 6320192-
-
NR
+
+
++
+
++
++
++
+
Lau etal., 2006, 1276159-
+
+
NR
+
+
++
++
++
+
+
Li etal., 2018, 5084746-
+
+
NR
++
+
+
+
++
+
Li etal., 2019, 5387402-
-
+
+*
+
-
+
-
-
-
Macon et al., 2011, 1276151 -
++
++
ft
+
+
+
+
+
+
NTP, 2020, 7330145-
++
++
NR
++
Salimi etal., 2019, 5381528^
4-
NR
NR
NR
+
+
+
+
-
-
Song etal., 2018, 5079725-
++
+
NR
++
+
+
+
++
+
+
Wolfetal., 2007, 1332672-
+
+~
NR
++
+
+
++
+
+
+
Zhang etal., 2021, 10176453-
++
+
NR
~
+
+
¦
+
+*
-
van Esterik et al., 2015, 2850288 -
++
NR
+*
++
+
¦
++
++
a
-
Legend
[ Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
B Critically deficient (metric) or Uninformative (overall)
Not reported
* Multiple judgments exist
Figure 3-65. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Developmental Effects
Interactive figure and additional study details available on HAWC.
Evidence suggests that PFOA exposure can adversely affect development. Oral studies in mice
and rats report effects in offspring including decreased survival, decreased body weights,
structural abnormalities (e.g., reduced skeletal ossification), delayed eye opening, and altered
3-233
-------
DRAFT FOR PUBLIC COMMENT
March 2023
mammary gland development. Doses that elicited responses were generally lower in mice than in
rats. Additionally, three studies of gestational PFOA exposure to mice reported effects on
placental weight and histopathological changes in placental tissue, suggesting that the placenta
may be a target of PFOA. In some cases, adverse developmental effects of PFOA exposure that
relate to other health outcomes may be discussed in the corresponding health outcome section
(e.g., neurodevelopmental effects are discussed in the Appendix; see PFOA Appendix).
3.4.4.2.1 Maternal Effects
Exposure to PFOA resulted in significant decreases in maternal body weight and/or weight gain
at doses >10 mg/kg/day in multiple strains of pregnant mice {Li, 2018, 5084746; Lau, 2006,
1276159; Yahia, 2010, 1332451} and at doses > 30 mg/kg/day in pregnant Sprague Dawley rats
{Butenhoff, 2004, 1291063; Hinderliter, 2005, 1332671}. The effect followed a dose-related
trend in some studies. PFOA exposure was also associated with significantly delayed parturition
at doses > 3 mg/kg/day in CD-I mice {Lau, 2006, 1276159} and at 10 mg/kg/day in ICR mice
{Yahia, 2010, 1332451}.
3.4.4.2.1.1 Studies in Mice
Li et al. (2018, 5084746) reported marked, dose-related decreases in maternal body weight gain
at > 10 mg/kg/day in pregnant Kunming mice exposed from gestation day 1 to 17 (GD 1 to GD
17; no statistical tests performed). Dose-related decreases in body weight gain were also seen in
pregnant CD-I mice exposed to 10, 20, or 40 mg/kg/day (significant at 20 and 40 mg/kg/day) by
Lau et al. (2006, 1276159); significantly delayed time to parturition was also seen at 3, 10, and
20 mg/kg/day in this study (all litters at 40 mg/kg/day were resorbed). Yahia et al. (2010,
1332451) dosed pregnant ICR mice with 0, 1, 5, or 10 mg/kg/day from GD 0 to GD 17
(sacrificed on GD 18) or GD 0 to GD 18 (allowed to give birth), and at 10 mg/kg/day, observed
significant decreases in body weight gain from GD 12 onward in dams allowed to give birth as
well as significantly decreased terminal body weight in dams sacrificed on GD 18. In the same
study, a significant decrease in food intake during early gestation was also reported for the dams
allowed to give birth, but data were not shown. Delayed parturition was also observed at
10 mg/kg/day (data not shown). Pregnant CD-I mice exposed to 25 mg/kg/day from GD 11 to
GD 16 exhibited significantly decreased body weight from GD 13 to GD 16 {Suh, 2011,
1402560}. Hu et al. (2010, 1332421) exposed pregnant C57BL/6N mouse dams to 0.5 or
1.0 mg/kg/day PFOA and found no significant differences relative to controls on GD 19. No
significant effects on maternal body weight were noted in C57BL/6N mouse dams exposed to
0.02, 0.2, or 2 mg/kg/day PFOA from time of mating through PND 21 (Hu, 2012, 1937235). In
contrast to the above-described findings, two studies in pregnant CD-I mice reported
significantly increased maternal body weight gain after exposure to 5 mg/kg/day {Blake, 2020,
6305864} or 3 or 5 mg/kg/day PFOA {Wolf, 2007, 1332672} from GD 1-17. Abbott et al.
(2007, 1335452) found no effects of 0.1, 0.3, 0.6, or 1 mg/kg/day PFOA on maternal weight
changes in 129Sl/SvlmJ wild-type mice (exposure to 5, 10, and 20 mg/kg/day PFOA led to
increased maternal death) (Figure 3-66).
3.4.4.2.1.2 Studies in Rats
A two-generation oral gavage reproductive toxicity study in Sprague-Dawley rats reported no
effect on parental generation (Po) maternal body weight or food consumption, but found
significantly decreased body weight in first-generation (Fi) parental females at 30 mg/kg/day
during pre-cohabitation, gestation (GD 0-GD 14), and lactation day 5 to 15 (LD 5-LD 15).
3-234
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Decreased absolute food consumption was reported, but data were not shown; relative feed
consumption was unaffected {Butenhoff, 2004, 1291063}. In pregnant Sprague-Dawley rats
dosed with 30 mg/kg/day from GD 4 to LD 21, body weight gain was decreased during gestation
and body weight was 4% lower than controls during lactation (statistical significance not
indicated) {Hinderliter, 2005, 1332671}.
In a two-year chronic toxicity/carcinogenicity assay conducted by the NTP (2020, 7330145),
female Sprague Dawley (Hsd:Sprague Dawley® SD®) rat dams were exposed to 0, 150, or
300 parts per million (ppm) PFOA in feed during the perinatal period. In study 1, Fi male rats
were administered 0, 150, or 300 ppm PFOA and Fi female rats were administered 0, 300, or
1,000 ppm PFOA in feed during the postweaning period. For study 2, lower postweaning
exposure levels (0, 20, 40, or 80 ppm) were utilized for males due to unexpected toxicity in male
offspring using the original exposure regime. Exposure for all Fi generations in both studies
occurred for 107 weeks or until the 16-week interim necropsy. The perinatal and postweaning
exposure regimes for females and males for both studies are presented in Table 3-9. Dose groups
for this study are referred to as "[perinatal exposure level]/[postweaning exposure level]" (e.g.
300/100).
Table 3-9. Study Design for Perinatal and Postweaning Exposure Levels for Fi Male and
Female Rats for the NTP (2020, 7330145) Study
Perinatal
Postweaning Dose
Dose
0 ppm
20 ppm
40 ppm 80 ppm
150 ppm
300 ppm
1,000 ppm
Study 1 Females
0 ppm
X
-
-
-
X
X
150 ppm
-
-
-
-
X
300 ppm
-
-
-
-
-
X
Study 1 Males
0 ppm
X
-
-
X
X
-
150 ppm
-
-
-
X
-
300 ppm
-
-
-
-
X
-
Study 2 Males
0 ppm
X
X
X X
-
-
-
300 ppm
X
X
X X
-
-
-
Notes: Fi = first generation; X = exposure level used.
In pregnant Sprague-Dawley rats exposed to 150 or 300 ppm via diet (equivalent to
approximately 11 and 22 mg/kg/day during gestation and 22 and 45 mg/kg/day from LD 1 to LD
14), no consistent effects were observed on body weight or body weight gain during gestation or
lactation (Figure 3-66). Food consumption was marginally but significantly decreased (up to 4%)
at one or both dose levels at various intervals. In a repeat of this study that tested a single dose
level of 300 ppm (approximately 21.8 mg/kg/day during gestation and 48.3 mg/kg/day from LD
1 to LD 14), no effects were observed on maternal body weight or body weight gain during
gestation; from LD 1 to LD 14, there was a marginal but significant decrease (2%-3%) in
maternal body weight and body weight gain and a significant decrease (5%) in food consumption
{NTP, 2020, 7330145}.
3-235
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Study Name
Study Design
Observation Time
Animal Description
Maternal Body Weight
Blake el al.. 2020, 6305864
developmental (GD1.5-11.5)
GD11.5
P0 Mouse. CD-1 (y, N=11)
developmental (GD1.5-17.5)
GD17.5
PO Mouse, CD-1 (9. N=11)
Abbott et al., 2007, 1335452
developmental (GD1-17)
GD17
PO Mouse, 129S1/SvlmJ (N=0-21)
P0 Mouse, 129S4/SvJae PPARa null (y. N=1-19)
Wolfet al., 2007,1332672
developmental (GD1-17)
GD17
P0 Mouse, CD-1 (9, N=25-39)
developmental (GD15-17)
GD17
P0 Mouse, CD-1 (y, N=4-10)
Li et al., 2018, 5084746
developmental (GD1-17)
GD18
PO Mouse, Kunming N=10)
Hu et al., 2010, 1332421
developmental (GD6-17)
GD19
PO Mouse, C57BU6n {9, N=16)
Hu et al.. 2012. 1937235
developmental {14d mating-PND21)
PND21
P0 Mouse, C57BI6/N (9, N=24)
Butenhoffetal.. 2004,1291063
reproductive (84d)
LD22
P0 Rat, Crl:CD{SD)IGS BR (J. N=26-29)
reproductive (GD1-PND106)
GD14
F1 Rat, Crl:CD(SD)IGS BR (•". N=28-29)
LD15
F1 Rat, Crl:CD(SD)IGS BR ( ', N=28-29)
NTP. 2020, 7330145
chronic (GD6-PND21)
GD21
PO Rat. Sprague-Dawley N=30-91)
LD21
P0 Rat, Sprague-Dawley (^. N=30-86)
Maternal Body Weight Change
Blake etal.. 2020. 6305864
developmental {GD1.5-11.5)
GD0.5-11.5
P0 Mouse, CD-1 (9. N=11)
developmental (GD1.5-17.5)
GDO.5-17.5
P0 Mouse. CD-1 (y, N=11)
Wolfet al.. 2007, 1332672
developmental (GD1-17)
GD1-17
PO Mouse, CD-1 (9, N=25-39)
developmental (GD13-17)
GD1-17
PO Mouse, CD-1 (9, N=8-12)
developmental {GD 15-17)
GD1-17
P0 Mouse, CD-1 (9. N=4-10)
Abbott et al., 2007,1335452
developmental (GD1-17)
GD1-17
P0 Mouse. 129S1/SvlmJ (». N=0-21)
PO Mouse. 129S4/SvJae PPARa null (y, N=1-19)
Lau et al.. 2006, 1276159
developmental (GD1-17)
GD18
PO Mouse. CD-1 (y, N=9-45)
NTP, 2020, 7330145
chronic (GD6-PND21)
GD6-21
PO Rat. Sprague-Dawley ( • , N=30-91)
LD1-21
P0 Rat, Sprague-Dawley U'. N=30-86)
PFOA Developmental Effects - Maternal Body Weight
) No significant change A Significant
Y Significant
V V V
A
Concentration (mg/kg/day)
Figure 3-66. Maternal Body Weight in Rodents Following Exposure to PFOA (logarithmic
scale)
PFOA concentration is presented in logarithmic scale to optimize the spatial presentation of data. Interactive figure and additional
study details available on HAWC.
GD= gestation day; PND = postnatal day; LD = lactation day; Po = parental generation; Fi = first generation.
3.4.4.2.2 Placenta Effects
Two oral gavage studies in CD-I mice reported significant decreases in embryo to placenta
weight ratio at 5 mg/kg/day PFOA {Blake, 2020, 6305864} or doses > 2 mg/kg/day {Suh, 2011,
1402560}, as well as treatment-related histopathological lesions at 5 mg/kg/day {Blake, 2020,
6305864} or doses > 10 mg/kg/day {Suh, 2011, 1402560}. A third study in Kunming mice
reported decreased placenta to body weight ratio at PFOA doses > 5 mg/kg/day and
histopathological changes in placental tissue at doses > 2.5 mg/kg/day {Jiang, 2020, 6320192}
(Figure 3-67).
Blake et al. (2020, 6305864) administered 0, 1, or 5 mg/kg/day to pregnant CD-I mice from GD
1.5 through sacrifice on GD 11.5 or GD 17.5, Suh et al. (2011, 1402560) administered 0, 2, 10,
or 25 mg/kg/day to CD-I mice from GD 11 through sacrifice on GD 16, and Jiang et al. (2020,
6320192) administered 0, 2.5, 5, or 10 mg/kg/day to Kunming mice from GD 1 through sacrifice
on GD 13. The embryo to placental weight ratio was significantly decreased at 5 mg/kg/day in
Blake et al. (2020, 6305864) and at doses > 2 mg/kg/day in Suh et al. (2011, 1402560). Blake et
al. (2020, 6305864) observed significantly increased placental weight at 5 mg/kg/day at GD 17.5
and no changes in the numbers of viable fetuses or resorptions, whereas Suh et al. (2011,
1402560) observed significantly decreased placental weight and increased numbers of
resorptions and dead fetuses at > 2 mg/kg/day. Jiang et al. (2020, 6320192) observed
significantly decreased relative placental weight at > 5 mg/kg/day (decreases were also seen at
lower dose levels, but they did not reach statistical significance). Histopathological changes in
placental tissue were also observed at PFOA doses > 2.5 mg/kg/day (increased area of
spongiotrophoblast, decreased blood sinusoidal area in labyrinth), > 5 mg/kg/day (increased
3-236
-------
DRAFT FOR PUBLIC COMMENT
March 2023
interstitial edema of spongiotrophoblast), or 10 mg/kg/day (decreased labyrinth area, increased
ratio of spongiotrophoblast to labyrinth area). Jiang et al. (2020, 6320192) found no effect on
fetus to maternal body weight ratio. Viable fetus weight was significantly decreased in Blake et
al. (2020, 6305864) at 5 mg/kg/day and in Suh et al. (2011, 1402560) at > 10 mg/kg/day and
corresponded with treatment-related lesions in the placenta. The incidence of GD 17.5 placentas
within normal limits was significantly lower in mice exposed to 5 mg/kg/day {Blake, 2020,
6305864}, and the lesions observed in placentas from that group included labyrinth atrophy
(3/40 placentas), labyrinth congestion (23/40), and early fibrin clot (1/40). In dams treated with
1 mg/kg/day, labyrinth necrosis was observed in 1/32 placentas and placental nodules were
observed in 2/32 placentas. Histopathologic examination by Suh et al. (2011, 1402560) showed
normal placental tissue in 0 and 2 mg/kg/day groups and dose-dependent necrotic changes in
placentas from the 10 and 25 mg/kg/day groups (incidences of specific lesions and statistical
significance not reported).
PFOA Developmental KITects - Placental Weight
F.ndpoint
Study Name
Study Design
Observation Time
Animal Description
^ No significant change A Significant increase v Significant decrease
1
Embryo:Placenta Weight Ratio
Blake et al., 2020,6305864
developmental (GDI .5-11.5)
-
PO Mouse, CD-1 (9, N=62)
GUI 1.5
developmental (GD1.5-I7.5)
GDI 7.5
P0 Mouse, CD-I (9, N=62)
Placenta Weight, Absolute
Blake et al.. 2020,6305864
developmental (GD1.5-11.5)
<• -
P0 Mouse, CD-I (9. N=62)
developmental (GDI.5-17.5)
.... .
P0 Mouse, CD-I (9. N=62)
OD17.5
Placenta Weight. Relative
Jiang et al.. 2020,6320192
developmental (GD1-13)
....
P0 Mouse, Kunming (9. N=6)
Placenta and F.mbryo Weight, Relative
Jiang et al., 2020, 6320192
developmental (GDI-13)
P0 Mouse, Kunming (9, N=6)
1
2
3 4 5 6 7 8 9
Concentration (mg/kg/day)
10
Figure 3-67. Placental Weights in Mice Following Exposure to PFOA
Interactive figure and additional study details available on HAWC.
GD = gestation day; Po = parental generation.
3.4.4.2.3 Offspring Mortality
Studies of oral PFOA exposure in mice reported significant increases in resorptions and dead
fetuses with PFOA dose levels as low as 2 mg/kg/day in prenatal evaluations {Li, 2018,
5084746; Suh, 2011, 1402560; Lau, 2006, 1276159}. Stillbirths, pup mortality, and total litter
loss were observed in several strains of mice at doses > 5 mg/kg/day {Lau, 2006, 1276159; Song,
2018, 5079725; White, 2011, 1276150; Wolf, 2007, 1332672; Yahia, 2010, 1332451}; increased
litter loss was seen as low as 0.6 mg/kg/day PFOA in one study in 129Sl/SvImJ mice {Abbott,
2007, 1335452}. Comparatively, rat pup mortality (pre- and post-weaning) was reported at a
higher dose of 30 mg/kg/day {Butenhoff, 2004, 1291063}. Maternal effects observed in some of
these studies were not sufficient to explain effects observed in the offspring, as some studies
reported effects on offspring survival at dose levels that did not produce maternal effects.
3.4.4.2.3.1 Mice, Prenatal Evaluations
In two studies of gestational PFOA exposure in pregnant Kunming mice, Li et al. (2018,
5084746) reported significantly decreased GD 18 fetal survival at 10 and 20 mg/kg/day and total
fetal resorption at 40 mg/kg/day (fetal survival was also decreased at 5 mg/kg/day, but the effect
did not reach statistical significance), and Chen et al. (2017, 3981369) reported a significant
increase in the number of resorbed fetuses at GD 13, but not GD 7, after exposure to
10 mg/kg/day PFOA beginning on GD 1 (there were no effects on the number of implantation
sites). Suh et al. (2011, 1402560) exposed pregnant CD-I mice to 0, 2, 10, or 25 mg/kg/day from
GD 11 to GD 16 (dams were sacrificed on GDI6) and observed significant increases in the
3-237
-------
DRAFT FOR PUBLIC COMMENT
March 2023
number of resorptions and dead fetuses at all dose levels; post-implantation loss was 3.87%,
8.83%, 30.98%), and 55.41% at 0, 2, 10, and 25 mg/kg/day, respectively. In pregnant CD-I mice
exposed from GD 1 to GD 17, Lau et al. (2006, 1276159) reported significant increases in the
number of full-litter resorptions at PFOA doses > 5 mg/kg/day, with complete loss of all
pregnancies at the high dose of 40 mg/kg/day (no effect was observed on the number of
implantation sites in litters that were fully resorbed). At 20 mg/kg/day, a significant increase in
the percentage of prenatal loss per live litter was observed. White et al. (2011, 1276150) reported
significantly fewer implants in Fi-generation CD-I mouse dams that had been exposed to
5 mg/kg/day PFOA.
3.4.4.2.3.2 Mice, Postnatal Evaluations
Wolf et al. (2007, 1332672) reported a significant increase in total litter loss following oral
PFOA exposure of pregnant CD-I mice to 5 mg/kg/day (no effect on the number of implantation
sites). In offspring exposed to 5 mg/kg/day PFOA in utero and throughout lactation, significantly
decreased pup survival was observed from postnatal day (PND) 4 to 22; this effect was not seen
in cross-fostered offspring exposed during gestation only or during lactation only. In a separate
study, these authors exposed pregnant CD-I mice to 5 mg/kg/day PFOA for different lengths of
time (GD 7-GD 17, GD 10-GD 17, GD 13-GD 17, or GD 15-GD 17) and to 20 mg/kg/day
from GD 15-17. Control mice received deionized water from GD 7 to GD 17. Although
gestational PFOA exposure from GD 1 to GD 6 was not required to elicit adverse developmental
responses in pups, the severity of postnatal responses, including decreased pup weight during
lactation and delayed eye opening, increased with earlier and longer exposure durations (i.e., GD
7-GD 17 exposure resulted in more severe decreases in pup body weight when compared to pups
exposed from GD 15 to GD 17). The authors could not attribute the observed adverse effects to a
sensitive window of development as the pups exposed for longer durations had higher serum
PFOA levels than pups exposed for shorter durations. Notably, significantly decreased offspring
survival was observed in pups exposed to 20 mg/kg/day with the shortest exposure duration from
GD 15 to GD 17.
Lau et al. (2006, 1276159) reported significant increases in the incidence of stillbirths and pup
mortality at 5, 10, and 20 mg/kg/day PFOA in CD-I mice exposed from GD 1 to GD 18 and
allowed to deliver naturally. Complete loss of all pregnancies was observed at the high dose of
40 mg/kg/day, though there were no effects on the number of implantation sites. At 10 and
20 mg/kg/day, most of the pups died on PND 1. After exposure of pregnant Kunming mice to 1,
2.5, or 5 mg/kg/day from GD 1 to GD 17, Song et al. (2018, 5079725) reported a significant
decrease in the number of surviving pups per litter on PND 7, 14, and 21 at 5 mg/kg/day (a dose-
related trend was observed, but statistical significance was achieved only at the high dose). Yahia
et al. (2010, 1332451) dosed pregnant ICR mice with 0, 1, 5, or 10 mg/kg/day PFOA from GD 0
to GD 18, and the dams were allowed to give birth naturally. Approximately 58%> of pups born
to high-dose dams were stillborn, and the remaining pups died within 6 hours of birth. Mean
PND 4 survival rate was 98%>, 100%), 84.4%>, and 0%> at 0, 1, 5, and 10 mg/kg/day, respectively
(with significant decreases at 5 and 10 mg/kg/day). In the same study, some of the pregnant mice
were exposed to the same dose levels from GD 0 to GD 17 and sacrificed on GD 18, and the
number of live GD 18 fetuses from these dams was not significantly affected at any dose level.
White et al. (2011, 1276150) conducted a multi-generational study and dosed pregnant CD-I
mice with 0, 1, or 5 mg/kg/day from GD 1 to GD 17. Exposure to 5 mg/kg/day significantly
increased prenatal loss, significantly decreased the number of live pups born, and significantly
3-238
-------
DRAFT FOR PUBLIC COMMENT
March 2023
reduced postnatal survival. In adult female Fi animals, no effects were observed on the prenatal
loss or postnatal pup survival of the second generation (F2) offspring.
Abbott et al. (2007, 1335452) exposed pregnant 129Sl/SvImJ wild-type and PPARa-null mice
from GD 1 to GD 17 to dose levels ranging from 0.1 to 20 mg/kg/day and allowed the mice to
deliver naturally. There were no treatment-related effects on the number of implantation sites,
but wild-type dams exposed to > 0.6 mg/kg/day PFOA and PPARa-null dams exposed to
> 5 mg/kg/day PFOA had significantly increased litter loss compared to their respective controls.
At doses > 5 mg/kg/day in wild-type dams and 20 mg/kg/day in PPARa-null dams, 100% litter
loss occurred. The percentage of dams with full litter resorptions significantly increased in the 5,
10, and 20 mg/kg/day groups, with 100% full litter resorption in the 20 mg/kg/day group. When
excluding dams with full litter resorptions, wild-type dams exposed to 1 mg/kg/day had a
significant increase in litter loss. Pup survival from birth to weaning was significantly decreased
in wild-type litters exposed to PFOA doses > 0.6 mg/kg/day. No effect was seen in PPARa-null
litters. Survival was significantly decreased for wild-type and heterozygous pups born to wild-
type dams dosed with 1 mg/kg/day and for heterozygous pups born to PPARa-null dams dosed
with 3 mg/kg/day. In the wild-type mice, the number of live and dead pups per litter were not
affected by PFOA. Similarly, the number of pups per litter in CD-I mice exposed to 0.1 or 1
mg/kg/day PFOA from GD 1.5-17.5 did not significantly differ from control groups {Cope,
2021, 10176465}.
3.4.4.2.3.3 Rats, Postnatal Evaluations
The NTP two-year carcinogenicity studies in Sprague-Dawley rats found no effects on offspring
survival {NTP, 2020, 7330145}, but Butenhoff et al. (2004, 1291063) reported an increase in the
total number of dead Fi rat pups during lactation (26/388 deaths at 30 mg/kg/day and 10/397 in
the control group; statistically significant only on LD 6-LD 8) and a significant increase in Fi
female pup deaths with 30 mg/kg/day on post-weaning days 2-8. F2 generation pup survival was
unaffected. In pregnant Sprague-Dawley rats dosed with 0, 3, 10, or 30 mg/kg/day from GD 4 to
LD 21, one dam at 3 mg/kg/day and two dams at 30 mg/kg/day delivered small litters (3-6
pups/litter compared to 12-19 pups/litter in the control group); however, statistical significance
was not indicated, and given the small sample size (5 dams/group), the biological significance of
this finding is unclear {Hinderliter, 2005, 1332671} (Figure 3-68).
3-239
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA Developmental Effects - Offspring Mortality
Endpoint
Study Name
Study Design
Observation Time
Animal Description
# No significant changed
Significant increase y
Significant decrease
Full Litter Resorption (%)
Li et al.. 2018.5084746
developmental (GD1-17)
GD18
P0 Mouse. Kunming (+'. N=10)
Lau et al., 2006, 1276159
developmental (GD1-17)
PO Mouse. CD-1 (y, N=9-45)
Wolf eta I.. 2007,1332672
developmental (GD1-17)
PO Mouse, CD-1 (y. N=25-39)
Kcsorptions, Moan. Linor
aeveiopmenia (isui.0-1 /.o)
P0 Mouse CD-1 (y N=11)
Resorptions. Early
Chenetal.. 2017.3981369
developmental (GD1-7)
P0 Mouse. Kunming (+. N=6)
Resorptions, Lale
Chen el al., 2017, 3981369
developmental (GD1-13)
GD13
PO Mouse, Kunming (y. N=6)
Resorptions. Percent Dams
Abbott et al,, 2007,1335452
developmental (GD1-17)
GD17
PO Mouse. 129S1/SvlmJ ( v. N=5-22)
¦ A
PO Mouse. 129S4/SvJae PPARa null (9, N=4-23)
it ¦ A
A
Prenatal Loss (% per Live Litter)
Lau et al.. 2006, 1276159
developmental (GD1-17)
GD18
PO Mouse. CD-1 (y. N=5-42)
A
Fetal Survival
Li ot al., 2018, 5084746
developmental (GD1-17)
GD18
PO Mouse. Kunming {'j\ N=10)
V7
T7
1
7
hi 11 mm 11
pn m rn 1 /• • w-n\
aeveiopmeniai (L>m .o-n.oj
developmental (GD1.5-17.5)
GD17.5
P0 Mouse. CD-1 (y. N=11)
Fetuses, Live (No. per Live Litter)
Lau et al., 2006, 1276159
developmental (GD1-17)
GD18
PO Mouse. CD-1 ('?, N=5-42)
Stillborn Pups. Mean,'Litter
Butenhoff et al,. 2004,1291063
reproductive <84d)
PNrvi
F1 Rat, Crl:CD(SD)IGS BR N=27-29)
reproductive (GD1-PND10B)
F2 Rat, Crt:CD(SD)lGS BR (J , N=29-30)
Dams with Whole Litter Loss (%)
Wolf el al., 2007, 1332672
developmental (GD1-17)
PND1
P0 Mouse. CD-1 (y, N=28~48>
Litter Loss (% per Live Litter)
Wolf et al,. 2007,1332672
developmental (GD1-17)
PND1
PO Mouse. CD-1 ('y. N=24-38)
Litter Loss (%)
Abbott et al., 2007, 1335452
developmental (GD1-17)
PO Mouse. 129Sl/SvlmJ (y, N=5-22)
P0 Mouse. 129S4.'SvJae PPARo null (y. N=4-23)
Wolf et al , 2007, 1332672
developmental (GD15-17)
PO Mouse. CD-1 (9, N=4-10)
Live Pups Born (No. per Live Litter)
Wolf el al.. 2007, 1332672
developmental (GD1-17)
PND1
P0 Mouse. CD-1 (y, N=25-39)
Offspring Survival
Song et al., 2018, 5079725
developmental (GD1-17)
PND0
F1 Mouse, Kunming N=10)
PND21
F1 Mouse, Kunming (•-';\ N=10)
Lau et al.. 2006,1276159
developmental (GD1-18)
PND0
F1 Mouse, CD-1 N=8)
• • "
V
Y
PND22
F1 Mouse, CD-1 , N=8)
• » T
w
V
Wolf et al.. 2007, 1332672
developmental (GD1-17)
PND4
F1 Mouse, CD-1 («S)
developmental (PND1-22)
PND4
F1 Mouse, CD-1 («••)
PND22
rn if
PND22
PND1-22
F1 Mouse CD-1 ('* N=7-13)
eve opmen
PND1-22
. I . ,rni, 17.
PND1-22
F1 Mouse CD-1 (¦ N=7-10)
developmental (GD15-17)
F1 Mouse, CD-1 N=3-10)
Pre-Weaning Mortality (%)
Butenhoff et al.. 2004,1291063
reproductive (84d)
F1 Rat, Cr1:CD(SD)IGS BR N=27-29)
Viability Index
Butenhoff et al.. 2004, 1291063
reproductive (84d)
PND5
F1 Rat, Cr1:CD(SD)IGS BR " , N=27-29)
reproductive (GD1-PND106)
PND5
F2 Rat. Crl:CD(SD)IGS BR (-Ti2. N=29-30)
NTP, 2020, 7330145
chronic (GD6-PND21)
F1 Rat, Sprague-Dawley ( , N=31-90)
Lactation Index
Butenhoff etai.. 2004. 1291063
reproductive <84d)
PND22
F1 Rat. Crl:CD(SD)IGS BR ( 5 mg/kg/day and postnatal evaluations
at dose levels as low as 0.5 mg/kg/day {Abbott, 2007, 1335452; Blake, 2020, 6305864; Hu,
2012, 1937235; Lau, 2006, 1276159; Li, 2018, 5084746; Suh, 2011, 1402560; Tucker, 2015,
2851046; White, 2011, 1276150; Wolf, 2007, 1332672; Yahia, 2010, 1332451; Hu, 2010,
1332421}. Offspring weight deficits in pups were observed to extend beyond weaning in three
studies in CD-I mice (at 1, > 3, and 5 mg/kg/day, respectively) {Tucker, 2015, 2851046; Lau,
2006, 1276159; White, 2011, 1276150} and in a multi-generation rat study at doses of
30 mg/kg/day {Butenhoff, 2004, 1291063}. In some studies, decreased fetal and/or pup body
weight was observed in the absence of maternal body weight effects.
3-240
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.4.2.4.1 Mice, Prenatal Evaluations
Blake et al. (2020, 6305864) reported significantly decreased GD 17.5 fetal weight with
5 mg/kg/day PFOA following gestational exposure in CD-I mice, despite significantly increased
maternal body weight gain. Lau et al. (2006, 1276159) reported a significant decrease in GD 18
fetal body weights after gestational exposure of CD-I mice to 20 mg/kg/day PFOA. In pregnant
Kunming mice, gestational exposure was associated with significantly decreased GD 18 fetal
weights at 5-40 mg/kg/day {Li, 2018, 5084746}. Suh et al. (2011, 1402560) reported a
significant decrease in GD 16 fetal weights at doses >10 mg/kg/day after exposure of pregnant
CD-I mice to 0, 2, 10, or 25 mg/kg/day from GD 11 to GD 16. Body weights of GD 18 ICR
mouse fetuses were significantly decreased following gestational exposure to 5 or 10 mg/kg/day
PFOA {Yahia, 2010, 1332451}.
3.4.4.2.4.2 Mice, Postnatal Evaluations
Wolf et al. (2007, 1332672) reported that CD-I mouse pup body weights were significantly
decreased after gestational exposure to 5 mg/kg/day PFOA from GD 1 to GD 17. The authors
also exposed pregnant mice to 20 mg/kg/day from GD 15 to GD 17 and to 5 mg/kg/day for
different lengths of time (GD 7-GD 17, GD 10-GD 17, GD 13-GD 17, or GD 15-GD 17). After
exposure to 5 mg/kg/day from GD 7 to GD 17 or GD 10 to GD 17 and to 20 mg/kg/day from GD
15 to GD 17, male pup body weights were significantly decreased. Additionally, with
5 mg/kg/day PFOA, male and female pup body weights were significantly decreased throughout
lactation in all exposure groups, and the magnitude of the effect increased with increasing
number of exposure days. Body weight deficits in male pups that had been exposed from GD 7
to GD 17 or GD 10 to GD 17 persisted for 10-11 weeks.
Hu et al. (2010, 1332421) exposed C57BL/6N pregnant mice with 0.5 or 1.0 mg/kg/day PFOA in
drinking water from GD 6 through GD 17. At PND 2, litter weights were significantly reduced in
the PFOA treatment groups (7%—12% less than the controls). At PND 7 and 14, the
0.5 mg/kg/day group litter weight was equivalent to the controls, but the 1.0 mg/kg/day group
was still significantly less than the controls (14% and 5%, respectively, by time point).
Body weights of live pups born to pregnant ICR mice dosed with 5 or 10 mg/kg/day during
gestation were significantly reduced {Yahia, 2010, 1332451}. At> 3 mg/kg/day, a dose-related
trend in growth retardation (body weight reductions of 25%-30%) was observed in neonates at
weaning; body weights reached control levels by 6 weeks of age for females and by 13 weeks of
age for males {Lau, 2006, 1276159}. Exposure of pregnant C57BL/6N mice to 2 mg/kg/day
from mating through lactation resulted in significantly decreased pup weights (32.6% lower than
controls, on average) from PND 1 to PND 21 (there were no effects on maternal body weights)
{Hu, 2012, 1937235}. Song et al. (2018, 5079725) observed significantly increased body
weights in PND 21 male offspring after gestational exposure to 2.5 or 5 mg/kg/day PFOA
(female data not provided). However, the authors did not report controlling for litter size in this
study; the significantly decreased litter size in the 5 mg/kg/day group could potentially result in
increased body weight in those pups due to reduced competition for maternal resources.
In a study in which pregnant 129Sl/SvImJ wild-type and PPARa-null mice were orally exposed
from GD 1 to GD 17 to dose levels ranging from 0.1 to 20 mg/kg/day {Abbott, 2007, 1335452},
decreased offspring body weight was seen in wild-type mice at 1 mg/kg/day (highest dose level
at which this effect was measured due to extensive litter loss at higher doses) beginning around
3-241
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PND 6, and this effect achieved statistical significance on PND 9, PND 10, and PND 22 (males)
and PND 7-PND 10 and PND 22 (females). No effects were observed on PPARa-null offspring
body weights. White et al. (2011, 1276150) exposed pregnant CD-I mice to 0, 1, or 5 mg/kg/day
from GD 1 to GD 17. A separate group of pregnant mice was dosed with either 0 or 1 mg/kg/day
from GD 1 to GD 17 and received drinking water containing 5 ppb PFOA beginning on GD 7. Fi
females and F2 offspring from the second group continued to receive drinking water that
contained 5 ppb PFOA until the end of the study, except during Fi breeding and early gestation,
to simulate a chronic low-dose exposure. Fi offspring body weight at PND 42 was significantly
reduced at 5 mg/kg/day; at PND 63, body weight was significantly reduced for offspring from
dams given 1 mg/kg/day plus 5 ppb in the drinking water compared to offspring from dams
given only 1 mg/kg/day. For the F2 pups, a significant reduction in body weight was observed in
control plus 5 ppb drinking water PFOA offspring on PND 1, but there was no difference by
PND 3. F2 offspring from the 1 mg/kg/day and 1 mg/kg/day plus 5 ppb drinking water PFOA
groups had increased body weights compared to controls on PND 14, PND 17, and PND 22.
Female CD-I mice that had been exposed gestationally to 1 mg/kg/day had significantly
decreased body weights at PND 21 and PND 35 but not at PND 56 {Tucker, 2015, 2851046}.
Macon et al. (2011, 1276151) found no effects on offspring body weights following exposure of
pregnant CD-I mice to PFOA from GD 1 to GD 17 with doses up to 1 mg/kg/day or from GD 10
to GD 17 with doses up to 3 mg/kg/day. Similarly, Cope et al. (2021, 10176465) exposed CD-I
dams to 0.1 or 1.0 mg/kg/day PFOA via oral gavage from GD 1.5 to GD 17.5 and did not find
treatment-related changes in pup weight at PND 0.5, PND 5, or PND 22.
3.4.4.2.4.3 Rats, Postnatal Evaluations
In two NTP 2-year carcinogenicity studies {NTP, 2020, 7330145}, dietary exposure of pregnant
Sprague-Dawley rats to 300 ppm PFOA (approximately 22 mg/kg/day during gestation and
45 mg/kg/day from LD 1 to LD 14) resulted in significantly decreased pup weights throughout
lactation (3%-8% lower than controls). In both studies, there were minimal to no effects on
maternal body weight.
Significantly decreased Fi pup weight (8%-l 1% lower than controls) during lactation was
observed following exposure of pregnant Sprague-Dawley rats to 30 mg/kg/day, in the absence
of effects on maternal body weight; F2 pup weight was slightly decreased at 30 mg/kg/day, but
the effect was not statistically significant {Butenhoff, 2004, 1291063}. At 30 mg/kg/day,
significant decreases in body weight and body weight gain were seen in Fi male offspring during
the juvenile and peripubertal phases and in Fi female offspring beginning on day 8 postweaning
and continuing through pre-cohabitation, gestation, and lactation (along with decreased food
consumption) (Figure 3-69).
3-242
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA Developmental Effects - Offspring Body Weight
Endpoint Study Name Study Design Observation Time Animal Description
Body Weight Change Wolf el al.. 2007. 1332672 developmental (GD1-17) PND1-22 F1 Mouse. CD-1 N-11-14)
Cope el al.. 2021, 10176465 developmental (GD1.5-17.5) PND22-PNW18 F1 Mouse, CD-1 N=8)
F1 Mouse, CD-1 (N=9)
PN D22-128 F1 Mouse, CD-1 K)
F1 Mouse, CD-1 (v)
Fetal Body Weight Blake etal.. 2020, 6305864 developmental (GD1.5-11.5) GD11.S F1 Mouse, CD-1 N=62)
developmental (GD1.5-17.5) GD17.5 F1 Mouse, CD-1 N=62)
Lau et al.. 2006,1276159 developmental (GD1-17) GD18 F1 Mouse, CD-1 (i-i, N=5-42)
Li et al.. 2018,5084746 developmental (GD1 -17) GD18 F1 Mouse, Kunming (^52-, N=10)
LitterWeight Cope et al.. 2021.10176465 developmental (GD1.5-17.5) PND0.5 F1 Mouse. CD-1 (¦#«•, N=9)
PND5.5 F1 Mouse, CD-1 (-s~. N=9)
Huetal.. 2010.1332421 developmental (GD6-17) PND2 F1 Mouse, C57BU6n N=10)
PND7 F1 Mouse, C57BU6n , N=10)
PND14 F1 Mouse, C57BL.'6n , N=10)
Pup Body Weight Cope et al,. 2021.10176465 developmental (GD1.5-17.5) PND0.5 P0 Mouse. CD-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
phalanges (significant at all dose levels except 5 mg/kg/day), hindlimb proximal phalanges
(significant at all dose levels except 3 and 5 mg/kg/day), calvaria (significant at 1, 3, and
20 mg/kg/day), enlarged fontanel (significant at 1, 3, and 20 mg/kg/day), and supraoccipital bone
(significant at 10 and 20 mg/kg/day). Significantly reduced ossification of caudal vertebrae,
metacarpals, metatarsals, and hyoid was observed at 20 mg/kg/day. Significant increases in
minor limb and/or tail defects were observed in fetuses at > 5 mg/kg/day (no defects were
observed at 0, 1, or 3 mg/kg/day) and significantly increased incidence of microcardia was
observed at 10 and 20 mg/kg/day (no incidences were observed in any other groups). Yahia et al.
(2010, 1332451) dosed pregnant ICR mice with 0, 1, 5, or 10 mg/kg/day from GD 0 to GD 17
(sacrificed on GD 18) and reported a significant increase in the incidence of cleft sternum and
ossification delays (phalanges) in GD 18 fetuses at 10 mg/kg/day. In the same study, some dams
were dosed from GD 0 to GD 18 and allowed to give birth, and pup lungs and brains were
examined at PND 4; no abnormalities were reported.
3.4.4.2.6 Altered Developmental Timing
Reduced postnatal growth leading to developmental delays was observed in both rats and mice.
Lau et al. (2006, 1276159) and Wolf et al. (2007, 1332672) reported delayed eye opening in CD-
1 mice offspring after gestational exposure to > 5 mg/kg/day PFOA. Additionally, Wolf et al.
(2007, 1332672) observed delayed eye-opening following gestational plus lactational exposure
to 3 or 5 mg/kg/day. Wolf et al. (2007, 1332672) also observed delayed body hair emergence
following gestational exposure to 5 mg/kg/day or gestational plus lactational exposure to 3 or
5 mg/kg/day. In pregnant 129Sl/SvImJ wild-type and PPARa-null mice orally exposed from GD
1 to GD 17 to 0.1-20 mg/kg/day PFOA {Abbott, 2007, 1335452}, offspring born to wild-type
dams showed a dose-related trend for delayed eye opening compared to controls at 0.6 and
1 mg/kg/day (significant at 1 mg/kg/day; extensive litter loss seen at the higher doses). In
PPARa-null offspring, none of the litters from dams exposed to 3 mg/kg/day had eyes open on
PND 13, but no significant difference between this group and the control was observed by PND
14. Yahia et al. (2010, 1332451) dosed pregnant ICR mice with 0, 1, 5, or 10 mg/kg/day PFOA
from GD 0 to GD 17 (sacrificed on GD 18) and reported a significant decrease in the percentage
of GD 18 fetuses with erupted incisors at 10 mg/kg/day.
3.4.4.2.7 Mammary Gland Development
Altered mammary gland development has been shown to result in later-life functional
reproductive consequences, such as reduced lactational efficacy and subsequent pup loss, and has
been linked to increased incidence of mammary and breast cancers {Fenton, 2006, 470286;
Macon, 2013, 3827893; Birnbaum, 2003, 197117}. Studies examining effects of PFOA exposure
on mammary gland development in CD-I mice reported delayed mammary gland development at
dose levels as low as 0.01 mg/kg/day {Macon, 2011, 1276151; Tucker, 2015,2851046}.
However, no differences in response to a lactation challenge were seen in PFOA-exposed CD-I
mouse dams with delayed mammary gland development, and no significant effects on body
weight gain were seen in pups nursing from dams with less fully developed mammary glands
{White, 2011, 1276150}.
Macon et al. (2011, 1276151) exposed pregnant CD-I mice to PFOA from GD 1 to GD 17 (full
gestation) or GD 10 to GD 17 (late gestation) to examine effects of PFOA exposure on
mammary gland morphology. Mammary gland whole mounts were scored on a 1 to 4 subjective,
age-adjusted, developmental scale. Quantitative measures also were made of longitudinal
3-244
-------
DRAFT FOR PUBLIC COMMENT
March 2023
growth, lateral growth, and number of terminal end buds. At all PFOA exposure levels in both
experiments (> 0.3 mg/kg/day in the full gestation study and > 0.01 mg/kg/day in the late-
gestation study), significantly stunted mammary epithelial growth was observed in female
offspring in the absence of effects on offspring body weight. Additionally, there were significant
differences from controls in quantitative measures of longitudinal and lateral growth and
numbers of terminal end buds at 1 mg/kg/day in the late-gestation experiment. The delayed
development was characterized by reduced epithelial growth and the presence of numerous
terminal end buds. Photographs of the mammary gland whole mounts at PND 21 and PND 84
from the full-gestation experiment showed differences in the duct development and branching
pattern of offspring from dams given 0.3 and 1 mg/kg/day PFOA (offspring from high-dose
dams not pictured). At PND 21, mammary glands from the 1 mg/kg/day late-gestation group had
significantly less longitudinal epithelial growth and fewer terminal end buds compared with
controls. In the late-gestation experiment, mammary gland development was delayed by
exposure to PFOA, especially longitudinal epithelial growth. At PND 21, all treatment groups
had significantly lower developmental scores. At the highest dose, poor longitudinal epithelial
growth and decreased number of terminal end buds were observed. The quantitative measures
were statistically significant only for the high dose compared to the controls, whereas the
qualitative scores at all doses were significantly different from controls.
CD-I mice were dosed with 5 mg/kg/day on GD 7-GD 17, GD 10-GD 17, GD 13-GD 17, or
GD 15-GD 17 or with 20 mg/kg/day on GD 15-GD 17 (controls were dosed GD 7-GD 17) and
mammary gland effects of this study were published by White et al. (2009, 194811). Mammary
gland developmental scores for all offspring of dams exposed to PFOA were significantly lower
at PND 29 and PND 32. Delayed ductal elongation and branching and delayed appearance of
terminal end buds were characteristic of delayed mammary gland development at PND 32. At
18 months of age, mammary tissues were not scored (due to the lack of a protocol applicable to
mature animals) but dark foci (composition unknown) in the mammary tissue were observed at a
higher frequency in exposed animals compared to controls. There was no consistent response
with respect to dosing interval. Qualitatively, mammary glands from treated dams on LD 1
appeared immature compared with control dams {White, 2009, 194811}. The authors also
exposed pregnant CD-I mice to 0, 3, or 5 mg/kg/day from GD 1 to GD 17 and offspring were
cross-fostered at birth to create seven treatment groups: control, in utero exposure only (3U and
5U), lactational exposure only (3L and 5L), and in utero + lactational exposure (3U+L and
5U+L). Mammary gland whole mounts from female offspring between PND 22 and PND 63
were scored. With the exception of females of the 3L group, all female offspring of PFOA-
exposed dams had reduced mammary gland developmental scores at PND 22. At PND 42,
mammary gland scores from females in the 3U+L group were the only ones not statistically
different from control scores. This might have been due to inter-individual variance and multiple
criteria used to calculate mammary gland development scores. All offspring of dams exposed to
PFOA exhibited delayed mammary gland development at PND 63, including those exposed only
through lactation (3L and 5L).
White et al. (2011, 1276150) dosed pregnant CD-I mice with 0, 1, or 5 mg/kg/day from GD 1 to
GD 17. A second group of pregnant mice was dosed with either 0 or 1 mg/kg/day from GD 1 to
GD 17 and also received drinking water containing 5 ppb PFOA beginning on GD 7. The Fi
females and F2 offspring from the second group continued to receive drinking water that
contained 5 ppb PFOA until the end of the study, except during Fi breeding and early gestation,
3-245
-------
DRAFT FOR PUBLIC COMMENT
March 2023
to simulate a chronic low-dose exposure. Only the Po dams were given PFOA by gavage. Po
females were sacrificed on PND 22. Fi offspring were weaned on PND 22 and bred at 7-8 weeks
of age. F2 litters were maintained through PND 63. Groups of Fi and F2 offspring were sacrificed
on PND 22, PND 42, and PND 63. A group of F2 offspring also was sacrificed on PND 10. A
lactational challenge experiment was performed on PND 10 with Fi dams and F2 offspring to
estimate the volume of milk produced during a discrete period of nursing. Mammary glands were
evaluated from Po dams on PND 22, from Fi dams on PND 10 and PND 22, and from Fi and F2
female offspring on PND 10 (F2 only), PND 22, PND 42, and PND 63. Mammary gland whole
mounts were scored qualitatively. At PND 22, control Po dams displayed weaning-induced
mammary involution. At PND 22, the mammary glands of all PFOA-exposed Po dams, including
the dams receiving 5 ppb PFOA via drinking water only, resembled glands of mice at or near the
peak of lactation (~PND 10). The Fi dams examined on PND 10 and PND 22 had significantly
lower developmental scores on PND 10, but that was no longer evident at PND 22, except for
those exposed in utero to 5 mg/kg/day. In the Fi female offspring not used for breeding, the
mammary glands of all PFOA-exposed mice were significantly delayed in development on PND
22, 42, and 63. For the F2 female offspring, some differences in mammary gland scores were
observed between the groups, but most were not significantly different from controls. No
differences in response to a lactational challenge were seen in PFOA-exposed dams with
morphologically delayed mammary gland development.
Tucker et al. (2015, 2851046) orally exposed pregnant CD-I and C57BL/6 mice to 0, 0.01, 0.1,
0.3, or 1 mg/kg/day from GD 1 to GD 17. After parturition, the number of pups was reduced so
that there were ultimately four to eight CD-I litters and three to seven C57BL/6 litters per
treatment. Different treatment blocks monitored for different endpoints at different times. There
was a dose-related trend towards decreasing mammary gland developmental scores for both
strains of mice. In CD-I mice, scores were significantly reduced at PFOA doses
> 0.01 mg/kg/day on PND 35 and >0.1 mg/kg/day on PND 21. In C57BL/6 mice, scores were
significantly reduced at 0.3 and 1.0 mg/kg/day on PND 21. The authors suggest that these
differences in responses between strains may be due to increased serum PFOA levels of the CD-
1 mice {Tucker, 2015, 2851046}. At 5 mg/kg/day, in mammary glands of C57BL/6 mice, there
was a significant increase in the number of terminal end buds and stimulated terminal ducts;
ductal length was not affected. Mammary gland development was inhibited in C57BL/6 mice
dosed with 10 mg/kg/day, with no terminal end buds or stimulated terminal ducts present and
very little ductal growth.
In a study of direct peripubertal exposure, Yang et al. (2009, 5085085) orally dosed 21-day-old
female BALB/c or C57BL/6 mice with 0, 1, 5, or 10 mg/kg/day PFOA for 5 days/week for
4 weeks. Mammary glands of BALB/c mice treated with 5 or 10 mg/kg/day had reduced ductal
length, decreased number of terminal end buds, and decreased stimulated terminal ducts;
injection with bromo-2'-deoxyuridine, a marker of cell proliferation, into the mammary gland
revealed a significantly lower number of proliferating cells in the ducts and terminal end
buds/terminal ducts at 5 mg/kg/day (not examined at 10 mg/kg/day).
3.4.43 Mechanistic Evidence
Mechanistic evidence linking PFOA exposure to adverse developmental outcomes is discussed
in Sections 3.2.6, 3.2.7, 3.3.4, 3.4.1, and 3.4.5 of the 2016 PFOAHESD {U.S. EPA, 2016,
3603279}. There are 19 studies from recent systematic literature search and review efforts
3-246
-------
DRAFT FOR PUBLIC COMMENT
March 2023
conducted after publication of the 2016 PFOA HESD that investigated the mechanisms of action
of PFOA that lead to developmental effects. A summary of these studies is shown in Figure 3-70.
Mechanistic Pathway Animal Human In Vitro Grand Total
Angiogenic. Antiangiogenic, Vascular Tissue Remodeling
1
0
0
1
Big Data, Non-Targeted Analysis
0
6
1
7
Cell Growth, Differentiation, Proliferation, Or Viability
5
1
2
e
Cell Signaling Or Signal Transduction
2
1
0
3
Fatty Add Synthesis, Metabolism, Storage, Transport, Binding, B-Oxidation
4
0
1
5
Hormone Function
2
0
0
2
Inflammation And Immune Response
0
1
0
1
Oxidative Stress
2
1
0
3
Xenobiotic Metabolism
3
0
1
4
Other
0
0
1
1
Not Applicable/Not Specified/Review Article
1
0
0
1
Grand Total
8
7
4
19
Figure 3-70. Summary of Mechanistic Studies of PFOA and Developmental Effects
Interactive figure and additional study details available on Tableau.
Mechanistic data available from in vitro, in vivo, and epidemiological studies were evaluated to
inform the mode of action of developmental effects of PFOA. Outcomes included early survival,
general development, and gross morphology; fetal growth and placental effects; metabolism;
hepatic development; cardiac development; and neurological development.
3.4.4.3.1 Early Survival, General Development, Gross Morphology
Mechanisms through which PFOA exposure may alter survival and development were studied in
several in vivo experimental animal models. In an in vivo mouse developmental study, pregnant
NMRI dams exposed to PFOA from GD 5—9 via intraperitoneal (IP) injection showed increased
fetal death in the offspring at the highest dose (20 mg/kg/day) of PFOA, as well as
histopathological abnormalities in the brain, liver, and heart, possibly due to the observed
mitochondrial toxicity/dysfunction (e.g., increased mitochondrial swelling, increased
mitochondrial membrane potential (MMP) collapse) or oxidative stress (e.g., increased
mitochondrial ROS formation) {Salimi, 2019, 5381528}. In another mouse developmental study
examining lower doses in the dams, embryo survival was not affected at up to 10 mg/kg/day
PFOA exposure in dams exposed from GD 1.5-11.5 or GD 1.5—17.5 via oral gavage {Blake,
2020, 6305864}. However, 5 and 10 mg/kg exposure via oral gavage from GD 1—17 decreased
survival rate in 5-day old pups, possibly due to hepatotoxicity; the authors observed significantly
3-247
-------
DRAFT FOR PUBLIC COMMENT
March 2023
increased liver index in pups and increased reactive oxygen species and changes in liver enzyme
function, mediated by the PPARa pathway {Li, 2019, 5387402}.
Several studies using zebrafish as a model organism that were identified in the current
assessment were included in a recent review of developmental effects of PFOA {Lee, 2020,
6323794}. In general, PFOA exposure was associated with developmental delays, reductions in
measures of embryo survival, and increased malformations in the head and tail that may be
related to perturbations in gene expression during critical windows of organism development.
The review by Lee et al. (2020, 6323794) included a zebrafish multigenerational study by
Jantzen et al. (2017, 3603831), in which embryos were exposed to PFOA from 3 to 120 hours
post-fertilization (hpf). Embryos were allowed to reach adulthood and breed. Although exposure
to PFOA did not decrease survival in the first exposed generation (Po), there were significantly
fewer eggs and viable embryos than the controls in the Po. Further, Fi embryos had significant
developmental delays and delayed hatching. Gene expression analysis of four solute carrier
organic anion transporter family members (slcoldl, slco2bl, slco3al, and slco4al) and the
growth factor transforming growth factor beta la (tgfbla) in the Po generation showed that
PFOA exposure led to decreased expression in slco2bl, slco3al, and slco4al and increased
expression in slcoldl. In the Fi embryos, there was a significant increase in expression of the
protein transporter adaptor related protein complex 1 subunit sigma 1 {aplsl). The authors
concluded that alterations in the expression of these genes during development likely contributed
to the delayed development and morphologic and toxic effects observed {Jantzen, 2017,
3603831}. The elevations in aplsl were in conflict with a prior publication from the same
research group that reported decreased aplsl at 120 hpf, which coincided with alterations in
morphometric parameters in zebrafish embryos, including increased interocular distance (a
metric of cranio-facial development), reduced total body length, and reduced yolk sac area
{Jantzen, 2016, 3860114}. Other alterations in gene expression at 120 hpf included elevations in
slco2bl (transport protein) and transcription factor 3a (tfc3a; involved in muscle development),
and c-fos (transcription factor complex). Altogether, results suggest that alterations in aplsl are
unlikely the result of a global upregulation or downregulation of genes and that PFOA may
differentially influence genes at certain points in development. However, the current data cannot
rule out the possibility that the observed alterations in gene expression are due to a delay or
acceleration in development.
In another zebrafish study by Bouwmeester et al. (2016, 3378942), embryos that were exposed to
10-320 |iM PFOA were examined for developmental toxicity and morphological effects. PFOA
did not induce embryotoxic effects at the exposure levels in the experiment; however, some
epigenome modifications were noted. When locus-specific methylation was assessed, PFOA
exposure was associated with hypomethylation on the CpG region of vasa, and hypermethylation
at CpGl in vitellogenin 1 (vtgl). Vasa is expressed in the germline and is active during
development, and vtgl is expressed in the liver of egg-laying vertebrates and encodes for the
estrogen responsive egg-yolk protein vitellogenin, although, interestingly, PFOA was included in
this study to demonstrate a "non-estrogenic PPARy/RXR agonist." These epigenetic
modifications early in life and development may play a role in the development of later life
adverse health outcomes {Bouwmeester, 2016, 3378942}.
In humans, epigenetic modification during development of the fetus can be measured via cord
blood at birth. Several human studies evaluated cord-blood DNA methylation patterns to
3-248
-------
DRAFT FOR PUBLIC COMMENT
March 2023
understand the epigenetic effects of PFOA exposure. Miura et al. (2018, 5080353) found that
increased PFOA in the cord blood was associated with global hypermethylation in a cohort from
Japan; however, two other cord blood studies of global methylation found no associations
between PFOA exposure and global methylation changes {Liu, 2018, 4926233; Leung, 2018,
4633577}. Similarly, Kingsley et al. (2017, 3981315) did not observe associations between
PFOA exposure in cord blood and epigenome-wide changes in global methylation status.
However, for the high PFOA exposure group, the authors found hypomethylation in seven CpG
sites located in several genes, including RAS P21 protein Activator 3 (RASA3) and Opioid
Receptor Delta 1 (OPRD1). RASA3 methylation changes could result in impaired cell growth and
differentiation, contributing to reduced fetal growth and birth weight. OPRD1 is involved in
weight and obesity, as well as morphine and heroin dependence, and could potentially be a
mechanistic pathway linking PFOA and obesity, an association that has previously been reported
{Kingsley, 2017, 3981315}. Cord blood samples from a prospective cohort in China were used
by Liu et al. (2018, 4239494) to evaluate potential associations between PFOA exposure and
leukocyte telomere lengths (LTLs). There was no association between PFOA exposure and LTLs
in this study.
3.4.4.3.2 Fetal Growth and Placental Effects
Growth was assessed in four mouse developmental studies. Blake et al. (2020, 6305864) found
decreased embryonic weights in CD-I mice at GD 17.5, with concurrent increases in placental
weights and placental lesions consistent with labyrinth congestion (Section 3.4.4.2.4.1).
Placentas also had higher thyroxine (T4) levels relative to controls, suggesting a possible
endocrine mechanistic pathway of effect. In NMRI mice exposed to 0, 1, 10, or 20 mg/kg/day
PFOA from GD 5—9, Salimi et al. (2019, 5381528) observed reduced fetal length and weight,
and decreased placental diameter at the highest dose group (20 mg/kg/day). The authors note that
toxicity was likely mediated through mitochondrial toxicity in the liver (described below), which
appeared to be isolated to the mouse fetus rather than the placenta. Li et al. (2019, 5387402)
reported a dose-dependent reduction in growth and weight gain in Kunming mouse pups exposed
to PFOA during gestation (GD 0—17). The authors attribute the stunted growth to hepatotoxicity
consequent to increased ROS and changes in liver enzyme function mediated by the PPARa
pathway {Li, 2019, 5387402}.
Perturbations in growth and corresponding changes in gene expression of key developmental
genes have been observed in several studies in zebrafish. In the multigenerational zebrafish study
by Jantzen et al. (2017, 3603831), Po generation fish exposed to PFOA had significantly shorter
body length and reduced body weight compared to controls. Offspring of PFOA exposed fish
were significantly developmentally delayed and had increased expression in the protein transport
gene aplsl at 48 hpf, possibly leading to the changes in growth {Jantzen, 2017, 3603831}. In
Jantzen et al. (2016, 3860114), several morphometric endpoints were measured in zebrafish
embryos exposed to 0.02, 0.2, or 2.0 |iM PFOA, including interocular distance, total body length,
and yolk sac area. The size of all three parameters was reduced in groups exposed to PFOA,
indicating slowed embryonic development) at values 5- to 25-fold below previously calculated
median lethal concentration (LCso) values. The authors also evaluated gene expression at 120 hpf
and 14 days post-fertilization (dpf). At 120 hpf, slco2bl (transport protein), tfc3a (involved in
muscle development), and c-fos (transcription factor complex) were upregulated, while apis
(involved in protein transport) was downregulated. At 14 dpf, slco2bl and Tcf3a (involved in
muscle development) were upregulated {Jantzen, 2016, 3860114}.
3-249
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Gorrochategui et al. (2014, 2324895) evaluated cytotoxicity and aromatase activity in a placental
cell line (JEG-3 cells). PFOA exposure was found to induce cytotoxicity and inhibit aromatase
(CYP19) activity {Gorrochategui, 2014, 2324895}. In a rhesus monkey trophoblast cell line,
PFOA treatment showed significant differences in gene expression, with possible affected
diseases/biological functions including cell movement, epithelial tissue growth, and
vasculogenesis. Pathways included cysteine metabolism, interleukin signaling, Toll-like receptor,
TGF-b, PDGF, PPAR, NFKB, MAPK, Endothelin 1, TNRF2, tight junctions, cytokines
including IFNY and IFNa, and possible FOS signaling {Midie, 2018, 4241048}.
Lastly, a longitudinal study by Ouidir et al. (2020, 6833759) examined global methylation in the
placenta at birth in women for whom PFOA levels in the plasma were determined in the first
trimester. The authors did not find any associations between PFOA exposure and DNA
methylation status of the placenta {Ouidir, 2020, 6833759}.
3.4.4.3.3 Metabolism
van Esterik et al. (2015, 2850288) examined metabolic effects of developmental exposure to 3-
3000 |ig/kg PFOA exposure in C57BL/6JxFVB hybrid mice. The authors found that PFOA
exposure during gestation and lactation resulted in reduction in weight that persisted to
adulthood. The weight loss was attenuated by a high-fat diet (from 21—25 days) in males, but
not females, suggesting that the weight reductions were mediated through metabolic mechanisms
that may exhibit a female bias. There were no significant changes in metabolic parameters (i.e.,
glucose homeostasis, basal glucose, energy expenditure, uncoupling protein 1 (ucpl; also known
as thermogenin) expression in brown adipose tissue) in either sex. However, in females,
cholesterol and triglycerides showed a dose-dependent decrease. The authors suggest that these
changes in lipid metabolism could be mediated by PPARa activation {van Esterik, 2015,
2850288}. Li et al. (2019, 5387402) examined PPARa activation pathways as a mechanism of
PFOA-induced liver and metabolic toxicity during development in mice. The authors found that
female mice exposed gestationally to PFOA had significantly downregulated gene expression of
PPARa in the 2.5 and 5 mg/kg/day groups, but not the highest dose group (i.e., 10 mg/kg/day).
PFOA exposure also increased gene expressions of Acoll and Acoxl (downstream regulatory
genes of PPARa), indicating that early PFOA exposure causes lasting changes in the PPARa
pathway. PPARa regulates fatty acid oxidative metabolism and energy consumption, through
peroxisome and mitochondrial P-oxidation and microsome co-oxidation {Li, 2019, 5387402}.
PFOA has been described as a weak PPARa ligand, but the role of PPARa in mediating the
developmental toxicity associated with PFOA exposure is not yet clear {Peraza, 2006, 509877}.
Metabolomic profiles in relation to PFOA exposure were analyzed in a human study. In a cross-
sectional study in 8-year-old children in Cincinnati, OH, the authors conducted untargeted, high-
resolution metabolomic profiling in relation to serum PFOA concentrations. They found that
PFOA exposure was associated with several lipid and amino acid metabolism pathways,
including that of arginine, proline, aspartate, asparagine, and butanoate {Kingsley, 2019,
5405904}.
3.4.4.3.4 Hepatic Development
Three developmental mouse studies examined the effect of PFOA on liver development and
function, van Esterik et al. (2015, 2850288) found that developmental exposure to PFOA
resulted in increased liver weights and abnormal liver histopathology, with toxicity possibly
3-250
-------
DRAFT FOR PUBLIC COMMENT
March 2023
mediated through the PPARa pathway. Salimi et al. (2019, 5381528) exposed pregnant mice to
PFOA from GD 5—9 and observed mitochondrial disruption in the fetal liver, including
mitochondrial swelling and mitochondrial membrane potential collapse. These effects
significantly increased at the highest (20 mg/kg/day) exposure group. Measures of oxidative
stress (hydrogen peroxide production) in the liver were also significantly higher in groups
exposed to 10 or 20 mg/kg/day PFOA in comparison to control animals. Li et al. (2019,
5387402) hypothesized that PFOA accumulation in pup liver may promote oxidative stress via
PPARa activation pathways that contribute to liver and metabolic toxicity in mice. The authors
found that female mice exposed gestationally to PFOA had increased liver weight and dose-
responsive morphological changes in the liver including swollen hepatocytes, blurred
architecture, and vacuolar degeneration. Liver enzymes (AST and ALT) were increased in the
serum, and oxidative stress biomarkers (Catalase (CAT), Superoxide dismutase (SOD), and 8-
OHdG) were increased. Liver histone acetyltransferase (HAT) activity was reduced, and histone
deacetylase (HDAC) activity was increased. Further, histone acetylation in the liver was reduced.
These effects suggest that PFOA can alter the epigenetic regulation of liver responses which may
contribute to adverse hepatic health outcomes (Section 3.4.1).
3.4.4.3.5 Cardiac Development
Data from one study in mice, one study in zebrafish, and one in vitro study provide insight into
the mechanism by which PFOA perturbs cardiac development. In a recent review that covered
PFOA toxicity in zebrafish, Lee et al. (2020, 6323794) reported that PFOA exposure has been
consistently associated with increases in pericardial edema and altered heart rates at various
stages of development in embryos. An in vivo mouse developmental study by Salimi et al. (2019,
5381528) also found that PFOA exposure was associated with cardiotoxicity in offspring. In this
study, pregnant dams were treated with PFOA, and fetuses were studied for tissue abnormalities.
Groups treated with PFOA showed increased histopathological abnormalities in the fetal heart,
including hepatomegaly. Mitochondrial swelling in mitochondrial suspension of fetal heart tissue
was also observed along with increased mitochondrial membrane potential collapse. Measures of
oxidative stress in the fetal heart were also significantly higher in exposed vs. control animals
{Salimi, 2019, 5381528}. An in vitro experiment by Zhou et al. (2017, 3981356) examined the
ability of mouse embryonic stem cells to differentiate into myocardiocytes following exposure to
2.5, 5, 10, 20, 40, 80, or 160 [j,g/mL PFOA. Differentiation was determined by the contractility
(i.e., contract rate) of the cells, as well as the upregulation of myh6, which is a regulatory gene
that is essential for cardiac muscle development. No effects on differentiation or myh6
expression were observed below 20 [j,g/mL.
3.4.4.3.6 Neurological Development
Salimi et al. (2019, 5381528) also reported teratogenic effects in the brain of fetal mice
following maternal exposures up to 20 mg/kg/day PFOA via IP injection from GD 5—9. The
histopathological abnormalities in the brain included anencephaly, microcephaly, and
hydrocephaly, all at the highest (20 mg/kg/day) exposure. Mitochondrial swelling in
mitochondrial suspension of fetal brain tissue was also observed along with increased
mitochondrial membrane potential collapse. Higher mitochondrial disruption was observed at
lower concentrations in the brain tissue than other fetal tissues (i.e., heart and liver), suggesting
that the brain was more susceptible to mitochondrial toxicity/dysfunction. Measures of oxidative
stress in the brain were also significantly higher in exposed animals in comparison to controls.
3-251
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The effects of PFOA on neurodevelopment and behavior in zebrafish were examined in two
studies. In the aforementioned zebrafish embryo assay by Jantzen et al. (2016, 3860114),
embryonic exposure to 0.02, 0.2, or 2.0 micromolar (|iM) PFOA during the first five dpf resulted
in hyperactive locomotor activity in larvae as evidenced by increased swimming velocity,
possibly mediated through altered expression of development-associated genes (c-fos, tfc3a,
slco2bl, and apis). Stengel et al. (2018, 4238489) developed a neurodevelopmental toxicity test
battery using zebrafish embryos. PFOA did not produce any changes in acetylcholinesterase
(AChE) inhibition, nor the neuromast assay, olfactory, or retinal toxicity assays {Stengel, 2018,
4238489}.
3.4.4.3.7 Conclusion
In the context of the available mechanistic studies, it appears that several mechanisms may be
involved in PFOA-driven developmental toxicity. In general, the observed effects suggest that
the developing liver, developing heart, and placenta may be affected by PFOA at the molecular
level (e.g., differential methylation of genes, gene expression changes), which may be reflected
in developmental health effects described in Section 3.4.4. The effects tend to vary by sex and
developmental timepoint of outcome evaluation. More research is needed to strengthen the
association between PFOA exposure to any one of the several possible contributing factors,
including fluctuations in transporter gene expression, epigenetic changes, oxidative stress, and
PPARa pathway activation, particularly in the placenta.
3.4.4.4 Evidence Integration
The evidence of an association between PFOA and developmental effects in humans is moderate
based on the recent epidemiological literature. As noted in the fetal growth restriction summary,
there is evidence that PFOA may impact fetal growth restriction across a variety of BWT-related
measures. Comparing the postnatal growth results in infants with birth-related measures is
challenging due to complex growth dynamics including rapid growth catch-up periods for those
with fetal restriction. Nonetheless, the evidence for postnatal weight deficits was comparable to
that seen for BWT. Collectively, the majority of LBW studies were supportive of an increased
risk with increasing PFOA exposures. Five medium or high confidence studies on LBW showed
increased risks with increased PFOA levels. Several meta-analyses also support evidence of
associations between maternal or cord blood serum PFOA and BWT or BWT-related measures
{Johnson, 2014, 2851237; Verner, 2015, 3150627; Negri, 2017, 3981320; Steenland, 2018,
5079861} (Table A-41, PFOA Appendix A).
Overall, there was mixed evidence of adverse associations between PFOA and both gestational
age (7 of the 18 studies) and preterm birth (6 of 11 studies). Most of the associations for either of
these gestational duration measures were reported in medium or high confidence studies. For
example, five of six studies were increased odds of PTB were high confidence. Few other
patterns were evident that explained any between study heterogeneity. For example, five of the
null studies were rated as having adequate sensitivity, and one was rated deficient. There was a
preponderance of associations related to sample timing possibly related to pregnancy
hemodynamic influences on the PFOA biomarkers, as five of the seven studies reporting adverse
associations were sampled later in pregnancy (i.e., trimester two onward).
There was less consistent evidence of PFOA impacts on rapid growth measures, postnatal height
and postnatal adiposity measures up to age 2. There was less evidence available for other
3-252
-------
DRAFT FOR PUBLIC COMMENT
March 2023
endpoints such as fetal loss and no evidence of associations in recent studies of PFOA and birth
defects such as cryptorchidism or hypospadias. Similarly, there was less consistent evidence of
an impact of PFOA exposure on gestational duration measures (i.e., either preterm birth or
gestational age measures) as many of studies did not show adverse associations.
However, as noted previously there is some uncertainty as to what degree the evidence may be
impacted by pregnancy hemodynamics and sample timing differences across studies as this may
result in either confounding or reverse causality {Steenland, 2018, 5079861}. Additional
uncertainty exists due to the potential for confounding by other PFAS. Very few of the existing
studies performed multipollutant modeling in comparison with single pollutant estimates of
PFOA associations. The multipollutant modeling results were often mixed from single pollutant
estimates with some estimates increasing and some decreasing. Unlike other PFAS, PFOA was
chosen amongst dimension-reducing statistical approaches from models with various PFAS and
or other environmental contaminants adjusted for two different studies {Lenters, 2016, 5617416;
Starling, 2017, 3858473}. Although these results are smaller in magnitude, they appear coherent
with single exposure model results. There is some concern that controlling for other highly
correlated co-exposures in the same model may amplify the potential confounding bias of
another co-exposure rather than removing it {Weisskopf, 2018, 7325521}. Given these
interpretation difficulties and potential for this co-exposure amplification bias, it remains unclear
whether certain mutually adjusted models give a more accurate representation of the independent
effect of specific pollutants for complex PFAS mixture scenarios.
The animal evidence of an association between PFOA and developmental toxicity is robust
based on 13 high or medium confidence animal toxicological studies, in concordance with the
data in humans, supporting that the developing fetus is a target of PFOA toxicity. Specifically,
several studies in rodents show decreased fetal and pup weight with gestational PFOA exposure,
similar to the evidence of LBW seen in infants. Oral studies in rodents consistently show that
gestational PFOA exposure results in pre- and postnatal effects on offspring, as well as maternal
effects in dams. Notably, mice appear to be more sensitive to developmental toxicity as a result
of gestational exposure compared to rats. In addition, studies in both rats and mice show that
effects on offspring (e.g., decreases in body weight, survival) occur at lower dose levels than
those that produce maternal body weight effects.
Evidence from mechanistic studies that relates to observed developmental effects of PFOA is
limited. Decreased survival in the offspring of pregnant mice exposed to PFOA was potentially
related to hepatotoxicity induced by PPARa activation, as discussed in detail in Section 3.4.1.3.
In human cord blood samples, evidence of epigenetic alterations within genes that are involved
in cell growth and differentiation and obesity was observed; however, these epigenetic
alterations were not evaluated in the context of postnatal outcomes and are inconsistent; two
other studies found no association between PFOA exposure and changes to the epigenome. In
zebrafish studies, the expression of several genes that are related to growth and development
(e.g., tfc3a, which is involved in muscle development) was altered by PFOA exposure, with
variable magnitude and, in some cases, the direction of change according to the timepoint
measured. Oxidative stress was observed in the developing brain and heart of mice exposed to
PFOA in utero, suggesting toxicity of PFOA during development. Overall, the data demonstrate
that PFOA may alter the expression of genes involved in growth and development, although
additional studies in mammals are needed to confirm such. Additionally, evidence exists that
3-253
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA can alter the epigenome, although the functional effects of the epigenetic effects are not
clear.
3.4.4.4.1 Evidence Integration Judgment
Overall, considering the available evidence from human, animal, and mechanistic studies, the
evidence indicates that PFOA exposure is likely to cause developmental toxicity in humans
under relevant exposure circumstances (Table 3-10). This conclusion is based primarily on
evidence of decreased birth weight from epidemiologic studies in which PFOA was measured
during pregnancy, primarily with median PFOA ranging from 1.1 to 5.2 ng/mL. The conclusion
is supported by coherent epidemiological evidence for biologically related effects
(e.g., decreased postnatal growth, birth length), as well as consistent findings of dose-dependent
decreases in fetal weight and other developmental effects observed in animal models
gestationally exposed to PFOA at doses as low as 0.5 mg/kg/day. Although there is available
mechanistic information that provides support for the biological plausibility of the phenotypic
effects observed in exposed animals, the data are too limited to sufficiently support the human
relevance of the animal findings.
3-254
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 3-10. Evidence Profile Table for PFOA Developmental Effects
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Studies and
Interpretation
Summary and Key
Findings
Factors that
Increase Certainty
Factors that Decrease
Certaintv
Evidence Stream
Judgment
Evidence from Studies of Exposed Humans (Section 3.4.4.1)
Fetal growth
restriction
22 High confidence
studies
14 Medium confidence
studies
10 Low confidence
studies
Some deficits in mean
birth weight were
observed in most studies
(21/32) in the overall
population, but evidence
for the exposure-
response relationship
was limited. The
majority of studies on
changes in standardized
birth weight measures
reported inverse
associations (9/15), with
most (6/9) of these being
high and medium
confidence. Similarly,
most studies (9/12)
observed either an
increased risk of low
birth weight or SGA.
Deficits in birth weight
were supported by
adverse findings for
related FGR outcomes
such as birth length
(9/26) and head
circumference (10/21) in
the overall population or
across sexes.
• High and medium
confidence studies
• Consistent
direction of effects
for most outcomes
• Coherence of
findings between
different measures
of FGR
• Limited evidence of
exposure-response
relationships based on
categorical data
• Potential bias due to
hemodynamic
differences noted in
studies using samples
from later pregnancy
®©o
Moderate
Epidemiological
evidence for
developmental effects is
based on consistent
adverse effects for FGR.
Consistent deficits in
birth weight and
standardized birth
weight were observed in
many high and medium
confidence cohort
studies. Birth weight
findings were supported
by adverse results
reported for other
measures of FGR,
including birth length
and head circumference,
and adverse effects on
gestational duration.
Some uncertainties
remain regarding the
shape of the exposure-
response relationship,
and the potential impact
0©O
" Evidence Indicates (likely)
Primary basis and cross-
stream coherence:
Evidence consisted of
decreased birth weight from
epidemiologic studies in which
PFOA was measured during
pregnancy. This is supported
by coherent epidemiological
evidence for biologically
related effects (e.g., decreased
postnatal growth, birth length)
and consistent findings of
dose-dependent decreases in
fetal weight observed in
animal models gestationally
exposed to PFOA.
Human relevance and other
inferences:
Although there is available
mechanistic information that
provides support for the
biological plausibility of the
phenotypic effects observed in
exposed animals, the data are
too limited to sufficiently
3-255
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Gestational duration
11 High confidence
studies
5 Medium confidence
studies
4 Low confidence
studies
Fetal Loss
2 High confidence
studies
2 Medium confidence
studies
1 Low confidence
study
Post-natal growth
6 High confidence
studies
4 Medium confidence
studies
3 Low confidence
studies
Birth Defects
2 Medium confidence
studies
Adverse effects on
gestational age were
observed (7/18), with
most (6/7) considered
high or medium
confidence. Increased
risk of preterm birth was
also observed in some
studies (5/11).
Increased risk of fetal
loss was reported in one
high (1/2) and one
medium (1/2) confidence
study. The response in
the high confidence
study was monotonic
across exposure
quartiles. One study
reported a decreased risk
of fetal loss, but the
study was considered
low confidence.
Increased risk of adverse
weight changes in
infancy was observed in
most studies (7/9)
examining the outcome.
Decreases in infant
height were observed in
a few studies (3/7). Only
one study (1/7) reported
increased adiposity,
which was in male
infants only.
Two low confidence
studies reported mixed
results for total or
' High and medium
confidence studies
• High and medium
confidence studies
• Good or adequate
sensitivity
• Consistent
magnitude of effect
• Dose-dependent
response
• High and medium
confidence studies
• Good or adequate
sensitivity for most
studies
> No factors noted
• Potential bias due to
hemodynamic
difference noted in
studies using samples
from later pregnancy
> No factors noted
of hemodynamics in
later pregnancy due to
use of biomonitoring
samples from the second
and third trimester or
post-partum.
support the human relevance
of the animal findings.
»Inconsistent timing of
follow-up evaluation
• Low confidence
studies
3-256
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
2 Low confidence
studies
combined birth defects.
No association with
cryptorchidism was
reported in one study;
one study reported
decreased odds of septal
defects, conotruncal
defects, and total
congenital heart
defects.
' Limited number of
studies examining
individual defects
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.4.2)
Studies and
Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that Decrease
Certainty
Evidence Stream
Judgment
Maternal body
weight
2 High confidence
studies
6 Medium confidence
studies
Offspring body
weight
2 High confidence
studies
9 Medium confidence
studies
Many rodent studies
observed a change in
maternal body weight or
weight gain following
PFOA exposure (5/8).
The direction of this
change was not
consistent among
studies, with some
rodent studies observing
a decrease in weight
(3/5), and some mouse
studies observing an
increase (2/5).
Many rodent studies
observed changes in
fetal or pup body weight
following PFOA
exposure (9/11). Most of
these show a decrease in
offspring weight (8/9).
One study observed an
increase in offspring
• High and medium
confidence studies
• Inconsistent direction
of effects
• High and medium
confidence studies
• Consistent
direction of effects
> No factors noted
©0©
Robust
Evidence based on 13
high or medium
confidence animal
toxicological studies
indicates that the
developing fetus is a
target of PFOA toxicity.
Several studies in
rodents show decreased
fetal and pup weight
with gestational PFOA
exposure, similar to the
evidence of FGR seen in
human infants. Oral
studies in rodents
consistently show that
gestational PFOA
exposure results in pre-
and postnatal effects on
offspring, as well as
3-257
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Offspring mortality
2 High confidence
studies
7 Medium confidence
studies
Placenta effects
2 Medium confidence
studies
body weight, but only in
male mice. Two mouse
studies showed no
change in offspring
body weight (2/11).
Many rodent studies
observed increases in
offspring mortality
following PFOA
exposure (6/9). A rat
study observed
increased post-weaning
mortality in female pups
but no pre-weaning
mortality or change in
stillborn pups. Five
mouse studies found
increased offspring
mortality including
increased resorption
(4/4), decreased live
fetuses or live pups born
(2/4), and decreased
postnatal survival
(2/3). Two studies found
no change in offspring
mortality or survival
(2/8). No change in litter
size was observed in any
rat or mouse study (3/3).
Two mouse studies
noted a decrease in
relative placenta weight
following gestational
PFOA exposure. In
these studies, lesions on
the placenta and other
• High and medium
confidence studies
• Consistent
direction of effects
> No factors noted
• Medium confidence
studies
• Limited number of
studies examining
outcomes
maternal effects in dams.
Notably, mice appear to
be more sensitive to
developmental toxicity
as a result of gestational
exposure compared to
rats. In addition, studies
in both rats and mice
show that effects on
offspring (e.g., decreases
in body weight, survival)
occur at lower dose
levels than those that
produced maternal body
weight effects.
Evidence Integration
Summary Judgment
3-258
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Offspring liver
weight
2 Medium confidence
studies
Developmental
timing
2 Medium confidence
studies
Structural
abnormalities
1 Medium confidence
study
Mammary gland
development
1 Medium confidence
study
histopathological
changes were observed
including changes to the
labyrinth (e.g., atrophy,
decreased area,
congestion, necrosis)
and early fibrin clot.
Fewer placentas were
determined to be within
normal limits (1/1).
Increases in offspring
relative liver weight
were noted in two
mouse studies following
gestational PFOA
exposure (2/2).
Delayed eye opening
(2/2) and delayed body
hair development (1/1)
was observed in both
sexes of mice.
One mouse study found
structural abnormalities
(e.g., reduced skeletal
ossification) after
developmental exposure
to PFOA.
One mouse study found
abnormal mammary
gland development in
animals exposed to
PFOA during gestation
(e.g., decreases in
terminal end buds,
mammary gland
developmental score).
• Medium confidence
studies
• Medium confidence
studies
»Medium confidence
study
• Medium confidence
study
• Limited number of
studies examining
outcomes
• Limited number of
studies examining
outcomes
> Limited number of
studies examining
outcomes
• Limited number of
studies examining
outcomes
Evidence Integration
Summary Judgment
3-259
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Lactation index
2 High confidence
studies
Of the two rat studies
that evaluated lactation
index, one noted a
decrease following
PFOA (1/2).
' High confidence
studies
> Limited number of
studies examining
outcomes
Mechanistic Evidence and Supplemental Information (Section 3.4.4.3)
Summary of Key Findings, Interpretation, and Limitations
Key findings and interpretation:
• Decreased survival in mice offspring exposed to PFOA in utero related to PPARa-related
hep ototoxicity.
• Alterations to the expression of genes related to growth and development in vivo in zebrafish.
• Inconsistent results for PFOA-related alterations to DNA methylation in human cord blood.
Limitations:
• Very limited database.
• The role of epigenetic mechanisms in changes at the mRNA level is not clear, nor is the
relationship between molecular changes and apical developmental outcomes.
The limited evidence
demonstrates that PFOA
exposure during
development can alter
the epigenome and the
expression of genes that
control regular growth
and development; it is
possible that such
changes are related,
although the relationship
has not been directly
measured.
Notes: DNA = deoxyribonucleic acid; FGR = fetal growth restriction; mRNA = messenger ribonucleic acid; PPARa = peroxisome proliferater-activated receptor alpha;
SGA = small-for-gestational-age.
3-260
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.4.5 Evidence for Other Health Outcomes
Consistent with the SAB's recommendation, EPA concluded that the non-cancer health
outcomes with the strongest evidence are hepatic, immune, cardiovascular, and developmental.
For all other health outcomes (e.g., reproductive and endocrine), EPA concluded that the
epidemiological and animal toxicological evidence available at this time and from the
preliminary scoping considered in the Proposed Approaches to the Derivation of a Draft
Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1) in
Drinking Water is either suggestive of associations or inadequate to determine associations
between PFOA and the health effects described. Based on this analysis, these outcomes were not
prioritized for the MCLG assessment. The evidence synthesis and integration for these outcomes
are presented in the PFOA appendix. In addition, Section 6.5 further describes rationale for
evidence integration judgments for health outcomes which EPA determined had evidence
suggestive of associations between PFOA and related adverse health effects, though the
databases for those health outcomes shared some characteristics with the evidence indicates
judgment.
3.5 Cancer Evidence Study Quality Evaluation, Synthesis,
Mode of Action Analysis and Weight of Evidence
EPA identified 16 epidemiological and 4 animal toxicological studies that investigated the
association between PFOA and cancer. Of the epidemiological studies, 8 were classified as
medium confidence, 7 as low confidence, and 1 was considered uninformative (Section 3.5.1). Of
the animal toxicological studies, 2 were classified as high confidence, 1 as medium confidence,
and 1 as low confidence (section 3.5.2). Though low confidence studies are considered
qualitatively in this section, they were not considered quantitatively for the dose-response
assessment (Section 4).
3.5.1 Human Evidence Study Quality Evaluation and
Synthesis
3.5.1.1 Introduction
There are 9 epidemiological studies from the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}
that investigated the association between PFOA and cancer effects. Study quality evaluations for
these 9 studies are shown in Figure 3-71.
The 2016 HESD for PFOA {U.S. EPA, 2016, 3603279} concluded there was suggestive
evidence of carcinogenic effects of PFOA for kidney and testicular cancer, based on two C8
Health Project studies and 2 occupational cohorts (Figure 3-71). Specifically, two studies
involving participants in the C8 Health Project showed a positive association between PFOA
levels (mean at enrollment 24 ng/mL) and kidney and testicular cancers {Barry, 2013, 2850946;
Vieira, 2013, 2919154}. There is some overlap in the cases included in these studies. As part of
the C8 Health Project, the C8 Science Panel {C8 Science Panel, 2012, 1430770} concluded that
a probable link existed between PFOA exposure and testicular and kidney cancer. Two
occupational cohorts in Minnesota and West Virginia {Raleigh, 2014, 2850270; Steenland, 2012,
2919168} also examined cancer mortality. Raleigh et al. (2014, 2850270) reported no evidence
of elevated risk for kidney cancer. In the West Virginia occupational cohort, Steenland and
3-261
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Woskie (2012, 2919168) observed significantly elevated risk of kidney cancer deaths in the
highest quartile of modeled PFOA exposure (> 2,384 ng/mL-years). However, each of these
studies is limited by a small number of observed cases (six kidney cancer deaths, sixteen incident
kidney cancer cases, and five incidence testicular cancer cases in Raleigh et al. (2014, 2850270);
twelve kidney cancer deaths and one testicular cancer death in Steenland and Woskie (2012,
2919168)). None of the general population studies reviewed for the 2016 PFOA Health Advisory
examined kidney or testicular cancer, and no associations were observed in the general
population between exposure to PFOA (mean serum PFOA levels up to 86.6 ng/mL) and
colorectal, breast, prostate, bladder, or liver cancer {Bonefeld-fergensen, 2014, 2851186;
Eriksen, 2009, 2919344; Hardell, 2014, 2968084; Innes, 2014, 2850898}. In the C8 Health
Project cohort, Barry et al. (2013, 2850946) observed a significant inverse association with
breast cancer for both untagged and 10-year lagged estimated cumulative PFOA serum
concentrations. Barry et al. (2013, 2850946) also observed positive and significant associations
between PFOA and thyroid cancer in DuPont workers at the Washington, West Virginia plant,
but not in community residents. However, Vieira et al. (2013, 2919154) found no association
between estimated serum concentrations of PFOA with thyroid cancer risk among residents
living near the DuPont Teflon-manufacturing plant in Parkersburg, West Virginia.
3-262
-------
DRAFT FOR PUBLIC COMMENT
March 2023
9^sV^'o^>,0">^"^sw6SVlS
>g®
Barry et al., 2013, 2850946-
+
+
+
+
++
+
+
+
Bonefeld-Jorgensen et al., 2011, 2150988-
+
+
+
+
+
+
+
+
Bonefeld-Jorgensen etal., 2014, 2851186-
+
+
-
+
-
-
+
-
Eriksen et al., 2009, 2919344 -
++
+
+
+
+
+
+
+
Hardell et al., 2014, 2968084 -
++
-
++
-
+
+
+
-
Raleigh et al., 2014, 2850270 -
+
-
+
-
+
+
-
-
Steenland and Woskie, 2012, 2919168-
+*
+
+
-
+
+
+
+*
Steenland et al., 2015, 2851015 -
-
+
-
+
++
+
+
-
Vieira et al., 2013, 2919154-
+
+
+
+
++
+
+
+
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
| Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-71. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cancer Effects Published Before 2016 (References from 2016 PFOA HESD)
Interactive figure and additional study details available on HAWC.
Since publication of the 2016 HESD {U.S. EPA, 2016, 3603279}, 17 epidemiological studies
have been published that investigated the association between PFOA and cancer (see PFOA
Appendix). Two of the publications {Girardi, 2019, 6315730; Steenland, 2015, 2851015} were
occupational studies and the remainder were conducted on the general population, with one in a
high-exposure community (C8 Health Project). Different study designs were also used including
four cohort studies {Fry, 2017, 4181820; Girardi, 2019, 6315730; Steenland, 2015, 2851015; Li,
2022, 9961926}, five case-control studies {Wielsoe, 2017, 3858479; Tsai, 2020, 6833693; Lin,
2020, 6835434; Itoh, 2021, 9959632; Liu, 2021, 10176563}, five nested case-control studies
{Mancini, 2020, 5381529; Ghisari, 2017, 3860243; Shearer, 2021, 7161466; Hurley, 2018,
5080646; Cohn, 2020, 5412451}, and three cross-sectional studies {Christensen, 2016, 3858533;
Ducatman, 2015, 3859843; Omoike, 2021, 7021502}. The studies were conducted in different
3-263
-------
DRAFT FOR PUBLIC COMMENT
March 2023
study populations including populations from China {Lin, 2020, 6835434; Liu, 2021,
10176563}, Denmark {Ghisari, 2017, 3860243}, France {Mancini, 2020, 5381529}, Greenland
{Wiels0e, 2017, 3858479}, Italy {Girardi, 2019, 6315730}, Japan {Itoh, 2021, 9959632},
Sweden {Li, 2022, 9961926}, Taiwan {Tsai, 2020, 6833693}, and the United States {Fry, 2017,
4181820; Christensen, 2016, 3858533; Ducatman, 2015, 3859843; Steenland, 2015, 2851015;
Shearer, 2021, 7161466; Hurley, 2018, 5080646; Cohn, 2020, 5412451; Omoike, 2021,
7021502}. All studies measured PFOA in study subjects' blood components (i.e., serum or
plasma) with two exceptions: one study measured PFOA in the maternal serum {Cohn, 2020,
5412451} and one study categorized exposure to any PFAS based on residence near highly
contaminated sources of drinking water {Li, 2022, 9961926}. Cancers evaluated included
bladder {Steenland, 2015, 2851015; Li, 2022, 9961926}, breast {Cohn, 2020, 5412451; Ghisari,
2017, 3860243; Hurley, 2018, 5080646; Itoh, 2021, 9959632; Li, 2022, 9961926; Mancini, 2020,
5381529; Omoike, 2021, 7021502; Tsai, 2020, 6833693; Wiels0e, 2017, 3858479}, colorectal
{Steenland, 2015, 2851015; Li, 2022, 9961926}, germ cell tumors {Lin, 2020, 6835434}, kidney
{Shearer, 2021, 7161466; Li, 2022, 9961926}, liver {Girardi, 2019, 6315730; Li, 2022,
9961926}, lung {Girardi, 2019, 6315730; Li, 2022, 9961926}, lymphatic or hematopoietic tissue
{Girardi, 2019, 6315730; Li, 2022, 9961926}, melanoma {Steenland, 2015, 2851015; Li, 2022,
9961926}, ovarian {Omoike, 2021, 7021502}, prostate {Steenland, 2015, 2851015; Ducatman,
2015, 3859843; Omoike, 2021, 7021502}, thyroid {Liu, 2021, 10176563} uterine {Omoike,
2021, 7021502}, and any cancer {Christensen, 2016, 3858533; Fry, 2017, 4181820; Girardi,
2019, 6315730; Li, 2022, 9961926}.
3.5.1.2 Study Quality
There are 16 studies from recent systematic literature search and review efforts conducted after
publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} that investigated the
association between PFOA and cancer effects. Study quality evaluations for these 16 studies are
shown in Figure 3-72.
Of the 16 studies identified since the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}, seven
were considered medium confidence, and seven were low confidence {Christensen, 2016,
3858533; Girardi, 2019, 6315730; Itoh, 2021, 9959632; Lin, 2020, 6835434; Liu, 2021,
10176563; Omoike, 2021, 7021502; Steenland, 2015, 2851015; Tsai, 2020,6833693}. One study
conducted in the high exposure to PFAS Ronneby Register Cohort in Sweden was uninformative
{Li, 2022, 9961926} because of concerns about exposure assessment and lack of data on
important covariates. One study conducted in Greenland was uninformative {Wielsoe, 2017,
3858479} because of concerns about selection bias and exposure assessment. As a result, these
two studies will not be further considered in this review. Concerns with the low confidence
studies included the possibility of outcome misclassification, confounding, or participation
selection methods. Residual confounding was also a concern, including lack of considering co-
exposures by other PFAS, and lack of appropriately addressing SES and other lifestyle factors,
which could be associated with both exposure and cancer diagnosis. The two low confidence
occupational studies {Girardi, 2019, 6315730; Steenland, 2015, 2851015} had several potential
sources of bias including potential selection bias, outcome measurement limitations which may
lead to survival bias, and poor/insufficient study sensitivity due to a small number of deaths.
Girardi et al. (2019, 6315730) had the potential for residual confounding because of use of
standardized mortality ratios (SMRs), which only account for gender, age, and calendar year.
3-264
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Confounders specific for cancer outcomes, besides age and gender, including factors such as
smoking or socioeconomic factors were not addressed in the study and behavioral risk factors
could have differed by outcome. Although PFOA has a long half-life in the blood, concurrent
measurements may not be appropriate for cancers with long latencies. Temporality of exposure
in terms of cancer development was noted to be an issue in several low confidence studies {Tsai,
2020, 6833693; Itoh, 2021, 9959632; Liu, 2021, 10176563; Omoike, 2021, 7021502}. Many of
the low confidence studies also had sensitivity issues due to limited sample sizes.
Avi® c,e^ a\cP
Christensen et al., 2016, 3858533 -
Cohn et al., 2020, 5412451
Ducatman et al., 2015, 3859843
Fryetal., 2017, 4181820
Ghisari et al., 2017, 3860243 -
Girardi et al., 2019, 6315730 -
Hurley etal.,2018, 5080646-
Itoh et al., 2021,9959632-
Li et al., 2022, 9961926-
Lin et al., 2020, 6835434 -
Liu et al., 2021, 10176563-
Mancinietal., 2019, 5381529
Omoike et al., 2020, 7021502
Shearer etal., 2021,7161466
Tsai et al., 2020, 6833693 -
Wielsoe et al., 2017, 3858479
I Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Figure 3-72. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cancer Effects
Interactive figure and additional study details available on HAWC.
3.5.1.3 Findings from Children
One low confidence study examined cancers in children {Lin 2020, 6835434} and reported a
statistically significant higher median PFOA concentration in 42 pediatric germ cell tumor cases
3-265
-------
DRAFT FOR PUBLIC COMMENT
March 2023
compared to 42 controls in blood samples collected from the children one week after diagnosis.
However, the study did not observe an increase in cancer risk when evaluated on a per ng/mL
increase in blood PFOA.
3.5.1.4 Findings from the General Adult Population
PFOA was associated with an increased risk of kidney cancer (i.e., renal cell carcinoma (RCC))
{Shearer, 2021, 7161466}. This large medium confidence case-control study nested within the
National Cancer Institute's (NCI) Prostate, Lung, Colorectal, and Ovarian Screening Trial
(PLCO) reported a statistically significant increase in risk of RCC with pre-diagnostic serum
levels of PFOA (OR = 2.63; 95% CI: 1.33, 5.20 for the highest vs. lowest quartiles; p-
trend = 0.007, or per doubling of PFOA: OR: 1.71; 95% CI: 1.23, 2.37) {Shearer, 2021,
7161466}. Even after adjusting for other PFAS the association remained significant in analyses
on a per doubling increase in PFOA. The increase in odds remained across the quartiles and the
magnitude was similar (i.e., OR = 2.63 without adjusting for other PFAS vs. 2.19 after adjusting
for other PFAS in the highest vs. lowest quartiles), although it was no longer statistically
significant. Statistically significant increased odds of RCC were observed in participants ages
55-59 years, and in men and in women, separately (See PFOA Appendix).
Seven general population studies published since the 2016 assessment examined breast cancer
{Cohn, 2020, 5412451; Ghisari, 2017, 3860243; Hurley, 2018, 5080646; Itoh, 2021, 9959632;
Mancini, 2020, 5381529; Omoike, 2021, 7021502; Tsai, 2020, 6833693}. Four were considered
medium confidence {Cohn, 2020, 5412451; Ghisari, 2017, 3860243; Hurley, 2018, 5080646;
Mancini, 2020, 5381529} and had mixed results. All studies were case-control studies (with
some nested designs), except for one cross-sectional NHANES-based study {Omoike, 2021,
7021502}. Two nested case-control studies did not observe an association between breast cancer
and PFOA concentrations measured in maternal serum throughout pregnancy and 1-3 days after
delivery ({Cohn, 2020, 5412451}; 75th percentile PFOA 0.6 ng/mL) or in in serum after case
diagnosis and breast cancer ({Hurley, 2018, 5080646}; max concentration of 39.1 ng/mL). Both
studies were conducted in California and most breast cancer cases were obtained from the cancer
registry. Two nested case-control studies found associations between PFOA and breast cancer,
but only in specific genotype or estrogen receptive groups of participants {Ghisari, 2017,
3860243; Mancini, 2020, 5381529}. Ghisari (2017, 3860243) reported an increased risk for
breast cancer identified from the cancer registry with increasing PFOA concentrations only in
participants with a CC genotype (n = 36 cases and 47 controls) in the CYP19 gene (cytochrome
P450 aromatase). A nested case-control study (194 pairs of breast cancer cases and controls)
within the French E3N cohort found an 86% higher risk of breast cancer in the 2nd quartile of
PFOA (4.8-6.8 ng/mL) compared to the first quartile (1.3-4.8 ng/mL) (OR = 1.86; 95% CI:
1.03, 3.36) in a partially adjusted model {Mancini, 2020, 5381529}. Mancini et al. (2020,
5381529) also reported that the risk for breast cancer (93% verified as pathologically confirmed
from medical records after self-reported cancer diagnosis) varied by type of cancer with a
statistically significant increase in estrogen receptor negative (ER-) and progesterone receptor
negative (PR-) breast cancers in the second quartile of PFOA only. The sample size was small
with 26 participants having ER- breast cancers and 57 having PR- breast cancers. No
association was observed between PFOA and receptor-positive breast cancer risk.
Three studies were considered low confidence {Tsai, 2020, 6833693; Itoh, 2021, 9959632;
Omoike, 2021, 7021502} because of concerns about temporality of exposure measurements and
3-266
-------
DRAFT FOR PUBLIC COMMENT
March 2023
breast cancer development, lack of confirmation of control status via examination or medical
records {Tsai, 2020, 6833693}, and potential for residual confounding due to SES, life-style
factors and other PFAS. One low confidence study {Tsai, 2020, 6833693} conducted in Taiwan
observed a statistically significant increase in risk of breast cancer only in women younger than
50 years (OR = 1.14; 95% CI: 0.66, 1.96). Tsai et al. (2020, 6833693) also reported an increase
in risk in ER+ participants aged 50 years or younger and a decrease in risk for ER- breast
cancers in participants aged 50 years or younger, but neither achieved statistical significance.
Statistically significant increased odds of breast cancer were also observed in a low confidence
NHANES study (2005-2012) {Omoike, 2021, 7021502} both per ng/mL increase in PFOA
(OR = 1.089; 95% CI: 1.089, 1.090) and across quartiles of exposure. One low confidence case-
control study conducted in Japanese women {Itoh, 2021, 9959632} observed a significant
inverse association across serum PFOA quartiles with a significant dose-response trend (p-
value < 0.0001) (see PFOA Appendix). Median PFOA levels ranged from 3.2 ng/mL in the
lowest quartile to 9.3 ng/mL in the highest quartile. The association was null in pre-menopausal
women and remained significantly inverse in postmenopausal women in the highest tertile of
exposure, with a significant dose-response trend (p-value for trend = 0.005).
One medium confidence study based on the C8 Health Project {Ducatman, 2015, 3859843}.
examined prostate-specific antigen (PSA) as a biomarker for prostate cancer in adult males over
age 20 years who lived, worked, or went to school in one of the six water districts contaminated
by the DuPont Washington Works facility. No association was observed between PSA levels in
either younger (i.e., 20-49 year old) or older (i.e., 50-69 year old) men and concurrent mean
serum PFOA concentration up to 46 ng/mL. In an NHANES population, Omoike et al. (2021,
7021502) observed a significantly inverse association with prostate cancer (OR = 0.944; 95% CI:
0.943, 0.944).
Omoike et al. (2021, 7021502) also observed statistically significant increased odds of ovarian
cancer both per ng/mL increase in PFOA (OR = 1.015; 95% CI: 1.013, 1.017) and for the highest
vs. lowest quartiles of exposure (OR = 1.77; 95% CI: 1.75, 1.79), although the association was
significantly inverse for the second and third quartiles of exposure (see PFOA Appendix). A
significantly inverse association was also observed for uterine cancer (OR = 0.912; 95% CI:
0.910, 0.914 per ng/mL increase in PFOA) {Omoike, 2021, 7021502}.
One low confidence study conducted in Shandong Province, in eastern China {Liu, 2021,
10176563} observed a statistically significant inverse association with thyroid cancer across
quartiles of serum PFOA (p-value for trend < 0.001). The median serum PFOA levels were
higher in controls than in cases (10.9 vs. 7.7 ng/mL, p-value < 0.001). However, there is some
concern about possible reverse causality. The ability to metabolize PFAS could change when the
thyroid becomes cancerous, thereby changing the PFAS concentrations. The abnormality of
thyroid hormones may also disturb the PFAS levels.
Two studies examined all cancers together, but collected different information on cancers (i.e.,
incidence vs. mortality) and obtained the information using different methods. Cancer mortality
based on Public-use Linked Mortality Files was observed with PFOA exposure in a medium
confidence study among subjects over 60 years of age from NHANES 2003-2006 with median
PFOA concentration 23.7 ng/g lipid {Fry, 2017, 4181820}. PFOA was associated with an
increase in self-reported cancer incidence in a low confidence study on male anglers over
50 years {Christensen, 2016, 3858533}. Christensen et al. (2016, 3858533) was considered low
3-267
-------
DRAFT FOR PUBLIC COMMENT
March 2023
confidence due to the potential of self-selection because subjects were recruited from flyers and
other methods and filled out an online survey including self-reported outcomes.
3.5.1.5 Findings from Occupational Studies
Two low confidence occupational studies examined cancer incidence {Steenland, 2015,
2851015} and mortality {Girardi, 2019, 6315730}. Issues of population selection, outcome
measurement and small number of deaths reducing the sensitivity were noted. In a retrospective
occupational cohort study based on the same DuPont cohort from West Virginia reported in the
2016 assessment {Steenland, 2012, 2919168}, Steenland et al. (2015, 2851015) observed no
significant associations with incidence of cancers of the bladder, colorectal, prostate, and
melanoma when compared to the general population (median serum levels in workers was
113 ng/mL in 2005 compared to 4 ng/mL in the general population). There was modest evidence
of a positive non-significant trend for prostate cancer (across quartiles) and a statistically
significant negative exposure-response trend for bladder cancers (p-value = 0.04).
Girardi et al. (2019, 6315730) conducted a retrospective cohort study at a factory in Italy where
PFOA was produced from 1968-2014 and observed statistically significant increases in liver
cancer mortality, malignant neoplasms of the lymphatic and hematopoietic tissue, and in all
malignant neoplasms with cumulative serum PFOA exposure of > 16,956 ng/mL-years. There
was no association observed with lung cancer in this occupational cohort. Mortality from cancers
in this cohort was low and supplemental data provided mortality for other cancers including
kidney, but no risk estimates were calculated.
3.5.2 Animal Evidence Study Quality Evaluation and
Synthesis
There are 2 studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and 2 studies from
recent systematic literature search and review efforts conducted after publication of the 2016
PFOA HESD that investigated the association between PFOA and cancer effects in animal
models. Study quality evaluations for these 4 studies are shown in Figure 3-73.
3-268
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Abdellatif et al., 1991, 1290862-
i
+
"
NR
"
NR
i
"
"
Biegel etal., 2001, 673581 -
++
++
NR
++ ++
+
Butenhoff et al., 2012, 2919192-
B
++
NR
-
+
++
NTP, 2020, 7330145-
++
++
NR
++
++
++
s
Legend
| Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NR Not reported
Figure 3-73. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA and Cancer Effects
Interactive figure and additional study details available on HAWC.
Three high or medium confidence animal carcinogenicity studies indicate that PFOA exposure
can lead to multiple types of neoplastic lesions including liver adenomas {Biegel, 2001, 673581;
NTP 2020, 7330145} or carcinomas {NTP, 2020, 7330145}, Leydig cell tumors (LCTs) {Biegel,
2001, 673581; Butenhoff, 2012, 2919192}, and pancreatic acinar cell tumors (PACTs) {Biegel,
2001, 673581; NTP 2020, 7330145} in male Sprague-Dawley rats. Neoplastic lesions were also
observed in female Sprague-Dawley rats, but the incidence was not as high as observed in the
males and often did not achieve statistical significance {Butenhoff, 2012, 2919192; NTP, 2020,
7330145}. NTP (2020, 7330145) reported increased incidences of neoplastic lesions in female
Sprague-Dawley rats, though these changes were not statistically significant, statistics could not
be computed (liver neoplasms and PACTs), or there was uncertainty regarding the strength of
response compared to controls (uterine adenocarcinomas). Another study {Filgo, 2015,
2851085} assessed hepatic tumor development in three strains of female mice after perinatal
exposures to PFOA. This study was not evaluated and is not further discussed here because of an
inadequate study design to assess lifetime/chronic carcinogenicity (i.e., the study did not include
exposure postweaning) and the results were equivocal (i.e., few significant findings that did not
display a dose-response relationship) and difficult to interpret due to small sample sizes (n = 6-
10 for some strains).
In the three rat carcinogenicity studies {Biegel, 2001, 673581; Butenhoff, 2012, 2919192; NTP,
2020, 7330145}, rats were fed diets containing similar concentrations of PFOA for
approximately 2 years. Butenhoff et al. (2012, 2919192) analyzed a variety of tissues collected
3-269
-------
DRAFT FOR PUBLIC COMMENT
March 2023
from male and female Sprague-Dawley rats fed diets containing 0, 30, or 300 ppm PFOA
(equivalent to 1.3 and 14.2 mg/kg for males and 1.6 and 16.1 mg/kg for females) for 2 years.
Similarly, Biegel et al. (2001, 673581) analyzed tissues collected from male Crl:CD® BR (CD)
rats fed diets containing 0 or 300 ppm PFOA (equivalent to 13.6 mg/kg/day) for 24 months.
Using a matrix-type exposure paradigm, NTP (2020, 7330145) administered PFOA in feed to
pregnant Sprague-Dawley (Hsd:Sprague Dawley® SD®) rats starting on GD 6 and analyzed
tissues of male and female offspring also fed postweaning diets containing PFOA for a total of
107 weeks. Dose groups for this report are referred to as "[perinatal exposure
level]/[postweaning exposure level]" (e.g. 300/1,000; see further study design details in Section
3.4.4.2.1.2).
Liver adenomas were observed in the Biegel et al. study (2001, 673581) at an incidence of 10/76
(13%) at 13.6 mg/kg/day, compared to 2/80 (3%) in controls. Liver adenomas were also
significantly increased in the NTP (2020, 7330145) in the 0/40, 0/80, and 300/80 ppm groups
(Table 3-11). Both the 0/0 and 300/0 ppm control groups had no observed liver adenomas.
Although no liver adenomas were observed in Butenhoff et al. (2012, 2919192), carcinomas
were identified in the male controls, males in the low-dose group (1.3 mg/kg/day), and male and
female rats in the high-dose group (14.2 and 16.1 mg/kg/day, respectively). NTP (2020,
7330145) reported increases in the incidence of hepatocellular carcinomas in the 300/80 ppm
group. The differences in carcinoma incidences from controls were not statistically significant in
either the Butenhoff et al. (2012, 2919192) or NTP (2020, 7330145) studies.
Table 3-11. Incidences of Liver Adenomas in Male Sprague-Dawley Rats as Reported by
NTP (2020, 7330145)
Perinatal Dose
Postweaning Dose
0 ppm
20 ppm
40 ppm
80 ppm
0 ppm
0/50 (0%)***
0/50 (0%)
7/50 (14%)*
11/50 (22%)**
300 ppm
0/50 (0%)***
1/50 (2%)
5/50 (10%)
10/50 (20%)**
Notes:
* Statistically significant compared to the respective control group (0/0 or 300/0 ppm) at p < 0.05.
**Statistically significant compared to the respective control group (0/0 or 300/0 ppm) at p < 0.01.
* "Statistically significant trend of response at p < 0.001.
Accompanying non-neoplastic/preneoplastic liver lesions were identified by Butenhoff et al.
(2012, 2919192) in males and females at the 1- and 2-year sacrifices. An increased incidence of
diffuse hepatomegalocytosis and hepatocellular necrosis occurred in the high-dose groups. At the
2-year sacrifice, hepatic cystic degeneration (characterized by areas of multilocular microcysts in
the liver parenchyma) was observed in males. Hyperplastic nodules in male livers were increased
in the 14.2 mg/kg/day group. NTP similarly reported a variety of non-neoplastic and
preneoplastic liver lesions in both male and female rats including increased incidences of liver
necrosis and mixed-cell foci, hepatocyte hypertrophy, and focal inflammation. These lesions
were more pronounced in males than females and were observed at both the 16-week interim and
107-week final necropsies.
Testicular LCTs were identified in both the Butenhoff et al. (2012, 2919192) and Biegel et al.
(2001, 673581) studies. The tumor incidence reported by Butenhoff et al. (2012, 2919192) was
3-270
-------
DRAFT FOR PUBLIC COMMENT
March 2023
0/50 (0%), 2/50 (4%), and 7/50 (14%) for the 0, 1.3, and 14.2 mg/kg/day dose groups,
respectively. The Biegel et al. study (2001, 673581) included one dose group (13.6 mg/kg/day);
the tumor incidence was 8/76 (11%) compared to 0/80 (0%) in the control group. LCT incidence
at similar dose levels was comparable between the two studies (11% and 14%). NTP (2020,
7330145) analyzed testicular tissue for LCTs but did not observe increased incidence due to
PFOA treatment. The authors noted that this inconsistency with other carcinogenicity studies
could be a result of differences in exposure concentrations or stock of Sprague-Dawley rat (i.e.
CD vs. Hsd:Sprague Dawley).
PACTs were observed in both the NTP (2020, 7330145) and Biegel et al. (2001, 673581)
studies. NTP (2020, 7330145) reported increased incidences of pancreatic acinar cell adenomas
in males in all treatment groups compared to their respective controls (Table 3-12). NTP (2020,
7330145) observed increases in pancreatic acinar cell adenocarcinoma incidence in males in
multiple dose groups and slight increases in the incidence of combined acinar cell adenoma or
carcinoma in females from the 300/1,000 ppm dose group, though these increases did not reach
statistical significance. In male rats, the incidence of PACTs in the Biegel et al. (2001, 673581)
study was 8/76 (11%; 7 adenomas, 1 carcinoma) at 13.6 mg/kg/day while none were observed in
the control animals. In a peer-reviewed pathological review of male pancreatic tissue collected
by Butenhoff et al. (2012, 2919192), Caverly Rae et al. (2014, 5079680) identified 1/47
carcinomas in the 300 ppm group (compared to 0/46 in the control and 30 ppm groups) and no
incidence of adenomas with any treatment. Pancreatic acinar hyperplasia was observed in males
of the control, 1.3, and 14.2 mg/kg/day groups at incidences of 3/46 (7%), 1/46 (2%), and 10/47
(2P/o), respectively. Butenhoff et al. (2012, 2919192) also reported increased incidences of
acinar atrophy in males (6/46 (13%>), 9/46 (20%>), and 11/49 (22%>) in 0, 1.3, and 14.2 mg/kg/day
dose groups, respectively), though this lesion was not discussed in the peer-reviewed pathology
report {Caverly Rae, 2014, 5079680}. NTP (2020, 7330145) similarly reported increased
incidences of acinus hyperplasia in males at incidence rates of 32/50 (64%>), 37/50 (74%>), 31/50
(62%) in the 0/20, 0/40, 0/80, and 27/50 (54%), 38/50 (76%), and 33/50 (66%) in the 300/20,
300/40, and 300/80 groups. The incidences in controls were 18/50 (36%) and 23/50 (46%) in the
0/0 and 300/0 groups, respectively. There were also low occurrences of acinus hyperplasia in the
exposed female groups, though not as frequently observed as in males. However, the authors
concluded that the incidence of pancreatic acinar cell neoplasms in males increased confidence
that the occurrence in females was due to PFOA exposure.
Table 3-12. Incidences of Pancreatic Acinar Cell Adenomas in Male Sprague-Dawley Rats
as Reported by NTP (2020, 7330145)
Perinatal Dose
Postweaning Dose
0 ppm
20 ppm 40 ppm
80 ppm
0 ppm
3/50 (6%)**
28/50 (56%)** 26/50 (52%)**
32/50 (64%)**
300 ppm
7/50 (14%)**
18/50 (36%)* 30/50 (60%)**
30/50 (60%)**
Notes:
* Statistically significant compared to the respective control group (0/0 or 300/0 ppm) at p < 0.05.
**Statistically significant compared to the respective control group (0/0 or 300/0 ppm) at p < 0.001. Asterisks on the control
group denotes a statistically significant trend of response.
3-271
-------
DRAFT FOR PUBLIC COMMENT
March 2023
NTP (2020, 7330145) observed increased incidences of uterine adenocarcinomas in female
Sprague-Dawley rats during the extended evaluation (i.e., uterine tissue which included cervical,
vaginal, and uterine tissue remnants). Incidence rates for this lesion are reported in Table 3-13.
The accompanying incidences of uterine hyperplasia did not follow a dose-response relationship.
Butenhoff et al. (2012, 2919192) identified mammary fibroadenomas and ovarian tubular
adenomas in female rats, though there were no statistical differences in incidence rates between
PFOA-treated dose groups and controls.
Table 3-13. Incidences of Uterine Adenocarcinomas in Female Sprague-Dawley Rats from
the Standard and Extended Evaluations (Combined) as Reported by NTP (2020, 7330145)
Perinatal Dose
Postweaning Dose
0 ppm
300 ppm
1,000 ppm
0 ppm
1/50 (2%)
5/49 (10%)
7/48 (15%)*
150 ppm
-
3/50 (6%)
-
300 ppm
-
-
5/48 (10%)
Notes:
* Statistically significant compared to the control group (0/0 ppm) at p = 0.050.
NTP concluded that under the exposure conditions presented, there was clear evidence of
carcinogenic activity of PFOA in male Sprague Dawley rats based on increased incidences of
hepatocellular neoplasms (predominately hepatocellular adenomas) and acinar cell neoplasms
(predominately acinar cell adenomas) of the pancreas {NTP, 2020, 7330145}. In females, NTP
concluded there was some evidence of carcinogenic activity of PFOA based on increased
incidences of pancreatic acinar cell adenoma or adenocarcinoma (combined) neoplasms. The
study authors also noted that the higher incidence of hepatocellular carcinomas and
adenocarcinomas of the uterus may have been related to exposure.
3.5.3 Mechanistic Evidence
Mechanistic evidence linking PFOA exposure to adverse cancer outcomes is discussed in
Sections 3.1.2, 3.2.9, 3.3.1, 3.4.2, 3.4.3, 3.4.4, and 4.2 of the 2016 PFOA HESD {U.S. EPA,
2016, 3603279}. There are 40 studies from recent systematic literature search and review efforts
conducted after publication of the 2016 PFOA HESD that investigated the mechanisms of action
of PFOA that lead to cancer effects. A summary of these studies is shown in Figure 3-74.
Evidence Stream
Animal Human In Vitro Grand Total
Figure 3-74. Summary of Mechanistic Studies of PFOA and Cancer Effects
Interactive figure and additional study details available on Tableau.
In 2016, ten key characteristics of carcinogens were selected by a multi-disciplinary working
group of the International Agency for Research on Cancer (IARC), based upon common
empirical observations of chemical and biological properties associated with human carcinogens
3-272
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(i.e., Group 1 carcinogens as determined by IARC) {Smith, 2016, 3160486}. In contrast to the
"Hallmarks of cancer" as presented by Hanahan and Weinberg {Hanahan, 2022, 10164687;
Hanahan, 2011, 758924; Hanahan, 2000, 188413}, the key characteristics focus on the properties
of human carcinogens that induce cancer, not the phenotypic or genotypic traits of cancers. The
ten key characteristics provide a framework to systematically identify, organize, and summarize
mechanistic information for cancer hazard evaluations {Smith, 2016, 3160486}.
To aid in the evaluation of the carcinogenic potential of PFOA, the studies containing
mechanistic data were organized by the proposed key characteristics of carcinogens for the
following section. Evidence related to eight of the ten key characteristics of carcinogens was
identified in the literature included in this assessment: 'Is Genotoxic', 'Induces Epigenetic
Effects', 'Induces Oxidative Stress', 'Induces Chronic Inflammation', 'Is Immunosuppressive',
'Modulates Receptor Mediated Effects', 'Alters Cells Proliferation, Cell Death, and Nutrient
Supply', and 'Causes Immortalization'. No studies from the 2016 PFOA HESD {U.S. EPA,
2016, 3603279} and recent systematic literature search and review efforts were identified for the
following key characteristics: 'Is Electrophilic or Can Be Metabolically Activated to
Electrophiles' and 'Alters DNA Repair and Causes Genomic Instability'.
3.5.3.1 Key Characteristic #2: Is Genotoxic
Genotoxicity is a well-studied mode of action for carcinogens, defined as alterations to DNA
through single or double strand breaks, alterations to DNA synthesis, and DNA adducts, all of
which can result in chromosomal aberrations, formation of micronuclei, and mutagenesis if not
effectively repaired.
3.53.1.1 Mutagenicity
All of the studies investigating the mutagenic potential of PFOA were conducted in in vitro
models. Of the available studies, most found that PFOA exposure did not induce mutagenicity
(Table 3-14). Studies involving Chinese hamster ovary (CHO) K-l cell lines presented primarily
negative results. Sadhu (2002, 10270882) reported PFOA exposure did not induce gene
mutations in CHO K-l cells when tested with or without metabolic activation. Zhao et al. (2011,
847496) also observed that PFOA did not induce mutagenesis in human-hamster hybrid (Al)
cells, which contain a standard set of CHO-K1 chromosomes and a single copy of human
chromosome 11, at sub-cytotoxic concentrations (<200 |iM). A subsequent experiment using
DMSO to quench oxidative stress found that PFOA was not mutagenic in the presence of
DMSO, suggesting that an increase in reactive oxygen species production may be required for
PFOA-induced mutagenicity (Section 3.5.3.3).
Of the six publications that tested PFOA mutagenicity in Salmonella typhimurium (S.
typhimurium) or Escherichia coli (E. coli) {NTP, 2019, 5400977; Butenhoff, 2014, 5079860;
Buhrke, 2015, 2850235; Fernandez Friere, 2008, 2919390; Lawlor, 1995, 10228128; Lawlor,
1996, 10228127}, two reported exposure-associated mutagenicity {NTP, 2019, 5400977;
Butenhoff, 2014, 5079860} (Table 3-14). Mutation was observed in S. typhimurium following
exposure to cytotoxic concentrations of PFOA in the presence of S9 metabolic activation
{Butenhoff, 2014, 5079860}. NTP (2019, 5400977) reported PFOA exposure caused a slight
increase in mutation in S. typhimurium TA98 cells, and Lawlor (1996, 10228127) reported that
one plate of S. typhimurium had a significant amount of mutagenicity in the absence of S9
metabolic activation. However, neither of these results were reproducible.
3-273
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.5.3.1.2 DNA Damage
¦ In Vivo Evidence
3.5.3.1.2.1.1 Human Studies
Two studies have reported on the genotoxic potential of PFOA exposure in humans (Table 3-15).
Franken et al. (2017, 3789256) measured blood PFOA concentrations in adolescents (14-15
years of age) that resided for >5 years within industrial areas of Belgium (near a stainless-steel
plant or a shredder factory). These data were then compared to age-matched controls. A
significant increase in DNA damage associated with PFOA exposure was observed, as evidenced
by an alkaline comet assay performed on the same blood samples. Urinary 8-
hydroxydeoxyguanosine (8-OHdG) was used as a biomarker for oxidative DNA damage. While
there was no significant change observed, a positive dose-response relationship with increasing
PFOA concentrations was noted. The authors attributed the DNA damage to oxidative stress, but
noted that urinary 8-OHdG can also indicate DNA repair. Governini et al. (2015, 3981589)
collected semen samples from healthy nonsmoking men and evaluated aneuploidy, diploidy, and
DNA fragmentation. The occurrence of aneuploidy and diploidy in sperm cells, which are
normally haploid, was significantly higher in the PFAS-positive samples (PFOA was detected in
75% of the samples) when compared to PF AS-negative samples. This suggests that PFAS
exposure is related to errors in cell division leading to aneugenicity. Additionally, fragmented
chromatin levels were also significantly increased for the PF AS-positive group compared with
the PFAS-negative group.
3.5.3.1.2.1.2 Animal Toxicological Studies
Studies of the genotoxicity related to PFOA exposure were conducted in rat and mouse models
(Table 3-15). All of the studies presented data from micronucleus tests of bone marrow,
peripheral blood, and splenocytes, with the exception of one study of DNA strand breaks. It is
important to note that rat models could be ineffective for determining micronucleus formation if
study authors do not use appropriate methodologies as the spleen will remove micronucleated
cells {Schlegel, 1984, 10368697}. However, this would generally bias studies towards the null,
not result in false positives.
With the exception of one micronucleus assay, there was no evidence for PFOA-induced
genotoxic effects after acute or subchronic exposures (Figure 3-15).The single study of DNA
strand breakage used a comet assay in tissues from male C57B1/6 mice administered <5
mg/kg/day for five weeks {Crebelli, 2019, 5381564}. Analysis of the liver and testis following
exposure indicated there was no change in DNA fragmentation. Acute and subchronic PFOA
exposures in mouse studies found no evidence of micronuclei formation, a measure of genotoxic
damage to DNA in proliferating cells and spindle formation {Hayashi, 2016, 9956921}, in either
peripheral blood cells or splenocytes {Crebelli, 2019, 5381564} or within erythrocytes of the
bone marrow {Butenhoff, 2014, 5079860; Murli, 1995, 10228120; Murli, 1996, 10228121}.
NTP (2019, 5400977) assessed micronuclei formation in Sprague Dawley rats using flow
cytometry to avoid the potential confounding effect of splenic filtration {Dertinger, 2004,
10328871; Schlegel, 1984, 10368697}. A subchronic study in Sprague Dawley rats noted that
PFOA exposure induced a slight increase in micronuclei formation in peripheral blood cells of
male rats administered 10 mg/kg/day; however, the micronuclei level was within the historical
control range, and there was no effect in females) {NTP, 2019, 5400977}.
3-274
-------
DRAFT FOR PUBLIC COMMENT
March 2023
¦ In Vitro Evidence
3.5.3.1.2.1.3 Chromosomal Aberrations
Measurements of chromosomal aberrations have been performed using human and animal cell
lines, and predominantly found that PFOA exposure does not cause alterations (Table 3-16). In
human lymphocytes, PFOA did not induce chromosomal aberrations in the presence of S9
activation (3 hours) or without the addition of S9 (<46 hours) at concentrations up to 600 |ig/ml
{Butenhoff, 2014, 5079860}. This evidence corroborates previous studies of human lymphocyte
cells that found similar results using non-cytotoxic concentrations of PFOA {Murli, 1996,
10228126; NOTOX, 2000, 10270878} as reported in the 2016 PFOA HESD {U.S. EPA, 2016,
3603279}.
In contrast, Butenhoff et al. (2014, 5079860) observed chromosomal aberrations after PFOA
exposure (>750 |ig/ml) with S9 metabolic activation in CHO cells. These results corroborate
with previously reported studies in S9 activated CHO cells {Murli, 1996, 10228125; Murli,
1996, 10228124}. Butenhoff et al. (2014, 5079860) and Murli (1996, 10228124) also reported
PFOA-induced chromosomal aberrations in CHO cells without S9 metabolic activation but were
unable to replicate their own results.
3.5.3.1.2.1.4 DNA Double Strand Breaks
Evaluation of DNA strand breakage using comet assays and histological analysis of
phosphorylated H2AX (yH2AX) yielded positive results in all of the studies reviewed (Table
3-16). PFOA exposure caused DNA breakage in a dose-dependent manner in human
lymphocytes exposed to >250 ppm PFOA for two hours {Yahia, 2016, 2851192} and in HepG2
cells exposed to >100 |iM PFOA for 24 hours in one study {Yao and Zhong, 2005, 5081563},
>10 |iM PFOA for 24 hours in another study {Wiels0e, 2015, 2533367}, and at 10 and 200 |iM
PFOA (but not 50 or 100 |iM PFOA) for 24 hours in a third study {Florentin et al., 2011,
2919235}. Paramecium caudatum (P. caudatum), a unicellular protozoa, exhibited DNA damage
after exposure to 100 |iM PFOA {Kawamoto, 2010, 1274162}. Peropadre et al. (2018, 5080270)
observed a 4.5-fold higher level of double strand breaks in human keratinocyte cells (HaCaT)
exposed to 50 |iM PFOA for 24 hours, compared to controls, as evidenced by yH2AX. Eight
days post-exposure, yH2AX levels were twice that of the controls, indicating that double strand
breaks were not fully repaired. In contrast, a study conducted in Syrian hamster embryo (SHE)
cells demonstrated no change in DNA strand breaks by the comet assay at 4.1 x 10~5 to 300 [xM
PFOA for 5 or 24 hours {Jacquet et al., 2012, 2124683}.
3.5.3.1.2.1.5 Micronuclei Formation
Three studies measured micronucleus formation in cells exposed to PFOA (Table 3-16). Buhrke
et al. (2013, 2325346) demonstrated that PFOA exposure (10 |iM, 24 hours) did not induce
micronuclei formation in Chinese hamster lung cells (V79). Studies conducted in human HepG2
cells reported conflicting results: in one study, PFOA induced micronuclei formation at
concentration of >100 |iM after 24 hours {Yao and Zhong, 2005, 5081563, while another study
reported no difference in micronuclei frequency in HepG2 cells exposed to concentrations of
PFOA up to 400 |iM for 24 hours compared to controls {Florentin et al., 2011, 2919235}. The
micronucleus assays were performed according to the same method {Natarajan, 1991, 5143588}.
3-275
-------
DRAFT FOR PUBLIC COMMENT
Table 3-14. Mutagenicity Data from In Vitro Studies
March 2023
Reference
Cell Line or
Results
Concentration
Bacterial Strain
S9-Activated
Non-Activated
(Duration of exposure)
NTP (2019,
5400977)
Salmonella
typhimurium (TA98,
TA100)
Equivocal3
(Not reproducible)
Equivocal3
(Not reproducible)
100-5,000 (ig/plate
Escherichia coli
(WP2wvrA/pkM101)
Negative
Negative
100-10,000 (ig/plate
Zhao et al.
Human-hamster
N/A
Positiveb
1-200 |iM
(2011,
847496)
hybrid (Al) cells
(1-16 days)
Mitochondrial DNA-
deficient human-
N/A
Negative
1-200 |iM
(1-16 days)
hamster hybrid
(P°Al) cells
Sadhu
(2002,
10270882)
CHOK-1
Negative
Negative
<39 ng/mL
(5 or 17 hours)
Butenhoff et Salmonella
Positive0
Negative
20-1,000 (ig/plate
al. (2014,
5079860)
typhimurium (TA98,
TA100, TA1535,
TA1537)
Buhrke et
al. (2015,
2850235)
Salmonella
typhimurium (TA98,
TA100, TA1535,
TA1537, TA1538)
Negative
Negative
5 |iM
Fernandez
Friere et al.
(2008,
Salmonella
typhimurium (TA98,
TA100, TA102,
Negative
Negative
100 or 500 (iM
2919390)
TA104)
Lawlor
(1995,
10228128)
Salmonella
typhimurium (TA98,
TA100, TA1535,
TA1537)
Negative
Negative
100-5,000 (ig/plate
Salmonella
typhimurium (TA98,
TA100, TA1535,
Negative
Negative
100-5,000 (ig/plate
TA1537)
Escherichia coli
(WP2uvrA)
Negative
Negative
100-5,000 (ig/plate
Escherichia coli
(WP2uvrA)
Negative
Negative
6.67-5,000 (ig/plate
Notes:
a Mutagens were present in 1 of 3 TA98 replicate plates only.
b Mutagens were present in cells that were exposed only to 200 (iM for 16 days.
c Mutagenicity found at cytotoxic concentrations only.
Table 3-15. DNA Damage Data from In Vivo Studies
Reference
Species, Strain
Tissue Results
PFOA Concentration
(Sex)
(Dosing Regimen)
DNA Strand Breakage
Franken et al.
Human
Peripheral Blood Positive
Average Blood Concentration of
(2017, 3789256)
(Male and
Cells
2.55 ng/L
Female)
3-276
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Governini et al.
Human
Semen
Positive
Average Seminal Plasma Concentration
(2015, 3981589)
(Male)
of 7.68 ng/g f.w.
Crebelli et al.
(2019, 5381564)
Mouse,
C57BL/6
(Male)
Liver, Testis
Negative
0.1-5 mg/kg/day
(daily via drinking water for 5 weeks)
Micronuclei Formation
Crebelli et al.
(2019, 5381564)
Mouse,
C57BL/6
(Male)
Peripheral Blood
Cells, Splenocytes
Negative
0.1-5 mg/kg/day
(daily via drinking water for 5 weeks)
Butenhoff et al.
Mouse, Crl:CD-
Bone Marrow
Negative
250 - 1,000 mg/kg
(2014, 5079860)
1
(Male and
Female)
(single dose via gavage)
NTP (2019,
5400977)
Rat, Sprague
Dawley
(Male and
Female)
Peripheral Blood
Cells
Positive3
6.25-100 mg/kg/day
(daily via gavage for 28 days)
Murli (1995,
10228120)
Mouse
Mouse
Bone Marrow
Bone Marrow
Negative
Negative
1,250-5,000 mg/kg
(Single dose delivered via gavage)
498-1,990 mg/kg
(Single dose delivered via gavage)
Notes: f.w. = formula weight.
a A slight increase in micronuclei in the male 10 mg/kg/day group was within the historical control range. No change in females.
Table 3-16. DNA Damage Data from In Vitro Studies
Reference
In Vitro Model
Results
Concentration
S9 Activated Non-Activated
(Duration of exposure)
Chromosomal Aberrations
Butenhoff et
Human
Negative
Negative
12.4-600 ng/mL
al. (2014,
Lymphocytes
(3-46 hours)
5079860)
Chinese Hamster
Positive
N/A
50-1,500 ng/mL
Ovarian Cells
(3 hours)
Chinese Hamster
N/A
Positive
25-1,000 ng/mL
Ovarian Cells
(Not reproducible)
(3-41.8 hours)
NOTOX
Human
Negative
Negative
< Cytotoxic concentration3
(2000,
Lymphocytes
10270878)
Murli (1996,
Human
Negative
Negative
125-4,010 ng/mL
10228126)
Lymphocytes
(3-43.3 hours)
Chinese Hamster
Positive
Negative
100-2,750 ng/mL
Ovarian Cells
(3-41.8 hours)
Chinese Hamster
Positive
Positive
125-5,000 ng/mL
Ovarian Cells
(Not reproducible)
(3 hours)
Cell Transformation
Jacquet et al.
Syrian Hamster
N/A
Negative
3.7 x 10"4-37 |iM
(2012,
Embryo Cells
(6 days)
2124683)
Garry and
C3H10T!/2
N/A
Negative
0.1-200 |ig/mL
Nelson (1981,
(24 hours)
10228130)
DNA Strand Breakage
3-277
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Reference
In Vitro Model
Results
Concentration
S9 Activated
Non-Activated
(Duration of exposure)
Peropadre et
al. (2018,
5080270)
Human
Keratinocyte
HaCaT cells
N/A
Positive
50 MM
(24 hours)
Yahia et al.
Human
N/A
Positive
125-500 ppm (2 hours)
(2016,
2851192)
Lymphocytes
Florentin et al. Human HepG2
(2011, Cells
N/A
Positiveb
5-400 nM
(1 or 24 hours)
2919235)
Wiclsoc et al.
(2014,
2533367)
Human HepG2
Cells
N/A
Positive
0.2- 20 nM
(24 hours)
Yao and
Zhong (2005,
5081563)
Human HepG2
Cells
N/A
Positive
50-400 nM
(24 hours)
Kawamoto et
Paramecia
N/A
Positive
10-100 nM
al. (2010,
1274162)
(1-24 hours)
Micronuclei Formation
Buhrke et al.
(2013,
2325346)
Chinese Hamster
Lung Fibroblast
Cells
Negative
Negative
10 nM
(3 hours)
Florentin et al. Human HepG2
(2011, Cells
N/A
Negative
5-400 nM
(1 or 24 hours)
2919235)
Yao and
Zhong (2005,
5081563)
Human HepG2
Cells
N/A
Positive0
50-400 nM
(24 hours)
Notes: N/A = not applicable.
aFindings based on the 2016 EPA's Health Effects Support Document {U.S. EPA, 2016, 3603279}, concentrations) unknown.
b Slight increase was observed at 10 and 200 (iM in a non-dose-dependent manner after 24-hour exposure only.
c Micronuclei were present in cells that were exposed only to > 100 (iM for 16 days.
3.5.3.2 Key Characteristic #4: Induces Epigenetic Alterations
Epigenetic alterations are modifications to the genome that do not change genetic sequence.
Epigenetic alterations include DNA methylation, histone modifications, changes in chromatin
structure, and dysregulated microRNA expression, all of which can affect the transcription of
individual genes and/or genomic stability 1 Smith, 2016, 3160486}.
3.5.3.2.1 In Vivo Evidence
3.5.3.2.1.1 Humans
A cohort of singleton term births were recruited from Faroese hospitals over an eighteen-month
period from 1986 to 1987 {Leung, 2018, 4633577}. At delivery, samples of umbilical cord
whole blood and scalp hair from the mothers were collected and used to measure toxicant levels
as well as evaluation of DNA methylation. No change in CpG island methylation was correlated
with PFOA levels, although changes in this epigenetic alteration were found to be significantly
correlated with several other toxicants in the blood samples. Two other studies evaluated global
DNA methylation patterns in cord-blood. Miura et al. (2018, 5080353) found that increased
3-278
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA in the cord blood was associated with a global DNA hypermethylation in a cohort from
Japan. Kingsley et al. (2017, 3981315) did not observe associations between PFOA exposure in
cord blood and epigenome-wide changes in methylation status. However, the authors found
significant changes in methylation in seven CpG sites located in several genes, including RAS
P21 Protein Activator 3 (RASA3) and Opioid Receptor Delta 1 (OPRD1). Three studies reviewed
herein found no association between maternal PFOA exposure and global methylation changes in
offspring {Liu, 2018, 4926233; Leung, 2018, 4633577} or placenta {Ouidir, 2020, 6833759}.
A subset of adults enrolled in the C8 Health Project between August 1, 2005 and August 31,
2006 were evaluated for exposure to perfluoroalkyl acids (PFAAs) via drinking water {Watkins,
2014, 2850906}. The cross-sectional survey consisted of residents within the mid-Ohio River
Valley. A second, short-term follow-up study including another sample collection was conducted
in 2010 to evaluate epigenetic alterations in relation to serum PFOA concentrations. Serum
concentrations of PFOA significantly decreased between enrollment (2005-2006) and follow-up
(2010). However, methylation of long interspersed nuclear elements (LINE-1) transposable DNA
elements in peripheral blood leukocytes was not associated with PFOA exposure at any
timepoint.
Several studies detail the influence of PFOA exposure on the epigenome in humans. Specifically,
in prenatal studies, PFOA exposure was associated with mixed results of increased methylation
in cord blood but not in placenta. However, consistently, studies found alterations in methylation
patterns in genes associated with fetal growth. For additional information, please see the
developmental mechanistic section (Section 3.4.4.3; refer to the interactive Tableau for
additional supporting information and study details).
3.5.3.2.1.2 Animals
An in vivo analysis of epigenetic modifications in an oral PFOA study (1-20 mg/kg/day; 10
days) was performed in female CD-I mice {Rashid, 2020, 6315778}. Measurement of 5-
methylcytosine (5mc) and 5- hydroxymethylcytosine (5hmc) indicated no alteration of global
CpG methylation levels in the kidneys. Downregulation of DNA methyltransferase 1 (Dnmtl)
mRNA was observed at <5 mg/kg/day PFOA, while Dnmtl expression increased by 4- and 7-
fold at doses of 10 and 20 mg/kg/day, respectively. Levels of Dmnt3a decreased at all doses, and
Dnmt3b expression increased at the highest dose (20 mg/kg/day). mRNA expression of ten
eleven translocation (Tet) 1/2/3 methylcytosine dioxygenases was decreased at low doses of
PFOA exposure compared to controls, with no change at higher doses.
3.5.3.2.2 In Vitro Evidence
In vitro PFOA exposures have yielded mixed results with evidence of both hyper- and
hypomethylation of DNA. Data presented here are categorized by global DNA methylation and
gene-specific modifications.
3.5.3.2.2.1 Global DNA Methylation
5mC expression can be used to indicate global DNA methylation. Pierozan et al. (2020,
6833637) treated MCF-10A cells with PFOA (100 |iM, 72 hours) and found elevated global
methylation levels in the first daughter cell subculture. However, methylation levels returned to
baseline after the second passage. This study contrasts with the results of Wen et al. (2020,
6302274) in a study conducted in HepG2 cells (20-400 |iM PFOA, 48 hours), and Liu and
3-279
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Irudayaraj (2020, 6512127) in a study of MCF7 cells (20-400 |iM PFOA, 24-48 hours). Both
studies found dose-dependent reductions in 5mC after PFOA exposure.
3.5.3.2.2.2 Modification to Gene Expression
Assays evaluating gene expression modified by enzymes that regulate DNA methylation levels,
such as DNMT and TET enzymes, and histone modifications have been used to assess the impact
of PFOA on the epigenome. Liu and Irudayaraj (2020, 6512127) reported significantly lower
levels of DNMT1 protein after PFOA exposure in both MCF7 (>100 |iM) and HepG2 (>200
|iM) cells. However, DNMT3 A expression was increased in a dose-dependent manner in MCF7
cells (>200 |iM). Authors attributed PFOA-induced global demethylation to alterations of
DNMT3A and subsequent enzymatic activity of DNMT. Levels of DNMT3B did not change
significantly in either cell line. Wen et al. (2020, 6302274) found no significant changes to
DNMT1/3A/3B gene profiles after PFOA exposure (20-400 |iM, 48 hours) in HepG2 cells.
Further analysis found PFOA (200 |iM) decreased TET1 expression, which is strongly associated
with DNA methylation, but increased TET2 and TET3. Pierozan et al. (2020, 6833637) noted
that PFOA-exposed MCF-10A cells and the direct daughter cell passages contained decreased
levels of histone 3 lysine 9 dimethylation (H3K9me2). H3K9me2 is a silencing epigenetic
marker; thus, a decrease in H3K9me2 is indicative of transcriptional activation, and has been
associated with altered gene expression in breast cancer transformation.
3.5.3.3 Key Characteristic #5: Induce Oxidative Stress
Reactive oxygen and nitrogen species (ROS and RNS, respectively) are byproducts of energy
production that occur under normal physiological conditions. An imbalance in the detoxification
of reactive such species can result in oxidative (or nitrosative) stress, which can play a role in a
variety of diseases and pathological conditions, including cancer. The primary mechanism by
which oxidative stress leads to the carcinogenic transformation of normal cells is by inducing
oxidative DNA damage that leads to genomic instability and/or mutations 1 Smith, 2016,
3160486 J.
3.5.3.3.1 In Vivo Evidence
3.5.3.3.1.1 Humans
Franken et al. (2017, 3789256) measured urinary 8-OHdG to evaluate DNA induced by oxidative stress,
in adolescents (14 - 15 years of age) that resided for >5 years in industrial areas of Belgium and
compared their findings to blood PFOA concentrations. While no significant change was observed in
urinary 8-OHdG in the subjects when compared to that of age-matched controls, a positive dose-response
relationship with increasing PFOA concentrations was noted. The authors attributed the DNA damage to
oxidative stress but noted that elevated 8-OHdG could also reflect aberrant DNA repair.
3.5.3.3.1.2 Animals
Several in vivo analyses of PFOA exposure in rodents found evidence that PFOA exposure
caused increased oxidative stress and markers of oxidative damage in a tissue-specific manner.
Takagi et al. (1991, 2325496) performed a two-week subchronic study (0.02% powdered PFOA
in the diet) in male Fischer 344 rats and evaluated the levels of 8-OHdG in the liver and kidneys
after exposure. While a significant increase was noted in liver and kidney weights, elevated levels
of 8-OHdG was observed only in the liver. A second subset of animals were given a single IP
3-280
-------
DRAFT FOR PUBLIC COMMENT
March 2023
injection of PFOA (100 mg/kg) and sacrificed at days 1, 3, 5, and 8. Results were comparable to
that of the dietary exposure study, as PFOA significantly increased liver (by day 1) and kidney
(on days 3 and 8) weights with elevated liver 8-OHdG levels (by day 3).
Minata et al. (2010, 1937251) exposed wild-type (129S4/SvlmJ) and Ppara-mA\ (129S4/SvJae-
PparatmlGon7J) mice to PFOA (<50 |imol/kg/day) for four weeks. Levels of 8-OHdG were
evaluated in the liver. No increase in oxidative stress levels was noted in exposed wild-type
mice. In contrast, l'paru-xwiW mice demonstrated a dose-dependent increase in 8-OHdG levels,
with a significance increase at 50 |imol/kg/day when compared to controls. The correlation
between PFOA exposure and 8-OHdG was associated with increased tumor necrosis factor a
(TNF-a) mRNA levels.
In a developmental toxicity study, Li et al. (2019, 5387402) exposed pregnant Kunming mice to
PFOA (<10 mg/kg/day) on gestational day (GD) 1-17. Female mice were sacrificed on postnatal
day (PND) 21 and livers were assessed for oxidative damage by quantification of 8-OHdG,
catalase, and superoxide dismutase (SOD). Findings indicate the PFOA caused a dose-dependent
increase in oxidative DNA damage levels, which were significantly elevated after 2.5 mg/kg/day.
These results were associated with increased superoxide dismutase and catalase protein levels.
Together, these findings suggest that the livers of exposed mice were producing antioxidant
enzymes to counteract PFOA-induced elevated oxidative stress.
The testes are particularly susceptible to oxidative stress due to high energy demand and
abundance of polyunsaturated fatty acids. Liu et al. (2015, 3981571) exposed male Kunming
mice to <10 mg/kg/day of PFOA for 14 days and examined oxidative stress in the testis and
epididymis. A dose-dependent increase in lipid peroxidation and oxidative stress was observed
with a significant increase at >5 mg/kg/day relative to controls. In contrast to the results of Li et
al. (2019, 5387402), levels of the antioxidant enzymes SOD and carnitine acyltransferase (CAT),
and Nrf2 expression (an oxidative stress response gene) decreased as PFOA exposure doses
increased.
Several other studies measuring oxidative stress in the liver have found that PFOA induces
damage through hydrogen peroxide production {Salimi, 2019, 5381528} and through PPARa
activation pathways {Li, 2019, 5387402}. For additional information that PFOA induces
oxidative stress in the liver, please see the hepatic mechanistic section (Section 3.4.1.3; refer to
the interactive Tableau for additional supporting information and study details).
Evidence that PFOA induces oxidative stress in the immune system have been reported. Wang et
al. (2014, 3860153) observed that the spleens of mice treated with PFOA had mitochondrial
swelling and cavitation as well as swollen and ruptured cristae, which suggests impaired
oxidative processes. For additional information that PFOA induces oxidative stress in immune
cells, please see the immune mechanistic section (Section 3.4.2.3; refer to the interactive
Tableau for additional supporting information and study details).
Mechanistic studies noted PFOA exposure increased oxidative stress in the heart and brain. For
additional information, please see the developmental (Section 3.4.4.3) and cardiovascular
(Section 3.4.3.3) mechanistic sections (refer to the interactive Tableau for additional supporting
information and study details).
3-281
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.5.3.3.2 In Vitro Evidence
The ability of PFOA to induce oxidative stress has been assessed in vitro in several human, non-
human primate, and animal cell lines.
PFOA exposure caused a dose-dependent increase in 8-OHdG in human lymphoblast cells
(TK6), with significant results noted at >250 ppm (2 hours) {Yahia, 2014, 2851192}. A similar
relationship was noted in HepG2 cells with significant increase in 8-OHdG levels found at PFOA
concentrations >100 |iM (3 hours) {Yao, 2005, 5081563}. Yao and Zhong (2005, 5081563)
measured ROS using a 2',7'-dichlorodihydrofluorescein diacetate (DCFH-DA) assay and
observed a dose-dependent increase associated with elevated 8-OHdG levels. Peropadre et al.
(2018, 5080270) found 8-OHdG levels were non-significantly elevated in human HaCaT cells
following 24-hour exposure to PFOA (50 |iM). However, measurements taken 8 days following
exposure found levels to be significantly elevated by 50%.
Panaretakis et al. (2001, 5081525) observed the peak in ROS generation three hours following
PFOA exposure in HepG2 cells exposed to concentrations of 200 and 400 |iM. Both
concentrations significantly increased hydrogen peroxide and superoxide anions. Wiels0e et al.
(2014, 2533367) noted non-significant elevated levels of ROS after HepG2 cells were exposed to
PFOA (0.2-20 |iM) for 24 hours. Additionally, total antioxidant capacities were reduced after
exposure to 0.02-2,000 |iM. These studies contrast with the findings of Florentin et al. (2011,
2919235), which found no change in ROS using a DCFH-DA test in HepG2 cells exposed to 5-
400 |iM PFOA for 1 or 24 hours.
Kidney cells isolated from the African green monkey (Vero) were used in a DCFH-DA assay to
measure ROS production {Fernandez Freire, 2008, 2919390}. Authors reported a dose-
dependent increase in ROS production that reached significance at 500 |iM after 24 hours. Vero
cells also displayed fragmentation of mitochondrial reticulum at >50 |iM, a morphological
change consistent with defective metabolism, indicating that irregular metabolic activity may
play a role in ROS production in this model and exposure scenario.
ROS production was significantly higher in Paramecium caudatum exposed to PFOA (100 |iM)
for 12 or 24 hours, while 8-OHdG was not affected by PFOA {Kawamoto, 2010, 1274162}.
Addition of the antioxidant glutathione attenuated the PFOA-induced ROS production but not
DNA damage (as measured by a comet assay), indicating that the PFOA-induced DNA damage
was not associated with oxidative stress is P. caudatum.
Hocevar et al. (2020, 6833720) exposed mouse pancreatic acinar cells to PFOA (<100 |ig/mL; 6
or 24 hours) and observed an increase in intracellular calcium-induced activation of the unfolded
protein response (UPR) in the endoplasmic reticulum at concentrations >50 |ig/mL. This is a
well-established oxidative stress-inducing pathway.
Zhao et al. (2010, 847496) exposed human-hamster hybrid (Al) cells to PFOA (1-200 |iM; 1-16
days) and found significantly increased intracellular ROS, NO, and O2 ~ levels at all timepoints
exposed to >100 |iM. These increases correlated with cytotoxicity, which was significant at all
timepoints at 100 and 200 |iM. DNA mutagenicity was only significant at the highest
concentration at the longest exposure (16 days). Effects were reversed when previously PFOA-
exposed cells were treated with oxidative stress inhibitors dimethyl sulfoxide (DMSO) and NG-
methyl-L-arginine (L-NMMA). When repeating the study using a mitochondrial deficient cell
3-282
-------
DRAFT FOR PUBLIC COMMENT
March 2023
line (p°Al), authors reported no mutagenesis, indicating that if the increase in DNA mutation
after PFOA exposure is related to ROS generation, the association is mitochondria dependent.
3.5.3.4 Key Characteristic #6: Induces Chronic Inflammation
The induction of chronic inflammation includes increased white blood cells, altered chemokine
and/or cytokine production, and myeloperoxidase activity 1 Smith, 2016, 3160486}. Chronic
inflammation has been associated with several forms of cancer, and a role of chronic
inflammation in the development of cancer has been hypothesized. However, there are biological
links between inflammation and oxidative stress and genomic instability, such that the
contribution of each in carcinogenic progression is not always clear.
3.5.3.4.1 In Vivo Evidence
Increased inflammation and/or inflammatory markers (i.e., inflammatory cytokines) has been
reported in animal toxicological studies of acute, subchronic, and chronic exposures to PFOA.
NTP (2020, 7330145) used a matrix-type exposure paradigm. Pregnant Sprague-Dawley rats
were administered PFOA via gavage beginning on GD 6 and exposure was continued in
offspring postweaning for a total of 107 weeks. Dose groups for this report are referred to as
(perinatal exposure level (ppm))/(postweaning exposure level (ppm)) and ranged from 0/0-
300/300 ppm in males and 0/0-300/1,000 ppm in females. At the 16-week interim sacrifice,
incidences of chronic active inflammation of the glandular stomach submucosa was significantly
higher in the male 0/300 ppm group compared to the control group. No effects were seen in
female rats at the interim sacrifice. At the 2-year evaluation, females in the 0/1,000 and
300/1,000 ppm groups exhibited increased incidences of ulcer, epithelial hyperplasia, and
chronic active inflammation of the submucosa of the forestomach when compared to controls.
Histopathological analysis of animals exposed to PFOA (0.625-10 mg/kg) by oral gavage for 28
day exhibited nasal respiratory epithelium inflammation in both males and females, though these
effects did not follow a linear dose-response {NTP, 2019, 5400977}. Similarly, olfactory
epithelial inflammation and degeneration were observed in females. Increases in nasal and
olfactory hyperplasia were thought to be a result of the observed epithelial degradation and/or
inflammation.
Activation of the NF-kB signaling pathway plays an important role in the regulation of
inflammation, including through expression of proinflammatory cytokines {Lee, 2017, 3981419;
Shane, 2020, 6316911; Zhong, 2020, 6315790; Zhang, 2014, 2851150}. Modification to NF-kB
expression has been observed in adult zebrafish after 7, 14, and 21 days of PFOA exposure
{Zhang, 2014, 2851150; Zhong, 2020, 6315790} and in female BALB/c mice dermally exposed
to PFOA for 14 days {Shane, 2020, 6316911}. Additionally, proinflammatory cytokines IL-ip,
TNF-a, and others were upregulated by PFOA exposure at doses ranging from 0.002% w/w in
the diet and 2.5-10 mg/kg/day by gavage for 10 or 14 days in various tissues across several
mouse studies {Qazi, 2009, 1276154; Wang, 2014, 3860153; Liu, 2016, 3981762; Yang, 2014,
2850321}.
3.5.3.4.2 In Vitro Evidence
Saejia et al. (2019, 5387114) noted that PFOA (1 nM, 72 hours) significantly increased
activation of NF-kB in FTC133 cells. Furthermore, translocation of the phosphorylated version
of NF-kB to the nucleus from the cytosol, a crucial step in inflammation cytokine production,
3-283
-------
DRAFT FOR PUBLIC COMMENT
March 2023
was observed. Inhibition of NF-kB activation was found to reduce invasive characteristics of
cells, likely through reduced expression of MMP-2 and MMP-9. PFOA increased the levels of
proinflammatory cytokines, such as TNF-a, IL-ip, IL-6, and IL-8, in a dose-responsive manner
in IgE-stimulated rat mast cells (RBL-2H3 cell line) {Lee, 2017, 3981419}. It is important to
note that in vitro models may be used for the evaluation of changes in inflammatory markers and
response, they are generally not effective in modeling the events that are associated with chronic
inflammation.
Several studies have identified the potential of PFOA to increase inflammation within various
testing systems. For additional information, please see the immune (Section 3.4.2.3), hepatic
(Section 3.4.1.3), and cardiovascular (Section 3.4.3.3) mechanistic sections (refer to the
interactive Tableau for additional supporting information and study details).
3.5.3.5 Key Characteristic #7: Is Immunosuppressive
Immunosuppression refers to the reduction in the response of the immune system to antigen,
which is important in cases of tumor antigens 1 Smith, 2016, 3 1604861. It is important to note
that immunosuppressive agents do not directly transform cells, but rather can facilitate immune
surveillance escape of cells transformed through other mechanisms (e.g., genotoxicity).
Studies have identified the immunosuppressive potential of PFOA in in vivo and in vitro testing
systems. The pleotropic immunomodulatory effects of PFOA, including impaired vaccine
response in humans and reduction in B and T cell populations in the thymus and spleen in
laboratory animals, may reflect perturbed function of B and/or T cells. At the molecular level,
dysregulation of the NF-kB pathway may contribute to the immunosuppressive effects of PFOA.
The NF-kB pathway facilitates initial T cell responses by supporting proliferation and regulating
apoptosis, participates in the regulation of CD4+ T cell differentiation, and is involved in
mediating inflammatory responses. Dysregulation of the NF-kB pathway by PFOA, potentially
consequent to the induction of oxidative stress, may be a key component of the underlying
mechanism of PFOA-mediated immunosuppression. Reduced NF-kB activation and consequent
elevation of apoptosis is consistent with increased apoptosis in multiple cell types, the reduction
of pre/pro B cell numbers, and dysregulation of pro-inflammatory cytokines and mediators of
inflammation. For additional information, please see the immune mechanistic section (Section
3.4.2.3; refer to the interactive Tableau for additional supporting information and study details).
3.5.3.6 Key Characteristic #8: Modulates Receptor-Mediated Effects
Modulation of receptor-mediated effects involves the activation or inactivation of receptors (e.g.,
PPAR, AhR) or the modification of endogenous ligands (including hormones) 1 Smith, 2016,
3160486 J.
3.5.3.6.1 In Vivo Evidence
Yan et al. (2015, 2851199) exposed adult male Balb/c mice to PFOA (0.08-20 mg/kg/day) via
oral gavage for four weeks. Livers were isolated and mRNA levels of several peroxisome
proliferator-activated receptors (PPARs) were evaluated using RT-PCR. PPARa was found to be
increased by 50% in the 0.08 and 0.31 mg/kg/day dose groups. This trend was not consistent as
PPARa levels diminished at higher doses. PPARy was found to increase in a dose-dependent
3-284
-------
DRAFT FOR PUBLIC COMMENT
March 2023
manner that reached significance at 1.25 mg/kg/day PFOA. No differences were observed in
PPARp/S mRNA expression after exposure.
Data from studies conducted in rodent models have demonstrated PPARa activation as a
mechanism for PFOA-induced hepatotoxicity, due to the association between hepatic lesions
and/or increased liver weight and peroxisome proliferation downstream of PPARa activation.
There is also growing evidence the PFOA activates other nuclear receptors (e.g., CAR/PXR,
ERa, HNF4a) in tandem with PPARa to enact its effects. For additional information, please see
the hepatic (Section 3.4.1.3) and cardiovascular (Section 3.4.3.3) mechanistic sections (refer to
the interactive Tableau for additional supporting information and study details).
3.5.3.6.2 In Vitro Evidence
PPARa and PPARy gene expression was assessed in hepatocellular carcinoma cells (Hepa 1-6)
exposed to PFOA (50-200 |iM; 72 hours) {Yan, 2015, 2851199}. While no significant changes
were observed for these genes, PPARa target genes were significantly increased, indicating that
PPARa was activated by PFOA.
Available mechanistic evidence demonstrates that PFOA has the potential to dysregulate
hormone levels in hepatic cells, particularly in regard to thyroid function. Furthermore, rodent
and human hepatocytes treated with PFOA demonstrated a concentration-dependent decrease in
lipid accumulation that was associated with PPARa activation. For additional information, please
see the hepatic mechanistic section (Section 3.4.1.3; refer to the interactive Tableau for
additional supporting information and study details).
3.5.3.7 Key Characteristic #9: Causes Immortalization
Immortalization leads to tumorigenesis when cells continue to divide after sustaining DNA
damage and/or shortened telomeres, events that cause cells to cease to divide in healthy or
normal states (i.e., the Hayflick limit). Immortalization is a key characteristic typically observed
in and associated with human DNA and RNA viruses, such as human papillomaviruses and
hepatitis C virus, among others. In vitro cell transformation assays have been historically used to
test carcinogenic potential of both genotoxic and non-genotoxic compounds {Creton, 2012,
8803671}, and is recognized as an assay related to key characteristic #9 {Smith et al., 2020,
6956443}.
In the case of PFOA, two studies reported no change in cell transformation in vitro in cells
exposed to PFOA relative to controls. Jacquet et al. (2012, 2124683) exposed SHE cells to
PFOA at concentrations ranging from 3.7x 10~4 - 37.2 |iM for 6 days with or without pre-
treatment with the tumor initiator benzo-a-pyrene (BaP). PFOA exposure alone did not induce
cell transformation, but PFOA did significantly induce transformation in BaP-sensitized cells,
indicating that PFOA does not alone initiate cell transformation, but may have tumor promoter-
like activity. A second in vitro cell transformation assay reported no evidence of transformation
in C3H 10T-1/2 mouse embryo cells exposed to 0.1-200 |ig/mL PFOA in a 14-day colony assay
for transformation nor in a 38-day foci transformation assay {Garry and Nelson, 1981,
10228130}.
3-285
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.5.3.8 Key Characteristic #10: Alters Cell Proliferation, Cell Death,
or Nutrient Supply
Aberrant cellular proliferation, cell death, and/or nutrient supply is a common mechanism among
carcinogens. This mechanism includes aberrant proliferation, decreased apoptosis or other
evasion of terminal programming, changes in growth factors, angiogenesis, and modulation of
energetics and signaling pathways related to cellular replication or cell cycle control {Smith,
2016, 3160486}.
3.5.3.8.1 In Vivo Evidence
To determine if PFOA exposure induced proliferation in cancer cells, Ma et al. (2016, 3981426)
xenografted human endometrial adenocarcinoma (Ishikawa cell line) cells into the flanks of six-
week-old female BALB/c mice. Animals were then treated with PFOA (20 mg/kg/day) by oral
gavage daily for three weeks beginning the same day of the xenograft. Tumor volume was
measured after five weeks, and data indicated that PFOA caused tumors to nearly triple in size.
Additionally, levels of proliferating cell nuclear antigen (PCNA) and vimentin protein were both
upregulated by PFOA, suggesting increased cell proliferation and invasion. E-cadherin
expression was downregulated after PFOA exposure, indicating that cells were more likely to
migrate and form metastases.
Treatment effects on apoptosis and cell cycle have also been observed in immune system cells of
animals exposed to PFOA. Wang et al. (2014, 3860153) exposed BALB/c mice to PFOA (5-20
mg/kg/day, 14 days) via gavage and reported that the percent of apoptotic cells increased in the
spleen (10-20 mg/kg/day) and in the thymus (20 mg/kg/day). Yang et al. (2002, 1332453)
reported significant reductions in the proportion of thymocytes in the S and G2/M phases and
significant increases in the G0/G1 phases of mice treated with PFOA, effects that were PPARa-
dependent.
Additional mechanistic studies, detailed elsewhere, noted PFOA exposure alters the number of
various B and T cell subsets in primary and secondary lymphoid organs, which may impact
immune system development, including dysregulation of proliferation, differentiation, and/or
apoptosis. For additional information, please see the immune mechanistic section (Section
3.4.2.3; refer to the interactive Tableau for additional supporting information and study details).
3.5.3.8.2 In Vitro Evidence
PFOA has been demonstrated to increase cell proliferation and apoptosis evasion in vitro.
Evidence presented here is organized into three categories: induced proliferation, apoptosis
evasion, and modification of cellular migration.
3.5.3.8.2.1 Proliferation
Exacerbation of proliferation in cancer cell lines is of particular concern to the development and
prognosis of cancer. Several studies have utilized MTT assays to measure cellular metabolic
activity to determine cell proliferation and cytotoxicity rates.
PFOA exposure (5-50 |iM) increased cellular proliferation in MCF-7 human breast cancer cells
and HepG2 human hepatoma (nontumorigenic) cells {Burhke, 2015, 2850235; Burhke, 2013,
2325346; Liu, 2020, 6512127}. However, predictably, proliferation rates decreased at cytotoxic
3-286
-------
DRAFT FOR PUBLIC COMMENT
March 2023
concentrations (>100 |iM PFOA) {Burhke, 2015, 2850235; Burhke, 2013, 2325346; Wen, 2020,
6302274}. Similar results were observed in the breast epithelial (nontumorigenic) cell line MCF-
10A, in which PFOA exposure (50 and 100 |iM; 24-72 hours) increased cell proliferation,
whereas proliferation rates decreased as the PFOA concentration was increased to a cytotoxic
level (250 |iM) {Pierozan, 2018, 4241050}. A subsequent study by Pierozan et al. (2020,
6833637) reported that PFOA-induced (100 |iM, 72 hours) proliferation persisted in MCF-10A
daughter subcultures that were not exposed to PFOA. PFOA exposure (1-100 nM) in colorectal
cancer cells (DLD-1) has also been shown to modify the cell cycle by causing more cells to enter
S-phase and less in Gi of mitosis {Miao, 2015, 3981523}.
Several studies of the effects of low exposure to PFOA found no evidence of modification to cell
proliferation rates. These studies include ovarian cancer cell line A2780 (1-200 nM, 48 hours)
{Li, 2018, 5079796} Ishikawa human endometrial adenocarcinoma cells (50 nM, 48 hours) {Ma,
2016, 3981426}, and human colorectal cancer cell line DLD-1 (1-10,000 nM, 72 hours) {Miao,
2015, 3981523}.
Insulin growth factor 1 (IGF-1) expression has been implicated in governing proliferation in
cancer cells. A series of experiments performed by Gogola et al. (2019, 5016947; 2020,
6316203; 2020, 6316206) used COV434 and KGN cells exposed to PFOA (0.02 ng/mL-2
|ig/mL; 72 hours). All studies found increased proliferation in both cell lines. Proliferation was
highest in COV434 and KGN cells at 0.02 ng/mL and 2 ng/mL, respectively. Interestingly,
proliferation returned to baseline levels in both cell lines at PFOA concentration of 2 |ig/mL,
indicating a bell-shaped dose response. These experiments were repeated after inhibition of IGF-
1 caused normalization in both cell lines after PFOA exposure. Together, these studies
demonstrate the potential pathway in which PFOA induces proliferation in cancer cells.
HepG2 cells were exposed to non-cytotoxic concentrations of PFOA for 24 hours before SHP-2, a tumor
suppressor protein, was immunoprecipitated from the cell lysates {Yang, 2017, 3981427}. PFOA (100
l_iM) slightly lowered SHP-2 mRNA expression and decreased SHP-2 enzyme activity in a concentration-
dependent manner. SHP-2 protein levels were increased only at 140 (.iM exposure, and unchanged at other
concentrations. These results indicate that PFOA inhibits SHP-2 by reducing enzyme activity, not protein
content.
Rainieri et al. (2017, 3860104) evaluated the effects of PFOA on cell proliferation by quantifying the
distribution of cells in different stages of the cell cycle in a human macrophage cell line (TLT cells).
Significantly more cells were in G2M phase following exposure to PFOA (50-500 mg/L; 12 hours) in
parallel with a lower proportion of cells in the G0/G1 phase, suggesting increased cell proliferation. For
additional evidence of the effect of PFOA on cell death and cell proliferation in the immune system,
please see the immune mechanistic section (Section 3.4.2.3; refer to the interactive Tableau for
additional supporting information and study details).
3.5.3.8.2.2 Apoptosis
Evasion of programmed cell death is a characteristic of cancer cells, allowing them to continue
proliferating, which can be enhanced by PFOA exposure. Dairkee et al. (2018, 4563919)
evaluated several human breast cancer cell lines for apoptosis following PFOA exposure (1 or
100 nM; 7 days). Using fluorescence activated cell sorting (FACS) of Annexin V-FITC, PFOA
concentrations were found to be inversely correlated with apoptosis rates. However, in HepG2
cells, PFOA exposure was found to increase metabolically-induced BAX apoptosis in a dose-
3-287
-------
DRAFT FOR PUBLIC COMMENT
March 2023
dependent manner {Wen, 2020, 6302274}. Apoptosis was also found to increase in HepG2 cells
after PFOA exposure (200 or 400 |iM; <24 hours) and was associated with an increase in
caspase-9 activation after 5 hours of exposure {Panaretakis, 2001, 5081525}. Additionally, the
murine spermatogonial cell line GC-1 exhibited a dose-dependent increase in apoptosis after
exposure to PFOA (>250 |iM) for 24 hours that reached significance at >500 |iM {Lin, 2020,
6833675}.
Caspase protease enzymes are essential in apoptotic cell death and are frequently used to assess
apoptosis. Gogola at al. (2020, 6316203; 2020, 6316206) found that PFOA (0.2-20 ng/mL; 72
hours) caused no changes to caspase 3/7 expression in COV434 and KGN cells. Additionally,
PFOA (<100 |iM) had no effect on caspase 3/7 activity in HepG2 cells. Lin et al. (2020,
6833675) reported a dose-dependent increase in caspase-3 activity that correlated with apoptosis
rates in GC-1 cells. Additionally, apoptosis and caspase activity were inversely correlated with
Bcl-2/Bax ratios. These results indicate that PFOA may induce apoptosis through an increase in
BAX expression. Hu and Hu (2009, 2919334) also suggested that PFOA could induce apoptosis
by overwhelming the homeostasis of antioxidative systems, increasing ROS, impacting
mitochondria, and changing expression of apoptosis gene regulators, based on their findings in
studies with HepG2 cells. Overall, data are conflicting on the ability of PFOA to induce or
inhibit apoptosis, with the variation likely dependent upon dose and duration of exposure.
3.5.3.8.2.3 Modulation of Migration
Cancer cells are invasive in nature due to their ability to increase mobility, reduce attachment to
neighboring cells, and express proteins that break down the extracellular matrix of tissues.
Wound healing assays are a common and reproducible way to inflict a 'wound' on a monolayer
plate of cells and measure the time for the cells to re-establish confluency. Two independent
studies concluded PFOA exposure increased the rate at which Ishikawa cells (50 nM, 48 hours)
{Ma, 2016, 3981426} and A2780 cells (>100 nM, 72 hours) {Li, 2018, 5079796} were able to
re-establish confluency in a dose-dependent manner.
Assays of migration and invasion measure the ability of a cell to travel either without inhibition
or through the extracellular matrix of plated cells, respectively. Two studies investigated cellular
migration after PFOA exposure and found no change after FTC133 cells were exposed to 1 nM
(72 hours) {Saejia 2019, 5387114} or 0-1 mM (24-72 hours) {Pierozan, 2018, 4241050}, while
an increase in migration was found at 100 nM (72 hours) in MCF-10A cells {Pierozan, 2018,
4241050}. All studies reviewed found an increase in the invasive nature of cancer cells lines
FTC133 (1 nM, 72hours) {Saejia, 2019, 5387114}, Ishikawa (>50 nM) {Ma, 2016, 3981426},
MCF-10A (100 nM, 72 hours) {Pierozan, 2018, 4241050}, A2780 (>100 nM, 72 hours) {Li,
2018, 5079796}, andDLD-1 (1 nM-1 |iM, 72 hours) {Miao, 2015, 3981523} after PFOA
exposure.
Pierozan et al. (2020, 6833637) exposed MCF-10A cells to PFOA (100 |iM, 72 hours) and found
that invasion and migration of daughter cell passages was elevated when compared to control.
Several reports noted cell invasion and upregulated MMP2 and MMP9 expression levels, which
help to break down the extracellular matrix allowing cells to move freely, indicating that cancer
cells could be more likely to become invasive or metastasize after exposure to PFOA {Li, 2018,
5079796; Miao, 2015, 3981523; Saejia, 2019, 5387114}.
3-288
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Additional mechanistic studies have identified the potential of PFOA to induce aberrant cellular
proliferation rates and increase apoptosis within in vitro testing systems. For additional
information, please see the immune (Section 3.4.2.3) and hepatic (Section 3.4.1.3) mechanistic
sections (refer to the interactive Tableau for additional supporting information and study details).
3.5.4 Weight Of Evidence for Carcinogenicity
3.5.4.1 Summary of Evidence
The carcinogenicity of PFOA has been documented in both epidemiological and animal
toxicological studies. The evidence in epidemiological studies is primarily based on the
incidence of kidney and testicular cancer, as well as potential incidence of breast cancer in
genetically susceptible subpopulations. Other cancer types have been observed in humans,
although the evidence for these is generally limited to low confidence studies. The evidence of
carcinogenicity in animal models is provided in three chronic oral animal bioassays in Sprague-
Dawley rats which identified neoplastic lesions of the liver, pancreas, and testes.
3.5.4.1.1 Evidence from Epidemiological Studies
As discussed in depth in the 2016 HESD {U.S. EPA, 2016, 3603279}, two medium confidence
studies involving members of the C8 Health Project showed a positive association between
PFOA levels (mean at enrollment 0.024 |ig/mL) and kidney and testicular cancers {Barry, 2013,
2850946; Vieira, 2013, 2919154}. Vieira et al. (2013, 2919154) investigated the relationship
between PFOA exposure and cancer among the residents living near the DuPont plant in
Parkersburg, West Virginia. The adjusted ORs were increased for testicular cancer and for
kidney cancer (OR: 5.1, 95% CI: 1.6, 15.6; n = 8 and OR: 1.7, 95% CI: 0.4, 3.3; n = 10,
respectively) in the Little Hocking water district of Ohio and for kidney cancer (OR: 2.0, 95%
CI: 1.3, 3.1; n = 23) in the Tuppers Plains water district of Ohio. Barry et al. (2013, 2850946)
extended this work and found significantly increased testicular cancer risk with an increase in the
estimated cumulative PFOA serum level (HR: 1.34, 95% CI: 1.00, 1.79; n = 17). Increases,
though nonsignificant, were also observed for kidney cancer (HR: 1.10, 95% CI: 0.98, 1.24;
n = 105) and thyroid cancer (HR: 1.10, 95% CI: 0.95, 1.26; n = 86). As part of the C8 Health
Project, the C8 Science Panel (2012, 9642155) concluded that a probable link existed between
PFOA exposure and testicular and kidney cancer {Steenland, 2020, 7161469}.
Since publication of the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}, eight medium
confidence epidemiological studies examining the carcinogenicity of PFOA have been
published. Six of those studies focused specifically on breast cancer risk. One study adds support
to the previous evidence of an association between PFOA and kidney cancer {Shearer, 2021,
7161466}. No new epidemiological studies on testicular cancer were identified. Shearer et al.
(2021, 7161466) is a multi-center case-control study nested within the NCI's PLCO. This
randomized clinical trial included all the participants of the screening arm of the PLCO trial who
were newly diagnosed with histopathologically confirmed RCC during the follow-up period
(n = 326). The authors reported a statistically significant increase in risk of RCC with pre-
diagnostic serum levels of PFOA (OR = 2.63; 95% CI: 1.33, 5.20 for the highest vs. lowest
quartiles; p-trend = 0.007, or per doubling of PFOA: OR: 1.71; 95% CI: 1.23, 2.37). After
adjusting for other PFAS the association remained significant in analyses on a per doubling
increase in PFOA. The increase in the highest exposure quartile remained and the magnitude was
similar (i.e., OR = 2.63 without adjusting for other PFAS vs. 2.19 after adjusting for other
3-289
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFAS), but it was no longer statistically significant. The analyses accounted for numerous
confounders including BMI, smoking, history of hypertension, eGFR, previous freeze-thaw
cycle, calendar and study year of blood draw, sex, race and ethnicity, study center. Statistically
significant increased odds of RCC were observed in participants ages 55-59 years, and in men
and in women, separately.
Notably, a recent critical review and meta-analysis of the epidemiological literature concluded
that there was an increased risk for kidney (16%) and testicular (3%) tumors for every 10 ng/mL
increase in serum PFOA {Bartell, 2021, 7643457}. Although the authors concluded that the
associations were likely causal, they noted that there were a limited number of studies and
additional studies with larger cohorts could strengthen the conclusion.
The majority of studies examining associations between PFOA and cancer outcomes were on
breast cancer, with five medium confidence epidemiological studies examining the
carcinogenicity of PFOA published since the 2016 assessment. Two nested case-control studies
did not report an association between breast cancer and PFOA concentrations measured in
maternal serum throughout pregnancy and 1-3 days after delivery {Cohn, 2020, 5412451} or in
in serum after case diagnosis and breast cancer {Hurley, 2018, 5080646}. Both studies were
conducted in California and most breast cancer cases were obtained from the cancer registry.
Two nested case-control studies found associations between PFOA and breast cancer, but only in
specific genotype or estrogen receptive groups of participants {Ghisari, 2017, 3860243; Mancini,
2020, 5381529}. Ghisari et al. (2017, 3860243) reported an increased risk for breast cancer
identified from the Danish cancer registry with increasing PFOA concentrations only in subjects
with a CC genotype (n = 36 cases and 47 controls) in the CYP19 gene (cytochrome P450
aromatase). A nested case-control study (194 pairs of breast cancer cases and controls) within the
French E3N cohort found an 86% higher risk of breast cancer in the 2nd quartile of PFOA
compared to the 1st quartile in a partially adjusted model {Mancini, 2020, 5381529}. Mancini et
al. (2020, 5381529) also reported that the risk for breast cancer (93% verified as pathologically
confirmed from medical records after self-reported cancer diagnosis) varied by type of cancer
with a statistically significant increase in ER- and progesterone receptor negative (PR-) breast
cancers in the second quartile of PFOA only. The sample size was small with 26 participants
having ER- breast cancers and 57 having PR- breast cancers. There was no association between
PFOA and receptor-positive breast cancer risk. In Taiwan, Tsai et al. (2020, 6833693) observed a
statistically significant increase in risk of breast cancer only in women younger than 50 years,
and in ER+ participants aged 50 years or younger, along with a decrease in risk for ER- breast
cancers in participants aged 50 years or younger. Significantly increased breast cancer risk was
also observed in an NHANES population across serum PFOA quartiles with a significant dose-
response trend {Omoike, 2021, 7021502}. A recent study in a Japanese population observed
inverse association across serum PFOA quartiles with a significant dose-response trend {Itoh,
2021, 9959632}. The association remained significantly inverse in postmenopausal women in the
highest tertile of exposure, with a significant dose-response trend. Overall, study design issues,
lack of replication of the results, and a lack of mechanistic understanding of PFOA on specific
breast cancer subtypes or in subpopulations limit firm conclusions regarding PFOA and breast
cancer.
These findings are supported by other recent assessments and reviews {IARC, 2016, 3982387;
AT SDR, 2021, 9642134; Steenland, 2021, 7491705; CalEPA, 2021, 9416932}. In their review
3-290
-------
DRAFT FOR PUBLIC COMMENT
March 2023
of the available epidemiological data, the International Agency for Research on Cancer (IARC)
(2016, 3982387) concluded that the evidence for testicular cancer was "considered credible and
unlikely to be explained by bias and confounding, however, the estimate was based on small
numbers." Similarly, IARC (2016, 3982387) concluded that the evidence for kidney cancer was
also credible but noted that chance, bias, and confounding could not be ruled out with reasonable
confidence. They considered that there was limited evidence in humans for the carcinogenicity of
PFOA.
3.5.4.1.2 Evidence from Animal Bioassays
In addition to the available epidemiological data, two multi-dose bioassays and one single-dose
chronic cancer bioassay are available that investigate the relationship between dietary PFOA
exposure and carcinogenicity in male and female rats {Butenhoff, 2012, 2919192; NTP, 2020,
7330145; Biegel, 2001, 673581}. Increased incidences of neoplastic lesions were primarily
observed in male rats, though results in females are supportive of potential carcinogenicity of
PFOA. Testicular LCTs were identified in both the Butenhoff et al. (2012, 2919192) and Biegel
et al. (2001, 673581) studies. LCT incidence at similar dose levels was comparable between the
two studies (11% and 14%). PACTs were observed in both the NTP (2020, 7330145) and Biegel
et al. (2001, 673581) studies. NTP (2020, 7330145) reported increased incidences of pancreatic
acinar cell adenomas in males in all treatment groups compared to their respective controls
(Table 3-12). This rare tumor type was also observed in female rats from the highest dose group,
though these increases did not reach statistical significance. Biegel et al. (2001, 673581)
similarly reported increases in the incidence of PACTs in male rats treated with PFOA, with zero
incidences observed in control animals. In addition, NTP (2020, 7330145) reported dose-
dependent increases in the incidence of liver adenomas in male rats (Table 3-11), which were
also observed by Biegel et al. (2001, 673581). Overall, NTP concluded that under the exposure
conditions of their report, there was clear evidence of carcinogenic activity of PFOA in male
Sprague Dawley rats and some evidence of carcinogenic activity of PFOA in female Sprague
Dawley rats based on the observed tumor types {NTP, 2020, 7330145}.
The report from NTP (2020, 7330145) provides evidence that chronic exposure accompanied by
perinatal exposure to PFOA does not increase cancer risk when compared to chronic exposure
scenarios alone. There were no differences in the incidences of all tumor types examined across
the treatment groups administered PFOA during both perinatal and postweaning periods
compared with the postweaning-only treatment groups (see further study design details in
Section 3.4.4.2.1.2). Age-dependent sensitivity to the carcinogenic effects of PFOA was
previously only addressed in the study by Filgo et al. (2015, 2851085) in mice which is limited
in terms of its gestational-only study design and small sample sizes.
3.5.4.2 Mode of Action Analysis
As PFOA has been associated with multi-site tumorigenesis in both epidemiological studies and
animal toxicological studies, not always with site concordance, it is reasonable to assume that it
may act through multiple carcinogenic MOAs. In the 2016 HESD {U.S. EPA, 2016, 3603279},
EPA suggested that the induction of tumors was likely due to interactions with nuclear receptors,
perturbations in the endocrine system, interruption of intercellular communication, mitochondrial
effects, and/or perturbations in the DNA replication and cell division processes. Since that time,
several health agencies have reviewed the available mechanistic literature. For example,
3-291
-------
DRAFT FOR PUBLIC COMMENT
March 2023
CalEPA's Office of Environmental Health Hazard Assessment recently concluded that PFOA
"possesses several of the key characteristics of carcinogens, including the ability to induce
oxidative stress, inflammation, and modulate receptor-mediated effects. Additionally, there is
suggestive evidence that PFOA... [is] genotoxic, thus a genotoxic MOA for cancer remains
plausible" {CalEPA, 2021, 9416932}. IARC (2016, 3982387) also concluded that there is
moderate evidence for many potential mechanisms for PFOA-induced toxicity.
As described in the following subsections, the available mechanistic data continue to suggest that
multiple MO As could play role in the renal, testicular, pancreatic, and hepatic tumorigenesis
associated with PFOA exposure in human populations as well as animal models.
3.5.4.2.1 Mode of Action for Renal Tumors
As discussed in Section 3.5.10, there is convincing evidence for an association between renal
carcinogenesis and serum PFOA concentrations in epidemiological studies from both the general
population and residents of high-exposure communities {Barry, 2013, 2850946; Shearer, 2021,
7161466}. However, there is limited mechanistic information from epidemiological studies
explaining the observed renal carcinogenicity. Additionally, many animal models are limited in
their ability to replicate kidney damage due to PFOA exposure that is observed in humans {Li,
2017, 3981403}. One factor that may be driving this inconsistency between humans and animals
is the difference in renal clearance rates between human and animal models. Regardless of
elimination differences, both animal toxicological studies and the limited available human
biomonitoring data suggest that the kidneys may be a site of enrichment upon PFOA exposure
and subsequent distribution {Shearer, 2021, 7161466}.
The few available studies focusing on PFOA-induced renal toxicity highlight several potential
MO As by which PFOA exposure may result in renal tumorigenesis, including altered cell
proliferation and apoptosis, epigenetic alterations, and oxidative stress. However, due to data
limitations, it is difficult to distinguish what mechanism(s) are the most relevant for PFOA-
induced kidney cancer. The renal-specific evidence supporting multiple MO As is described in
the subsections below, though as described in Section 3.5.3, PFOA generally exhibits evidence
of multiple key characteristics of carcinogens which may also be relevant to the MOA for
PFOA-induced renal cell carcinoma.
3.5.4.2.1.1 Altered Cell Death, Cell Proliferation, or Nutrient Supply
There is evidence that relative kidney weight, particularly in male rats, is increased after PFOA
treatment (see PFOA Appendix) {NTP, 2019, 5400977; Butenhoff, 2004, 1291063; NTP, 2020,
7330145}. However, these increases in kidney weight and presumably increases in cell
proliferation may be due to increased need for renal transporters and not necessarily an indicator
of the initial stages of carcinogenesis {U.S. EPA, 2016, 3603278}. Though there is conflicting
evidence of alterations in relative kidney weight in female rats, NTP (2020, 7330145) reported
increased hyperplasia of urothelium that lines the renal papilla in female rats from the 0/1,000
and 300/1,000 ppm (63.4 and 63.5 mg/kg/day, respectively) dose groups at the interim sacrifice
timepoint (16 weeks) and in female rats from the 0/300 (18.2 mg/kg/day), 0/1,000, and
300/1,000 ppm dose groups at the terminal sacrifice (107 weeks). These changes were
accompanied by increased incidence of renal papilla necrosis at terminal sacrifice in both
1,000 ppm postweaning groups. Though NTP (2020, 7330145) did not explore the mechanisms
of toxicity underlying the observed renal effects, they note that prolonged exposure and
3-292
-------
DRAFT FOR PUBLIC COMMENT
March 2023
relatively high dose levels along with the enhanced efficiency of excretion and increased urinary
concentrations of PFOA in female rats (compared to males) may have resulted in cytotoxicity
and hyperplasia of the papilla.
Evidence of cytotoxicity and cell cycle disruption was also provided by a single in vitro study in
Vero cells (cell line derived from monkey kidney epithelial cells) {Fernandez Freire, 2008,
2919390}. Fernandez Freire et al. (2008, 2919390) assessed potential cytotoxic effects and
alterations in cell cycle progression in Vero cells treated with PFOA at concentrations of 50-
500 |iM for 24 hours. Cells treated with PFOA exhibited decreases in viability and proliferation,
as indicated by alterations in mitochondrial metabolism (MTT assay) and the total number of
cells (Bradford/TPC assay), though both assays exhibited a plateau in cytotoxicity at PFOA
concentrations of approximately 200 |iM and higher. The study also reported dose-dependent
increases in the percentage of apoptotic cells with increasing PFOA concentrations. Flow
cytometric analysis demonstrated G0/G1 cell cycle arrest in Vero cells treated with the
maximum concentration of 500 |iM PFOA. The percentage of cells in the G0-G1 stage was
increased whereas the percentages of cells in the S and G2-M stages were decreased. The authors
hypothesized that the observed cell cycle arrest may be linked to increased ROS and oxidative
stress, further described below.
3.5.4.2.1.2 Oxidative Stress
The increases in cytotoxicity and apoptosis in Vero cells treated with up to 500 |iM PFOA for
24 hours observed by Fernandez Freire et al. (2008, 2919390) were accompanied by a dose-
dependent increase in ROS which was statistically significant in the cells treated with 500 |iM.
The authors noted that severe oxidative stress could induce cell cycle arrest and apoptosis, as
described previously {Fernandez Freire, 2008, 2919390}. However, in the only available animal
toxicological study assessing oxidative damage in the kidney, levels of 8-
hydroxydeoxyguanosine (8-OH-dG) DNA damage in the kidney were unchanged in male
Fischer 344 rats administered PFOA via the diet (0.02% for 2 weeks) or by IP injection
(100 mg/kg single injection) {Takagi, 1991, 2325496}. Though the renal-specific evidence of
PFOA-induced oxidative stress is limited, further discussion on oxidative stress in other organ
systems is discussed below, as well as in Section 3.5.3.
3.5.4.2.1.3 Epigenetics
Rashid et al. (2020, 6315778) investigated epigenetic markers that could contribute to the kidney
dysfunction associated with PFOA exposure. CD-I mice were orally exposed to 1-20 mg/kg/day
PFOA for 10 days and kidney tissues were evaluated for epigenetic alterations (DNA
methylation and histone acetylation). Though no PFOA-induced changes in global methylation
were noted (by measurements of 5-methyl cytosine and 5-hydroxy methylation levels), the study
reported specific methylation changes with reduced representation bisulfite sequencing (RRBS).
Overall, 879 genes were differentially methylated in in the 20 mg/kg/day dose group vs. control.
PFOA exposure also altered mRNA expression of several proteins that regulate DNA
methylation, including DNA methyl transferases and ten eleven translocation enzymes, as well
as mRNA expression of several histone deacetylases. Combined, these results suggest that PFOA
exposure triggered epigenetic alterations, including DNA methylation changes and potentially
histone modifications, in the kidney {Rashid, 2020, 6315778}. However, further study is needed
to explore connections between the observed epigenetic changes and subsequent regulation of
genes associated with kidney tumorigenesis.
3-293
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.5.4.2.2 Mode of Action for Testicular Tumors
There is both epidemiological evidence and evidence from animal bioassays of an association
between increased PFOA serum concentrations or doses and testicular carcinogenesis. Testicular
cancer was observed in epidemiological studies from the C8 Health Project {Barry, 2013,
2850946; Vieira, 2013, 2919154}. In addition, a recent meta-analysis concluded that there is a
3% increase in risk for testicular cancer with every 10 ng/mL increase in serum PFOA
concentrations {Bartell, 2021, 7643457}. In animal models, testicular tumors (Leydig cell
tumors (LCTs)) were reported in two chronic studies in male Sprague-Dawley rats {Butenhoff,
2012, 2919192; Biegel, 2001, 673581}. Combined, these results indicate that the testes are a
common site of PFOA-induced tumorigenesis.
The available literature highlights several potential MO As by which PFOA exposure may result
in increased incidence of LCTs in animals, though it is unclear whether these MO As are relevant
to testicular cancers associated with PFOA exposure in humans. In a review of LCTs published
by Clegg et al. (1997, 224277), a workgroup identified seven non-genotoxic hormonal MOAs
(i.e., androgen receptor antagonism; testosterone biosynthesis inhibition; 5a-reductase inhibition;
aromatase inhibition; estrogen agonism; GnRH agonism; and dopamine agonism), the majority
of which involved downstream increases in luteinizing hormone (LH) levels and subsequent
Leydig cell hyperplasia/tumorigenesis. It has been proposed that PPARa agonism potentially
mediates these effects, though the evidence supporting this claim is not as strong as for other
tumor types (i.e., hepatic tumors) {Klaunig, 2003, 5772415; Klaunig, 2012, 1289837}.
The testes-specific evidence for various MOAs are described in the subsections below, though,
as described in Section 3.5.3, PFOA generally exhibits evidence of multiple key characteristics
of carcinogens which may also be relevant to the MOA for testicular cancers associated with
increased serum PFOA concentrations in humans.
3.5.4.2.2.1.1 Hormone-mediated MOAs
Clegg et al. (1997, 224277) identified five human-relevant MOAs for LCTs that involve
alterations in hormone balances, steroid receptor activity, or enzymes involved in steroid
metabolism (androgen receptor antagonism; testosterone biosynthesis inhibition; 5a-reductase
inhibition; aromatase inhibition; estrogen agonism). In addition, some compounds have been
shown to influence Leydig cell function, including steroidogenesis, via hormone-mediated
MOAs that are initiated upon PPARa activation {Gazouli, 2002, 674161; Klaunig, 2003,
5772415}. Klaunig et al. (2003, 5772415) described two proposed hormone-mediated MOAs
and key events by which PPARa agonists could induce LCTs in rats: one MOA which is
secondary to liver PPARa induction and one MOA which involves direct inhibition of
testosterone biosynthesis in the testes. These two MOAs involve associative key events such as
increased aromatase activity, increased serum estradiol (E2) levels, increased TGFa levels,
decreased testosterone levels, increased LH levels, and/or Leydig cell proliferation.
There is no evidence in the identified literature of PFOA treatment resulting in 5a-reductase
inhibition. Similarly, the majority of the limited available in vitro studies report that PFOA does
not act as an androgen receptor antagonist {McComb, 2019, 6304412; Kang, 2016, 3749062;
Du, 2013, 2850983; Rosenmei, 2013, 2919164}. In vivo studies in male rats and mice generally
found no effect of oral PFOA exposure on testicular aromatase activity or mRNA expression,
though there was some evidence for increased hepatic microsomal aromatase activity or mRNA
3-294
-------
DRAFT FOR PUBLIC COMMENT
March 2023
expression {Li, 2011, 1294081; Biegel, 1995, 1307447; Liu, 1996, 1307751}. This hepatic
aromatase activity provides some support for the MOA that is secondary to liver PPARa
induction {Klaunig, 2003, 5772415}.
Although increased aromatase activity was observed indicating potential increases in the
conversion of androgens to estrogens, further evidence of estrogen agonism in rodents was not
robust. Biegel et al. (2001, 673581) reported consistent increases in serum E2 in male rats treated
with the same concentration of PFOA that induced LCTs (300 ppm; approximately
13.6 mg/kg/day); however, the estrogen levels were too low to be accurately measured with the
radioimmunoassay methods utilized in the study. Cook et al. (1992, 1306123) observed similar
increases in serum E2 concentrations in male rats gavaged with 10, 25, or 50 mg/kg/day PFOA
for 14 days, though the authors also used a radioimmunoassay and reported similarly low E2
concentrations. Perkins et al. (2004, 1291118) additionally reported suggestive increases in
serum E2 concentrations in male rats treated with up to 6.5 mg/kg/day PFOA for 13 weeks,
though this response was not statistically significant. Overall, there is not sufficient evidence to
support estrogen agonism as the MOA for PFOA-induced LCTs.
Several of the available studies support an impact of PFOA on testosterone production in male
rodents {Biegel, 1995, 1307447; Cook, 1992, 1306123; Zhang, 2014, 2850230; Song, 2019,
5079725; Li, 2011, 1294081; Eggert, 2019, 5381535; Lu, 2019, 5381625; Martin, 2007,
758419}, as well as in men from the general population or high-exposure communities from
epidemiological studies {Cui, 2020, 6833614; Petersen, 2018, 5080277; Lopez-Espinosa, 2016,
3859832}. However, neither the subchronic nor the chronic study in male rats that measured
serum testosterone reported decreases across multiple time points ranging from 1-21 months
{Perkins, 2004, 1291118; Biegel, 2001, 673581}. Though there is evidence of PFOA-induced
inhibition of testosterone biosynthesis, this lack of response in the only study that both observed
LCTs and measured testosterone serum levels limits potential conclusions about whether
decreased testosterone plays a role in the MOA for LCTs {Biegel, 2001, 673581}.
Support for at least partial PPARa mediation of testosterone production inhibition due to PFOA
administration is available from one study in mice {Li, 2011, 1295081}. Significantly reduced
plasma testosterone concentrations were observed in male wildtype PPARa mice and humanized
PPARa transgenic mice. These decreases were visible but no longer statistically significant in
PPARa-null mice. In addition, reduced reproductive organ weights and increased sperm
abnormalities were also observed in PFOA-treated male PPARa wild-type and humanized
PPARa mice but not in PPARa-null mice {Li, 2011, 1295081}. However, data are not currently
sufficient to demonstrate that the other key steps in the postulated PPARa-mediated MO As are
present in PFOA-treated animals following exposures that lead to tumor formation. Studies are
needed to demonstrate the increase of GnRH and LH in concert with the changes in aromatase
and further study is needed to confirm the potential increases in serum E2. There was also no
indication of increased Ley dig cell proliferation at the doses that caused adenomas in the Biegel
et al. (2001, 673581) study. Thus, additional research is needed to determine if the hormone
testosterone-E2 imbalance is a key factor in development of LCTs as a result of PFOA exposure.
3.5.4.2.3 Mode of Action for Pancreatic Tumors
As discussed in Section 3.5.20, PACTs were identified in male rats in two 2-year chronic cancer
bioassays {Biegel, 2001, 673581; NTP, 2020, 7330145}. In fact, NTP (2020, 7330145) reported
3-295
-------
DRAFT FOR PUBLIC COMMENT
March 2023
increased incidences of pancreatic acinar cell adenomas in males in all treatment groups, as well
as increased incidence, though non-significant, in female rats from the highest dose group. A
subchronic drinking water exposure study in the LSL-KRasG12D; Pdx-1 Cre (KC) mouse model
for pancreatic cancer also provides evidence that PFOA exposure promotes the growth of
pancreatic lesions {Kamendulis, 2022, 10176439}.
The literature provides two hypothetical MO As for PFOA-induced pancreatic tumors in animal
models, including one study that utilizes a transgenic mouse model to mimic the histologic
progression of pancreatic cancer that occurs in humans {Kamendulis, 2022, 10176439; Klaunig,
2003, 5772415; Klaunig, 2012, 1289837}. The proposed MOAs are: 1) changes in bile acid,
potentially linked to activation of hepatic PPARa, leading to cholestasis, a positive
cholecystokinin (CCK) feedback loop, and acinar cell proliferation; and 2) oxidative stress.
However, the existing database is limited in its ability to determine the relationship between
PFOA exposure and these MOAs, particularly for the PACTs observed in chronic rat studies.
3.5.4.2.3.1 Gastric Bile Alterations
Gastric bile compositional changes or flow alterations can lead to cholestasis, a reduction or
stoppage of bile flow. Cholestasis may cause an increase in CCK, a peptide hormone that
stimulates digestion of fat and protein, causes increased production of hepatic bile, and
stimulates contraction of the gall bladder. There is some research to suggest that pancreatic
acinar cell adenomas may result from increased CCK levels resulting from blocked bile flow
{Obourn, 1997, 3748746}, which may result in a CCK-activated feedback loop that leads to
increased proliferation of secretory pancreatic acinar cells.
PFOA may change bile composition by competing with bile acids for biliary transport.
Upregulation of MRP3 and MRP4 transporters {Maher, 2008, 2919367} and downregulation of
OATPs {Cheng, 2008, 758807} linked to PPARa activation in mice may favor excretion of
PFOA from liver via bile. Minata et al. (2010, 1937251) found that PFOA levels in bile were
much higher in wild-type male mice vs. PPARa-null mice, suggesting a link to PPARa. In this
study, male mice were dosed with 0, 5.4, 10.8, and 21.6 mg/kg/day PFOA for 4 weeks, resulting
in increased total bile acid in PPARa-null mice at the highest dose, which indicated that PFOA-
induced activation of PPARa may result in increased PFOA excretion. This may, in turn, result
in decreased flow of bile acids that compete for the same transporters. Notably, however, these
alterations in male mice occurred at relatively high dose levels compared to those that resulted in
PACTs in male rats {NTP, 2020, 7330145}. There was no further evidence of cholestasis
reported in the literature.
Additionally, there is no evidence of alterations in CCK in animal models or human studies. In
fact, medical surveillance data from male workers at 3M's Cottage Grove plant demonstrated a
significant negative association between CCK levels and serum PFOA {Olsen, 1998, 9493903;
Olsen, 2000, 1424954}. Further, cholestasis was not observed in the workers {Olsen, 2000,
1424954}. It has been suggested that the lack of a positive association may be due to PFOA
levels being too low to increase CCK in humans, although it has been demonstrated that PFOA is
not an agonist for the CCKA receptor that activates CCK release {Obourn, 1997, 3748746}.
Overall, there is not sufficient evidence to determine whether bile acid alterations contribute to
the MOA for PACTs observed after chronic PFOA exposure.
3-296
-------
DRAFT FOR PUBLIC COMMENT
March 2023
3.5.4.2.3.2 Oxidative Stress
More recent literature has suggested a potential role for oxidative stress in pancreatic
carcinogenesis associated with PFOA exposure. Hocevar et al. (2020, 6833720) and Kamendulis
et al. (2022, 10176439) suggest that pancreatic cancer is induced through the activation of the
UPR pathway, which leads to the activation of nuclear factor erythroid 2-related factor 2 (Nrf2),
a regulator of the oxidative stress response, and protein kinase-like endoplasmic reticulum kinase
(PERK), a signaler of endoplasmic reticulum (ER) stress, and subsequent upregulation
antioxidant responses (e.g., SOD gene expression). Activation of the UPR pathway can also
stimulate ROS production. Activation of Sodl in the mouse by the Nrf2 or PERK signaling
pathways can stimulate cell proliferation through increased production of hydrogen peroxide
which can then, in turn, act as a second messenger in mitogen signaling or through its
elimination of ROS, leading to prevention of ROS-stimulated apoptosis {Kamendulis 2022,
10176439}. Activation of PERK through the UPR pathway may also result in increased cytosolic
calcium levels through activation of the inositol 1,4,5-trisphosphate receptor (IP3R), leading to
ER stress and generation of ROS {Hocevar, 2020, 6833720}.
Induction of tumors by PFOA through the UPR signaling pathway is supported by two studies.
Hocevar et al. (2020, 6833720) evaluated PFOA-induced oxidative stress in mouse pancreatic
acinar cells (266-6 cells) treated with 50 |ag/mL PFOA for various durations. PFOA-exposed
cells exhibited increased ER stress as well as activation of PERK, inositol-requiring
kinase/endonuclease la (IREla), and activating transcription factor 6 (ATF6) signaling cascades
of the UPR pathway. Exposure to PFOA at concentrations of 20, 50, or 100 |ag/m L was also
shown to result in time- and dose-dependent increases in cytosolic calcium levels, an effect that
occurred predominantly through activation of IP3R. Altogether, results in this study
demonstrated that PFOA increased intracellular calcium levels through activation of the IP3R,
leading to ER stress, the generation of ROS and oxidative stress and subsequent PERK-
dependent induction of antioxidant genes. The oxidative stress and ROS generated in response to
PFOA may serve as a mechanism through which PFOA may induce pancreatic tumors.
Kamendulis et al. (2022, 10176439) evaluated the ability for PFOA to promote pancreatic cancer
using the KC mouse model, which has a mutation in the KRas gene. KRAS mutations are present
in over 90% of human pancreatic cancers; mutating this gene in mice results in a histologic
progression of pancreatic cancer that mirrors human pancreatic cancer progression, including
formation of pancreatic intraepithelial neoplasia (PanIN). KC mice were exposed to 5 ppm
PFOA in drinking water for up to 7 months and effects were monitored after 4 months and
7 months of exposure. PFOA treatment was demonstrated to increase PanIN at 4 months, as well
as result in a 2-fold increase in pancreatic lesion number. Significant increases in inflammation
were observed in the pancreas after 7 months of exposure. PFOA exposure was associated with
enhanced desmoplasia (collagen deposition) in the pancreas compared to controls and also with
progressive exposure duration.
Oxidative stress was also apparent in the PFOA-treated mice {Kamendulis, 2022, 10176439}.
The authors reported increases in Sod enzyme activity at 4 and 7 months, along with 3-fold
increases in Sodl protein and mRNA levels and increased pancreatic catalase and thioredoxin
reductase activities at 4 months relative to control. Pancreatic malondialdehyde, a product of
oxidized lipids, was increased at 7 months of exposure but not 4 months, indicating a potential
accumulation of oxidative damage over time. Altogether, the results of this study demonstrated
3-297
-------
DRAFT FOR PUBLIC COMMENT
March 2023
that PFOA increased PanIN area and number at 4 months, indicating early lesion formation. The
increased desmoplasia and inflammation (MDA levels) at 7-month exposure suggest PFOA
exposure increased disease severity over time, potentially through prolonged oxidative stress,
resulting in pancreatic cancer progression.
3.5.4.2.4 Mode of Action for Hepatic Tumors
Two high confidence chronic studies on PFOA reported an increased incidence of hepatocellular
adenomas in male rats {Biegel, 2001, 673581; NTP, 2020, 7330145}, one of which also
demonstrated increased incidence of hepatocellular carcinomas specific to male rats exposed to
PFOA perinatally. As described in the subsections below, the available mechanistic evidence
across different in vivo and in vitro models establishes that multiple modes of action (MOA) are
plausible for PFOA-induced liver cancer, including PPARa activation, activation of other
nuclear receptors such as CAR, and an oxidative stress-mediated MOA.
EPA previously concluded that liver tumor development in rats exposed to PFOA was not
relevant to human health because it was determined to be mediated through PPARa activation. A
substantial body of evidence exists suggesting that although PPARa activators cause liver tumors
in rodents, they are unlikely to result in liver tumors in humans due to comparatively low hepatic
PPARa expression, as well as biological differences between rodents and humans in the
responses of events that are downstream of PPARa activation {Corton, 2018, 4862049; U.S.
EPA, 2016, 3603279}. Specifically, there is strong consensus that the MOA for liver tumor
induction by PPARa activators in rodents has limited-to-no relevance to humans, due to
differences in cellular expression patterns of PPARa and related proteins (e.g., cofactors and
chromatin remodelers), as well as differences in binding site affinity and availability {Corton,
2018, 4862049; Klaunig, 2003, 5772415}. However, there is also evidence that other MOAs are
operative in PFOA-induced hepatic tumorigenesis (e.g., cytotoxicity {Felter, 2018, 9642149}
and liver necrosis in PFOA-exposed mice and rats; see Section 3.5.2). Recently published data
suggest that oxidative stress and other mechanistic key characteristics associated with
carcinogens may play a role in liver tumor development, as described further below. The
existence of multiple MOAs in addition to PPARa activation suggest that PFOA-induced liver
cancer in rats may be more relevant to humans than previously thought.
The available literature on mechanisms related to PFOA-induced hepatic tumor development
also supports EPA's prior conclusion that PFOA-induced tumors are likely due to nongenotoxic
mechanisms involving nuclear receptor activation, perturbations of the endocrine system, and/or
DNA replication and cell division {U.S. EPA, 2016, 3603729}.
3.5.4.2.4.1 PPARa Activation
Exposure to several PFAS have been shown to activate PPARa, which is characterized by
downstream cellular or tissue alterations in peroxisome proliferation, cell cycle control (e.g.,
apoptosis and cell proliferation), and lipid metabolism {U.S. EPA, 2016, 3603279}. Notably,
human expression of PPARa mRNA and protein is only a fraction of what is expressed in rodent
models, though there are functional variant forms of PPARa that are expressed in human liver to
a greater extent than rodent models {Klaunig, 2003, 5772415; Corton, 2018; 4862049}.
Therefore, for PPARa activators that act solely or primarily through PPARa-dependent
mechanisms (e.g., Wyeth-14,643 or di-2-ethyl hexyl phthalate), the hepatic tumorigenesis
3-298
-------
DRAFT FOR PUBLIC COMMENT
March 2023
observed in rodents is expected to be infrequent and/or less severe in humans, or not observed at
all {Klaunig, 2003, 5772415; Corton, 2014, 2215399; Corton, 2018, 4862049}.
The MO A for PPARa activator-induced rodent hepatocarcinogenesis consists of the following
sequence of key events: 1) PPARa activation in hepatic cells; 2) alterations in cell growth
signaling pathways (e.g., increases in Kupffer cell activation leading to increases in TNFa); 3)
perturbations of hepatocyte growth and survival (i.e., increased cell proliferation and inhibition
of apoptosis); and 4) selective clonal expansion of preneoplastic foci cells leading to increases in
hepatocellular adenomas and carcinomas {Klaunig, 2003, 5772415; Corton, 2014, 2215399;
Corton, 2018, 4862049}. Modulating factors in this MO A include increased oxidative stress and
activation of NF-kB {Corton, 2018, 4862049}, both of which have been demonstrated for PFOA.
This MOA is associated with, but not necessarily causally related to, non-neoplastic effects
including peroxisome proliferation, hepatocellular hypertrophy, Kupffer cell-mediated events,
and increased liver weight. There is also some overlap between signaling pathways and adverse
outcomes, including turn oogenesis, associated with PPARa activation and the activation or
degradation of other nuclear receptors, such as CAR, PXR, HNF4a, and PPARy {Rosen, 2017,
3859803; Huck, 2018, 5079648; Beggs, 2016, 3981474; Corton, 2018, 4862049}.
The key events underlying PFOA-induced hepatic tumor development through the PPARa MOA
have been demonstrated in both in vivo and in vitro studies and have been discussed in detail
previously {U.S. EPA, 2016, 3603729}, as well as in Sections 3.5.2 and 3.5.3 of this document.
A number of studies illustrate the potential of PFOA to activate human and rodent PPARa. For
example, Buhrke et al. (2013, 232534) demonstrated PPARa activation in human Hep2G cells
after 24-hour exposure to PFOA at a concentration of 25 |iM. PFOA also activated mouse
{Maloney, 1999, 630744; Takacs, 2007, 7922012; Li, 2019, 5387402; Yan, 2015, 3981567} and
human PPARa {Takacs, 2007, 7922012} in cell transfection studies. Gene expression analyses
showed that PPARa activation was required for most transcriptional changes observed in livers
of mice exposed to either PFOA or the known PPARa agonist Wyeth-14,643, demonstrating
PFOA's ability to act as a PPARa agonist {Rosen, 2008, 1290828; 2008, 1290832}. Non-
neoplastic (or pre-neoplastic) events that are associated with PPARa activation include
peroxisome proliferation, hepatocellular hypertrophy, and increases in liver weight. Studies of
PFOA exposure in rodents have reported one or more of these non-neoplastic effects (Section
3.5.2). For example, hepatocellular hypertrophy was observed in one of the two available chronic
carcinogenicity studies of PFOA in rats {NTP, 2020, 7330145}, and both studies observed
increases in liver weights {Biegel, 2001, 673581; NTP, 2020, 7330145}.
There is evidence from in vivo animal bioassays and in vitro studies of Kupffer cell activation, an
indicator of alterations in cell growth, in response to PFOA treatment. Though this mechanism is
itself PPARa-independent, factors secreted upon Kupffer cell activation may be required for
increased cell proliferation by PPARa activators {Corton, 2018, 4862049}. Minata et al. (2010,
1937251) observed a correlation between PFOA exposure and increased tumor necrosis factor-
alpha (TNF-a) mRNA levels in the livers of wild-type (129S4/SvlmJ) and Ppara-nu\\
(129S4/SvJae-PparatmlGonz/J) mice treated with PFOA (<50 |imol/kg/day) for four weeks. TNFa
is a pro-inflammatory cytokine that can be released upon activation of Kupffer cells {Corton,
2018, 4862049}. Further study is needed to understand the potential role of other mediators of
Kupffer cell activation since, unlike PPARa, PPARy is expressed in Kupffer cells and can also
be activated by PFOA.
3-299
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Studies in both rats and mice have demonstrated that PFOA induces peroxisome proliferation in
the liver, an indication of PPARa activation {Elcombe, 2010, 2850034; Minata, 2010, 1937251;
Pastoor, 1987, 3748971; Wolf, 2008, 1290827; Yang, 2001, 1014748}. Gene expression
profiling of livers of PFOA-exposed rats showed changes indicative of hepatocyte cell growth
and proliferation {Martin, 2007, 758419}; exposure of HepG2 cells to low PFOA concentrations
(0.1 and 1 |iM) was associated with increased expression of cell cycle regulators (e.g., Cyclin
Dl, Cyclin El). Higher PFOA concentrations generally had no effect on these genes, but were
associated with increased expression of p53, pl6, and p21 cell cycle regulators {Buhrke, 2013,
232534}. Evidence for cell proliferation in the form of increased mitotic figures and/or bile duct
hyperplasia as observed in PFOA-exposed male mice {Loveless, 2008, 988599}, pregnant mice
{Yahia, 2010, 1332451}, male rats {Elcombe, 2010, 2850034}, and female rats {NTP, 2020,
7330145}. Buhrke et al. (2013, 2325346) also reported increased proliferation in HepG2 cells
exposed to PFOA, in addition to PPARa activation. With respect to inhibition of apoptosis, there
are conflicting reports, with some studies reported decreases in apoptosis following PFOA
exposure {Son, 2008, 1276157}, while others report no effect or an increase in apoptosis {Blake,
2020, 6305864; Elcombe, 2010, 2850034; Minata, 2010, 1937251}. There is also evidence to
support the clonal expansion key event. In an initiation-promotion study of liver tumor induction
in rats, Abdellatif et al. (1990, 2328171) reported that PFOA had promoting activity and
increased the incidence of hepatocellular carcinomas. Jacquet et al. (2012, 2124683) exposed
SHE cells to PFOA at concentrations ranging from 3.7x 10~4 - 37.2 [xM for 6 days with or
without pre-treatment with the tumor initiator benzo-a-pyrene (BaP). PFOA exposure alone did
not induce cell transformation, but PFOA did significantly induce transformation in BaP-
sensitized cells, indicating that PFOA does not alone initiate cell transformation, but may have
tumor promoter-like activity.
Two modulating factors have been proposed as part of the PPARa activation MOA that are
relevant to PFOA: increased ROS and activation of NF-kB. Although there is not enough
evidence to designate these effects as key events in the MOA, they have the potential to alter the
ability of PPARa activators to increase liver cancer, and are thus defined as modulating factors.
PFOA exposure has been demonstrated to cause oxidative stress (detailed below).
3.5.4.2.4.2 Other Nuclear Receptors
In addition to PPARa, there is some evidence that other nuclear receptors, such as CAR, PXR,
PPARy, and ER, can be activated by PFOA. CAR, which has an established adverse outcome
pathway of key events similar to that of PPARa, has been implicated in hepatic tumorigenesis in
rodents. The key events of CAR-mediated hepatic tumors are: 1) CAR activation; 2) altered gene
expression specific to CAR activation; 3) increased cell proliferation; and 4) clonal expansion
leading to altered hepatic foci, leading to 5) liver tumors {Felter, 2018, 9642149}. Non-
neoplastic events associated with this pathway include hypertrophy, induction of CAR-specific
CYP enzymes (e.g., CYP2B), and inhibition of apoptosis. There is evidence that PFOA can
activate CAR and initiate altered gene expression and associative events {Martin, 2007, 758419;
Elcombe, 2010, 2850034; Rosen, 2008, 1290828; Rosen, 2008, 1290832; Rosen, 2017,
3859803}. For example, Martin et al. (2007, 758419) and Elcombe et al. (2010, 2850034)
observed evidence of activation of CAR-related genes in rats following PFOA exposure, and
Wen et al. (2019, 5080582) observed increased CAR activation in PFOA-exposed PPARa
knock-out mice compared to wild-type. Other studies have shown altered gene expression of
transcriptional targets related to CAR in wild type and PPARa knock-out mice {Rosen, 2008,
3-300
-------
DRAFT FOR PUBLIC COMMENT
March 2023
1290828; Rosen, 2008, 1290832; Rosen, 2017, 3859803}. As with PPARa-mediated
turn oogenesis, there are claims that CAR-mediated tumorigenesis seen in animals is not relevant
to humans because CAR activators (e.g., phenobarbital) induce cell proliferation and tumors in
rodents but not in human cell lines {Elcombe, 2014, 2343661}. Hall et al. {2012, 2718645}
noted that there is evidence that CAR in humans is more resistant to mitogenic effects (e.g.,
studies showing that human hepatocytes are resistant to induction of replicative DNA synthesis).
There is also evidence that PFOA can activate other nuclear receptors, such as PXR, PPARy, and
ERa. Martin et al. (2007, 758419) and Elcombe et al. (2010, 2850034) observed evidence of
PPARy agonism and/or activation of PXR-related genes in rats following PFOA exposure, and
Wen et al. (2019, 5080582) reported evidence suggesting increased ERa and PXR activation in
PFOA-exposed PPARa knock-out mice compared to wild-type. PFOA has also been shown to
activate PXR in human HepG2 cells {Zhang, 2017, 3604013}. Buhrke et al. (2013, 2325346)
demonstrated PPARy and PPAR8 activation at PFOA concentrations of >100 |iM in transfected
HEK293 cells, and activation of PPARy by PFOA in HepG2 cells {Buhrke, 2015, 2850235}.
An evaluation of high-throughput screening (HTS) assay data from the ToxCast/Tox21 program
provides further evidence that PFOA activates other nuclear receptors in addition to PPARa.
Chiu et al. (2018, 3981309) evaluated HTS data for PFOA in the context of the ten key
characteristics of carcinogens as described in Smith et al. (2016, 3160486). The assay results
demonstrated PFOA activity in four ER assays (ERa, ERE, ERA LUC, ERa BLA), seven PPAR
and PXR assays (PPARa, PPARy, PPRE, hRRAg, PXR, PXRE, hPXR), two androgen receptor
assays (rAR, AR LUC), five enzyme assays (hBACE, hTie2, gLTB4, hORLl, hPY2), and six
other assays (Nrf2, RXRb, hCYP2C9, AhR, ELG1, and TR LUC Via.) The results suggest a
broad range of PFOA-induced receptor-mediated effects that were not exclusively receptor
effects.
Many of the above-described nuclear receptors are known to play a role in liver homeostasis and
disease and may be driving factors in the hepatotoxicity observed after PFOA exposure;
however, their role in hepatic tumorigenesis is less clear.
3.5.4.2.4.3 Cytotoxicity
There is suggestive evidence that PFOA may act through a cytotoxic MOA. Felter et al. (2018,
9642149) identified the following key events for establishing a cytotoxicity MOA: 1) the
chemical is not DNA reactive; 2) clear evidence of cytotoxicity by histopathology such as the
presence of necrosis and/or increased apoptosis; 3) evidence of toxicity by increased serum
enzymes indicative of cellular damage that are relevant to humans; 4) presence of increased cell
proliferation as evidenced by increased labeling index and/or increased number of hepatocytes;
5) demonstration of a parallel dose response for cytotoxicity and formation of tumors; and 6)
reversibility upon cessation of exposure. As discussed above in the genotoxicity section, there is
some evidence that PFOA can induce DNA damage, although most of the genotoxicity data
indicate that PFOA is not genotoxic. These data indicate that PFOA may be DNA reactive (either
directly or indirectly), but it is not clear if DNA reactivity plays a role in tumorigenicity of
PFOA. Quantitative liver histopathology is available in one study {NTP, 2020, 7330145}.
Significantly increased single cell (hepatocyte) death and in necrosis in male and female was
reported in Sprague-Dawley rats, with a significant dose-response trend.
3-301
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In vitro results regarding apoptosis are variable. Wiels0e et al. (2015, 2533367) observed no
change in LDH release, a marker for cytotoxicity, in HepG2 cells after 24-hour exposure to
PFOA doses as high as 2E"5M, while Panaretakis et al. (2001, 5081525) demonstrated that PFOA
exposure increased ROS generation, which led to activation of caspase-9 and induction of the
apoptotic pathway in HepG2 cells.
Increased cell proliferation or markers of cell proliferation has been reported. Buhrke et al.
(2013, 2325346) determined that PFOA exposures of 10 |iM and 25 |iM for 24 hours resulted in
increased proliferation of HepG2 cells. Increases in metabolic activity were also detected at 10,
25, and 50 |iM exposures. Low PFOA concentrations (0.1 and 1 |iM) were associated with
increased expression of cell cycle regulators Cyclin Dl, Cyclin El, and Cyclin B1 whereas
higher concentrations generally had no effect on these genes (except for increased expression of
Cyclin El at 100 |iM). The higher PFOA concentration of 100 |iM was associated with increased
expression of p53, pl6, and p21 regulators (a non-significant increase was observed at 25 |iM).
Although Wen et al. (2020, 6302274) observed decreasing cell viability with increasing PFOA
exposure in HepG2 cells after 48 hours of exposure (20 to 600 |iM), no change in metabolic
activity was observed. Wen et al. (2020, 6302274) evaluated the impact of PFOA on several
genes involved in cell cycle regulation, proliferation, and apoptosis and found that the expression
of the BAX gene, a regulator of apoptosis, increased at 20, 50, and 150 |iM, and decreased at 100
and 200 |iM. The expression of cell cycle genes CCNA2, CCNE1, and CCNB1 was altered, as
was that of several genes related to cell proliferation (CDKN1A and CDK4)\ at lower
concentrations (50 |iM) of PFOA exposure, a minor increase in expression was observed, while
significant decreases in expression was observed in a dose-dependent manner at concentrations
>50 |iM. Lipid metabolism and transport genes were also altered in the study: increased
expression of lipid anabolism geneACSLl, decreased expression of cholesterol synthesis enzyme
gene HMGCR, decreased expression of fatty acid binding protein gene (FABP1), decreased
expression ACOX2. There was no change in expression in the beta-oxidation acyl-CoA
dehydrogenase enzyme encoding genes ACAD 11 and ACADM.
3.5.4.2.4.4 Genotoxicity
Evidence of PFOA genotoxicity (e.g., chromosomal aberrations, DNA breakage, micronuclei
formation) is mixed, whereas most of the evidence for mutagenicity is consistently negative
(Table 3-16). In an in vivo study in humans, Franken et al. (2017, 3789256) observed an increase
in DNA damage with increasing PFOA exposure, but the effect did not achieve statistical
significance. The authors suggest that the DNA damage may have resulted from induction of
oxidative stress. Additionally, Governini et al. (2015, 3981589) reported that incidence of
aneuploidy and diploidy was increased in PFAS-positive semen samples from non-smokers
(PFOA detected in 75% of the samples) compared to PFAS-negative samples. Of the five
available animal toxicological studies that evaluated PFOA genotoxicity in vivo, only one
yielded a positive result (micronuclei formation in peripheral blood cells from PFOA-exposed
rats {NTP, 2019, 5400977}. A number of studies assessing genotoxicity of PFOA in vitro in
both animal and human cell lines were reviewed. Results for chromosomal aberrations were
negative for PFOA in human lymphocytes both with and without metabolic activation; results in
CHO cells were mostly positive, both with and without activation, but the authors reported that
the positive results were not reproducible. PFOA exposure induced DNA breakage in all in vitro
DNA strand break assays that were reviewed, across three different human cell types. As noted
3-302
-------
DRAFT FOR PUBLIC COMMENT
March 2023
in U.S. EPA (2016, 3603279) and Fenton et al. (2021, 6988520), the clastogenic effects observed
in some PFOA studies may arise from an indirect mechanism related to the physical-chemical
properties of PFOA (specifically, PFOA is not subject to metabolism, it binds to proteins, it
carries a net-negative electrostatic surface charge) and/or as a consequence of oxidative stress.
PFOA is non-mutagenic both with and without activation in several bacterial assays. Although
three positive or equivocal results have been reported, these positive results were either
exclusively at cytotoxic concentrations or were not reproducible (Table 3-16).
The available evidence suggests that PFOA is not mutagenic, but that PFOA exposure may cause
DNA damage, although there is currently no known mechanistic explanation for the interaction
between PFOA and genetic material. Although unlikely, genotoxicity cannot be ruled out as a
potential MO A for PFOA-inducted hepatic tumor formation.
3.5.4.2.4.4.1 Consideration of Other Plausible Mode of Actions
In addition to the evidence supporting modulation of receptor-mediated effects, and potential
genotoxicity, PFOA also exhibits several other key characteristics (KCs) of carcinogens (Section
3.5.3), some of which are similarly directly evident in hepatic tissues.
For example, PFOA appears to induce oxidative stress, another KC of carcinogens, particularly
in hepatic tissues (Section 3.4.1.3.7). Several studies in rats and mice showed evidence of
increased oxidative stress and reduced capacity for defense against oxidants and oxidative
damage in hepatic tissues.
3.5.4.2.4.4.2
3.5.4.2.4.4.3 Epigenetics
There is limited in vivo and in vitro evidence that PFOA induces epigenetic changes, (e.g., DNA
methylation; Section 3.5.3.2) with very little liver-specific data. Two studies conducted with
human cord blood reported associations between PFOA concentration and changes in DNA
methylation {Miura, 2018, 5080353; Kingsley, 2017, 3981315}, whereas an additional three
studies reported no association between maternal PFOA exposure and global DNA methylation
changes in the blood of the children or placenta {Leung, 2018, 4633577; Ouidir, 2020, 6833759;
Liu, 2018, 4926233}. Leung et al. (2018, 4633577), however, did report some evidence of
changes in methylation at CpG sites associated with PFOA exposure in a subset of a Faroese
birth cohort with a mean cord blood PFOA concentration of 2.57 |ig/L. Watkins et al. (2014,
2850906) found no association between DNA methylation and PFOA in adults from the C8
Health Project.
Li et al. (2019, 5387402) observed PFOA-associated epigenetic alterations in the liver of female
mouse pups following maternal exposure to PFOA. Histone acetyltransferase (HAT) levels were
decreased, while histone deacetylase (HDAC) levels were increased at all dose levels. These
results suggest that PFOA inhibits HAT and enhances HDAC activity, which was further
demonstrated by a dose-dependent decrease in acetylation of histones H3 and H4 in the livers of
PFOA-treated mice. The authors proposed that increased HDAC may activate PPARa, based
upon known interactions between specific HDACs and PPARa (specifically, the class III HDAC
SIRT1 deacetylates PPARa resulting in its activation), representing a regulatory role of an event
included in the PPARa MOA.
3-303
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In vitro studies have yielded mixed results with evidence of both hyper- and hypo-methylation of
DNA in response to PFOA exposure (Section 3.5.3.2). For example, Pierozan et al. (2020,
6833637) observed increased global methylation in the first daughter cell subculture of breast
epithelial MCF-10A cells exposed to PFOA, although levels returned to baseline after the second
passage. Two other studies found inverse relationships between global methylation and PFOA
concentration in HepG2 and MCF7 cell lines {Wen, 2020, 6302274; Liu, 2020, 6512127,
respectively}.
3.5.4.2.4.4.4 Oxidative Stress
Results vary regarding the effect of PFOA exposure on markers of oxidative stress in in vitro and
in vivo studies, both with and without a demonstrated relationship to PPARa activation.
Li et al. (2019, 5387402) observed a dose-dependent increase in 8-OHdG, as well as increases in
the antioxidants catalase (CAT) and superoxide dismutase (SOD) (also indicative of oxidative
stress) in the liver of female offspring of Kunming mice exposed to 1, 2.5, 5, or 10 mg/kg/day
PFOA from GD 0-17, with pups sacrificed at PND 21. Serum AST and ALT levels were
significantly increased in the PFOA-treated groups, indicating liver damage. Liver CAT content
significantly increased in the 5 and 10 mg/kg/day dose groups. The authors propose that
oxidative stress occurred through PPARa activation pathways and demonstrated changes in the
mRNA level of PPARa-target genes in the same study. One such target gene is Acoxl, which
was significantly increased in livers of offspring of dams exposed to >2.5 mg/kg/day PFOA.
Overexpression of Acoxl has been reported to generate excess ROS, as ACOX1 is involved in
fatty acid P-oxidation and produces hydrogen peroxide as a byproduct {Kim et al., 2014,
4318185}. This aligns with oxidative stress being proposed as a modulating factor in the
PPARa-activation MOA for rodent hepatic tumors {Corton, 2018, 4862049}, as discussed
above. Another study observed an increase in hydrogen peroxide in the liver of PFOA-exposed
NMRI mice exposed to PFOA in utero (GD 5-9) {Salimi, 2019, 5381528}. Although they did
not measure PPARa targets or PPARa itself, the type of oxidative stress observed aligns with the
modulating factor in the MOA.
In contrast, Minata et al. (2010, 1937251) did not observe an increase in a biomarker of oxidative
stress in wild-type mice exposed to PFOA. The authors treated wild-type (129S4/SvlmJ) and
Ppara-mi\\ (129S4/SvJae-PparatmlGonz/J) mice with PFOA (<50 |amol/kg/day) for four weeks,
after which no changes in 8-OHdG were observed in the wild-type mice. In contrast, a dose-
dependent increase in 8-OHdG levels was observed in the Ppara-rwiW mice, with a significant
increase at 50 |imol/kg/day when compared to controls. The correlation between PFOA exposure
and 8-OHdG was associated with increased tumor necrosis factor-alpha (TNF-a) mRNA levels.
Takagi et al. (1991, 2325496) performed a two-week subchronic (0.02% powdered PFOA in the
diet) in male Fischer 344 rats and evaluated the levels of 8-OHdG in the liver and kidneys after
exposure. The 8-OHdG level was significantly higher in the liver of exposed rats relative to
controls, while there was no change in the kidneys, despite increased weights of both organs.
Another group of rats were administered a single IP injection of PFOA (100 mg/kg) and
sacrificed at days 1, 3, 5, and 8. Results were comparable to that of the dietary exposure study,
with a significant increase in 8-OHdG levels in the liver (by Day 1 following injection) as well
as increased liver weight (by Day 3).
3-304
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA exposure caused increases in 8-OHdG, a biomarker of oxidative stress, in human
lymphoblast cells (TK6) andHepG2 hepatocytes {Yahia, 2014, 2851192; Yao, 2005, 5081563}.
Peropadre et al. (2018, 5080270) observed a slight elevation in 8-OHdG levels in PFOA-exposed
human p53-deficient keratinocytes (HaCaT), and significantly elevated levels eight days
following cessation of PFOA exposure. Several other in vitro studies reported increases in ROS
in PFOA-exposed cells, including HepG2, non-human primate kidney, and human-hamster
hybrid (AL) cells {Panaretakis, 2001, 5081525; Wiels0e, 2014, 2533367; Fernandez Freire,
2008, 2919390; Zhao, 2010, 847496}. In contrast, Florentinet et al. (2011, 2919235) did not
observe increased ROS in HepG2 cells exposed to 5-400 |iM PFOA for 24-hours, despite
increased cytotoxicity at 200 |iM PFOA and higher.
Some of the in vitro studies reported oxidative stress in relation to cell death and/or DNA
damage. For example, Panaretakis et al. (2001, 5081525) investigated ROS, mitochondrial
damage, and caspase-9 following PFOA exposure and determined that PFOA-induced apoptosis
involved a ROS- and mitochondria-mediated pathway. ROS generation (H2O2 and superoxide
anions) was detected in HepG2 cells following exposure to 200 and 400 |iM PFOA. PFOA
treatment also resulted in depolarization of the mitochondria and loss of mitochondrial
transmembrane potential. A population of sub-G0/G2 phase of cell cycle was also observed.
PFOA treatment was also associated with an increase in cells undergoing apoptotic DNA
degradation. Caspase-9 activation was evident in cells exposed to 200 |iM PFOA. The results of
this study suggested that PFOA exposure increased ROS generation, which led to activation of
caspase-9 and induction of the apoptotic pathway in HepG2 cells.
Wiels0e et al. (2015, 2533367) observed a significant increase in ROS production in HepG2 cells
exposed to 2.0E-7, 2.0E-6, and 2.0E-5M PFOA for 24 hours, along with a dose-dependent
increase in DNA damage. Total antioxidant concentration was significantly decreased after 24
hours of exposure to all PFOA concentrations tested. This study demonstrated that genotoxic
effects in vitro are the result of oxidative DNA damage following excess ROS production.
3.5.4.2.4.5 Conclusions
PFOA exposure is associated with several mechanisms that can contribute to carcinogenicity.
There is robust evidence that PFOA activates PPARa and initiates downstream events that lead
to hepatic turn oogenesis, including key events and modulating factors of the PPARa activator-
induced MOA for rodent hepatocarcinogenesis {Klaunig, 2003, 5772415; Corton, 2014,
2215399; Corton, 2018, 4862049}.
Additionally, PFOA exposure is associated with several mechanisms that can contribute to
carcinogenicity, including epigenetic changes and oxidative stress, which may occur in
conjunction with or independently of PPARa activation. It is plausible that these mechanisms
may occur independently of PPARa-dependent mechanisms. These observations are consistent
with literature reviews recently published by state health agencies which concluded that the
hepatotoxic effects of PFOA may not entirely depend on PPARa activation {CalEPA, 2021,
9416932; NJDWQI, 2017, 5024840}. The existence of multiple MOAs in addition to PPARa
activation suggest that PFOA-induced liver cancer in rats may be more relevant to humans than
previously thought. Additional research is warranted to better characterize the MOAs for PFOA-
induced hepatic tumorigenesis.
3-305
-------
DRAFT FOR PUBLIC COMMENT
March 2023
As described in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329},
"[i]n the absence of sufficiently, scientifically justifiable mode of action information, EPA
generally takes public health-protective, default positions regarding the interpretation of
toxicologic and epidemiologic data; animal tumor findings are judged to be relevant to humans,
and cancer risks are assumed to conform with low dose linearity." For the available data
regarding the MOA of PFOA-induced hepatic carcinogenesis, there is an absence of definitive
information supporting a single, scientifically justified MOA; in fact, there is evidence
supporting the potential for multiple plausible MO As. Therefore, EPA takes the health-protective
approach and concludes that the hepatic tumors observed by Biegel et al. (2001, 673581) and
NTP (2020, 7330145) can be relevant to human health.
3.5.4.3 Conclusions
The available data is limited in its ability to provide enough evidence to support conclusions
about potential MO As for PFOA-induced kidney and testicular tumors in humans. Similarly,
there is limited data to support specific MO As for PFOA-induced testicular and pancreatic
tumors in rats. While there is robust evidence that PFOA activates PPARa and initiates
downstream events that lead to hepatic turn oogenesis, there are also reports of PPARa-
independent MO As that could be the underlying the observed hepatocellular adenomas and
carcinomas in rodents treated with PFOA. Additionally, the available in vivo and in vitro assays
provide considerable support that PFOA may induce tumorigenesis through multiple
mechanisms that are considered key characteristics of carcinogens.
3.5.5 Cancer Classification
Under the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}, EPA
reviewed the weight of the evidence and determined that PFOA is Likely to Be Carcinogenic to
Humans, as "the evidence is adequate to demonstrate carcinogenic potential to humans but does
not reach the weight of evidence for the descriptor Carcinogenic to HumansThis determination
is based on the evidence of kidney and testicular cancer in humans and LCTs, PACTs, and
hepatocellular adenomas in rats.
Since publication of the 2016 HESD {U.S. EPA, 2016, 3603279}, the evidence supporting the
carcinogenicity of PFOA has been strengthened. In particular, the evidence of kidney cancer
from high-exposure community studies {Vieira, 2013, 2919154; Barry, 2013, 2850946} is now
supported by evidence of RCC from a nested case-control study in the general population
{Shearer, 2021, 7161466}. In animal models, the evidence of multi-site tumorigenesis reported
in two chronic bioassays in rats {Butenhoff, 2012, 2919192; Biegel, 2001, 673581} is now
supported by evidence from a second chronic bioassay in rats similarly reporting multi-site
tumorigenesis {NTP, 2020, 7330145}.
The Guidelines provide descriptions of data that may support the Likely to Be Carcinogenic to
Humans descriptor; the available PFOA data are consistent with the following factors:
• "an agent demonstrating a plausible (but not definitively causal) association between
human exposure and cancer, in most cases with some supporting biological, experimental
evidence, though not necessarily carcinogenicity data from animal experiments;
• an agent that has tested positive in animal experiments in more than one species, sex,
strain, site, or exposure route, with or without evidence of carcinogenicity in humans;
3-306
-------
DRAFT FOR PUBLIC COMMENT
March 2023
• a rare animal tumor response in a single experiment that is assumed to be relevant to
humans;
• a positive tumor study that is strengthened by other lines of evidence, for example, either
plausible (but not definitively causal) association between human exposure and cancer or
evidence that the agent or an important metabolite causes events generally known to be
associated with tumor formation (such as DNA reactivity or effects on cell growth
control) likely to be related to the tumor response in this case" {U.S. EPA, 2005,
6324329}.
The available evidence indicates that PFOA has carcinogenic potential in humans and at least
one animal model. A plausible, though not definitively causal, association exists between human
exposure to PFOA and kidney and testicular cancers in the general population and highly
exposed populations. As stated in the Guidelines for Carcinogen Risk Assessment, "an inference
of causality is strengthened when a pattern of elevated risks is observed across several
independent studies." Two medium confidence independent studies provide evidence of an
association between kidney cancer and elevated PFOA serum concentrations {Shearer, 2021,
7161466; Vieira, 2013, 2919154}, while two studies in the same cohort provide evidence of an
association between testicular cancer and elevated PFOA serum concentrations {Vieira, 2013,
2919154; Barry, 2013, 2850946}. Additionally, though the Shearer et al. (2021, 7161466) study
showed an increased risk of kidney cancer in the highest PFOA exposure quartile (OR = 2.63),
the relationship was slightly attenuated (OR = 2.19) and not statistically significant after
adjusting for other PFAS. The PFOA cancer database would benefit from additional large high
confidence cohort studies in independent populations.
The evidence of carcinogenicity in animals is limited to three studies using the same strain of rat.
However, the results provide evidence of increased incidence of three tumor types (LCTs,
PACTs, and hepatocellular adenomas) in males administered diets contaminated with PFOA.
Additionally, pancreatic acinar cell adenocarcinomas are a rare tumor type {NTP, 2020,
7330145}. Importantly, site concordance is not always assumed between humans and animal
models; agents observed to produce tumors may do so at the same or different sites in humans
and animals, which appears to be the case for PFOA {U.S. EPA, 2005, 6324329}.
Table 3-17. Comparison of the PFOA Carcinogenicity Database with the Likely Cancer
Descriptor as Described in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329}
Likely to be Carcinogenic to Humans
An agent demonstrating a plausible (but not
definitively causal) association between human
exposure and cancer, in most cases with some
supporting biological, experimental evidence, though
not necessarily carcinogenicity data from animal
experiments
An agent that has tested positive in animal
experiments in more than one species, sex, strain, site,
or exposure route, with or without evidence of
carcinogenicity in humans
PFOA data are consistent with this description.
Epidemiological evidence supports a plausible
association between exposure and cancer, though there
are significant uncertainties regarding the MO As for
tumor types observed in humans. There is supporting
experimental evidence, including carcinogenicity data
from animal experiments.
PFOA data are consistent with this description.
PFOA has tested positive in one species (rat), both sexes,
and multiple sites (liver, pancreas, testes, uterus). There
is also evidence of carcinogenicity in humans.
3-307
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Likely to be Carcinogenic to Humans
A positive tumor study that raises additional biological
concerns beyond that of a statistically significant
result, for example, a high degree of malignancy, or an
early age at onset
A rare animal tumor response in a single experiment
that is assumed to be relevant to humans
A positive tumor study that is strengthened by other
lines of evidence, for example, either plausible (but
not definitively causal) association between human
exposure and cancer or evidence that the agent or an
important metabolite causes events generally known to
be associated with tumor formation (such as DNA
reactivity or effects on cell growth control) likely to be
related to the tumor response in this case
This description is not applicable to PFOA. The report
by NTP (2020, 7330145) does not indicate that perinatal
exposure exacerbates the carcinogenic potential of
PFOA.
PFOA data are consistent with this description. The
pancreatic adenocarcinomas observed in multiple male
dose groups are a rare tumor type in this strain {NTP,
2020, 7330145}.
PFOA data are consistent with this description.
Multiple positive tumor studies in the same strain of rat
are supported by plausible associations between human
exposure and kidney and testicular cancer.
While reviewing the weight of evidence for PFOA, EPA evaluated consistencies of the
carcinogenicity database with other cancer descriptors according to the Guidelines for
Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}. A discussion on these findings is
presented in Section 6.4.
3-308
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4 Dose-Response Assessment
4.1 Non-Cancer
4.1.1 Study and End point Selection
There is evidence from both epidemiological and animal toxicological studies that oral PFOA
exposure may result in adverse health effects across many health outcomes (Section 3.4). Per
recommendations made by the SAB and the conclusions presented in EPA's preliminary
analysis, Proposed Approaches to the Derivation of a Draft Maximum Contaminant Level Goal
for Perfluorooctanoic Acid (PFOA) in Drinking Water, EPA has focused its toxicity value
derivation efforts "on those health outcomes that have been concluded to have the strongest
evidence" {U.S. EPA, 2022, 10476098}. EPA prioritized health outcomes and endpoints with
the strongest overall weight of evidence (evidence demonstrates or evidence indicates) based on
human and animal evidence (Section 3.4 and 3.5) for POD derivation using the systematic
review methods described in Section 2 and the Appendix (see PFOA Appendix). For PFOA,
these health outcomes are immunological, developmental, cardiovascular (serum lipids), and
hepatic effects. EPA considered both epidemiological and animal toxicological studies for POD
derivation.
In the previous section, for hazard judgment decisions (Section 3.4 and 3.5), EPA qualitatively
considered high, medium, and, at times, low confidence studies to characterize the weight of
evidence for each health outcome. However, given the robust database for PFOA, only well-
conducted high or medium confidence human and animal toxicological studies were considered
for POD derivation, as recommended in the IRIS Handbook {U.S. EPA, 2022, 10476098}. Such
human epidemiological studies were available for immunotoxicity, developmental, serum lipid,
and hepatic effects. Preferred animal toxicological studies consisted of medium and high
confidence studies of longer exposure duration (e.g., chronic or subchronic studies vs. 28-day
studies) or with exposure during sensitive windows of development (i.e., perinatal periods) with
exposure levels near the lower dose range of doses tested across the evidence base, along with
medium or high confidence animal toxicological studies evaluating exposure periods relevant to
developmental outcomes. These types of animal toxicological studies increase the confidence in
the RfD relative to other animal toxicological studies because they are based on data with
relatively low risk of bias and are associated with less uncertainty related to low-dose and
exposure duration extrapolations. See Section 6.3 for a discussion of animal toxicological studies
and endpoints selected for POD derivation for this updated assessment compared to those
selected for the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}.
For all other health outcomes (e.g., reproductive, endocrine, nervous, hematological,
musculoskeletal), the evidence integration summary judgment for the human and animal
evidence was suggestive or inadequate and these outcomes were not assessed quantitatively.
Uncertainties related to health outcomes for which the results were suggestive are discussed in
the evidence profile tables provided in the Appendix (See PFOA Appendix), as well as Section
6.5.
4-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.1.1 Hepatic effects
As reviewed in Section 3.4.1.4, evidence indicates that elevated exposures to PFOA are
associated with hepatic effects in humans. As described in Table 3-3, the majority of
epidemiological studies assessed endpoints related to serum biomarkers of hepatic injury (9
medium confidence studies), while several studies also reported on liver disease or injury (4
medium confidence studies) and other serum markers of liver function (3 medium confidence
studies). EPA prioritized endpoints related to serum biomarkers of injury for quantitative
analyses as the reported effects on these endpoints, particularly ALT, were well-represented
within the database and were generally consistent across the available medium confidence
studies. Specifically, all 5 medium confidence studies in the recent literature reported positive
associations between PFOA serum concentrations and ALT in adults, three of which reported
statistically significant responses. Findings for AST and GGT in adults were generally positive
and are supportive of the selection of ALT as an endpoint for dose-response modeling.
Serum ALT measures are considered a reliable indicator of impaired liver function because
increased serum ALT is indicative of leakage of ALT from damaged hepatocytes {Boone, 2005,
782862; Liu, 2014, 10473988; U.S. EPA, 2002, 625713}. Additionally, evidence from both
human epidemiological and animal toxicological studies indicates that increased serum ALT is
associated with liver disease {Ioannou, 2006, 10473853; Ioannou, 2006, 10473854; Kwo, 2017,
10328876; Roth, 2021, 9960592}. Human epidemiological studies have demonstrated that even
low magnitude increases in serum ALT can be clinically significant. For example, a
Scandinavian study in people with no symptoms of liver disease observed that relatively small
increases in serum ALT were associated with liver diseases such as steatosis and chronic
hepatitis C {Mathiesen, 1999, 10293242}. Additionally, a study in Korea found that the use of
lowered thresholds for "normal" serum ALT values showed good prediction power for liver-
related adverse outcomes such as mortality and hepatocellular carcinoma {Park, 2019,
10293238}. Others have questioned the biological significance of relatively small increases in
serum ALT (i.e., less than 2-fold) reported in animal toxicological studies {Hall, 2012,
2718645}, though measures of ALT in these studies can be supported by histopathological
evidence of liver damage.
Additionally, numerous studies have demonstrated an association between elevated ALT and
liver-related mortality (reviewed by Kwo et al. (2017, 10328876)). Furthermore, the American
Association for the Study of Liver Diseases (AASLD) recognizes serum ALT as an indicator of
overall human health and mortality {Kim, 2008, 7757318}. For example, as reported by Kwo et
al. (2017, 10328876), Kim et al. (2004, 10473876) observed that higher serum ALT
concentrations corresponded to an increased risk of liver-related death in Korean men and
women; similarly, Ruhl and Everhart (2009, 3405056; 2013, 2331047) analyzed NHANES data
and observed an association between elevated serum ALT and increased mortality, liver-related
mortality, coronary heart disease in Americans, and Lee et al. (2008, 10293233) found that
higher serum ALT was associated with higher mortality in men and women in Olmstead County,
Minnesota. Furthermore, the American College of Gastroenterology (ACG) recommends that
people with ALT levels greater than 33 (men) or 25 IU/L (women) undergo screenings and
assessments for liver diseases, alcohol use, and hepatotoxic medication use {Kwo, 2017,
10328876}. Results of human and animal toxicological studies as well as the positions of the
AASLD and the ACG demonstrate the clinical significance of increased serum ALT. It is also
4-2
-------
DRAFT FOR PUBLIC COMMENT
March 2023
important to note that while evaluation of direct liver damage is possible in animal studies, it is
difficult to obtain biopsy-confirmed histological data in humans. Therefore, liver injury is
typically assessed using serum biomarkers of hepatotoxicity {Costello et al, 2022, 10285082}.
Results reported in animal toxicological studies are consistent with the observed elevated ALT
indicative of hepatic damage in epidemiological studies. Specifically, studies in rodents found
that oral PFOA treatment resulted in increased relative liver weight (17/20 high and medium
confidence studies), biologically significant alterations in levels of at least one serum biomarker
of liver injury (i.e., ALT, AST, and ALP) (6/9 high and medium confidence studies), and
evidence of histopathological alterations including hepatocyte degenerative or necrotic changes
(12/12 high and medium confidence studies). These hepatic effects, particularly the increases in
serum enzymes and histopathological evidence of liver damage are supportive of elevated ALT
observed in human populations. Mechanistic studies in rodents and limited evidence from in
vitro studies and animal models provide additional support for the biological plausibility and
human relevance of the apical effects observed in animals and suggest possible PPARa-
dependent and -independent MOA for PFOA induced liver toxicity. EPA prioritized studies that
quantitatively reported histopathological evidence of hepatic damage for dose-response modeling
because these endpoints are more direct measures of liver injury than serum biomarkers.
However, the observed increases in liver enzymes in rodents are supportive of the hepatic
damage confirmed during histopathological examinations in several studies.
Four medium confidence epidemiological studies {Gallo, 2012, 1276142; Darrow, 2016,
3749173; Lin, 2010, 1291111; Nian, 2019, 5080307} and one high and one medium confidence
animal toxicological study were considered for POD derivation {NTP, 2020, 7330145; Loveless,
2008, 988599} (Table 4-1). The two largest studies of PFOA and ALT in adults are Gallo et al.
(2012, 1276142) and Darrow et al. (2016, 3749173), both conducted in over 30,000 adults from
the C8 Study. Gallo et al. (2012, 1276142) demonstrated a statistically significant trend in
elevated ALT across deciles and Darrow et al. (2016, 3749173) provides an exposure-response
gradient for PFOA. Two additional studies {Lin, 2010, 1291111; Nian, 2019, 5080307} were
considered by EPA for POD derivation because they reported statistically significant
associations in general populations in the United States and a highly exposed population in
China, respectively. Nian et al. (2019, 5080307) examined a large population of adults in
Shenyang (one of the largest fluoropolymer manufacturing centers in China) as part of the
Isomers of C8 Health Project. Lin et al. (2010, 1291111) observed elevated ALT levels per log-
unit increase in PFOA in an NHANES adult population and these associations remained after
accounting for other PFAS in the regression models. While Lin et al. (2010, 1291111) was
considered for POD derivation, several methodological limitations, including lack of clarity
about base of logarithmic transformation applied to PFOS concentrations in regression models
and the choice to model ALT as an untransformed variable ultimately preclude its use for POD
derivation.
EPA identified two studies in male rodents, NTP (2020, 7330145), a chronic dietary study in
Sprague Dawley rats (see study design details in Section 3.4.4.2.1.2)., and Loveless et al. (2008,
988599), a 29-day gavage dosing study in CD-I mice, for POD derivation. NTP (2020, 7330145)
conducted histopathological examinations of liver tissue in male rats and reported dose-
dependent increases in the incidence of hepatocellular single cell death and hepatocellular
necrosis. As this is one of the few available chronic PFOA toxicity studies with a large sample
4-3
-------
DRAFT FOR PUBLIC COMMENT
March 2023
size (n = 50), numerous and relatively low dose levels, and a comprehensive suite of endpoints,
both the single cell death and necrosis endpoints in males from the 107-week time point were
considered for derivation of PODs.
Similar to the NTP study (2020, 7330145), Loveless et al. (2008, 988599) provides a
comprehensive report of hepatotoxicity, with a low dose range resulting in dose-dependent
increases in histopathological outcomes indicating adversity in male mice gavaged with PFOA
for 29 days. Therefore, the incidences of focal cell necrosis and individual cell necrosis in male
mice from Loveless et al. (2008, 988599) were also considered for the derivation of PODs.
4.1.1.2 Immunological Effects
As reviewed in Section 3.4.2.4, evidence indicates that elevated exposures to PFOA are
associated with immunological effects in humans. As described in Table 3-7, the majority of
epidemiological studies assessed endpoints related to immunosuppression (1 high and 15
medium confidence studies) and immune hypersensitivity (1 high and 16 medium confidence
studies), while several studies (2 medium confidence) also reported on endpoints related to
autoimmune disease. Endpoints related to autoimmune diseases were not further considered for
quantitative assessments as there were a limited number of studies that assessed specific diseases
(e.g., rheumatoid arthritis, celiac disease). Endpoints related to immune hypersensitivity were
also not considered for dose-response analyses. Although the majority (6/9) of the available
medium confidence studies reported consistent increases in the odds of asthma, there were
inconsistencies in effects reported in the same or similar subgroups across these different studies.
These inconsistencies limited the confidence needed to select particular studies and populations
for dose-response modeling. Other immune hypersensitivity endpoints, such as odds of allergies
and rhinoconjunctivitis, had less consistent results reported across medium and high confidence
studies and were therefore excluded from further consideration, though they are supportive of an
association between PFOA and altered immune function.
Evidence of immunosuppression in children reported by epidemiological studies was consistent
across studies and endpoints. Specifically, epidemiological studies reported reduced humoral
immune response to routine childhood immunizations, including lower levels of tetanus and anti-
diphtheria antibodies {Timmerman, 2021, 9416315; Abraham, 2020, 6506041; Grandjean, 2012,
1248827; Budtz-j0rgensen, 2018, 5083631} and rubella {Granum, 2013, 1937228; Pilkerton,
2018, 5080265; Stein, 2016, 3108691} antibody titers. Reductions in antibody response were
observed at multiple timepoints throughout childhood, using both prenatal and childhood
exposure levels, and were consistent across study populations from medium confidence studies.
Measurement of antigen-specific antibodies following vaccinations is an overall measure of the
ability of the immune system to respond to a challenge. The antigen-specific antibody response
is extremely useful for evaluating the entire cycle of adaptive immunity and is a sweeping
approach to detect immunosuppression across a range of cells and signals {Myers, 2018,
10473136}. The SAB's PFAS review panel noted that reduction in the level of antibodies
produced in response to a vaccine represents a failure of the immune system to respond to a
challenge and is considered an adverse immunological health outcome {U.S. EPA, 2022,
10476098}. This is in line with a review by Selgrade (2007, 736210) who suggested that specific
immunotoxic effects observed in children may be broadly indicative of developmental
immunosuppression impacting these children's ability to protect against a range of immune
4-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
hazards—which has the potential to be a more adverse effect that just a single immunotoxic
effect. Thus, decrements in the ability to maintain effective levels of antitoxins following
immunization may be indicative of wider immunosuppression in these children exposed to
PFOA.
As noted by Dewitt et al. (2017, 5926400; 2019, 5080663) as well as subject matter experts on
the SAB's PFAS review panel {U.S. EPA, 2022, 10476098}, the clinical manifestation of a
disease is not a prerequisite for a chemical to be classified as an immunotoxic agent and the
ability to measure clinical outcomes as a result of mild to moderate immunosuppression from
exposure to chemicals in traditional epidemiological studies can be challenging. Specifically, the
SAB noted that "[djecreased antibody responses to vaccines is relevant to clinical health
outcomes and likely to be predictive of risk of disease" {U.S. EPA, 2022, 10476098}. The WHO
Guidance for immunotoxicity risk assessment for chemicals similarly recommends measures of
vaccine response as a measure of immune effects as "childhood vaccine failures represent a
significant public health concern" {WHO, 2012, 10633091}. This response is also translatable
across multiple species, including rodents and humans, and extensive historical data indicate that
suppression of antigen-specific antibody responses by exogenous agents is predictive of
immunotoxicity.
When immunosuppression occurs during immune system development, the risks of developing
infectious diseases and other immunosuppression-linked diseases may increase {Dietert, 2010,
644213}. Immunosuppression linked with chemical stressors is not the same as an
immunodeficiency associated with, for example, genetic-based diseases, but still is an endpoint
associated with potential health risks. Studies of individuals exposed at the extremes of age,
those with existing immunodeficiencies, and those exposed to chronic stress, show that what
may be considered mild to moderate immunosuppression in the general population could result
in increased risk of infections in these more susceptible populations {Selgrade, 2007, 736210}.
Finally, the immune system continues developing after birth; because of this continued
development, exposures to PFAS may have serious and long-lasting consequences {DeWitt,
2019, 5080663; MacGillivray, 2014, 6749084; Selgrade, 2007, 736210}. Hessel et al. (2015,
5750707) reviewed the effect of exposure to nine toxicants on the developing immune system
and found that the developing immune system was at least as sensitive or more sensitive than the
general (developmental) toxicity parameters. Immunotoxicity that occurs in the developing
organism generally occurs at doses lower than required to affect the adult immune system, thus
providing a more sensitive endpoint for assessing risk {vonderEmbse, 2018, 6741321}. Luster et
al. (2005, 2174509) similarly noted that responses to childhood vaccines may be sensitive
enough to detect changes in populations with moderate degrees of immunosuppression, such as
those exposed to an immunotoxic agent.
Results reported in animal toxicological studies are coherent with the observed
immunosuppression in epidemiological studies. Specifically, studies in rodents found that oral
PFOA treatment resulted in reduced immune response (i.e., reduced globulin and
immunoglobulin levels upon immune challenges) (4 medium confidence studies) and altered
immune cell populations (e.g., altered white blood cell counts, altered splenic and thymic
cellularity) (1 high and 4 medium confidence studies). Immunosuppression evidenced by
functional assessments of the immune responses, such as analyses of globulin and
immunoglobulin levels after challenges, are supportive of decreased antibody responses seen in
4-5
-------
DRAFT FOR PUBLIC COMMENT
March 2023
human populations and were therefore prioritized for quantitative assessment. Studies assessing
alterations in immune cell populations were not considered further as there were a limited
number of studies assessing specific endpoints of interest or because effects were observed in
only a single species or sex. Other immune effects observed in animals, such as altered organ
weights and histopathology, were not considered for dose-response assessments as these effects
may be confounded by changes in body weight, effects were not consistent across species, sexes,
or studies, and/or a limited number of studies assessed specific outcomes, though overall, the
results from these endpoints are consistent with evidence indicating alterations in immune
function and response from animal toxicological studies.
Two medium confidence epidemiologic studies {Budtz-j0rgensen, 2018, 5083631; Timmerman,
2021, 9416315} and two medium confidence animal toxicological studies {DeWitt, 2008,
1290826; Loveless, 2008, 988599} were considered for POD derivation (Table 4-1). The
candidate epidemiological studies offer data characterizing antibody responses to vaccinations in
children using a variety of PFOA exposure measures across various populations and
vaccinations. Budtz-j0rgensen and Grandjean (2018, 5083631) investigated anti-tetanus and
anti-diphtheria responses in Faroese children aged 5-7 and Timmerman et al. (2021, 9416315)
investigated anti-tetanus and anti-diphtheria responses in Greenlandic children aged 7-12. Given
the mixed epidemiological results for hypersensitivity and autoimmune disease, these outcomes
were not considered for derivation of PODs. In addition to the results from epidemiological
studies, impaired IgM response reported in Dewitt et al. (2008, 1290826), a 15-day drinking
water exposure study in female mice, and Loveless et al. (2008, 988599), a 29-day study in male
mice, supported the evidence of immunosuppression in humans and were also considered for
POD derivation.
4.1.1.3 Cardiovascular effects
As reviewed in Section 3.4.3.4, evidence indicates that elevated exposures to PFOA are
associated with cardiovascular effects in humans. As described in Table 3-8, the majority of
epidemiological studies assessed endpoints related to serum lipids (1 high and 20 medium
confidence studies) and blood pressure and hypertension (3 high and 14 medium confidence
studies), while several studies also reported on cardiovascular disease (1 high and 5 medium
confidence studies) and atherosclerosis (1 high and 3 medium confidence studies). Endpoints
related to cardiovascular disease and atherosclerosis were not prioritized for dose-response as
they reported mixed or primarily null results. Endpoints related to blood pressure and
hypertension were also not prioritized for quantitative analyses because studies reported no
effects or generally mixed associations, though there was some evidence of associations between
PFOA exposure and continuous measures of blood pressure and risk of hypertension in adults
from the general population and high-exposure communities.
The majority of studies in adults from the general population, including high-exposure
communities, reported positive associations between PFOA serum concentrations and serum
lipids. Specifically, medium confidence epidemiological studies in adults reported positive
associations between PFOA exposure and total cholesterol (TC) (8/10 studies), low-density
lipoprotein (LDL) (6/6 studies), and triglycerides (5/9 studies). Of these three endpoints, EPA
prioritized TC for quantitative assessments because the association was consistently positive in
adults, with some studies reporting statistically significant ORs, this response was more
consistently positive in other populations (i.e., children and pregnant women) compared to LDL
4-6
-------
DRAFT FOR PUBLIC COMMENT
March 2023
and triglycerides, and elevations in TC were reported in a marginally larger number of studies.
Additionally, the positive associations with TC were supported by a recent meta-analysis
restricted to 14 general population studies in adults {U.S. EPA, 2022, 10369698}.
Increased serum cholesterol is associated with changes in incidence of cardiovascular disease
events such as myocardial infarction (MI, i.e., heart attack), ischemic stroke (IS), and
cardiovascular mortality occurring in populations without prior CVD events {D'Agostino, 2008,
10694408; Goff, 2014, 3121148; Lloyd-Jones, 2017, 10694407}. Additionally, disturbances in
cholesterol homeostasis contribute to the pathology of non-alcoholic fatty liver disease (NAFLD)
and to accumulation of lipids in hepatocytes {Malhotra, 2020, 10442471}. Cholesterol is made
and metabolized in the liver, and thus the evidence indicating that PFOA exposure disrupts lipid
metabolism, suggests that toxic disruptions of lipid metabolism by PFOA are indications of
hepatoxicity.
Though results reported in animal toxicological studies support the alterations in lipid
metabolism observed in epidemiological studies, variations in the direction of effect with dose
increases the uncertainty of the biological relevance of these responses in rodents to humans.
Additionally, the available mechanistic data does not help to explain the non-monotonicity of
serum lipid levels and decreased serum lipid levels at higher dose levels in rodents (Section
3.4.3.2). EPA did not derive PODs for animal toxicological studies reporting cardiovascular
effects, such as altered serum lipid levels, due to uncertainties about the human relevance of
these responses.
Three medium confidence epidemiologic studies were considered for POD derivation (Table 4-1)
{Dong, 2019, 5080195; Lin, 2019, 5187597; Steenland, 2009, 1291109}. These candidate
studies offer a variety of PFOA exposure measures across various populations. Dong et al.
(2019, 5080195) investigated the NHANES population (2003-2014), while Steenland et al.
(2009, 1291109) investigated effects in a high-exposure community (the C8 Health Project study
population). Lin et al. (2019, 5187597) collected data from prediabetic adults from the DPP and
DPPOS at baseline (1996-1999). Dong et al. (2019, 5080195) and Steenland et al. (2009,
1291109) excluded individuals prescribed cholesterol medication from their analyses, a potential
confounder for the total cholesterol endpoint, while Lin et al. (2019, 5187597) did not.
4.1.1.4 Developmental effects
As reviewed in Section 3.4.4.4, evidence indicates that elevated exposures to PFOA are
associated with developmental effects in humans. As described in Table 3-10, the majority of
epidemiological studies assessed endpoints related to fetal growth restriction (22 high and 14
medium confidence studies) and gestational duration (11 high and 5 medium confidence studies),
while several studies also reported on endpoints related to fetal loss (2 high and 2 medium
confidence studies) and birth defects (2 medium confidence studies). Findings from the small
number of studies reporting on birth defects were mixed and therefore not prioritized for
quantitative assessments. Although half of the available high and medium confidence studies
reported increased incidence of fetal loss (2/4), EPA did not prioritize this endpoint for dose-
response analyses as there were a relatively limited number of studies compared to endpoints
related to gestational duration and fetal growth restriction and the evidence from high confidence
studies was mixed. The impacts observed on fetal loss are supportive of an association between
PFOA exposure and adverse developmental effects.
4-7
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Approximately half of the available studies reporting metrics of gestational duration observed
increased risk associated with PFOA exposure. Six of the fourteen medium or high confidence
studies reported adverse effects on gestational age at birth and five of the eleven medium or high
confidence studies reported an association with preterm birth. Gestational age was not prioritized
for quantitative analyses because several studies did not report statistically significant results and
some studies reported inconsistent responses across sexes. There were generally more consistent
associations with adverse effects on preterm birth, particularly from the high confidence studies.
However, there were some concerns with sample timing and potential influence of pregnancy
hemodynamics on the observed outcomes, as the majority of studies reporting increased odds of
preterm birth sampled PFOA concentrations later in pregnancy. While overall there appears to be
associations between PFOA exposure and gestational duration, the inconsistencies in the
database and lack of studies sampling in the first trimester of pregnancy reduce the level of
confidence in the responses preferred for endpoints prioritized for dose-response modeling.
The adverse effects on gestational duration were consistent with effects on fetal growth
restriction. The majority of high and medium confidence epidemiological studies (17/25)
reported associations between PFOA and decreased mean birth weight in infants. Studies on
changes in standardized birth weight measures (i.e., z-scores) also generally reported inverse
associations (6/12 studies; 5 high and 1 medium confidence). Low birth weight is clinically
defined as birth weight less than 2,500 g (approximately 5.8 lbs) and can include babies born
SGA (birth weight below the 10th percentile for gestational age, sex, and parity) {JAMA, 2002,
10473200; Mclntire, 1999, 15310; U.S. EPA, 2013, 4158459}. Low birth weight is widely
considered a useful measure of public health {Cutland, 2017, 10473225; Lira, 1996, 10473966;
Vilanova, 2019, 10474271; WHO, 2004, 10473140} and is on the World Health Organization's
(WHO's) global reference list of core health indicators {WHO, 2014, 10473141; WHO, 2018,
10473143}.
Substantial evidence links low birth weight to a variety of adverse health outcomes at various
stages of life. It has been shown to predict prenatal mortality and morbidity {Cutland, 2017,
10473225; U.S. EPA, 2013, 4158459; WHO, 2014, 10473141} and is a leading cause of infant
mortality in the United States {CDC, 2021, 10473144}. Low-birth-weight infants are also more
likely to have underdeveloped and/or improperly-functioning organ systems (e.g., respiratory,
hepatic, cardiovascular), clinical manifestations of which can include breathing problems, red
blood cell disorders (e.g., anemia), and heart failure {Guyatt, 2004, 10473298; JAMA, 2002,
10473200; U.S. EPA, 2013, 4158459; WHO, 2004, 10473140; Zeleke, 2012, 10474317}.
Additionally, low-birth-weight infants evaluated at 18 to 22 months of age demonstrated
impaired mental development {Laptook, 2005, 3116555}.
Low birth weight is also associated with increased risk for diseases in adulthood, including
obesity, diabetes, and cardiovascular disease {Gluckman, 2008, 10473269; Osmond, 2000,
3421656; Risnes, 2011, 2738398; Smith, 2016, 10474151; Ong, 2002, 10474127, as reported in
Yang \, 2022, 10176603}. Poor academic performance, cognitive difficulties {Hack, 2002,
3116212; Larroque, 2001, 10473940}, and depression {Loret de Mola, 2014, 10473992} in
adulthood have also been linked to low birth weight. These associations between low birth
weight and infant mortality, childhood disease, and adult disease establish low birth weight as an
adverse effect. Given the known consequences of this effect, as well as the consistency of the
database and large number of high confidence studies reporting statistically significant odds of
4-8
-------
DRAFT FOR PUBLIC COMMENT
March 2023
this effect, the endpoint of low birth weight in humans was considered for dose-response
modeling.
Results reported in animal toxicological studies are consistent with the observed developmental
toxicity in epidemiological studies. Specifically, studies in rodents found that gestational PFOA
treatment resulted in reduced offspring weight (8/11 studies; 2 high and 6 medium confidence),
decreased offspring survival (6/9 studies; 1 high and 5 medium confidence), reduced maternal
weight (3/8 studies; 2 high and 1 medium confidence), developmental delays (2/2 studies; both
medium confidence), physical abnormalities (2/2 studies; both medium confidence), and altered
placental parameters (2/2 studies; both medium confidence). The developmental effects seen in
the offspring of rodents treated with PFOA are supportive of low birth weight and potential
consequences of low birth weight observed in human populations.
Given the large number of adverse effects identified in the animal toxicological database for the
developmental health outcome, EPA considered only the most sensitive effects in pups supported
by multiple studies for derivation of PODs. EPA focused on the animal toxicological studies
with effects in offspring, as opposed to placental or maternal effects, because these effects
provide concordance with the approximate timing of low birth weight observed in human infants.
Though several studies measured birth weight in dams, there were inconsistencies across the
database, with some studies reporting decreased maternal weight, some reporting no effect, and
some reporting increased maternal weight due to PFOA treatment. These inconsistencies may
stem from the potential confounding effect of reduced offspring weight observed in those same
studies. EPA also focused on endpoints for which multiple animal toxicological studies
corroborated the observed effect, thereby increasing the confidence in that effect. Multiple
animal toxicological studies observed effects at low dose levels and demonstrated a dose-related
response in decreased offspring weight, decreased offspring survival, and developmental delays,
which therefore, were prioritized for dose-response analyses.
Six high confidence epidemiologic studies {Chu, 2020, 6315711; Govarts, 2016, 3230364;
Sagiv, 2018, 4238410; Starling, 2017, 3858473; Wikstrom, 2020, 6311677; Yao, 2021,
9960202} and 3 medium confidence animal toxicological studies {Li, 2018, 5084746; Song,
2018, 5079725; Lau, 2006, 1276159} were considered for POD derivation (Table 4-1). The
candidate epidemiological studies offer a variety of PFOA exposure measures across the fetal
and neonatal window. All six studies reported their exposure metric in units of ng/mL and
reported the P coefficients per ng/mL or ln(ng/mL), along with 95% confidence intervals,
estimated from linear regression models. In addition, the endpoints of decreased fetal body
weights in mice gestationally treated with PFOA from GD 1-17 {Li, 2018, 5084746}, decreased
pup survival at PND 22 in male mice gestationally treated with PFOA from GD 1-17 {Song,
2018, 5079725}, and delayed eye opening in mice gestationally treated with PFOA from GD 1-
17 {Lau, 2006, 1276159} were considered for the derivation of PODs.
Table 4-1 summarizes the studies and endpoints considered for POD derivation.
4-9
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 4-1. Summary of Endpoints and Studies Considered for Dose-Response Modeling and Derivation of Points of Departure
for All Effects in Humans and Rodents
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Notes
Immune Effects
Reduced Antibody
Concentrations for
Diphtheria and
Tetanus
Reduced Antibody
Concentrations for
Rubella
Budtz-Jorgensen and
Grandjean (2018,
508363 If
Medium
Timmerman et al.
(2021, 9416315)
Medium
Human (male Yes Evidence for immune effects is based on decreases in childhood antibody
and female responses to pathogens such as diphtheria and tetanus. Reductions in antibody
children) response were observed at multiple timepoints in childhood, using both prenatal
and childhood exposure levels. Effect was large in magnitude and generally
coherent with epidemiological evidence for other antibody effects.
Granum et al. (2013,
1937228)
Medium
Human (male
and female
children)
No Effect was large in magnitude and generally coherent with epidemiological
evidence for other antibody effects, however, the data were not suitable for
application of a BMR of 1 SD and Vi SD to provide a reasonably good estimate
of 10% and 5% extra risk. The Benchmark Dose Technical Guidance {U.S.
EPA, 2012, 1239433} explains that in a control population where 1.4%are
considered to be at risk of having an adverse effect, a downward shift in the
control mean of one SD results in about 10% extra risk of being at risk of
having an adverse effect. Using a cut off value of 0.1 IU/mL resulted in 0.003%
of the control population at risk of having an adverse effect, a value much
smaller than 1.4% which in turn did not result in 10% extra risk, (see PFOA
Appendix).
Reduced Antibody
Concentrations for
Influenza
Reduced
immunoglobulin M
(IgM) Response
Looker etal. (2014,
2850913)
Medium
Human (male
and female
adults)
No Effect observed in adults coherent with evidence for other antibody effects in
children. The study was not amenable to modeling because of inconsistent
results by vaccine type, lack of dose-response trend, and imprecision of most
results.
Loveless et al. (2008,
988599)
Medium
DeWitt et al. (2008,
1290826)
Medium
C57BL/6N mice Yes Functional assessment indicative of immunosuppression. Immune effects were
(females), consistently observed across multiple studies including reduced spleen and
Crl:CD- thymus weights, altered immune cell populations, and decreased splenic and
1(ICR)BR mice thymic cellularity. Reduced IgM response is coherent with epidemiological
(males) evidence of reduced immune response to vaccinations.
Developmental Effects
4-10
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Notes
Decreased Birth
Weight
Chu et al. (2020,
6315711)
High
Govarts et al. (2016,
3230364)
High
Sagiv et al. (2018,
4238410)
High
Starling et al. (2017,
3858473)
High
Wikstrom et al.
(2020, 6311677)
High
Human (male
and female
infants)
Yes
Evidence for developmental effects is based on consistent adverse effects for
FGR including birthweight measures which are the most accurate endpoint.
Some deficits were consistently reported for birth weight and standardized birth
weight in many high and medium confidence cohort studies. Effect was
generally large in magnitude and coherent with epidemiological evidence for
other biologically related effects.
Decreased Birth
Weight
Yao et al. (2021,
9960202)
High
Human (male
and female
infants)
No
Effect was supportive of epidemiological evidence for this effect, but the
exposure median in this study was at least lOx higher than the other studies
considered (see PFOA Appendix).
Decreased Offspring
Survival
Song et al. (2018,
5079725)
Medium
Kunming mice
(Fi males and
females)
Yes
Effect was consistently observed across multiple studies and species. Supported
by the prenatal loss observed by Lau et al. (2006, 1276159) and Wolf et al.
(2007, 1332672). Lau et al. (2006, 1276159) was not amenable to benchmark
dose modeling (See PFOA appendix). Song et al. (2018, 5079725) was selected
over Wolf et al. (2007, 1332672) because Song et al. (2018, 5079725) had more
dose groups and tested a lower dose range. BMD modeling results for Wolf et
al. (2007, 1332672) are provided in the PFOA appendix for comparison
purposes.
Effect was consistently observed across multiple studies and species. Supported
by reduced offspring weight observed by Lau et al. (2006, 1276159), Wolf et al.
(2007, 1332672), Blake et al. (2020, 6305864), and Butenhoff et al. (2004,
1291063). While the data from Lau et al. (2006, 1276159), Wolf et al. (2007,
1332672) and Li et al. (2018, 5084746) were not amenable to BMD modeling
(see PFOA appendix), Li et al. (2018, 5084746) was selected because the study
tested a relatively large number of dose groups and had decreased variability
compared to the other studies. Note that decreases in maternal body weight
were also considered for POD derivation but was not a selected endpoint
because the decreased fetal body weight could be a potential confounder and
was found to be a more sensitive effect.
Decreased Fetal Li et al. (2018, Kunming mice Yes
Body Weight 5084746) (Fi males and
Medium females)
4-11
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Notes
Delayed Time to Eye Lau et al. (2006, CD-I mice (Fi
Opening 1276159) males and
Medium females)
Yes Effect also observed in Wolf et al. (2007, 1332672). Lau et al. (2006, 1276159)
was prioritized for dose response because this study tested more dose groups (5)
with a lower dose range (1, 3, 5,10 and 20 mg/kg/day) than Wolf et al. (2007,
1332672) (2 dose groups - 3 and 5 mg/kg/day). Additionally, the data from
Wolf et al. (2007, 1332762) were not amenable to BMD modeling (see PFOA
Appendix).
Serum Lipid Effects
Increased Total
Cholesterol
Dong et al. (2019,
5080195)
Medium
Lin et al. (2019,
5187597)
Medium
Steenland et al.
(2009, 1291109)b
Medium
Human (male Yes Effect was consistent and observed across multiple populations including
and female general population adults {Dong, 2019, 5080195; Lin, 2019, 5187597}
adults) (NHANES) and the C8 Health project high-exposure community {Steenland,
2009, 1291109}, as well as when study designs excluded individuals prescribed
cholesterol medication, minimizing concerns of bias due to medical
intervention {Dong, 2019, 5080195; Steenland, 2009, 1291109}.
Hepatic Effects
Increased ALT
Increased ALT
Galloetal. (2012,
1276142)
Medium
Darrow et al. (2016,
3749173)b
Medium
Nian et al. (2019,
5080307)
Medium
Human (male Yes Effect was consistent and observed across multiple populations including
and female general population adults {Lin, 2010, 1291111} (NHANES) and high-exposure
adults) communities including the C8 Health Project {Darrow, 2016, 3749173; Gallo,
2012, 1276142} and Isomers of C8 Health Project in China {Nian, 2019,
5080307}.
Linetal. (2010,
1291111)
Medium
Human (male
and female
adults)
No While this is a large nationally representative population, several
methodological limitations preclude its use for POD derivation. Limitations
include lack of clarity about base of logarithmic transformation applied to
PFOA concentrations in regression models, and the choice to model ALT as an
untransformed variable, a departure from the typically lognormality assumed in
most of the ALT literature.
Necrosis (focal, Loveless et al. (2008, Crl:CD-
individual cell, both) 988599) l(ICR)BRmice
in the Liver Medium (males),
Yes Effect was accompanied in both studies by other liver lesions including
cytoplasmic alteration and apoptosis. Necrotic liver cells were also observed in
male mice in Crebelli et al. (2019, 5381564) and pregnant dams in Blake et al.
4-12
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived? Notes
NTP (2020,
7330145)
High
Sprague-
Dawley rats
(males)
(2020, 6305864). Effect is further supported by changes in serum ALT levels in
animals and humans. Data from females were not considered for POD
derivation as they appear to be less sensitive, potentially due to toxicokinetic
differences between the sexes in rats.
Notes: ALT = alanine transaminase; BMD = benchmark dose; F1 = first generation; NHANES = National Health and Nutrition Examination Survey; POD = point of departure.
a Supported by Grandjean et al. (2012, 1248827), Grandjean et al. (2017, 3858518), and Grandjean et al. (2017,4239492).
b See Section 6.6.3 for discussion on the approach to estimating BMDs from regression coefficients.
4-13
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.2 Estimation or Selection of Points of Departure (PODs)
for RfD Derivation
Consistent with EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433}, the
BMD and 95% lower confidence limit on the BMD (BMDL) were estimated using a BMR
intended to represent a minimal, biologically significant level of change. Thq Benchmark Dose
Technical Guidance {U.S. EPA, 2012, 1239433} describes a hierarchy by which BMRs are
selected, with the first and preferred approach being the use of a biological or toxicological basis
to define what minimal level of response or change is biologically significant. If that biological
or toxicological information is lacking, the guidance document recommends BMRs that could be
used in the absence of information about a minimal clinical or biological level of change
considered to be adverse—specifically, a BMR of one standard deviation (SD) change from the
control mean for continuous data or a BMR of 10% extra risk for dichotomous data. When
severe or frank effects are modeled, a lower BMR can be adopted. For example, developmental
effects are frequently serious effects, and the Benchmark Dose Technical Guidance suggests that
studies of developmental effects can support lower BMRs. BMDs for these effects may employ a
BMR of 0.5 SD change from the control mean for continuous data or a BMR of 5% for
dichotomous data {U.S. EPA, 2012, 1239433}. A lower BMR can also be used if it can be
justified on a biological and/or statistical basis. The Benchmark Dose Technical Guidance (page
23; {U.S. EPA, 2012, 1239433}) shows that in a control population where 1.4% are considered
to be at risk of having an adverse effect, a downward shift in the control mean of one SD results
in a -10%) extra risk of being at risk of having an adverse effect. A BMR smaller than 0.5 SD
change from the control mean is generally used for severe effects (e.g., 1% extra risk of cancer
mortality).
Based on rationales described in EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433, the IRIS Handbook {U.S. EPA, 2022, 10476098} and past IRIS assessment precedent,
BMRs were selected for dose-response modeling of PFOA-induced health effects for individual
study endpoints as described below and summarized in Table 4-2 along with the rationales for
their selection. For this assessment, EPA took statistical and biological considerations into
account to select the BMR. For dichotomous responses, the general approach was to use 10%
extra risk as the BMR for borderline or minimally adverse effects and either 5% or 1% extra risk
for adverse effects, with 1% reserved for the most severe effects. For continuous responses, the
preferred approach for defining the BMR was to use a preestablished cutoff for the minimal level
of change in the endpoint at which the effect is generally considered to become biologically
significant (e.g., greater than or equal to 42 IU/L serum ALT in human males {Valenti, 2021,
10369689}) In the absence of an established cutoff, a BMR of 1 SD change from the control
mean, or 0.5 SD for effects considered to be severe, was generally selected. Specific
considerations for BMR selection for endpoints under each of the priority non-cancer health
outcomes are described in the subsections below. Considerations for BMR selection for cancer
endpoints are described in Section 4.2.
4.1.2.1 Hepatic Effects
Modeling elevated human ALT used cutoff levels of 42 IU/L for males and 30 IU/L for females,
based on the most recent sex-specific upper reference limits {Valenti, 2021, 10369689}. The
baseline prevalence of elevated ALT is estimated as 14% and 13% in U.S. male and female
4-14
-------
DRAFT FOR PUBLIC COMMENT
March 2023
adults (aged 20 and older), respectively (see PFOA Appendix). Therefore, the BMR was defined
as a 5% increase in the number of people with ALT values above the cutoffs. Although the
Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} recommends a BMR of 10%
extra risk for dichotomous data when biological information is not sufficient to identify the
BMR, in this situation, such a BMR would result in a doubling of risk.
For the adverse effect of single cell and focal liver necrosis observed in rats following PFOA
exposure, there is currently inadequate available biological or toxicological information to permit
determination of a minimal biologically significant response level. Therefore, in accordance with
EP A's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433}, a BMR of 10% extra
risk was used (dichotomous data; see Table 4-2).
4.1.2.2 Immune Effects
For the developmental immune endpoint of decreased diphtheria and tetanus antibody response
in children found to be associated with PFOA exposure, the BMD and the BMDL were estimated
using a BMR of 0.5 SD change from the control mean (see Table 4-2). Consistent with EPA's
Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433}, EPA typically selects a 5%
or 0.5 standard deviation (SD) benchmark response (BMR) when performing dose response
modeling of data from an endpoint resulting from developmental exposure. Because Budtz-
J0rgensen and Grandjean (2018, 5083631) and Timmerman et al. (2021, 9416315) assessed
antibody response after PFAS exposure during gestation and childhood, these are considered
developmental studies {U.S. EPA, 1991, 732120} based on EPA's Guidelines for Developmental
Toxicity Risk Assessment, which includes the following definition:
"Developmental toxicology - The study of adverse effects on the developing
organism that may result from exposure prior to conception (either parent), during
prenatal development, or postnatally to the time of sexual maturation. Adverse
developmental effects may be detected at any point in the lifespan of the
organism."
EPA guidance recommends the use of a 1 or 0.5 SD change in cases where there is no accepted
definition of an adverse level of change or clinical cut-off for the health outcome {U.S. EPA,
2012, 1239433}. A 0.5 SD was selected since the health outcome is developmental and there is
no accepted definition of an adverse level of change or clinical cut-off for reduced antibody
concentrations in response to vaccination. Therefore, EPA performed the BMDL modeling using
a BMR equivalent to a 0.5 SD change in log2-transformed antibody concentrations, as opposed
to a fixed change in the antibody concentration distributions {U.S. EPA, 2012, 1239433}.
For the effect of reduced IgM response, there is currently inadequate available biological or
toxicological information to permit determination of a minimal biologically significant response
level. Therefore, in accordance with EPA's Benchmark Dose Technical Guidance {U.S. EPA,
2012, 1239433}, a BMR of 1 SD change from the control mean was employed (see Table 4-2).
4.1.2.3 Cardiovascular Effects
Modeling human cholesterol used a cutoff level of 240 mg/dL for elevated serum total
cholesterol, consistent with the American Heart Association's definition of hypercholesterolemia
{NCHS, 2019, 10369680}. Recent data (for years 2015-2018) show that the percentage of U.S.
adults aged 20 and older with total cholesterol >240 mg/dL is 11.5% {NCHS, 2019, 10369680}.
4-15
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Therefore, the BMR was defined as a 5% increase in the number of people with total cholesterol
values above 240 mg/dL. Although the Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433} recommends a BMR of 10% extra risk for dichotomous data when biological
information is not sufficient to identify the BMR, in this situation, such a BMR would result in a
doubling of risk.
4.1.2.4 Developmental Effects
For the developmental endpoint of decreased birth weight in infants associated with PFOA
exposure, the BMD and the BMDL were estimated using a BMR of 5% extra risk, given that this
level of response is typically used when modeling developmental responses from animal
toxicology studies, and that low birthweight confers increased risk for adverse health effects
throughout life {Hack, 1995, 8632216; Reyes, 2005, 1065677; Tian, 2019, 8632212}. Low birth
weight is clinically defined as birth weight less than 2,500 g (approximately 5.8 lbs) and can
include babies born SGA (birth weight below the 10th percentile for gestational age, sex, and
parity) {JAMA, 2002, 10473200; Mclntire, 1999, 15310; U.S. EPA, 2013, 4158459}.
For decreased fetal and pup weights and decreased pup survival observed in animal studies, a
BMR of 5% relative deviation and 0.5 SD from the control was employed, respectively (see
Table 4-2). This is consistent with EPA's Benchmark Dose Technical Guidance {U.S. EPA,
2012, 1239433} and the IRIS Handbook {U.S. EPA, 2022, 10476098}, which note that studies
of adverse developmental effects represent a susceptible lifestage and can support BMRs that are
lower than 10% extra risk (dichotomous data) and 1 SD change from the control mean
(continuous data).
A 5% relative deviation in markers of growth in gestational exposure studies (e.g., fetal weight)
that do not lead to death has generally been considered an appropriate biologically significant
response level and has been used as the BMR in final IRIS assessments (e.g., U.S. EPA (2003,
1290574), U.S. EPA (2004, 198783), and U.S. EPA (2012, 3114808)). Additionally, the 5%
BMR selection is statistically supported by data which compared a BMR of 5% relative
deviation for decreased fetal weight to NOAELs and other BMR measurements, including 0.5
standard deviation, and found they were statistically similar {Kavlock, 1995, 75837}.
For the effects time to eye opening, there is currently inadequate available biological or
toxicological information to permit determination of minimal biologically significant response
levels. Therefore, in accordance with EP A's Benchmark Dose Technical Guidance {U.S. EPA,
2012, 1239433}, a BMR of 1 SD change from the control mean was employed (results for this
endpoints is averaged across a dose group and are therefore continuous data).
Table 4-2. Benchmark Response Levels Selected for BMD Modeling of Health Outcomes
Endpoint
BMR Rationale
Immune Effects
Reduced antibody
0.5 SD Consistent with EPA guidance. EPA typically selects a
concentrations for diphtheria and
5% or 0.5 standard deviation (SD) benchmark response
tetanus
(BMR) when performing dose response modeling of data
from an endpoint resulting from developmental exposure
and selects a 1 or 0.5 SD change in cases where there is no
4-16
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
BMR
Rationale
Reduced immunoglobulin M
(IgM) response
accepted definition of an adverse level of change or clinical
cut-off for the health outcome {U.S. EPA, 2012, 1239433}
1 SD Insufficient information available to determine
minimal biologically significant response level. The
available biological or toxicological information does not
allow for determination of a minimal biologically
significant response level for this adverse effect, and so a
BMR of one SD was used as per EPA guidance {U.S.
EPA, 2012,1239433}
Developmental Effects
Decreased Birth Weight in
Infants or Decreased Fetal Body
Weight in Rodent Offspring
Decreased Pup Survival
Delayed Time to Eye Opening
5% extra risk of
exceeding adversity
cutoff (hybrid
approach)
0.5 SD
1 SD
Consistent with EPA guidance. EPA typically selects a
5% or 0.5 standard deviation (SD) benchmark response
(BMR) when performing dose response modeling of data
from an endpoint resulting from developmental exposure
{U.S. EPA, 2012, 1239433}. There is biological
Consistent with EPA guidance. EPA typically selects a
5% or 0.5 standard deviation (SD) benchmark response
(BMR) when performing dose response modeling of data
from an endpoint resulting from developmental exposure
{U.S. EPA, 2012, 1239433}
Insufficient information available to determine
minimal biologically significant response level. The
available biological or toxicological information does not
allow for determination of a minimal biologically
significant response level for this adverse effect, and so a
BMR of one SD was used as per EPA guidance {U.S.
EPA, 2012, 1239433}
Serum Lipids
Increased Cholesterol
5% extra risk of
exceeding adversity
cutoff (hybrid
approach)
Response rate of 5% extra risk is reasonable, whereas
a 10% BMR would result in a doubling of risk.
Although EPA's Benchmark Dose Technical Guidance
{U.S. EPA, 2012, 1239433} recommends a BMR based on
a 10% extra risk for dichotomous endpoints when
biological information is not sufficient to identify the
BMR, in this situation such a BMR would result in a
highly improbable doubling of risk
Hepatic Effects
Increased ALT
Single Cell and Focal Liver
Necrosis
5% extra risk of
exceeding adversity
cutoff (hybrid
approach)
10%
Response rate of 5% extra risk is reasonable, whereas
a 10% BMR would result in a doubling of risk.
Although EPA's Benchmark Dose Technical Guidance
{U.S. EPA, 2012, 1239433} recommends a BMR based on
a 10% extra risk for dichotomous endpoints when
biological information is not sufficient to identify the
BMR, in this situation such a BMR would result in a
highly improbable doubling of risk
Insufficient information available to determine minimal
biologically significant response level. The available
biological or toxicological information does not allow for
determination of a minimal biologically significant
response level for this adverse effect, and so a BMR of
4-17
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
BMR
Rationale
10% was used as per EPA guidance {U.S. EPA, 2012,
1239433}
Notes: ALT = alanine transaminase; BMD = benchmark dose; BMR = benchmark response; CDC = Centers for Disease
Control; SD = standard deviation.
4.1.3 Pharmacokinetic Modeling Approaches to Convert
Administered Dose to Internal Dose in Animals and Humans
4.1.3.1 Pharmacokinetic Model for Animal Internal Dosimetry
Following review of the available models in the literature (see Section 3.3.2), EPA chose the
Wambaugh et al. (2013, 2850932) model to describe PFOA dosimetry in experimental animals
based on the following criteria:
• availability of model parameters across the species of interest,
• agreement with out-of-sample datasets (see PFOA Appendix), and
• flexibility to implement life-course modeling.
These criteria originated from the goal of accurately predicting internal dose metrics for
toxicology studies that were selected for dose-response analysis. These studies involved rats,
mice, and non-human primates, and these were the species of interest necessary to have available
model parameters. Good agreement with out-of-sample datasets shows that the model
performance is good compared to both the data used to identify model parameters and to external
data. This increases confidence that the model can be used to make accurate predictions of
internal dose metrics for the toxicology studies, which can also be seen as external. The ability to
implement life-course modeling was necessary to properly predict internal dose metrics for
developmental studies and endpoints as the animal transitioned through numerous life-stages.
In this case, an oral dosing version of the original model structure introduced by Andersen et al.
(2006, 818501) and summarized in Section 3.3.2 was selected for having the fewest number of
parameters that would need estimation. In addition, the Wambaugh et al. (2013, 2850932)
approach allowed for a single model structure to be used for all species in the toxicological
studies allowing for model consistency for the predicted dose metrics associated with LOAELs
and NOAELs from 13 animal toxicological studies of PFOA.
The Wambaugh et al. (2013, 2850932) model was selected for pharmacokinetic modeling for
animal internal dosimetry for several important reasons: 1) it allowed for sex-dependent
concentration-time predictions for PFOA across all three species of interest, 2) it adequately
predicted dosimetry of newer datasets published after model development, and 3) it was
amendable to addition of a life stage component for predicting developmental study designs.
These analyses are further described in the subsections below. Uncertainties and limitations of
the selected modeling approach are described in Section 6.6.1.
4-18
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.3.1.1 Animal Model Parameters
Pharmacokinetic parameters for different species represented in the Wambaugh et al. (2013, 2850932) model are presented in Table
4-3.
Table 4-3. PK Parameters from Wambaugh et al. (2013, 2850932) Meta-Analysis of Literature Data for PFOA
Parameter
Units
CD1 Mouse
(F)a
C57BL/6 Mouse
(F)a
Sprague-Dawley Rat
(F)a
Sprague-Dawley Rat
(M)a
Cynomolgus Monkey
(M/F)a
Body Weightb (BW)
kg
0.02
0.02
0.20
0.24
7 (M), 4.5 (F)
(0.16-0.23)
(0.21-0.28)
Cardiac Output0 (Qcc)
L/h/kg074
8.68
8.68
12.39
12.39
19.8
Absorption Rate (ka)
1/h
290
340
1.7
1.1
230
(0.6-73,000)
(0.53 -69,000)
(1.1-3.1)
(0.83-1.3)
(0.27-73,000)
Central Compartment
L/kg
0.18
0.17
0.14
0.15
0.4
Volume (Vcc)
(0.16-2.0)
(0.13-2.3)
(0.11-0.17)
(0.13-0.16)
(0.29-0.55)
Intercompartment
1/h
0.012
0.35
0.098
0.028
0.0011
Transfer Rate (ki2)
(3.1 x e-10-38,000)
(0.058-52)
(0.039-0.27)
(0.0096-0.08)
(2.4 xe-10- 35,000)
Intercompartment
Unitless
1.07
53
9.2
8.4
0.98
Ratio (Rv2:V2i)
(0.26 - 5.84)
(11-97)
(3.4-28)
(3.1-23)
(0.25-3.8)
Maximum Resorption
|imol/h
4.91
2.7
1.1
190
3.9
Rate (Tmaxc)
(1.75-2.96)
(0.95 - 22)
(0.25 - 9.6)
(5.5 - 50,000)
(0.65 - 9,700)
Renal Resorption
|imol
0.037
0.12
1.1
0.092
0.043
Affinity (Kt)
(0.0057-0.17)
(0.033 -0.24)
(0.27-4.5)
(3.4 x e~4- 1.6)
(4.3 x e-5- 0.29)
Free Fraction
Unitless
0.011
0.034
0.086
0.08
0.01
(0.0026-0.051)
(0.014-0.17)
(0.031 -0.23)
(0.03-0.22)
(0.0026-0.038)
Filtrate Flow Rate
Unitless
0.077
0.017
0.039
0.22
0.15
(Qfiic)
(0.015-0.58)
(0.01-0.081)
(0.014-0.13)
(0.011-58)
(0.02 - 24)
Filtrate Volume (Vr,ic)
L/kg
0.00097
7.6 x e~5
2.6 x e~5
0.0082
0.0021
(3.34 x e-9- 7.21)
(2.7 x e~10- 6.4)
(2.9 x e~10- 28)
(1.3 x e-8- 7.6)
(3.3 x e-9- 6.9)
Notes: F = female; M = male.
Means and 95% credible intervals (in parentheses) from Bayesian analysis are reported. For some parameters, the distributions are quite wide, indicating uncertainty in that
parameter (i.e., the predictions match the data equally well for a wide range of values).
aData sets modeled for the CD1 mouse were from Lou et al. (2009, 2919359), for the C57BL/6 mouse were from DeWitt et al. (2008, 1290826), for
the rat were from Kemper (2003, 6302380), and for the monkey from Butenhoff et al. (2004, 3749227).
b Estimated average body weight for species used except with Kemper (2003, 6302380) where individual rat weights were available and assumed to be constant.
c Cardiac outputs obtained from Davies and Morris (1993, 192570).
4-19
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.3.1.2 Out-of-Sample Comparisons
To evaluate the model's ability to predict PFOA concentration-time data in the species of
interest, EPA compared model fits to in vivo datasets either not considered in or published after
the 2016 HESD (Table 4-4). For rats, this included Kudo et al. (2002, 2990271), Kim et al.
(2016, 3749289), and Dzierlenga et al. (2020, 5916078). Model simulations demonstrated good
agreement with available data for adult time-course PFOA PK predictions in the rat. For mice
however, only one adult PFOA study was available for comparison {Fujii, 2015, 2816710} and
that study only tracked PFOA concentrations through 24 hours. As mentioned when this study is
discussed in Section 3.3.2.1, a 24 hr observation window is insufficient to accurately estimate the
terminal excretion half-life of PFOA. Therefore, only the original study used for parameter
determination, Lou et al. (2009, 2919359), was compared to model simulations. This comparison
approach demonstrated agreement with the in vivo data.
Using the Wambaugh et al. (2013, 2850932) model, EPA predicted the half-life, Vd, and
clearance and compared these species-specific predictions to values obtained from in vivo studies
when data were available.
Because male mouse parameters are not available for PFOA, only female parameters are used for
all PFOA modeling in mice. This assumption is addressed in Wambaugh et al. (2013, 2850932)
and is based on a lack of evidence for sex-dependent PK differences in the mouse.
Table 4-4. Model Predicted and Literature PK Parameter Comparisons for PFOA
Male
Female
tl/2,a
(days)
tl/2,P
(days)
Vd,a Vd,p
(L/kg) (L/kg)
CL
(L/d/kg)
tl/2,a
(days)
tl/2,P
(days)
Vd,a
(L/kg)
vd,p
(L/kg)
CL
(L/d/kg)
Rat
Model
Literature
5.8
1.64a,
2.8b
16.5
10.25b
0.12 0.35
0.11^°, 0.15b c
0.0147
0.0473,
0.013b
0.16
0.19a,
0.028b
2.84
0.22b
0.16 2.81
0.17a,c, 0.12b,c
0.686
0.6133,
0.81b
Mouse
Model
Literature
—
—
— —
—
17.8
18.9
0.18
0.19
0.007
Notes: CL = clearance; PK = pharmacokinetic; ti/2,a = initial-phase elimination half-life; ti/2,p = terminal-phase elimination half-
life; Vd,a = volume of distribution during the initial phase; Vd, p = volume of distribution during the terminal phase.
a Information obtained from Kim et al. (2016, 3749289).
bInformation obtained from Dzierlenga et al. (2020, 5916078).
c Literature volumes of distribution represent central compartment volumes from a one-compartment or two-compartment model.
4.1.3.1.3 Life Course Modeling
The Wambaugh et al. (2013, 2850932) model was modified to account for gestation, lactation,
and post-weaning phases (Figure 4-1). Using the original model structure and published
parameters, simulations assumed that dams were dosed prior to conception and up to the date of
parturition. Following parturition, a lactational phase involved PFOA transfer from the
breastmilk to the suckling pup where the pup was modeled with a simple one-compartment PK
model. Finally, a post-weaning phase utilized the body-weight scaled Wambaugh model to
D-4
-------
DRAFT FOR PUBLIC COMMENT March 2023
simulate dosing to the growing pup and accounted for filtrate rate as a constant fraction of
cardiac output.
Gestation Lactation Post-weaning
Figure 4-1. Model Structure for Life Stage Modeling
Model parameters for three-compartment model are the same as those described earlier. Pup-specific parameters include milk
consumption in kgmiik/day (Rmiik), infant-specific volume of distribution (Vd), and infant-specific half-life (ti/2).
This methodology was adapted from Kapraun et al. (2022, 9641977) and relies on the following
assumptions for gestation/lactation modeling:
• During gestation and up through the instant birth occurs, the ratio of the fetal
concentration (mg of substance per mL of tissue) to the maternal concentration is
constant.
• Infant animal growth during the lactional period is governed by the infant growth curves
outlined in Kapraun et al. (2022, 9641977).
• Rapid equilibrium between maternal serum PFOA and milk PFOA is assumed and
modeled using a serum:milk partition coefficient.
• All (100%) of the substance in the breast milk ingested by the offspring is absorbed by
the offspring.
• The elimination rate of the substance in offspring is proportional to the amount of
substance in the body and is characterized by an infant-specific half-life that is a fixed
constant for any given animal species as described in Table 4-5 below.
• Following the lactation period, infant time course concentrations are tracked using the
more physiologically-based Wambaugh model to model post-weaning exposure and
infant growth.
A simple one-compartment model for infant lactational exposure was chosen because of
differences between PFOA Vd reported in the literature and Wambaugh et al. (2013, 2850932)
model-predicted Vd following extrapolation to a relatively low infant body weight. Because Vd
is assumed to be extracellular water in human, Goeden et al. (2019, 5080506) adjusts for life
stage-specific changes in extracellular water using an adjustment factor where infants have 2.1
times more extracellular water than adults resulting in a larger Vd. However, this large difference
in extracellular water is not observed in rats {Johanson, 1979, 9641334}. Johanson (1979,
9641334) demonstrated a 5% decrease in blood water content from early postnatal life
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(-0.5 weeks) to adulthood (> 7 weeks) in the rat. Therefore, EPA used the literature reported Vd
{Dzierlenga, 2020, 5916078; Lou, 2009, 2919359} for the one compartment model to describe
infant toxicokinetics. Finally, the Wambaugh et al. (2013, 2850932) model was not
parameterized on a post-partum infant, and it was not possible to evaluate the mechanistic
assumptions for renal elimination with postnatal toxicokinetic data. While there is one study that
doses PFOA in young, post-weaning, juvenile animals {Hinderliter, 2006, 3749132},
concentrations at only two time points are reported for each age group making it not possible to
estimate infant/juvenile pharmacokinetic parameters such as half-life. Therefore, the parameters
listed in Table 4-5 in a one-compartment gestation/lactation model were used in conjunction with
the parameters published in Wambaugh et al. (2013, 2850932) to predict developmental dose
metrics for PFOA.
Table 4-5. Additional PK Parameters for Gestation/Lactation for PFOA
Parameter
Units
Rat
Mouse
Maternal Milk:Blood Partition Coefficient (Pmiik) Unitless
0.1lab
0.32e
Fetus:Mother Concentration Ratio (Rfm)
Unitless
0.42b
0.25f
Elimination Half-Life (ti/2)
Days
2.23°
18.5s
Volume of Distribution (Vd)
L/kg
0.18d
0.28
Starting Milk Consumption Rate C r"m,n )
kgmiiv/day
0.001h
0.000P
Week 1 Milk Consumption Rate (rVik)
kgmiiv/day
0.003h
0.00031
Week 2 Milk Consumption Rate (r^nk)
kgmiiv/day
0.0054h
0.000541
Week 3 Milk Consumption Rate (r\nin:)
kgmiik/dav
0.0059h
0.000591
Notes: PK = pharmacokinetic.
information obtained from Loccisano et al. (2013, 1326665) (derived from Hinderliter et al. (2005,1332671)).
b Information obtained from Hinderliter et al. (2005, 1332671).
c Average of male/female half-lives reported in Dzierlenga et al. (2020, 5916078), Kim et al. (2016, 3749289), and Kemper et al.
(2003, 6302380).
information obtained from Kim et al. (2016, 3749289) and Dzierlenga et al. (2020, 5916078).
information obtained from Fujii et al. (2020, 6512379).
fInformation obtained from Blake et al. (2020, 6305864).
g Information obtained from Lou et al. (2009, 2919359).
information obtained from Kapraun et al. (2022, 9641977) (adapted from Lehmann et al. (2014, 2447276)).
information obtained from Kapraun et al. (2022, 9641977) (mouse value is 10% of rat based on assumption that milk ingestion
rate is proportional to body mass).
These developmental-specific parameters include the maternal milk:blood PFOA partition
coefficient (Pmiik), the ratio of the concentrations in the fetus(es) and the mother during
pregnancy (Rfm), the species-specific in vivo determined half-life (ti/2) and Vd for PFOA, and the
species-specific milk consumption rate during lactation (r'miik) for the ith week of lactation. Milk
rate consumptions are defined as:
• r°miik, the starting milk consumption rate in kg milk per day (kg/d);
• ^miik, the (average) milk consumption rate (kg/d) during the first week of lactation (and
nursing);
• r2miik, the (average) milk consumption rate (kg/d) during the second week of lactation; and
• r3miik, the (average) milk consumption rate (kg/d) during the third week of lactation.
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
where Rmiik used in the model is a piecewise linear function comprising each r'miik depending on
the week of lactation.
Using this gestation/lactation model, EPA simulated two studies for PFOA exposure (one in
mice and one in rats) to ensure the model predicted the time-course concentration curves for both
the dam and the pup. For all gestation/lactation studies, time zero represents conception followed
by a gestational window (21 days for the rat, 17 days for the mouse). Dosing prior to day zero
represents pre-mating exposure to PFOA.
Figure 4-2 demonstrates the model's ability to predict gestation/lactation study design in the rat
for dams exposed to 30 mg/kg/day PFOA giving birth to pups who are exposed through lactation
{Hinderliter, 2005, 1332671}. Comparatively, Figure 4-3 demonstrates model fits for PFOA
exposure in mice from a cross-fostering study {White, 2009, 194811}. In each case, the original
Wambaugh et al. (2013, 2850932) model with increasing maternal weight predicts dam
concentrations in female rats and mice while the one-compartmental lactational transfer model
predicts infant concentrations for pups exposed both in utero and through lactation only.
PFOA: 30 mg/kg/day
I 102
u
§ 10*
<
Q
£ 1Q-2
0
15 20 25 30 35
time [days]
0 5 10 15 20 25 30 35
Time [days]
Figure 4-2. Gestation/Lactation Predictions of PFOA in the Rat
Top panel represents time-course model predicted dam concentrations (solid line) where open diamonds (0) represent the in vivo
dam concentrations reported in Hinderliter et al. (2005, 1332671) and x's represent the model-predicted value at the reported
time. Bottom panel demonstrates the model predicted pup concentrations (solid line) where open diamonds (0) represent the
reported pup concentrations in Hinderliter et al. (2005,1332671) with PFOA exposure is from the breast milk. Vertical dashed
line represents birth.
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
PFOA: 5 img/kg/day
Time [days]
Figure 4-3. Gestation/Lactation Predictions of PFOA in the Mouse in a Cross-Fostering
Study
Top panel represents predicted dam concentrations while bottom panel represents the predicted pup concentrations from White et
al. (2009, 194811). Solid lines (-) represent a 5 mg/kg/day maternal dose paired with nursing pups that were exposed to PFOA in
ntero and open diamonds (0) represent the reported dam and infant concentrations for this exposure scenario. Comparatively, dot-
dashed lines (• -) represent the simulations from the cross-fostering study where dams were exposed to 5 mg/kg/day PFOA and
pups bom to the control dam were exposed through lactation. Open triangles (V) represent the reported dam and infant
concentrations for this cross-foster study.
The purpose of the animal PBPK model is to make predictions of internal dose in lab animals
used in toxicity studies and extrapolate these internal dose points-of-departure to humans.
Therefore, to evaluate its predictive utility for risk assessment, a number of dose-metrics across
life stages were selected for simulation in a mouse, rat, monkey, or human. Concentrations of
PFOA in blood were considered for all the dose-metrics. For studies in adult animals the dose-
metric options were generally a maximum blood concentration (Cmax, mg/L) and a time averaged
blood concentration i.e., the area under the curve over the duration of the study (AUC,
mg * day/L) or the blood concentration over the last 7 days (Ciast7, mg/L). In developmental
studies, dose-metrics were developed for the dam, the fetus (during gestation), and the pup
(during lactation) for both time Cmax and averaged blood concentrations (Cavg). In the dam, the
Cmax and Cavg, were calculated over a range of life stages: during gestation (CaVg dam gest), during
lactation (Cavg damjact), or combined gestation and lactation (Cavg dam gest iact). In pups for Cmax,
two different life stages were calculated either during gestation or lactation (Cmax_pup gest,
Cmax pup jact). In pups for time averaged metrics, a Cavg was calculated during gestation, lactation,
or combined gestation and lactation (Cavg_pup_gest, Cavg PuP jact, and Cavg PuP gest jact). Finally, for
NTP, 2020, 7330145, an additional dose metric was derived which averages out the
concentration in the pup from conception to the end of the 2 years (Cavg PuP total). Specifically, it
adds the area under the curve in gestation/lactation to the area under the curve from diet (post-
weaning) and then divides by two years.
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.3.2 Pharmacokinetic Model for Human Dosimetry
The key factors considered in model determination were to implement a human model from the
literature that was able to model gestational and lactational exposure to infants, that was able to
describe time course changes in serum concentration due to changes in bodyweight during
growth, and that required minimal new development. Previous modeling efforts suggest that
limiting model complexity helps to prevent errors and facilitates rapid implementation
{Bernstein, 2021, 9639956}. For the human and animal endpoints of interests, serum
concentration was identified as a suitable internal dosimetry target, which provides support for
using a simpler model that did not have individual tissue dosimetry. For these reasons, EPA
selected the one-compartment human developmental model published by Verner et al. (2016,
3299692). Several alternative models to EPA's updated version of the Verner et al. (2016,
3299692) model for the calculation of PODhed from an internal POD were considered. This
included consideration of full PBPK models (i.e., the Loccisano family of models {Loccisano,
2011, 787186; Loccisano, 2012, 1289830; Loccisano, 2012, 1289833; Loccisano, 2013,
1326665}), as well as other one-compartment PK models (e.g., Goeden et al. (2019, 5080506)).
Discussion on the justification for selection of the Verner et al. (2016, 3299629) model as the
basis for the pharmacokinetic modeling approach used for PFOA is available in Sections 6.6.2
and 6.7.
Several adjustments were undertaken to facilitate the application of the model for this use. First,
the model was converted from acslX language to an R/MCSim framework. This allows the code
to be more accessible to others by updating it to a contemporary modeling language, as acslX
software is no longer available or supported. The starting point for the conversion to R/MCSim
was another model with a similar structure that was in development by EPA at that time
{Kapraun, 2022, 9641977}. Second, the modeling language conversion body weight curves for
non-pregnant adults were revised based on CDC growth data for juveniles and values from
EPA's Exposure Factors Handbook in adults {Kuczmarski, 2002, 3490881; U.S. EPA, 2011,
786546}. Linear interpolation was used to connect individual timepoints from these two sources
to produce a continuous function over time. Body weight during pregnancy was defined based on
selected studies of maternal body weight changes during pregnancy {Portier, 2007, 192981;
Carmichael, 1997, 1060457; Thorsdottir, 1998, 4940407; Dewey, 1993, 1335605; U.S. EPA,
2011, 786546}. Age-dependent breastmilk intake rates were based on the 95th percentile
estimates from EPA's Exposure Factors Handbook and was defined relative to the infant's
bodyweight {U.S. EPA, 2011, 786546}.
A third modification was the update of parameters: the half-life, the volume of distribution (Vd),
the ratio of PFOA concentration in cord blood to maternal serum, and the ratio of PFOA
concentration in breastmilk and maternal serum. Details for how these parameters were updated
are given in the following paragraphs. In the model, half-life and Vd are used to calculate the
clearance, which is used in the model directly and is also used for calculation of steady-state
concentrations in adults. Other than half-life and, because of that, clearance, the updated
parameters were similar to the original parameters (Table 4-6). The results of the new R model
and updated acslX model with the original parameters were essentially identical (see PFOA
Appendix). With the updated parameters, the predicted PFOA serum concentrations are
approximately 70% of the original values during pregnancy, and the child's serum concentration
is approximately 60% of the original values during the first year of life.
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The use of the Verner model in humans presents a substantial advancement in approach for
endpoints in children compared to the previous EPA assessment of PFOA {U.S. EPA, 2016,
3603279}. The previous assessment did not explicitly model children, but instead applied an
uncertainty factor to an RfD based on long-term adult exposure to account for the potential for
increased susceptibility. The current approach explicitly models PFOA exposure to infants
during nursing and the rapid growth of children, who do not reach steady state until near
adulthood. This allows for a more accurate estimation of exposures associated with either serum
levels in children or dose metric from developmental animal toxicological studies. The Verner
model also explicitly models the mother from her birth through the end of breastfeeding which
allows for the description of accumulation in the mother prior to pregnancy followed by
decreasing maternal levels during pregnancy. Detailed modeling of this period is important for
dose metrics based on maternal levels during pregnancy, especially near term, and on cord blood
levels.
Application of the updated Verner model to three cohorts with paired maternal measurements
and subsequent samples in children between ages of 6 months and 6 years showed good
agreement between reported and predicted serum levels in the children (see PFOA Appendix).
This suggests that the assumptions made governing lactational transfer and the selected half-life
value are reasonable. A local sensitivity analysis was also performed to better understand the
influence of each parameter on model output (see PFOA Appendix).
Table 4-6. Updated and Original Chemical-Specific Parameters for PFOA in Humans
Parameter
Updated Value
Original Value"
Volume of Distribution (mL/kg)
170b
170
Half-life (yr)
2.7°
3.8
Clearance (mL/kg/d)
0.120d
0.085
Cord Serum:Maternal Serum Ratio
0.83e
0.79
Milk: Serum Partition Coefficient
0.049f
0.058
Notes:
a Verner et al. (2016, 3299692).
b Thompson et al. (2010,2919278).
cLietal. (2017, 9641333).
d Calculated from half-life and volume of distribution. CI = Vd * ln(2)/ti/2.
e Average values for total PFOA Cord Serum:Maternal Serum ratios (see PFOA Appendix). This is a similar approach to that
used by Verner et al. (2016, 3299692), but also includes studies made available after the publication of that model.
f Average value of studies as reported in Table 4-7. This is a similar approach to that used by Verner et al. (2016, 3299692), but
also includes studies made available after the publication of that model.
EPA selected a reported half-life value from an exposure to a study population that is
demographically representative of the general population, with a clear decrease in exposure at a
known time, with a high number of participants and a long follow-up time. Based on these
criteria, a half-life of 2.7 years was determined for PFOA as reported in Li et al. (2017, 9641333;
2018, 4238434). This value comes from a large population (n = 455) who originally had
contaminated drinking water for which the study documents the decrease in exposure levels after
the installation of filtration with an average final serum sample taken 3.9 years after the
beginning of water filtration. Li et al. (2018, 4238434) also reported a similar half-life of
2.7 years for PFOA in a separate community with a similar study design. In that study, serial
blood samples were collected from participants after the beginning of drinking water filtration
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
after a long period of exposure to drinking water contaminated with PFOA. The second study
involved 106 participants with a median number of 6 samples per person but with only a two-
year observation period Li et al. (2017, 9641333). The good agreement between the second study
and the previous, larger study in diverse populations support the use of this value as a good
estimate of the PFOA elimination half-life.
A summary of PFOA half-life values is presented in the Appendix (see PFOA Appendix).
Uncertainties related to EPA's selected half-life are discussed in Section 6.6.2.
The updated value for human Vd of PFOA, 170 mL/kg, was sourced from Thompson et al.
(2010, 2919278) who used a one-compartment PK model. This calculation involves several
assumptions: that the participants' serum concentrations are at steady-state, their exposure can be
estimated from the drinking water concentration in their community, there is 91% bioavailability
for exposure from drinking water, and the half-life of PFAS is 2.3 years, which comes from the
report of Bartell et al. (2010, 379025). EPA considered updating this parameter to 200 mL/kg,
which is the value that would be calculated using the EPA chosen half-life value of 2.7 years.
However, the value of 2.3 years was calculated under very similar conditions as the other data in
the Thompson et al. (2010, 2919278) population and 2.3 years may better reflect the clearance
rate in that specific population at that time. This calculation was performed in a population with
PFOA contamination. Vd is a parameter that is relatively easily obtained from an analysis of PK
data from controlled experimental studies, as it is related to the peak concentration observed after
dosing and is expected to be similar between human and non-human primates {Mordenti, 1991,
9571900}. For comparison, the optimized Vd for PFOA from oral dosing in monkeys was
140 mL/kg {Andersen, 2006, 818501}.
Another group has approached the calculation of Vd by taking the average of reported animal and
human values and reported values of 200 mL/kg for PFOA {Gomis, 2017, 3981280}. This
calculation included the Vd value from Thompson et al. (2010, 2919278) and did not include
additional values derived from human data. The resulting average value shows that the value
from Thompson et al. (2010, 2919278) is reasonable; EPA selected the Thompson et al. (2010,
2919278) result based on the fact that it is the only value derived from human data that EPA
considers to be reliable for risk estimation in the general population.
A summary of PFOA Vd values is presented in the Appendix (see PFOA Appendix).
Uncertainties related to EPA's selected Vd are discussed in Section 6.6.2.
In the original model, the ratio of PFOA concentration in cord blood to maternal serum, and the
ratio of PFOA concentration in breastmilk and maternal serum were based on an average of
values available in the literature; here, EPA identified literature made available since the original
model was published and updated those parameters with the averages of all identified values
(Table 4-7). The values for cord blood to maternal serum ratio are presented in the Appendix
(see PFOA Appendix). One restriction implemented on the measurements of the cord blood to
maternal serum ratio was to only include reports where the ratio was reported, and not to
calculate the ratio from reported mean cord and maternal serum values.
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 4-7. Summary of Studies Reporting the Ratio of PFOA Levels in Breastmilk and
Maternal Serum or Plasma
Source
HERO ID
Milk: Maternal
Plasma Ratio
Included in Verner et al.
(2016,3299692) Analysis
Haug et al. (2011,2577501)
2577501
0.038
No
Seung-Kyu Kim etal. (2011, 2919258)
2919258
0.025
No
Liu et al. (2011,2919240)
2919240
0.11
No
Cariou et al. (2015, 3859840)3
3859840
0.034
Yes
Sunmi Kim et al. (2011, 1424975)b
1424975
0.04
Yes
Verner et al. (2016, 3299692)
3299692
0.058°
-
Additional Studies
-
0.049d
-
Notes: Whether studies were included in the analysis of Verner et al. (2016, 3299692) is noted. The reported values were based
on the mean of ratios in the study populations except when noted otherwise.
a Median result based on the report of Pizzurro et al. (2019, 5387175).
b Median result as reported by the authors.
c Average value of milk:maternal plasma ratio used by Verner et al. (2016, 3299692).
d Average value of milk:maternal plasma ratio with the inclusion of additional studies not in the original analysis. This value was
used in the human PK model.
This updated model was used to simulate the HED from the animal PODs that were obtained
from BMD modeling of the animal toxicological studies (see PFOA Appendix). It was also used
to simulate selected epidemiological studies (Section 4.1.1.2) to obtain a chronic dose that would
result in the internal POD obtained from dose-response modeling (see PFOA Appendix). For
PODs resulting from chronic exposure, such as a long-term animal toxicological study or an
epidemiological study on an adult cohort, the steady state approximation was used to calculate a
PODhed that would result in the same dose metric after chronic exposure. For PODs from
exposure to animals in developmental scenarios, the updated Verner model was used to calculate
a PODhed that results in the same dose metric during the developmental window selected. The
updated Verner model was also used to calculate a PODhed for PODs based on epidemiological
observations of maternal serum concentration during pregnancy, cord blood concentration, and
serum concentrations in children.
The pharmacokinetic modeling code for both the updated Wambaugh et al. (2013, 850932) and
Verner et al. (2013, 299692) models that was used to calculate human equivalence doses is
available in an online repository (http s: // github. com/LJ SEP A./O W -PF O S-1 VtCLG-support-
PK-modelsY The model code was thoroughly QA'd through the established EPA Quality
Assurance Project Plan (QAPP) for PBPK models {U.S. EPA, 2018, 4326432}.
4.1.4 Application of Pharmacokinetic Modeling for Animal-
Human Extrapolation of PFOA Toxicological End points and
Dosimetric Interpretation of Epidemiological End points
Table 4-8 displays the POD and estimated internal and PODheds for immune, developmental,
cardiovascular (serum lipids), and hepatic endpoints from animal and/or human studies selected
for the derivation of candidate RfDs. The PODs from epidemiological studies (immune,
developmental, hepatic, and serum lipid endpoints) were derived using benchmark dose
modeling (see PFOA Appendix) which provided an internal serum concentration in mg/L. The
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
internal dose PODs were converted to a PODhed using the modified Verner model described in
Section 4.1.3.1.3 to calculate the dose that results in the same serum concentrations. Specifically,
reverse dosimetry was performed by multiplying an internal dose POD by a model predicted
ratio of a standard exposure and the internal dose for that standard exposure. This expedited
procedure can be performed because the human model is linear, that is, the ratio of external and
internal dose is constant with dose. Additional details are provided below and in
D-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 4-8.
The PODs from the animal toxicological studies were derived by first converting the
administered dose to an internal dose as described in Section 4.1.3.1.1. The rationale for the
internal dosimetric selected for each endpoint is described in the Appendix (see PFOA
Appendix). Because a toxicological endpoint of interest results from the presence of chemical at
the organ-specific site of action, dose response modeling is preferentially performed on internal
doses rather than administered doses and assumes the internal dose metric is proportional to the
target tissue dose In addition, the non-linear elimination described in Wambaugh et al. (2013,
2850932) requires conversion to an internal dose as the relationship between internal and
external dose will not scale linearly. The internal doses were then modeled using the Benchmark
Dose Software (BMDS) (see PFOA Appendix for additional modeling details). The internal dose
animal PODs were converted to a PODhed using the model described in Section 4.1.3.1.3.
Reverse dosimetry for the animal PODs used the ratio of standard exposure and internal dose as
was applied to PODs from epidemiological data. For animal toxicological studies using the
average concentration over the final week of the study (Ciast7), the PODhed is the human dose
that would result in the same steady-state concentration in adults. When a concentration internal
dose metric in the pup during lactation and/or gestation was selected, the PODhed is the dose to
the mother that results in the same average concentration in the fetus/infant over that period.
This approach for interspecies extrapolation follows the EPA's guidance to prefer the use of a
PK or PBPK model over the use of a data-derived extrapolation factor (DDEF) {EPA, 2014,
2520260}. A PK model allows for predictions of dosimetry for specific exposure scenarios in
animals and humans and can incorporate PK details such as maternal accumulation and
subsequent gestation/lactational transfer to a fetus/infant. Using a hierarchical decision-making
framework, a DDEF approach is only considered when a validated PK or PBPK model is not
available. Furthermore, EPA considers DDEF values based on the ratio of maximum blood
concentration from acute, high-dose exposures to likely not be protective for typical exposure
scenarios to humans, chronic low-dose exposure or lactational exposure to a nursing infant
{Dourson, 2019, 6316919}. While a repeat dose DDEF has been presented {Dourson, 2019,
6316919}, this factor relied on maximum concentrations from Elcombe et al. (2013, 10494295),
for which the results are not considered relevant to the general population as discussed in Section
4.1.3.2.
4-30
-------
DRAFT FOR PUBLIC COMMENT March 2023
Table 4-8. PODheds Considered for the Derivation of Candidate RfD Values
Endpoint
Reference,
Confidence
Strain/
Species/Sex
POD Type,
Model
POD
POD Internal
Dose/Internal
Dose Metric3
PODhed
(mg/kg/day)
Immunological Effects
Decreased serum
Budtz-Jorgensen and Grandjean (2018,
Human, male and female;
BMDLo.5sd,
3.47 ng/mL
3.05 xlO-7
anti-tetanus
508363 l)b
PFOA concentrations at age
Linear
antibody
Medium
five years and anti-tetanus
concentration in
antibody serum
children
concentrations at age
seven years
Budtz-Jorgensen and Grandjean (2018,
Human, male and female;
BMDLo.5sd,
3.31 ng/mL
5.21 xlO"7
508363 l)b
PFOA concentrations in the
Linear
Medium
mother0 and anti-tetanus
antibody serum
concentrations at age 5 years
Timmerman et al. (2021, 9416315)
Human, male and female;
BMDLo.5sd,
2.26 ng/mL
3.34xl0-7
Medium
PFOA concentrations and
anti-tetanus antibody
concentrations at ages 7-
10 years
Linear
Decreased serum
Budtz-Jorgensen and Grandjean (2018,
Human, male and female;
BMDLo.5sd,
3.32 ng/mL
2.92xl0-7
anti-diphtheria
508363 l)b
PFOA concentrations at age
Linear
antibody
Medium
five years and anti-
concentration in
diphtheria antibody serum
children
concentrations at age
seven years
Budtz-Jorgensen and Grandjean (2018,
Human, male and female;
BMDLo.5sd,
1.24 ng/mL
1.95xl0"7
508363 l)b
PFOA concentrations in the
Piecewise
Medium
mother0 and anti-diphtheria
antibody serum
concentrations at age 5 years
Timmerman et al. (2021, 9416315)
Human, male and female;
BMDLo.5sd,
1.49 ng/mL
2.20 xlO"7
Medium
PFOA concentrations and
anti-diphtheria antibody
Linear
4-31
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Reference,
Confidence
Strain/
Species/Sex
POD Type,
Model
POD
POD Internal
Dose/Internal
Dose Metric"
PODhed
(mg/kg/day)
concentrations at ages 7-
10 years
Decreased IgM
Dewitt et al. (2008, 1290826)
Mouse, female Study 1
BMDLisd,
18.2 mg/L
2.18>
<10"3
response to SRBC
Medium
Polynominal
4
Clast7
Decreased IgM
Dewitt et al. (2008, 1290826)
Mouse, female Study 2
NOAELd
1.88 mg/kg/d
45.3 mg/L
5.43>
<10"3
response to SRBC
Medium
ay
Clast7
Decreased IgM
Loveless et al. (2008, 988599)
Mouse, male
BMDLisd,
57.6 mg/L
6.91>
<10"3
response to SRBC
Medium
Exponential 3
Clast7
Developmental Effects
Low Birth Weight
Chu et al. (2020, 6315711)
High
Human, male and female;
PFOA serum concentrations
in third trimester
BMDL5RD,
Hybrid
2.0 ng/mL
3.15 >
<10-7
Govarts et al. (2016, 3230364)
Human, male and female;
BMDL5RD,
1.2 ng/mL
2.28>
<10-7
High
PFOA concentrations in
umbilical cord
Hybrid
Sagivetal. (2018, 4238410)
Human, male and female;
BMDL5RD,
9.1 ng/mL
1.21 >
<10^
High
PFOA serum concentrations
in first trimester
Hybrid
Starling et al. (2017, 3858473)
Human, male and female;
BMDL5RD,
1.8 ng/mL
2.65>
<10-7
High
PFOA serum concentrations
in second and third
trimesters
Hybrid
Wikstrom et al. (2020, 6311677)
Human, male and female;
BMDL5RD,
2.2 ng/mL
2.92>
<10-7
High
PFOA serum concentrations
in first and second trimesters
Hybrid
Decreased
Song et al. (2018, 5079725)
Kunming Mice, Fi males
BMDL0.5SD,
12.3 mg/L
6.40>
<10^
Offspring Survival
Medium
and females
Polynomial
3rd degree
C(iyg pup lact
Decreased Fetal
Li et al. (2018, 5084746)
Kunming Mice, Fi males
NOAELd
1 mg/kg/day
8.5 mg/L
1.44>
<10-3
Body Weight
Medium
and females
Cavg pup gest
Delayed Time to
Lau et al. (2006, 1276159)
CD-I Mice, Fi males and
BMDLisd,
10.1 mg/L
1.71 >
<10-3
Eye Opening
Medium
females
Polynomial 2
Cavg_pup gest
4-32
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Reference,
Confidence
Strain/
Species/Sex
POD Type,
Model
POD
POD Internal
Dose/Internal
Dose Metric"
PODhed
(mg/kg/day)
Cardiovascular Effects (Serum Lipids)
Increased Total
Dongetal. (2019, 5080195)
Human, male and female;
BMDL5RD,
2.29 ng/mL
2.75 xlO-7
Cholesterol
Medium
excluding individuals
prescribed cholesterol
medication
Hybrid
Steenland et al. (2009, 1291109)
Human, male and female;
BMDL5RD,
4.25 ng/mL
5.10xl0-7
Medium
excluding individuals
prescribed cholesterol
medication
Hybrid
Lin et al. (2019,5187597)
Human, male and female
BMDL5RD,
5.28 ng/mL
6.34xl0-7
Medium
Hybrid
Hepatic Effects
Elevated ALT
Galloetal. (2012, 1276142)
Medium
Human, female
BMDL5RD,
Hybrid
17.9 ng/mL
2.15x10^
Darrow et al. (2016, 3749173)
Human, female
BMDL5RD,
66.0 ng/mL
7.92x10^
Medium
Hybrid
Nian et al. (2019, 5080307)
Human, female
BMDL5RD,
3.76 ng/mL
4.51x10-7
Medium
Hybrid
Increased Focal
Loveless et al. (2008, 988599)
Crl:CD-l(ICR)BR mice,
BMDLiord,
10.0 mg/L
1.20xl0-3
Necrosis
Medium
male
Dichotomous
Hill
Clast7
Increased Individual
Loveless et al. (2008, 988599)
Crl:CD-l(ICR)BR mice,
BMDLiord,
36.0 mg/L
4.32xl0-3
Cell Necrosis
Medium
male
Probit
Clast7
Increased Necrosis
NTP (2020, 7330145)
Sprague-Dawley rats, males;
BMDLiord,
26.9 mg/L
3.23X10-3
High
perinatal and postweaning
Multistage
Degree 1
Cavg_pup total
Notes: ALT = alanine aminotransferase; AUC = area under the curve; BMDLo.5sd = lower bound on the dose level corresponding to the 95% lower confidence limit for a change in
the mean response equal to 0.5 standard deviation from the control mean; BMDLsrd = lower bound on the dose level corresponding to the 95% lower confidence limit for a 5%
change in response; BMDLiord = lower bound on the dose level corresponding to the 95% lower confidence limit for a 10% change in response; Ciast7 = blood concentration over
4-33
-------
DRAFT FOR PUBLIC COMMENT
March 2023
the last 7 days; Fi = first generation; IgM = immunoglobulin M; NOAEL = no-observed-adverse-effect level; NTP = National Toxicology Program; PODhed = point-of-departure
human equivalence dose; RiD = reference dose; SRBC = sheep red blood cell.
a see PFOA Appendix for additional details on BMD modeling.
b Supported by Grandjean et al. (2012, 1248827), Grandjean et al. (2017, 3858518), and Grandjean et al. (2017, 4239492).
c Maternal serum concentrations were taken either in the third trimester (32 weeks) or about two weeks after the expected term date.
dNo models provided adequate fit; therefore, aNOAEL/LOAEL approach was selected.
4-34
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.4.
Increased ALT in individuals aged 18 and older {Gallo, 2012,1276142; Darrow, 2016,
3749173; Nian,2019, 5080307}
The POD for increased ALT in adults was derived by quantifying a benchmark dose using a
hybrid modeling approach (see PFOA Appendix) on the measured {Gallo, 2012, 1276142; Nian,
2019, 5080307} or modeled (Darrow, 2016, 3749173} PFOA serum concentrations collected
from adults aged 18 years and older, which provided an internal serum concentration POD in
mg/L. The internal serum POD was converted to an external dose (PODhed), in mg/kg/day.
Specifically, the PODhed was calculated as the external dose that would result in a steady-state
serum concentration equal to the internal serum POD. This calculation is simply the POD
multiplied by the selected human clearance value (0.120 mL/kg/day; calculated from half-life
and volume of distribution; CI = Vd * ln(2)/ti/2)).
Focal Necrosis, Crl:CD-l(ICR)BR mice, male, Ciast7 {Loveless, 2008, 7330145}
Increased incidence of focal necrosis of the liver was observed in male ICR mice. Dichotomous
models were used to fit dose-response data. A BMR of 10% extra risk was chosen. Ciast7 was
selected for this model rather than alternate metrics such as Cmax because the average blood
concentration is expected to better correlate with an accumulation of focal necrosis in the liver.
The BMDS produced a BMDL in mg/L. A PODhed was calculated as the external dose that
would result in a steady-state serum concentration in humans equal to the POD from the animal
analysis. This calculation is simply the POD multiplied by the selected human clearance value
(0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Vd * ln(2)/ti/2)).
Individual Cell Necrosis, Crl:CD-l(ICR)BR mice, male, Ciast7 {Loveless, 2008, 7330145}
Increased incidence of individual cell necrosis of the liver was observed in male ICR mice.
Dichotomous models were used to fit dose-response data. A BMR of 10% extra risk was chosen.
Ciast7 was selected for this model rather than alternate metrics such as Cmax because the average
blood concentration is expected to better correlate with an accumulation of individual cell
necrosis of the liver. The BMDS produced a BMDL in mg/L. A PODhed was calculated as the
external dose that would result in a steady-state serum concentration in humans equal to the POD
from the animal analysis. This calculation is simply the POD multiplied by the selected human
clearance value (0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Yd
Necrosis, Sprague-Dawley rats, males, perinatal and postweaning, male rats,
C avg pup total {NTP, 2020, 7330145}
Increased incidence of necrosis of the liver was observed in adult male Sprague-Dawley rats.
Dichotomous models were used to fit dose-response data. A BMR of 10% extra risk was chosen.
The Cavg pup total was selected for this model rather than alternate metrics such as Cmax because the
average blood concentration is expected to better correlate with an accumulation of necrosis in
the liver. The BMDS produced a BMDL in mg/L. A PODhed was calculated as the external dose
that would result in a steady-state serum concentration in humans equal to the POD from the
animal analysis. This calculation is simply the POD multiplied by the selected human clearance
* ln(2)/ti/2)).
4-35
-------
DRAFT FOR PUBLIC COMMENT
March 2023
value (0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Vd *
ln(2)/ti/2)).
4.1.4.2 Immune Effects
Decreased Diphtheria and Tetanus antibody response in vaccinated children at age 7
{Budtz-Jorgensen, 2018, 5083631}
The POD for decreased antibody production at age 7 was derived by quantifying a benchmark
dose (see PFOA Appendix) on the measured PFOA serum concentrations at age 5, which
provided an internal serum concentration POD in mg/L. The internal serum POD was converted
to an external dose (PODhed), in mg/kg/day, using the updated Verner model (described in
Section 4.1.3.1.3). For this, the model was run starting at the birth of the mother, with constant
exposure relative to bodyweight. Pregnancy began at 24.25 years maternal age and birth
occurred at 25 years maternal age. The initial concentration in the child is governed by the
observed ratio between maternal serum and cord blood at delivery. Then the model is run
through the 1 year breastfeeding period, where the exposure to the child is only through
lactation, which is much greater than the exposure to the mother. After 1 year, the exposure to
the child, relative to bodyweight, is set to the same value as the mother. The model provides
predictions up to a child age of 5 years, when the serum concentrations used to determine the
POD were collected, and reverse dosimetry was used to determine the PODhed that results in the
POD serum concentration. Because of different growth curves used for male and female children
used in the model, the model predicted slightly different (less than 5%) serum concentrations for
them. The lower HED was then selected as it was the most health protective.
Decreased Diphtheria and Tetanus antibody response in vaccinated children at age 5
{Budtz-Jorgensen, 2018, 5083631}
The POD for decreased antibody production at age 5 was derived by quantifying a benchmark
dose (see PFOA Appendix) on the measured PFOA serum concentrations collected from the
mother either in the third trimester (32 weeks) or about two weeks after the expected term date,
which provided an internal serum concentration POD in mg/L. The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.1.3). For this, the model was run similarly to the endpoint based on
antibodies at age 7, except that the model was only run until the maternal age of 25 years, when
delivery occurs in the model. As the POD was based on maternal serum concentrations taken
before and after birth, the time of delivery was chosen as an average of the two. Reverse
dosimetry was performed on model predicted maternal serum concentration at that time to
calculate the PODhed. This metric is independent of the sex of the child in the model.
Decreased Diphtheria and Tetanus antibody response in vaccinated children at ages 7-12
{Timmerman, 2021, 9416315}
The POD for decreased antibody production in children aged 7-12 was derived by quantifying a
benchmark dose (see PFOA Appendix) on the measured PFOA serum concentrations at ages 7-
12, which provided an internal serum concentration POD in mg/L. The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.1.3). For this, the model was run similarly to the endpoint based on
antibodies at age 7 {Budtz-Jorgensen, 2018, 5083631}, but the model was run until the median
4-36
-------
DRAFT FOR PUBLIC COMMENT
March 2023
age of this cohort at blood collection, 9.9 years. Reverse dosimetry was used to calculate the
PODHEDthat resulted in a serum level equal to the POD at that age. Because of different growth
curves used for male and female children, the model predicted slightly different serum
concentrations for them. The lower HED was then selected as it was the most health protective.
Decreased IgM response to SRBC, Mouse, Female, Studies 1 and 2, Ciast7 {Dewitt, 2008,
1290826}
Decreased mean response of SRBC-specific IgM antibody titers was observed in female
C57BL/6N mice (Studies 1 and 2). Using the Wambaugh et al. (2013, 2850932) model, daily
exposure to PFOA in the drinking water was simulated for 15 days using female C57BL/6 mice
parameters. An average concentration over the last 7 days of treatment (Ciast7) was calculated as
the internal dose metric for each dose group. Continuous models were used to fit dose-response
data. A BMR of a change in the mean equal to 1 SD from the control mean was chosen per
EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433}. The average
concentration over the final week of study (Ciast7) was selected for this model rather than
alternate metrics such as Cmax because the average blood concentration is expected to better
correlate with an accumulation of effects leading to decreased response of SRBC-specific IgM
antibody titers. The BMDS produced a BMDL in mg/L. A PODhed was calculated as the
external dose that would result in a steady-state serum concentration in humans equal to the POD
from the animal analysis. This calculation is simply the POD multiplied by the selected human
clearance value (0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Vd
* ln(2)/ti/2)).
Decreased IgM response to SRBC, Mouse, Male, Ciast7 {Loveless, 2008, 988599}
Decreased mean response of IgM serum titer was observed in male Crl:CD-l(ICR)BR mice.
Using the Wambaugh et al. (2013, 2850932) model, daily oral gavage exposure to PFOA was
simulated for 29 days using male CD1 mice parameters. An average concentration over the last 7
days of treatment (Ciast7) was calculated as the internal dose metric for each dose group.
Continuous models were used to fit dose-response data. A BMR of a change in the mean equal to
1 SD from the control mean was chosen per EPA's Benchmark Dose Technical Guidance {U.S.
EPA, 2012, 1239433}. Ciast7 was selected for this model rather than alternate metrics such as
Cmax because the average blood concentration is expected to better correlate with an
accumulation of effect resulting in decreased mean response of IgM serum titer. The BMDS
produced a BMDL in mg/L. A PODhed was calculated as the external dose that would result in a
steady-state serum concentration in humans equal to the POD from the animal analysis. This
calculation is simply the POD multiplied by the selected human clearance value (0.120
mL/kg/day; calculated from half-life and volume of distribution; CI = Vd * ln(2)/ti/2)).
4.1.4.3 Cardiovascular Effects
Increased total cholesterol in adults aged 20-80, excluding individuals prescribed
cholesterol medication {Dong, 2019, 5080195}
The POD for increased TC in adults was derived by quantifying a benchmark dose using a
hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum concentrations
collected from adults aged 20-80 years not prescribed cholesterol medication through the
NHANES, which provided an internal serum concentration POD in mg/L. The internal serum
4-37
-------
DRAFT FOR PUBLIC COMMENT
March 2023
POD was converted to an external dose (PODhed), in mg/kg/day. Specifically, the PODhed was
calculated as the external dose that would result in a steady-state serum concentration equal to
the internal serum POD. This calculation is simply the POD multiplied by the selected human
clearance value (0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Vd
* ln(2)/ti/2)).
Increased total cholesterol in individuals aged 18 and older, excluding individuals
prescribed cholesterol medication {Steenland, 2009,1291109}
The POD for increased TC in adults was derived by quantifying a benchmark dose using a
hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum concentrations
collected from adults aged 18 years and older not prescribed cholesterol medication from the C8
study population, which provided an internal serum concentration POD in mg/L. The internal
serum POD was converted to an external dose (PODhed), in mg/kg/day. Specifically, the
PODhed was calculated as the external dose that would result in a steady-state serum
concentration equal to the internal serum POD. This calculation is simply the POD multiplied by
the selected human clearance value (0.120 mL/kg/day; calculated from half-life and volume of
distribution; CI = Vd * ln(2)/ti/2)).
Increased total cholesterol in individuals aged 25 and older {Lin, 2019, 5187597}
The POD for increased TC in adults was derived by quantifying a benchmark dose using a
hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum concentrations
collected in adults 25 years and older who were at high risk of developing type 2 diabetes and
hyperlipidemia from the Diabetes Prevention Program (DPP) and Outcomes Study (DPPOS),
which provided an internal serum concentration POD in mg/L. The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day. Specifically, the PODhed was calculated
as the external dose that would result in a steady-state serum concentration equal to the internal
serum POD. This calculation is simply the POD multiplied by the selected human clearance
value (0.120 mL/kg/day; calculated from half-life and volume of distribution; CI = Vd *
ln(2)/ti/2)).
4.1.4.4 Developmental Effects
Decreased birthweight using the mother's serum PFOA concentration collected in third
trimester {Chu, 2020, 6315711}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum
concentrations collected from the mother in the third trimester (blood was collected within
3 days after delivery), which provided an internal serum concentration POD in mg/L. The
internal serum POD was converted to an external dose (PODhed), in mg/kg/day, using the
updated Verner model (described in Section 4.1.3.1.3). This calculation was performed similarly
for each of the birthweight endpoints. The model was run starting at the birth of the mother, with
constant exposure relative to bodyweight. Pregnancy began at 24.25 years maternal age. The
model was stopped at a time to match the median gestational age of the cohort at sample time for
samples taken during pregnancy, or at delivery (25 years maternal age) in the case of maternal
samples at delivery or samples of cord blood. Reverse dosimetry was performed to calculate the
PODhed resulting in serum levels matching the POD at the model end time. For this study,
4-38
-------
DRAFT FOR PUBLIC COMMENT
March 2023
maternal blood was drawn within a few days of the birth of the child, so delivery was chosen as
the model end time. This metric is independent of the sex of the child in the model.
Decreased birthweight using the serum PFOA concentrations collected from umbilical cord
samples {Govarts, 2016, 3230364}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum
concentrations collected from an umbilical cord sample, which provided an internal serum
concentration POD in mg/L. The internal serum POD was converted to an external dose
(PODhed), in mg/kg/day, using the updated Verner model (described in Section 4.1.3.1.3). This
was performed as described for the Chu et al. (2020, 6315711) study. The model was stopped at
delivery and reverse dosimetry was performed to calculate the PODhed that resulted in the POD
serum level in cord serum. This metric is independent of the sex of the child in the model.
Decreased birthweight using the mother's serum PFOA concentration collected in in first
trimester {Sagiv, 2018, 4238410}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum
concentrations collected from the mother in the first trimester (median gestational age of 9
weeks), which provided an internal serum concentration POD in mg/L. The internal serum POD
was converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.1.3). This was performed as described for the Chu et al. (2020,
6315711) study. The model was stopped at the median gestational age of this cohort, 9 weeks.
The time after conception was calculated as the fraction of pregnancy competed after 9 weeks
(9/39 weeks), times the pregnancy duration of 0.75 year. Reverse dosimetry was performed to
calculate the PODhed that resulted in the POD in maternal serum at that time. This metric is
independent of the sex of the child in the model.
Decreased birthweight using the mother's serum PFOA concentration collected in second
and third trimesters {Starling, 2017, 3858473}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum
concentrations collected from the mother in the trimesters 2 and 3 (median gestational age of 27
weeks), which provided an internal serum concentration POD in mg/L. The internal serum POD
was converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.1.3). This was performed as described for the Chu et al. (2020,
6315711) study. The model was stopped at the median gestational age of this cohort, 27 weeks.
The time after conception was calculated as the fraction of pregnancy competed after 27 weeks
(27/39 weeks), times the pregnancy duration of 0.75 year. Reverse dosimetry was performed to
calculate the PODhed that resulted in the POD in maternal serum at that time. This metric is
independent of the sex of the child in the model.
Decreased birthweight using the mother's serum PFOA concentration collected in first and
second trimesters {Wikstrom, 2020, 6311677}
4-39
-------
DRAFT FOR PUBLIC COMMENT
March 2023
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see PFOA Appendix) on the measured PFOA serum
concentrations collected from the mother in the trimesters 1 and 2 (median gestational age of 10
weeks), which provided an internal serum concentration POD in mg/L. The internal serum POD
was converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.1.3). This was performed as described for the Chu et al. (2020,
6315711) study. The model was stopped at the median gestational age of this cohort, 10 weeks.
The time after conception was calculated as the fraction of pregnancy completed at 10 weeks
(10/39 weeks), times the pregnancy duration of 0.75 year. Reverse dosimetry was performed to
calculate the PODhed that resulted in the POD in maternal serum at that time. This metric is
independent of the sex of the child in the model.
Decreased Offspring Survival, Kunming Mice, Fi males and females, Cavg_pup_gest_iact {Song,
2018, 5079725}
Decreased mean response of number of offspring survival was observed in Fl male and female
Kunming mice. Continuous models were used to fit dose-response data. BMR of a change in the
mean equal to 0.1 and 0.5 standard deviations from the control mean were chosen. The
Cavg,pup,gest, Cavg,pup,lact, Cavg,pup,gest,lact, Cmax,pup,gest, and Cmax,pup,lact were
considered because prenatal loss could be a result of exposure during a sensitive window of
development where a Cmax metric is expected to better correlate with the effect or an
accumulation of exposure and an average concentration metric is expected to better correlate
with the effect and this could occur during the gestation or lactation lifestages. The
Cavg,pup,gest,lact was selected for this model since an average concentration metric is expected
to better correlate with the effect and this could occur during the gestation or lactation lifestages.
A 0.5 standard deviation BMR was ultimately selected. The BMDS produced a BMDL in mg/L.
The internal serum POD, based on the predicted average serum concentration in the pup during
gestation, was converted to an external dose (PODhed), in mg/kg/day, using the updated Verner
model (described in Section 4.1.3.1.3). For this, the model was run starting at the birth of the
mother, with constant exposure relative to bodyweight. Pregnancy began at 24.25 years maternal
age and birth occurred at 25 years maternal age. The model was run up to the birth of the child.
The average serum concentration in the infant during gestation was determined for this scenario
and reverse dosimetry was used to calculate the exposure that results in the same value as the
POD. Before birth, model predictions for male and female children are equivalent.
Decreased Fetal Body Weight, Kunming Mice, Fi males and females, Cavgjmp_gest {Li, 2018,
5084746}
Decreased mean response of fetal body weight was observed in Fi male and female Kunming
mice. Continuous models were used to fit dose-response data. A BMR of 5% extra risk was
selected as described in Section 4.1.2, and a change in the mean equal to 0.5 standard deviations
from the control mean was provided for comparison purposes (See PFOA Appendix). The
Cavg,pup,gest was selected for this model rather than alternate metrics such as Cmax because the
average concentration normalized per day during gestation is expected to better correlate with an
accumulation of effect resulting in decreased fetal body weight. The BMDS did not produce a
model with adequate fit, so a NOAEL approach was taken. The internal serum POD, based on
the predicted average serum concentration in the pup during gestation, was converted to an
external dose (PODhed), in mg/kg/day, using the updated Verner model (described in Section
4-40
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.3.1.3). For this, the model was run starting at the birth of the mother, with constant exposure
relative to bodyweight. Pregnancy began at 24.25 years maternal age and birth occurred at
25 years maternal age. The model was run up to the birth of the child. The average serum
concentration in the infant during gestation was determined for this scenario and reverse
dosimetry was used to calculate the exposure that results in the same value as the POD. Before
birth, model predictions for male and female children are equivalent.
Delayed Time to Eye Opening, CD-I Mice, Fi males and females, Cavgjmp_gest {Lau, 2006,
1276159}
Decreased mean response of time to eye opening was observed in Fl male and female CD-I
mice. Continuous models were used to fit dose-response data. BMR of a change in the mean
equal to 1 standard deviations from the control mean was selected, and a BMR of a change in the
mean equal to 0.5 standard deviations from the control mean is provided for comparison
purposes (See PFOA Appendix). The average concentration normalized per day during gestation,
(Cavg,pup,gest), average concentration normalized per day during lactation (Cavg,pup,lact),
maximum fetal concentration during gestation (Cmax,pup,gest), and maximum pup
concentration during lactation (Cmax,pup,lact) were all considered because time to eye opening
could be a result of exposure during a sensitive window of development where a Cmax metric is
expected to better correlate with the effect or an accumulation of exposure where an average
concentration metric is expected to better correlate with the effect and time to eye opening could
be due to exposure during the gestation or lactation lifestages. The Cavg,pup,gest was selected
for this model. The BMDS produced a BMDL in mg/L. The internal serum POD, based on the
predicted average serum concentration in the pup during gestation and lactation, was converted
to an external dose (PODhed), in mg/kg/day, using the updated Verner model (described in
Section 4.1.3.1.3). For this, the model was run starting at the birth of the mother, with constant
exposure relative to bodyweight. Pregnancy began at 24.25 years maternal age and birth
occurred at 25 years maternal age. The initial concentration in the child was governed by the
observed ratio between maternal serum and cord blood at delivery. Then the model was run
through the 1 year breastfeeding period. The average serum concentration in the infant through
gestation and lactation was determined for this scenario and reverse dosimetry was used to
calculate the exposure that results in the same value as the POD. Because of different growth
curves used for male and female children, the model predicted slightly different serum
concentrations for them. The lower HED was selected to be more health protective.
4.1.5 Derivation of Candidate Chronic Oral Reference Doses
(RfDs)
Though multiple PODheds were derived for multiple health systems from both epidemiological
and animal toxicological studies, EPA selected the PODheds with the greatest strength of
evidence and the lowest risk of bias represented by high or medium confidence studies for
candidate RfD derivation, as described below. As presented in Table 4-9, epidemiological data
representing the four prioritized health outcomes represented the most sensitive effects after
PFOA exposure in the lower dose range. Four endpoints from epidemiological studies
representing the four health outcomes were considered for candidate RfD derivation. These
endpoints are decreased antibody response, low birth weight, increased total cholesterol, and
4-41
-------
DRAFT FOR PUBLIC COMMENT
March 2023
elevated ALT. As described in the subsections below, EPA further evaluated studies within each
endpoint to determine those most suitable for candidate RfD derivation.
EPA also further evaluated animal toxicological studies to determine which were the most
suitable for candidate RfD derivation. Factors considered included study confidence (i.e., high
confidence studies were prioritized over medium confidence studies), amenability to benchmark
dose modeling, and health effects observed after exposure in the lower dose range among the
animal toxicological studies. As described in the subsections below, this examination led to the
exclusion a number of studies considered for POD derivation, including both epidemiological
and animal toxicological studies, from further consideration.
4.1.5.1 Hepatic Effects
Three medium confidence epidemiological studies were carried forward for candidate RfD
determination {Gallo, 2012, 1276142; Darrow, 2016, 3749173; Nian, 2019, 5080307}. EPA
considered all three studies as they represented the low-dose range of effects across hepatic
endpoints and provided data from relatively large populations, including U.S. populations.
One high confidence animal toxicological study was carried forward for candidate RfD
determination {NTP, 2020, 7330145}. NTP (2020, 7330145) was prioritized for candidate RfD
development because it was determined to be a high confidence study and it used a chronic
exposure duration that encompassed sensitive periods of development whereas Loveless et al.
(2008, 988599) was a medium confidence study that used a short-term (28 day) exposure
duration and predated current criteria for hepatic histopathological assessment of cell death
{Elmore, 2016, 10671182}.
4.1.5.2 Immune Effects
Two medium confidence epidemiological studies were carried forward for candidate RfD
determination {Budtz-j0rgensen, 2018, 5083631; Timmerman, 2021, 9416315}. EPA considered
both studies as they both represented the low-dose range of effects across immunological
endpoints and provided data regarding sensitive populations (i.e., children). Although EPA
derived PODheds for two time points reported by Budtz-j0rgensen and Grandjean (2018,
5083631) (i.e., PFOA serum concentrations at age 5 and antibody concentrations at age 7; PFOA
serum concentrations in the mother during the third trimester or approximately 2 weeks after the
expected term date and antibody concentrations at age 5), EPA did not carry forward PODheds
based on serum PFOA concentrations measured in the mother for candidate RfD derivation
because of concerns surrounding bias due to pregnancy-related hemodynamic effects.
One medium confidence animal toxicological study was carried forward for candidate RfD
determination {Dewitt, 2008, 1290826}. Study quality evaluations and further consideration did
not identify notable characteristics distinguishing the two candidate studies {Dewitt, 2008,
1290826; Loveless, 2008, 988599}, but because the PODheds of reduced IgM response in
rodents represented effects at the highest dose range of responses and because the observed
effects were from medium confidence less-than-chronic studies, EPA selected the most health
protective PODHEDfor candidate RfD derivation.
4-42
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.5.3 Cardiovascular Effects
Two medium confidence epidemiological studies were carried forward for candidate RfD
determination {Dong, 2019, 5080195; Steenland, 2009, 1291109}. Of the three studies for which
PODheds were derived, Dong et al. (2019, 5080195) and Steenland et al. (2009, 1291109)
exclude individuals who were prescribed cholesterol medication, minimizing concerns
surrounding confounding due to the medical intervention altering serum total cholesterol levels.
Therefore, these two studies were considered further for candidate RfD derivation.
4.1.5.4 Developmental Effects
Two high confidence epidemiological studies were carried forward for candidate RfD
determination for the endpoint of low birth weight {Sagiv, 2018, 4238410; Wikstrom, 2020,
6311677}. Of the six epidemiological studies for which PODheds were derived, Sagiv et al.
(2018, 4238410) and Wikstrom et al. (2020, 6311677) assessed maternal PFOA serum
concentrations primarily or exclusively in the first trimester, minimizing concerns surrounding
bias due to pregnancy-related hemodynamic effects. Therefore, these two studies were
considered further for candidate RfD derivation.
Two medium confidence animal toxicological studies were carried forward for candidate RfD
determination {Lau, 2006, 1276159; Song, 2018, 5079725}. These two studies were amenable to
benchmark dose modeling, unlike Li et al. (2018, 5084746), which had aNOAEL as the basis of
the PODhed. As the endpoints reported by Lau et al. (2006, 1276159) and Song et al. (2018,
5079725) encompass sensitive populations (i.e., fetuses and pups), these two studies were
considered further for candidate RfD derivation.
4.1.5.5 Application of Uncertainty Factors (UFs)
To calculate the candidate RfD values, EPA applied UFs to the PODheds derived from selected
epidemiological and animal toxicological studies (Table 4-9 and Table 4-10). UFs were applied
according to methods described in EPA's Review of the Reference Dose and Reference
Concentration Processes {U.S. EPA, 2002, 88824}.
Table 4-9. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
from Epidemiological Studies {U.S. EPA, 2002, 88824}
UF
Value
Justification
UFa
1
A UFa of 1 is applied to effects observed in epidemiological studies as the study
population is humans.
UFh
10
A UFh of 10 is applied when information is not available relative to variability in
the human population.
UFs
1
A UFS of 1 is applied when effects are observed in adult human populations that
are assumed to have been exposed to a contaminant over the course of many years.
A UFS of 1 is applied for developmental effects because the developmental period
is recognized as a susceptible life stage when exposure during a time window of
development is more relevant to the induction of developmental effects than
lifetime exposure {U.S. EPA, 1991, 732120}.
UFl
1
A UFl of 1 is applied for LOAEL to NOAEL extrapolation when the POD is a
BMDL or a NOAEL.
4-43
-------
DRAFT FOR PUBLIC COMMENT
March 2023
UF
Value
Justification
UFd
1
A UFd of 1 is applied when the database for a contaminant contains a multitude of
studies of adequate quality that encompass a comprehensive array of endpoints in
various life stages and populations and allow for a complete characterization of the
contaminant's toxicity.
UFC
10
Composite UFC = UFA x UFH x UFS x UFL x UFD
Notes'. BMDL = benchmark dose level; LOAEL = lowest-observed-adverse-effect level; NOAEL = no-observed-adverse-effect
level; POD = point of departure; LTFa = interspecies uncertainty factor; UFd = database uncertainty factor; UFh = intraspecies
uncertainty factor; UFl = LOAEL-to-NOAEL extrapolation uncertainty factor; UFs = uncertainty factor for extrapolation from a
subchronic to a chronic exposure duration; UFc= composite UF.
An interspecies UF (UFa) of 1 was applied to PODheds derived from epidemiological studies
because the dose response information from these studies is directly relevant to humans. There is
no need to account for uncertainty in extrapolating from laboratory animals to humans.
An intraspecies UF (UFh) of 10 was applied to PODheds derived from epidemiological studies to
account for variability in the responses within the human populations because of both intrinsic
(toxicokinetic, toxicodynamic, genetic, life stage, and health status) and extrinsic (lifestyle)
factors that can influence the response to dose. No information to support a UFh other than 10
was available to quantitatively characterize interindividual and age-related variability in the
toxicokinetics or toxicodynamics.
A LOAEL-to-NOAEL extrapolation UF (UFl) of 1 was applied to PODheds derived from
epidemiological studies because a BMDL is used as the basis for the PODhed derivation. When
the POD type is a BMDL, the current approach is to address this factor as one of the
considerations in selecting a BMR for BMD modeling.
A UF for extrapolation from a subchronic to a chronic exposure duration (UFs) of 1 was applied
to PODheds derived from epidemiological studies. A UFS of 1 was applied to the hepatic and
cardiovascular endpoints because the effects were observed in adult populations that were
assumed to have been exposed to PFOA over the course of many years. A UFS of 1 was applied
to the developmental endpoints because the developmental period is recognized as a susceptible
life stage when exposure during a time window of development is more relevant to the induction
of developmental effects than lifetime exposure {U.S. EPA, 1991, 732120}. A UFs of 1 was also
applied to the immune endpoints in children because the developing immune system is
recognized as a susceptible lifestage; therefore, exposure during this time window can be
considered more relevant than lifetime exposure {U.S. EPA, 1991, 732120}. According to the
WHO/International Programme on Chemical Safety (IPCS) Immunotoxicity Guidance for Risk
Assessment, developmental immunotoxicity encompasses the prenatal, neonatal, juvenile and
adolescent life stages and should be viewed differently from the immune system of adults from a
risk assessment perspective {IPCS, 2012, 1249755}.
A database UF (UFd) of 1 was applied to account for deficiencies in the database for PFOA. In
animals, comprehensive oral short term, subchronic, and chronic studies in three species and
several strains of laboratory animals have been conducted and published in the peer reviewed
literature. Additionally, there are several neurotoxicity studies (including developmental
neurotoxicity) and several reproductive (including one- and two-generation reproductive toxicity
studies) and developmental toxicity studies including assessment of immune effects following
developmental exposure. Moreover, there is a robust epidemiological database which was used
4-44
-------
DRAFT FOR PUBLIC COMMENT
March 2023
quantitatively in this assessment. Typically, the specific study types lacking in a chemical's
database that influence the value of the UFd to the greatest degree are developmental toxicity and
multigenerational reproductive toxicity studies. Effects identified in developmental and
multigenerational reproductive toxicity studies have been quantitatively considered in this
assessment.
The total UF that was applied to candidate RfDs derived from all of the epidemiological studies
were the same value (UFc = 10) (Table 4-9).
Increased uncertainty is associated with the use of animal toxicological studies as the basis of
candidate RfDs. The UFs applied to animal toxicological studies considered for candidate RfD
derivation were either one of two values, depending on the duration of exposure (i.e., chronic vs.
subchronic) or exposure window (e.g., gestational) (Table 4-10).
Table 4-10. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
from Animal Toxicological Studies {U.S. EPA, 2002, 88824}
UF
Value
Justification
UFa
3
A UFa of 3 is applied for the extrapolation from animal models to humans due to
the implementation of a PK model for animal PODhed derivation.
UFh
10
A UFh of 10 is applied when information is not available relative to variability in
the human population.
UFs
1 or 10
A UFS of 10 is applied for the extrapolation of subchronic to chronic exposure
durations. A UFS of 1 is applied to studies with chronic exposure durations or that
encompass a developmental period (i.e., gestation). The developmental period is
recognized as a susceptible life stage when exposure during a time window of
development is more relevant to the induction of developmental effects than
lifetime exposure {U.S. EPA, 1991, 732120}.
UFl
1
A UFl of 1 is applied for LOAEL to NOAEL extrapolation when the POD is a
BMDL or a NOAEL.
UFd
1
A UFd of 1 is applied when the database for a contaminant contains a multitude of
studies of adequate quality that encompass a comprehensive array of endpoints in
various life stages and populations and allow for a complete characterization of the
contaminant's toxicity.
UFC
30 or 300
Composite UFC = UFA x UFH x UFS x UFL x UFD
Notes: BMDL = benchmark dose level; LOAEL = lowest-observed-adverse-effect level; NOAEL = no-observed-adverse-effect
level; POD = point of departure; UFa = interspecies uncertainty factor; UFd = database uncertainty factor; UFh = intraspecies
uncertainty factor; UFl = LOAEL-to-NOAEL extrapolation uncertainty factor; UFs = uncertainty factor for extrapolation from a
subchronic to a chronic exposure duration; UFc = total uncertainty factors.
A UFa of 3 was applied to PODheds derived from animal toxicological studies to account for
uncertainty in extrapolating from laboratory animals to humans (i.e., interspecies variability).
The 3-fold factor is applied to account for toxicodynamic differences between the animals and
humans. The HEDs were derived using a model that accounted for PK differences between
animals and humans.
A UFh of 10 was applied to PODheds derived from animal toxicological studies to account for
variability in the responses within human populations because of both intrinsic (toxicokinetic,
toxicodynamic, genetic, life stage, and health status) and extrinsic (lifestyle) factors can
4-45
-------
DRAFT FOR PUBLIC COMMENT
March 2023
influence the response to dose. No information to support a UFh other than 10 was available to
characterize interindividual and age-related variability in the toxicokinetics or toxicodynamics.
A UFl of 1 was applied to PODheds derived from animal toxicological studies because a BMDL
is used as the basis for the PODhed derivation. When the POD type is a BMDL, the current
approach is to address this factor as one of the considerations in selecting a BMR for BMD
modeling.
A UFs of 1 was applied to PODheds derived from chronic animal toxicological studies as well as
animal toxicological studies that encompass a developmental period (i.e., gestation). A UFS of 1
was applied to developmental endpoints because the developmental period is recognized as a
susceptible life stage when exposure during a time window of development is more relevant to
the induction of developmental effects than lifetime exposure {U.S. EPA, 1991, 732120}. A UFS
of 10 was applied to PODheds derived from studies that implemented a less-than-chronic
exposure duration because extrapolation is required to translate from a subchronic PODhed to a
chronic RfD.
A database UF (UFd) of 1 was applied to account for deficiencies in the database for PFOA. In
animals, comprehensive oral short term, subchronic, and chronic studies in three species and
several strains of laboratory animals have been conducted and published in the peer reviewed
literature. Additionally, there are several neurotoxicity studies (including developmental
neurotoxicity) and several reproductive (including one- and two-generation reproductive toxicity
studies) and developmental toxicity studies including assessment of immune effects following
developmental exposure. Moreover, there is a robust epidemiological database which was used
quantitatively in this assessment. Typically, the specific study types lacking in a chemical's
database that influence the value of the UFd to the greatest degree are developmental toxicity and
multigenerational reproductive toxicity studies. Effects identified in developmental and
multigenerational reproductive toxicity studies have been quantitatively considered in this
assessment.
4.1.5.6 Candidate RfDs
Table 4-11 shows the UFs applied to each candidate study to subsequently derive the candidate
RfDs.
4-46
-------
DRAFT FOR PUBLIC COMMENT March 2023
Table 4-11. Candidate Reference Doses (RfDs)
Endpoint
Study,
Confidence
Strain/Species/Sex
PODhed
(mg/kg/day)
UFa
UFh
UFs
UFl
UFd
UFc
Candidate RfDa
(mg/kg/day)
Immune Effects
Decreased serum anti-
Budtz-Jorgensen and
Human, male and
3.05 xKT7
1
10
1
1
1
10
3.05xl0-8 = 3xl0-8
tetanus antibody
Grandjean (2018,
female
concentration in
children
508363 l)b
Medium
Timmerman et al.
Human, male and
3.34xKT7
1
10
1
1
1
10
3.34xl0~8 = 3xl0-8
(2021, 9416315)
Medium
female
Decreased serum anti-
Budtz-Jorgensen and
Human, male and
2.92xKT7
1
10
1
1
1
10
2.92xl0-8 = 3xl0-8
diphtheria antibody
concentration in
children
Grandjean (2018,
508363 l)b
Medium
female
Timmerman et al.
Human, male and
2.20x10-'
1
10
1
1
1
10
2.20X10-8 = 2xl0-8
(2021, 9416315)
Medium
female
Decreased IgM
Dewitt et al. (2008,
Mouse, Female Study 1
2.18xKT3
3
10
10
1
1
300
7.27x10-® = 7xl0-6
response to SRBC
1290826)
Medium
Developmental Effects
Low Birth Weight
Sagiv et al. (2018,
4238410)
High
Human, male and
female
1.21 xKT6
1
10
1
1
1
10
1.21x10-' = lxlO-7
Wikstrom et al. (2020,
Human, male and
2.92xKT7
1
10
1
1
1
10
2.92x10-8 = 3x10-8
6311677)
High
female
Decreased Offspring
Survival
Song et al. (2018,
5079725)
Medium
Kunming Mice, Fi
males and females
6.40 xKT4
3
10
1
1
1
30
2.13xl0-5 = 2xl0-5
Delayed Time to Eye
Lau et al. (2006,
CD-I Mice, Fi males
1.71 xKT3
3
10
1
1
1
30
5.70xl0-5 = 6xl0-5
Opening
1276159)
Medium
and females
4-47
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Endpoint
Study,
Confidence
Strain/Species/Sex
PODhed
(mg/kg/day)
UFa
UFh
UFs
UFl
UFd
UFc
Candidate RfDa
(mg/kg/day)
Cardiovascular Effects
Increased Serum Total
Cholesterol
Dong et al. (2019,
5080195)
Medium
Human, male and
female, excluding
individuals prescribed
cholesterol medication
2.75 xKT7
1
10
1
1
1
10
2.75xl0-8 = 3xl0-8
Steenland et al. (2009,
1291109)
Medium
Human, male and
female, excluding
individuals prescribed
cholesterol medication
5.10xKT7
1
10
1
1
1
10
5.10X10-8 = 5xl0-8
Hepatic Effects
Increased Serum ALT
Galloetal. (2012,
1276142)
Medium
Human, female
2.15X10"6
1
10
1
1
1
10
2.15xl0-7 = 2xl0-7
Darrow et al. (2016,
3749173)
Medium
Human, female
7.92x10-®
1
10
1
1
1
10
7.92X10-7 = 8xl0-7
Nian et al. (2019,
5080307)
Medium
Human, female
4.51X107
1
10
1
1
1
10
4.51x10-8=5x10-8
Necrosis
NTP (2020, 7330145)
High
Sprague-Dawley rats,
perinatal and
postweaning, male
3.23X10-3
3
10
1
1
1
30
1.08xl0-4= lxlQ-4
Notes: ALT = alanine aminotransferase; NTP = National Toxicology Program; PODhed = point-of-departure human equivalence dose; RiD = reference dose; SRBC = sheep red
blood cells; UTa = interspecies uncertainty factor; UTh = intraspecies uncertainty factor; UTs = subchronic-to-chronic extrapolation uncertainty factor; UTl = extrapolation from a
LOAEL to NOAEL uncertainty factor; UTd = database uncertainty factor; UTc = composite uncertainty factor.
aRfDs were rounded to one significant figure.
b Supported by Grandjean et al. (2012,1248827), Grandjean et al. (2017, 3858518), and Grandjean et al. (2017, 4239492).
4-48
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.6 RfD Selection
As presented in Section 4.1.5 (Table 4-11), EPA derived and considered multiple candidate RfDs
across the four non-cancer health outcomes that EPA determined had the strongest weight of
evidence (i.e., immune, cardiovascular, hepatic, and developmental). EPA derived candidate
RfDs based on both epidemiological and animal toxicological studies. As depicted in Figure 4-4,
the candidate RfDs derived from epidemiological studies were all within 1 order of magnitude of
each other (10"7 to 10"8 mg/kg/day), regardless of endpoint, health outcome, or study population.
Candidate RfDs derived from animal toxicological studies were generally 2-3 orders of
magnitude higher than candidate RfDs derived from epidemiological studies. However, EPA
does not necessarily expect concordance between animal and epidemiological studies in terms of
the adverse effect(s) observed, as well as the dose level that elicits the adverse effect(s). For
example, EPA's Guidelines for Developmental Toxicity Risk Assessment states that "the fact that
every species may not react in the same way could be due to species-specific differences in
critical periods, differences in timing of exposure, metabolism, developmental patterns,
placentation, or mechanisms of action" {U.S. EPA, 1991, 732120}. Additionally, for
developmental effects, the guidance says that "the experimental animal data were generally
predictive of adverse developmental effects in humans, but in some cases, the administered dose
or exposure level required to achieve these adverse effects was much higher than the effective
dose in humans" {U.S. EPA, 1991, 732120}.
As shown in Table 4-11 and Figure 4-4, there is greater uncertainty associated with the use of
animal toxicological studies as the basis of RfDs than human epidemiological studies. Though
there are some uncertainties in the use of epidemiological studies for quantitative dose-response
analyses (see Section 6.1), human data eliminate the uncertainties associated with interspecies
extrapolation and the toxicokinetic differences between species which are major uncertainties
associated with the PFOA animal toxicological studies due to the half-life differences and sex-
specific toxicokinetic differences in rodent species. These uncertainties may explain why the
candidate RfDs derived from animal toxicological studies were several orders of magnitude
higher in value than the candidate RfDs derived from epidemiological studies. Moreover, the
human epidemiological studies also have greater relevance of exposure to human exposure
because they directly measure environmental or serum concentrations of PFOA. In accordance
with EPA's current best practices for systematic review, "animal studies provide supporting
evidence when adequate human studies are available, and they are considered to be the studies of
primary interest when adequate human studies are not available" {U.S. EPA, 2022, 10476098}.
For these reasons, EPA determined that candidate RfDs based on animal toxicological studies
would not be further considered for health outcome-specific RfD selection or overall RfD
selection. See Section 6.2 for further comparisons between toxicity values derived from
epidemiological and animal toxicological studies.
4-49
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Human Animal
Decreased
serum
anti-tetanus
antibody
concentration
in children
Decreased
serum
anti-diptheria
antibody
concentration
in children
Decreased
IgM response
to SRBC
Timmerman et al. (2021, 9416315);
Medium confidence
Budtz-Jorgensen and Grandjean
(2018, 5083631);
Medium confidence
Timmerman et al. (2021, 9416315);
Medium confidence
Budtz-Jorgensen and Grandjean
(2018, 5083631);
Medium confidence
Dewitt et al. (2008, 1290826);
Medium confidence
¦o
RfD
¦o
-o
-o
UF
PODHED
O
-o
Decreased
Birth Weight
Delayed Time
to Eye
Opening
Decreased
Offspring
Survival
Sagiv et al. (2018, 4238410);
High confidence
Wikstrom et al. (2019, 6311677);
High confidence
Lau et al. (2006, 1276159);
Medium confidence
Song et al. (2018, 5079725);
Medium confidence
-o
-o
-o
-o
Increased
Serum Total
Cholesterol
Dong et al. (2019, 5080195);
Medium confidence
Steenland et al. (2009, 1291109);
Medium confidence
¦o
¦o
Gallo et al. (2012, 1276142);
Medium confidence
•—o
Increased
Serum ALT
Darrow et al. (2016, 3749173);
Medium confidence
•—
-o
Nian et al. (2019, 5080307);
Medium confidence
•—o
Necrosis
NTP, 2020, 7330145;
High confidence
•—
O
1(
)-8 -10-7 10"®
105
10-4
o
CO
o
PFOA Concentration (mg/kg-d)
Figure 4-4. Comparison of Candidate RfDs Resulting from the Application of Uncertainty
Factors to PODheds Derived from Epidemiological and Animal Toxicological Studies
4-50
-------
DRAFT FOR PUBLIC COMMENT
March 2023
As described in the subsections below, EPA selected amongst the candidate RfDs to identify an
RfD representative of each of the four prioritized health outcomes (i.e., health outcome-specific
RfDs), as well as an overall RfD that is protective of the effects of PFOA on all health outcomes
and endpoints (Figure 4-5).
4.1.6.1 Health Outcome-Specific RfDs
Three medium confidence epidemiological studies were selected as candidates for RfD
derivation for the endpoint of elevated ALT {Gallo, 2012, 1276142; Darrow, 2016, 3749173;
Nian, 2019, 5080307}. The two largest studies of PFOA and ALT in adults, Gallo et al. (2012,
1276142) and Darrow et al. (2016, 3749173), were both conducted in over 30,000 adults from
the C8 Study. Gallo et al. (2012) reported measured serum ALT levels, unlike Darrow et al.
(2016) which reported a modeled regression coefficient as the response variable. Another
difference between the two studies is reflected in exposure assessment: Gallo et al. (2012,
1276142) includes measured PFOA serum concentrations, while Darrow et al. (2016, 3749173)
based PFOA exposure on modeled PFOA serum levels. The third study by Nian et al. (2019,
5080307) examined a large population of adults in Shenyang (one of the largest fluoropolymer
manufacturing centers in China) as part of the Isomers of C8 Health Project and observed
significant increases in lognormal ALT per each ln-unit increase in PFOA, as well significant
increases in ORs of elevated ALT. While both Nian et al. (2019, 5080307) and Gallo et al.
(2012, 1276142) provide measured PFOA serum concentrations and a measure of serum ALT
levels, the RfD for increased ALT from Gallo et al. (2012, 1276142) was ultimately selected for
the hepatic outcome as it was conducted in a community exposed predominately to PFOA,
whereas Nian et al. (2019, 5080307) was in a community exposed predominately to PFOS,
which reduces concerns about confounding from other PFAS. The resulting health outcome-
specific RfD is 2 x 10"7 mg/kg/day (Figure 4-5).
4.1.6.1.2 Immune Effects
Two medium confidence epidemiological studies were considered for RfD derivation for the
endpoint of decreased antibody production in response to various vaccinations in children
{Budtz-j0rgensen, 2018, 5083631; Timmerman, 2021, 9416315}. These candidate studies offer
a variety of PFOA exposure measures across various populations and various vaccinations.
Budtz-j0rgensen and Grandjean (2018, 5083631) investigated anti-tetanus and anti-diphtheria
responses in Faroese children aged 5-7 and Timmerman et al. (2021, 9416315) investigated anti-
tetanus and anti-diphtheria responses in Greenlandic children aged 7-12. Though the
Timmerman et al. (2021, 9416315) study is also a medium confidence study, the study by Budtz-
J0rgensen and Grandjean (2018, 5083631) has two additional features that strengthen the
confidence in this RfD: 1) the response reported by this study was more precise in that it reached
statistical significance, and 2) the analysis considered co-exposures of other PFAS. The RfDs for
anti-tetanus response in 7-year-old Faroese children and anti-diphtheria response in 7-year-old
Faroese children, both from Budtz-j0rgensen and Grandjean (2018, 5083631) were ultimately
selected for the immune outcome as they are the same value and have no distinguishing
characteristics that would facilitate selection of one over the other. The resulting health outcome-
specific RfD is 3 x 10"8 mg/kg/day (Figure 4-5). Note that all candidate RfDs based on
epidemiological studies for the immune outcome were within one order of magnitude of the
selected health outcome-specific RfD.
4-51
-------
DRAFT FOR PUBLIC COMMENT
March 2023
4.1.6.1.3 Cardiovascular Effects
Two medium confidence epidemiological studies were considered for RfD derivation for the
endpoint of increased TC {Dong, 2019, 5080195; Steenland, 2009, 1291109}. These candidate
studies offer a variety of PFOA exposure measures across various populations. Dong et al.
(2019, 5080195) investigated the NHANES population (2003-2014), while Steenland et al.
(2009, 1291109) investigated effects in a high-exposure community (the C8 Health Project study
population). Both of these studies excluded individuals prescribed cholesterol medication which
minimizes concerns of confounding due to medical intervention. The RfD for increased TC from
Dong et al. (2019, 5080195) was ultimately selected for the health outcome-specific RfD for
cardiovascular effects as there is marginally increased confidence in the modeling from this
study. Steenland et al. (2009, 1291109) presented analyses using both PFOA and TC as
categorical and continuous variables. The results using the natural log transformed TC and the
natural log transformed PFOA were stated to fit the data slightly better than the ones using
untransformed PFOA. However, the dramatically different changes in regression slopes between
the two analyses by Steenland et al. (2009, 1291109) resulting in extremely different PODs raise
concerns about the appropriateness of using this data. Therefore, the resulting health outcome-
specific RfD based on results from Dong et al. (2019, 5080195) is 3 x 10"8 mg/kg/day (Figure
4-5). Note that both candidate RfDs for the cardiovascular outcome were within one order of
magnitude of the selected health outcome-specific RfD.
4.1.6.1.4 Developmental Effects
Two high confidence epidemiological studies were considered for RfD derivation for the
endpoint of low birth weight {Sagiv, 2018, 4238410; Wikstrom, 2020, 6311677}. These
candidate studies assessed maternal PFOA serum concentrations primarily or exclusively in the
first trimester, minimizing concerns surrounding bias due to pregnancy-related hemodynamic
effects. Both were high confidence prospective cohort studies with many study strengths
including sufficient study sensitivity and sound methodological approaches, analysis, and design,
as well as no evidence of bias. The RfD for low birth weight from Wikstrom et al. (2020,
6311677) was selected as the basis for the health outcome-specific RfD for developmental
effects as it was the lowest and therefore most health protective candidate RfD from these two
studies. The resulting health outcome-specific RfD is 3 x 10"8 mg/kg/day (Figure 4-5). Note that
both candidate RfDs based on epidemiological studies for the developmental outcome were
within one order of magnitude of the selected health outcome-specific RfD.
4-52
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Immune
Anti-tetanus
antibody
response
Anti-diphtheria
antibody
response
Budtz-j0rgensen
and Grandjean -
(2018, 5083631)
sTimmerman et al.
(2021, 9416315)"
Budtz-J0rgensen
and Grandjean -
(2018, 5083631)
sTimmerman et al.
(2021, 9416315)"
3 x 10'8
3 x 10-8
3 x 108
2 x 10"8
3 x 10~e
Developmental
Low birth weight 1
Sagivetal.
(2018,4238410) '
Wikstrom et al.
(2019,6311677) '
1 x 10"7
3 x 10 s
3 x 10 8
3x 10 s
Cardiovascular
Increased total
cholesterol
Dong et al.
(2019, 5080195) '
Steenland et al.
(2009, 1291109)'
3 x 108
5x 10 s
3 x 10'8
Health Outcome
Gallo et al.
(2012, 1276142)'
Endpoint
Nian et al.
(2019, 5080307) '
Study
2 x 10'7
Hepatic
Elevated ALT
/
Darrow et al.
(2016, 3749173)
8 x 10"7
N 1
5 x 10"8
Candidate RfD
(mg/kg/day)
2 x 10"T
Health Outcome
Specific RfD
(mg/kg/day)
Overall RfD
(mg/kg/day)
Figure 4-5. Schematic depicting selection of the overall RfD for PFOA
RfD = reference dose.
Blue highlighted boxes indicate outcomes, endpoints, studies, candidate RlDs, and health outcome-specific RfDs that were
selected as the basis of the overall RfD.
4.1.6.2 Overall RfD
The available evidence indicates there are effects across immune, developmental, cardiovascular,
and hepatic organ systems at the same or approximately the same level of PFOA exposure. In
fact, candidate RfDs within the immune, developmental, and cardiovascular outcomes are the
same value (i.e., 3 x 10_f> mg/kg/day). Therefore, EPA has selected an overall RfD for PFOA of
3 x 10"8 mg/kg/day. The immune, developmental, and cardiovascular RfDs based on endpoints of
decreased anti-tetanus and anti-diphtheria antibody concentrations in children, low birth weight,
and increased total cholesterol, respectively, serve as co-critical effects for this RfD. Notably, the
RfD is protective of effects that may occur in sensitive populations (i.e., infants and children; see
Section 6.8), as well as hepatic effects in adults that may result from PFOA exposure. As two of
the co-critical effects identified for PFOA are developmental endpoints and can potentially result
4-53
-------
DRAFT FOR PUBLIC COMMENT
March 2023
from a short-term exposure during critical periods of development, EPA concludes that the
overall RfD for PFOA is applicable to both short-term and chronic risk assessment scenarios.
The critical studies that serve as the basis of the RfD are all medium or high confidence
epidemiological studies. The critical studies are supported by multiple other medium or high
confidence studies in both humans and animal models and have health outcome databases for
which EPA determined that either evidence indicates or evidence demonstrates that oral PFOA
exposure is associated with adverse effects. Additionally, the selected critical effects can lead to
clinical outcomes in a sensitive lifestage (children) and/or yield the lowest PODHEDand
candidate RfDs and therefore, is expected to be protective of all other health effects in humans.
4.2 Cancer
4.2.1 Animal Toxicological Studies
4.2.1.1 Study and End point Selection
Three chronic studies are available that investigate the relationship between dietary PFOA
exposure and carcinogenicity in male and female rats {Butenhoff, 2012, 2919192; NTP, 2020,
7330145; Biegel, 2001, 673581}. Biegel et al. (2001, 673581) was not considered for dose-
response modeling because it is a single-dose study. Butenhoff et al. (2012, 2919192) and NTP
(2020, 7330145) are medium and high confidence multi-dose chronic cancer bioassays,
respectively, and were used for the cancer dose-response assessment.
Increased incidences of neoplastic lesions were primarily observed in male rats, though results in
females are supportive of potential carcinogenicity of PFOA. Butenhoff et al. (2012, 2919192)
and Biegel et al. (2001, 673581) reported dose-dependent increases in testicular LCTs.
Additionally, LCT incidence at similar dose levels was comparable between the two studies (11
and 14%, respectively). PACTs were observed in both the NTP (2020, 7330145) and Biegel et
al. (2001, 673581) studies. NTP (2020, 7330145) reported increased incidences of pancreatic
acinar cell adenomas in males in all treatment groups compared to their respective controls. This
rare tumor type was also observed in female rats from the highest dose group, though the
increased incidence did not reach statistical significance. Biegel et al. (2001, 673581) reported
increases in the incidence of PACTs in male rats treated with PFOA, with zero incidences
observed in control animals. In addition, both NTP (2020, 7330145) and Biegel et al. (2001,
673581) reported dose-dependent increases in the incidence of liver adenomas in male rats.
Therefore, testicular LCTs, pancreatic acinar cell adenomas, and liver adenomas in male rats
were considered for CSF derivation.
NTP (2020, 7330145) observed marginally increased incidences of uterine adenocarcinomas in
female Sprague-Dawley rats during the extended evaluation (i.e., uterine tissue which included
cervical, vaginal, and uterine tissue remnants). The accompanying incidences of uterine
hyperplasia did not follow a dose-response relationship. Uterine adenocarcinomas were not
considered for CSF derivation because "the strength of the response was marginal and there was
a low concurrent control incidence that lowered confidence in the response" {NTP, 2020,
7330145}. Butenhoff et al. (2012, 2919192) identified mammary fibroadenomas and ovarian
tubular adenomas in female rats, though there were no statistical differences in incidence rates
between PFOA-treated groups and controls. These tumor types were also not considered for CSF
4-54
-------
DRAFT FOR PUBLIC COMMENT
March 2023
derivation because the incidences were not statistically significant and the tumors were not
observed by NTP (2020, 7330145).
4.2.1.2 CSF Derivation
In the 2016 HESD for PFOA {U.S. EPA, 2016, 3603279}, a CSF based on LCTs reported by
Butenhoff et al. (2012, 2919192) was calculated to determine if a lifetime Health Advisory
derived from the RfD would be protective for the cancer endpoint. At that time, the dose-
response relationship for the LCTs observed by Butenhoff et al. (2012, 2919192) was modeled
using EPA's Benchmark Dose Software (BMDS) Version 2.3.1. The multistage cancer model
predicted the dose at which a 4% increase in tumor incidence would occur. The 4% increase was
chosen as the low-end of the observed response range within the Butenhoff et al. (2012,
2919192) results. The CSF presented in the 2016 PFOA HESD of 0.07 (mg/kg/day)"1 was
derived from the BMDL04 of 1.99 mg/kg/day after converting the animal BMDL to an HED
using body weights to the % power. The resultant 0.5 [j,g/L value was greater than the lifetime
Health Advisory (0.070 (J,g/L) based on noncancer effects {U.S. EPA, 2016, 3982042},
indicating that the Health Advisory derived based on the developmental endpoint was protective
for the cancer endpoint.
EPA has reevaluated the LCTs reported by Butenhoff et al. (2012, 2919192) in the current effort
using the updated animal and human PK models described in Section 4.1.3. These modeling
results are described in the Appendix (see PFOA Appendix). To duplicate the findings from the
2016 PFOA HESD, a BMR of 4% was chosen as the low end of the observed response range
within the study results. EPA also derived CSFs for the tumor types observed in the NTP study
that provide further evidence of carcinogenic activity of PFOA in male Hsd:Sprague DawleySD
rats: hepatocellular neoplasms (hepatocellular adenomas and carcinomas) and acinar cell
adenomas of the pancreas {NTP, 2020, 7330145} (Table 4-12). A BMR of 10% was selected for
these tumor types, consistent with the BMD Technical Guidance {U.S. EPA, 2012, 1239433}.
For all tumor types, dichotomous models were used to fit dose-response data. For LCTs reported
by Butenhoff et al. (2012, 2919192), the area under the curve (AUC) for duration of the study
was selected for this model because the AUC accounts for the accumulation of effects expected
to precede the increased incidence of Ley dig cell adenomas. For tumor types reported by NTP
(2020, 7330145), the Cavg_pup total was selected for this model to account for the perinatal window
of exposure. The Cavg_pup total metric averages out the concentration in the pup from conception to
the end of the 2 years by adding the area under the curve in gestation/lactation to the area under
the curve from diet (post-weaning) and dividing by two years. The BMDS produced BMDLs in
mg/L for all tumor types. The animal PODs were converted to PODheds by multiplying the POD
by the human clearance value (Table 4-6). This PODhed is equivalent to the constant exposure,
per body weight, that would result in serum concentration equal to the POD at steady state. The
CSF is then calculated by dividing the BMR by the PODhed. These modeling results are
described further in the Appendix (see PFOA Appendix).
4-55
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Table 4-12. Cancer Slope Factors based on Animal Toxicological Data
Tumor Type
Reference,
Confidence
Strain/
Species/Sex
POD Type,
Model
POD Internal
Dose/Internal
Dose Metric3
PODhed
CSF
(BMR/PODhed)
Ley dig Cell
Adenomas in
the Testes
Butenhoff et al.
(2012, 2919192)
Medium
Male Sprague-
Dawley Rats
BMDL4RD,
Multistage
Degree 1
27,089.3
(AUC
(mgxd/L)
4.75 xlO"3
mg/kg/day
8.42
(mg/kg/day)"1
Hepatocellular NTP (2020,
Adenomas or 7330145)
Carcinoma High
Fi Male Sprague- BMDLiord,
Dawley Rats, Multistage
Perinatal and Degree 2
Postweaning
Exposure
88.7
(Cavg_pup_total in
mg/L)
1.06X10-2
mg/kg/day
9.4
(mg/kg/day)1
1.88X103
mg/kg/day
53.2
(mg/kg/day)"1
Pancreatic NTP (2020, Fi Male Sprague- BMDLiord, 15.7
Acinar Cell 7330145) Dawley Rats, Multistage (CaVg_pup total in
Adenoma High Perinatal and Degree 1 mg/L)
Postweaning
Exposure
Notes: AUC = area under the curve;
change; BMDLiord = lower bound
BMR = benchmark response; CSF
a see PFOA Appendix for additional
BMDL4RD = benchmark dose level corresponding to the 95% lower confidence limit of a 4%
on the dose level corresponding to the 95% lower confidence limit for a 10% change;
= cancer slope factor; NTP = National Toxicology Program,
details on benchmark dose modeling.
4.2.2 Epidemiological Studies
4.2.2.1 Study and End point Selection
This updated review indicated that there is an increase in risk for kidney or RCC and testicular
cancers and PFOA exposure {Shearer, 2021, 7161466; Chang, 2014, 2850282; Bartell, 2021,
7643457}. Although newer studies generally show no association with breast cancer in the
general population, there is some evidence that PFOA may be associated with breast cancer risk
in participants with specific polymorphisms or specific types of tumors {Ghisari, 2017, 3860243;
Mancini, 2019, 5381529}. Two occupational studies {Steenland, 2015, 2851015; Girardi, 2019,
6315730} support an increase in risk for liver cancer and malignant neoplasm of the lymphatic
and hematopoietic tissue, as well as an increasing trend in prostate cancer that did not reach
statistical significance. No associations were found for colorectal cancer in either the general
population or occupational studies, or for lung cancer in occupational studies.
Results are most consistent for kidney cancer in adults based on a new nested case-control study
{Shearer, 2021, 7161466}, two C8 Health Project studies {Barry, 2013, 2850946; Vieira, 2013,
2919154}, and an occupational mortality study {Steenland, 2012, 2919168} described in the
2016 PFOA Health Advisory {U.S. EPA, 2016, 3982042}.
For dose-response modeling, Shearer et al. (2021, 7161466) was selected as the key study.
Shearer et al. (2021, 7161466) is a multi-center case-control study nested within NCI's PLCO.
The PLCO is a randomized clinical trial of the use of serum biomarkers for cancer screening.
The cases in this study {Shearer 2021, 7161466} included all the participants of the screening
arm of the PLCO trial who were newly diagnosed with RCC during the follow-up period
(N = 326). All cases were histopathologically confirmed. Controls were selected from among
participants of the PLCO trial screening arm who had never had RCC. Controls were
individually matched to the RCC cases by age at enrollment, sex, race/ethnicity, study center,
4-56
-------
DRAFT FOR PUBLIC COMMENT
March 2023
and year of blood draw. PFOA concentrations were measured in the baseline serum samples
collected between 1993 and 2002. Median PFOA levels in controls was 5.0 ng/mL, comparable
with 4.8 ng/mL in adults 60 and over from NHHANES 1999-2000. The analyses accounted for
numerous confounders including BMI, smoking, history of hypertension, eGFR, previous freeze-
thaw cycle, calendar and study year of blood draw, sex, race and ethnicity, study center. Socio-
economic status was not explicitly considered in the analyses.
There was a statistically significant increase in odds of RRC per doubling of PFOA (OR = 1.71,
95% CI: 1.23, 2.37) and in the highest vs. lowest quartile (OR = 2.63, 95% CI: 1.33, 5.2).
Although non-significant elevated risks were observed in the second and third quartiles, there
was a statistically significant increasing trend with increasing PFOA exposure across quartiles
(p-trend = 0.007). Statistically significant increased odds of RCC were observed in participants
ages 55-59 years, and in men and in women, separately.
For sensitivity analyses, EPA also considered the C8 Health Project study {Vieira, 2013,
2919154} and those results can be found in the Appendix (see PFOA Appendix). Shearer et al.
(2021, 7161466) was selected as the critical study over Vieira et al. (2013, 2919154) due to
multiple study design considerations. Specifically, study design advantages of Shearer et al.
(2021, 7161466) compared with the Vieira et al. (2013, 2919154) include specificity in the
health outcome considered (RCC vs. any kidney cancer), the type of exposure assessment (serum
biomarker vs. modeled exposure), source population (multi-center vs. Ohio and West Virginia
regions), and study size (324 cases and 324 matched controls vs. 59 cases and 7,585 registry-
based controls).
The high exposure occupational study by Steenland and Woskie (2012, 2919168) evaluated
kidney cancer mortality in workers from West Virginia and observed significant elevated risk of
kidney cancer death in the highest exposure quartile (> 2,384 ppm-years). This study was limited
by the small number of observed cancer cases (six kidney cancer deaths). This study was not
used for dose-response analysis because information on a range of exposures more relevant to
the general population were available from the Shearer et al. (2021, 7161466) and Vieira et al.
(2013, 2919154). The study by Barry et al. (2013, 2850946) was not used for dose-response
analysis because it was performed in the same study area as the Vieira et al. (2013, 2919154)
study and these two studies likely involved a number of the same participants. In addition, Barry
et al. (2013, 2850946) could not be used in the sensitivity analysis because it lacked the
necessary exposure measurements for CSF calculation. In this study, estimated PFOA
concentrations are provided in Table 2 for community level and worker level. However,
combined exposure levels of each quartile of the overall study population were not reported.
Without overall exposure data in each quartile, CSF calculations are not feasible.
The study by Raleigh et al. (2014, 2850270) was not selected because of the concerns of
exposure assessment methods and study quality. This study used modeled estimates of PFOA air
concentrations in the workplace rather than biomonitoring measurements. This is a concern
because the assessment lack of information about the degree to which inhaled PFOA is absorbed
in humans and factors that may affect the absorption, as well as PFOA exposure data in non-
production workers was not based on actual measurements. In addition, this study did not
observe an association between PFOA and kidney cancer. The possible reasons of this study
could have missed to identify the association between PFOA, and kidney cancer include
4-57
-------
DRAFT FOR PUBLIC COMMENT
March 2023
relatively small numbers of cases, lack of information adjustment on risk factors of kidney
cancer such as smoking status and BMI, and the methods for exposure assessment
4.2.2.2 CSF Derivation
EPA calculated CSFs for RCC from Shearer et al. (2021, 7161466) based on the method used in
CalEPA (2021, 9416932) and for its Public Health Goals for Arsenic in Drinking Water
{OEHHA, 2004, 10369748}. Details are provided in the Appendix (see PFOA Appendix). The
underlying model involves a linear regression between PFOA exposure and cancer relative risk
used to estimate the dose-response between PFOA and RCC risk. This was calculated using a
weighted linear regression of the quartile specific RRs, with the weights defined as the inverse of
the variance of each RR. Since the incidence of kidney cancer is relatively low and because the
cases and controls were matched on age, the ORs represent a good approximation of the
underlying RRs. The CSF is then calculated as the excess cancer risk associated with each ng/mL
increase in serum PFOA (internal CSF). The internal CSF was calculated by first converting the
linear regression model discussed above from the RR scale to the absolute risk scale. This was
done assuming a baseline risk (Ro) of RCC or kidney cancer in an unexposed or lower exposure
reference group. Since this is not available in a case-control study, the lifetime risk of RCC in
U.S. males is used. The lifetime RCC risk was estimated by multiplying the lifetime risk of
kidney cancer in U.S. males {American Cancer Society, 2020, 9642148} by the percentage of all
kidney cancers that are the RCC subtype (90%). This gives an Ro of 0.0202 x 90% = 0.0182. The
internal CSF was then calculated as either the product of the upper 95% CI or the central
tendency of the dose-response slope and Ro and represents the excess cancer risk associated with
each ng/mL increase in serum PFOA. The internal serum CSF was converted to an external dose
CSF, which describes the increase in cancer risk per 1 ng/(kg-day) increase in dose. This was
done by dividing the internal serum CSF by the selected clearance value, which is equivalent to
dividing by the change in external exposure that results in a 1 ng/mL increase in serum
concentration at steady-state. The clearance value used was the same as that used in the updated
Verner model for endpoints related to developmental exposure (Table 4-6). The results of the
modeling and the CSFs derived are presented in Table 4-13.
Table 4-13. Cancer Slope Factors based on Epidemiological Data
Internal CSF -
CSF - Increase in
Tumor Type
Reference,
Strain/
POD Type,
Increase in cancer risk
cancer risk per
Confidence
Species/Sex
Model
per 1 ng/mL serum
increase
1 ng/(kg*d) increase
in dose
Renal cell
Shearer et al.
Human, male
CSF serum in
3.52X10-3
0.0293 (ng/kg/day)"1
carcinoma
(2021,
and female 55-
adults (per
(ng/mL)"1
(RCC)
7161466)
Medium
74 years
ng/mL of
serum PFOA);
upper limit of
the 95% CI
(see PFOA Appendix
for additional detail)
Kidney cancer
Vieira et al.
Human, male
CSF serum in
4.81xl0-4
0.00401 (ng/kg/day)1
(2013,
and female
adults (per
(ng/mL)"1
2919154)
ng/mL of
(see PFOA Appendix
Medium
serum PFOA);
upper limit of
the 95% CI,
highest
for additional detail)
4-58
-------
DRAFT FOR PUBLIC COMMENT
March 2023
„ _ Reference,
Tumor Type „
Confidence
Strain/
Species/Sex
POD Type,
Model
Internal CSF -
Increase in cancer risk
per 1 ng/mL serum
increase
CSF - Increase in
cancer risk per
1 ng/(kg*d) increase
in dose
exposure group
excluded
Notes: CI = Confidence Interval; CSF = cancer slope factor; POD = point of departure.
4.2.3 CSF Selection
Overall, new data and the candidate CSFs indicate that PFOA is a more potent carcinogen than
previously understood and described in the 2016 HESD {U.S. EPA, 2016, 3603279}. To select
an overall CSF, EPA focused on the CSFs derived from the epidemiological data consistent with
the draft ORD Staff Handbook which states "when both laboratory animal data and human data
with sufficient information to perform exposure-response modeling are available, human data are
generally preferred for the derivation of toxicity values" {U.S. EPA, 2022, 10476098}. EPA
selected the critical effect of renal cell carcinomas in human males reported by Shearer et al.
(2021, 7161466) as the basis of the CSF for PFOA because it is a multi-center case-control
epidemiological study nested within NCI's PLCO with median PFOA levels relevant to the
general U.S. population.
Study design advantages of the Shearer et al. (2021, 7161466) study compared with the Vieira et
al. (2013, 2919154) study include specificity in the health outcome considered (RCC vs. any
kidney cancer), the type of exposure assessment (serum biomarker vs. modeled exposure),
source population (multi-center vs. Ohio and West Virginia regions), and study size (324 cases
and 324 matched controls vs. 59 cases and 7,585 registry-based controls). The resulting CSF is
0.0293 (ng/kg/day)"1.
Selection of renal cell carcinomas as the critical effect is supported by other studies of a highly
exposed community {Barry, 2013, 2850946; Vieira, 2013, 2919154}, an occupational kidney
cancer mortality study {Steenland, 2012, 2919168}, as well as a meta-analysis of
epidemiological literature that concluded that there was an increased risk of kidney tumors
correlated with increased PFOA serum concentrations {Bartell, 2021, 7643457}.
4.2.4 Application of Age-Dependent Adjustment Factors
EPA's Guidelines for Carcinogen Risk Assessment and Supplemental Guidance for Assessing
Susceptibility from Early-Life Exposure to Carcinogens require the consideration of applying
age-dependent adjustment factors (ADAFs) to CSFs to address potential increased risk for cancer
due to early life stage susceptibility to chemical exposure {U.S. EPA, 2005, 6324329; U.S. EPA
2005, 88823}. ADAFs are only to be used for carcinogenic chemicals with a mutagenic MOA
when chemical-specific data about early-life susceptibility are lacking. For carcinogens with any
MOA, including mutagens and non-mutagens, but with available chemical specific data for
early-life exposure, those data should be used.
As described in Section 3.5.3.1.1, most of the studies assessing mutagenicity following PFOA
exposure were negative and therefore, PFOA is unlikely to cause tumorigenesis via a mutagenic
MOA. Given the lack of evidence of a mutagenic MOA, EPA does not recommend applying
ADAFs when quantitatively determining the cancer risk for PFOA {U.S. EPA, 2011, 783747}.
4-59
-------
DRAFT FOR PUBLIC COMMENT
March 2023
EPA additionally evaluated whether there are chemical specific data for early-life exposure to
PFOA and determined that there is insufficient information available from epidemiological and
animal toxicological studies to adequately determine whether exposure during early-life periods,
per EPA's above-referenced supplemental guidance, may increase incidence or reduce latency
for cancer compared with adult-only exposure. No current studies allow for comparisons of
cancer incidence after early-life vs. adult-only PFOA exposure. However, there are two studies
that assessed cancer risk after PFOA exposure during various developmental stages.
An NTP cancer bioassay in rats chronically exposed to PFOA both perinatally and post-weaning
did not report an increased cancer risk compared to chronic postweaning-only exposure (see
further study design details in Section 3.4.4.2.1.2 and study results in Section 3.5.2), which
suggests no increased cancer risk as a result of lifetime exposure compared to postweaning-only
exposure. The NTP cancer bioassay does not include dose groups that were only exposed during
early-life stages (i.e., only during development), the findings of the NTP cancer bioassay do not
provide a basis for quantitatively estimating the difference in susceptibility between early-life
and adult exposures. The other study, by Filgo et al. (2015, 2851085), reported equivocal
evidence of hepatic tumors in three strains of F1 female mice perinatally treated with PFOA
(gestational exposure from GD 1-17 and potential exposure through lactation) and necropsied at
18 months of age. This study is also limited in that there was no adult-only exposure comparison
group, and the authors only assessed female mice and only histopathologically examined the
liver {Filgo, 2015, 2851085}. In summary, the available studies do not provide information on
whether early-life PFOA exposures result in increased cancer incidence compared with adult-
only exposure. Due to the lack of evidence supporting postnatal early life susceptibility to PFOA
exposure, EPA did not adjust the risk value using chemical-specific data.
4-60
-------
DRAFT FOR PUBLIC COMMENT
March 2023
5 MCLG Derivation
Consistent with the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329},
EPA reviewed the weight of the evidence and determined that PFOA is Likely to Be
Carcinogenic to Humans, as "the evidence is adequate to demonstrate carcinogenic potential to
humans but does not reach the weight of evidence for the descriptor Carcinogenic to Humans
This determination is based on the evidence of kidney and testicular cancer in humans and
Leydig cell tumors, pancreatic acinar cell tumors, and hepatocellular adenomas in rats as
described in Section 3.5. As previously noted, there is additional evidence supporting the
carcinogenicity of PFOA since the publication of the prior PFOA 2016 HESD {U.S. EPA, 2016,
3603729}. In particular, the evidence of kidney cancer in humans from high-exposure
community studies {Vieira, 2013, 2919154; Barry, 2013, 2850946} is corroborated by
associations between PFOA serum concentrations and risk of renal cell carcinoma from a nested
case-control study in the general population {Shearer, 2021, 7161466}. In addition, in animal
models, the evidence of multi-site tumorigenesis reported in two chronic bioassays in rats
{Butenhoff, 2012, 2919192; Biegel, 2001, 673581} is now further supported by similar findings
of multi-site tumorigenesis from a third chronic bioassay in rats {NTP, 2020, 7330145}.
Unless a non-linear mode of action is determined, EPA establishes MCLGs of zero for
carcinogens classified as Carcinogenic to Humans or Likely to be Carcinogenic to Humans
consistent with the statutory definition of MCLG, which requires EPA to establish MCLGs at a
level where there are "no known or anticipated adverse effects" on public health and with "an
adequate margin of safety." Under SDWA, where there is insufficient information to determine
that a carcinogen has a threshold below which there are no carcinogenic effects, EPA takes the
health-protective approach of assuming that there is no such threshold and that carcinogenic
effects should therefore be extrapolated linearly to zero {U.S. EPA, 1985, 9207; U.S. EPA, 1991,
5499; U.S. EPA, 2016, 6557097}. This approach, known as the linear default extrapolation
approach, ensures that the MCLG is set at a level where there are no adverse health effects with a
margin of safety. EPA has determined that PFOA is Likely to be Carcinogenic to Humans based
on sufficient evidence of carcinogenicity in humans and animals, that there is not sufficient
evidence of a threshold for PFOA, and that therefore a linear default extrapolation approach is
appropriate {U.S. EPA, 2005, 6324329}. Based upon a consideration of the best available peer
reviewed science and a consideration of an adequate margin of safety, EPA proposes a MCLG of
zero for PFOA in drinking water.
5-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6 Effects Characterization
6.1 Addressing Uncertainties in the Use of Epidemiological
Studies for Quantitative Dose-Response Analyses
In the 2016 PFOA HESD and Drinking Water Health Advisory {U.S. EPA, 2016, 3603279; U.S.
EPA, 2016, 3982042}, EPA qualitatively considered epidemiological studies as a supporting line
of evidence but did not quantitatively consider them for POD derivation, citing the following as
reasons to exclude the epidemiological data that were available at that time from quantitative
analyses:
• inconsistencies in the epidemiological database,
• the use of mean serum PFOA concentrations rather than estimates of exposure,
• declining serum PFOA values in the U.S. general population over time {CDC, 2017,
4296146},
• uncertainties related to potential exposure to additional PFAS, telomer alcohols that
metabolically break down into PFOA, and other bio-persistent contaminants, and
• uncertainties related to the clinical significance of effects observed in epidemiological
studies.
Since 2016, EPA has identified many additional epidemiology studies that have increased the
database of information for PFOA (see Sections 3.1.1, 3.4, and 3.5). Further, new tools that have
facilitated the use of study quality evaluation as part of systematic review have enabled EPA to
systematically assess study quality in a way that includes consideration of confounding. As a
result, EPA is now in a position to be able to quantitatively consider epidemiological studies for
POD derivation in this assessment.
In this assessment EPA has assessed the strength of epidemiological and animal evidence
systematically, a process that was not followed in 2016. By performing an updated assessment
using systematic review methods, EPA determined that five health outcomes and five
epidemiological endpoints within these outcomes (i.e., decreased antibody response to
vaccination in children, decreased birthweight, elevated total cholesterol, elevated ALT, and
increased risk of kidney cancer) have sufficient weight of evidence to consider quantitatively.
Each endpoint quantified in this assessment has consistent evidence from multiple medium
and/or high confidence epidemiological and animal toxicological studies supporting an
association between PFOA exposure and the adverse effect. Each of the endpoints were also
specifically supported by multiple epidemiological studies in different populations, including
general and highly exposed populations. Several of these supporting studies have been published
since 2016 and have strengthened the weight of evidence for this assessment.
As described in Section 4.1.3, EPA has improved upon the pharmacokinetic modeling technique
used in 2016. Though there are challenges in estimations of human dosimetry from measured or
modeled serum concentrations (see Section 6.6.2), EPA has evaluated the available literature and
developed a pharmacokinetic model that estimates PFOA exposure concentrations from the
6-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
serum PFOA concentrations provided in epidemiological studies, which reduces uncertainties
related to exposure estimations in humans. This new approach is supplemented with the UF
accounting for intraspecies variation of lOx applied to each PODhed, which accounts for the
sensitivities of specific populations, including those that may have increased susceptibility to
PFOA toxicity due to differential toxicokinetics.
An additional source of uncertainty in using epidemiological data for POD derivation is the
documented declined in human serum PFOA levels over time, which raises concerns about
whether one-time serum PFOA measurements are a good representation of lifetime peak
exposure. Because of PFOA's long half-life in serum, however, one-time measurements likely
reflect several years of exposure {Lorber, 2011, 2914150}. Importantly, EPA considered
multiple time periods when estimating PFOA exposure, ranging from the longest period with
available data on PFOA serum levels within the U.S. population (1999-2018) to the shortest and
most recent period (2017-2018) (see Table E-17 in PFOA Appendix), when performing dose-
response modeling of the ALT and TC endpoints in the epidemiological data. EPA selected
PODs for these two endpoints using PFOA exposure estimates based on the serum PFOA data
for 1999-2018, which is likely to capture the peak PFOA exposures in the U.S. which occurred
in the 1990s {Dong, 2019, 5080195}. The modeling results show that the BMDL estimates for
increased TC derived using these exposure data are consistently lower than those based on the
2017-2018 PFOA exposure data whereas for ALT, the BMDL estimates using data from the
longest exposure period are consistently higher than those based on the 2017-2018 PFOA
exposure data. Based on these analyses, it appears that selection of one exposure time-period
over another does not predictably impact the modeling results. Therefore, for this assessment,
EPA decided to consistently select the time periods more likely to capture peak PFOA exposures
(e.g., 1999-2018) as the basis of BMDL estimates for all endpoints of interest (see PFOA
Appendix E).
It is plausible that observed associations between adverse health effects and PFOA exposure
could be explained in part by confounding from other PFAS exposures, including the metabolism
of precursor compounds to PFOA in the human body. However, for four of the five priority
health outcomes, at least one available study performed multi-pollutant modeling. For example,
for the decreased antibody production endpoint, Budtz-Jorgensen and Grandjean (2018,
5083631) performed a follow-up analysis of the study by Grandjean et al. (2012, 1248827) in
which results were additionally adjusted for PFOS, and there was no notable attenuation of the
observed association between PFOA exposure and decreased antibody response. Similarly, Lin
et al. (2010, 1291111) performed multipollutant modeling for the effects on serum ALT and
found that when PFOS, PFHxS, and PFNA were simultaneously added in the fully adjusted
regression models, the associations remained and were slightly larger; one unit increase in serum
log-PFOA concentration was associated with a 2.19 unit (95% CI: 1.40, 2.98) increase in serum
ALT concentration compared to an increase of 1.86 units (95% CI: 1.24, 2.48). For an extended
review of the uncertainties associated with PFAS co-exposures, see Systematic Review Protocol
for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (anionic and acid forms) IRIS Assessments
{U.S. EPA, 2020, 8642427}.
Additionally, there is uncertainty about the magnitude of the contribution of PFAS precursors to
PFOA serum concentrations, especially as biotransformation efficiency appears to vary
depending on the precursor of interest {Lorber, 2011, 2914150; Mcdonough, 2022, 10412593;
6-2
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Vestergren, 2008, 2558842; D'eon, 2011, 2903650}. The contributions of PFAS precursors to
serum concentrations also varies between populations with differing PFAS exposure histories
(i.e., individuals living at or near sites with AFFF use may have different precursor PFOA
contributions than the general population).
In addition, some populations may be disproportionately exposed to other contaminants, such as
polychlorobiphenyls and methylmercury. To address this, EPA quantified associations between
PFOA serum concentrations and endpoints of interest in populations with varying exposure
histories, including the general population and high-exposure communities. EPA observed
associations for several endpoints in populations known to have been predominantly exposed to
PFOA (e.g., C8 Health Project participants), reducing the uncertainty related to potential
confounding of other contaminants, including PFAS precursor compounds. These sensitivity
analyses are supportive of EPA's conclusions regarding the effects of PFOA reported across
many epidemiological studies.
In this assessment, studies were not excluded from consideration based primarily on lack of or
incomplete adjustments for potential confounders including socioeconomic status (SES) or
race/ethnicity. A small number of studies examining PFAS serum levels across SES and
racial/ethnic groups were identified, many of which reported on a national scale (e.g., using
NHANES data). The identified studies (most from the early-mid 2000s) reported that serum
concentrations of PFOA were lower among people of color on average nationwide {Buekers,
2018, 5080471; Kato, 2014, 2851230; Nelson, 2012, 4904674; Calafat, 2007, 1290899}.
However, certain races/ethnicities may have relatively higher serum concentrations than others
depending on the geographic location and the specific PFAS sampled {Park, 2019, 5381560}.
EPA acknowledges that in observational epidemiological studies, potential residual confounding
may result from SES and racial/ethnic disparities. Additional racially and ethnically diverse
studies in multiple U.S. communities are needed to fill this important data gap. The PFOA
Appendix provides detailed information on the available epidemiological studies and identifies
the study-specific confounding variables that were considered, such as SES.
Lastly, the potential uncertainty related to the clinical significance of effects observed in the
PFOA epidemiological studies is sometimes cited for dismissing the epidemiological data
quantitatively. However, as described in section 4.1.1, increased ALT levels, decreased antibody
responses in children, increased serum cholesterol levels, and decreased birthweight are
clinically meaningful effects, and EPA's A Review of the Reference Dose and Reference
Concentration Processes, states that a RfD should be based on an adverse effect or a precursor to
an adverse effect (e.g., increased risk of an adverse effect occuring) {U.S. EPA, 2002, 88824}.
Briefly, evidence from both human epidemiological and animal toxicological studies indicates
that increased serum ALT is associated with increased risk for liver disease {Ioannou, 2006,
10473853; Ioannou, 2006, 10473854; Kwo, 2017, 10328876; Roth, 2021, 9960592}. Human
epidemiological studies have also demonstrated that even low magnitude increases in serum
ALT can be clinically significant (See section 4.1.1.1). It is also important to note that while
evaluation of direct liver damage is possible in animal studies, it is difficult to obtain biopsy-
confirmed histological data in humans. Therefore, liver injury is typically assessed using serum
biomarkers of hepatotoxicity {Costello et al, 2022, 10285082}. The SAB's PFAS review panel
noted that reduction in the level of antibodies produced in response to a vaccine represents a
failure of the immune system to respond to a challenge and is considered an adverse
6-3
-------
DRAFT FOR PUBLIC COMMENT
March 2023
immunological health outcome {U.S. EPA, 2022, 10476098}. Further, a review by Selgrade
(2007, 736210) suggests that specific immunotoxic effects, such as antibody response, observed
in children may be broadly indicative of developmental immunosuppression impacting these
children's ability to protect against a range of immune hazards.
Additionally, increased serum cholesterol is associated with changes in incidence of
cardiovascular disease events such as myocardial infarction (MI, i.e., heart attack), ischemic
stroke (IS), and cardiovascular mortality occurring in populations without prior CVD events
{D'Agostino, 2008, 10694408; Goff, 2014, 3121148; Lloyd-Jones, 2017, 10694407}. Moreover,
disturbances in cholesterol homeostasis contribute to the pathology of non-alcoholic fatty liver
disease (NAFLD) and to accumulation of lipids in hepatocytes {Malhotra, 2020, 10442471},
providing further evidence of effects in the liver. Finally, substantial evidence links low birth
weight to a variety of adverse health outcomes at various stages of life. It has been shown to
predict prenatal mortality and morbidity {Cutland, 2017, 10473225; U.S. EPA, 2013, 4158459;
WHO, 2014, 10473141} and is a leading cause of infant mortality in the United States {CDC,
2020, 10473144}.Low birth weight is also associated with increased risk for diseases in
adulthood, including obesity, diabetes, and cardiovascular disease {Gluckman, 2008, 10473269;
Osmond, 2000, 3421656; Risnes, 2011, 2738398; Smith, 2016, 10474151; Ong, 2002,
10474127, as reported in Yang et al. (2022, 10176603).
There are challenges associated with quantitative use of epidemiological data for risk assessment
{Deener, 2018, 6793519} as described above; however, improvements such as methodological
advancements that minimize bias and confounding, strengthened methods to estimate and
measure exposure, and updated systematic review practices facilitate the use of epidemiological
studies to quantitatively inform risk.
6.2 Comparisons Between Toxicity Values Derived from
Animal Toxicological Studies and Epidemiological studies
As recommended by the SAB {U.S. EPA, 2022, 10476098}, EPA derived candidate RfDs and
CSFs for multiple health outcomes using data from both epidemiological and animal
toxicological studies. Candidate RfDs from epidemiological and animal toxicological studies
within a health outcome differed by approximately two to three orders of magnitude (see Figure
4-4. Comparison of Candidate RfDs Resulting from the Application of Uncertainty Factors to
PODheds Derived from Epidemiological and Animal Toxicological Studies, with
epidemiological studies producing lower values. EPA does not necessarily expect concordance
between animal and epidemiological studies in terms of the adverse effect(s) observed, as well as
the dose level that elicits the adverse effect(s). For example, EPA's Guidelines for
Developmental Toxicity Risk Assessment states that "the fact that every species may not react in
the same way could be due to species-specific differences in critical periods, differences in
timing of exposure, metabolism, developmental patterns, placentation, or mechanisms of action"
{U.S. EPA, 1991, 732120}. EPA further describes these factors in relation to this assessment
below.
First, there are well-established differences in the toxicokinetics between humans and animal
models such as rats and mice. As described in Section 3.3.1.4.5, PFOA half-life estimates vary
considerably by species, being lowest in rodents (hours to days) and several orders of magnitude
higher in humans (years). All candidate toxicity values based on animal toxicological studies
6-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
were derived from studies conducted in rats or mice, adding a potential source of uncertainty
related to toxicokinetic differences in these species compared to humans. To address this
potential source of uncertainty, EPA utilized a PK model to estimate the internal dosimetry of
each animal model and convert the values into predicted levels of human exposure that would
result in the corresponding observed health effects. However, the outputs of these models are
estimates and may not fully account for species-specific toxicokinetic differences, particularly
differences in excretion. The application of uncertainty factors (i.e., UFa) also may not precisely
reflect animal-human toxicokinetic differences.
Second, candidate toxicity values derived from epidemiological studies are based on responses
associated with actual environmental exposure levels, whereas animal toxicological studies are
limited to the tested dose levels which are often several orders of magnitude higher than the
ranges of exposure levels in humans. Extrapolation from relatively high experimental doses to
environmental exposure levels introduces a potential source of uncertainty for toxicity values
derived from animal toxicological studies; exposures at higher dose levels could result in
different responses, perhaps due to differences in mechanisms activated, compared to responses
to lower dose levels. One example of this is the difference between epidemiological and animal
toxicological studies in the effect of PFOA exposure on serum lipid levels (i.e., potential non-
monotonic dose-response relationships that are not easily assessed in animal studies due to low
dose levels needed to elicit the same response observed in humans).
Third, there may be differences in mechanistic responses between humans and animal models.
One example of this is the PPARa response. It is unclear to what extent PPARa influences the
responses to PFOA exposure observed in humans, though it has been shown that the rodent
PPARa response is greater than that observed in humans (see Section 3.4.1.3.1). Mechanistic
differences could influence dose-response relationships and subsequently result in differences
between toxicity values derived from epidemiological and animal toxicological studies. There
may be additional mechanisms that differ between humans and animal models that could
contribute to the magnitude of responses and doses required to elicit responses across species.
The factors described above represent some but not all potential contributors that may explain
the differences between toxicity values derived from epidemiological and animal toxicological
studies. In this assessment, EPA prioritized epidemiological studies of medium or high
confidence for the selection of health outcome-specific and overall RfDs and CSFs (see Section
4.1.6). The use of human data to derive toxicity values removes uncertainties and assumptions
about human relevance inherent in extrapolating from and interpreting animal toxicological data
in quantitative risk assessment.
6.3 Updated Approach to Animal Toxicological RfD Derivation
Compared to the 2016 PFOA HESD
For POD derivation in this assessment, EPA considered the studies identified in the recent
literature searches and also re-examined the candidate RfDs derived in the 2016 PFOA HESD
{U.S. EPA, 2016, 3603279} and the animal toxicological studies and endpoints on which they
were based. The updated approach used for hazard identification and dose response in the current
assessment as compared to the 2016 HESD led to some differences between animal toxicological
studies and endpoints used as the basis of candidate RfDs for each assessment. These updates
and the resulting differences are further described below.
6-5
-------
DRAFT FOR PUBLIC COMMENT
March 2023
For the 2016 PFOA HESD, EPA did not use BMD modeling to derive PODs, and instead relied
on the NOAEL/LOAEL approach for all candidate studies and endpoints {U.S. EPA, 2016,
3603279}. The NOAEL/LOAEL approach allows for the incorporation of multiple endpoints
from a single study to derive a single POD, if the endpoints have the same NOAEL and/or
LOAEL. For example, in the 2016 PFOA HESD, EPA derived a candidate RfD based on the
endpoints of decreased parental body weight and increased parental absolute and relative kidney
weight reported by Butenhoff et al. (2004, 1291063), all of which shared a common POD
(LOAEL). For the current assessment, EPA preferentially used BMD modeling to derive PODs
because it allows for greater precision than the NOAEL/LOAEL approach and considers the
entire dose-response curve. This approach requires the consideration of endpoints on an
individual basis and further examination of the weight of evidence for particular endpoints, as
well as the dose-response trend reported for each endpoint, in order to derive a BMDL. When
considering an effect on a standalone basis rather than together with other effects occurring at the
same exposure level, EPA sometimes determined the weight of evidence was not sufficient to
consider an individual endpoint for POD derivation. For the current assessment, EPA used a
systematic review approach consistent with the IRIS Handbook {U.S. EPA, 2022, 10476098} to
consider the weight of evidence for both the health outcomes as well as for individual endpoints
of interest when selecting endpoints and studies for dose-response modeling. In the case of the
endpoints selected in 2016 from the Butenhoff et al. (2004, 1291063) study, systemic effects
such as body weight and renal effects such as kidney weight were reevaluated and determined to
have evidence suggestive of an association with PFOA exposure. As described in Section 4.1.1
of this assessment, EPA derived PODs only for endpoints from health outcomes with evidence
indicating or evidence demonstrating an association with PFOA exposure.
Additionally, for the current assessment, EPA preferentially selected endpoints for which there
were a greater number of studies supporting the observed effect. For example, for the 2016
PFOA HESD, EPA derived a candidate RfD based on the co-critical effect of accelerated male
puberty reported by Lau et al. (2006, 1276159). Results of the current assessment's literature
search showed that no high or medium confidence studies supporting that observed effect have
been published since 2016. As Lau et al. (2006, 1276159) was also the only study identified in
2016 that reported an acceleration of male puberty (a second study reported a delay in male
puberty {Butenhoff, 2004, 1291063}) and there were several other developmental endpoints
(e.g., reduced offspring weight and survival; delayed eye opening) that were supported by
multiple studies, EPA did not further consider this endpoint from Lau et al. (2006, 1276159) for
POD derivation in the present assessment. Similarly, upon further evaluation during the current
assessment of the co-critical effects of reduced forelimb and hindlimb ossification in pups
reported by Lau et al. (2006, 1276159), it was determined that an unexplained non-linear dose-
response trend adds uncertainty to selection of the LOAEL as the POD. As reduced ossification
was only observed at the highest dose tested (10 mg/kg/day) by the one other study {Yahia,
2010, 1332451} that tested dose levels close to the LOAEL from Lauetal. (2006, 1276159) (1
mg/kg/day) and because no studies identified during literature searches for the current
assessment reported this effect, EPA relied on other endpoints from Lau et al. (2006, 1276159)
that were amenable to BMD modeling, exhibited dose-dependent response trends, and were
supported by at least one other study in the available literature. For several studies considered in
the 2016 PFOA HESD, EPA performed BMD modeling for specific endpoints, but the efforts
did not produce viable model fits (see PFOA Appendix).
6-6
-------
DRAFT FOR PUBLIC COMMENT
March 2023
For some health effects that served as the basis for candidate RfDs in the 2016 PFOA HESD,
new studies published since 2016 provide more information about these same endpoints. For
example, in 2016, EPA derived a candidate RfD based on increased liver weight and necrosis in
rats reported by Perkins et al. (2004, 1291118). Since that time, NTP (2020, 7330145) published
an animal bioassay that has more strengths than the older study based on study design and study
quality evaluation results. Specifically, the NTP (2020, 7330145) study was identified as a high
confidence study that used a chronic (rather than 14-week) exposure duration, larger sample
sizes (n = 50), and a dose range that was more sensitive to the histopathological effects in both
male and female rats. Therefore, EPA considered liver necrosis data as reported by NTP (2020,
7330145) for POD derivation rather than data from the medium confidence study by Perkins et
al. (2004, 1291118).
For transparency, EPA has provided a comparison of studies and endpoints used to derive
candidate RfDs for both the 2016 PFOA HESD and the present assessment (Table 6-1).
Table 6-1. Comparison of Candidate RfDs Derived from Animal Toxicological Studies for
Priority Health Outcomes"
Studies and Effects Used in 2016 for Candidate RfD
Derivationb
Dewitt et al. (2008, 1290826), medium confi
reduced immunoglobulin M (IgM) response
Lau et al. (2006, 1276159), medium confidence -
reduced pup ossification (forelimb and hindlimb) and
accelerated male puberty (preputial separation)
Wolf et al. (2007, 1332672), medium confidence -
decreased pup body weight
Studies and Effects Used in 2023 for Candidate RfD
Derivation
;t al. (2008, 1290826), medium confidence -
immunoglobulin M (IgM) response
Lau et al. (2006, 1276159), medium confidence - delayed
time to eye opening
Song et al. (2018, 5079725), medium confidence -
decreased offspring survival
(2020, 7330145), high confidence - liver necrosis
Hepatic
Perkins et al. (2004, 1291118), medium confidence - NTP
increased liver weight and necrosis
Immune
idence - Dewitt e
reduced
Developmental
Notes: RfD = reference dose; IgM = immunoglobulin M; NTP = National Toxicology Program.
a Note that candidate RfDs for the fourth priority health outcome (i.e., cardiovascular) are not presented in this table because
candidate RfDs based on animal toxicological studies representing this health outcome were not derived in the 2016 HESD or
the current assessment.
b Candidate RfDs from the 2016 PFOA HESD that correspond to non-prioritized health outcomes (e.g., renal) are not presented
here.
6.4 Consideration of Alternative Conclusions Regarding the
Weight of Evidence of PFOA Carcinogenicity
In the 2016 PFOA HESD, EPA determined that the available carcinogenicity database for PFOA
at that time was consistent with the descriptions for Suggestive Evidence of Carcinogenic
Potential {U.S. EPA, 2016, 3603279}. Upon reevaluation for this assessment, EPA identified
several new studies reporting on cancer outcomes and subsequently determined the currently
6-7
-------
DRAFT FOR PUBLIC COMMENT
March 2023
available carcinogenicity database is consistent with the descriptions for Likely to be
Carcinogenic to Humans according to the Guidelines for Carcinogen Risk Assessment {U.S.
EPA, 2005, 6324329} (see Section 3.5.5). More specifically, the available data for PFOA surpass
many of the descriptions for Suggestive Evidence of Carcinogenic Potential provided in the
Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}. The examples for
which the PFOA database exceeds the descriptions include:
• "a small, and possibly not statistically significant, increase in tumor incidence observed in
a single animal or human study that does not reach the weight of evidence for the
descriptor "Likely to Be Carcinogenic to Humans." The study generally would not be
contradicted by other studies of equal quality in the same population group or
experimental system (see discussions of conflicting evidence and differing results, below);
• a small increase in a tumor with a high background rate in that sex and strain, when there
is some but insufficient evidence that the observed tumors may be due to intrinsic factors
that cause background tumors and not due to the agent being assessed;
• evidence of a positive response in a study whose power, design, or conduct limits the
ability to draw a confident conclusion (but does not make the study fatally flawed), but
where the carcinogenic potential is strengthened by other lines of evidence (such as
structure-activity relationships); or
• a statistically significant increase at one dose only, but no significant response at the other
doses and no overall trend" {U.S. EPA, 2005, 6324329}.
There are multiple medium or high confidence human and animal toxicological studies that
provide evidence of multiple tumor types resulting from exposure to PFOA. The observed tumor
types are generally consistent across human subpopulations (i.e., kidney and testicular) or studies
within the same species of rat (i.e., testicular, pancreatic, and hepatic) and there is no indication
that a high background incidence or other intrinsic factors related to these tumor types are
driving the observed responses. The SAB PFAS Review Panel agreed that: "a) the evidence for
potential carcinogenicity of PFOA has been strengthened since the 2016 HESD; b) the results of
human and animal studies of PFOA are consistent with the examples provided above and support a
designation of "likely to be carcinogenic to humans"; and c) the data exceed the descriptors for the
three designations lower than "likely to be carcinogenic" {U.S. EPA, 2022, 10476098}.
While the SAB panel agreed that the data for PFOA exceed a Suggestive cancer descriptor, the
final report also recommends "explicit description of how the available data for PFOA do not
meet the criteria for the higher designation as "carcinogenic" {U.S. EPA, 2022, 10476098}.
After reviewing the descriptions of the descriptor Carcinogenic to Humans, EPA has determined
that at this time, the evidence supporting the carcinogenicity of PFOA does not warrant a
descriptor exceeding Likely to be Carcinogenic to Humans. As discussed in Section 3.5.5, there
is not convincing epidemiological evidence supporting a causal association between human
exposure to PFOA and cancer. The Guidelines also indicate that a chemical agent can be deemed
Carcinogenic to Humans if it meets all of the following conditions:
• "there is strong evidence of an association between human exposure and either cancer or
the key precursor events of the agent's mode of action but not enough for a causal
association, and
6-8
-------
DRAFT FOR PUBLIC COMMENT
March 2023
• there is extensive evidence of carcinogenicity in animals, and
• the mode(s) of carcinogenic action and associated key precursor events have been
identified in animals, and
• there is strong evidence that the key precursor events that precede the cancer response in
animals are anticipated to occur in humans and progress to tumors, based on available
biological information" {U.S. EPA, 2005, 6324329}.
Though the available evidence indicates that there are positive associations between PFOA and
multiple cancer types, there is significant uncertainty regarding the carcinogenic MOA(s) of
PFOA, particularly for renal cell carcinomas and testicular cancer in humans. The evidence of
carcinogenicity in animals is limited to a single strain of rat, although PFOA tested positive for
multi-site tumorigenesis. The animal database does not provide significant clarity on the MOA of
PFOA in humans, though there is some evidence supporting hormone-mediated MOAs for
testicular tumors and oxidative stress-mediated MOAs for pancreatic tumors. Overall, there is
significant uncertainty that key events in animals are anticipated to occur in humans and progress
to tumors. The SAB similarly concluded that "the available epidemiologic data do not provide
convincing evidence of a causal association but rather provide evidence of a plausible association,
and thus do not support a higher designation of 'carcinogenic to humans'" {U.S. EPA, 2022,
10476098}.
Table 6-2 depicts comparisons of the PFOA carcinogenicity database with the Suggestive and
Known cancer descriptors.
Table 6-2. Comparison of the PFOA Carcinogenicity Database with Cancer Descriptors as
Described in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}
Comparison of Evidence for Carcinogenic and Suggestive Cancer Descriptors
Carcinogenic to Humans
This descriptor is appropriate when there is convincing PFOA data are not consistent with this description,
epidemiologic evidence of a causal association There is evidence of a plausible association between
between human exposure and cancer. PFOA exposure and cancer in humans, however, the
database is limited to only two independent populations,
there is uncertainty regarding the potential confounding
of other PFAS, and there is limited mechanistic
information that could contribute to the determination of
a causal relationship. The database would benefit from
additional large high confidence cohort studies in
independent populations.
Or, this descriptor may be equally appropriate with a lesser weight of epidemiologic evidence that is
strengthened by other lines of evidence. It can be used when all of the following conditions are met:
There is strong evidence of an association between PFOA data are not consistent with this description,
human exposure and either cancer or the key precursor There is evidence of an association between human
events of the agent's mode of action but not enough for exposure and cancer, however, there is limited
a causal association mechanistic information that could contribute to the
determination of a causal relationship.
There is extensive evidence of carcinogenicity in PFOA data are not consistent with this description,
animals While there are three chronic cancer bioassays available,
each testing positive in at least one tumor type, they
were all conducted in the same strain of rat. The
6-9
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Comparison of Evidence for Carcinogenic and Suggestive Cancer Descriptors
database would benefit from high confidence chronic
studies in other species and/or strains.
The mode(s) of carcinogenic action and associated key PFOA data are not consistent with this description. A
precursor events have been identified in animals definitive MOA has not been identified for each of the
PFOA-induced tumor types identified in rats.
There is strong evidence that the key precursor events PFOA data are not consistent with this description,
that precede the cancer response in animals are The animal database does not provide significant clarity
anticipated to occur in humans and progress to tumors, on the MOA(s) of PFOA in humans, though there is
based on available biological information some evidence supporting hormone-mediated MO As for
testicular tumors and oxidative stress-mediated MO As
for pancreatic tumors.
Suggestive Evidence of Carcinogenic Potential
A small, and possibly not statistically significant,
increase in tumor incidence observed in a single
animal or human study that does not reach the weight
of evidence for the descriptor "Likely to Be
Carcinogenic to Humans." The study generally would
not be contradicted by other studies of equal quality in
the same population group or experimental system
PFOA data exceed this description. Statistically
significant increases in tumor incidence of multiple
tumor types were observed across several human and
animal toxicological studies.
A small increase in a tumor with a high background
rate in that sex and strain, when there is some but
insufficient evidence that the observed tumors may be
due to intrinsic factors that cause background tumors
and not due to the agent being assessed.
This description is not applicable to the tumor types
observed after PFOA exposure.
Evidence of a positive response in a study whose
power, design, or conduct limits the ability to draw a
confident conclusion (but does not make the study
fatally flawed), but where the carcinogenic potential is
strengthened by other lines of evidence (such as
structure-activity relationships)
PFOA data exceed this description. The studies from
which carcinogenicity data are available were
determined to be high or medium confidence during
study quality evaluation.
A statistically significant increase at one dose only,
but no significant response at the other doses and no
overall trend
PFOA data exceed this description. Increases in
kidney cancer in humans were statistically significant in
two exposure groups in one study {Vieira, 2013,
2919154} and there was a statistically significant
increasing trend across exposure quartiles in a second
study {Shearer, 2021, 7161466}. Increases in hepatic
and pancreatic tumors in male rats were observed in
multiple dose groups with a statistically significant trend
overall {NTP, 2020, 7330145}.
Notes: MOA = mode of action.
6.5 Health Outcomes with Evidence Integration Judgments of
Evidence Suggests Bordering on Evidence indicates
EPA evaluated sixteen non-cancer health outcomes as part of this assessment. In accordance with
recommendations from the SAB {U.S. EPA, 2022, 10476098} and the IRIS Handbook {U.S.
EPA, 2022, 10476098}, for both quantitative and qualitative analyses in the current assessment,
EPA prioritized health outcomes with either evidence demonstrating or evidence indicating
associations between PFOA exposure and adverse health effects. Health outcomes reaching these
tiers of judgment were the hepatic, immune, developmental, cardiovascular, and cancer
6-10
-------
DRAFT FOR PUBLIC COMMENT
March 2023
outcomes. Some other health outcomes were determined to have evidence suggestive of
associations between PFOA and adverse health effects as well as some characteristics associated
with the evidence indicates tier, and EPA made judgments on these health outcomes as described
below.
For PFOA, two health outcomes that had characteristics of both evidence suggests and evidence
indicates were the reproductive and endocrine outcomes. Endpoints relevant to these two health
outcomes had been previously considered for POD derivation in the Proposed Approaches to the
Derivation of a Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA)
(CASRN 335-67-1) in Drinking Water. However, upon further examination using the protocols
for evidence integration outlined in the PFOA Appendix and Section 2.1.5, EPA concluded that
the available epidemiological and animal toxicological evidence did not meet the criteria
necessary for subsequent quantitative dose-response analyses. Although these health outcomes
were not prioritized in the current assessment, based on the available data, EPA concluded that
PFOA exposure may cause adverse reproductive or endocrine effects.
Epidemiological studies considered for evidence integration for adverse endocrine effects
included many high and medium confidence studies. There was slight evidence to suggest human
endocrine toxicity, including associations between PFOA exposure and changes in serum
thyroxine (T4) in children, though there was considerable uncertainty in the results due to
inconsistencies across sexes and age groups and a limited number of studies. Animal
toxicological studies considered for evidence integration included 8 high or medium confidence
studies. Collectively, the animal evidence for an association between PFOA exposure and effects
on the endocrine system was considered moderate, based on observed alterations in thyroid and
adrenocortical hormone levels, increased thyroid gland weight, and increased thyroid follicular
cell hypertrophy. Overall, the available evidence was suggestive but not indicative of adverse
endocrine effects due to PFOA exposure. Therefore, EPA did not prioritize this health outcome
for dose-response modeling. See Appendix C for a detailed description of endocrine evidence
synthesis and integration.
Epidemiological studies of reproductive effects in males that were considered for evidence
integration included three medium confidence studies {Cui, 2020, 6833614; Lopez-Espinosa,
2016, 3859832; Petersen, 2018, 5080277} and one low confidence study {Di Nisio, 2019,
5080655}. Although there was slight evidence to suggest human male reproductive toxicity,
including for effects on testosterone levels and sperm parameters, the associations were
inconsistent and it was difficult to assess the impacts of the alterations. Animal toxicological
studies considered for evidence integration included 3 high confidence studies {Biegel, 2001,
673581; NTP, 2019, 5400977; NTP, 2020, 7330145} and 5 medium confidence studies
{Butenhoff, 2012, 2919192; Li, 2011, 1294081; Lu, 2016, 3981459; Song, 2018, 5079725;
Zhang, 2014, 2850230}. The available animal data provided slight evidence that exposure to
PFOA results in adverse effects to the male reproductive system, including changes to the testes
and epididymis. However, the evidence from animal studies was inconsistent. Therefore, this
health outcome was not prioritized for dose-response modeling. See Appendix C for a detailed
description of male reproductive evidence synthesis and integration.
Female reproductive epidemiological studies published since the 2016 HESD that were
considered for evidence integration included 1 high confidence study {Ding, 2020, 6833612}
and 10 medium confidence studies {Lum, 2017, 3858516; Crawford, 2017, 3859813; Wang,
6-11
-------
DRAFT FOR PUBLIC COMMENT
March 2023
2017, 3856459; Kim, 2020, 6833596; Timmermann, 2017, 3981439; Ernst, 2019, 5080529;
Wang, 2019, 5080598; Lopez-Espinosa, 2016, 3859832; Donley, 2019, 5381537; Romano, 2016,
3981728}. Although there was slight evidence to suggest human female reproductive toxicity,
including preeclampsia and gestational hypertension, there was conflicting evidence on altered
puberty onset and limited data suggesting reduced fertility and fecundity. The associations were
inconsistent across reproductive hormone parameters, and it was difficult to assess the adversity
of these alterations. Animal toxicological studies considered for evidence integration included 1
high confidence study {NTP, 2019, 5400977} and 3 medium confidence studies {Zhang, 2020,
6505878; Chen, 2017, 3981369; Butenhoff, 2012, 2919192}. The available animal data provided
slight evidence that exposure to PFOA can result in alterations in ovarian physiology and
hormonal parameters in adult female rodents following exposure to doses as low as 1 mg/kg/day.
However, as with the available epidemiological studies, the evidence from animal studies was
inconsistent. Therefore, this health outcome was not prioritized for dose-response modeling. See
Appendix C for a detailed description of female reproductive evidence synthesis and integration.
Similar adverse reproductive and endocrine effects have been observed among the family of
PFAS. For example, the developing fetus and thyroid were identified as targets following oral
exposure to PFBS {U.S. EPA, 2021, 7310530}, though the observed reproductive effects were
considered equivocal. Additionally, EPA's 2021 assessment of GenX chemicals identified the
reproductive system as a potential toxicological target {U.S. EPA, 2021, 9960186}.
Additionally, the draft IRIS Toxicological Review ofPFBA concluded that the available evidence
indicates that the observed thyroid effects were likely due to PFBA exposure {U.S. EPA, 2021,
10064222}. Given the similarities across PFAS, these findings support potential associations
between PFOA and reproductive and endocrine effects.
As the databases for endocrine and reproductive outcomes were suggestive of human health
effects resulting from PFOA exposure, they were not prioritized during the updated literature
review conducted in February 2022. However, EPA acknowledges that future studies of these
currently "borderline" associations could impact the strength of the association and the weight of
evidence for these health outcomes. The currently available studies indicate the potential for
endocrine and reproductive effects after PFOA exposure. Studies on endocrine and reproductive
health outcomes represent two important research needs.
6.6 Challenges and Uncertainty in Modeling
6.6.1 Modeling of Animal Internal Dosimetry
There are several limitations and uncertainties associated with using pharmacokinetic models in
general and estimating animal internal dosimetry. In this assessment, EPA utilized the
Wambaugh et al. (2013, 2850932) animal internal dosimetry model because it had availability of
model parameters across all species of interest, agreement with out-of-sample datasets (see
PFOA Appendix), and flexibility to implement life-course modeling (see Section 4.1.3.1).
However, there were some limitations to this approach.
First, posterior parameter distributions summarized in Table 4-3 for each sex/species
combination were determined using a single study. Therefore, uncertainty in these parameters
represents only uncertainty in fitting that single study; any variability between studies or
differences in study design were not accounted for in the uncertainty of these parameters.
6-12
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Second, issues with parameter identifiability for some sex/species combinations resulted in
substantial uncertainty for some parameters. For example, filtrate volume (Vfil) represents a
parameter with poor identifiability when determined using only serum data, due to lack of
sensitivity to serum concentrations (see PFOA Appendix). Measurements in additional matrices,
such as urine, would help inform this parameter and reduce the uncertainty reflected in the wide
confidence intervals of the posterior distribution. These parameters with wide posterior CIs
represent parameters that are not sensitive to the concentration-time datasets on which the model
was trained (see PFOA Appendix). However, these uncertain model parameters did not impact
the median prediction used for BMD modeling and simply demonstrate that the available data
are unable to identify all parameters across every species over the range of doses used for model
calibration. Finally, the model is only parameterized using adult, single dose, PFOA study
designs. Gestational and lactational PK modeling parameters were later identified from
numerous sources (Table 4-5) to allow for the modeling of these life stages, with a more detailed
description of the life course modeling in Section 4.1.3.1.3.
The Wambaugh et al. (2013, 2850932) model fit the selected PFOA developmental study data
well, though there are additional limitations to using this method to model developmental life
stages. First, perinatal fetal concentrations assume instantaneous equilibration across the placenta
and do not account for the possibility of active transporters mediating distribution to the fetus.
Second, clearance in the pup during lactation is assumed to be a first-order process governed by
a single half-life. At low doses, this assumption is in line with adult clearance, but it is unclear
how physiological changes during development impact the infant half-life. Finally, PFOA
concentrations in breast milk are assumed to partition passively from the maternal blood. This
assumption does not account for the presence of active transport in the mammary gland or time-
course changes for PFOA uptake to the milk. Despite these limitations, the incorporation of
model parameters related to developmental life stages is a significant improvement over the
model used in the 2016 PFOA HESD which did not implement life course modeling {U.S. EPA,
2016, 3603279}.
6.6.2 Modeling of Human Dosimetry
Uncertainties may stem from efforts to model human dosimetry. One limitation is that the
clearance parameter, which is a function of the measured half-life and Vd values, is difficult to
estimate in the human general population. Specifically for PFOA, the measurement of half-life is
hindered by slow excretion and ongoing exposure. Additionally, it is unclear whether some of
the variability in measured half-life values reflects actual variability in the population as opposed
to uncertainty in the measurement of the value.
In the Verner et al. (2016, 3299692) model, half-life, Vd, and hence clearance values are assumed
to be constant across ages and sexes. The excretion of PFOA in children and infants is not well
understood. The ontogeny of renal transporters, age-dependent changes in overall renal function,
and the amount of protein binding (especially in serum) could all play a role in PFOA excretion
and could vary between children and adults. It is even difficult to predict the overall direction of
change in excretion in children (higher or lower than in adults) without a clear understanding of
these age-dependent differences. Vd is also expected to be different in children. Children have a
higher body water content, which results in a greater distribution of hydrophilic chemicals to
tissues compared to blood in neonates and infants compared to adults {Fernandez, 2011,
9641878}. This is well known for pharmaceuticals, but PFOA is unlike most pharmaceuticals in
6-13
-------
DRAFT FOR PUBLIC COMMENT
March 2023
that it undergoes extensive protein interaction, such that its distribution in the body is driven
primarily by protein binding and active transport. Hence, it is difficult to infer the degree to
which increased body water content might impact the distribution of PFOA.
The updated half-life value was developed based upon a review of recent literature (see Section
3.3.1.4.5). Many half-life values have been reported for the clearance of PFOA in humans (see
PFOA Appendix). The slow excretion of PFOA requires measurement of a small change in
serum concentration over a long time; the difficulties associated with making these
measurements may represent one reason for the variance in reported values. Another challenge is
the ubiquity of PFOA exposure. Ongoing exposure will result in a positive bias in observed half-
life values if not considered {Russell, 2015, 2851185}. In studies that calculate the half-life in a
population with greatly decreased PFOA exposures, typically due to the end of occupational
exposure or the introduction of drinking water filtration, the amount of bias due to continuing
exposure will depend on the ratio of the prior and ongoing exposures. That is, for a given
ongoing exposure, a higher prior exposure may be less likely to overestimate half-life compared
to a lower prior exposure. However, a half-life value determined from a population with very
high exposure may not be informative of the half-life in typical exposure scenarios because of
non-linearities in PK that may occur due to the saturation of PFAS-protein interactions. This will
likely take the form of an under-estimation of the half-life that is relevant to lower levels, which
are more representative of the general population due to saturation of renal resorption and
increased urinary clearance in the study population. One probable example of this is the
elimination half-life of approximately 120 or 200 days reported by Dourson and Gadagbui
{2021, 9641867}, who analyzed a clinical trial with exposures to PFOA of between 50 and 1200
mg weekly for a period of 6 weeks. In this study, the average plasma level after 6 weeks ranged
from 34 ug/mL at 0.1 mg/kg/day to 492.7 ug/mL at 2.3 mg/kg/day {Dourson, 2019, 6316919}.
This is orders of magnitude greater than the blood levels seen in the general population (the
NHANES 2007-08 95th percentile serum PFOA concentration was 9.7 ng/mL {Kato, 2011,
1290883}) and is in the range of the maximum values seen at the peak of PFOA manufacturing
{Post, 2012, 1290868}. The high exposure and short follow-up time may be the source of the
short half-life observed in this population. In addition, this study was also carried out in patients
with advanced cancer, which may have an effect on the rate of PFOA excretion.
A recent review publication {Campbell, 2022, 10492319} addressing the variation in reported
half-life values for PFOA promoted a half-life value of 1.3 years, based on the authors' analysis
of half-life values estimated from paired blood and urine samples {Zhang, 2013, 3859849}. The
rationale for this was the exclusion of studies that may be biased upward by ongoing exposure,
and studies that did not analyze linear and branched isomers of PFOA separately. A commentary
in response to the review disputed this conclusion and the approach used to make it {Post, 2022,
10492320}. The authors pointed out two citations that explore the effect of explicitly correcting
for background exposure: Russell et al. (2015, 2851185) and Bartell (2012, 2919207). Both
estimated half-lives >2 years after accounting for ongoing exposure. They go on to list several
high-quality studies that estimated half-lives much longer than the value calculated from Zhang
et al. (2013, 3859849). They also pointed out methodological limitations of Zhang et al. (2013,
3859849) and noted that another estimate of renal clearance using the same approach resulted in
a considerably different value {Gao, 2015, 2850134}. EPA is aware of two other studies
estimating renal clearance of PFOA from measurements in urine, and both estimated longer half-
lives than the value calculated by Zhang et al. (2013, 3859849). Fu et al. (2016, 3859819)
6-14
-------
DRAFT FOR PUBLIC COMMENT
March 2023
estimated a half-life of 4.1 years and Fujii et al. (2015, 2816710) estimated a renal clearance
value of 0.044 mL/kg/day, equivalent to a half-life of 7.3 years. These additional measurements
of PFOA half-life using a similar study design show that Campbell et al. (2022, 10492319)
selected an outlier study, both from other urinary clearance studies and from direct-observation
studies.
Another factor EPA considered when evaluating Zhang et al. (2013, 3859849) was that the
estimated value for the half-life of PFOS, geometric mean of 5.8 years for young females and 18
years for males and older females, is higher than is typically estimated. This result for PFOS
illustrates that there are uncertainties in any single estimate. Campbell et al. (2022, 10492319)
selected an outlier study for the half-life of PFOA, both from other urinary clearance studies and
from direct-observation studies. The range of results from among various studies represents a
range of uncertainty and EPA has chosen a half-life based on study quality (i.e., representative
population, environmentally relevant exposure, and multiple sampling of each individual) that
results in a value intermediate among the published estimates.
There are few reported Vd values for humans because this parameter requires knowledge of the
total dose or exposure, and Vd values are difficult to determine from environmental exposures. In
addition to the Vd reported by Thompson et al. (2010, 5082271), which was selected by EPA for
model parameterization, Dourson and Gadagbui (2021, 9641867) reported a human Vd of
91 mL/kg from a clinical trial on PFOA. This value is much lower than other reported values
across mammalian species and may reflect an earlier initial distribution step rather than the
distribution observed after chronic exposure. Chronic exposure may result in a greater
distribution to tissues relative to the plasma, and this process may be slowed by extensive
binding of PFOA to plasma proteins. Additionally, the exposure levels used in the clinical trial
are much higher than typically seen in the general population, which could result in a different
distribution profile.
Lastly, the description of breastfeeding in the updated Verner et al. (2016, 3299692) model relied
on a number of assumptions: that infants were exclusively breastfed for one year, that there was
a constant relationship between maternal serum and breastmilk PFOA concentrations, and that
weaning was an immediate process with the infant transitioning from a breastmilk-only diet to
the background exposure at one year. This is a relatively long duration of breastfeeding; only
27% of children in the U.S. are being breastfed at one year of age {CDC, 2013, 1936457}. Along
with using the 95th percentile of breastmilk consumption, this provides a scenario of high but
realistic lactational exposure. Lactational exposure to the infant is much greater than background
exposure, so the one-year breastfeeding duration is a conservative approach and will result in a
lower PODhed than a scenario with earlier weaning. Children in the U.S. are very unlikely to be
exclusively breastfed for up to one year, and this approach does not account for potential PFOA
exposure via the introduction to solid foods. However, since lactational exposure is much greater
than exposure after weaning, a breastfeeding scenario that does not account for potential PFOA
exposure from introduction of infants to solid foods is not expected to introduce substantial error.
6.6.3 Approach of Estimating a Benchmark Dose from a
Regression Coefficient
EPA identified several epidemiological studies (e.g., Steenland et al. (2009, 1291109), Darrow et
al. (2016, 3749173)) that reported associations between PFOA exposure and diseases or clinical
6-15
-------
DRAFT FOR PUBLIC COMMENT
March 2023
outcomes as regression coefficients. BMD modeling of regression coefficients results in a non-
traditional BMD, where the BMR is associated with a change in the regression coefficient of the
response variable rather than the measured biological response variable. As a result, there is
some uncertainty about the biological relevance of this non-traditional BMD associated with a
regression coefficient. However, as this regression coefficient is associated with a change in the
biological response variable, it is biological meaningful and EPA concluded that it can therefore
be used for POD derivation. EPA modeled these regression coefficients using the same approach
that EPA used to model for studies that reported measured response variables which is similar to
the approach followed by CalEPA in their draft Public Health Goal for PFOA {CalEPA, 2021,
9416932}.
To evaluate this potential uncertainty, EPA obtained the measured dose response data across
exposure deciles from Steenland et al. (2009, 1291109) (kindly provided to EPA on June 30,
2022 via email communication with the corresponding study author) and conducted sensitivity
analyses to compare BMDs produced by the reported regression coefficients with the measured
response variable (i.e., mean total cholesterol and odds ratios of elevated total cholesterol). For
PFOA, the analyses did not generate viable models and therefore the comparison could not be
made. These analyses are presented in detail in the PFOA Appendix.
For PFOS, however, BMDLs values estimated using the regression coefficient and using the
measured response variable were 9.52 ng/L and 26.39 ng/L, respectively. The two BMDL
estimates from the two approaches are within an order of magnitude, less than a 3-fold
difference, and the RfD allows for an order of magnitude (10-fold or 1,000%) uncertainty in the
estimate. Therefore, EPA is confident in its use regression coefficients as the basis of
PODHEDs.
6.7 Human Dosimetry Models: Consideration of Alternate
Modeling Approaches
PBPK models are typically preferred over a one-compartment approach because they can
provide individual tissue information and have a one-to-one correspondence with the biological
system that can be used to incorporate additional features of pharmacokinetics, including tissue-
specific internal dosimetry and local metabolism. In addition, though PBPK models are more
complex than one-compartment models, many of the additional parameters are chemical-
independent and have widely accepted values. Even some of the chemical-dependent values can
be extrapolated from animal toxicological studies when parameterizing a model for humans,
where data are typically scarcer.
The decision to select a non-physiologically based model as opposed to one of the PBPK models
was influenced in part by past issues identified during evaluation of the application of PBPK
models to other PFAS for the purpose of risk assessment. During the process of adapting a
published PBPK model for EPA needs, models are subjected to an extensive EPA internal QA
review. During initial review of the Loccisano family of models {Loccisano, 2011, 787186;
Loccisano, 2012, 1289830; Loccisano, 2012, 1289833; Loccisano, 2013, 1326665}, an unusual
implementation of PFOA plasma binding appeared to introduce a mass balance error. Due to the
stated goal of minimizing new model development (see Section 4.1.3.2), EPA did not pursue
resolution of the discrepancies, which would have required modifications to one of these models
for application in this assessment.
6-16
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Due to the previous issues that EPA encountered for other PFAS when implementing PBPK
models and the known issue with the Loccisano model and the models based upon it, EPA
selected a one-compartment model because it was the most robust available approach for this
effort. Based on the consideration and analysis of different models, EPA concluded that a one-
compartment model is sufficient to predict blood (or serum/plasma) concentrations.
Serum/plasma is a good biomarker for exposure, because a major proportion of the PFOA in the
body is found in serum/plasma due to albumin binding {Forsthuber, 2020, 6311640}. There were
no other specific tissues that were considered essential to describing the dosimetry of PFOA.
The two one-compartment approaches identified in the literature for PFOA was the model of
Verner et al. (2016, 3299692) and the model developed by the Minnesota Department of Health
(MDH model) {Goeden, 2019, 5080506}. These two models are structurally very similar, with a
single compartment each for mother and child, first-order excretion from those compartments,
and a similar methodology for describing lactational transfer from mother to child. The following
paragraphs describe the slight differences in model implementations, but it is first worth
emphasizing the similarity in the two approaches.
One advantage of the Verner model is that it explicitly models the mother from birth through the
end of breastfeeding. The MDH model, however, is limited to predictions for the time period
after the birth of the child with maternal levels set to an initial steady-state level. An explicit
description of maternal blood levels allows for the description of accumulation in the mother
prior to pregnancy followed by decreasing maternal levels during pregnancy, as has been
observed for serum PFOA in serial samples from pregnant women {Glynn, 2012, 1578498}.
This decrease occurs due to the relatively rapid increase in body weight during pregnancy
(compared to the years preceding pregnancy) and the increase in blood volume that occurs to
support fetal growth { Sibai, 1995, 1101373}. Detailed modeling of this period is important for
dose metrics based on maternal levels during pregnancy, especially near term, and on cord blood
levels.
Another distinction of the Verner model is that it is written in terms of rates of change in mass
rather than concentrations, as in the MDH model. This approach includes the effect of dilution of
PFOA during childhood growth without the need for an explicit term in the equations. Not
accounting for growth will result in the overprediction of serum concentrations in individuals
exposed during growth. Despite this, PFOA concentration in infants at any specific time is driven
more by recent lactational exposure than by earlier exposure (either during pregnancy or early
breastfeeding), which tends to minimize the impact of growth dilution. Additionally, this
structural consideration best matches the approach taken in our animal model, presenting a
harmonized approach. These structural considerations favor the application of the updated
Verner model over the MDH model.
EPA evaluated two other factors that were present in the MDH model: the application of a
scaling factor to increase the Vd in children and the treatment of exposure as a drinking water
intake rather than a constant exposure relative to bodyweight. After testing these features within
the updated Verner model structure, EPA determined that neither of these features were
appropriate for this assessment, primarily because they did not meaningfully improve the
comparison of model predictions to validation data.
6-17
-------
DRAFT FOR PUBLIC COMMENT
March 2023
In the MDH model, Vd in children starts at 2.4 times the adult Vd and decreases relatively
quickly to 1.5 times the adults Vd between 6 and 12 months, reaching the adult level at 10 years
of age. These scaling values originated from measurements of body water content relative to
weight compared to the adult value. There is no chemical-specific information to suggest that Vd
is larger in children compared to adults for PFOA. However, it is generally accepted in
pharmaceutical research that hydrophilic chemicals have greater Vd in children {Batchelor, 2015,
3223516}, which is attributed to increased body water. Still, PFOA is amphiphilic, not simply
hydrophilic, and its distribution is driven by interactions with binding proteins and transporters,
not by passive diffusion with body water. While it is plausible that Vd is larger in children, it is
unknown to what degree.
Since increased Vd in children is plausible but neither supported nor contradicted by direct
evidence, EPA evaluated the effect of variable Vd by implementing this change the updated
Verner model and comparing the results with constant and variable Vd (see PFOA Appendix).
This resulted in reduced predictions of serum concentrations, primarily during their peak in early
childhood. The model with variable Vd did not decrease the average relative error or the average
absolute value of relative error compared to the model with constant Vd (with PFOA and PFOS
results combined). Since the model with constant Vd had marginally better performance and was
an overall simpler solution, EPA did not implement variable Vd in the application of the model
for PODhed calculation.
The other key difference between the MDH model and the updated Verner model is that instead
of constant exposure relative to body weight, exposure in the MDH model was based on drinking
water consumption, which is greater relative to bodyweight in young children compared to
adults. Drinking water consumption is also greater in lactating women. To evaluate the potential
impact of calculating a drinking water concentration directly, bypassing the RfD step, EPA
implemented drinking water consumption in the modified Verner model (See PFOA Appendix).
EPA evaluated this decision for PFOA and PFOS together because the choice of units used for
human exposure represents a substantial difference in risk assessment methodology. For reasons
explained below, EPA ultimately decided to continue to calculate an RfD in terms of constant
exposure, with an MCLG calculated thereafter using life-stage specific drinking water
consumption values.
When comparing exposure based on drinking water consumption to the traditional RfD
approach, the impact on the serum concentrations predicted by the updated Verner model
differed between PFOA and PFOS. For PFOA, the predicted serum concentration in the child
was qualitatively similar, with the main effect seen in overprediction of timepoints that occur
later in childhood. These timepoints are more susceptible to changes in exposure, as early
childhood exposure is dominated by lactational exposure. Lactational exposure is slightly
increased in this scenario, because of increased drinking water consumption during lactation.
However, the main source of PFOA or PFOS in breastmilk in the model with exposure based on
drinking water consumption is that which accumulated over the mother's life prior to childbirth,
not that which was consumed during lactation. For PFOS, the increased exposure predicted
based on children's water intake results in much greater levels in later childhood compared to the
model with constant exposure relative to bodyweight. Use of water ingestion rates to adjust for
dose in the Verner model fails to match the decrease in PFOS concentration present in the
reported data with multiple timepoints and overestimates the value for the Norwegian Mother,
6-18
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Father, and Child Cohort Study (MoBa) cohort with a single timepoint. There is a much greater
effect on PFOS model results relative to PFOA. This comparison suggests that incorporating
variations in drinking water exposure in this way is not appropriate for the updated Verner
model.
In addition to the comparison with reported data, EPA's decision to use the Verner model was
also considered in the context of the effect on the derivation of MCLGs. The epidemiological
endpoints can be placed into three categories based on the age of the individuals: adults,
children, and pregnant women. Because increased drinking water exposure is only applied to
children and lactating women, the group of endpoints in children are the only ones that would be
affected. While the RfD estimated using the updated Verner model assumed constant exposure,
the MCLG is an algebraic calculation that incorporates the RfD, RSC, and drinking water intake.
The drinking water intake used for the MCLG calculation is chosen based on the target
population relevant to the critical effect that serves as the basis of the RfD. Therefore, even if the
RfD does not incorporate increased drinking water intake in certain lifestages, the subsequent
MCLG calculation does take this into account. Furthermore, the derivation of an RfD is useful
for general assessment of risk and not limited to drinking water exposure.
For these reasons and based on EPA's analyses, EPA determined that the updated Verner model
was the most appropriate model structure for PODhed calculation for PFOA. Including the
determination that assuming Vd in children equal to the adult values was appropriate, and that
calculating an RfD assuming a constant dose (mg/kg/day) was appropriate for this assessment.
6.8 Sensitive Populations
Some populations may be more susceptible to the potential adverse health effects of toxic
substances such as PFOA. These potentially susceptible populations include populations
exhibiting a greater response than others despite similar PFOA exposure due to increased
biological sensitivity, as well as populations exhibiting a greater response due to higher PFOA
exposure and/or exposure to other chemicals or non-chemical stressors. Populations with greater
biological sensitivity may include pregnant women and their developing fetuses, lactating
women, the elderly, and people with certain underlying medical conditions (see Section 6.8.1).
Additionally, some available data indicates that there may be sex-specific differences in
sensitivity to potential effects of PFOA (see Section 6.8.2). Populations that could exhibit a
greater response to PFOA exposure due to higher exposures to PFOA or other chemicals include
communities overburdened by chemical exposures or nonchemical stressors such as communities
with environmental justice concerns (see Section 6.8.3).
The potential health effects after PFOA exposure have been evaluated in some sensitive
populations (e.g., pregnant women, children) and a small number of studies have assessed
differences in exposure to PFOA across populations to assess whether racial/ethnic or
socioeconomic differences are associated with greater PFOA exposure. However, the available
research on PFOA's potential impacts on sensitive populations is limited and more research is
needed. Health effects differences in sensitivity to PFOA exposure have not allowed for the
identification or characterization of all potentially sensitive subpopulations. This lack of
knowledge about susceptibility to PFOA represents a potential source of uncertainty in the
assessment of PFOA.
6-19
-------
DRAFT FOR PUBLIC COMMENT
March 2023
6.8.1 Fetuses, Infants, Children
One of the more well-studied sensitive populations to PFOA exposure is developing fetuses,
infants, and children. Both animal toxicological and epidemiological data suggest that the
developing fetus is particularly sensitive to PFOA-induced toxicity. As described in Section
3.4.4.1, results of some epidemiological studies indicate an association between PFOA exposure
during pregnancy and adverse birth outcomes such as low birth weight, and studies of PFOA
exposure during early childhood, which may also reflect in utero exposure, suggest an
association between PFOA exposure and effects on development, including immune system
development (Section 3.4.2.1). The available animal toxicological data lend support to these
findings; as described in Section 3.4.4.2, numerous studies in rodents report effects similar to
those seen in humans (e.g., decreased body weights in offspring exposed to PFOA during
gestation). Additionally, PFOA exposure to humans during certain life stages or exposure
windows (e.g., prenatal or early postnatal exposure windows) may be more consequential than
others, and these potentially different effects in different populations and/or exposure windows
have not been fully characterized. More research is needed to fully understand the specific
critical windows of exposure during development.
With respect to the decreased antibody production endpoint, children who have autoimmune
diseases (e.g., juvenile arthritis) or are taking medications that weaken the immune system would
be expected to be more likely to mount a low antibody response and would therefore represent
potentially susceptible populations for PFOA exposure. There are also concerns about declines in
vaccination status {Smith, 2011, 9642143; Bramer, 2020, 9642145} for children overall, and the
possibility that diseases which are considered eradicated (such as diphtheria or tetanus) could
return to the United States {Hotez, 2019, 9642144}. As noted by Dietert et al. (2010, 644213),
the risks of developing infectious diseases may increase if immunosuppression occurs in the
developing immune system.
6.8.2 Sex Differences
In humans, potential sex differences in the disposition of PFOA in the body, as well as in the
potential for adverse health effects in response to PFOA exposure, have not been fully
elucidated. With respect to sex differences in the development of adverse health effects in
response to PFOA exposure, the available epidemiological data are insufficient to draw
conclusions, although some studies reported sex differences (e.g., an association between PFOA
exposure and serum ALT in girls but not boys {Attanasio, 2019, 5412069; Mora, 2018,
4239224}). In some studies in rats, males appeared to be more sensitive to some effects than
females, even when females received much higher PFOA doses {Butenhoff, 2004, 1291063;
NTP, 2020, 7330145}.
With respect to ADME, research in humans indicates that PFOA half-lives in males are generally
longer than those in females {Fu, 2016, 3859819; Gomis, 2017, 3981280; Li, 2017, 4238434}.
Some animal studies (in rats in particular) show the same sex difference, but additional research
is needed to determine whether the underlying mechanisms identified in rats are relevant to
humans. Female rats have been shown to absorb PFOA faster than male rats {Kim, 2016,
3749289}, and PFOA may distribute to some compartments (i.e., liver cytosol) to a greater
extent in female rats compared to males {Han, 2005, 5081570}. Several studies have
demonstrated that female rats and rabbits eliminate PFOA from the body faster than males
6-20
-------
DRAFT FOR PUBLIC COMMENT
March 2023
{Hinderliter, 2006, 3749132; Hundley, 2006, 3749054; NTP, 2019, 5400977; Dzierlenga, 2019,
5916078}. These studies and others are further described in Section 3.3.1 and the PFOA
Appendix (Appendix B).
Several studies have been conducted to elucidate the cause of the sex difference in the
elimination of PFOA by rats {Kudo, 2002, 2990271; Cheng, 2006, 6551310; Hinderliter, 2006,
3749132}. Many of the studies have focused on the role of transporters in the kidney tubules,
especially the OATs and OATPs located in the proximal portion of the descending tubule
{Nakagawa, 2007, 2919370; Nakagawa, 2009, 2919342; Yang, 2009, 2919328; Yang, 2010,
2919288}. Generally, both in vivo and in vitro studies reported differences in renal transporters
that are regulated by sex hormones and show consistent results indicating increased resorption of
PFOA in male rats (see Section 3.3.1 and Appendix B). Hinderliter et al. (2006, 3749132) found
that a developmental change in renal transport occurs in rats between 3 and 5 weeks of age that
allows for expedited excretion of PFOA in females and an inverse development in males. When
considered together, the studies of the transporters suggest that female rats are efficient in
transporting PFOA across the basolateral and apical membranes of the proximal kidney tubules
into the glomerular filtrate, but male rats are not.
Although sex differences in rats have been relatively well studied, sex differences observed in
mice were less pronounced {Lau, 2006, 1276159; Lou, 2009, 2919359} and were actually
reversed in cynomolgus monkeys and hamsters {Butenhoff, 2004, 3749227; Hundley, 2006,
3749054}, indicating species-specific factors impacting elimination across sexes.
Although there is some evidence to suggest sex differences in humans exposed to PFOA, the
mechanisms for these potential differences have not been fully explored. For example,
postmenopausal females and adult males have longer PFOA elimination half-lives than
premenopausal adult females {Zhang, 2013, 3859849}. Partitioning to the placenta, amniotic
fluid, fetus, menstruation, and breast milk represent important routes of elimination in humans
and may account for some of the sex differences observed for blood and urinary levels of PFOA
by sex and age. It is unclear whether hormone-dependent renal transporters play an additional
role in the observed sex differences in PFOA half-life in humans. Additional research is needed
to further elucidate these sex differences and their implications, and to ascertain whether the sex
differences observed in some animal species are relevant to humans. This data gap represents a
source of uncertainty in the elucidation of the risks of PFOA to humans.
6.8.3 Other Susceptible Populations
As noted in the SAB PFAS review panel's final report {U.S. EPA, 2022, 10476098}, there is
uncertainty about whether there are susceptible populations, such as certain racial/ethnic groups,
that might be more sensitive to the health effects of PFOA exposure because of either greater
biological sensitivity or higher exposure to PFOA and/or other environmental chemicals.
Although some studies have evaluated differences in PFAS exposure levels across SES and
racial/ethnic groups (see Section 6.1), studies of differential health effects incidence and PFOA
exposure are limited. To fully address equity and environmental justice concerns about PFOA,
these data gaps regarding differential exposure and health effects after PFOA exposure need to
be addressed. In the development of the proposed PFAS NPDWR, EPA conducted an analysis to
evaluate potential environmental justice impacts of the proposed regulation (See Chapter 8 of the
Economic Analysis for the Proposed PFAS National Primary Drinking Water Regulation {U.S.
6-21
-------
DRAFT FOR PUBLIC COMMENT
March 2023
EPA, 2023, 10692765}). EPA acknowledges that exposure to PFOA, and PFAS in general, may
have a disproportionate impact on certain communities (e.g., low SES communities; tribal
communities; minority communities; communities in the vicinity of areas of historical PFOA
manufacturing and/or contamination) and that studies of these communities are high priority
research needs.
6-22
-------
DRAFT FOR PUBLIC COMMENT
March 2023
7 References
3M. (2000). Determination of serum half-lives of several fluorochemicals, Interim report #1, June 8, 2000
[TSCA Submission], In TSCA 8(e) Supplemental Notice for Sulfonate-based and Carboxylic-
based Fluorochemicals-DocketNumbers 8EHQ-1180-373; 8EHQ-1180-374; 8EHQ-0381- 0394;
8EHQ-0598-373. (8EHQ-80-373. 8EHQ-0302-00373. 89(811844Q). AR226-0611).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8568548
3M. (2002). Determination of serum half-lives of several fluorochemicals, Interim report #2, January 11,
2002 [TSCA Submission], In TSCA 8(e) Supplemental Notice for Sulfonate-based and
Carboxylic-based Fluorochemicals-DocketNumbers 8EHQ-1180-373; 8EHQ-1180-374; 8EHQ-
0381- 0394; 8EHQ-0598-373. (8EHQ-80-373. 8EHQ-0302-00373. 89(811844Q). AR226-1086).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6574114
3M Company. (2000). Voluntary Use and Exposure Information Profile Perfluorooctanic Acid and Salts;
U.S. EPA Administrative Record AR226-0595 [EPA Report]. Washington, D.C.: U. S.
Environmental Protection Agency.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419083
Abbott, BD; Wolf, CJ; Schmid, JE; Das, KP; Zehr, RD; Helfant, L; Nakayama, S; Lindstrom, AB;
Strynar, MJ; Lau, C. (2007). Perfluorooctanoic acid induced developmental toxicity in the mouse
is dependent on expression of peroxisome proliferator activated receptor-alpha. Toxicol Sci 98:
571-581. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1335452
Abdallah, MA; Wemken, N; Drage, DS; Tlustos, C; Cellarius, C; Cleere, K; Morrison, JJ; Daly, S;
Coggins, MA; Harrad, S. (2020). Concentrations of perfluoroalkyl substances in human milk
from Ireland: Implications for adult and nursing infant exposure. Chemosphere 246: 125724.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316215
Abdullah Soheimi, SS; Abdul Rahman, A; Abd Latip, N; Ibrahim, E; Sheikh Abdul Kadir, SH. (2021).
Understanding the impact of perfluorinated compounds on cardiovascular diseases and their risk
factors: A meta-analysis study [Review]. Int J Environ Res Public Health 18: 8345.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959584
Abe, T; Takahashi, M; Kano, M; Amaike, Y; Ishii, C; Maeda, K; Kudoh, Y; Morishita, T; Hosaka, T;
Sasaki, T; Kodama, S; Matsuzawa, A; Kojima, H; Yoshinari, K. (2017). Activation of nuclear
receptor CAR by an environmental pollutant perfluorooctanoic acid. Arch Toxicol 91: 2365-
2374. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981405
Abraham, K; Mielke, H; Fromme, H; Volkel, W; Menzel, J; Peiser, M; Zepp, F; Willich, SN; Weikert, C.
(2020). Internal exposure to perfluoroalkyl substances (PFASs) and biological markers in 101
healthy 1-year-old children: associations between levels of perfluorooctanoic acid (PFOA) and
vaccine response. Arch Toxicol 94: 2131-2147.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6506041
Agier, L; Basagana, X; Maitre, L; Granum, B; Bird, PK; Casas, M; Oftedal, B; Wright, J; Andrusaityte,
S; de Castro, M; Cequier, E; Chatzi, L; Donaire-Gonzalez, D; Grazuleviciene, R; Haug, LS;
Sakhi, AK; Leventakou, V; Mceachan, R; Nieuwenhuijsen, M; Petraviciene, I; Robinson, O;
Roumeliotaki, T; Sunyer, J; Tamayo-Uria, I; Thomsen, C; Urquiza, J; Valentin, A; Slama, R;
Vrijheid, M; Siroux, V. (2019). Early-life exposome and lung function in children in Europe: an
analysis of data from the longitudinal, population-based HELIX cohort. The Lancet Planetary
Health 3: e81-e92. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5043613
Ahrens, L. (2011). Polyfluoroalkyl compounds in the aquatic environment: a review of their occurrence
and fate [Review]. J Environ Monit 13: 20-31.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2657780
Ahrens, L; Shoeib, M; Harner, T, om; Lee, S; Guo, R, ui; Reiner, EJ. (2011). Wastewater Treatment Plant
and Landfills as Sources of Polyfluoroalkyl Compounds to the Atmosphere. Environ Sci Technol
45: 8098-8105. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2325317
7-1
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Aimuzi, R; Luo, K; Chen, Q; Wang, H; Feng, L; Ouyang, F; Zhang, J. (2019). Perfluoroalkyl and
polyfluoroalkyl substances and fetal thyroid hormone levels in umbilical cord blood among
newborns by prelabor caesarean delivery. Environ Int 130: 104929.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387078
Aimuzi, R; Luo, K; Huang, R; Huo, X; Nian, M; Ouyang, F; Du, Y; Feng, L; Wang, W; Zhang, J. (2020).
Perfluoroalkyl and polyfluroalkyl substances and maternal thyroid hormones in early pregnancy.
Environ Pollut 264: 114557.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6512125
Ait Bamai, Y; Goudarzi, H; Araki, A; Okada, E; Kashino, I; Miyashita, C; Kishi, R. (2020). Effect of
prenatal exposure to per- and polyfluoroalkyl substances on childhood allergies and common
infectious diseases in children up to age 7 years: The Hokkaido study on environment and
children's health. Environ Int 143: 105979.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833636
Alderete, TL; Jin, R; Walker, DI; Valvi, D; Chen, Z; Jones, DP; Peng, C; Gilliland, FD; Berhane, K;
Conti, DV; Goran, MI; Chatzi, L. (2019). Perfluoroalkyl substances, metabolomic profiling, and
alterations in glucose homeostasis among overweight and obese Hispanic children: A proof-of-
concept analysis. Environ Int 126: 445-453.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5080614
American Cancer Society. (2020). Cancer facts and figures 2019 key statistics about kidney cancer.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9642148
Ammitzboll. C; Bornsen, L; Petersen, ER; Oturai, AB; Sondergaard. HB; Grandjean, P; Sellebjerg, F.
(2019). Perfluorinated substances, risk factors for multiple sclerosis and cellular immune
activation. J Neuroimmunol 330: 90-95.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5080379
Andersen, ME; Butenhoff, JL; Chang, SC; Farrar, DG; Kennedy, GL; Lau, C; Olsen, GW; Seed, J;
Wallace, KB. (2008). Perfluoroalkyl acids and related chemistries—toxicokinetics and modes of
action. Toxicol Sci 102: 3-14.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3749214
Andersen, ME; Clewell, HJ; Tan, YM; Butenhoff, JL; Olsen, GW. (2006). Pharmacokinetic modeling of
saturable, renal resorption of perfluoroalkylacids in monkeys—probing the determinants of long
plasma half-lives. Toxicology 227: 156-164.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/818501
Anzai, N; Kanai, Y; Endou, H. (2006). Organic anion transporter family: current knowledge. 100: 411-
426. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/9642039
Apelberg, BJ; Witter, FR; Herbstman, JB; Calafat, AM; Halden, RU; Needham, LL; Goldman, LR.
(2007). Cord serum concentrations of perfluorooctane sulfonate (PFOS) and perfluorooctanoate
(PFOA) in relation to weight and size at birth. Environ Health Perspect 115: 1670-1676.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/1290833
Arbuckle, TE; Macpherson, S; Foster, WG; Sathyanarayana, S; Fisher, M; Monnier, P; Lanphear, B;
Muckle, G; Fraser, WD. (2020). Prenatal Perfluoroalkyl Substances and Newborn Anogenital
Distance in a Canadian Cohort. Reprod Toxicol 94: 31-39.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6356900
Armstrong, LE; Guo, GL. (2019). Understanding environmental contaminants' direct effects on non-
alcoholic fatty liver disease progression. Curr Environ Health Rep 6: 95-104.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6956799
Arrebola, JP; Ramos, JJ; Bartolome, M; Esteban, M; Huetos, O; Canas, AI; Lopez-Herranz, A; Calvo, E;
Perez-Gomez, B; Castano, A; BIOAMBIENT.ES. (2019). Associations of multiple exposures to
persistent toxic substances with the risk of hyperuricemia and subclinical uric acid levels in
BIOAMBIENT.ES study. Environ Int 123: 512-521.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080503
Ashley-Martin, J; Dodds, L; Arbuckle, TE; Bouchard, MF; Fisher, M; Morriset, AS; Monnier, P; Shapiro,
7-2
-------
DRAFT FOR PUBLIC COMMENT
March 2023
GD; Ettinger, AS; Dallaire, R; Taback, S; Fraser, W; Piatt, RW. (2017). Maternal concentrations
of perfluoroalkyl substances and fetal markers of metabolic function and birth weight. Am J
Epidemiol 185: 185-193.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981371
Ashley-Martin, J; Dodds, L; Arbuckle, TE; Morisset, AS; Fisher, M; Bouchard, MF; Shapiro, GD;
Ettinger, AS; Monnier, P; Dallaire, R; Taback, S; Fraser, W. (2016). Maternal and Neonatal
Levels of Perfluoroalkyl Substances in Relation to Gestational Weight Gain. Int J Environ Res
Public Health 13: 146.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859831
ATSDR. (2021). Toxicological profile for perfluoroalkyls [ATSDR Tox Profile]. Atlanta, GA: U.S.
Department of Health and Human Services, Public Health Service.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9642134
Attanasio, R. (2019). Sex differences in the association between perfluoroalkyl acids and liver function in
US adolescents: Analyses of NHANES 2013-2016. Environ Pollut 254: 113061.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5412069
Avanasi, R; Shin, HM; Vieira, VM; Bartell, SM. (2016). Impacts of geocoding uncertainty on
reconstructed PFOA exposures and their epidemiological association with preeclampsia. Environ
Res 151: 505-512. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3981413
Avanasi, R; Shin, HM; Vieira, VM; Bartell, SM. (2016). Variability and epistemic uncertainty in water
ingestion rates and pharmacokinetic parameters, and impact on the association between
perfluorooctanoate and preeclampsia in the C8 Health Project population. Environ Res 146: 299-
307. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/3981510
Averina, M; Brox, J; Huber, S; Furberg, AS. (2021). Exposure to perfluoroalkyl substances (PFAS) and
dyslipidemia, hypertension and obesity in adolescents. The Fit Futures study. Environ Res 195:
110740. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/7410155
Averina, M; Brox, J; Huber, S; Furberg, AS; Sorensen. M. (2019). Serum perfluoroalkyl substances
(PFAS) and risk of asthma and various allergies in adolescents. The Tromso study Fit Futures in
Northern Norway. Environ Res 169: 114-121.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5080647
Aylward, LL; Hays, SM; Kirman, CR; Marchitti, SA; Kenneke, JF; English, C; Mattison, DR; Becker,
RA. (2014). Relationships of chemical concentrations in maternal and cord blood: a review of
available data [Review]. J Toxicol Environ Health B Crit Rev 17: 175-203.
https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/2920555
Bach, C; Matthiesen, B; Olsen; Henriksen, B. (2018). Conditioning on parity in studies of perfluoroalkyl
acids and time to pregnancy: an example from the danish national birth cohort. Environ Health
Perspect 126: 117003.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5080557
Bach, CC; Bech, BH; Nohr, EA; Olsen, J; Matthiesen, NB; Bossi, R; Uldbjerg, N; Bonefeld-Jorgensen,
EC; Henriksen, TB. (2015). Serum perfluoroalkyl acids and time to pregnancy in nulliparous
women. Environ Res 142: 535-541.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3981559
Balk, FGP; Winkens Piitz, K; Ribbenstedt, A; Gomis, MI; Filipovic, M; Cousins, IT. (2019). Children's
exposure to perfluoroalkyl acids - a modelling approach. Environ Sci Process Impacts 21: 1875-
1886. https://hero.epa.gOv/hero/index.cftn/reference/details/reference_id/5 918617
Bangma, J; Eaves, LA; Oldenburg, K; Reiner, JL; Manuck, T; Fry, RC. (2020). Identifying Risk Factors
for Levels of Per- and Polyfluoroalkyl Substances (PFAS) in the Placenta in a High-Risk
Pregnancy Cohort in North Carolina. Environ Sci Technol 54: 8158-8166.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6833725
Bao, WW; Qian, ZM; Geiger, SD; Liu, E; Liu, Y; Wang, SQ; Lawrence, WR; Yang, BY; Hu, LW; Zeng,
XW; Dong, GH. (2017). Gender-specific associations between serum isomers of perfluoroalkyl
substances and blood pressure among Chinese: Isomers of C8 Health Project in China. Sci Total
7-3
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Environ 607-608: 1304-1312.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860099
Barry, V; Winquist, A; Steenland, K. (2013). Perfluorooctanoic acid (PFOA) exposures and incident
cancers among adults living near a chemical plant. Environ Health Perspect 121: 1313-1318.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850946
Bartell, SM. (2012). Bias in half-life estimates using log concentration regression in the presence of
background exposures, and potential solutions. J Expo Sci Environ Epidemiol 22: 299-303.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919207
Bartell, SM; Calafat, AM; Lyu, C; Kato, K; Ryan, PB; Steenland, K. (2010). Rate of Decline in Serum
PFOA Concentrations after Granular Activated Carbon Filtration at Two Public Water Systems in
Ohio and West Virginia. Environ Health Perspect 118: 222-228.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/379025
Bartell, SM; Vieira, VM. (2021). Critical Review on PFOA, Kidney Cancer, and Testicular Cancer. J Air
Waste Manag Assoc. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7643457
Bassler, J; Ducatman, A; Elliott, M; Wen, S; Wahlang, B; Barnett, J; Cave, MC. (2019). Environmental
perfluoroalkyl acid exposures are associated with liver disease characterized by apoptosis and
altered serum adipocytokines. Environ Pollut 247: 1055-1063.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080624
Batchelor, HK; Marriott, JF. (2015). Paediatric pharmacokinetics: key considerations. Br J Clin
Pharmacol 79: 395-404.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3223516
Beach, SA; Newsted, JL; Coady, K; Giesy, JP. (2006). Ecotoxicological evaluation of
perfluorooctanesulfonate (PFOS) [Review]. Rev Environ Contam Toxicol 186: 133-174.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290843
Beck, IH; Timmermann, CAG; Nielsen, F; Schoeters, G; Johnk. C; Kyhl, HB; Host, A; Jensen, TK.
(2019). Association between prenatal exposure to perfluoroalkyl substances and asthma in 5-year-
old children in the Odense Child Cohort. Environ Health 18: 97.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5922599
Beesoon, S; Martin, JW. (2015). Isomer-Specific Binding Affinity of Perfluorooctanesulfonate (PFOS)
and Perfluorooctanoate (PFOA) to Serum Proteins. Environ Sci Technol 49: 5722-5731.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850292
Beesoon, S; Webster, GM; Shoeib, M; Harner, T, om; Benskin, JP; Martin, JW. (2011). Isomer Profiles
of Perfluorochemicals in Matched Maternal, Cord, and House Dust Samples: Manufacturing
Sources and Transplacental Transfer. Environ Health Perspect 119: 1659-1664.
https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/2050293
Beggs, KM; Mcgreal, S. R.; McCarthy, A; Gunewardena, S; Lampe, JN; Lau, C; Apte, U. (2016). The
role of hepatocyte nuclear factor 4-alpha in perfluorooctanoic acid- and perfluorooctanesulfonic
acid-induced hepatocellular dysfunction. Toxicol Appl Pharmacol 304: 18-29.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981474
Behr, AC; Kwiatkowski, A; Stahlman, M; Schmidt, FF; Luckert, C; Braeuning, A; Buhrke, T. (2020).
Impairment of bile acid metabolism by perfluorooctanoic acid (PFOA) and
perfluorooctanesulfonic acid (PFOS) in human HepaRG hepatoma cells. Arch Toxicol 94: 1673-
1686. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505973
Behr, AC; Plinsch, C; Braeuning, A; Buhrke, T. (2020). Activation of human nuclear receptors by
perfluoroalkylated substances (PFAS). Toxicol In Vitro 62: 104700.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6305866
Benskin, JP; De Silva, AO; Martin, LJ; Arsenault, G; McCrindle, R; Riddell, N; Mabury, SA; Martin,
JW. (2009). Disposition of perfluorinated acid isomers in Sprague-Dawley rats; part 1: single
dose. Environ Toxicol Chem 28: 542-554.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1617974
Benskin, JP; Muir, DCG; Scott, BF; Spencer, C; De Silva, AO; Kylin, H; Martin, JW; Morris, A;
7-4
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Lohmann, R; Tomy, G; Rosenberg, B; Taniyasu, S; Yamashita, N. (2012). Perfluoroalkyl Acids
in the Atlantic and Canadian Arctic Oceans. Environ Sci Technol 46: 5815-5823.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1274133
Berg, V; Nost, TH; Hansen, S; Elverland, A; Veyhe, AS; Jorde, R; Odland, J0; Sandanger, TM. (2015).
Assessing the relationship between perfluoroalkyl substances, thyroid hormones and binding
proteins in pregnant women; a longitudinal mixed effects approach. Environ Int 77: 63-69.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851002
Berk, M; Williams, LJ; Andreazza, A; Pasco, JA; Dodd, S; Jacka, FN; Moylan, S; Reiner, EJ; Magalhaes,
PVS. (2014). Pop, heavy metal and the blues: secondary analysis of persistent organic pollutants
(POP), heavy metals and depressive symptoms in the NHANES National Epidemiological
Survey. BMJ Open 4: e005142.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2713574
Bernstein, AS; Kapraun, DF; Schlosser, PM. (2021). A Model Template Approach for Rapid Evaluation
and Application of Physiologically Based Pharmacokinetic Models for Use in Human Health
Risk Assessments: A Case Study on Per- and Polyfluoroalkyl Substances. Toxicol Sci 182: 215—
228. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9639956
Biegel, LB; Hurtt, ME; Frame, S. R.; O'Connor, JC; Cook, JC. (2001). Mechanisms of extrahepatic tumor
induction by peroxisome proliferators in male CD rats. Toxicol Sci 60: 44-55.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/673581
Biegel, LB; Liu, RC; Hurtt, ME; Cook, JC. (1995). Effects of ammonium perfluorooctanoate on Leydig
cell function: in vitro, in vivo, and ex vivo studies. Toxicol Appl Pharmacol 134: 18-25.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1307447
Birnbaum, LS; Fenton, SE. (2003). Cancer and developmental exposure to endocrine disruptors [Review].
Environ Health Perspect 11: 389-394.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/197117
Bjorke-Monsen, AL; Varsi, K; Averina, M; Brox, J; Huber, S. (2020). Perfluoroalkyl substances (PFASs)
and mercury in never-pregnant women of fertile age: association with fish consumption and
unfavorable lipid profile. 3: 277-284.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/7643487
Blain, A; Tiwari, TSP. (2020). Chapter 16: Tetanus. In SW Roush; LM Baldy; MA Kirkconnell Hall
(Eds.), Manual for the Surveillance of Vaccine-Preventable Diseases. [Atlanta, GA]: Department
of Health and Human Services, Centers for Disease Control and Prevention, National Center for
Immunization and Respiratory Diseases.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642140
Blake, BE; Cope, HA; Hall, SM; Keys, RD; Mahler, BW; Mccord, J; Scott, B; Stapleton, HM; Strynar,
MJ; Elmore, SA; Fenton, SE. (2020). Evaluation of Maternal, Embryo, and Placental Effects in
CD-I Mice following Gestational Exposure to Perfluorooctanoic Acid (PFOA) or
Hexafluoropropylene Oxide Dimer Acid (HFPO-DA or GenX). Environ Health Perspect 128:
27006. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6305864
Blake, BE; Pinney, SM; Hines, EP; Fenton, SE; Ferguson, KK. (2018). Associations between longitudinal
serum perfluoroalkyl substance (PFAS) levels and measures of thyroid hormone, kidney function,
and body mass index in the Fernald Community Cohort. Environ Pollut 242: 894-904.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080657
Blomberg, AJ; Shih, YH; Messerlian, C; Jorgensen. LH; Weihe, P; Grandjean, P. (2021). Early-life
associations between per- and polyfluoroalkyl substances and serum lipids in a longitudinal birth
cohort. Environ Res 200: 111400.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8442228
Bloom, MS; Kannan, K; Spliethoff, HM; Tao, L; Aldous, KM; Vena, JE. (2010). Exploratory assessment
of perfluorinated compounds and human thyroid function. Physiol Behav 99: 240-245.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/757875
Bogdanska, J; Borg, D; Bergstrom, U; Mellring, M; Bergman, A; Depierre, J; Nobel, S. (2020). Tissue
7-5
-------
DRAFT FOR PUBLIC COMMENT
March 2023
distribution of 14C-labelled perfluorooctanoic acid in adult mice after 1-5 days of dietary
exposure to an experimental dose or a lower dose that resulted in blood levels similar to those
detected in exposed humans. Chemosphere 239: 124755.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315801
Bonefeld-Jorgensen, EC; Long, M; Bossi, R; Ayotte, P; Asmund, G; Kriiger, T; Ghisari, M; Mulvad, G;
Kern, P; Nzulumiki, P; Dewailly, E. (2011). Perfluorinated compounds are related to breast
cancer risk in Greenlandic Inuit: a case control study. Environ Health 10: 88.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2150988
Bonefeld-Jorgensen, EC; Long, M; Fredslund, SO; Bossi, R; Olsen, J. (2014). Breast cancer risk after
exposure to perfluorinated compounds in Danish women: a case-control study nested in the
Danish National Birth Cohort. Cancer Causes Control 25: 1439-1448.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851186
Boone, L; Meyer, D; Cusick, P; Ennulat, D; Bolliger, AP; Everds, N; Meador, V; Elliott, G; Honor, D;
Bounous, D; Jordan, H. (2005). Selection and interpretation of clinical pathology indicators of
hepatic injury in preclinical studies [Review]. Vet Clin Pathol 34: 182-188.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/782862
Borg, D; Ivarsson, J. (2017). Analysis of PFASs and TOF in Products. (TemaNord 2017:543). Nordic
Council of Ministers, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416541
Borghese, MM; Walker, M; Helewa, ME; Fraser, WD; Arbuckle, TE. (2020). Association of
perfluoroalkyl substances with gestational hypertension and preeclampsia in the MIREC study.
Environ Int 141: 105789.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833656
Botelho, SC; Saghafian, M; Pavlova, S; Hassan, M; Depierre, JW; Abedi-Valugerdi, M. (2015).
Complement activation is involved in the hepatic injury caused by high-dose exposure of mice to
perfluorooctanoic acid. Chemosphere 129: 225-231.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851194
Boulanger, B; Vargo, J; Schnoor, JL; Hornbuckle, KC. (2004). Detection of perfluorooctane surfactants
in Great Lakes water. Environ Sci Technol 38: 4064-4070.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289983
Bouwmeester, MC; Ruiter, S; Lommelaars, T; Sippel, J; Hodemaekers, HM; van den Brandhof, EJ;
Pennings, JL; Kamstra, JH; Jelinek, J; Issa, JP; Legler, J; van der Ven, LT. (2016). Zebrafish
embryos as a screen for DNA methylation modifications after compound exposure. Toxicol Appl
Pharmacol 291: 84-96.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3378942
Braissant, O; Wahli, W. (1998). Differential expression of peroxisome proliferator-activated receptor-
alpha, -beta, and -gamma during rat embryonic development. Endocrinology 139: 2748-2754.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/729555
Bramer, CA; Kimmins, LM; Swanson, R; Kuo, J; Vranesich, P; Jacques-Carroll, LA; Shen, AK. (2020).
Decline in Child Vaccination Coverage During the COVID-19 Pandemic - Michigan Care
Improvement Registry, May 2016-May 2020. 69: 630-631.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642145
Braun, JM; Chen, A; Romano, ME; Calafat, AM; Webster, GM; Yolton, K; Lanphear, BP. (2016).
Prenatal perfluoroalkyl substance exposure and child adiposity at 8 years of age: The HOME
study. Obesity (Silver Spring) 24: 231-237.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859836
Braun, JM; Kalkbrenner, AE; Just, AC; Yolton, K; Calafat, AM; Sjodin, A; Hauser, R; Webster, GM;
Chen, A; Lanphear, BP. (2014). Gestational exposure to endocrine-disrupting chemicals and
reciprocal social, repetitive, and stereotypic behaviors in 4- and 5-year-old children: the HOME
study. Environ Health Perspect 122: 513-520.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2345999
Brede, E; Wilhelm, M; Goen, T; Miiller, J; Rauchfuss, K; Kraft, M; Holzer, J. (2010). Two-year follow-
7-6
-------
DRAFT FOR PUBLIC COMMENT
March 2023
up biomonitoring pilot study of residents' and controls' PFC plasma levels after PFOA reduction
in public water system in Arnsberg, Germany. Int J Hyg Environ Health 213: 217-223.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859855
Brieger, A; Bienefeld, N; Hasan, R; Goerlich, R; Haase, H. (2011). Impact of perfluorooctanesulfonate
and perfluorooctanoic acid on human peripheral leukocytes. Toxicol In Vitro 25: 960-968.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937244
Brochot, C; Casas, M; Manzano-Salgado, C; Zeman, FA; Schettgen, T; Vrijheid, M; Bois, FY. (2019).
Prediction of maternal and foetal exposures to perfluoroalkyl compounds in a Spanish birth
cohort using toxicokinetic modelling. Toxicol Appl Pharmacol 379: 114640.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381552
Buck, CO; Eliot, MN; Kelsey, KT; Calafat, AM; Chen, A; Ehrlich, S; Lanphear, BP; Braun, JM. (2018).
Prenatal exposure to perfluoroalkyl substances and adipocytokines: the HOME Study. Pediatr Res
84: 854-860. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080288
Buck Louis, GM; Chen, Z; Schisterman, EF; Kim, S; Sweeney, AM; Sundaram, R; Lynch, CD; Gore-
Langton, RE; Barr, DB. (2015). Perfluorochemicals and human semen quality: The LIFE Study.
Environ Health Perspect 123: 57-63.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851189
Buck, RC; Franklin, J; Berger, U; Conder, JM; Cousins, IT; de Voogt, P; Jensen, AA; Kannan, K;
Mabury, SA; van Leeuwen, SP. (2011). Perfluoroalkyl and polyfluoroalkyl substances in the
environment: terminology, classification, and origins [Review]. Integr Environ Assess Manag 7:
513-541. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4771046
Buck, RC; Korzeniowski, SH; Laganis, E; Adamsky, F. (2021). Identification and classification of
commercially relevant per- and poly-fluoroalkyl substances (PFAS). Integr Environ Assess
Manag 17: 1045-1055.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9640864
Budtz-Jorgensen. E; Grandjean, P. (2018). Application of benchmark analysis for mixed contaminant
exposures: Mutual adjustment of perfluoroalkylate substances associated with immunotoxicity.
PLoS ONE 13: e0205388.
https ://hero .epa.gov/hero/index. cfm/reference/details/reference_id/5 083 631
Buekers, J; Colles, A; Cornelis, C; Morrens, B; Govarts, E; Schoeters, G. (2018). Socio-economic status
and health: evaluation of human biomonitored chemical exposure to per- and polyfluorinated
substances across status [Review]. Int J Environ Res Public Health 15: 2818.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080471
Buhrke, T; Kibellus, A; Lampen, A. (2013). In vitro toxicological characterization of perfluorinated
carboxylic acids with different carbon chain lengths. Toxicol Lett 218: 97-104.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2325346
Buhrke, T; Kriiger, E; Pevny, S; RoBlcr. M; Bitter, K; Lampen, A. (2015). Perfluorooctanoic acid (PFOA)
affects distinct molecular signalling pathways in human primary hepatocytes. Toxicology 333:
53-62. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850235
Bulka, CM; Avula, V; Fry, RC. (2021). Associations of exposure to perfluoroalkyl substances
individually and in mixtures with persistent infections: Recent findings from NHANES 1999-
2016. Environ Pollut 275: 116619.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7410156
Burkemper, JL; Aweda, TA; Rosenberg, AJ; Lunderberg, DM; Peaslee, GF; Lapi, SE. (2017).
Radiosynthesis and biological distribution of F-18-labeled perfluorinated alkyl substances.
Environ Sci Technol Lett 4: 211-215.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858622
Buser, MC; Scinicariello, F. (2016). Perfluoroalkyl substances and food allergies in adolescents. Environ
Int 88: 74-79. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859834
Butenhoff, J; Costa, G; Elcombe, C; Farrar, D; Hansen, K; Iwai, H; Jung, R; Kennedy, G; Lieder, P;
Olsen, G; Thomford, P. (2002). Toxicity of ammonium perfluorooctanoate in male cynomolgus
7-7
-------
DRAFT FOR PUBLIC COMMENT
March 2023
monkeys after oral dosing for 6 months. Toxicol Sci 69: 244-257.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1276161
Butenhoff, JL; Kennedy, GL; Chang, SC; Olsen, GW. (2012). Chronic dietary toxicity and
carcinogenicity study with ammonium perfluorooctanoate in Sprague-Dawley rats. Toxicology
298: 1-13. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919192
Butenhoff, JL; Kennedy, GL; Frame, S. R.; O'Connor, JC; York, RG. (2004). The reproductive
toxicology of ammonium perfluorooctanoate (APFO) in the rat. Toxicology 196: 95-116.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291063
Butenhoff, JL; Kennedy, GL; Hinderliter, PM; Lieder, PH; Jung, R; Hansen, KJ; Gorman, GS; Noker,
PE; Thomford, PJ. (2004). Pharmacokinetics of perfluorooctanoate in cynomolgus monkeys.
Toxicol Sci 82: 394-406.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749227
Butenhoff, JL; Kennedy, GL; Jung, R; Chang, SC. (2014). Evaluation of perfluorooctanoate for potential
genotoxicity. Toxicol Rep 1: 252-270.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079860
Butt, CM; Berger, U; Bossi, R; Tomy, GT. (2010). Levels and trends of poly- and perfluorinated
compounds in the arctic environment [Review]. Sci Total Environ 408: 2936-2965.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1291056
Byrne, S; Seguinot-Medina, S; Miller, P; Waghiyi, V; von Hippel, FA; Buck, CL; Carpenter, DO. (2017).
Exposure to polybrominated diphenyl ethers and perfluoroalkyl substances in a remote population
of Alaska Natives. Environ Pollut 231: 387-395.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/4165183
Byrne, SC; Miller, P; Seguinot-Medina, S; Waghiyi, V; Buck, CL; von Hippel, FA; Carpenter, DO.
(2018). Exposure to perfluoroalkyl substances and associations with serum thyroid hormones in a
remote population of Alaska Natives. Environ Res 166: 537-543.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 079678
C8 Science Panel. (2012). C8 portable link reports.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642155
C8 Science Panel. (2012). C8 study results - Status reports. Available online at
http://www.c8sciencepanel.org/study_results.html 1430770
Cai, D; Li, QQ; Chu, C; Wang, SZ; Tang, YT; Appleton, AA; Qiu, RL; Yang, BY; Hu, LW; Dong, GH;
Zeng, XW. (2020). High trans-placental transfer of perfluoroalkyl substances alternatives in the
matched maternal-cord blood serum: Evidence from a birth cohort study. Sci Total Environ 705:
135885. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6318671
Calafat, AM; Kato, K; Hubbard, K; Jia, T; Botelho, JC; Wong, LY. (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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381304
Calafat, AM; Wong, LY; Kuklenyik, Z; Reidy, 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. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290899
CalEPA. (2021). Public Health Goals: Perfluorooctanoic Acid and Perfluorooctane Sulfonic Acid in
Drinking Water (First Public Review Draft ed.). California Environmental Protection Agency,
Office of Environmental Health Hazard Assessment, Pesticide and Environmental Toxicology
Branch, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416932
Campbell, J; Clewell, H; Cox, T; Dourson, M; Ethridge, S; Forsberg, N; Gadagbui, B; Hamade, A; Naidu,
R; Pechacek, N; Peixe, TS; Prueitt, R; Rachamalla, M; Rhomberg, L; Smith, J; Verma, N. (2022).
The Conundrum of the PFOA human half-life, an international collaboration. Regul Toxicol
Pharmacol 132: 105185.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10492319
7-8
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Campbell, S; Raza, M; Pollack, AZ. (2016). Perfluoroalkyl substances and endometriosis in US women in
NHANES 2003-2006. Reprod Toxicol 65: 230-235.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860110
Canova, C; Barbieri, G; Zare Jeddi, M; Gion, M; Fabricio, A; Dapra, F; Russo, F; Fletcher, T; Pitter, G.
(2020). Associations between perfluoroalkyl substances and lipid profile in a highly exposed
young adult population in the Veneto Region. Environ Int 145: 106117.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/7021512
Canova, C; Di Nisio, A; Barbieri, G; Russo, F; Fletcher, T; Batzella, E; DallaZuanna, T; Pitter, G.
(2021). PFAS Concentrations and Cardiometabolic Traits in Highly Exposed Children and
Adolescents. Int J Environ Res Public Health 18.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/10176518
Cao, T; Qu, A; Li, Z; Wang, W; Liu, R; Wang, X; Nie, Y; Sun, S; Zhang, X; Liu, X. (2021). The
relationship between maternal perfluoroalkylated substances exposure and low birth weight of
offspring: a systematic review and meta-analysis. Environ Sci Pollut Res Int 28: 67053-67065.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/9959525
Cao, Y; Ng, C. (2021). Absorption, distribution, and toxicity of per- and polyfluoroalkyl substances
(PFAS) in the brain: a review [Review]. Environ Sci Process Impacts 23: 1623-1640.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/9959613
Cardenas, A; Gold, DR; Hauser, R; Kleinman, KP; Hivert, MF; Calafat, AM; Ye, X; Webster, TF;
Horton, ES; Oken, E. (2017). Plasma concentrations of per- and polyfluoroalkyl substances at
baseline and associations with glycemic indicators and diabetes incidence among high-risk adults
in the Diabetes Prevention Program trial. Environ Health Perspect 125: 107001.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/4167229
Cardenas, A; Hivert, MF; Gold, DR; Hauser, R; Kleinman, KP; Lin, PD; Fleisch, AF; Calafat, AM; Ye,
X; Webster, TF; Horton, ES; Oken, E. (2019). Associations of perfluoroalkyl and polyfluoroalkyl
substances with incident diabetes and microvascular disease. Diabetes Care 42: 1824-1832.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/5381549
Cariou, R; Veyrand, B; Yamada, A; Berrebi, A; Zalko, D; Durand, S; Pollono, C; Marchand, P; Leblanc,
JC; Antignac, JP; Le Bizec, B. (2015). Perfluoroalkyl acid (PFAA) levels and profiles in breast
milk, maternal and cord serum of French women and their newborns. Environ Int 84: 71-81.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3859840
Carmichael, S; Abrams, B; Selvin, S. (1997). The pattern of maternal weight gain in women with good
pregnancy outcomes. Am J Public Health 87: 1984-1988.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1060457
Caron-Beaudoin, E; Ayotte, P; Laouan Sidi, EA; Simon, CoL; Nation, CoWLPF; Nutashkuan, CTKo;
Shipu, CoU; Gros-Louis McHugh, N; Lemire, M. (2019). Exposure to perfluoroalkyl substances
(PFAS) and associations with thyroid parameters in First Nation children and youth from Quebec.
Environ Int 128: 13-23.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 097914
Caserta, D; Ciardo, F; Bordi, G; Guerranti, C; Fanello, E; Perra, G; Borghini, F; La Rocca, C; Tait, S;
Bergamasco, B; Stecca, L; Marci, R; Lo Monte, G; Soave, I; Focardi, S; Mantovani, A;
Moscarini, M. (2013). Correlation of endocrine disrupting chemicals serum levels and white
blood cells gene expression of nuclear receptors in a population of infertile women. International
Journal of Endocrinology 2013: 510703.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2000966
Caserta, D; Pegoraro, S; Mallozzi, M; Di Benedetto, L; Colicino, E; Lionetto, L; Simmaco, M. (2018).
Maternal exposure to endocrine disruptors and placental transmission: a pilot study. Gynecol
Endocrinol 34: 1-4. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/4728855
Cavallini, G; Donati, A; Taddei, M; Bergamini, E. (2017). Peroxisomes proliferation and pharmacological
stimulation of autophagy in rat liver: evidence to support that autophagy may remove the
"older" peroxisomes. Mol Cell Biochem 431: 97-102.
7-9
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981367
Caverly Rae, JM; Frame, S. R.; Kennedy, GL; Butenhoff, JL; Chang, SC. (2014). Pathology review of
proliferative lesions of the exocrine pancreas in two chronic feeding studies in rats with
ammonium perfluorooctanoate. Toxicol Rep 1: 85-91.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079680
CDC. (2011). Tetanus surveillance — United States, 2001-2008. MMWR Morb Mortal Wkly Rep 60:
365-369. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9998272
CDC. (2013). Breastfeeding report card : United States 2013. Atlanta, GA.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/193 645 7
Cellesi, C; Michelangeli, C; Rossolini, GM; Giovannoni, F; Rossolini, A. (1989). Immunity to diphtheria,
six to 15 years after a basic three-dose immunization schedule. Journal of Biological
Standardization 17: 29-34.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642154
Chang, CJ; Barr, DB; Ryan, PB; Panuwet, P; Smarr, MM; Liu, K; Kannan, K; Yakimavets, V; Tan, Y;
Ly, V; Marsit, CJ; Jones, DP; Corwin, EJ; Dunlop, AL; Liang, D. (2022). Per- and
polyfluoroalkyl substance (PFAS) exposure, maternal metabolomic perturbation, and fetal growth
in African American women: A meet-in-the-middle approach. Environ Int 158: 106964.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959688
Chang, ET; Adami, HO; Boffetta, P; Cole, P; Starr, TB; Mandel, JS. (2014). A critical review of
perfluorooctanoate and perfluorooctanesulfonate exposure and cancer risk in humans [Review].
Crit Rev Toxicol 44 Suppl 1: 1-81.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850282
Chang, S; Parker, GA; Kleinschmidt, SE; Olsen, GW; Ley, CA; Taiwo, OA. (2020). A Pathology Review
of the Lower Gastrointestinal Tract in Relation to Ulcerative Colitis in Rats and Cynomolgus
Macaques Treated With Ammonium Perfluorooctanoate. Toxicol Patholl92623320911606.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6320656
Chang, SC; Noker, PE; Gorman, GS; Gibson, SJ; Hart, JA; Ehresman, DJ; Butenhoff, JL. (2012).
Comparative pharmacokinetics of perfluorooctanesulfonate (PFOS) in rats, mice, and monkeys.
Reprod Toxicol 33: 428-440.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289832
Chen, A; Jandarov, R; Zhou, L; Calafat, AM; Zhang, G; Urbina, EM; Sarac, J; Augustin, DH; Caric, T;
Bockor, L; Petranovic, MZ; Novokmet, N; Missoni, S; Rudan, P; Deka, R. (2019). Association of
perfluoroalkyl substances exposure with cardiometabolic traits in an island population of the
eastern Adriatic coast of Croatia. Sci Total Environ 683: 29-36.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387400
Chen, F; Yin, S; Kelly, BC; Liu, W. (2017). Isomer-specific transplacental transfer of perfluoroalkyl
acids: Results from a survey of paired maternal, cord sera, and placentas. Environ Sci Technol 51:
5756-5763. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859806
Chen, H; Wang, Q; Cai, Y; Yuan, R; Wang, F; Zhou, B. (2020). Investigation of the Interaction
Mechanism of Perfluoroalkyl Carboxylic Acids with Human Serum Albumin by Spectroscopic
Methods. Int J Environ Res Public Health 17.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6324256
Chen, MH; Ha, EH; Liao, HF; Jeng, SF; Su, YN; Wen, TW; Lien, GW; Chen, CY; Hsieh, WS; Chen, PC.
(2013). Perfluorinated compound levels in cord blood and neurodevelopment at 2 years of age.
Epidemiology 24: 800-808.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850933
Chen, Q; Zhang, X; Zhao, Y; Lu, W; Wu, J; Zhao, S; Zhang, J; Huang, L. (2019). Prenatal exposure to
perfluorobutanesulfonic acid and childhood adiposity: A prospective birth cohort study in
Shanghai, China. Chemosphere 226: 17-23.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080578
Chen, Y; Zhou, L; Xu, J; Zhang, L; Li, M; Xie, X; Xie, Y; Luo, D; Zhang, D; Yu, X; Yang, B; Kuang, H.
7-10
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(2017). Maternal exposure to perfluorooctanoic acid inhibits luteal function via oxidative stress
and apoptosis in pregnant mice. Reprod Toxicol 69: 159-166.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981369
Cheng, J; Fujimura, M; Zhao, W; Wang, W. (2013). Neurobehavioral effects, c-Fos/Jun expression and
tissue distribution in rat offspring prenatally co-exposed to MeHg and PFOA: PFOA impairs Hg
retention. Chemosphere 91: 758-764.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2304777
Cheng, W; Ng, CA. (2017). A permeability-limited physiologically based pharmacokinetic (PBPK)
model for perfluorooctanoic acid (PFOA) in male rats. Environ Sci Technol 51: 9930-9939.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981304
Cheng, W; Ng, CA. (2018). Predicting relative protein affinity of novel per- and polyfluoroalkyl
substances (PFASs) by an efficient molecular dynamics approach. Environ Sci Technol 52: 7972-
7980. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5024207
Cheng, X; Klaassen, CD. (2008). Critical role of PPAR-alpha in perfluorooctanoic acid- and
perfluorodecanoic acid-induced downregulation of Oatp uptake transporters in mouse livers.
Toxicol Sci 106: 37-45.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/758807
Cheng, X; Klaassen, CD. (2008). Perfluorocarboxylic acids induce cytochrome P450 enzymes in mouse
liver through activation of PPAR-alpha and CAR transcription factors. Toxicol Sci 106: 29-36.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850410
Cheng, X; Klaassen, CD. (2009). Tissue distribution, ontogeny, and hormonal regulation of xenobiotic
transporters in mouse kidneys. Drug Metab Dispos 37: 2178-2185.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4116789
Cheng, X; Maher, J; Lu, H; Klaassen, CD. (2006). Endocrine regulation of gender-divergent mouse
organic anion-transporting polypeptide (Oatp) expression. Mol Pharmacol 70: 1291-1297.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6551310
Christensen, KY; Maisonet, M; Rubin, C; Holmes, A; Calafat, AM; Kato, K; Flanders, WD; Heron, J;
Mcgeehin, MA; Marcus, M. (2011). Exposure to polyfluoroalkyl chemicals during pregnancy is
not associated with offspring age at menarche in a contemporary British cohort. Environ Int 37:
129-135. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290803
Christensen, KY; Raymond, M; Meiman, J. (2019). Perfluoroalkyl substances and metabolic syndrome.
Int J Hyg Environ Health 222: 147-153.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080398
Christensen, KY; Raymond, M; Thompson, BA; Anderson, HA. (2016). Perfluoroalkyl substances in
older male anglers in Wisconsin. Environ Int 91: 312-318.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858533
Christensen, KY; Raymond, MR; Thompson, BA; Anderson, HA. (2016). Fish consumption, levels of
nutrients and contaminants, and endocrine-related health outcomes among older male anglers in
Wisconsin. J Occup Environ Med 58: 668-675.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3350721
Christenson, B; Bottiger, M. (1986). Serological immunity to diphtheria in Sweden in 1978 and 1984.
Scand J Infect Dis 18: 227-233.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9978484
Chu, C; Zhou, Y; Li, QQ; Bloom, MS; Lin, S; Yu, YJ; Chen, D; Yu, HY; Hu, LW; Yang, BY; Zeng,
XW; Dong, GH. (2020). Are perfluorooctane sulfonate alternatives safer? New insights from a
birth cohort study. Environ Int 135: 105365.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315711
Cichoz-Lach, H; Michalak, A. (2014). Oxidative stress as a crucial factor in liver diseases [Review].
World J Gastroenterol 20: 8082-8091.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2996796
Clegg, ED; Cook, JC; Chapin, RE; Foster, PMD; Daston, GP. (1997). Leydig cell hyperplasia and
7-11
-------
DRAFT FOR PUBLIC COMMENT
March 2023
adenoma formation: Mechanisms and relevance to humans [Review]. Reprod Toxicol 11: 107-
121. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/224277
Cluett, R; Seshasayee, SM; Rokoff, LB; Rifas-Shiman, SL; Ye, X; Calafat, AM; Gold, DR; Coull, B;
Gordon, CM; Rosen, CJ; Oken, E; Sagiv, SK; Fleisch, AF. (2019). Per- and Polyfluoroalkyl
Substance Plasma Concentrations and Bone Mineral Density in Midchildhood: A Cross-Sectional
Study (Project Viva, United States). Environ Health Perspect 127: 87006.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412438
Cochran, RS. (2015) Evaluation of Perfluorinated Compounds in Sediment, Water, and Passive Samplers
Collected from the Barksdale Air Force Base. (Master's Thesis). Texas Tech University, [n.l.].
Retrieved from https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416545
Cohn, BA; La Merrill, MA; Krigbaum, NY; Wang, M; Park, JS; Petreas, M; Yeh, G; Hovey, RC;
Zimmermann, L; Cirillo, PM. (2020). In utero exposure to poly- and perfluoroalkyl substances
(PFASs) and subsequent breast cancer. Reprod Toxicol 92: 112-119.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412451
Collier, RJ. (1975). Diphtheria toxin: mode of action and structure. 39: 54-85.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642066
Cong, J; Chu, C; Li, QQ; Zhou, Y; Qian, ZM; Geiger, SD; Vaughn, MG; Zeng, XW; Liu, RQ; Hu, LW;
Yang, BY; Chen, G; Zeeshan, M; Sun, X; Xiang, M; Dong, GH. (2021). Associations of
perfluorooctane sulfonate alternatives and serum lipids in Chinese adults. Environ Int 155:
106596. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8442223
Convertino, M; Church, TR; Olsen, GW; Liu, Y; Doyle, E; Elcombe, CR; Barnett, AL; Samuel, LM;
Macpherson, IR; Evans, TRJ. (2018). Stochastic pharmacokinetic-pharmacodynamic modeling
for assessing the systemic health risk of perfluorooctanoate (pfoa). Toxicol Sci 163: 293-306.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080342
Conway, B; Innes, KE; Long, D. (2016). Perfluoroalkyl substances and beta cell deficient diabetes. J
Diabetes Complications 30: 993-998.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859824
Conway, BN; Badders, AN; Costacou, T; Arthur, JM; Innes, KE. (2018). Perfluoroalkyl substances and
kidney function in chronic kidney disease, anemia, and diabetes. Diabetes, Metabolic Syndrome
and Obesity 11: 707-716.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080465
Cook, JC; Murray, SM; Frame, SR; Hurtt, ME. (1992). Induction of Leydig cell adenomas by ammonium
perfluorooctanoate: a possible endocrine-related mechanism. Toxicol Appl Pharmacol 113: 209-
217. https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/13 06123
Cook, TM; Protheroe, RT; JM, H. (2001). Tetanus: a review of the literature. Br J Anaesth 87: 477-487.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642139
Cope, HA; Blake, BE; Love, C; McCord, J; Elmore, SA; Harvey, JB; Chappell, VA; Fenton, SE. (2021).
Latent, sex-specific metabolic health effects in CD-I mouse offspring exposed to PFOA or
HFPO-DA (GenX) during gestation. Emerging Contaminants 7: 219-235.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10176465
Corley, RA; Mendrala, AL; Smith, FA; Staats, DA; Gargas, ML; Conolly, RB; Andersen, ME; Reitz, RH.
(1990). Development of a physiologically based pharmacokinetic model for chloroform. Toxicol
Appl Pharmacol 103: 512-527.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10123
Cornelis, C; D'Hollander, W; Roosens, L; Covaci, A; Smolders, R; Van Den Heuvel, R; Govarts, E; Van
Campenhout, K; Reynders, H; Bervoets, L. (2012). First assessment of population exposure to
perfluorinated compounds in Flanders, Belgium. Chemosphere 86: 308-314.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2569108
Corton, JC; Cunningham, ML; Hummer, BT; Lau, C; Meek, B; Peters, JM; Popp, JA; Rhomberg, L;
Seed, J; Klaunig, JE. (2014). Mode of action framework analysis for receptor-mediated toxicity:
The peroxisome proliferator-activated receptor alpha (PPARa) as a case study [Review]. Crit Rev
7-12
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Toxicol 44: 1-49. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2215399
Costa, G; Sartori, S; Consonni, D. (2009). Thirty years of medical surveillance in perfluooctanoic acid
production workers. J Occup Environ Med 51: 364-372.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1429922
Costello, E; Rock, S; Stratakis, N; Eckel, SP; Walker, DI; Valvi, D; Cserbik, D; Jenkins, T; Xanthakos,
SA; Kohli, R; Sisley, S; Vasiliou, V; La Merrill, MA; Rosen, H; Conti, DV; Mcconnell, R;
Chatzi, L. (2022). Exposure to per- and Polyfluoroalkyl Substances and Markers of Liver Injury:
A Systematic Review and Meta-Analysis [Review]. Environ Health Perspect 130: 46001.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10285082
Cox, GG; Haigh, D; Hindley, RM; Miller, DJ; Moody, CJ. (1994). COMPETING O-H INSERTION
AND BETA-ELIMINATION IN RHODIUM CARBENOID REACTIONS - SYNTHESIS OF 2-
ALKOXY-3-ARYLPROPANOATES. Tetrahedron Lett 35: 3139-3142.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1201708
Crawford, NM; Fenton, SE; Strynar, M; Hines, EP; Pritchard, DA; Steiner, AZ. (2017). Effects of
perfluorinated chemicals on thyroid function, markers of ovarian reserve, and natural fertility.
Reprod Toxicol 69: 53-59.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859813
Crebelli, R; Caiola, S; Conti, L; Cordelli, E; De Luca, G; Dellatte, E; Eleuteri, P; Iacovella, N; Leopardi,
P; Marcon, F; Sanchez, M; Sestili, P; Siniscalchi, E; Villani, P. (2019). Can sustained exposure to
PFAS trigger a genotoxic response? A comprehensive genotoxicity assessment in mice after
subacute oral administration of PFOA and PFBA. Regul Toxicol Pharmacol 106: 169-177.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381564
Creton, S; Aardema, MJ; Carmichael, PL; Harvey, JS; Martin, FL; Newbold, RF; O'Donovan, MR; Pant,
K; Poth, A; Sakai, A; Sasaki, K; Scott, AD; Schechtman, LM; Shen, RR; Tanaka, N; Yasaei, H.
(2012). Cell transformation assays for prediction of carcinogenic potential: State of the science
and future research needs. Mutagenesis 27: 93-101.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8803671
Crissman, JW; Goodman, DG; Hildebrandt, PK; Maronpot, RR; Prater, DA; Riley, JH; Seaman, WJ;
Thake, DC. (2004). Best practices guideline: Toxicologic histopathology. Toxicol Pathol 32: 126-
131. https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/51763
Cropp, C, .D.; Komori, T, .; Shima, J, .E.; Urban, T, .J.; Yee, S, .W.; More, S, .S.; Giacomini, K, .M.
(2008). Organic anion transporter 2 (SLC22A7) is a facilitative transporter of cGMP. 73: 1151-
1158. https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/9641964
Crump, KS. (1995). Calculation of benchmark doses from continuous data. Risk Anal 15: 79-89.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2258
Cui, L; Liao, CY; Zhou, QF; Xia, TM; Yun, ZJ; Jiang, GB. (2010). Excretion of PFOA and PFOS in male
rats during a subchronic exposure. Arch Environ Contam Toxicol 58: 205-213.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919335
Cui, L; Zhou, QF; Liao, CY; Fu, JJ; Jiang, GB. (2009). Studies on the toxicological effects of PFOA and
PFOS on rats using histological observation and chemical analysis. Arch Environ Contam
Toxicol 56: 338-349. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/757868
Cui, Q; Pan, Y; Wang, J; Liu, H; Yao, B; Dai, J. (2020). Exposure to per- and polyfluoroalkyl substances
(PFASs) in serum versus semen and their association with male reproductive hormones. Environ
Pollut 266 Pt.2: 115330.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833614
Cui, R; Li, C; Wang, J; Dai, J. (2019). Induction of hepatic miR-34a by perfluorooctanoic acid regulates
metabolism-related genes in mice. Environ Pollut 244: 270-278.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080384
Cui, Y; Liu, W; Xie, W; Yu, W; Wang, C; Chen, H. (2015). Investigation of the Effects of
Perfluorooctanoic Acid (PFOA) and Perfluorooctane Sulfonate (PFOS) on Apoptosis and Cell
Cycle in a Zebrafish (Danio rerio) Liver Cell Line. Int J Environ Res Public Health 12: 15673-
7-13
-------
DRAFT FOR PUBLIC COMMENT
March 2023
15682. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981517
D'Agostino, RB; Vasan, RS; Pencina, MJ; Wolf, PA; Cobain, M; Massaro, JM; Kannel, WB. (2008).
General cardiovascular risk profile for use in primary care: the Framingham Heart Study.
Circulation 117: 743-753.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10694408
Dairkee, SH; Luciani-Torres, G; Moore, DH; Jaffee, IM; Goodson, WH. (2018). A ternary mixture of
common chemicals perturbs benign human breast epithelial cells more than the same chemicals
do individually. Toxicol Sci 165: 131-144.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4563919
Dalla Zuanna, T; Savitz, DA; Barbieri, G; Pitter, G; Zare Jeddi, M; Dapra, F; Fabricio, ASC; Russo, F;
Fletcher, T; Canova, C. (2021). The association between perfluoroalkyl substances and lipid
profile in exposed pregnant women in the Veneto region, Italy. Ecotoxicol Environ Saf 209:
111805. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/7277682
Dalsager, L; Christensen, N; Halekoh, U; Timmermann, CAG; Nielsen, F; Kyhl, HB; Husby, S;
Grandjean, P; Jensen, TK; Andersen, HR. (2021). Exposure to perfluoroalkyl substances during
fetal life and hospitalization for infectious disease in childhood: A study among 1,503 children
from the Odense Child Cohort. Environ Int 149: 106395.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/7405343
Dalsager, L; Christensen, N; Husby, S; Kyhl, H; Nielsen, F; Host, A; Grandjean, P; Jensen, TK. (2016).
Association between prenatal exposure to perfluorinated compounds and symptoms of infections
at age l-4years among 359 children in the Odense Child Cohort. Environ Int 96: 58-64.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3858505
Darrow, LA; Groth, AC; Winquist, A; Shin, HM; Bartell, SM; Steenland, K. (2016). Modeled
perfluorooctanoic acid (PFOA) exposure and liver function in a mid-Ohio valley community.
Environ Health Perspect 124: 1227-1233.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3749173
Darrow, LA; Stein, CR; Steenland, K. (2013). Serum perfluorooctanoic acid and perfluorooctane
sulfonate concentrations in relation to birth outcomes in the Mid-Ohio Valley, 2005-2010.
Environ Health Perspect 121: 1207-1213.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2850966
Das, KP; Wood, CR; Lin, MT; Starkov, AA; Lau, C; Wallace, KB; Corton, JC; Abbott, BD. (2017).
Perfluoroalkyl acids-induced liver steatosis: Effects on genes controlling lipid homeostasis.
Toxicology 378: 37-52.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3859817
Dassuncao, C; Hu, XC; Nielsen, F; Weihe, P; Grandjean, P; Sunderland, EM. (2018). Shifting Global
Exposures to Poly- and Perfluoroalkyl Substances (PFASs) Evident in Longitudinal Birth Cohorts
from a Seafood-Consuming Population. Environ Sci Technol 52: 3738-3747.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4563862
Daston, GP; Kimmel, CA. (1998). An evaluation and interpretation of reproductive endpoints for human
health risk assessment. In An evaluation and interpretation of reproductive endpoints for human
health risk assessment. Washington, DC: ILSI Press.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3393032
Davies, B; Morris, T. (1993). Physiological parameters in laboratory animals and humans [Review].
PharmRes 10: 1093-1095.
https://hero .epa.gov/hero/index.cfin/reference/details/reference_id/1925 70
de Cock, M; de Boer, MR; Lamoree, M; Legler, J; van de Bor, M. (2014). Prenatal exposure to endocrine
disrupting chemicals in relation to thyroid hormone levels in infants - a Dutch prospective cohort
study. Environ Health 13: 106.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2718059
De Guise, S; Levin, M. (2021). Suppression of Th2 cytokines as a potential mechanism for reduced
antibody response following PFOA exposure in female B6C3F1 mice. Toxicol Lett 351: 155-162.
7-14
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959746
De Silva, AO; Armitage, JM; Bruton, TA; Dassuncao, C; Heiger-Bernays, W; Hu, XC; Karrman, A;
Kelly, B; Ng, C; Robuck, A; Sun, M; Webster, TF; Sunderland, EM. (2021). PFAS Exposure
Pathways for Humans and Wildlife: A Synthesis of Current Knowledge and Key Gaps in
Understanding [Review]. Environ Toxicol Chem 40: 631-657.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7542691
De Toni, L; Radu, CM; Sabovic, I; Di Nisio, A; DallAcqua, S; Guidolin, D; Spampinato, S; Campello, E;
Simioni, P; Foresta, C. (2020). Increased cardiovascular risk associated with chemical sensitivity
to perfluoro-octanoic acid: role of impaired platelet aggregation. International Journal of
Molecular Sciences 21: 399.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316907
Deener, KC; Sacks, JD; Kirrane, EF; Glenn, BS; Gwinn, MR; Bateson, TF; Burke, TA. (2018).
Epidemiology: a foundation of environmental decision making [Review]. J Expo Sci Environ
Epidemiol 28: 515-521.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6793519
Deji, Z; Liu, P; Wang, X; Zhang, X; Luo, Y; Huang, Z. (2021). Association between maternal exposure to
perfluoroalkyl and polyfluoroalkyl substances and risks of adverse pregnancy outcomes: A
systematic review and meta-analysis [Review]. Sci Total Environ 783: 146984.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7564388
Del Campo, JA; Gallego, P; Grande, L. (2018). Role of inflammatory response in liver diseases:
Therapeutic strategies [Review]. World Journal ofHepatology 10: 1-7.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365729
Dertinger, SD; Camphausen, K; Macgregor, JT; Bishop, ME; Torous, DK; Avlasevich, S; Cairns, S;
Tometsko, CR; Menard, C; Muanza, T; Chen, Y; Miller, RK; Cederbrant, K; Sandelin, K; Ponten,
I; Bolcsfoldi, G. (2004). Three-color labeling method for flow cytometric measurement of
cytogenetic damage in rodent and human blood. Environ Mol Mutagen 44: 427-435.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328871
Dewey, KG; Heinig, MJ; Nommsen, LA. (1993). Maternal weight-loss patterns during prolonged
lactation. Am J Clin Nutr 58: 162-166.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1335605
Dewitt, J; Cox, K; Savitz, D. (2019). Health based drinking water value recommendations for PFAS in
Michigan. Lansing, MI: Michigan Science Advisory Work Group.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6982827
Dewitt, JC; Blossom, SJ; Schaider, LA. (2019). Exposure to per-fluoroalkyl and polyfluoroalkyl
substances leads to immunotoxicity: epidemiological and toxicological evidence [Review]. J
Expo Sci Environ Epidemiol 29: 148-156.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080663
Dewitt, JC; Copeland, CB; Luebke, RW. (2009). Suppression of humoral immunity by perfluorooctanoic
acid is independent of elevated serum corticosterone concentration in mice. Toxicol Sci 109: 106-
112. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937261
Dewitt, JC; Copeland, CB; Strynar, MJ; Luebke, RW. (2008). Perfluorooctanoic acid-induced
immunomodulation in adult C57BL/6J or C57BL/6N female mice. Environ Health Perspect 116:
644-650. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290826
Dewitt, JC; Keil, DE. (2017). Current issues in developmental immunotoxicity. In G Parker (Ed.),
Immunopathology in Toxicology and Drug Development: Volume 1, Immunobiology,
Investigative Techniques, and Special Studies (pp. 601-618). Totowa, NJ: Humana Press.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5926400
Dewitt, JC; Williams, WC; Creech, NJ; Luebke, RW. (2016). Suppression of antigen-specific antibody
responses in mice exposed to perfluorooctanoic acid: Role of PPARa and T- and B-cell targeting.
J Immunotoxicol 13: 38-45.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2851016
7-15
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Dhingra, R; Darrow, LA; Klein, M; Winquist, A; Steenland, K. (2016). Perfluorooctanoic acid exposure
and natural menopause: A longitudinal study in a community cohort. Environ Res 146: 323-330.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981508
Dhingra, R; Lally, C; Darrow, LA; Klein, M; Winquist, A; Steenland, K. (2016). Perfluorooctanoic acid
and chronic kidney disease: Longitudinal analysis of a Mid-Ohio Valley community. Environ Res
145: 85-92. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981521
Dhingra, R; Winquist, A; Darrow, LA; Klein, M; Steenland, K. (2017). A study of reverse causation:
Examining the associations of perfluorooctanoic acid serum levels with two outcomes. Environ
Health Perspect 125: 416-421.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981432
Di Nisio, A; Sabovic, I; Valente, U; Tescari, S; Rocca, MS; Guidolin, D; Dall'Acqua, S; Acquasaliente,
L; Pozzi, N; Plebani, M; Garolla, A; Foresta, C. (2019). Endocrine disruption of androgenic
activity by perfluoroalkyl substances: clinical and experimental evidence. J Clin Endocrinol
Metab 104: 1259-1271.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080655
Dietert, RR; Dewitt, JC; Germolec, DR; Zelikoff, JT. (2010). Breaking patterns of environmentally
influenced disease for health risk reduction: Immune perspectives [Review]. Environ Health
Perspect 118: 1091-1099.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/644213
Ding, N; Harlow, SD; Randolph, JF; Calafat, AM; Mukheijee, B; Batterman, S; Gold, EB; Park, SK.
(2020). Associations of perfluoroalkyl substances with incident natural menopause: The study of
women's health across the nation. J Clin Endocrinol Metab 105: E3169-E3182.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833612
Ding, N; Karvonen-Gutierrez, CA; Mukheijee, B; Calafat, AM; Harlow, SD; Park, SK. (2022). Per- and
Polyfluoroalkyl Substances and Incident Hypertension in Multi-Racial/Ethnic Women: The Study
of Women's Health Across the Nation. Hypertension 79:
101161HYPERTENSIONAHA12118 809.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328874
Ding, N; Park, SK. (2020). Perfluoroalkyl substances exposure and hearing impairment in US adults.
Environ Res 187: 109686.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6711603
Dinglasan-Panlilio, MJ; Prakash, SS; Baker, JE. (2014). Perfluorinated compounds in the surface waters
of Puget Sound, Washington and Clayoquot and Barkley Sounds, British Columbia. Mar Pollut
Bull 78: 173-180. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2545254
Dixon, LJ; Barnes, M; Tang, H; Pritchard, MT; Nagy, LE. (2013). Kupffer cells in the liver [Review].
Compr Physiol 3: 785-797.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365841
Domazet, SL; Grontved. A; Timmermann, AG; Nielsen, F; Jensen, TK. (2016). Longitudinal associations
of exposure to perfluoroalkylated substances in childhood and adolescence and indicators of
adiposity and glucose metabolism 6 and 12 years later: The European Youth Heart Study.
Diabetes Care 39: 1745-1751.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981435
Domazet, SL; Jensen, TK; Wedderkopp, N; Nielsen, F; Andersen, LB; Grontved, A. (2020). Exposure to
perfluoroalkylated substances (PFAS) in relation to fitness, physical activity, and adipokine levels
in childhood: The European youth heart study. Environ Res 191: 110110.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833700
Donat-Vargas, C; Bergdahl, IA; Tornevi, A; Wennberg, M; Sommar, J; Kiviranta, H; Koponen, J;
Rolandsson, O; Akesson, A. (2019). Perfluoroalkyl substances and risk of type II diabetes: A
prospective nested case-control study. Environ Int 123: 390-398.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5083542
Donat-Vargas, C; Bergdahl, IA; Tornevi, A; Wennberg, M; Sommar, J; Koponen, J; Kiviranta, H;
7-16
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Akesson, A. (2019). Associations between repeated measure of plasma perfluoroalkyl substances
and cardiometabolic risk factors. Environ Int 124: 58-65.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080588
Dong, GH; Tung, KY; Tsai, CH; Liu, MM; Wang, D; Liu, W; Jin, YH; Hsieh, WS; Lee, YL; Chen, PC.
(2013). Serum polyfluoroalkyl concentrations, asthma outcomes, and immunological markers in a
case-control study of Taiwanese children. Environ Health Perspect 121: 507-513, 513e501-508.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937230
Dong, Z; Wang, H; Yu, YY; Li, YB; Naidu, R; Liu, Y. (2019). Using 2003-2014 U.S. NHANES data to
determine the associations between per- and polyfluoroalkyl substances and cholesterol: Trend
and implications. Ecotoxicol Environ Saf 173: 461-468.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080195
Donley, GM; Taylor, E; Jeddy, Z; Namulanda, G; Hartman, TJ. (2019). Association between in utero
perfluoroalkyl substance exposure and anti-Miillerian hormone levels in adolescent females in a
British cohort. Environ Res 177: 108585.
https://hero.epa.gov/hero/index. cfm/reference/details/reference_id/5 381537
Dourson, M; Gadagbui, B. (2021). The Dilemma of perfluorooctanoate (PFOA) human half-life. 126:
105025. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641867
Dourson, ML; Gadagbui, B; Onyema, C; Mcginnis, PM; York, RG. (2019). Data derived Extrapolation
Factors for developmental toxicity: A preliminary research case study with perfluorooctanoate
(PFOA). Regul Toxicol Pharmacol 108: 104446.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316919
Dreyer, AF; Jensen, RC; Glintborg, D; Schmedes, AV; Brandslund, I; Nielsen, F; Kyhl, HB; Jensen, TK;
Andersen, MS. (2020). Perfluoroalkyl substance exposure early in pregnancy was negatively
associated with late pregnancy cortisone levels. J Clin Endocrinol Metab 105: E2834-E2844.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833676
Du, G; Huang, H; Hu, J; Qin, Y; Wu, D; Song, L; Xia, Y; Wang, X. (2013). Endocrine-related effects of
perfluorooctanoic acid (PFOA) in zebrafish, H295R steroidogenesis and receptor reporter gene
assays. Chemosphere 91: 1099-1106.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850983
Duan, Y; Sun, H; Yao, Y; Meng, Y; Li, Y. (2020). Distribution of novel and legacy per-/polyfluoroalkyl
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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5918597
Ducatman, A; Zhang, J; Fan, H. (2015). Prostate-specific antigen and perfluoroalkyl acids in the C8
health study population. J Occup Environ Med 57: 111-114.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3859843
Dufour, P; Pirard, C; Seghaye, MC; Charlier, C. (2018). Association between organohalogenated
pollutants in cord blood and thyroid function in newborns and mothers from Belgian population.
Environ Pollut 238: 389-396.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4354164
Dzierlenga, AL; Robinson, VG; Waidyanatha, S; Devito, MJ; Eifrid, MA; Gibbs, ST; Granville, CA;
Blystone, CR. (2019). Toxicokinetics of perfluorohexanoic acid (PFHxA), perfluorooctanoic acid
(PFOA) and perfluorodecanoic acid (PFDA) in male and female Hsd:Sprague dawley SD rats
following intravenous or gavage administration. Xenobiotica 50: 1-11.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5916078
Dzierlenga, M, .W.; Crawford, L, .; Longnecker, M, .P. (2020). Birth weight and perfluorooctane sulfonic
acid: a random-effects meta-regression analysis. Environmental Epidemiology 4: e095.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7643488
Dzierlenga, MW; Allen, BC; Clewell, HJ; Longnecker, MP. (2020). Pharmacokinetic bias analysis of an
association between clinical thyroid disease and two perfluoroalkyl substances. Environ Int 141:
105784. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833691
7-17
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Dzierlenga, MW; Allen, BC; Ward, PL; Clewell, HJ; Longnecker, MP. (2019). A model of functional
thyroid disease status overthe lifetime. PLoS ONE 14: e0219769.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7947729
Dzierlenga, MW; Moreau, M; Song, G; Mallick, P; Ward, PL; Campbell, JL; Housand, C; Yoon, M;
Allen, BC; Clewell, HJ; Longnecker, MP. (2020). Quantitative bias analysis of the association
between subclinical thyroid disease and two perfluoroalkyl substances in a single study. Environ
Res 182: 109017. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315786
East, A; Egeghy, PP; Hubal, E; Slover, R; Vallero, DA. (2021). Computational estimates of daily
aggregate exposure to PFOA/PFOS from 2011 to 2017 using a basic intake model. J Expo Sci
Environ Epidemiol Online ahead of print.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416543
Ebert, A; Allendorf, F; Berger, U; Goss, KU; Ulrich, N. (2020). Membrane/water partitioning and
permeabilities of perfluoroalkyl acids and four of their alternatives and the effects on
toxicokinetic behavior. Environ Sci Technol 54: 5051-5061.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505873
EFSA. (2020). Risk to human health related to the presence ofperfluoroalkyl substances in food. EFSA J
18. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6984182
Eggert, A; Cisneros-Montalvo, S; Anandan, S; Musilli, S; Stukenborg, JB; Adamsson, A; Nurmio, M;
Toppari, J. (2019). The effects of perfluorooctanoic acid (PFOA) on fetal and adult rat testis.
Reprod Toxicol 90: 68-76.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/5 381535
Ehresman, DJ; Froehlich, JW; Olsen, GW; Chang, SC; Butenhoff, JL. (2007). Comparison of human
whole blood, plasma, and serum matrices for the determination of perfluorooctanesulfonate
(PFOS), perfluorooctanoate (PFOA), and other fluorochemicals. Environ Res 103: 176-184.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1429928
Eick, SM; Enright, EA; Geiger, SD; Dzwilewski, KLC; DeMicco, E; Smith, S; Park, JS; Aguiar, S;
Woodruff, TJ; Morello-Frosch, R; Schantz, SL. (2021). Associations of maternal stress, prenatal
exposure to per- and polyfluoroalkyl substances (PFAS), and demographic risk factors with birth
outcomes and offspring neurodevelopment: An overview of the ECO.CA.IL prospective birth
cohorts [Review]. Int J Environ Res Public Health 18: 742.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/7510968
Elcombe, BM. Compositions Comprising Perfluorooctanoic Acid, (World Intellectual Property
Organization2013). https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10494295
Elcombe, CR; Elcombe, BM; Farrar, DG; Foster, JR. (2007). Characterization of ammonium
perfluorooctanoic acid (APFO) induced hepatomegaly in rats. Toxicology 240: 172-173.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5085376
Elcombe, CR; Elcombe, BM; Foster, JR; Farrar, DG; Jung, R; Chang, SC; Kennedy, GL; Butenhoff, JL.
(2010). Hepatocellular hypertrophy and cell proliferation in Sprague-Dawley rats following
dietary exposure to ammonium perfluorooctanoate occurs through increased activation of the
xenosensor nuclear receptors PPARa and CAR/PXR. Arch Toxicol 84: 787-798.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850034
Eldasher, LM; Wen, X; Little, MS; Bircsak, KM; Yacovino, LL; Aleksunes, LM. (2013). Hepatic and
renal Bcrp transporter expression in mice treated with perfluorooctanoic acid. Toxicology 306:
108-113. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850979
Elmore, SA; Dixon, D; Hailey, JR; Harada, H; Herbert, RA; Maronpot, RR; Nolte, T; Rehg, JE;
Rittinghausen, S; Rosol, TJ; Satoh, H; Vidal, JD; Willard-Mack, CL; Creasy, DM. (2016).
Recommendations from the INHAND Apoptosis/Necrosis Working Group. Toxicol Pathol 44.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10671182
EMEA. (2008). Non-clinical guideline on drug induced hepatotoxicity. (Doc. Ref.
EMEA/CHMP/SWP/150115/2006). London, UK.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/3 05 6793
7-18
-------
DRAFT FOR PUBLIC COMMENT
March 2023
EMEA. (2010). Reflection paper on non-clinical evaluation of drug-induced liver injury (DILI).
(EMEA/CHMP/SWP/150115/2006). London, UK.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3056796
Emmett, EA; Zhang, H; Shofer, FS; Freeman, D; Rodway, NV; Desai, C; Shaw, LM. (2006). Community
exposure to perfluorooctanoate: Relationships between serum levels and certain health
parameters. J Occup Environ Med 48: 771-779.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290905
Ericson, I; Gomez, M; Nadal, M; van Bavel, B; Lindstrom, G; Domingo, JL. (2007). Perfluorinated
chemicals in blood of residents in Catalonia (Spain) in relation to age and gender: a pilot study.
Environ Int 33: 616-623.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858652
Eriksen, KT; Raaschou-Nielsen, O; Mclaughlin, JK; Lipworth, L; Tjonncland. A; Overvad, K; Sorensen.
M. (2013). Association between plasma PFOA and PFOS levels and total cholesterol in a middle-
aged Danish population. PLoS ONE 8: e56969.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919150
Eriksen, KT; Sorensen, M; Mclaughlin, JK; Lipworth, L; Tjonneland, A; Overvad, K; Raaschou-Nielsen,
O. (2009). Perfluorooctanoate and perfluorooctanesulfonate plasma levels and risk of cancer in
the general Danish population. J Natl Cancer Inst 101: 605-609.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919344
Ernst, A; Brix, N; Lauridsen, LLB; Olsen, J; Parner, ET; Liew, Z; Olsen, LH; Ramlau-Hansen, CH.
(2019). Exposure to perfluoroalkyl substances during fetal life and pubertal development in boys
and girls from the danish national birth cohort. Environ Health Perspect 127: 17004.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080529
Erol, E; Kumar, LS; Cline, GW; Shulman, GI; Kelly, DP; Binas, B. (2004). Liver fatty acid binding
protein is required for high rates of hepatic fatty acid oxidation but not for the action of
PPARalpha in fasting mice. FASEB J 18: 347-349.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5212239
Eryasa, B; Grandjean, P; Nielsen, F; Valvi, D; Zmirou-Navier, D; Sunderland, E; Weihe, P; Oulhote, Y.
(2019). Physico-chemical properties and gestational diabetes predict transplacental transfer and
partitioning of perfluoroalkyl substances. Environ Int 130: 104874.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412430
Etzel, TM; Braun, JM; Buckley, JP. (2019). Associations of serum perfluoroalkyl substance and vitamin
D biomarker concentrations in NHANES, 2003-2010. Int J Hyg Environ Health 222: 262-269.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5043582
Fabrega, F; Kumar, V; Benfenati, E; Schuhmacher, M; Domingo, JL; Nadal, M. (2015). Physiologically
based pharmacokinetic modeling of perfluoroalkyl substances in the human body. Toxicol
Environ Chem 97: 814-827.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3223669
Fabrega, F; Kumar, V; Schuhmacher, M; Domingo, JL; Nadal, M. (2014). PBPK modeling for PFOS and
PFOA: validation with human experimental data. Toxicol Lett 230: 244-251.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850904
Fan, H; Ducatman, A; Zhang, J. (2014). Perfluorocarbons and Gilbert syndrome (phenotype) in the C8
Health Study Population. Environ Res 135: 70-75.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2967086
Fan, Y; Lu, C; Li, X; Xu, Q; Zhang, Y; Yang, X; Han, X; Du, G; Xia, Y; Wang, X. (2020). Serum
albumin mediates the effect of multiple per- and polyfluoroalkyl substances on serum lipid levels.
Environ Pollut266 Pt2: 115138.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7102734
Fasano, M; Curry, S; Terreno, E; Galliano, M; Fanali, G; Narciso, P; Notari, S; Ascenzi, P. (2005). The
extraordinary ligand binding properties of human serum albumin [Review]. IUBMB Life 57: 787-
796. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1023584
7-19
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Fasano, WJ; Kennedy, GL; Szostek, B; Farrar, DG; Ward, RJ; Haroun, L; Hinderliter, PM. (2005).
Penetration of ammonium perfluorooctanoate through rat and human skin in vitro. Drug Chem
Toxicol 28: 79-90. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749187
Fassler, CS; Pinney, SE; Xie, C; Biro, FM; Pinney, SM. (2019). Complex relationships between
perfluorooctanoate, body mass index, insulin resistance and serum lipids in young girls. Environ
Res 176: 108558. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315820
FDA. (2002). Guidance for industry: immunotoxicology evaluation of investigational new drugs.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/88170
FDA. (2009). Drug-induced liver injury: Premarketing clinical evaluation. (Docket no. FDA-2008-D-
0128). Rockville, MD: U.S. Department of Health and Human Services, Food and Drug
Administration, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6987952
FDA. (2016). Analytical Results for PFAS in 2016 Carbonated Water and Non-Carbonated Bottled Water
Sampling (parts per trillion). Retrieved from
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419013
FDA. (2018). Analytical Results for PFAS in 2018 Produce Sampling (Parts Per Trillion). Retrieved from
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419064
FDA. (2020). Authorized Uses of PFAS in Food contact Applications.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419078
FDA. (2020). FDA Announces Voluntary Agreement with Manufacturers to Phase Out Certain Short-
Chain PFAS Used in Food Packaging.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419079
FDA. (2021). Analytical Results of Testing Food for PFAS from Environmental Contamination.
Retrieved from https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419076
Fei, C; Mclaughlin, JK; Lipworth, L; Olsen, J. (2008). Prenatal exposure to perfluorooctanoate (PFOA)
and perfluorooctanesulfonate (PFOS) and maternally reported developmental milestones in
infancy. Environ Health Perspect 116: 1391-1395.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290822
Fei, C; Mclaughlin, JK; Lipworth, L; Olsen, J. (2010). Prenatal exposure to PFOA and PFOS and risk of
hospitalization for infectious diseases in early childhood. Environ Res 110: 773-777.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290805
Fei, C; Mclaughlin, JK; Tarone, RE; Olsen, J. (2007). Perfluorinated chemicals and fetal growth: A study
within the Danish National Birth Cohort. Environ Health Perspect 115: 1677-1682.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1005775
Fei, C; Olsen, J. (2011). Prenatal exposure to perfluorinated chemicals and behavioral or coordination
problems at age 7 years. Environ Health Perspect 119: 573-578.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/758428
Fei, CY; Mclaughlin, JK; Lipworth, L; Olsen, J. (2009). Maternal levels of perfluorinated chemicals and
subfecundity. Hum Reprod 24: 1200-1205.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291107
Feng, Y; Bai, Y; Lu, Y; Chen, M; Fu, M; Guan, X; Cao, Q; Yuan, F; Jie, J; Li, M; Meng, H; Wang, C;
Hong, S; Zhou, Y; Zhang, X; He, M; Guo, H. (2022). Plasma perfluoroalkyl substance exposure
and incidence risk of breast cancer: A case-cohort study in the Dongfeng-Tongji cohort. Environ
Pollut 306: 119345. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328872
Fenton, SE. (2006). Endocrine-disrupting compounds and mammary gland development: Early exposure
and later life consequences [Review]. Endocrinology 147: S18-S24.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/470286
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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988520
Fenton, SE; Reiner, JL; Nakayama, SF; Delinsky, AD; Stanko, JP; Hines, EP; White, SS; Lindstrom, AB;
7-20
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Strynar, MJ; Petropoulou, SSE. (2009). Analysis of PFOA in dosed CD-I mice. Part 2:
Disposition of PFOA in tissues and fluids from pregnant and lactating mice and their pups.
Reprod Toxicol 27: 365-372.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/194799
Fernandez, E, .; Perez, R, .; Hernandez, A, .; Tejada, P, .; Arteta, M, .; Ramos, J, .T. (2011). Factors and
Mechanisms for Pharmacokinetic Differences between Pediatric Population and Adults.
Pharmaceutics 3: 53-72.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641878
Fernandez Freire, P; Perez Martin, JM; Herrero, O; Peropadre, A; de La Pena, E; Hazen, MJ. (2008). In
vitro assessment of the cytotoxic and mutagenic potential of perfluorooctanoic acid. Toxicol In
Vitro 22: 1228-1233. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919390
Filgo, AJ; Quist, EM; Hoenerhoff, MJ; Brix, AE; Kissling, GE; Fenton, SE. (2015). Perfluorooctanoic
acid (PFOA)-induced liver lesions in two strains of mice following developmental exposures:
PPARalpha is not required. Toxicol Pathol 43: 558-568.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851085
Fisher, M; Arbuckle, TE; Wade, M; Haines, DA. (2013). Do perfluoroalkyl substances affect metabolic
function and plasma lipids?~Analysis of the 2007-2009, Canadian Health Measures Survey
(CHMS) Cycle 1. Environ Res 121: 95-103.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/2919156
Fitz-Simon, N; Fletcher, T; Luster, MI; Steenland, K; Calafat, AM; Kato, K; Armstrong, B. (2013).
Reductions in serum lipids with a 4-year decline in serum perfluorooctanoic acid and
perfluorooctanesulfonic acid. Epidemiology 24: 569-576.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850962
Fleisch, AF; Rifas-Shiman, SL; Mora, AM; Calafat, AM; Ye, X; Luttmann-Gibson, H; Gillman, MW;
Oken, E; Sagiv, SK. (2017). Early-life exposure to perfluoroalkyl substances and childhood
metabolic function. Environ Health Perspect 125: 481-487.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3 85 8513
Fletcher, T; Galloway, TS; Melzer, D; Holcroft, P; Cipelli, R; Pilling, LC; Mondal, D; Luster, M; Harries,
LW. (2013). Associations between PFOA, PFOS and changes in the expression of genes involved
in cholesterol metabolism in humans. Environ Int 57-58: 2-10.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850968
Florentin, A; Deblonde, T; Diguio, N; Hautemaniere, A; Hartemann, P. (2011). Impacts of two
perfluorinated compounds (PFOS and PFOA) on human hepatoma cells: cytotoxicity but no
genotoxicity? Int J Hyg Environ Health 214: 493-499.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919235
Foley, GL. (2001). Overview of male reproductive pathology [Review]. Toxicol Pathol 29: 49-63.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4003913
Forns, J; Iszatt, N; White, RA; Mandal, S; Sabaredzovic, A; Lamoree, M; Thomsen, C; Haug, LS;
Stigum, H; Eggcsbo. M. (2015). Perfluoroalkyl substances measured in breast milk and child
neuropsychological development in a Norwegian birth cohort study. Environ Int 83: 176-182.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3228833
Forsthuber, M; Kaiser, AM; Granitzer, S; Hassl, I; Hengstschlager, M; Stangl, H; Gundacker, C. (2020).
Albumin is the major carrier protein for PFOS, PFOA, PFHxS, PFNA and PFDA in human
plasma. Environ Int 137: 105324.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311640
Fragki, S; Dirven, H; Fletcher, T; Grasl-Kraupp, B; Bjerve Giitzkow, K; Hoogenboom, R; Kersten, S;
Lindeman, B; Louisse, J; Peijnenburg, A; Piersma, AH; Princen, HMG; Uhl, M; Westerhout, J;
Zeilmaker, MJ; Luijten, M. (2021). Systemic PFOS and PFOA exposure and disturbed lipid
homeostasis in humans: what do we know and what not? Crit Rev Toxicoll41-164.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8442211
Franco, ME; Fernandez-Luna, MT; Ramirez, AJ; Lavado, R. (2020). Metabolomic-based assessment
7-21
-------
DRAFT FOR PUBLIC COMMENT
March 2023
reveals dysregulation of lipid profiles in human liver cells exposed to environmental obesogens.
Toxicol Appl Pharmacol 398: 115009.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6507465
Franco, ME; Sutherland, GE; Fernandez-Luna, MT; Lavado, R. (2020). Altered expression and activity of
phase I and II biotransformation enzymes in human liver cells by perfluorooctanoate (PFOA) and
perfluorooctane sulfonate (PFOS). Toxicology 430: 152339.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/6315712
Franken, C; Koppen, G; Lambrechts, N; Govarts, E; Bruckers, L; Den Hond, E; Loots, I; Nelen, V; Sioen,
I; Nawrot, TS; Baeyens, W; Van Larebeke, N; Boonen, F; Ooms, D; Wevers, M; Jacobs, G;
Covaci, A; Schettgen, T; Schoeters, G. (2017). Environmental exposure to human carcinogens in
teenagers and the association with DNA damage. Environ Res 152: 165-174.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3789256
Fraser, AJ; Webster, TF; Watkins, DJ; Strynar, MJ; Kato, K; Calafat, AM; Vieira, VM; Mcclean, MD.
(2013). Polyfluorinated compounds in dust from homes, offices, and vehicles as predictors of
concentrations in office workers' serum. Environ Int 60: 128-136.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2325338
Frisbee, SJ; Shankar, A; Knox, SS; Steenland, K; Savitz, DA; Fletcher, T; Ducatman, AM. (2010).
Perfluorooctanoic acid, perfluorooctanesulfonate, and serum lipids in children and adolescents:
results from the C8 Health Project. Arch Pediatr Adolesc Med 164: 860-869.
https://hero .epa.gov/hero/index.cfin/reference/details/reference_id/1430763
Fromme, H; Mosch, C; Morovitz, M; Alba-Alejandre, I; Boehmer, S; Kiranoglu, M; Faber, F; Hannibal,
I; Genzel-Boroviczeny, O; Koletzko, B; Volkel, W. (2010). Pre- and postnatal exposure to
perfluorinated compounds (PFCs). Environ Sci Technol 44: 7123-7129.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1290877
Fromme, H; Tittlemier, SA; Volkel, W; Wilhelm, M; Twardella, D. (2009). Perfluorinated compounds -
Exposure assessment for the general population in western countries [Review]. Int J Hyg Environ
Health 212: 239-270. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1291085
Fry, K; Power, MC. (2017). Persistent organic pollutants and mortality in the United States, NHANES
1999-2011. Environ Health 16: 105.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4181820
Fu, J; Gao, Y; Cui, L; Wang, T; Liang, Y; Qu, G; Yuan, B; Wang, Y; Zhang, A; Jiang, G. (2016).
Occurrence, temporal trends, and half-lives of perfluoroalkyl acids (PFAAs) in occupational
workers in China. Sci Rep 6: 38039.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3859819
Fujii, Y; Harada, KH; Kobayashi, H; Haraguchi, K; Koizumi, A. (2020). Lactational transfer of long-
chain perfluorinated carboxylic acids in mice: A method to directly collect milk and evaluate
chemical transferability. Toxics 8.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/6512379
Fujii, Y; Niisoe, T; Harada, KH; Uemoto, S; Ogura, Y; Takenaka, K; Koizumi, A. (2015). Toxicokinetics
of perfluoroalkyl carboxylic acids with different carbon chain lengths in mice and humans. J
Occup Health 57: 1-12.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2816710
Gabriel, K. (1976). Primary eye irritation study in rabbits, Report 226-0422. Biosearch, Inc.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4442370
Gabrielsson, J; Weiner, D. (2000). Pharmacokinetic and pharmacodynamic data analysis: concepts and
applications (3rd ed.). Stockholm: Swedish Pharmaceutical Press.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/9642135
Galazka, A; Kardymowicz, B. (1989). Immunity against diphtheria in adults in Poland. Epidemiol Infect
103: 587-593. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/9642152
Galazka, AM; Milstien, JB; Robertson, SE; Cutts, FT. (1993). The immunological basis for immunization
module 2 : Diphtheria. (WHO/EPI/Gen/93.11-18). Galazka, AM; Milstien, JB; Robertson, SE;
7-22
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Cutts, FT. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228565
Gallo, V; Leonardi, G; Brayne, C; Armstrong, B, en; Fletcher, T. (2013). Serum perfluoroalkyl acids
concentrations and memory impairment in a large cross-sectional study. BMJ Open 3.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2272847
Gallo, V; Leonardi, G; Genser, B; Lopez-Espinosa, MJ; Frisbee, SJ; Karlsson, L; Ducatman, AM;
Fletcher, T. (2012). Serum perfluorooctanoate (PFOA) and perfluorooctane sulfonate (PFOS)
concentrations and liver function biomarkers in a population with elevated PFOA exposure.
Environ Health Perspect 120: 655-660.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1276142
Gannon, SA; Fasano, WJ; Mawn, MP; Nabb, DL; Buck, RC; Buxton, LW; Jepson, GW; Frame, SR.
(2016). Absorption, distribution, metabolism, excretion, and kinetics of 2,3,3,3-tetrafluoro-2-
(heptafluoropropoxy)propanoic acid ammonium salt following a single dose in rat, mouse, and
cynomolgus monkey. Toxicology 340: 1-9.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3810188
Gao, B; He, X; Liu, W; Zhang, H; Saito, N; Tsuda, S. (2015). Distribution of perfluoroalkyl compounds
in rats: Indication for using hair as bioindicator of exposure. J Expo Sci Environ Epidemiol 25:
632-638. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851191
Gao, K, e; Zhuang, T; Liu, X; Fu, J; Zhang, J; Fu, J, ie; Wang, L; Zhang, A; Liang, Y; Song, M; Jiang, G.
(2019). Prenatal Exposure to Per- and Polyfluoroalkyl Substances (PFASs) and Association
between the Placental Transfer Efficiencies and Dissociation Constant of Serum Proteins-PFAS
Complexes. Environ Sci Technol 53: 6529-6538.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387135
Gao, X; Ni, W; Zhu, S; Wu, Y; Cui, Y; Ma, J; Liu, Y; Qiao, J; Ye, Y; Yang, P; Liu, C; Zeng, F. (2021).
Per- and polyfluoroalkyl substances exposure during pregnancy and adverse pregnancy and birth
outcomes: A systematic review and meta-analysis. Environ Res 201: 111632.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959601
Gao, Y; Fu, J; Cao, H; Wang, Y; Zhang, A; Liang, Y; Wang, T; Zhao, C; Jiang, G. (2015). Differential
Accumulation and Elimination Behavior of Perfluoroalkyl Acid Isomers in Occupational Workers
in a Manufactory in China. Environ Sci Technol 49: 6953-6962.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850134
Garry, VF; Nelson, RL. (1981). An assay of cell transformation and cytotoxicity in C3H 10T 1/2 clonal
cell line for the test chemical T-2942 CoC. (EPA-AR-226-0428). Minneapolis, MN: Stone
Research Laboratories.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228130
Gaylord, A; Berger, KI; Naidu, M; Attina, TM; Gilbert, J; Koshy, TT; Han, X; Marmor, M; Shao, Y;
Giusti, R; Goldring, RM; Kannan, K; Trasande, L. (2019). Serum perfluoroalkyl substances and
lung function in adolescents exposed to the World Trade Center disaster. Environ Res 172: 266-
272. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080201
Gaylord, A; Trasande, L; Kannan, K; Thomas, KM; Lee, S; Liu, M; Levine, J. (2020). Persistent organic
pollutant exposure and celiac disease: A pilot study. Environ Res 186: 109439.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833754
Gazouli, M; Yao, ZX; Boujrad, N; Corton, JC; Culty, M; Papadopoulos, V. (2002). Effect of peroxisome
proliferators on Leydig cell peripheral-type benzodiazepine receptor gene expression, hormone-
stimulated cholesterol transport, and steroidogenesis: Role of the peroxisome proliferator-
activator receptor alpha. Endocrinology 143: 2571-2583.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/674161
Gebbink, WA; Berger, U; Cousins, IT. (2015). Estimating human exposure to PFOS isomers and PFCA
homologues: the relative importance of direct and indirect (precursor) exposure. Environ Int 74:
160-169. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850068
Geiger, SD; Xiao, J; Ducatman, A; Frisbee, S; Innes, K; Shankar, A. (2014). The association between
PFOA, PFOS and serum lipid levels in adolescents. Chemosphere 98: 78-83.
7-23
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850925
Geiger, SD; Xiao, J; Shankar, A. (2013). Positive association between perfluoroalkyl chemicals and
hyperuricemia in children. Am J Epidemiol 177: 1255-1262.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919148
Geiger, SD; Xiao, J; Shankar, A. (2014). No association between perfluoroalkyl chemicals and
hypertension in children. Integrated Blood Pressure Control 7: 1-7.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851286
Genser, B; Teles, CA; Barreto, ML; Fischer, JE. (2015). Within- and between-group regression for
improving the robustness of causal claims in cross-sectional analysis. Environ Health 14: 60.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3271854
Genuis, SJ; Beesoon, S; Birkholz, D. (2013). Biomonitoring and Elimination of Perfluorinated
Compounds and Polychlorinated Biphenyls through Perspiration: Blood, Urine, and Sweat Study.
ISRN Toxicology 2013: 483832.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/2149530
Genuis, SJ; Birkholz, D; Ralitsch, M; Thibault, N. (2010). Human detoxification of perfluorinated
compounds. Public Health 124: 367-375.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2583643
Genuis, SJ; Liu, Y; Genuis, QI; Martin, JW. (2014). Phlebotomy treatment for elimination of
perfluoroalkyl acids in a highly exposed family: a retrospective case-series. PLoS ONE 9:
e 114295. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851045
Getz, GS; Reardon, CA. (2012). Animal models of atherosclerosis [Review]. Arterioscler Thromb Vase
Biol 32: 1104-1115. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1065480
Ghassabian, A; Bell, EM; Ma, WL; Sundaram, R; Kannan, K; Buck Louis, GM; Yeung, E. (2018).
Concentrations of perfluoroalkyl substances and bisphenol A in newborn dried blood spots and
the association with child behavior. Environ Pollut 243: 1629-1636.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080189
Ghisari, M; Long, M; R0ge, DM; Olsen, J; Bonefcld-Jorgenscn. EC. (2017). Polymorphism in xenobiotic
and estrogen metabolizing genes, exposure to perfluorinated compounds and subsequent breast
cancer risk: A nested case-control study in the Danish National Birth Cohort. Environ Res 154:
325-333. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860243
Gibson, SJ; Johnson, JD. (1979). Absorption of FC-143-14C in Rats After a Single Oral Dose. (USEPA
Public Docket AR-226-0455). St. Paul, MN: Riker Laboratories, Inc. Subsidiary of 3M company.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641813
Gilliland, FD; Mandel, JS. (1993). Mortality among employees of a perfluorooctanoic acid production
plant. J Occup Med 35: 950-954.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290858
Gimenez-Bastida, JA; Surma, M; Zielinski, H. (2015). In vitro evaluation of the cytotoxicity and
modulation of mechanisms associated with inflammation induced by perfluorooctanesulfonate
and perfluorooctanoic acid in human colon myofibroblasts CCD-I8C0. Toxicol In Vitro 29:
1683-1691. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981569
Girardi, P; Merler, E. (2019). A mortality study on male subjects exposed to polyfluoroalkyl acids with
high internal dose of perfluorooctanoic acid. Environ Res 179: 108743.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315730
Glassmeyer, ST; Furlong, ET; Kolpin, DW; Batt, AL; Benson, R; Boone, JS; Conerly, O; Donohue, MJ;
King, DN; Kostich, MS; Mash, HE; Pfaller, SL; Schenck, KM; Simmons, JE; Varughese, EA;
Vesper, SJ; Villegas, EN; Wilson, VS. (2017). Nationwide reconnaissance of contaminants of
emerging concern in source and treated drinking waters of the United States. Sci Total Environ
581-582: 909-922. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3454569
Gleason, JA; Cooper, KR; Klotz, JB; Post, GB; Van Orden, G; New Jersey Drinking Water Quality
Institute (NJDWQI). (2017). Health-based maximum contaminant level support document:
Perfluorooctanoic acid (PFOA): Appendix A.
7-24
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5024840
Gleason, JA; Post, GB; Fagliano, JA. (2015). Associations of perfluorinated chemical serum
concentrations and biomarkers of liver function and uric acid in the US population (NHANES),
2007-2010. Environ Res 136: 8-14.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2966740
Glynn, A; Berger, U; Bignert, A; Ullah, S; Aune, M; Lignell, S; Darnerud, PO. (2012). Perfluorinated
alkyl acids in blood serum from primiparous women in Sweden: serial sampling during
pregnancy and nursing, and temporal trends 1996-2010. Environ Sci Technol 46: 9071-9079.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1578498
Goeden, HM; Greene, CW; Jacobus, JA. (2019). A transgenerational toxicokinetic model and its use in
derivation of Minnesota PFOA water guidance. J Expo Sci Environ Epidemiol 29: 183-195.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080506
Goff, DC; Lloyd-Jones, DM; Bennett, G; Coady, S; DAgostino, RB; Gibbons, R; Greenland, P;
Lackland, DT; Levy, D; O'Donnell, CJ; Robinson, JG; Schwartz, JS; Shero, ST; Smith, SC;
Sorlie, P; Stone, NJ; Wilson, PW. (2014). 2013 ACC/AHA guideline on the assessment of
cardiovascular risk: a report of the American College of Cardiology/American Heart Association
Task Force on Practice Guidelines. Circulation 129: S49-S73.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3121148
Gogola, J; Hoffmann, M; Nimpsz, S; Ptak, A. (2020). Disruption of 17(3-estradiol secretion by persistent
organic pollutants present in human follicular fluid is dependent on the potential of ovarian
granulosa tumor cell lines to metabolize estrogen. Mol Cell Endocrinol 503: 110698.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316203
Gogola, J; Hoffmann, M; Ptak, A. (2019). Persistent endocrine-disrupting chemicals found in human
follicular fluid stimulate the proliferation of granulosa tumor spheroids via GPR30 and IGF1R
but not via the classic estrogen receptors. Chemosphere 217: 100-110.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5016947
Gogola, J; Hoffmann, M; Ptak, A. (2020). Persistent endocrine-disrupting chemicals found in human
follicular fluid stimulate IGF1 secretion by adult ovarian granulosa cell tumor spheroids and
thereby increase proliferation of non-cancer ovarian granulosa cells. Toxicol In Vitro 65: 104769.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316206
Goldenthal, E; Jessup, DC; Geil, RG; Mehring, JS. (1978). Ninety-day subacute rhesus monkey toxicity
study: Fluorad " Fluorochemical FC-143. (Study No. 137-090). St. Paul, MN: Report prepared for
3M by Institutional Research and Devlopment Corporation (Mattawan, MN).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291068
Goldenthal, EI; Jessup, DC; Geil, RB; Mehring, JS. (1979). 90-day subacute Rhesus monkey toxicity
study (FC-95). (Study No. 137-087). Mattawan, MI: International Research and Development
Corporation, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9573133
Goldenthal, EI; Jessup, DC; Geil, RB; Mehring, JS. (1979). Ninety-Day Subacute Rhesus Monkey
Toxicity Study. (Study No. 137-087). Mattawan, MI: International Research and Development
Corporation, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9573133
Gomis, MI; Vestergren, R; Macleod, M; Mueller, JF; Cousins, IT. (2017). Historical human exposure to
perfluoroalkyl acids in the United States and Australia reconstructed from biomonitoring data
using population-based pharmacokinetic modelling. Environ Int 108: 92-102.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981280
Gomis, MI; Vestergren, R; Nilsson, H; Cousins, IT. (2016). Contribution of Direct and Indirect Exposure
to Human Serum Concentrations of Perfluorooctanoic Acid in an Occupationally Exposed Group
of Ski Waxers. Environ Sci Technol 50: 7037-7046.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749264
Goodrich, JA; Walker, D; Lin, X; Wang, H; Lim, T; Mcconnell, R; Conti, DV; Chatzi, L; Setiawan, VW.
(2022). Exposure to perfluoroalkyl substances and risk of hepatocellular carcinoma in a
multiethnic cohort. 4: 100550.
7-25
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369722
Gorrochategui, E; Perez-Albaladejo, E; Casas, J; Lacorte, S; Porte, C. (2014). Perfluorinated chemicals:
Differential toxicity, inhibition of aromatase activity and alteration of cellular lipids in human
placental cells. Toxicol Appl Pharmacol 277: 124-130.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2324895
Goudarzi, H; Araki, A; Itoh, S; Sasaki, S; Miyashita, C; Mitsui, T; Nakazawa, H; Nonomura, K; Kishi, R.
(2017). The association of prenatal exposure to perfluorinated chemicals with glucocorticoid and
androgenic hormones in cord blood samples: The Hokkaido study. Environ Health Perspect 125:
111-118. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981462
Goudarzi, H; Miyashita, C; Okada, E; Kashino, I; Chen, CJ; Ito, S; Araki, A; Kobayashi, S; Matsuura, H;
Kishi, R. (2017). Prenatal exposure to perfluoroalkyl acids and prevalence of infectious diseases
up to 4 years of age. Environ Int 104: 132-138.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859808
Goudarzi, H; Miyashita, C; Okada, E; Kashino, I; Kobayashi, S; Chen, CJ; Ito, S; Araki, A; Matsuura, H;
Ito, YM; Kishi, R. (2016). Effects of prenatal exposure to perfluoroalkyl acids on prevalence of
allergic diseases among 4-year-old children. Environ Int 94: 124-132.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859523
Goudarzi, H; Nakajima, S; Ikeno, T; Sasaki, S; Kobayashi, S; Miyashita, C; Ito, S; Araki, A; Nakazawa,
H; Kishi, R. (2016). Prenatal exposure to perfluorinated chemicals and neurodevelopment in early
infancy: The Hokkaido Study. Sci Total Environ 541: 1002-1010.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981536
Goulding, DR; White, SS; Mcbride, SJ; Fenton, SE; Harry, GJ. (2017). Gestational exposure to
perfluorooctanoic acid (PFOA): Alterations in motor related behaviors. Neurotoxicology 58: 110-
119. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981400
Govarts, E; Remy, S; Bruckers, L; Den Hond, E; Sioen, I; Nelen, V; Baeyens, W; Nawrot, TS; Loots, I;
Van Larebeke, N; Schoeters, G. (2016). Combined effects of prenatal exposures to environmental
chemicals on birth weight. Int J Environ Res Public Health 13: n/a.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3230364
Governini, L; Guerranti, C; De Leo, V; Boschi, L; Luddi, A; Gori, M; Orvieto, R; Piomboni, P. (2015).
Chromosomal aneuploidies and DNA fragmentation of human spermatozoa from patients
exposed to perfluorinated compounds. Andrologia 47: 1012-1019.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981589
Graber, JM; Alexander, C; Laumbach, RJ; Black, K; Strickland, PO; Georgopoulos, PG; Marshall, EG;
Shendell, DG; Alderson, D; Mi, Z; Mascari, M; Weisel, CP. (2019). Per and polyfluoroalkyl
substances (PFAS) blood levels after contamination of a community water supply and
comparison with 2013-2014 NHANES. J Expo Sci Environ Epidemiol 29: 172-182.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080653
Grandjean, P; Andersen, EW; Budtz-Jorgensen. E; Nielsen, F; Molbak. K; Weihe, P; Heilmann, C.
(2012). Serum vaccine antibody concentrations in children exposed to perfluorinated compounds.
JAMA 307: 391-397. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1248827
Grandjean, P; Bateson, T. (2021). RE: Benchmark analysis for PFAS immunotoxicity. Available online
9959716
Grandjean, P; Heilmann, C; Weihe, P; Nielsen, F; Mogensen, UB; Budtz-Jorgensen, E. (2017). Serum
Vaccine Antibody Concentrations in Adolescents Exposed to Perfluorinated Compounds. Environ
Health Perspect 125: 077018.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858518
Grandjean, P; Heilmann, C; Weihe, P; Nielsen, F; Mogensen, UB; Timmermann, A; Budtz-Jorgensen, E.
(2017). Estimated exposures to perfluorinated compounds in infancy predict attenuated vaccine
antibody concentrations at age 5-years. J Immunotoxicol 14: 188-195.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239492
Granum, B; Haug, LS; Namork, E; Stolevik. SB; Thomsen, C; Aaberge, IS; van Loveren, H; Lovik. M;
7-26
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Nygaard, UC. (2013). Pre-natal exposure to perfluoroalkyl substances may be associated with
altered vaccine antibody levels and immune-related health outcomes in early childhood. J
Immunotoxicol 10: 373-379.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937228
Greenland, S; Longnecker, MP. (1992). Methods for trend estimation from summarized dose-response
data, with applications to meta-analysis. Am J Epidemiol 135: 1301-1309.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5069
Gremmel, C; Fromel, T; Knepper, TP. (2016). Systematic determination of perfluoroalkyl and
polyfluoroalkyl substances (PFASs) in outdoor jackets. Chemosphere 160: 173-180.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858525
Gui, SY; Chen, YN; Wu, KJ; Liu, W; Wang, WJ; Liang, HR; Jiang, ZX; Li, ZL; Hu, CY. (2022).
Association Between Exposure to Per- and Polyfluoroalkyl Substances and Birth Outcomes: A
Systematic Review and Meta-Analysis. Front Public Health 10: 855348.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365824
Guo, H; Chen, J; Zhang, H; Yao, J; Sheng, N; Li, Q; Guo, Y; Wu, C; Xie, W; Dai, J. (2021). Exposure to
genX and its novel analogs disrupts hepatic bile acid metabolism in male mice. Environ Sci
Technol. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9963377
Guo, H; Wang, J; Yao, J; Sun, S; Sheng, N; Zhang, X; Guo, X; Guo, Y; Sun, Y; Dai, J. (2019).
Comparative hepatotoxicity of novel PFOA alternatives (perfluoropolyether carboxylic acids) on
male mice. Environ Sci Technol 53: 3929-3937.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080372
Guo, H; Zhang, H; Sheng, N; Wang, J; Chen, J; Dai, J. (2021). Perfluorooctanoic acid (PFOA) exposure
induces splenic atrophy via overactivation of macrophages in male mice. J Hazard Mater 407:
124862. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7542749
Guruge, KS; Yeung, LW; Yamanaka, N; Miyazaki, S; Lam, PK; Giesy, JP; Jones, PD; Yamashita, N.
(2006). Gene expression profiles in rat liver treated with perfluorooctanoic acid (PFOA). Toxicol
Sci 89: 93-107. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/1937270
Gutzkow, KB; Haug, LS; Thomsen, C; Sabaredzovic, A; Becher, G; Brunborg, G. (2012). Placental
transfer of perfluorinated compounds is selective - A Norwegian Mother and Child sub-cohort
study. Int J Hyg Environ Health 215: 216-219.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290878
Gyllenhammar, I; Benskin, JP; Sandblom, O; Berger, U; Ahrens, L; Lignell, S; Wiberg, K; Glynn, A.
(2018). Perfluoroalkyl acids (PFAAs) in serum from 2-4-month-old infants: Influence of maternal
serum concentration, gestational age, breast-feeding, and contaminated drinking water. Environ
Sci Technol 52: 7101-7110.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4778766
Gyllenhammar, I; Benskin, JP; Sandblom, O; Berger, U; Ahrens, L; Lignell, S; Wiberg, K; Glynn, A.
(2019). Perfluoroalkyl Acids (PFAAs) in Children's Serum and Contribution from PFAA-
Contaminated Drinking Water. Environ Sci Technol 53: 11447-11457.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5919402
Gyllenhammar, I; Diderholm, B; Gustafsson, J; Berger, U; Ridefelt, P; Benskin, JP; Lignell, S; Lampa, E;
Glynn, A. (2018). Perfluoroalkyl acid levels in first-time mothers in relation to offspring weight
gain and growth. Environ Int 111: 191-199.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238300
Hack, M; Klein, NK; Taylor, HG. (1995). Long-term developmental outcomes of low birth weight
infants. Future Child 5: 176-196.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8632216
Hagenaars, A; Vergauwen, L; Benoot, D; Laukens, K; Knapen, D. (2013). Mechanistic toxicity study of
perfluorooctanoic acid in zebrafish suggests mitochondrial dysfunction to play a key role in
PFOA toxicity. Chemosphere 91: 844-856.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850980
7-27
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Hall, AP; Elcombe, CR; Foster, JR; Harada, T; Kaufmann, W; Knippel, A; Kiittler, K; Malarkey, DE;
Maronpot, RR; Nishikawa, A; Nolte, T; Schulte, A; Strauss, V; York, 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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2718645
Hammer, T; Lophaven, SN; Nielsen, KR; Petersen, MS; Munkholm, P; Weihe, P; Burisch, J; Lynge, E.
(2019). Dietary risk factors for inflammatory bowel diseases in a high-risk population: Results
from the Faroese IBD study. 7: 924-932.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8776815
Han, R; Zhang, F; Wan, C; Liu, L; Zhong, Q; Ding, W. (2018). Effect of perfluorooctane sulphonate-
induced Kupffer cell activation on hepatocyte proliferation through the NF-KB/TNF-a/IL-6-
dependent pathway. Chemosphere 200: 283-294.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/43 5 5066
Han, W; Gao, Y; Yao, Q; Yuan, T; Wang, Y; Zhao, S; Shi, R; Bonefeld-Jorgensen, EC; Shen, X; Tian, Y.
(2018). Perfluoroalkyl and polyfluoroalkyl substances in matched parental and cord serum in
Shandong, China. Environ Int 116: 206-213.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080230
Han, X. (2003). Ammonium Perfluorooctanoate: Age Effect on the Plasma Concentration in Post-
Weaning Rats Following Oral Gavage [EPA Report]. (Study No. Dupont-13267, December 15,
2003; US EPA Administrative Record 226-1553). Haskell Laboratory for Health and
Environmental Sciences.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9978263
Han, X; Kemper, RA; Jepson, GW. (2005). Subcellular distribution and protein binding of
perfluorooctanoic acid in rat liver and kidney. Drug Chem Toxicol 28: 197-209.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081570
Han, X; Meng, L; Zhang, G; Li, Y; Shi, Y; Zhang, Q; Jiang, G. (2021). Exposure to novel and legacy per-
and polyfluoroalkyl substances (PFASs) and associations with type 2 diabetes: A case-control
study in East China. Environ Int 156: 106637.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7762348
Han, X; Snow, TA; Kemper, RA; Jepson, GW. (2003). Binding of perfluorooctanoic acid to rat and
human plasma proteins. Chem Res Toxicol 16: 775-781.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081471
Hanahan, D. (2022). Hallmarks of cancer: New dimensions [Review]. Cancer Discovery 12: 31-46.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10164687
Hanahan, D; Weinberg, RA. (2000). The hallmarks of cancer [Review]. Cell 100: 57-70.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/188413
Hanahan, D; Weinberg, RA. (2011). Hallmarks of cancer: The next generation [Review]. Cell 144: 646-
674. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/758924
Hanhijarvi, H; Ophaug, RH; Singer, L. (1982). THE SEX-RELATED DIFFERENCE IN
PERFLUOROOCTANOATE EXCRETION IN THE RAT. Proc Soc Exp Biol Med 171: 50-55.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5085525
Hansen, KJ; Johnson, HO; Eldridge, JS; Butenhoff, JL; Dick, LA. (2002). Quantitative characterization of
trace levels of PFOS and PFOA in the Tennessee River. Environ Sci Technol 36: 1681-1685.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1424808
Hanssen, L; Dudarev, AA; Huber, S; Odland, J0; Nieboer, E; Sandanger, TM. (2013). Partition of
perfluoroalkyl substances (PFASs) in whole blood and plasma, assessed in maternal and
umbilical cord samples from inhabitants of arctic Russia and Uzbekistan. Sci Total Environ 447:
430-437. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859848
Hanssen, L; Rollin, H; Odland, J0; Moe, MK; Sandanger, TM. (2010). Perfluorinated compounds in
maternal serum and cord blood from selected areas of South Africa: results of a pilot study. J
Environ Monit 12: 1355-1361.
7-28
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919297
Hanvatananukul, P; Prasarakee, C; Sarachai, S; Aurpibul, L; Sintupat, K; Khampan, R; Saheng, J;
Sudjaritruk, T. (2020). Seroprevalence of antibodies against diphtheria, tetanus, and pertussis
among healthy Thai adolescents. Int J Infect Dis 96: 422-430.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642158
Harada, K; Inoue, K; Morikawa, A; Yoshinaga, T; Saito, N; Koizumi, A. (2005). Renal clearance of
perfluorooctane sulfonate and perfluorooctanoate in humans and their species-specific excretion.
Environ Res 99: 253-261.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4564250
Hardell, E; Karrman, A; van Bavel, B; Bao, J; Carlberg, M; Hardell, L. (2014). Case-control study on
perfluorinated alkyl acids (PFAAs) and the risk of prostate cancer. Environ Int 63: 35-39.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2968084
Harkness, JE; Wagner, JE. (1983). The Biology and Medicine of Rabbits and Rodents (2nd ed.).
Philadelphia, PA: Lea & Febiger.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641985
Harris, MH; Oken, E; Rifas-Shiman, SL; Calafat, AM; Ye, X; Bellinger, DC; Webster, TF; White, RF;
Sagiv, SK. (2018). Prenatal and childhood exposure to per- and polyfluoroalkyl substances
(PFASs) and child cognition. Environ Int 115: 358-369.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4442261
Hartman, TJ; Calafat, AM; Holmes, AK; Marcus, M; Northstone, K; Flanders, WD; Kato, K; Taylor, EV.
(2017). Prenatal exposure to perfluoroalkyl substances and body fatness in girls. Child Obes 13:
222-230. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3859812
Haug, LS; Huber, S; Becher, G; Thomsen, C. (2011). Characterisation of human exposure pathways to
perfluorinated compounds—comparing exposure estimates with biomarkers of exposure. Environ
Int 37: 687-693. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2577501
Hayashi, M. (2016). The micronucleus test-most widely used in vivo genotoxicity test [Review]. Genes
Environ 38: 18. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/9956921
He, X; Liu, Y; Xu, B; Gu, L; Tang, W. (2018). PFOA is associated with diabetes and metabolic alteration
in US men: National Health and Nutrition Examination Survey 2003-2012. Sci Total Environ
625: 566-574. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4238388
Heffernan, AL; Cunningham, TK; Drage, DS; Aylward, LL; Thompson, K; Vijayasarathy, S; Mueller, JF;
Atkin, SL; Sathyapalan, T. (2018). Perfluorinated alkyl acids in the serum and follicular fluid of
UK women with and without polycystic ovarian syndrome undergoing fertility treatment and
associations with hormonal and metabolic parameters. Int J Hyg Environ Health 221: 1068-1075.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5079713
Hessel, EVS; Tonk, ECM; Bos, PMJ; Van Loveren, H; Piersma, AH. (2015). Developmental
immunotoxicity of chemicals in rodents and its possible regulatory impact [Review]. Crit Rev
Toxicol 45: 68-82. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5750707
Hill, AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/71664
Hinderliter, PM; Delorme, MP; Kennedy, GL. (2006). Perfluorooctanoic acid: Relationship between
repeated inhalation exposures and plasma PFOA concentration in the rat. Toxicology 222: 80-85.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/135732
Hinderliter, PM; Han, X; Kennedy, GL; Butenhoff, JL. (2006). Age effect on perfluorooctanoate (PFOA)
plasma concentration in post-weaning rats following oral gavage with ammonium
perfluorooctanoate (APFO). Toxicology 225: 195-203.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3749132
Hinderliter, PM; Mylchreest, E; Gannon, SA; Butenhoff, JL; Kennedy, GL. (2005). Perfluorooctanoate:
Placental and lactational transport pharmacokinetics in rats. Toxicology 211: 139-148.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1332671
Hines, EP; White, SS; Stanko, JP; Flournoy, EAG; Lau, C; Fenton, SE. (2009). Phenotypic dichotomy
7-29
-------
DRAFT FOR PUBLIC COMMENT
March 2023
following developmental exposure to perfluorooctanoic acid (PFOA) in female CD-I mice: low
doses induce elevated serum leptin and insulin, and overweight in mid-life. Mol Cell Endocrinol
304: 97-105. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/194816
Hocevar, SE; Kamendulis, LM; Hocevar, BA. (2020). Perfluorooctanoic acid activates the unfolded
protein response in pancreatic acinar cells. J Biochem Mol Toxicol 34: e22561.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833720
Hoffman, K; Webster, TF; Weisskopf, MG; Weinberg, J; Vieira, VM. (2010). Exposure to
polyfluoroalkyl chemicals and attention deficit/hyperactivity disorder in U.S. children 12-15
years of age. Environ Health Perspect 118: 1762-1767.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291112
Holder, C; Deluca, N; Luh, J; Soleymani, P; Vallero, D; Cohen Hubal, E. (2021 in prep.). (In Press)
Systematic evidence mapping of potential exposure pathways for per- and poly-fluoroalkyl
(PFAS) chemicals based on measured occurrence in multiple media.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419128
Honda-Kohmo, K; Hutcheson, R; Innes, KE; Conway, BN. (2019). Perfluoroalkyl substances are
inversely associated with coronary heart disease in adults with diabetes. J Diabetes Complications
33: 407-412. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080551
Hossein-Khannazer, N; Zian, Z; Bakkach, J; Kamali, AN; Hosseinzadeh, R; Anka, AU; Yazdani, R;
Azizi, G. (2021). Features and roles of T helper 22 cells in immunological diseases and
malignancies [Review]. Scand J Immunol 93: e 13030.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365738
Hotez, P. (2019). America and Europe's new normal: the return of vaccine-preventable diseases. 85: 912-
914. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642144
Hover. BB; Ramlau-Hansen, CH; Obel, C; Pedersen, HS; Hernik, A; Ogniev, V; Jonsson, BA; Lindh,
CH; Rylander, L; Rignell-Hydbom, A; Bonde, JP; Toft, G. (2015). Pregnancy serum
concentrations of perfluorinated alkyl substances and offspring behaviour and motor development
at age 5-9 years—a prospective study. Environ Health 14: 2.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851038
Hu, J; Li, J; Wang, J; Zhang, A; Dai, J. (2014). Synergistic effects of perfluoroalkyl acids mixtures with
J-shaped concentration-responses on viability of a human liver cell line. Chemosphere 96: 81-88.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2325340
Hu, Q; Franklin, JN; Bryan, I; Morris, E; Wood, A; Dewitt, JC. (2012). Does developmental exposure to
perflurooctanoic acid (PFOA) induce immunopathologies commonly observed in
neurodevelopmental disorders? Neurotoxicology 33: 1491-1498.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937235
Hu, Q; Strynar, MJ; Dewitt, JC. (2010). Are developmentally exposed C57BL/6 mice insensitive to
suppression of TDARby PFOA? J Immunotoxicol 7: 344-349.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332421
Hu, XDC; Tokranov, AK; Liddie, J; Zhang, XM; Grandjean, P; Hart, JE; Laden, F; Sun, Q; Yeung,
LWY; Sunderland, EM. (2019). Tap Water Contributions to Plasma Concentrations of Poly- and
Perfluoroalkyl Substances (PFAS) in a Nationwide Prospective Cohort of U.S. Women. Environ
Health Perspect 127: 67006.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381562
Hu, XZ; Hu, DC. (2009). Effects of perfluorooctanoate and perfluorooctane sulfonate exposure on
hepatoma Hep G2 cells. Arch Toxicol 83: 851-861.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919334
Hu, Y; Liu, G; Rood, J; Liang, L; Bray, GA; de Jonge, L; Coull, B; Furtado, JD; Qi, L; Grandjean, P;
Sun, Q. (2019). Perfluoroalkyl substances and changes in bone mineral density: A prospective
analysis in the POUNDS-LOST study. Environ Res 179: 108775.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315798
Huang, H; Yu, K; Zeng, X; Chen, Q; Liu, Q; Zhao, Y; Zhang, J; Zhang, X; Huang, L. (2020). Association
7-30
-------
DRAFT FOR PUBLIC COMMENT
March 2023
between prenatal exposure to perfluoroalkyl substances and respiratory tract infections in
preschool children. Environ Res 191: 110156.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988475
Huang, JS; Borensztajn, J; Reddy, JK. (2011). Hepatic lipid metabolism. In SPS Monga (Ed.), Molecular
pathology of liver diseases (pp. 133-146). Boston, MA: Springer.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10284973
Huang, M; Jiao, J; Zhuang, P; Chen, X; Wang, J; Zhang, Y. (2018). Serum polyfluoroalkyl chemicals are
associated with risk of cardiovascular diseases in national US population. Environ Int 119: 37-46.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5024212
Huang, Q; Zhang, J; Martin, FL; Peng, S; Tian, M; Mu, X; Shen, H. (2013). Perfluorooctanoic acid
induces apoptosis through the p53-dependent mitochondrial pathway in human hepatic cells: a
proteomic study. Toxicol Lett 223: 211-220.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850934
Huang, Q; Zhang, J; Peng, S; Du, M; Ow, S; Pu, H; Pan, C; Shen, H. (2014). Proteomic analysis of
perfluorooctane sulfonate-induced apoptosis in human hepatic cells using the iTRAQ technique. J
Appl Toxicol 34: 1342-1351.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851292
Huang, R; Chen, Q; Zhang, L; Luo, K; Chen, L; Zhao, S; Feng, L; Zhang, J. (2019). Prenatal exposure to
perfluoroalkyl and polyfluoroalkyl substances and the risk of hypertensive disorders of
pregnancy. Environ Health 18:5.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5083564
Huhtaniemi, I; Toppari, J. (1995). Endocrine, paracrine and autocrine regulation of testicular
steroidogenesis. In AK Mukhopadhyay; MK Raizada (Eds.), Tissue renin-angiotensin systems:
Current concepts of local regulators in reproductive and endocrine organs (pp. 33-54). New York,
NY: Springer, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7420539
Hui, Z; Li, R; Chen, L. (2017). The impact of exposure to environmental contaminant on hepatocellular
lipid metabolism. Gene 622: 67-71.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981345
Humblet, O; Diaz-Ramirez, LG; Balmes, JR; Pinney, SM; Hiatt, RA. (2014). Perfluoroalkyl chemicals
and asthma among children 12-19 years of age: NHANES (1999-2008). Environ Health Perspect
122: 1129-1133. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851240
Hundley, SG; Sarrif, AM; Kennedy, GL. (2006). Absorption, distribution, and excretion of ammonium
perfluorooctanoate (APFO) after oral administration to various species. Drug Chem Toxicol 29:
137-145. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749054
Huo, X; Huang, R; Gan, Y; Luo, K; Aimuzi, R; Nian, M; Ao, J; Feng, L; Tian, Y; Wang, W; Ye, W;
Zhang, J. (2020). Perfluoroalkyl substances in early pregnancy and risk of hypertensive disorders
of pregnancy: A prospective cohort study. Environ Int 138: 105656.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505752
Hurley, S; Goldberg, D; Wang, M; Park, JS; Petreas, M; Bernstein, L; Anton-Culver, H; Nelson, DO;
Reynolds, P. (2018). Breast cancer risk and serum levels of per- and poly-fluoroalkyl substances:
a case-control study nested in the California Teachers Study. Environ Health 17: 83.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080646
Hutcheson, R; Innes, K; Conway, B. (2020). Perfluoroalkyl substances and likelihood of stroke in persons
with and without diabetes. Diab Vase Dis Res 17: 1-8.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6320195
IARC (International Agency for Research on Cancer). (2016). Some chemicals used as solvents and in
polymer manufacture
Perfluorooctanoic acid. In IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans. Lyons, France: World Health Organization.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3982387
ILEPA. Title 35: Environmental Protection. Subtitle: Public Water Supplies. Chapter I: Pollution Control
7-31
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Board. Part 620 Groundwater Quality, (2019).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9417528
Impinen, A; Longnecker, MP; Nygaard, UC; London, SJ; Ferguson, KK; Haug, LS; Granum, B. (2019).
Maternal levels of perfluoroalkyl substances (PFASs) during pregnancy and childhood allergy
and asthma related outcomes and infections in the Norwegian Mother and Child (MoBa) cohort.
Environ Int 124: 462-472.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080609
Impinen, A; Nygaard, UC; Lodrup Carlsen, KC; Mowinckel, P; Carlsen, KH; Haug, LS; Granum, B.
(2018). Prenatal exposure to perfluoralkyl substances (PFASs) associated with respiratory tract
infections but not allergy- and asthma-related health outcomes in childhood. Environ Res 160:
518-523. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238440
Innes, KE; Wimsatt, JH; Frisbee, S; Ducatman, AM. (2014). Inverse association of colorectal cancer
prevalence to serum levels of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA)
in a large Appalachian population. BMC Cancer 14: 45.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850898
Inoue, K; Ritz, B; Andersen, SL; Ramlau-Hansen, CH; Hover. BB; Bech, BH; Henriksen, TB; Bonefeld-
Jorgensen. EC; Olsen, J; Liew, Z. (2019). Perfluoroalkyl Substances and Maternal Thyroid
Hormones in Early Pregnancy; Findings in the Danish National Birth Cohort. Environ Health
Perspect 127: 117002.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5918599
Ioannou, GN; Boyko, EJ; Lee, SP. (2006). The prevalence and predictors of elevated serum
aminotransferase activity in the United States in 1999-2002. Am J Gastroenterol 101: 76-82.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10473853
Ioannou, GN; Weiss, NS; Boyko, EJ; Mozaffarian, D; Lee, SP. (2006). Elevated serum alanine
aminotransferase activity and calculated risk of coronary heart disease in the United States.
Hepatology 43: 1145-1151.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10473854
IPCS. (2012). Harmonization project document no. 10: Guidance for immunotoxicity risk assessment for
chemicals. Geneva, Switzerland: World Health Organization.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/124975 5
Ipsen, J. (1946). Circulating antitoxin at the onset of diphtheria in 425 patients. J Immunol 54: 325-347.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228563
Ito, S; Alcorn, J. (2003). Xenobiotic transporter expression and function in the human mammary gland.
Adv Drug Deliv Rev 55: 653-665.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641803
Itoh, H; Harada, KH; Kasuga, Y; Yokoyama, S; Onuma, H; Nishimura, H; Kusama, R; Yokoyama, K;
Zhu, J; Harada Sassa, M; Tsugane, S; Iwasaki, M. (2021). Serum perfluoroalkyl substances and
breast cancer risk in Japanese women: A case-control study. Sci Total Environ 800: 149316.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959632
Itoh, S; Araki, A; Mitsui, T; Miyashita, C; Goudarzi, H; Sasaki, S; Cho, K; Nakazawa, H; Iwasaki, Y;
Shinohara, N; Nonomura, K; Kishi, R. (2016). Association of perfluoroalkyl substances exposure
in utero with reproductive hormone levels in cord blood in the Hokkaido Study on Environment
and Children's Health. Environ Int 94: 51-59.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981465
Itoh, S; Araki, A; Miyashita, C; Yamazaki, K; Goudarzi, H; Minatoya, M; Ait Bamai, Y; Kobayashi, S;
Okada, E; Kashino, I; Yuasa, M; Baba, T; Kishi, R. (2019). Association between perfluoroalkyl
substance exposure and thyroid hormone/thyroid antibody levels in maternal and cord blood: The
Hokkaido Study. Environ Int 133: 105139.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 915990
Iwabuchi, K; Senzaki, N; Mazawa, D; Sato, I; Hara, M; Ueda, F; Liu, W; Tsuda, S. (2017). Tissue
toxicokinetics of perfluoro compounds with single and chronic low doses in male rats. J Toxicol
7-32
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Sci 42: 301-317. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859701
Jaacks, LM; Boyd Barr, D; Sundaram, R; Grewal, J; Zhang, C; Buck Louis, GM. (2016). Pre-Pregnancy
Maternal Exposure to Persistent Organic Pollutants and Gestational Weight Gain: A Prospective
Cohort Study. Int J Environ Res Public Health 13.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981711
Jackson-Browne, MS; Eliot, M; Patti, M; Spanier, AJ; Braun, JM. (2020). PFAS (per- and
polyfluoroalkyl substances) and asthma in young children: NHANES 2013-2014. Int J Hyg
Environ Health 229: 113565.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833598
Jain, R. (2013). Association between thyroid profile and perfluoroalkyl acids: Data from NHNAES 2007-
2008. Environ Res 126: 51-59.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2168068
Jain, RB. (2019). Concentration of selected liver enzymes across the stages of glomerular function: The
associations with PFOA and PFOS. Heliyon 5: e02168.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381541
Jain, RB. (2020). Associations between selected perfluoroalkyl acids in serum and hemoglobin in whole
blood, a biomarker of anemia: Impact of deteriorating kidney function. Environ Pollut 263:
114458. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6333438
Jain, RB. (2020). Impact of the co-occurrence of obesity with diabetes, anemia, hypertension, and
albuminuria on concentrations of selected perfluoroalkyl acids. Environ Pollut 266 Pt. 2: 115207.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833623
Jain, RB. (2020). Variabilities in concentrations of selected perfluoroalkyl acids among normotensives
and hypertensives across various stages of glomerular function. Arch Environ Occup Health 76:
1-11. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6311650
Jain, RB; Ducatman, A. (2018). Associations between lipid/lipoprotein levels and perfluoroalkyl
substances among US children aged 6-11 years. Environ Pollut 243: 1-8.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 07965 6
Jain, RB; Ducatman, A. (2019). Dynamics of associations between perfluoroalkyl substances and uric
acid across the various stages of glomerular function. Environ Sci Pollut Res Int 26: 12425-
12434. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080378
Jain, RB; Ducatman, A. (2019). Perfluoroalkyl acids and thyroid hormones across stages of kidney
function. Sci Total Environ 696: 133994.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315816
Jain, RB; Ducatman, A. (2019). Perfluoroalkyl acids serum concentrations and their relationship to
biomarkers of renal failure: Serum and urine albumin, creatinine, and albumin creatinine ratios
across the spectrum of glomerular function among US adults. Environ Res 174: 143-151.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5381566
Jain, RB; Ducatman, A. (2019). Roles of gender and obesity in defining correlations between
perfluoroalkyl substances and lipid/lipoproteins. Sci Total Environ 653: 74-81.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080642
Jain, RB; Ducatman, A. (2019). Selective associations of recent low concentrations of perfluoroalkyl
substances with liver function biomarkers: nhanes 2011 to 2014 data on us adults aged >20 years.
J Occup Environ Med 61: 293-302.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 0 80621
Jain, RB; Ducatman, A. (2020). Associations between apolipoprotein B and selected perfluoroalkyl
substances among diabetics and nondiabetics. Environ Sci Pollut Res Int 2020: 13819-13828.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988488
James, K; Peters, RE; Laird, BD; Ma, WK; Wickstrom, M; Stephenson, GL; Siciliano, SD. (2011).
Human exposure assessment: a case study of 8 PAH contaminated soils using in vitro digestors
and the juvenile swine model. Environ Sci Technol 45: 4586-4593.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6718854
7-33
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Janku, I. (1993). Physiological modelling of renal drug clearance. Eur J Clin Pharmacol 44: 513-519.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8630776
Jantzen, CE; Annunziato, KA; Bugel, SM; Cooper, KR. (2016). PFOS, PFNA, and PFOA sub-lethal
exposure to embryonic zebrafish have different toxicity profiles in terms of morphometries,
behavior and gene expression. Aquat Toxicol 175: 160-170.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860114
Jantzen, CE; Annunziato, KM; Cooper, KR. (2016). Behavioral, morphometric, and gene expression
effects in adult zebrafish (Danio rerio) embryonically exposed to PFOA, PFOS, and PFNA.
Aquat Toxicol 180: 123-130.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860109
Jantzen, CE; Toor, F; Annunziato, KA; Cooper, KR. (2017). Effects of chronic perfluorooctanoic acid
(PFOA) at low concentration on morphometries, gene expression, and fecundity in zebrafish
(Danio rerio). Reprod Toxicol 69: 34-42.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/3603831
Jeddy, Z; Hartman, TJ; Taylor, EV; Poteete, C; Kordas, K. (2017). Prenatal concentrations of
perfluoroalkyl substances and early communication development in British girls. Early Hum Dev
109: 15-20. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/3859807
Jeddy, Z; Tobias, JH; Taylor, EV; Northstone, K; Flanders, WD; Hartman, TJ. (2018). Prenatal
concentrations of perfluoroalkyl substances and bone health in British girls at age 17. Archives of
Osteoporosis 13: 84. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/5079850
Jensen, RC; Andersen, MS; Larsen, PV; Glintborg, D; Dalgard, C; Timmermann, CAG; Nielsen, F;
Sandberg, MB; Andersen, HR; Christesen, HT; Grandjean, P; Jensen, TK. (2020). Prenatal
Exposures to Perfluoroalkyl Acids and Associations with Markers of Adiposity and Plasma
Lipids in Infancy: An Odense Child Cohort Study. Environ Health Perspect 128: 77001.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/6833719
Jensen, RC; Glintborg, D; Gade Timmermann, CA; Nielsen, F; Kyhl, HB; Frederiksen, H; Andersson,
AM; Juul, A; Sidelmann, JJ; Andersen, HR; Grandjean, P; Andersen, MS; Jensen, TK. (2020).
Prenatal exposure to perfluorodecanoic acid is associated with lower circulating concentration of
adrenal steroid metabolites during mini puberty in human female infants. The odense child
cohort. Environ Res 182: 109101.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311643
Jensen, RC; Glintborg, D; Timmermann, CAG; Nielsen, F; Kyhl, HB; Andersen, HR; Grandjean, P;
Jensen, TK; Andersen, M. (2018). Perfluoroalkyl substances and glycemic status in pregnant
Danish women: The Odense Child Cohort. Environ Int 116: 101-107.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/4354143
Jensen, TK; Andersen, LB; Kyhl, HB; Nielsen, F; Christesen, HT; Grandjean, P. (2015). Association
between Perfluorinated Compound Exposure and Miscarriage in Danish Pregnant Women. PLoS
ONE 10: e0123496. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/2850253
Ji, J; Song, L; Wang, J; Yang, Z; Yan, H; Li, T; Yu, L; Jian, L; Jiang, F; Li, J; Zheng, J; Li, K. (2021).
Association between urinary per- and poly-fluoroalkyl substances and COVID-19 susceptibility.
Environ Int 153: 106524.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/7491706
Jiang, H; Liu, H; Liu, G; Yu, J; Liu, N; Jin, Y; Bi, Y; Wang, H. (2022). Associations between
Polyfluoroalkyl Substances Exposure and Breast Cancer: A Meta-Analysis [Review]. Toxics 10.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328207
Jiang, W; Deng, Y; Song, Z; Xie, Y; Gong, L; Chen, Y; Kuang, H. (2020). Gestational Perfluorooctanoic
Acid Exposure Inhibits Placental Development by Dysregulation of Labyrinth Vessels and uNK
Cells and Apoptosis in Mice. Front Physiol 11: 51.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/6320192
Jiang, W; Zhang, Y; Zhu, L; Deng, J. (2014). Serum levels of perfluoroalkyl acids (PFAAs) with isomer
analysis and their associations with medical parameters in Chinese pregnant women. Environ Int
7-34
-------
DRAFT FOR PUBLIC COMMENT
March 2023
64: 40-47. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850910
Jin, H; Zhang, Y; Jiang, W; Zhu, L; Martin, JW. (2016). Isomer-Specific Distribution of Perfluoroalkyl
Substances in Blood. Environ Sci Technol 50: 7808-7815.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859825
Jin, R; Mcconnell, R; Catherine, C; Xu, S; Walker, DI; Stratakis, N; Jones, DP; Miller, GW; Peng, C;
Conti, DV; Vos, MB; Chatzi, L. (2020). Perfluoroalkyl substances and severity of nonalcoholic
fatty liver in Children: An untargeted metabolomics approach. Environ Int 134: 105220.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315720
Joensen, UN; Bossi, R; Leffers, H; Jensen, AA; Skakkebsek, NE; Jorgensen. N. (2009). Do perfluoroalkyl
compounds impair human semen quality? Environ Health Perspect 117: 923-927.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1405085
Joensen, UN; Veyrand, B; Antignac, JP; Jensen, MB; Petersen, JH; Marchand, P; Skakkebaek, NE;
Andersson, AM; Le Bizec, B; Jorgensen, N. (2014). PFOS (perfluorooctanesulfonate) in serum is
negatively associated with testosterone levels, but not with semen quality, in healthy men. Hum
Reprod 29: 1600-1600.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851244
Johanson, CE. (1979). Distribution of fluid between extracellular and intracellular compartments in the
heart, lungs, liver and spleen of neonatal rats. 36: 282-289.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641334
Johansson, N; Eriksson, P; Viberg, H. (2009). Neonatal exposure to PFOS and PFOA in mice results in
changes in proteins which are important for neuronal growth and synaptogenesis in the
developing brain. Toxicol Sci 108: 412-418.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/757874
Johansson, N; Fredriksson, A; Eriksson, P. (2008). Neonatal exposure to perfluorooctane sulfonate
(PFOS) and perfluorooctanoic acid (PFOA) causes neurobehavioural defects in adult mice.
Neurotoxicology 29: 160-169.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1276156
Johnson, PI; Sutton, P; Atchley, DS; Koustas, E; Lam, J; Sen, S; Robinson, KA; Axelrad, DA; Woodruff,
TJ. (2014). The navigation guide - evidence-based medicine meets environmental health:
Systematic review of human evidence for PFOA effects on fetal growth [Review]. Environ
Health Perspect 122: 1028-1039.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851237
Jusko, TA; Oktapodas, M; Palkovicova Murinova, L; Babinska, K; Babjakova, J; Verner, MA; Dewitt,
JC; Thevenet-Morrison, K; Conka, K; Drobna, B; Chovancova, J; Thurston, SW; Lawrence, BP;
Dozier, AM; Jarvinen, KM; Patayova, H; Trnovec, T; Legler, J; Hertz-Picciotto, I; Lamoree, MH.
(2016). Demographic, reproductive, and dietary determinants of perfluorooctane sulfonic (PFOS)
and perfluorooctanoic acid (PFOA) concentrations in human colostrum. Environ Sci Technol 50:
7152-7162. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981718
Kamendulis, LM; Hocevar, JM; Stephens, M; Sandusky, GE; Hocevar, BA. (2022). Exposure to
perfluorooctanoic acid leads to promotion of pancreatic cancer. Carcinogenesis.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1017643 9
Kamendulis, LM; Wu, Q; Sandusky, GE; Hocevar, BA. (2014). Perfluorooctanoic acid exposure triggers
oxidative stress in the mouse pancreas. Toxicol Rep 1: 513-521.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080475
Kang, H; Choi, K; Lee, HS; Kim, DH; Park, NY; Kim, S; Kho, Y. (2016). Elevated levels of short
carbon-chain PFCAs in breast milk among Korean women: Current status and potential
challenges. Environ Res 148: 351-359.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859603
Kang, H; Lee, HK; Moon, HB; Kim, S; Lee, J; Ha, M; Hong, S; Kim, S; Choi, K. (2018). Perfluoroalkyl
acids in serum of Korean children: Occurrences, related sources, and associated health outcomes.
Sci Total Environ 645: 958-965.
7-35
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/493 7567
Kang, JS; Choi, JS; Park, JW. (2016). Transcriptional changes in steroidogenesis by perfluoroalkyl acids
(PFOA and PFOS) regulate the synthesis of sex hormones in H295R cells. Chemosphere 155:
436-443. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749062
Kang, Q; Gao, F; Zhang, X; Wang, L; Liu, J; Fu, M; Zhang, S; Wan, Y; Shen, H; Hu, J. (2020).
Nontargeted identification of per- and polyfluoroalkyl substances in human follicular fluid and
their blood-follicle transfer. Environ Int 139: 105686.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6356899
Kapraun, DF; Zurlinden, TJ; Verner, M-A; Chiang, C; Dzierlenga, MW; Carlson, LM; Schlosser, PM;
Lehmann, GM. (2022). A Generic Pharmacokinetic Model for Quantifying Mother-to-Offspring
Transfer of Lipophilic Persistent Environmental Chemicals.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641977
Karlsen, M; Grandjean, P; Weihe, P; Steuerwald, U; Oulhote, Y; Valvi, D. (2017). Early-life exposures to
persistent organic pollutants in relation to overweight in preschool children. Reprod Toxicol 68:
145-153. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858520
Karrman, A; Domingo, JL; Llebaria, X; Nadal, M; Bigas, E; van Bavel, B; Lindstrom, G. (2010).
Biomonitoring perfluorinated compounds in Catalonia, Spain: concentrations and trends in
human liver and milk samples. Environ Sci Pollut Res Int 17: 750-758.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2732071
Karrman, A; van Bavel, B; Jarnberg, U; Hardell, L; Lindstrom, G. (2006). Perfluorinated chemicals in
relation to other persistent organic pollutants in human blood. Chemosphere 64: 1582-1591.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2159543
Kataria, A; Trachtman, H; Malaga-Dieguez, L; Trasande, L. (2015). Association between perfluoroalkyl
acids and kidney function in a cross-sectional study of adolescents. Environ Health 14: 89.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859835
Kato, K; Wong, LY; Chen, A; Dunbar, C; Webster, GM; Lanphear, BP; Calafat, AM. (2014). Changes in
serum concentrations of maternal poly- and perfluoroalkyl substances over the course of
pregnancy and predictors of exposure in a multiethnic cohort of Cincinnati, Ohio pregnant
women during 2003-2006. Environ Sci Technol 48: 9600-9608.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851230
Kato, K; Wong, LY; Jia, LT; Kuklenyik, Z; Calafat, AM. (2011). Trends in exposure to polyfluoroalkyl
chemicals in the US population: 1999-2008. Environ Sci Technol 45: 8037-8045.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290883
Kato, S; Itoh, S; Yuasa, M; Baba, T; Miyashita, C; Sasaki, S; Nakajima, S; Uno, A; Nakazawa, H;
Iwasaki, Y; Okada, E; Kishi, R. (2016). Association of perfluorinated chemical exposure in utero
with maternal and infant thyroid hormone levels in the Sapporo cohort of Hokkaido Study on the
Environment and Children's Health. Environ Health Prev Med 21: 334-344.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981723
Kavlock, RJ; Allen, BC; Faustman, EM; Kimmel, CA. (1995). Dose-response assessments for
developmental toxicity. IV. Benchmark doses for fetal weight changes. Toxicol Sci 26: 211-222.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/75837
Kawabata, K; Matsuzaki, H; Nukui, S; Okazaki, M; Sakai, A; Kawashima, Y; Kudo, N. (2017).
Perfluorododecanoic acid induces cognitive deficit in adult rats. Toxicol Sci 157: 421-428.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858489
Kawamoto, K; Oashi, T; Oami, K; Liu, W; Jin, YH; Saito, N; Sato, I; Tsuda, S. (2010). Perfluorooctanoic
acid (PFOA) but not perfluorooctane sulfonate (PFOS) showed DNA damage in comet assay on
Paramecium caudatum. J Toxicol Sci 35: 835-841.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1274162
Keller & Heckman LLP. (2021). Attack on PFAS extends to food packaging. National Law Review X.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419081
Kemper, R. (2003). Perfluorooctanoic acid: Toxicokinetics in the rat. (DuPont 7473; US EPA Public
7-36
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Docket Administrative Record AR-226-1499). E.I. du Pont de Nemours and Company, Haskell
Laboratory for Health and Environmental Sciences.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6302380
Kennedy, GL. (1985). Dermal toxicity of ammonium perfluorooctanoate. Toxicol Appl Pharmacol 81:
348-355. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3797585
Kennedy, GL; Jr; Butenhoff, JL; Olsen, GW; Connor, JCO; Seacat, AM; Perkins, RG; Biegel, LB;
Murphy, S. R.; Farrar, DG. (2004). The toxicology of perfluorooctanoate [Review]. Crit Rev
Toxicol 34: 351-384. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/724950
Kerstner-Wood, C; Coward, L; Gorman, G; Southern Research Institute. (2003). Protein binding of
perfluorobutane sulfonate, perfluorohexane sulfonate, perfluorooctane sulfonate and
perfluorooctanoate to plasma (human, rat, and monkey), and various human-derived plasma
protein fractions. Study ID 9921.7 [TSCA Submission], (8EHQ-04-15845A; 88040000364). St.
Paul, MN: 3M Company.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4771364
Khalil, N; Chen, A; Lee, M; Czerwinski, SA; Ebert, JR; Dewitt, JC; Kannan, K. (2016). Association of
Perfluoroalkyl Substances, Bone Mineral Density, and Osteoporosis in the US Population in
NHANES 2009-2010. Environ Health Perspect 124: 81-87.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3229485
Khalil, N; Ducatman, AM; Sinari, S; Billheimer, D; Hu, C; Littau, S; Burgess, JL. (2020). Per- and
polyfluoroalkyl substance and cardio metabolic markers in firefighters. J Occup Environ Med 62:
1076-1081. https://hero.epa.gOv/hero/index.cfin/reference/details/reference_id/7021479
Khalil, N; Ebert, JR; Honda, M; Lee, M; Nahhas, RW; Koskela, A; Hangartner, T; Kannan, K. (2018).
Perfluoroalkyl substances, bone density, and cardio-metabolic risk factors in obese 8-12 year old
children: A pilot study. Environ Res 160: 314-321.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4238547
Khetsuriani, N; Zakikhany, K; Jabirov, S; Saparova, N; Ursu, P; Wannemuehler, K; Wassilak, S;
Efstratiou, A; Martin, R. (2013). Seroepidemiology of diphtheria and tetanus among children and
young adults in Tajikistan: nationwide population-based survey, 2010. 31: 4917-4922.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9642159
Ki, SH; Park, O; Zheng, M; Morales-Ibanez, O; Kolls, JK; Bataller, R; Gao, B. (2010). Interleukin-22
treatment ameliorates alcoholic liver injury in a murine model of chronic-binge ethanol feeding:
role of signal transducer and activator of transcription 3. Hepatology 52: 1291-1300.
https://hero .epa.gov/hero/index.cfim/reference/details/reference_id/10365730
Kielsen, K; Shamim, Z; Ryder, LP; Nielsen, F; Grandjean, P; Budtz-Jorgensen. E; Heilmann, C. (2016).
Antibody response to booster vaccination with tetanus and diphtheria in adults exposed to
perfluorinated alkylates. J Immunotoxicol 13: 270-273.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4241223
Kim, DH; Kim, UJ; Kim, HY; Choi, SD; Oh, JE. (2016). Perfluoroalkyl substances in serum from South
Korean infants with congenital hypothyroidism and healthy infants - Its relationship with thyroid
hormones. Environ Res 147: 399-404.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3351917
Kim, DH; Lee, JH; Oh, JE. (2019). Assessment of individual-based perfluoroalkly substances exposure
by multiple human exposure sources. J Hazard Mater 365: 26-33.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5080673
Kim, HC; Nam, CM; Jee, SH; Han, KH; Oh, DK; Suh, I. (2004). Normal serum aminotransferase
concentration and risk of mortality from liver diseases: prospective cohort study. Br Med J (Clin
Res Ed) 328: 983. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/10473876
Kim, HY; Kim, KN; Shin, CH; Lim, YH; Kim, JI; Kim, BN; Hong, YC; Lee, YA. (2020). The
relationship between perfluoroalkyl substances concentrations and thyroid function in early
childhood: A prospective cohort study. Thyroid 30: 1556-1565.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6833758
7-37
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Kim, JH; Park, HY; Jeon, JD; Kho, Y; Kim, SK; Park, MS; Hong, YC. (2015). The modifying effect of
vitamin C on the association between perfluorinated compounds and insulin resistance in the
Korean elderly: a double-blind, randomized, placebo-controlled crossover trial. Eur J Nutr 55:
1011-1020. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850129
Kim, MJ; Moon, S; Oh, BC; Jung, D; Ji, K; Choi, K; Park, YJ. (2018). Association between
perfluoroalkyl substances exposure and thyroid function in adults: A meta-analysis. PLoS ONE
13: eO 197244. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079795
Kim, RB. (2003). Organic anion-transporting polypeptide (OATP) transporter family and drug disposition
[Review]. Eur J Clin Invest 33 Suppl 2: 1-5.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641809
Kim, S; Choi, K; Ji, K; Seo, J; Kho, Y; Park, J; Kim, S; Park, S; Hwang, I; Jeon, J; Yang, H; Giesy, JP.
(2011). Trans-placental transfer of thirteen perfluorinated compounds and relations with fetal
thyroid hormones. Environ Sci Technol 45: 7465-7472.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1424975
Kim, SJ; Heo, SH; Lee, DS; Hwang, IG; Lee, YB; Cho, HY. (2016). Gender differences in
pharmacokinetics and tissue distribution of 3 perfluoroalkyl and polyfluoroalkyl substances in
rats. Food Chem Toxicol 97: 243-255.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749289
Kim, SK; Kannan, K. (2007). Perfluorinated acids in air, rain, snow, surface runoff, and lakes: relative
importance of pathways to contamination of urban lakes. Environ Sci Technol 41: 8328-8334.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289790
Kim, SK; Lee, KT; Kang, CS; Tao, L; Kannan, K; Kim, KR; Kim, CK; Lee, JS; Park, PS; Yoo, YW; Ha,
JY; Shin, YS; Lee, JH. (2011). Distribution of perfluorochemicals between sera and milk from
the same mothers and implications for prenatal and postnatal exposures. Environ Pollut 159: 169-
174. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919258
Kim, WR; Flamm, SL; Di Bisceglie, AM; Bodenheimer, HC. (2008). Serum activity of alanine
aminotransferase (ALT) as an indicator of health and disease. Hepatology 47: 1363-1370.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7757318
Kim, YR; White, N; Braunig, J; Vijayasarathy, S; Mueller, JF; Knox, CL; Harden, FA; Pacella, R; Toms,
LL. (2020). Per- and poly-fluoroalkyl substances (PFASs) in follicular fluid from women
experiencing infertility in Australia. Environ Res 190: 109963.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833596
Kimura, O; Fujii, Y; Haraguchi, K; Kato, Y; Ohta, C; Koga, N; Endo, T. (2017). Uptake of
perfluorooctanoic acid by Caco-2 cells: Involvement of organic anion transporting polypeptides.
Toxicol Lett 277: 18-23.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981330
Kingsley, SL; Kelsey, KT; Butler, R; Chen, A; Eliot, MN; Romano, ME; Houseman, A; Koestler, DC;
Lanphear, BP; Yolton, K; Braun, JM. (2017). Maternal serum PFOA concentration and DNA
methylation in cord blood: A pilot study. Environ Res 158: 174-178.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981315
Kingsley, SL; Walker, DI; Calafat, AM; Chen, A; Papandonatos, GD; Xu, Y; Jones, DP; Lanphear, BP;
Pennell, KD; Braun, JM. (2019). Metabolomics of childhood exposure to perfluoroalkyl
substances: across-sectional study. Metabolomics 15: 95.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5405904
Kishi, R; Nakajima, T; Goudarzi, H; Kobayashi, S; Sasaki, S; Okada, E; Miyashita, C; Itoh, S; Araki, A;
Ikeno, T; Iwasaki, Y; Nakazawa, H. (2015). The association of prenatal exposure to
perfluorinated chemicals with maternal essential and long-chain polyunsaturated fatty acids
during pregnancy and the birth weight of their offspring: the hokkaido study. Environ Health
Perspect 123: 1038-1045.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850268
Klaassen, CD. (2013). Casarett & Doull's toxicology: The basic science of poisons. In CD Klaassen (Ed.),
7-38
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(8th ed.). New York, NY: McGraw-Hill Education.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2993368
Klaassen, CD; Aleksunes, LM. (2010). Xenobiotic, bile acid, and cholesterol transporters: function and
regulation. Pharmacol Rev 62: 1-96.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641804
Klaassen, CD; Lu, H. (2008). Xenobiotic transporters: ascribing function from gene knockout and
mutation studies. Toxicol Sci 101: 186-196.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642044
Klamt, A; Huniar, U; Spycher, S; Keldenich, J. (2008). COSMOmic: a mechanistic approach to the
calculation of membrane-water partition coefficients and internal distributions within membranes
and micelles. J Phys Chem B 112: 12148-12157.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641966
Klaunig, JE; Babich, MA; Baetcke, KP; Cook, JC; Corton, JC; David, RM; Deluca, JG; Lai, DY; Mckee,
RH; Peters, JM; Roberts, RA; Fenner-Crisp, PA. (2003). PPARalpha agonist-induced rodent
tumors: modes of action and human relevance [Review]. Crit Rev Toxicol 33: 655-780.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5772415
Klaunig, JE; Hocevar, BA; Kamendulis, LM. (2012). Mode of Action analysis of perfluorooctanoic acid
(PFOA) tumorigenicity and Human Relevance [Review]. Reprod Toxicol 33: 410-418.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289837
Knox, SS; Jackson, T; Javins, B; Frisbee, SJ; Shankar, A; Ducatman, AM. (2011). Implications of early
menopause in women exposed to perfluorocarbons. J Clin Endocrinol Metab 96: 1747-1753.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1402395
Kobayashi, S; Azumi, K; Goudarzi, H; Araki, A; Miyashita, C; Kobayashi, S; Itoh, S; Sasaki, S; Ishizuka,
M; Nakazawa, H; Ikeno, T; Kishi, R. (2017). Effects of prenatal perfluoroalkyl acid exposure on
cord blood IGF2/H19 methylation and ponderal index: The Hokkaido Study. J Expo Sci Environ
Epidemiol 27: 251-259.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981430
Kobayashi, S; Sata, F; Goudarzi, H; Araki, A; Miyashita, C; Sasaki, S; Okada, E; Iwasaki, Y; Nakajima,
T; Kishi, R. (2021). Associations among perfluorooctanesulfonic/perfluorooctanoic acid levels,
nuclear receptor gene polymorphisms, and lipid levels in pregnant women in the Hokkaido study.
Sci Rep 11: 9994. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8442188
Konwick, BJ; Tomy, GT; Ismail, N; Peterson, JT; Fauver, RJ; Higginbotham, D; Fisk, AT. (2008).
Concentrations and patterns of perfluoroalkyl acids in Georgia, USA surface waters near and
distant to a major use source. Environ Toxicol Chem 27: 2011-2018.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291088
Koshy, TT; Attina, TM; Ghassabian, A; Gilbert, J; Burdine, LK; Marmor, M; Honda, M; Chu, DB; Han,
X; Shao, Y; Kannan, K; Urbina, EM; Trasande, L. (2017). Serum perfluoroalkyl substances and
cardiometabolic consequences in adolescents exposed to the World Trade Center disaster and a
matched comparison group. Environ Int 109: 128-135.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238478
Koskela, A; Ducatman, A; Schousboe, JT; Nahhas, RW; Khalil, N. (2022). Perfluoroalkyl Substances and
Abdominal Aortic Calcification. J Occup Environ Med.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/10176386
Kotlarz, N; Mccord, J; Collier, D; Lea, CS; Strynar, M; Lindstrom, AB; Wilkie, AA; Islam, JY; Matney,
K; Tarte, P; Polera, ME; Burdette, K; Dewitt, J; May, K; Smart, RC; Knappe, DRU; Hoppin, JA.
(2020). Measurement of Novel, Drinking Water-Associated PFAS in Blood from Adults and
Children in Wilmington, North Carolina. Environ Health Perspect 128: 77005.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833715
Kotthoff, M; Miiller, J; Jiirling, H; Schlummer, M; Fiedler, D. (2015). Perfluoroalkyl and polyfluoroalkyl
substances in consumer products. Environ Sci Pollut Res Int 22: 14546-14559.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850246
7-39
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Kuczmarski, RJ; Ogden, CL; Guo, SS; Grummer-Strawn, LM; Flegal, KM; Mei, Z; Wei, R; Curtin, LR;
Roche, AF; Johnson, CL. (2002). 2000 CDC growth charts for the United States: Methods and
development. (Vital and Health Statistics: Series 11, No. 246). Hyattsville, MD: National Center
for Health Statistics, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3490881
Kudo, N; Katakura, M; Sato, Y; Kawashima, Y. (2002). Sex hormone-regulated renal transport of
perfluorooctanoic acid. Chem Biol Interact 139: 301-316.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2990271
Kullak-Ublick, G, .A.; Hagenbuch, B, .; Stieger, B, .; Schteingart, C, .D.; Hofmann, A, .F.; Wolkoff, A,
.W.; Meier, P, .J. (1995). Molecular and functional characterization of an organic anion
transporting polypeptide cloned from human liver. 109: 1274-1282.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641965
Kummu, M; Sieppi, E; Koponen, J; Laatio, L; VaoHaoKangas. K; Kiviranta, H; Rautio, A; Myllynen, P.
(2015). Organic anion transporter 4 (OAT 4) modifies placental transfer of perfluorinated alkyl
acids PFOS and PFOA in human placental ex vivo perfusion system. Placenta 36: 1185-1191.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3789332
Kusuhara, H; Sugiyama, Y. (2009). In vitro-in vivo extrapolation of transporter-mediated clearance in the
liver and kidney [Review]. Drug Metab Pharmacokinet 24: 37-52.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641810
Kvalem, HE; Nygaard, UC; Lodrup Carlsen, KC; Carlsen, KH; Haug, LS; Granum, B. (2020).
Perfluoroalkyl substances, airways infections, allergy and asthma related health outcomes -
Implications of gender, exposure period and study design. Environ Int 134: 105259.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316210
Kvist, L; Giwercman, YL; Jonsson, BA; Lindh, CH; Bonde, JP; Toft, G; Strucinski, P; Pedersen, HS;
Zvyezday, V; Giwercman, A. (2012). Serum levels of perfluorinated compounds and sperm Y:X
chromosome ratio in two European populations and in Inuit from Greenland. Reprod Toxicol 34:
644-650. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919170
Kwo, PY; Cohen, SM; Lim, JK. (2017). ACG Clinical Guideline: Evaluation of Abnormal Liver
Chemistries. Am J Gastroenterol 112: 18-35.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328876
Lange, F; Schmidt, C; Brauch, H-J. (2006). Perfluoroalkylcarboxylates and -sulfonates: Emerging
Contaminants for Drinking Water Supplies? Nieuwegein, The Netherlands: Association of River
Waterworks - RIWA.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/104113 76
Lau, C; Thibodeaux, JR; Hanson, RG; Narotsky, MG; Rogers, JM; Lindstrom, AB; Strynar, MJ. (2006).
Effects of perfluorooctanoic acid exposure during pregnancy in the mouse. Toxicol Sci 90: 510-
518. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1276159
Launay-Vacher, V; Izzedine, H; Karie, S; Hulot, JS; Baumelou, A; Deray, G. (2006). Renal tubular drug
transporters. Nephron Physiol 103: p97-106.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641802
Lauritzen, HB; Larose, TL; 0ien, T; Sandanger, TM; Odland, J0; van de Bor, M; Jacobsen, GW. (2018).
Prenatal exposure to persistent organic pollutants and child overweight/obesity at 5-year follow-
up: a prospective cohort study. Environ Health 17: 9.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4217244
Lawlor, TE. (1995). Mutagenicity test with T-6342 in the Salmonella-Escherichia coli/mammalian-
microsome reverse mutation assay. (CHV Study No. 17073-0-409; EPA-AR-226-0436). Vienna,
VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228128
Lawlor, TE. (1996). Mutagenicity test with T-6564 in the Salmonella-Escherichia coli/mammalian-
microsome reverse mutation assay with a confirmatory assay. (CHV Study No. 17750-0-409R;
EPA-AR-226-0432). Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228127
7-40
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Laws, SC; Stoker, TE; Ferrell, JM; Hotchkiss, MG; Cooper, RL. (2007). Effects of altered food intake
during pubertal development in male and female wistar rats. Toxicol Sci 100: 194-202.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1411456
Leary, DB; Takazawa, M; Kannan, K; Khalil, N. (2020). Perfluoroalkyl substances and metabolic
syndrome in firefighters a pilot study. J Occup Environ Med 62: 52-57.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7240043
Lebeaux, RM; Doherty, BT; Gallagher, LG; Zoeller, RT; Hoofnagle, AN; Calafat, AM; Karagas, MR;
Yolton, K; Chen, A; Lanphear, BP; Braun, JM; Romano, ME. (2020). Maternal serum
perfluoroalkyl substance mixtures and thyroid hormone concentrations in maternal and cord sera:
The HOME Study. Environ Res 185: 109395.
https://hero .epa.gov/hero/index.cfim/reference/details/reference_id/63 5 63 61
Lee, J; Oh, S; Kang, H; Kim, S; Lee, G; Li, L; Kim, CT; An, JN; Oh, YK; Lim, CS; Kim, DK; Kim, YS;
Choi, K; Lee, JP. (2020). Environment-wide association study of CKD. Clin J Am Soc Nephrol
15: 766-775. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/6833761
Lee, JK; Lee, S; Baek, MC; Lee, BH; Lee, HS; Kwon, TK; Park, PH; Shin, TY; Khang, D; Kim, SH.
(2017). Association between perfluorooctanoic acid exposure and degranulation of mast cells in
allergic inflammation. J Appl Toxicol 37: 554-562.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3981419
Lee, JS; Ward, WO; Liu, J; Ren, H; Vallanat, B; Delker, D; Corton, JC. (2011). Hepatic xenobiotic
metabolizing enzyme and transporter gene expression through the life stages of the mouse. PLoS
ONE 6: e24381. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3114850
Lee, JW; Choi, K; Park, K; Seong, C; Yu, SD; Kim, P. (2020). Adverse effects of perfluoroalkyl acids on
fish and other aquatic organisms: A review [Review]. Sci Total Environ 707: 135334.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6323794
Lee, S; Kim, S; Park, J; Kim, HJ; Choi, G; Choi, S; Kim, S; Kim, SY; Kim, S; Choi, K; Moon, HB.
(2017). Perfluoroalkyl substances (PFASs) in breast milk from Korea: Time-course trends,
influencing factors, and infant exposure. Sci Total Environ 612: 286-292.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3983576
Lee, TH; Kim, WR; Benson, JT; Therneau, TM; Melton, LJ. (2008). Serum aminotransferase activity and
mortality risk in a United States community. Hepatology 47: 880-887.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/10293233
Lee, YJ; Kim, MK; Bae, J; Yang, JH. (2013). Concentrations of perfluoroalkyl compounds in maternal
and umbilical cord sera and birth outcomes in Korea. Chemosphere 90: 1603-1609.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3859850
Lehmann, GM; Verner, MA; Luukinen, B; Henning, C; Assimon, SA; Lakind, JS; Mclanahan, ED;
Phillips, LJ; Davis, MH; Powers, CM; Hines, EP; Haddad, S; Longnecker, MP; Poulsen, MT;
Farrer, DG; Marchitti, SA; Tan, YM; Swartout, JC; Sagiv, SK; Welsh, C; Campbell, JL; Foster,
WG; Yang, RS; Fenton, SE; Tornero-Velez, R; Francis, BM; Barnett, JB; El-Masri, HA;
Simmons, JE. (2014). Improving the risk assessment of lipophilic persistent environmental
chemicals in breast milk [Review]. Crit Rev Toxicol 44: 600-617.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2447276
Lehner, R; Quiroga, AD. (2016). Chapter 5: Fatty acid handling in mammalian cells. In ND Ridgway; RS
McLeod (Eds.), Biochemistry of lipids, lipoproteins and membranes (6th ed., pp. 149-184).
Amsterdam, Netherlands: Elsevier.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/10284974
Lenters, V; Iszatt, N; Forns, J; Cechova, E; Kocan, A; Legler, J; Leonards, P; Stigum, H; Eggesbo. M.
(2019). Early-life exposure to persistent organic pollutants (OCPs, PBDEs, PCBs, PFASs) and
attention-deficit/hyperactivity disorder: A multi-pollutant analysis of a Norwegian birth cohort.
Environ Int 125: 33-42.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5080366
Lenters, V; Portengen, L; Rignell-Hydbom, A; Jonsson, BA; Lindh, CH; Piersma, AH; Toft, G; Bonde,
7-41
-------
DRAFT FOR PUBLIC COMMENT
March 2023
JP; Heederik, D; Rylander, L; Vermeulen, R. (2016). Prenatal phthalate, perfluoroalkyl acid, and
organochlorine exposures and term birth weight in three birth cohorts: multi-pollutant models
based on elastic net regression. Environ Health Perspect 124: 365-372.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5617416
Leonard, RC; Kreckmann, KH; Sakr, CJ; Symons, JM. (2008). Retrospective cohort mortality study of
workers in a polymer production plant including a reference population of regional workers. Ann
Epidemiol 18: 15-22. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291100
Leter, G; Consales, C; Eleuteri, P; Uccelli, R; Specht, 10; Toft, G; Moccia, T; Budillon, A; Jonsson, BA;
Lindh, CH; Giwercman, A; Pedersen, HS; Ludwicki, JK; Zviezdai, V; Heederik, D; Bonde, JP;
Spano, M. (2014). Exposure to perfluoroalkyl substances and sperm DNA global methylation in
Arctic and European populations. Environ Mol Mutagen 55: 591-600.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2967406
Leung, YK; Ouyang, B; Niu, L; Xie, C; Ying, J; Medvedovic, M; Chen, A; Weihe, P; Valvi, D;
Grandjean, P; Ho, SM. (2018). Identification of sex-specific DNA methylation changes driven by
specific chemicals in cord blood in a Faroese birth cohort. Epigenetics 13: 290-300.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4633577
Lewis, RC; Johns, LE; Meeker, JD. (2015). Serum Biomarkers of Exposure to Perfluoroalkyl Substances
in Relation to Serum Testosterone and Measures of Thyroid Function among Adults and
Adolescents from NHANES 2011-2012. Int J Environ Res Public Health 12: 6098-6114.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3749030
Li, D; Song, P; Liu, L; Wang, X. (2018). Perfluorooctanoic acid exposure during pregnancy alters the
apoptosis of uterine cells in pregnant mice. Int J Clin Exp Pathol 11: 5602-5611.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5084746
Li, D; Zhang, L; Zhang, Y; Guan, S; Gong, X; Wang, X. (2019). Maternal exposure to perfluorooctanoic
acid (PFOA) causes liver toxicity through PPAR-a pathway and lowered histone acetylation in
female offspring mice. Environ Sci Pollut Res Int 26: 18866-18875.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387402
Li, H; Hammarstrand, S; Midberg, B; Xu, Y; Li, Y; Olsson, DS; Fletcher, T; Jakobsson, K; Andersson,
EM. (2022). Cancer incidence in a Swedish cohort with high exposure to perfluoroalkyl
substances in drinking water. Environ Res 204: 112217.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9961926
Li, J; Cai, D; Chu, C; Li, QQ; Zhou, Y; Hu, LW; Yang, BY; Dong, GH; Zeng, XW; Chen, D. (2020).
Transplacental Transfer of Per- and Polyfluoroalkyl Substances (PFASs): Differences between
Preterm and Full-Term Deliveries and Associations with Placental Transporter mRNA
Expression. Environ Sci Technol 54: 5062-5070.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505874
Li, K; Gao, P; Xiang, P; Zhang, X; Cui, X; Ma, LQ. (2017). Molecular mechanisms of PFOA-induced
toxicity in animals and humans: Implications for health risks [Review]. Environ Int 99: 43-54.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981403
Li, K; Li, C; Yu, NY; Juhasz, AL; Cui, XY; Ma, LQ. (2015). In vivo bioavailability and in vitro
bioaccessibility of perfluorooctanoic acid (PFOA) in food matrices: correlation analysis and
method development. Environ Sci Technol 49: 150-158.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851033
Li, K; Sun, J; Yang, J; Roberts, SM; Zhang, X; Cui, X; Wei, S; Ma, LQ. (2017). Molecular Mechanisms
of Perfluorooctanoate-Induced Hepatocyte Apoptosis in Mice Using Proteomic Techniques.
Environ Sci Technol 51: 11380-11389.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238518
Li, L; Zheng, H; Wang, T; Cai, M; Wang, P. (2018). Perfluoroalkyl acids in surface seawater from the
North Pacific to the Arctic Ocean: Contamination, distribution and transportation. Environ Pollut
238: 168-176. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080354
Li, MC. (2020). Serum Per- and Polyfluoroalkyl Substances Are Associated with Increased Hearing
7-42
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Impairment: A Re-Analysis of the National Health and Nutrition Examination Survey Data. Int J
Environ Res Public Health 17.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833686
Li, N; Liu, Y; Papandonatos, GD; Calafat, AM; Eaton, CB; Kelsey, KT; Cecil, KM; Kalkwarf, HJ;
Yolton, K; Lanphear, BP; Chen, A; Braun, JM. (2021). Gestational and childhood exposure to
per- and polyfluoroalkyl substances and cardiometabolic risk at age 12 years. Environ Int 147:
106344. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7404102
Li, X; Bao, C; Ma, Z; Xu, B; Ying, X; Liu, X; Zhang, X. (2018). Perfluorooctanoic acid stimulates
ovarian cancer cell migration, invasion via ERK/NF-kB/MMP-2/-9 pathway. Toxicol Lett 294:
44-50. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079796
Li, X; Wang, Z; Klaunig, JE. (2019). The effects of perfluorooctanoate on high fat diet induced non-
alcoholic fatty liver disease in mice. Toxicology 416: 1-14.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080362
Li, Y; Barregard, L; Xu, Y; Scott, K; Pineda, D; Lindh, CH; Jakobsson, K; Fletcher, T. (2020).
Associations between perfluoroalkyl substances and serum lipids in a Swedish adult population
with contaminated drinking water. Environ Health 19: 33.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315681
Li, Y; Cheng, 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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3856460
Li, Y; Fletcher, T; Mucs, D; Scott, K; Lindh, CH; Tallving, P; Jakobsson, K. (2018). Half-lives of PFOS,
PFHxS and PFOA after end of exposure to contaminated drinking water. Occup Environ Med 75:
46-51. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238434
Li, Y, ing; Mucs, D, aniel; Scott, K, ristin; Lindh, C, hristian; Tallving, P, ia; Fletcher, T, ony; Jakobsson,
K, ristina. (2017). Half-lives of PFOS, PFHxS and PFOA after end of exposure to contaminated
drinking water. (2:2017). Gothenburg, Sweden: Gothenburg University, Unit for Occupational &
Environmental Medicine.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641333
Li, Y; Oliver, DP; Kookana, RS. (2018). A critical analysis of published data to discern the role of soil
and sediment properties in determining sorption of per and polyfluoroalkyl substances (PFASs)
[Review]. Sci Total Environ 628-629: 110-120.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238331
Li, Y; Ramdhan, DH; Naito, H; Yamagishi, N; Ito, Y; Hayashi, Y; Yanagiba, Y; Okamura, A; Tamada,
H; Gonzalez, FJ; Nakajima, T. (2011). Ammonium perfluorooctanoate may cause testosterone
reduction by adversely affecting testis in relation to PPARa. Toxicol Lett 205: 265-272.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1294081
Li, Y; Yu, N; Du, L; Shi, W; Yu, H; Song, M; Wei, S. (2020). Transplacental Transfer of Per- and
Polyfluoroalkyl Substances Identified in Paired Maternal and Cord Sera Using Suspect and
Nontarget Screening. Environ Sci Technol 54: 3407-3416.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6506038
Li, ZR; Hromchak, R; Mudipalli, A; Bloch, A. (1998). Tumor suppressor proteins as regulators of cell
differentiation. Cancer Res 58: 4282-4287.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/598342
Liang, JL; Tiwari, T; Moro, P; Messonnier, NE; Reingold, A; Sawyer, M; Clark, TA. (2018). Prevention
of pertussis, tetanus, and diphtheria with vaccines in the United States: Recommendations of the
Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 67: 1-44.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9978483
Liang, L; Pan, Y; Bin, L; Liu, Y; Huang, W; Li, R; Lai, KP. (2021). Immunotoxicity mechanisms of
perfluorinated compounds PFOA and PFOS [Review]. Chemosphere 291: 132892.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959458
Liang, X; Xie, G; Wu, X; Su, M; Yang, B. (2019). Effect of prenatal PFOS exposure on liver cell
7-43
-------
DRAFT FOR PUBLIC COMMENT
March 2023
function in neonatal mice. Environ Sci Pollut Res Int 26: 18240-18246.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412467
Liao, S; Yao, W; Cheang, I; Tang, X; Yin, T; Lu, X; Zhou, Y; Zhang, H; Li, X. (2020). Association
between perfluoroalkyl acids and the prevalence of hypertension among US adults. Ecotoxicol
Environ Saf 196: 110589.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6356903
Lien, GW; Huang, CC; Shiu, JS; Chen, MH; Hsieh, WS; Guo, YL; Chen, PC. (2016). Perfluoroalkyl
substances in cord blood and attention deficit/hyperactivity disorder symptoms in seven-year-old
children. Chemosphere 156: 118-127.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860112
Liew, Z; Luo, J; Nohr, EA; Bech, BH; Bossi, R; Arah, OA; Olsen, J. (2020). Maternal Plasma
Perfluoroalkyl Substances and Miscarriage: A Nested Case-Control Study in the Danish National
Birth Cohort. Environ Health Perspect 128: 47007.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6387285
Liew, Z; Ritz, B; Bach, CC; Asarnow, RF; Bech, BH; Nohr, EA; Bossi, R; Henriksen, TB; Bonefeld-
Jorgensen. EC; Olsen, J. (2018). Prenatal exposure to perfluoroalkyl substances and iq scores at
age 5; a study in the danish national birth cohort. Environ Health Perspect 126: 067004.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079744
Liew, Z; Ritz, B; Bonefeld-Jorgensen, EC; Henriksen, TB; Nohr, EA; Bech, BH; Fei, C; Bossi, R; von
Ehrenstein, OS; Streja, E; Uldall, P; Olsen, J. (2014). Prenatal exposure to perfluoroalkyl
substances and the risk of congenital cerebral palsy in children. Am J Epidemiol 180: 574-581.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2852208
Liew, Z; Ritz, B; von Ehrenstein, OS; Bech, BH; Nohr, EA; Fei, C; Bossi, R; Henriksen, TB; Bonefeld-
Jorgensen. EC; Olsen, J. (2015). Attention deficit/hyperactivity disorder and childhood autism in
association with prenatal exposure to perfluoroalkyl substances: A nested case-control study in
the Danish National Birth Cohort. Environ Health Perspect 123: 367-373.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851010
Lin, CY; Chen, PC; Lin, YC; Lin, LY. (2009). Association among serum perfluoroalkyl chemicals,
glucose homeostasis, and metabolic syndrome in adolescents and adults. Diabetes Care 32: 702-
707. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290820
Lin, CY; Chen, PC; Lo, SC; Torng, PL; Sung, FC; Su, TC. (2016). The association of carotid intima-
media thickness with serum Level of perfluorinated chemicals and endothelium-platelet
microparticles in adolescents and young adults. Environ Int 94: 292-299.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981457
Lin, CY; Lee, HL; Hwang, YT; Su, TC. (2020). The association between total serum isomers of per- and
polyfluoroalkyl substances, lipid profiles, and the DNA oxidative/nitrative stress biomarkers in
middle-aged Taiwanese adults. Environ Res 182: 109064.
https://hero .epa.gov/hero/index.cfim/reference/details/reference_id/6315756
Lin, CY; Lin, LY; Chiang, CK; Wang, WJ; Su, YN; Hung, KY; Chen, PC. (2010). Investigation of the
Associations Between Low-Dose Serum Perfluorinated Chemicals and Liver Enzymes in US
Adults. Am J Gastroenterol 105: 1354-1363.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/1291111
Lin, CY; Lin, LY; Wen, TW; Lien, GW; Chien, KL; Hsu, SH; Liao, CC; Sung, FC; Chen, PC; Su, TC.
(2013). Association between levels of serum perfluorooctane sulfate and carotid artery intima-
media thickness in adolescents and young adults. Int J Cardiol 168: 3309-3316.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2850967
Lin, CY; Wen, LL; Lin, LY; Wen, TW; Lien, GW; Hsu, SH; Chien, KL; Liao, CC; Sung, FC; Chen, PC;
Su, TC. (2013). The associations between serum perfluorinated chemicals and thyroid function in
adolescents and young adults. J Hazard Mater 244-245: 637-644.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332458
Lin, HW; Feng, HX; Chen, L; Yuan, XJ; Tan, Z. (2020). Maternal exposure to environmental endocrine
7-44
-------
DRAFT FOR PUBLIC COMMENT
March 2023
disruptors during pregnancy is associated with pediatric germ cell tumors. Nagoya J Med Sci 82:
323-333. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6835434
Lin, LY; Wen, LL; Su, TC; Chen, PC; Lin, CY. (2014). Negative association between serum
perfluorooctane sulfate concentration and bone mineral density in US premenopausal women:
NHANES, 2005-2008. J Clin Endocrinol Metab 99: 2173-2180.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079772
Lin, P; Cardenas, A; Hauser, R; Gold, DR; Kleinman, K; Hivert, MF; Fleisch, AF; Calafat, AM; Webster,
TF; Horton, ES; Oken, E. (2019). Per- and polyfluoroalkyl substances and blood lipid levels in
pre-diabetic adults-longitudinal analysis of the diabetes prevention program outcomes study.
Environ Int 129: 343-353.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5187597
Lin, PD; Cardenas, A; Hauser, R; Gold, DR; Kleinman, KP; Hivert, MF; Calafat, AM; Webster, TF;
Horton, ES; Oken, E. (2020). Per- and polyfluoroalkyl substances and blood pressure in pre-
diabetic adults-cross-sectional and longitudinal analyses of the diabetes prevention program
outcomes study. Environ Int 137: 105573.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311641
Lin, T; Zhang, Y; Ding, X; Huang, T; Zhang, W; Zou, W; Kuang, H; Yang, B; Wu, L; Zhang, D. (2020).
Perfluorooctanoic acid induces cytotoxicity in spermatogonial GC-1 cells. Chemosphere 260:
127545. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833675
Lin, TW; Chen, MK; Lin, CC; Chen, MH; Tsai, MS; Chan, DC; Hung, KY; Chen, PC. (2020).
Association between exposure to perfluoroalkyl substances and metabolic syndrome and related
outcomes among older residents living near a Science Park in Taiwan. Int J Hyg Environ Health
230: 113607. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988476
Lind, DV; Priskorn, L; Lassen, TH; Nielsen, F; Kyhl, HB; Kristensen, DM; Christesen, HT; Jorgensen.
JS; Grandjean, P; Jensen, TK. (2017). Prenatal exposure to perfluoroalkyl substances and
anogenital distance at 3 months of age in a Danish mother-child cohort. Reprod Toxicol 68: 200-
206. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858512
Lind, L; Zethelius, B; Salihovic, S; van Bavel, B; Lind, PM. (2014). Circulating levels of perfluoroalkyl
substances and prevalent diabetes in the elderly. Diabetologia 57: 473-479.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/2215376
Lind, PM; Salihovic, S; van Bavel, B; Lind, L. (2017). Circulating levels of perfluoroalkyl substances
(PFASs) and carotid artery atherosclerosis. Environ Res 152: 157-164.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858504
Lindstrom, AB; Strynar, MJ; Libelo, EL. (2011). Polyfluorinated compounds: past, present, and future
[Review]. Environ Sci Technol 45: 7954-7961.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290802
Liu, CY; Chen, PC; Lien, PC; Liao, YP. (2018). Prenatal perfluorooctyl sulfonate exposure and Alu DNA
hypomethylation in cord blood. Int J Environ Res Public Health 15: 1066.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4926233
Liu, G; Dhana, K; Furtado, JD; Rood, J; Zong, G; Liang, L; Qi, L; Bray, GA; Dejonge, L; Coull, B;
Grandjean, P; Sun, Q. (2018). Perfluoroalkyl substances and changes in body weight and resting
metabolic rate in response to weight-loss diets: A prospective study. PLoS Med 15: el002502.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238396
Liu, G; Zhang, B; Hu, Y; Rood, J; Liang, L; Qi, L; Bray, GA; Dejonge, L; Coull, B; Grandjean, P;
Furtado, JD; Sun, Q. (2020). Associations of Perfluoroalkyl substances with blood lipids and
Apolipoproteins in lipoprotein subspecies: the POUNDS-lost study. Environ Health 19: 5.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6318644
Liu, H; Chen, Q; Lei, L; Zhou, W; Huang, L; Zhang, J; Chen, D. (2018). Prenatal exposure to
perfluoroalkyl and polyfluoroalkyl substances affects leukocyte telomere length in female
newborns. Environ Pollut235: 446-452.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239494
7-45
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Liu, H; Pan, Y; Jin, S; Li, Y; Zhao, L; Sun, X; Cui, Q; Zhang, B; Zheng, T; Xia, W; Zhou, A; Campana,
AM; Dai, J; Xu, S. (2020). Associations of per-/polyfluoroalkyl substances with glucocorticoids
and progestogens in newborns. Environ Int 140: 105636.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6569227
Liu, H; Wang, J; Sheng, N; Cui, R; Pan, Y; Dai, J. (2017). Acotl is a sensitive indicator for PPARa
activation after perfluorooctanoic acid exposure in primary hepatocytes of Sprague-Dawley rats.
Toxicol In Vitro 42: 299-307.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981337
Liu, HS; Wen, LL; Chu, PL; Lin, CY. (2018). Association among total serum isomers of perfluorinated
chemicals, glucose homeostasis, lipid profiles, serum protein and metabolic syndrome in adults:
NHANES, 2013-2014. Environ Pollut 232: 73-79.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4238514
Liu, J; Li, J; Liu, Y; Chan, HM; Zhao, Y; Cai, Z; Wu, Y. (2011). Comparison on gestation and lactation
exposure of perfluorinated compounds for newborns. Environ Int 37: 1206-1212.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2919240
Liu, J; Liu, S; Huang, Z; Fu, Y; Fei, J; Liu, X; He, Z. (2020). Associations between the serum levels of
PFOS/PFOA and IgG N-glycosylation in adult or children. Environ Pollut 265: 114285.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6833599
Liu, M; Zhang, G; Meng, L; Han, X; Li, Y; Shi, Y; Li, A; Turyk, ME; Zhang, Q; Jiang, G. (2021).
Associations between novel and legacy per- and polyfluoroalkyl substances in human serum and
thyroid cancer: A case and healthy population in Shandong Province, East China. Environ Sci
Technol. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/10176563
Liu, P; Yang, F; Wang, Y; Yuan, Z. (2018). Perfluorooctanoic acid (PFOA) exposure in early life
increases risk of childhood adiposity: a meta-analysis of prospective cohort studies. Int J Environ
Res Public Health 15. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/5079852
Liu, QS; Hao, F; Sun, Z; Long, Y; Zhou, Q; Jiang, G. (2018). Perfluorohexadecanoic acid increases
paracellular permeability in endothelial cells through the activation of plasma kallikrein-kinin
system. Chemosphere 190: 191-200.
https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/4238499
Liu, QS; Sun, Y; Qu, G; Long, Y; Zhao, X; Zhang, A; Zhou, Q; Hu, L; Jiang, G. (2017). Structure-
Dependent Hematological Effects of Per- and Polyfluoroalkyl Substances on Activation of
Plasma Kallikrein-Kinin System Cascade. Environ Sci Technol 51: 10173-10183.
https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/4238579
Liu, RC; Hurtt, ME; Cook, JC; Biegel, LB. (1996). Effect of the peroxisome proliferator, ammonium
perfluorooctanoate (C8), on hepatic aromatase activity in adult male Crl:CD BR (CD) rats.
Fundam Appl Toxicol 30: 220-228.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/1307751
Liu, W; Irudayaraj, J. (2020). Perfluorooctanoic acid (PFOA) exposure inhibits DNA methyltransferase
activities and alters constitutive heterochromatin organization. Food Chem Toxicol 141: 111358.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6512127
Liu, W; Xu, C; Sun, X, i; Kuang, H; Kuang, X; Zou, W; Yang, B, ei; Wu, L, ei; Liu, F; Zou, T; Zhang, D.
(2016). Grape seed proanthocyanidin extract protects against perfluorooctanoic acid-induced
hepatotoxicity by attenuating inflammatory response, oxidative stress and apoptosis in mice.
Toxicology Research 5: 224-234.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3981762
Liu, W; Yang, B; Wu, L; Zou, W; Pan, X; Zou, T; Liu, F; Xia, L; Wang, X; Zhang, D. (2015).
Involvement of NRF2 in Perfluorooctanoic Acid-Induced Testicular Damage in Male Mice. Biol
Reprod 93: 41. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/3981571
Liu, X; Guo, Z; Krebs, KA; Pope, RH; Roache, NF. (2014). Concentrations and trends of perfluorinated
chemicals in potential indoor sources from 2007 through 2011 in the US. Chemosphere 98: 51-
57. https://hero.epa.gov/hero/index.cftn/reference/details/reference_id/2324799
7-46
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Liu, X; Zhang, L; Chen, L; Li, J; Wang, Y; Wang, J; Meng, G; Chi, M; Zhao, Y; Chen, H; Wu, Y. (2019).
Structure-based investigation on the association between perfluoroalkyl acids exposure and both
gestational diabetes mellitus and glucose homeostasis in pregnant women. Environ Int 127: 85-
93. https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5881135
Liu, Z; Que, S; Xu, J; Peng, T. (2014). Alanine aminotransferase-old biomarker and new concept: a
review. Int J Med Sci 11: 925-935.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10473988
Lloyd-Jones, DM; Huffman, MD; Karmali, KN; Sanghavi, DM; Wright, JS; Pelser, C; Gulati, M;
Masoudi, FA; Goff, DC. (2017). Estimating Longitudinal Risks and Benefits From
Cardiovascular Preventive Therapies Among Medicare Patients: The Million Hearts Longitudinal
ASCVD Risk Assessment Tool: A Special Report From the American Heart Association and
American College of Cardiology [Review]. Circulation 135: e793-e813.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10694407
Loccisano, AE; Campbell, JL, Jr; Andersen, ME; Clewell, HJ, III. (2011). Evaluation and prediction of
pharmacokinetics of PFOA and PFOS in the monkey and human using a PBPK model. Regul
Toxicol Pharmacol 59: 157-175.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/787186
Loccisano, AE; Campbell, JL, Jr; Butenhoff, JL; Andersen, ME; Clewell, HJ, III. (2012). Comparison and
evaluation of pharmacokinetics of PFOA and PFOS in the adult rat using a physiologically based
pharmacokinetic model. Reprod Toxicol 33: 452-467.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289830
Loccisano, AE; Campbell, JL; Butenhoff, JL; Andersen, ME; Clewell, HJ. (2012). Evaluation of placental
and lactational pharmacokinetics of PFOA and PFOS in the pregnant, lactating, fetal and neonatal
rat using a physiologically based pharmacokinetic model. Reprod Toxicol 33: 468-490.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289833
Loccisano, AE; Longnecker, MP; Campbell, JL, Jr; Andersen, ME; Clewell, HJ, III. (2013). Development
of pbpk models for pfoa and pfos for human pregnancy and lactation life stages. J Toxicol
Environ Health A 76: 25-57.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1326665
Long, M; Ghisari, M; Kjeldsen, L; Wielsoe. M; Norgaard-Pedcrsen. B; Mortensen, EL; Abdallah, MW;
Bonefcld-Jorgenscn. EC. (2019). Autism spectrum disorders, endocrine disrupting compounds,
and heavy metals in amniotic fluid: a case-control study. Molecular autism 10: 1.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080602
Looker, C; Luster, MI; Calafat, AM; Johnson, VJ; Burleson, GR; Burleson, FG; Fletcher, T. (2014).
Influenza vaccine response in adults exposed to perfluorooctanoate and perfluorooctanesulfonate.
Toxicol Sci 138: 76-88.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850913
Lopez-Espinosa, MJ; Carrizosa, C; Luster, MI; Margolick, JB; Costa, O; Leonardi, GS; Fletcher, T.
(2021). Perfluoroalkyl substances and immune cell counts in adults from the Mid-Ohio Valley
(USA). Environ Int 156: 106599.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7751049
Lopez-Espinosa, MJ; Fletcher, T; Armstrong, B, en; Genser, B; Dhatariya, K; Mondal, D; Ducatman, A;
Leonardi, G. (2011). Association of Perfluorooctanoic Acid (PFOA) and Perfluorooctane
Sulfonate (PFOS) with Age of Puberty among Children Living near a Chemical Plant. Environ
Sci Technol 45: 8160-8166.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1424973
Lopez-Espinosa, MJ; Mondal, D; Armstrong, B; Bloom, MS; Fletcher, T. (2012). Thyroid function and
perfluoroalkyl acids in children living near a chemical plant. Environ Health Perspect 120: 1036-
1041. https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1291122
Lopez-Espinosa, MJ; Mondal, D; Armstrong, BG; Eskenazi, B; Fletcher, T. (2016). Perfluoroalkyl
Substances, Sex Hormones, and Insulin-like Growth Factor-1 at 6-9 Years of Age: A Cross-
7-47
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Sectional Analysis within the C8 Health Project. Environ Health Perspect 124: 1269-1275.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859832
Lorber, M; Eaglesham, GE; Hobson, P; Toms, LM; Mueller, JF; Thompson, JS. (2015). The effect of
ongoing blood loss on human serum concentrations of perfluorinated acids. Chemosphere 118:
170-177. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851157
Lorber, M; Egeghy, PP. (2011). Simple intake and pharmacokinetic modeling to characterize exposure of
Americans to perfluoroctanoic acid, PFOA. Environ Sci Technol 45: 8006-8014.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2914150
Lou, I; Wambaugh, JF; Lau, C; Hanson, RG; Lindstrom, AB; Strynar, MJ; Zehr, RD; Setzer, RW; Barton,
HA. (2009). Modeling single and repeated dose pharmacokinetics of PFOA in mice. Toxicol Sci
107: 331-341. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919359
Louis, GM; Sapra, KJ; Barr, DB; Lu, Z; Sundaram, R. (2016). Preconception perfluoroalkyl and
polyfluoroalkyl substances and incident pregnancy loss, LIFE Study. Reprod Toxicol 65: 11-17.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858527
Louis, GMB; Peterson, CM; Chen, Z; Hediger, ML; Croughan, MS; Sundaram, R; Stanford, JB;
Fujimoto, VY; Varner, MW; Giudice, LC; Kennedy, A; Sun, L; Wu, Q; Kannan, K. (2012).
Perfluorochemicals and endometriosis: The ENDO study. Epidemiology 23: 799-805.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1597490
Louisse, J; Rijkers, D; Stoopen, G; Janssen, A; Staats, M; Hoogenboom, R; Kersten, S; Peijnenburg, A.
(2020). Perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), and
perfluorononanoic acid (PFNA) increase triglyceride levels and decrease cholesterogenic gene
expression in human HepaRG liver cells. Arch Toxicol 94: 3137-3155.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833626
Loveless, SE; Hoban, D; Sykes, G; Frame, SR; Everds, NE. (2008). Evaluation of the immune system in
rats and mice administered linear ammonium perfluorooctanoate. Toxicol Sci 105: 86-96.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/988599
Lu, H; Zhang, H; Gao, J; Li, Z; Bao, S; Chen, X; Wang, Y; Ge, R; Ye, L. (2019). Effects of
perfluorooctanoic acid on stem Leydig cell functions in the rat. Environ Pollut 250: 206-215.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 3 81625
Lu, Y; Luo, B; Li, J; Dai, J. (2016). Perfluorooctanoic acid disrupts the blood-testis barrier and activates
the TNFa/p38 MAPK signaling pathway in vivo and in vitro. Arch Toxicol 90: 971-983.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850390
Lu, Y; Pan, Y; Sheng, N; Zhao, AZ; Dai, J. (2016). Perfluorooctanoic acid exposure alters
polyunsaturated fatty acid composition, induces oxidative stress and activates the AKT/AMPK
pathway in mouse epididymis. Chemosphere 158: 143-153.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981459
Luebker, DJ; Hansen, KJ; Bass, NM; Butenhoff, JL; Seacat, AM. (2002). Interactions of fluorochemicals
with rat liver fatty acid-binding protein. Toxicology 176: 175-185.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291067
Lum, KJ; Sundaram, R; Barr, DB; Louis, TA; Buck Louis, GM. (2017). Perfluoroalkyl Chemicals,
Menstrual Cycle Length, and Fecundity: Findings from a Prospective Pregnancy Study.
Epidemiology 28: 90-98.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3 858516
Lundin, JI; Alexander, BH; Olsen, GW; Church, TR. (2009). Ammonium Perfluorooctanoate Production
and Occupational Mortality. Epidemiology 20: 921-928.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291108
Luo, J; Ramlau-Hansen, CH; Kesmodel, US; Xiao, J; Vasiliou, V; Deziel, NC; Zhang, Y; Olsen, J; Liew,
Z. (2022). Prenatal Exposure to Per- and Polyfluoroalkyl Substances and Facial Features at 5
Years of Age: A Study from the Danish National Birth Cohort. Environ Health Perspect 130:
17006. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10273290
Luster, MI; Johnson, VJ; Yucesoy, B; Simeonova, PP. (2005). Biomarkers to assess potential
7-48
-------
DRAFT FOR PUBLIC COMMENT
March 2023
developmental immunotoxicity in children. Toxicol Appl Pharmacol 206: 229-236.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2174509
Lv, D; Gu, Y; Guo, M; Hou, P; Li, Y; Wu, R. (2019). Perfluorooctanoic acid exposure induces apoptosis
in SMMC-7721 hepatocellular cancer cells. Environ Pollut247: 509-514.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080368
Lyall, K; Yau, VM; Hansen, R; Kharrazi, M; Yoshida, CK; Calafat, AM; Windham, G; Croen, LA.
(2018). Prenatal maternal serum concentrations of per- and polyfluoroalkyl substances in
association with autism spectrum disorder and intellectual disability. Environ Health Perspect
126: 017001. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239287
Lyngso, J; Ramlau-Hansen, CH; Hover. BB; Stovring. H; Bonde, JP; Jonsson, BA; Lindh, CH; Pedersen,
HS; Ludwicki, JK; Zviezdai, V; Toft, G. (2014). Menstrual cycle characteristics in fertile women
from Greenland, Poland and Ukraine exposed to perfluorinated chemicals: a cross-sectional
study. Hum Reprod 29: 359-367.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850920
Ma, S; Xu, C; Ma, J; Wang, Z; Zhang, Y; Shu, Y; Mo, X. (2019). Association between perfluoroalkyl
substance concentrations and blood pressure in adolescents. Environ Pollut 254: 112971.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5413104
Ma, Z; Liu, X; Li, F; Wang, Y; Xu, Y; Zhang, M; Zhang, X; Ying, X; Zhang, X. (2016).
Perfluorooctanoic acid induces human Ishikawa endometrial cancer cell migration and invasion
through activation of ERK/mTOR signaling. Onct 7: 66558-66568.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3 9 81426
Macgillivray, DM; Kollmann, TR. (2014). The role of environmental factors in modulating immune
responses in early life [Review]. Front Immunol 5: 434.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6749084
Macmanus-Spencer, LA; Tse, ML; Hebert, PC; Bischel, HN; Luthy, RG. (2010). Binding of
perfluorocarboxylates to serum albumin: a comparison of analytical methods. Anal Chem 82:
974-981. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850334
Macneil, J; Steenland, NK; Shankar, A; Ducatman, A. (2009). A cross-sectional analysis of type II
diabetes in a community with exposure to perfluorooctanoic acid (PFOA). Environ Res 109: 997-
1003. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919319
Macon, MB; Fenton, SE. (2013). Endocrine disruptors and the breast: Early life effects and later life
disease [Review]. J Mammary Gland Biol Neoplasia 18: 43-61.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3827893
Macon, MB; Villanueva, LR; Tatum-Gibbs, K; Zehr, RD; Strynar, MJ; Stanko, JP; White, SS; Helfant, L;
Fenton, SE. (2011). Prenatal perfluorooctanoic acid exposure in CD-I mice: low-dose
developmental effects and internal dosimetry. Toxicol Sci 122: 134-145.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1276151
Maekawa, R; Ito, R; Iwasaki, Y; Saito, K; Akutsu, K; Takatori, S; Ishii, R; Kondo, F; Arai, Y; Ohgane, J;
Shiota, K; Makino, T; Sugino, N. (2017). Evidence of exposure to chemicals and heavy metals
during pregnancy in Japanese women. Reproductive Medicine and Biology 16: 337-348.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/423 8291
Magnuson, K; Lemeris, C; McGill, R; Sibrizzi, C; Rooney, ASK; Taylor, K; Walker, V. (2022). Using
Interactive Literature Flow Diagrams to Increase Transparency in Systematic Reviews
(unpublished work), https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10442900
Maher, JM; Aleksunes, LM; Dieter, MZ; Tanaka, Y; Peters, JM; Manautou, JE; Klaassen, CD. (2008).
Nrf2- and PPAR alpha-mediated regulation of hepatic Mrp transporters after exposure to
perfluorooctanoic acid and perfluorodecanoic acid. Toxicol Sci 106: 319-328.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919367
Maisonet, M; Calafat, AM; Marcus, M; Jaakkola, JJ; Lashen, H. (2015). Prenatal exposure to
perfluoroalkyl acids and serum testosterone concentrations at 15 years of age in female ALSPAC
study participants. Environ Health Perspect 123: 1325-1330.
7-49
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859841
Makey, CM; Webster, TF; Martin, JW; Shoeib, M; Harner, T; Dix-Cooper, L; Webster, GM. (2017).
Airborne precursors predict maternal serum perfluoroalkyl acid concentrations. Environ Sci
Technol 51: 7667-7675.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860102
Malhotra, P; Gill, RK; Saksena, S; Alrefai, WA. (2020). Disturbances in Cholesterol Homeostasis and
Non-alcoholic Fatty Liver Diseases [Review]. Front Med 7: 467.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10442471
Mamsen, LS; Bjorvang, RD; Mucs, D; Vinnars, MT; Papadogiannakis, N; Lindh, CH; Andersen, CY;
Damdimopoulou, P. (2019). Concentrations of perfluoroalkyl substances (PFASs) in human
embryonic and fetal organs from first, second, and third trimester pregnancies. Environ Int 124:
482-492. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080595
Mamsen, LS; Jonsson, BAG; Lindh, CH; Olesen, RH; Larsen, A; Ernst, E; Kelsey, TW; Andersen, CY.
(2017). Concentration of perfluorinated compounds and cotinine in human foetal organs,
placenta, and maternal plasma. Sci Total Environ 596-597: 97-105.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3858487
Mancini, FR; Cano-Sancho, G; Gambaretti, J; Marchand, P; Boutron-Ruault, MC; Severi, G; Arveux, P;
Antignac, JP; Kvaskoff, M. (2020). Perfluorinated alkylated substances serum concentration and
breast cancer risk: Evidence from a nested case-control study in the French E3N cohort. Int J
Cancer 146: 917-928. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5381529
Mancini, FR; Rajaobelina, K; Praud, D; Dow, C; Antignac, JP; Kvaskoff, M; Severi, G; Bonnet, F;
Boutron-Ruault, MC; Fagherazzi, G. (2018). Nonlinear associations between dietary exposures to
perfluorooctanoic acid (PFOA) or perfluorooctane sulfonate (PFOS) and type 2 diabetes risk in
women: Findings from the E3N cohort study. Int J Hyg Environ Health 221: 1054-1060.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5079710
Mann, PC; Frame, SR. (2004). FC-143: Two year oral toxicity-oncogenicity study in rats. Peer review of
ovaries. (Project ID 15261). Newark, DE: E.I. du Pont de Nemours and Company.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/6569580
Manzano-Salgado, CB; Casas, M; Lopez-Espinosa, MJ; Ballester, F; Basterrechea, M; Grimalt, JO;
Jimenez, AM; Kraus, T; Schettgen, T; Sunyer, J; Vrijheid, M. (2015). Transfer of perfluoroalkyl
substances from motherto fetus in a Spanish birth cohort. Environ Res 142: 471-478.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3448674
Manzano-Salgado, CB; Casas, M; Lopez-Espinosa, MJ; Ballester, F; Iniguez, C; Martinez, D;
Romaguera, D; Fernandez-Barres, S; Santa-Marina, L; Basterretxea, M; Schettgen, T; Valvi, D;
Vioque, J; Sunyer, J; Vrijheid, M. (2017). Prenatal exposure to perfluoroalkyl substances and
cardiometabolic risk in children from the Spanish INMA birth cohort study. Environ Health
Perspect 125: 097018.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/4238509
Manzano-Salgado, CB; Granum, B; Lopez-Espinosa, MJ; Ballester, F; Iniguez, C; Gascon, M; Martinez,
D; Guxens, M; Basterretxea, M; Zabaleta, C; Schettgen, T; Sunyer, J; Vrijheid, M; Casas, M.
(2019). Prenatal exposure to perfluoroalkyl substances, immune-related outcomes, and lung
function in children from a Spanish birth cohort study. Int J Hyg Environ Health 222: 945-954.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5412076
Marks, KJ; Jeddy, Z; Flanders, WD; Northstone, K; Fraser, A; Calafat, AM; Kato, K; Hartman, TJ.
(2019). Maternal serum concentrations of perfluoroalkyl substances during pregnancy and
gestational weight gain: The Avon Longitudinal Study of Parents and Children. Reprod Toxicol
90: 8-14. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5381534
Martin, JA; Hamilton, BE; Osterman, MJ; Driscoll, AK; Mathews, TJ. (2018). Births: Final data for 2017.
Natl Vital Stat Rep 67.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/8632225
Martin, MT; Brennan, RJ; Hu, W; Ayanoglu, E; Lau, C; Ren, H; Wood, CR; Corton, JC; Kavlock, RJ;
7-50
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Dix, DJ. (2007). Toxicogenomic study of triazole fungicides and perfluoroalkyl acids in rat livers
predicts toxicity and categorizes chemicals based on mechanisms of toxicity. Toxicol Sci 97: 595-
613. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/758419
Martinsson, M; Nielsen, C; Bjork, J; Rylander, L; Malmqvist, E; Lindh, C; Rignell-Hydbom, A. (2020).
Intrauterine exposure to perfluorinated compounds and overweight at age 4: A case-control study.
PLoS ONE 15: e0230137.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311645
Mashayekhi, V; Tehrani, KH; Hashemzaei, M; Tabrizian, K; Shahraki, J; Hosseini, MJ. (2015).
Mechanistic approach for the toxic effects of perfluorooctanoic acid on isolated rat liver and brain
mitochondria. Hum Exp Toxicol 34: 985-996.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851019
MassDEP. (2019). Per- and polyfluoroalkyl substances (pfas):an updated subgroup approach to
groundwater and drinking water values. Boston, MA.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6983120
Mathiesen, UL; Franzen, LE; Fryden, A; Foberg, U; Bodemar, G. (1999). The clinical significance of
slightly to moderately increased liver transaminase values in asymptomatic patients. Scand J
Gastroenterol 34: 85-91.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10293242
Matilla-Santander, N; Valvi, D; Lopez-Espinosa, MJ; Manzano-Salgado, CB; Ballester, F; Ibarluzea, J;
Santa-Marina, L; Schettgen, T; Guxens, M; Sunyer, J; Vrijheid, M. (2017). Exposure to
Perfluoroalkyl Substances and Metabolic Outcomes in Pregnant Women: Evidence from the
Spanish INMA Birth Cohorts. Environ Health Perspect 125: 117004.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238432
Matkowskyj, KA; Bai, H; Liao, J; Zhang, W; Li, H; Rao, S; Omary, R; Yang, GY. (2014).
Aldoketoreductase family 1B10 (AKR1B10) as abiomarkerto distinguish hepatocellular
carcinoma from benign liver lesions. Hum Pathol 45: 834-843.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/10365736
Mattsson, K; Rignell-Hydbom, A; Holmberg, S; Thelin, A; Jonsson, BA; Lindh, CH; Sehlstedt, A;
Rylander, L. (2015). Levels of perfluoroalkyl substances and risk of coronary heart disease:
Findings from a population-based longitudinal study. Environ Res 142: 148-154.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859607
McComb, J; Mills, IG; Berntsen, HF; Ropstad, E; Verhaegen, S; Connolly, L. (2019). Human-based
exposure levels of perfluoroalkyl acids may induce harmful effects to health by disrupting major
components of androgen receptor signalling in vitro. Exposure and Health 12: 527-538.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6304412
Mccoy, JA; Bangma, JT; Reiner, JL; Bowden, JA; Schnorr, J; Slowey, M; O'Leary, T; Guillette, LJ;
Parrott, BB. (2017). Associations between perfluorinated alkyl acids in blood and ovarian
follicular fluid and ovarian function in women undergoing assisted reproductive treatment. Sci
Total Environ 605-606: 9-17.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858475
Mcdonough, CA; Li, W; Bischel, HN; De Silva, AO; Dewitt, JC. (2022). Widening the Lens on PFASs:
Direct Human Exposure to Perfluoroalkyl Acid Precursors (pre-PFAAs). Environ Sci Technol 56:
6004-6013. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10412593
McQuillan, GM; Kruszon-Moran, D; Deforest, A; Chu, SY; Wharton, M. (2002). Serologic immunity to
diphtheria and tetanus in the United States. 136: 660-666.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/9642142
MDH. (2020). Toxicological Summary for: Perfluorooctanoate.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9418094
Melnikov, F; Botta, D; White, CC; Schmuck, SC; Winfough, M; Schaupp, CM; Gallagher, E; Brooks,
BW; Williams, ES; Coish, P; Anastas, PT; Voutchkova, A; Kostal, J; Kavanagh, TJ. (2018).
Kinetics of Glutathione Depletion and Antioxidant Gene Expression as Indicators of Chemical
7-51
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Modes of Action Assessed in vitro in Mouse Hepatocytes with Enhanced Glutathione Synthesis.
Chem Res Toxicol 32: 421-436.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5031105
Mi, X; Yang, YQ; Zeeshan, M; Wang, ZB; Zeng, XY; Zhou, Y; Yang, BY; Hu, LW; Yu, HY; Zeng, XW;
Liu, RQ; Dong, GH. (2020). Serum levels of per- and polyfluoroalkyl substances alternatives and
blood pressure by sex status: Isomers of C8 health project in China. Chemosphere 261: 127691.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833736
Miao, C; Ma, J; Zhang, Y; Chu, Y; Li, J; Kuai, R; Wang, S; Peng, H. (2015). Perfluorooctanoic acid
enhances colorectal cancer DLD-1 cells invasiveness through activating NF-kB mediated matrix
metalloproteinase-2/-9 expression. Int J Clin Exp Pathol 8: 10512-10522.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981523
Midasch, O; Drexler, H; Hart, N; Beckmann, MW; Angerer, J. (2007). Transplacental exposure of
neonates to perfluorooctanesulfonate and perfluorooctanoate: a pilot study. Int Arch Occup
Environ Health 80: 643-648.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290901
Midic, U; Goheen, B; Vincent, KA; Vandevoort, CA; Latham, KE. (2018). Changes in gene expression
following long-term in vitro exposure of macaca mulatta trophoblast stem cells to biologically
relevant levels of endocrine disruptors. Reprod Toxicol 77: 154-165.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4241048
Mills, KH; Dungan, LS; Jones, SA; Harris, J. (2013). The role of inflammasome-derived IL-1 in driving
IL-17 responses [Review]. J Leukoc Biol 93: 489-497.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2556647
Min, JY; Lee, KJ; Park, JB; Min, KB. (2012). Perfluorooctanoic acid exposure is associated with elevated
homocysteine and hypertension in US adults. Occup Environ Med 69: 658-662.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919181
Minata, M; Harada, KH; Karrman, A; Hitomi, T; Hirosawa, M; Murata, M; Gonzalez, FJ; Koizumi, A.
(2010). Role of peroxisome proliferator-activated receptor-alpha in hepatobiliary injury induced
by ammonium perfluorooctanoate in mouse liver. Ind Health 48: 96-107.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937251
Minatoya, M; Itoh, S; Miyashita, C; Araki, A; Sasaki, S; Miura, R; Goudarzi, H; Iwasaki, Y; Kishi, R.
(2017). Association of prenatal exposure to perfluoroalkyl substances with cord blood adipokines
and birth size: The Hokkaido Study on environment and children's health. Environ Res 156: 175-
182. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981691
Mitro, SD; Sagiv, SK; Fleisch, AF; Jaacks, LM; Williams, PL; Rifas-Shiman, SL; Calafat, AM; Hivert,
MF; Oken, E; James-Todd, TM. (2020). Pregnancy per- and polyfluoroalkyl substance
concentrations and postpartum health in project viva: A prospective cohort. J Clin Endocrinol
Metab 105: e3415-e3426.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833625
Miura, R; Araki, A; Miyashita, C; Kobayashi, S; Kobayashi, S; Wang, SL; Chen, CH; Miyake, K;
Ishizuka, M; Iwasaki, Y; Ito, YM; Kubota, T; Kishi, R. (2018). An epigenome-wide study of cord
blood DNA methylations in relation to prenatal perfluoroalkyl substance exposure: The Hokkaido
study. Environ Int 115: 21-28.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080353
Mobacke, I; Lind, L; Dunder, L; Salihovic, S; Lind, PM. (2018). Circulating levels of perfluoroalkyl
substances and left ventricular geometry of the heart in the elderly. Environ Int 115: 295-300.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4354163
Mogensen, UB; Grandjean, P; Heilmann, C; Nielsen, F; Weihe, P; Budtz-Jorgensen. E. (2015). Structural
equation modeling of immunotoxicity associated with exposure to perfluorinated alkylates.
Environ Health 14: 47.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981889
Mogensen, UB; Grandjean, P; Nielsen, F; Weihe, P; Budtz-Jorgensen, E. (2015). Breastfeeding as an
7-52
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Exposure Pathway for Perfluorinated Alkylates. Environ Sci Technol 49: 10466-10473.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859839
Mondal, D; Weldon, RH; Armstrong, BG; Gibson, LJ; Lopez-Espinosa, MJ; Shin, HM; Fletcher, T.
(2014). Breastfeeding: a potential excretion route for mothers and implications for infant
exposure to perfluoroalkyl acids. Environ Health Perspect 122: 187-192.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850916
Monge Brenes, AL; Curtzwiler, G; Dixon, P; Harrata, K; Talbert, J; Vorst, K. (2019). PFOA and PFOS
levels in microwave paper packaging between 2005 and 2018. Food Addit Contam Part B
Surveill 12: 191-198. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080553
Monroy, R; Morrison, K; Teo, K; Atkinson, S; Kubwabo, C; Stewart, B; Foster, WG. (2008). Serum
levels of perfluoroalkyl compounds in human maternal and umbilical cord blood samples.
Environ Res 108: 56-62.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2349575
Mora, AM; Fleisch, AF; Rifas-Shiman, SL; Woo Baidal, JA; Pardo, L; Webster, TF; Calafat, AM; Ye, X;
Oken, E; Sagiv, SK. (2018). Early life exposure to per- and polyfluoroalkyl substances and mid-
childhood lipid and alanine aminotransferase levels. Environ Int 111: 1-13.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239224
Mora, AM; Oken, E; Rifas-Shiman, SL; Webster, TF; Gillman, MW; Calafat, AM; Ye, X; Sagiv, SK.
(2017). Prenatal exposure to perfluoroalkyl substances and adiposity in early and mid-childhood.
Environ Health Perspect 125: 467-473.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859823
Mora, S. (2016). Nonfasting for Routine Lipid Testing: From Evidence to Action. JAMA Intern Med 176:
1005-1006. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9564968
Mordenti, J; Chen, SA; Moore, JA; Ferraiolo, BL; Green, JD. (1991). Interspecies scaling of clearance
and volume of distribution data for five therapeutic proteins. Pharm Res 8:1351-1359.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9571900
MPCA. (2008). PFCs in Minnesota's Ambient Environment: 2008 Progress Report.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9419086
Murli, H. (1995). Mutagenicity test on T-6342 in an in vivo mouse micronucleus assay. (EPA-AR-226-
0435). Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228120
Murli, H. (1996). Mutagenicity test on T-6342 measuring chromosomal aberrations in Chinese hamster
ovary (CHO) cells with a confirmatory assay with multiple harvests. (EPA-AR-226-0434).
Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228124
Murli, H. (1996). Mutagenicity test on T-6342 measuring chromosomal aberrations in human whole
blood lymphocytes with a confirmatory assay with multiple harvests. (EPA-AR-226-0433).
Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228126
Murli, H. (1996). Mutagenicity test on T-6564 in an in vivo mouse micronucleus assay. (CHV Study No.
17750-0-455; EPA-AR-226-0430). Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228121
Murli, H. (1996). Mutagenicity test on T-6564 measuring chromosomal aberrations in Chinese hamster
ovary (CHO) cells with a confirmatory assay with multiple harvests. (CHV Study No. 17750-0-
437CO; EPA-AR-226-0431). Vienna, VA: Corning Hazleton.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228125
Myers, LP. (2018). Clinical immunotoxicology. In JC DeWitt; CE Rockwell; CC Bowman (Eds.),
Immunotoxicity testing: Methods and protocols (2nd ed., pp. 15-26). Totowa, NJ: Humana Press.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10473136
Mylchreest, E. (2003). PFOA: Lactational and Placental Transport Pharmacokinetic Study in Rats.
(DuPont-13309). Newark, DE: Haskell Laboratory for Health and Environmental Sciences.
7-53
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642031
Nakagawa, H; Hirata, T; Terada, T; Jutabha, P; Miura, D; Harada, KH; Inoue, K; Anzai, N; Endou, H;
Inui, K; Kanai, Y; Koizumi, A. (2008). Roles of organic anion transporters in the renal excretion
of perfluorooctanoic acid. Basic & Clinical Pharmacology & Toxicology Online Pharmacology
Online 103: 1-8. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919370
Nakagawa, H; Terada, T; Harada, KH; Hitomi, T; Inoue, K; Inui, K; Koizumi, A. (2009). Human organic
anion transporter hOAT4 is a transporter of perfluorooctanoic acid. Basic & Clinical
Pharmacology & Toxicology Online Pharmacology Online 105: 136-138.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919342
Nakayama, S; Strynar, MJ; Helfant, L; Egeghy, P; Ye, X; Lindstrom, AB. (2007). Perfluorinated
compounds in the Cape Fear Drainage Basin in North Carolina. Environ Sci Technol 41: 5271-
5276. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2901973
NASEM. (2021). Review of U.S. EPA's ORD staff handbook for developing IRIS assessments: 2020
version. Washington, DC: National Academies Press.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959764
Natarajan, AT; Darroudi, F. (1991). Use of human hepatoma cells for in vitro metabolic activation of
chemical mutagens/carcinogens. Mutagenesis 6: 399-404.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/5143588
NCBI. (2022). PubChem: Compound summary: Perfluorooctanoic acid. Available online at
https://pubchem.ncbi.nlm.nih.gov/compound/Perfluorooctanoic-acid (accessed April 21,
2022) .https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10411459
NCHS. (2019). Health, United States - Data Finder. 2019: Table 23. Hyattsville, MD.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369680
Needham, LL; Grandjean, P; Heinzow, B; Jorgensen. PJ; Nielsen, F; Patterson, DG; Sjodin, A; Turner,
WE; Weihe, P. (2011). Partition of environmental chemicals between maternal and fetal blood
and tissues. Environ Sci Technol 45: 1121-1126.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1312781
Negri, E; Metruccio, F; Guercio, V; Tosti, L; Benfenati, E; Bonzi, R; La Vecchia, C; Moretto, A. (2017).
Exposure to PFOA and PFOS and fetal growth: a critical merging of toxicological and
epidemiological data [Review]. Crit Rev Toxicol 47: 482-508.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981320
Nelson, JW; Hatch, EE; Webster, TF. (2010). Exposure to Polyfluoroalkyl Chemicals and Cholesterol,
Body Weight, and Insulin Resistance in the General US Population. Environ Health Perspect 118:
197-202. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291110
Nelson, JW; Scammell, MK; Hatch, EE; Webster, TF. (2012). Social disparities in exposures to bisphenol
A and polyfluoroalkyl chemicals: a cross-sectional study within NHANES 2003-2006. Environ
Health 11: 10. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4904674
Neumann, J; Rose-Sperling, D; Hellmich, UA. (2017). Diverse relations between ABC transporters and
lipids: An overview [Review]. Biochim Biophys ActaBiomembr 1859: 605-618.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365731
New Hampshire DES. (2019). Technical background report for the june 2019 proposed maximum
contaminant levels (MCLs) and ambient groundwater quality standards (AGQSs) for
perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), perfluorononanoic acid
(PFNA), and perfluorohexane sulfonic acid (PFHXs).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5949029
Ngueta, G; Longnecker, MP; Yoon, M; Ruark, CD; Clewell, HJ; Andersen, ME; Verner, MA. (2017).
Quantitative bias analysis of a reported association between perfluoroalkyl substances (PFAS)
and endometriosis: The influence of oral contraceptive use. Environ Int 104: 118-121.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860773
Nguyen, TMH; Braunig, J; Thompson, K; Thompson, J; Kabiri, S; Navarro, DA; Kookana, RS;
Grimison, C; Barnes, CM; Higgins, CP; Mclaughlin, MJ; Mueller, JF. (2020). Influences of
7-54
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Chemical Properties, Soil Properties, and Solution pH on Soil-Water Partitioning Coefficients of
Per- and Polyfluoroalkyl Substances (PFASs). Environ Sci Technol 54: 15883-15892.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7014622
Nian, M; Li, QQ; Bloom, M; Qian, ZM; Syberg, KM; Vaughn, MG; Wang, SQ; Wei, Q; Zeeshan, M;
Gurram, N; Chu, C; Wang, J; Tian, YP; Hu, LW; Liu, KK; Yang, BY; Liu, RQ; Feng, D; Zeng,
XW; Dong, GH. (2019). Liver function biomarkers disorder is associated with exposure to
perfluoroalkyl acids in adults: Isomers of C8 Health Project in China. Environ Res 172: 81-88.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080307
Nielsen, C; Andersson Hall, U; Lindh, C; Ekstrom, U; Xu, Y; Li, Y; Holmang, A; Jakobsson, K. (2020).
Pregnancy-induced changes in serum concentrations of perfluoroalkyl substances and the
influence of kidney function. Environ Health 19: 80.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833687
Niu, J; Liang, H; Tian, Y; Yuan, W; Xiao, H; Hu, H; Sun, X; Song, X; Wen, S; Yang, L; Ren, Y; Miao,
M. (2019). Prenatal plasma concentrations of Perfluoroalkyl and polyfluoroalkyl substances and
neuropsychological development in children at four years of age. Environ Health 18: 53.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381527
NLM. (2022). ChemlDplus: Perfluorooctanoic acid. Available online at
https://chem.nlm.nih.gov/chemidplus/name/pfoa 10369702
NLM. (2022). PubChem Hazardous Substances Data Bank (HSDB) Annotation Record for
Perfluorooctanoic acid. Washington, DC: National Institutes of Health, Department of Health and
Human Services. Retrieved from
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369700
Noorlander, CW; van Leeuwen, SP; Te Biesebeek, JD; Mengelers, MJ; Zeilmaker, MJ. (2011). Levels of
perfluorinated compounds in food and dietary intake of PFOS and PFOA in the Netherlands. J
Agric Food Chem 59: 7496-7505.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919242
NOTOX. (2000). Evaluation of the ability of T-7524 to induce chromosome aberrations in cultured
peripheral human lymphocytes. (NOTOX Project Number 292062). Hertogenbosch, The
Netherlands: NOTOX.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10270878
NRC. (2011). Review of the Environmental Protection Agency's draft IRIS assessment of formaldehyde
(pp. 1-194). Washington, DC: The National Academies Press.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/710724
NTP. (2016). NTP Monograph: Immunotoxicity associated with exposure to perfluorooctanoic acid
(PFOA) or perfluorooctane sulfonate (PFOS). Research Triangle Park, NC: U.S. Department of
Health and Human Services, Office of Health Assessment and Translation.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4613766
NTP. (2019). NTP technical report on the toxicity studies of perfluoroalkyl carboxylates
(perfluorohexanoic acid, perfluorooctanoic acid, perfluorononanoic acid, and perfluorodecanoic
acid) administered by gavage to Sprague Dawley (Hsd: Sprague Dawley SD) rats [NTP].
(Toxicity Report 97). Research Triangle Park, NC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5400977
NTP. (2019). NTP technical report on the toxicity studies of perfluoroalkyl sulfonates (perfluorobutane
sulfonic acid, perfluorohexane sulfonate potassium salt, and perfluorooctane sulfonic acid)
administered by gavage to Sprague Dawley (Hsd: Sprague Dawley SD) rats. (Toxicity Report 96).
Research Triangle Park, NC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5400978
NTP. (2020). NTP technical report on the toxicology and carcinogenesis studies of perfluorooctanoic acid
(CASRN 335-67-1) administered in feed to Sprague Dawley (Hsd:Sprague Dawley SD) rats
[NTP], (Technical Report 598). Research Triangle Park, NC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7330145
7-55
-------
DRAFT FOR PUBLIC COMMENT
March 2023
NYSDOH. (2018). NYSDOH Drinking Water Quality Council (DWQC) - October 17, 2018. Available
online at https://www.youtube.com/watch?v=2JIXCla6cHM&feature=youtu.be 6984171
O'Malley, KD; Ebbins, KL. (1981). Repeat application 28 day percutaneous absorption study with T-
2618CoC in albino rabbits. (USEPA Administrative Record 226-0446). St. Paul, MN: Riker
Laboratories, Inc. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4471529
Obourn, JD; Frame, SR; Bell, RH; Longnecker, DS; Elliott, GS; Cook, JC. (1997). Mechanisms for the
pancreatic oncogenic effects of the peroxisome proliferator Wyeth-14,643. Toxicol Appl
Pharmacol 145: 425-436.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3748746
Ode, A; Kallen, K; Gustafsson, P; Rylander, L; Jonsson, BA; Olofsson, P; Ivarsson, SA; Lindh, CH;
Rignell-Hydbom, A. (2014). Fetal exposure to perfluorinated compounds and attention deficit
hyperactivity disorder in childhood. PLoS ONE 9: e95891.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851245
OECD. (2001). Test no. 416: Two-generation reproduction toxicity. In OECD guidelines for the testing of
chemicals, Section 4: Health effects. Paris, France.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3421602
OECD. (2018). Toward a new comprehensive global database of per- and polyfluoroalkyl substances
(PFASs): Summary report on updating the OECD 2007 list of per- and polyfluoroalkyl
substances (PFASs). (ENV/JM/MONO(2018)7).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5099062
OEHHA. (2004). Public Health Goal for Arsenic in Drinking Water: Arsenic. Sacramento, CA: Office of
Environmental Health Hazard Assessment, California Environmental Protection Agency.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369748
Ojo, AF; Peng, C; Ng, JC. (2020). Combined effects and toxicological interactions of perfluoroalkyl and
polyfluoroalkyl substances mixtures in human liver cells (HepG2). Environ Pollut 263: 114182.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6333436
Okada, E; Sasaki, S; Saijo, Y; Washino, N; Miyashita, C; Kobayashi, S; Konishi, K; Ito, YM; Ito, R;
Nakata, A; Iwasaki, Y; Saito, K; Nakazawa, H; Kishi, R. (2012). Prenatal exposure to
perfluorinated chemicals and relationship with allergies and infectious diseases in infants.
Environ Res 112: 118-125.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332477
Olsen, G, .W., B.,urlew, M.,.M.; Burris, J, .M.; Mandel, J, .H. (2001). A Longitudinal Analysis of Serum
Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoate (PFOA) Levels in Relation to Lipid
and Hepatic Clinical Chemistry Test Results from Male Employee Participants of the 1994/95,
1997 and 2000 Fluorochemical Medical Surveillance Program. Final Report. (Epidemiology,
220-3W-05). St. Paul, MN: 3M Company.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10228462
Olsen, G; Ehresman, D; Froehlich, J; Burris, J; Butenhoff, J. (2005). Evaluation of the Half-life (Tl/2) of
Elimination of Perfluorooctanesulfonate (PFOS), Perfluorohexanesulfonate (PFHS) and
Perfluorooctanoate (PFOA) from Human Serum. St. Paul, MN: 3M Company.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642064
Olsen, GW; Burris, JM; Burlew, MM; Mandel, JH. (1998). 3M final report: an epidemiologic
investigation of plasma cholecystokinin, hepatic function and serum perfluorooctanoic acid levels
in production workers. (U.S. Environmental Protection Agency Administrative Record 226-
0476). 3M Company, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9493903
Olsen, GW; Burris, JM; Burlew, MM; Mandel, JH. (2000). Plasma cholecystokinin and hepatic enzymes,
cholesterol and lipoproteins in ammonium perfluorooctanoate production workers. Drug Chem
Toxicol 23: 603-620. https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/1424954
Olsen, GW; Burris, JM; Burlew, MM; Mandel, JH. (2003). Epidemiologic assessment of worker serum
perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) concentrations and medical
surveillance examinations. J Occup Environ Med 45: 260-270.
7-56
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290020
Olsen, GW; Ehresman, DJ; Buehrer, BD; Gibson, BA; Butenhoff, JL; Zobel, LR. (2012). Longitudinal
assessment of lipid and hepatic clinical parameters in workers involved with the demolition of
perfluoroalkyl manufacturing facilities. J Occup Environ Med 54: 974-983.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919185
Olsen, GW; Gilliland, FD; Burlew, MM; Burris, JM; Mandel, JS; Mandel, JH. (1998). An epidemiologic
investigation of reproductive hormones in men with occupational exposure to perfluorooctanoic
acid. J Occup Environ Med 40: 614-622.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290857
Olsen, GW; Hansen, KJ; Clemen, LA; Burris, JM; Mandel, JH. (2001). Identification of Fluorochemicals
in Human Tissue. (U.S. Environmental Protection Agency Administrative Record 226-
1030a022). 3M. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641811
Olsen, GW; Zobel, LR. (2007). Assessment of lipid, hepatic, and thyroid parameters with serum
perfluorooctanoate (PFOA) concentrations in fluorochemical production workers. Int Arch Occup
Environ Health 81: 231-246.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290836
Omoike, OE; Pack, RP; Mamudu, HM; Liu, Y; Strasser, S; Zheng, S; Okoro, J; Wang, L. (2020).
Association between per and polyfluoroalkyl substances and markers of inflammation and
oxidative stress. Environ Res 196: 110361.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988477
Omoike, OE; Pack, RP; Mamudu, HM; Liu, Y; Wang, L. (2021). A cross-sectional study of the
association between perfluorinated chemical exposure and cancers related to deregulation of
estrogen receptors. Environ Res 196: 110329.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7021502
Onishchenko, N; Fischer, C; Wan Ibrahim, WN; Negri, S; Spulber, S; Cottica, D; Ceccatelli, S. (2011).
Prenatal exposure to PFOS or PFOA alters motor function in mice in a sex-related manner.
NeurotoxRes 19: 452-461.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/758427
Oppi, S; Liischer, TF; Stein, S. (2019). Mouse models for atherosclerosis research- Which is my line?
[Review]. Front Cardiovasc Med 6: 46.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5926372
Orbach, SM; Ehrich, MF; Rajagopalan, P. (2018). High-throughput toxicity testing of chemicals and
mixtures in organotypic multi-cellular cultures of primary human hepatic cells. Toxicol In Vitro
51: 83-94. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079788
Oshida, K; Vasani, N; Jones, C; Moore, T; Hester, S; Nesnow, S; Auerbach, S; Geter, DR; Aleksunes,
LM; Thomas, RS; Applegate, D; Klaassen, CD; Corton, JC. (2015). Identification of chemical
modulators of the constitutive activated receptor (CAR) in a gene expression compendium.
Nuclear Receptor Signaling 13: e002.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850125
Oshida, K; Vasani, N; Thomas, RS; Applegate, D; Rosen, M; Abbott, B; Lau, C; Guo, G; Aleksunes, LM;
Klaassen, C; Corton, JC. (2015). Identification of modulators of the nuclear receptor peroxisome
proliferator-activated receptor a (PPARa) in a mouse liver gene expression compendium. PLoS
ONE 10: eOl 12655. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5386121
Oshida, K; Waxman, DJ; Corton, JC. (2016). Chemical and hormonal effects on STAT5b-dependent
sexual dimorphism of the liver transcriptome. PLoS ONE 11: eO 150284.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6781228
Osorio-Yanez, C; Sanchez-Guerra, M; Cardenas, A; Lin, PID; Hauser, R; Gold, DR; Kleinman, KP;
Hivert, MF; Fleisch, AF; Calafat, AM; Webster, TF; Horton, ES; Oken, E. (2021). Per- and
polyfluoroalkyl substances and calcifications of the coronary and aortic arteries in adults with
prediabetes: Results from the diabetes prevention program outcomes study. Environ Int 151:
106446. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7542684
7-57
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Ouidir, M; Mendola, P; Louis, GMB; Kannan, K; Zhang, C; Tekola-Ayele, F. (2020). Concentrations of
persistent organic pollutants in maternal plasma and epigenome-wide placental DNA
methylation. Clinical Epigenetics 12: 103.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/683 3 759
Oulhote, Y; Coull, B; Bind, MA; Debes, F; Nielsen, F; Tamayo, I; Weihe, P; Grandjean, P. (2019). Joint
and independent neurotoxic effects of early life exposures to a chemical mixture: A multi-
pollutant approach combining ensemble learning and g-computation. Environmental
Epidemiology 3: e063.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316905
Oulhote, Y; Steuerwald, U; Debes, F; Weihe, P; Grandjean, P. (2016). Behavioral difficulties in 7-year
old children in relation to developmental exposure to perfluorinated alkyl substances [Review].
Environ Int 97: 237-245.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3789517
Pan, Y; Cui, Q; Wang, J; Sheng, N; Jing, J; Yao, B; Dai, J. (2019). Profiles of Emerging and Legacy Per-
/Polyfluoroalkyl Substances in Matched Serum and Semen Samples: New Implications for
Human Semen Quality. Environ Health Perspect 127: 127005.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/6315783
Pan, Y; Zhu, Y; Zheng, T; Cui, Q; Buka, SL; Zhang, B; Guo, Y; Xia, W; Yeung, LW; Li, Y; Zhou, A;
Qiu, L; Liu, H; Jiang, M; Wu, C; Xu, S; Dai, J. (2017). Novel Chlorinated Polyfluorinated Ether
Sulfonates and Legacy Per-/Polyfluoroalkyl Substances: Placental Transfer and Relationship with
Serum Albumin and Glomerular Filtration Rate. Environ Sci Technol 51: 634-644.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3981900
Panaretakis, T; Shabalina, IG; Grander, D; Shoshan, MC; Depierre, JW. (2001). Reactive oxygen species
and mitochondria mediate the induction of apoptosis in human hepatoma HepG2 cells by the
rodent peroxisome proliferator and hepatocarcinogen, perfluorooctanoic acid. Toxicol Appl
Pharmacol 173: 56-64.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/5081525
Papadopoulou, E; Poothong, S; Koekkoek, J; Lucattini, L; Padilla-Sanchez, JA; Haugen, M; Herzke, D;
Valdersnes, S; Maage, A; Cousins, IT; Leonards, PEG; Smastuen Haug, L. (2017). Estimating
human exposure to perfluoroalkyl acids via solid food and drinks: Implementation and
comparison of different dietary assessment methods. Environ Res 158: 269-276.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/3859798
Papadopoulou, E; Stratakis, N; Basagana, X; Brantsseter, AL; Casas, M; Fossati, S; Grazuleviciene, R;
Smastuen Haug, L; Heude, B; Maitre, L; Mceachan, RRC; Robinson, O; Roumeliotaki, T;
Sabido, E; Borras, E; Urquiza, J; Vafeiadi, M; Zhao, Y; Slama, R; Wright, J; Conti, DV; Vrijheid,
M; Chatzi, L. (2021). Prenatal and postnatal exposure to PFAS and cardiometabolic factors and
inflammation status in children from six European cohorts. Environ Int 157: 106853.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/9960593
Park, JH; Choi, J; Jun, DW; Han, SW; Yeo, YH; Nguyen, MH. (2019). Low Alanine Aminotransferase
Cut-Off for Predicting Liver Outcomes; A Nationwide Population-Based Longitudinal Cohort
Study. J Clin Med 8. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/10293238
Park, MH; Gutierrez-Garcia, AK; Choudhury, M. (2019). Mono-(2-ethylhexyl) phthalate aggravates
inflammatory response via sirtuin regulation and inflammasome activation in RAW 264.7 cells.
Chem Res Toxicol 32: 935-942.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/5412425
Pastoor, TP; Lee, KP; Perri, MA; Gillies, PJ. (1987). Biochemical and morphological studies of
ammonium perfluorooctanoate-induced hepatomegaly and peroxisome proliferation. Exp Mol
Pathol 47: 98-109. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3748971
Patel, JC; Mehta, BC. (1999). Tetanus: Study of 8,697 cases. Indian J Med Sci 53: 393-401.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/10176842
Pecquet, AM; Maier, A; Kasper, S; Sumanas, S; Yadav, J. (2020). Exposure to perfluorooctanoic acid
7-58
-------
DRAFT FOR PUBLIC COMMENT
March 2023
(PFOA) decreases neutrophil migration response to injury in zebrafish embryos. BMC Research
Notes 13: 408. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833701
Peng, S; Yan, L; Zhang, J; Wang, Z; Tian, M; Shen, H. (2013). An integrated metabonomics and
transcriptomics approach to understanding metabolic pathway disturbance induced by
perfluorooctanoic acid. J Pharm Biomed Anal 86: 56-64.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850948
Penland, TN; Cope, WG; Kwak, TJ; Strynar, MJ; Grieshaber, CA; Heise, RJ; Sessions, FW. (2020).
Trophodynamics of Per- and Polyfluoroalkyl Substances in the Food Web of a Large Atlantic
Slope River. Environ Sci Technol 54: 6800-6811.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6512132
Pennati, G; Corno, C; Costantino, ML; Bellotti, M. (2003). Umbilical flow distribution to the liver and
the ductus venosus in human fetuses during gestation: an anatomy-based mathematical modeling.
MedEngPhys 25: 229-238.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642023
Pennings, JLA; Jennen, DGJ; Nygaard, UC; Namork, E; Haug, LS; van Loveren, H; Granum, B. (2016).
Cord blood gene expression supports that prenatal exposure to perfluoroalkyl substances causes
depressed immune functionality in early childhood. J Immunotoxicol 13: 173-180.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3352001
Peraza, MA; Burdick, AD; Marin, HE; Gonzalez, FJ; Peters, JM. (2006). The toxicology of ligands for
peroxisome proliferator-activated receptors (PPAR) [Review]. Toxicol Sci 90: 269-295.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/509877
Perez, F; Nadal, M; Navarro-Ortega, A; Fabrega, F; Domingo, JL; Barcelo, D; Farre, M. (2013).
Accumulation of perfluoroalkyl substances in human tissues. Environ Int 59: 354-362.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2325349
Perkins, RG; Butenhoff, JL; Kennedy, GL; Palazzolo, MJ. (2004). 13-week dietary toxicity study of
ammonium perfluorooctanoate (APFO) in male rats. Drug Chem Toxicol 27: 361-378.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291118
Peropadre, A; Freire, PF; Hazen, MJ. (2018). A moderate exposure to perfluorooctanoic acid causes
persistent DNA damage and senescence in human epidermal HaCaT keratinocytes. Food Chem
Toxicol 121: 351-359.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080270
Petersen, MS; Hailing, J; Jorgensen, N; Nielsen, F; Grandjean, P; Jensen, TK; Weihe, P. (2018).
Reproductive function in a population of young Faroe se men with elevated exposure to
polychlorinated biphenyls (pcbs) and perfluorinated alkylate substances (pfas). Int J Environ Res
Public Health 15: n/a. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080277
Petro, EM; D'Hollander, W; Covaci, A; Bervoets, L; Fransen, E; De Neubourg, D; De Pauw, I; Leroy, JL;
Jorssen, EP; Bols, PE. (2014). Perfluoroalkyl acid contamination of follicular fluid and its
consequence for in vitro oocyte developmental competence. Sci Total Environ 496: 282-288.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850178
Pierozan, P; Cattani, D; Karlsson, O. (2020). Perfluorooctane sulfonate (PFOS) and perfluorooctanoic
acid (PFOA) induce epigenetic alterations and promote human breast cell carcinogenesis in vitro.
Arch Toxicol 94: 3893-3906.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833637
Pierozan, P; Jerneren, F; Karlsson, O. (2018). Perfluorooctanoic acid (PFOA) exposure promotes
proliferation, migration and invasion potential in human breast epithelial cells. Arch Toxicol 92:
1729-1739. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4241050
Pilkerton, CS; Hobbs, GR; Lilly, C; Knox, SS. (2018). Rubella immunity and serum perfluoroalkyl
substances: Sex and analytic strategy. PLoS ONE 13: e0203330.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080265
Pinney, SM; Windham, GC; Xie, C; Herrick, RL; Calafat, AM; Mcwhorter, K; Fassler, CS; Hiatt, RA;
Kushi, LH; Biro, FM. (2019). Perfluorooctanoate and changes in anthropometric parameters with
7-59
-------
DRAFT FOR PUBLIC COMMENT
March 2023
age in young girls in the Greater Cincinnati and San Francisco Bay Area. Int J Hyg Environ
Health 222: 1038-1046.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315819
Pirali, B; Negri, S; Chytiris, S; Perissi, A; Villani, L; La Manna, L; Cottica, D; Ferrari, M; Imbriani, M;
Rotondi, M; Chiovato, L. (2009). Perfluorooctane sulfonate and perfluorooctanoic acid in
surgical thyroid specimens of patients with thyroid diseases. Thyroid 19: 1407-1412.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/757881
Pitter, G; Zare Jeddi, M; Barbieri, G; Gion, M; Fabricio, ASC; Dapra, F; Russo, F; Fletcher, T; Canova,
C. (2020). Perfluoroalkyl substances are associated with elevated blood pressure and
hypertension in highly exposed young adults. Environ Health 19: 102.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988479
Pizzurro, DM; Seeley, M; Kerper, LE; Beck, BD. (2019). Interspecies differences in perfluoroalkyl
substances (PFAS) toxicokinetics and application to health-based criteria [Review]. Regul
Toxicol Pharmacol 106: 239-250.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387175
Podder, A, .; Sadmani, A, .; Reinhart, D, .; Chang, N, . B.; Goel, R, . (2021). Per and poly-fluoroalkyl
substances (PFAS) as a contaminant of emerging concern in surface water: A transboundary
review of their occurrences and toxicity effects. J Hazard Mater 419: 126361.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9640865
Poothong, S; Padilla-Sanchez, JA; Papadopoulou, E; Giovanoulis, G; Thomsen, C; Haug, LS. (2019).
Hand Wipes: A Useful Tool for Assessing Human Exposure to Poly- and Perfluoroalkyl
Substances (PFASs) through Hand-to-Mouth and Dermal Contacts. Environ Sci Technol 53:
1985-1993. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080584
Poothong, S; Papadopoulou, E; Padilla-Sanchez, JA; Thomsen, C; Haug, LS. (2020). Multiple pathways
of human exposure to poly- and perfluoroalkyl substances (PFASs): From external exposure to
human blood. Environ Int 134: 105244.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311690
Poothong, S; Thomsen, C; Padilla-Sanchez, JA; Papadopoulou, E; Haug, LS. (2017). Distribution of
novel and well-known poly- and perfluoroalkyl substances (PFASs) in human serum, plasma, and
whole blood. Environ Sci Technol 51: 13388-13396.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239163
Porpora, MG; Lucchini, R; Abballe, A; Ingelido, AM; Valentini, S; Fuggetta, E; Cardi, V; Ticino, A;
Marra, V; Fulgenzi, AR; Felip, ED. (2013). Placental transfer of persistent organic pollutants: a
preliminary study on mother-newborn pairs. Int J Environ Res Public Health 10: 699-711.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2150057
Portier, K; Tolson, JK; Roberts, SM. (2007). Body weight distributions for risk assessment. Risk Anal 27:
11-26. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/192981
Post, GB; Birnbaum, LS; Dewitt, JC; Goeden, H; Heiger-Bernays, WJ; Schlezinger, JJ. (2022). Letter to
the editors regarding "The conundrum of the PFOA human half-life, an international
collaboration" [Letter], Regul Toxicol Pharmacol 134: 105240.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10492320
Post, GB; Cohn, PD; Cooper, KR. (2012). Perfluorooctanoic acid (PFOA), an emerging drinking water
contaminant: a critical review of recent literature [Review]. Environ Res 116: 93-117.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290868
Pouwer, MG; Pieterman, EJ; Chang, SC; Olsen, GW; Caspers, MPM; Verschuren, L; Jukema, JW;
Princen, HMG. (2019). Dose effects of ammonium perfluorooctanoate on lipoprotein metabolism
in apoe*3-leiden.cetp mice. Toxicol Sci 168: 519-534.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080587
Predieri, B; Iughetti, L; Guerranti, C; Bruzzi, P; Perra, G; Focardi, SE. (2015). High Levels of
Perfluorooctane Sulfonate in Children at the Onset of Diabetes. International Journal of
Endocrinology 2015: 234358.
7-60
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3889874
Preston, EV; Rifas-Shiman, SL; Hivert, MF; Zota, AR; Sagiv, SK; Calafat, AM; Oken, E; James-Todd, T.
(2020). Associations of per- and polyfluoroalkyl substances (PFAS) with glucose tolerance
during pregnancy in project viva. J Clin Endocrinol Metab 105: E2864-E2876.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833657
Preston, EV; Webster, TF; Oken, E; Claus Henn, B; Mcclean, MD; Rifas-Shiman, SL; Pearce, EN;
Braverman, LE; Calafat, AM; Ye, X; Sagiv, SK. (2018). Maternal plasma per- and
polyfluoroalkyl substance concentrations in early pregnancy and maternal and neonatal thyroid
function in a prospective birth cohort: Project Viva (USA). Environ Health Perspect 126: 027013.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4241056
Pritchard, JA. (1965). Changes in the blood volume during pregnancy and delivery [Review].
Anesthesiology 26: 393-399.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641812
Puttige Ramesh, N; Arora, M; Braun, JM. (2019). Cross-sectional study of the association between serum
perfluorinated alkyl acid concentrations and dental caries among US adolescents (NHANES
1999-2012). BMJ Open 9: e024189.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080517
Qazi, MR; Abedi, MR; Nelson, BD; Depierre, JW; Abedi-Valugerdi, M. (2010). Dietary exposure to
perfluorooctanoate or perfluorooctane sulfonate induces hypertrophy in centrilobular hepatocytes
and alters the hepatic immune status in mice. Int Immunopharmacol 10: 1420-1427.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1276154
Qazi, MR; Bogdanska, J; Butenhoff, JL; Nelson, BD; Depierre, JW; Abedi-Valugerdi, M. (2009). High-
dose, short-term exposure of mice to perfluorooctanesulfonate (PFOS) or perfluorooctanoate
(PFOA) affects the number of circulating neutrophils differently, but enhances the inflammatory
responses of macrophages to lipopolysaccharide (LPS) in a similar fashion. Toxicology 262: 207-
214. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937259
Qazi, MR; Nelson, BD; Depierre, JW; Abedi-Valugerdi, M. (2012). High-dose dietary exposure of mice
to perfluorooctanoate or perfluorooctane sulfonate exerts toxic effects on myeloid and B-
lymphoid cells in the bone marrow and these effects are partially dependent on reduced food
consumption. Food Chem Toxicol 50: 2955-2963.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937236
Qin, P; Liu, R; Pan, X; Fang, X; Mou, Y. (2010). Impact of carbon chain length on binding of
perfluoroalkyl acids to bovine serum albumin determined by spectroscopic methods. J Agric
Food Chem 58: 5561-5567.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858631
Qin, XD; Qian, Z; Vaughn, MG; Huang, J; Ward, P; Zeng, XW; Zhou, Y; Zhu, Y; Yuan, P; Li, M; Bai,
Z; Paul, G; Hao, YT; Chen, W; Chen, PC; Dong, GH; Lee, YL. (2016). Positive associations of
serum perfluoroalkyl substances with uric acid and hyperuricemia in children from Taiwan.
Environ Pollut 212: 519-524.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981721
Qin, XD; Qian, ZM; Dharmage, SC; Perret, J; Geiger, SD; Rigdon, SE; Howard, S; Zeng, XW; Hu, LW;
Yang, BY; Zhou, Y; Li, M; Xu, SL; Bao, WW; Zhang, YZ; Yuan, P; Wang, J; Zhang, C; Tian,
YP; Nian, M; Xiao, X; Chen, W; Lee, YL; Dong, GH. (2017). Association of perfluoroalkyl
substances exposure with impaired lung function in children. Environ Res 155: 15-21.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3869265
Qu, A; Cao, T; Li, Z; Wang, W; Liu, R; Wang, X; Nie, Y; Sun, S; Liu, X; Zhang, X. (2021). The
association between maternal perfluoroalkyl substances exposure and early attention deficit
hyperactivity disorder in children: a systematic review and meta-analysis. Environ Sci Pollut Res
Int 28: 67066-67081. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959569
Quaak, I; de Cock, M; de Boer, M; Lamoree, M; Leonards, P; van de Bor, M. (2016). Prenatal Exposure
to Perfluoroalkyl Substances and Behavioral Development in Children. Int J Environ Res Public
7-61
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Health 13. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981464
Quist, EM; Filgo, AJ; Cummings, CA; Kissling, GE; Hoenerhoff, MJ; Fenton, SE. (2015). Hepatic
mitochondrial alteration in CD-I mice associated with prenatal exposures to low doses of
perfluorooctanoic acid. Toxicol Pathol 43: 546-557.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6570066
Rahman, ML; Zhang, C; Smarr, MM; Lee, S; Honda, M; Kannan, K; Tekola-Ayele, F; Buck Louis, GM.
(2019). Persistent organic pollutants and gestational diabetes: A multi-center prospective cohort
study of healthy US women. Environ Int 124: 249-258.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5024206
Rainieri, S; Conlledo, N; Langerholc, T; Madorran, E; Sala, M; Barranco, A. (2017). Toxic effects of
perfluorinated compounds at human cellular level and on a model vertebrate. Food Chem Toxicol
104: 14-25. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860104
Raleigh, KK; Alexander, BH; Olsen, GW; Ramachandran, G; Morey, SZ; Church, TR; Logan, PW; Scott,
LL; Allen, EM. (2014). Mortality and cancer incidence in ammonium perfluorooctanoate
production workers. Occup Environ Med 71: 500-506.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850270
Rantakokko, P; Mannisto, V; Airaksinen, R; Koponen, J; Viluksela, M; Kiviranta, H; Pihlajamaki, J.
(2015). Persistent organic pollutants and non-alcoholic fatty liver disease in morbidly obese
patients: A cohort study. Environ Health 14: 79.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3 351439
Rashid, F; Ahmad, S; Irudayaraj, JMK. (2020). Effect of Perfluorooctanoic Acid on the Epigenetic and
Tight Junction Genes of the Mouse Intestine. Toxics 8: 64.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833711
Rashid, F; Ramakrishnan, A; Fields, C; Irudayaraj, J. (2020). Acute PFOA exposure promotes
epigenomic alterations in mouse kidney tissues. Toxicol Rep 7: 125-132.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315778
Reardon, AJF; Khodayari Moez, E; Dinu, I; Goruk, S; Field, CJ; Kinniburgh, DW; Macdonald, AM;
Martin, JW; Study, A. (2019). Longitudinal analysis reveals early-pregnancy associations
between perfluoroalkyl sulfonates and thyroid hormone status in a Canadian prospective birth
cohort. Environ Int 129: 389-399.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412435
Rebholz, SL; Jones, T; Herrick, RL; Xie, C; Calafat, AM; Pinney, SM; Woollett, LA. (2016).
Hypercholesterolemia with consumption of PFOA-laced Western diets is dependent on strain and
sex of mice. Toxicol Rep 3: 46-54.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981499
Reece, PA; Stafford, I; Russell, J; Gill, PG. (1985). Nonlinear renal clearance of ultrafilterable platinum
in patients treated with cis-dichlorodiammineplatinum (II). 15: 295-299.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642054
Remucal, CK. (2019). Spatial and temporal variability of perfluoroalkyl substances in the Laurentian
Great Lakes [Review]. Environ Sci Process Impacts 21: 1816-1834.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5413103
Ren, Y; Jin, L; Yang, F; Liang, H; Zhang, Z; Du, J; Song, X; Miao, M; Yuan, W. (2020). Concentrations
of perfluoroalkyl and polyfluoroalkyl substances and blood glucose in pregnant women. Environ
Health 19: 88. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833646
Reyes, L; Manalich, R. (2005). Long-term consequences of low birth weight [Review]. Kidney Int Suppl
68: S107-S111. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1065677
Rigden, M; Pelletier, G; Poon, R; Zhu, J; al, e. (2015). Assessment of Urinary Metabolite Excretion After
Rat Acute Exposure to Perfluorooctanoic Acid and Other Peroxisomal Proliferators. Arch
Environ Contam Toxicol 68: 148.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7907801
Rigden, M; Pelletier, G; Poon, R; Zhu, J; Auray-Blais, C; Gagnon, R; Kubwabo, C; Kosarac, I; Lalonde,
7-62
-------
DRAFT FOR PUBLIC COMMENT
March 2023
K; Cakmak, S; Xiao, B; Leingartner, K; Ku, KL; Bose, R; Jiao, J. (2015). Assessment of urinary
metabolite excretion after rat acute exposure to perfluorooctanoic acid and other peroxisomal
proliferators. Arch Environ Contam Toxicol 68: 148-158.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2519093
Robinson, MW; Harmon, C; O'Farrelly, C. (2016). Liver immunology and its role in inflammation and
homeostasis [Review]. Cell Mol Immunol 13: 267-276.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10284350
Romano, ME; Xu, Y; Calafat, AM; Yolton, K; Chen, A; Webster, GM; Eliot, MN; Howard, CR;
Lanphear, BP; Braun, JM. (2016). Maternal serum perfluoroalkyl substances during pregnancy
and duration of breastfeeding. Environ Res 149: 239-246.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981728
Rooney, JP; Oshida, K; Kumar, R; Baldwin, WS; Corton, JC. (2019). Chemical Activation of the
Constitutive Androstane Receptor Leads to Activation of Oxidant-Induced Nrf2. Toxicol Sci 167:
172-189. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988236
Rose, K; Wolf, KK; Ukairo, O; Moore, A; Gaffney, J; Bradford, BU; Wetmore, BA; Andersen, ME;
LeCluyse, EL. (2016). Nuclear Receptor-Mediated Gene Expression Changes in a Human
Hepatic Micropatterned Coculture Model After Treatment with Hepatotoxic Compounds. Applied
in Vitro Toxicology 2: 8-16.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959775
Rosen, MB; Das, KP; Rooney, J; Abbott, B; Lau, C; Corton, JC. (2017). PPARa-independent
transcriptional targets of perfluoroalkyl acids revealed by transcript profiling. Toxicology 387:
95-107. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/3859803
Rosen, MB; Das, KP; Wood, CR; Wolf, CJ; Abbott, BD; Lau, C. (2013). Evaluation of perfluoroalkyl
acid activity using primary mouse and human hepatocytes. Toxicology 308: 129-137.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/2919147
Rosen, MB; Lee, JS; Ren, H; Vallanat, B; Liu, J; Waalkes, MP; Abbott, BD; Lau, C; Corton, JC. (2008).
Toxicogenomic dissection of the perfluorooctanoic acid transcript profile in mouse liver:
evidence for the involvement of nuclear receptors PPAR alpha and CAR. Toxicol Sci 103: 46-56.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1290832
Rosenmai, AK; Ahrens, L; le Godec, T; Lundqvist, J; Oskarsson, A. (2018). Relationship between
peroxisome proliferator-activated receptor alpha activity and cellular concentration of 14
perfluoroalkyl substances in HepG2 cells. J Appl Toxicol 38: 219-226.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4220319
Rosenmai, AK; Nielsen, FK; Pedersen, M; Hadrup, N; Trier, X; Christensen, JH; Vinggaard, AM. (2013).
Fluorochemicals used in food packaging inhibit male sex hormone synthesis. Toxicol Appl
Pharmacol 266: 132-142.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2919164
Rosner, B. (2015). Fundamentals of biostatistics (8th ed.). Boston, MA: Brooks/Cole, Cengage Learning.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10406286
Rotander, A; Toms, LM; Aylward, L; Kay, M; Mueller, JF. (2015). Elevated levels of PFOS and PFHxS
in firefighters exposed to aqueous film forming foam (AFFF). Environ Int 82: 28-34.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3859842
Roth, K; Yang, Z; Agarwal, M; Liu, W; Peng, Z; Long, Z, e; Birbeck, J; Westrick, J; Liu, W; Petriello,
MC. (2021). Exposure to a mixture of legacy, alternative, and replacement per- and
polyfluoroalkyl substances (PFAS) results in sex-dependent modulation of cholesterol
metabolism and liver injury. Environ Int 157: 106843.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/9960592
Rothman, K; Greenland, S; Lash, T. (2008). Modern epidemiology. In Modern Epidemiology (3 ed.).
Philadelphia, PA: Lippincott, Williams & Wilkins.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1260377
Ruark, CD; Song, G; Yoon, M; Verner, MA; Andersen, ME; Clewell, HJ; Longnecker, MP. (2017).
7-63
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Quantitative bias analysis for epidemiological associations of perfluoroalkyl substance serum
concentrations and early onset of menopause. Environ Int 99: 245-254.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981395
Ruggiero, MJ; Miller, H; Idowu, JY; Zitzow, JD; Chang, SC; Hagenbuch, B. (2021). Perfluoroalkyl
Carboxylic Acids Interact with the Human Bile Acid Transporter NTCP. Livers 1: 221 -229.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641806
Ruhl, CE; Everhart, JE. (2009). Elevated Serum Alanine Aminotransferase and gamma-
Glutamyltransferase and Mortality in the United States Population. Gastroenterology 136: 477-
485. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3405056
Ruhl, CE; Everhart, JE. (2013). The Association of Low Serum Alanine Aminotransferase Activity With
Mortality in the US Population. Am J Epidemiol 178: 1702-1711.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2331047
Russell, MH; Waterland, RL; Wong, F. (2015). Calculation of chemical elimination half-life from blood
with an ongoing exposure source: The example of perfluorooctanoic acid (PFOA). Chemosphere
129: 210-216. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851185
Rylander, L; Lindh, CH; Hansson, S. R.; Broberg, K; Kallen, K. (2020). Per- and polyfluoroalkyl
substances in early pregnancy and risk for preeclampsia: a case-control study in Southern
Sweden. Toxics 8: 43.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833607
Sadhu, D. (2002). CHO/HGPRT forward mutation assay - ISO (T6.889.7). (US EPA AR-226-1101).
Bedford, MA: Toxicon Corporation.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10270882
Saejia, P; Lirdprapamongkol, K; Svasti, J; Paricharttanakul, NM. (2019). Perfluorooctanoic Acid
Enhances Invasion of Follicular Thyroid Carcinoma Cells Through NF-kB and Matrix
Metalloproteinase-2 Activation. Anticancer Res 39: 2429-2435.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387114
Sagiv, SK; Rifas-Shiman, SL; Fleisch, AF; Webster, TF; Calafat, AM; Ye, X; Gillman, MW; Oken, E.
(2018). Early Pregnancy Perfluoroalkyl Substance Plasma Concentrations and Birth Outcomes in
Project Viva: Confounded by Pregnancy Hemodynamics? Am J Epidemiol 187: 793-802.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238410
Sakolish, C; Chen, Z; Dalaijamts, C; Mitra, K; Liu, Y; Fulton, T; Wade, TL; Kelly, EJ; Rusyn, I; Chiu,
WA. (2020). Predicting tubular reabsorption with a human kidney proximal tubule tissue-on-a-
chip and physiologically-based modeling. Toxicol In Vitro 63: 104752.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6320196
Sakr, CJ; Kreckmann, KH; Green, JW; Gillies, PJ; Reynolds, JL; Leonard, RC. (2007). Cross-sectional
study of lipids and liver enzymes related to a serum biomarker of exposure (ammonia
perfluorooctanoate or APFO) as part of a general health survey in a cohorent of occupational
exposed workers. J Occup Environ Med 49: 1086-1096.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291103
Sakr, CJ; Leonard, RC; Kreckmann, KH; Slade, MD; Cullen, MR. (2007). Longitudinal study of serum
lipids and liver enzymes in workers with occupational exposure to ammonium
perfluorooctanoate. J Occup Environ Med 49: 872-879.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1430761
Sakr, CJ; Symons, JM; Kreckmann, KH; Leonard, RC. (2009). Ischaemic heart disease mortality study
among workers with occupational exposure to ammonium perfluorooctanoate. Occup Environ
Med 66: 699-703. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2593135
Salihovic, S; Stubleski, J; Karrman, A; Larsson, A; Fall, T; Lind, L; Lind, PM. (2018). Changes in
markers of liver function in relation to changes in perfluoroalkyl substances - A longitudinal
study. Environ Int 117: 196-203.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5083555
Salimi, A; Nikoosiar Jahromi, M; Pourahmad, J. (2019). Maternal exposure causes mitochondrial
7-64
-------
DRAFT FOR PUBLIC COMMENT
March 2023
dysfunction in brain, liver, and heart of mouse fetus: An explanation for perfluorooctanoic acid
induced abortion and developmental toxicity. Environ Toxicol 34: 878-885.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381528
Salvalaglio, M; Muscionico, I; Cavallotti, C. (2010). Determination of energies and sites of binding of
PFOA and PFOS to human serum albumin. J Phys Chem B 114: 14860-14874.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919252
Sanchez Garcia, D; Sjodin, M; Hellstrandh, M; Norinder, U; Nikiforova, V; Lindberg, J; Wincent, E;
Bergman, A; Cotgreave, I; Munic Kos, V. (2018). Cellular accumulation and lipid binding of
perfluorinated alkylated substances (PFASs) - A comparison with lysosomotropic drugs. Chem
Biol Interact 281: 1-10.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4234856
Savitz, DA; Stein, CR; Bartell, SM; Elston, B; Gong, J; Shin, HM; Wellenius, GA. (2012).
Perfluorooctanoic acid exposure and pregnancy outcome in a highly exposed community.
Epidemiology 23: 386-392.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/1276141
Savitz, DA; Stein, CR; Elston, B; Wellenius, GA; Bartell, SM; Shin, HM; Vieira, VM; Fletcher, T.
(2012). Relationship of perfluorooctanoic Acid exposure to pregnancy outcome based on birth
records in the mid-ohio valley. Environ Health Perspect 120: 1201-1207.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1424946
Schaaf, MJM. (2017). Nuclear receptor research in zebrafish [Review]. J Mol Endocrinol 59: R65-R76.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/103 65 760
Schaider, LA; Balan, SA; Blum, A; Andrews, DQ; Strynar, MJ; Dickinson, ME; Lunderberg, DM; Lang,
JR; Peaslee, GF. (2017). Fluorinated compounds in US fast food packaging. Environ Sci Technol
Lett 4: 105-111. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981864
Schecter, A; Colacino, J; Haffner, D; Patel, K; Opel, M; Papke, O; Birnbaum, L. (2010). Perfluorinated
compounds, polychlorinated biphenyls, and organochlorine pesticide contamination in composite
food samples from Dallas, Texas, USA. Environ Health Perspect 118: 796-802.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/729962
Schiller, JS; Lucas, JW; Ward, BW; Peregoy, JA. (2012). Summary health statistics for U.S. adults:
National Health Interview Survey, 2010. (DHHS Publication No. (PHS) 2012-1580). Hyattsville,
MD: National Center for Health Statistics.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1798736
Schlegel, R, MacGregor, J.,.T. (1984). The persistence of micronucleated erythrocytes in the peripheral
circulation of normal and splenectomized Fischer 344 rats: Implications for cytogenetic
screening. Mutat Res 127: 169-174.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10368697
Schlezinger, JJ; Puckett, H; Oliver, J; Nielsen, G; Heiger-Bernays, W; Webster, S. (2020).
Perfluorooctanoic acid activates multiple nuclear receptor pathways and skews expression of
genes regulating cholesterol homeostasis in liver of humanized PPARa mice fed an American
diet. Toxicol Appl Pharmacol 405: 115204.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833593
Schlummer, M; Gruber, L; Fiedler, D; Kizlauskas, M; Miiller, J. (2013). Detection of fluorotelomer
alcohols in indoor environments and their relevance for human exposure. Environ Int 57-58: 42-
49. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2552131
Schreder, E; Dickman, J. (2018). Take Out Toxics: PFAS Chemicals in Food Packaging. Schreder, Erika
Schreder; Dickman, Jennifer.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419077
Schumann, G; Bonora, R; Ceriotti, F; Ferard, G; Ferrero, CA; Franck, PFH; Gella, FJ; Hoelzel, W;
Jorgensen. PJ; Kanno, T; Kessner, A; Klauke, R; Kristiansen, N; Lessinger, JM; Linsinger, TPJ;
Misaki, H; Panteghini, M; Pauwels, J; Schiele, F; Schimmel, HG. (2002). IFCC Primary
Reference Procedures for the Measurement of Catalytic Activity Concentrations of Enzymes at
7-65
-------
DRAFT FOR PUBLIC COMMENT
March 2023
37°C. Part 4. Reference Procedure for the Measurement of Catalytic Concentration of Alanine
Aminotransferase. 40: 718-724.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369681
Scinicariello, F; Buser, MC; Balluz, L; Gehle, K; Murray, HE; Abadin, HG; Attanasio, R. (2020).
Perfluoroalkyl acids, hyperuricemia and gout in adults: Analyses of NHANES 2009-2014.
Chemosphere 259: 127446.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833670
Seals, R; Bartell, SM; Steenland, K. (2011). Accumulation and clearance of perfluorooctanoic acid
(PFOA) in current and former residents of an exposed community. Environ Health Perspect 119:
119-124. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919276
Selgrade, MK. (2007). Immunotoxicity: The risk is real [Review]. Toxicol Sci 100: 328-332.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/736210
Seo, SH; Son, MH; Choi, SD; Lee, DH; Chang, YS. (2018). Influence of exposure to perfluoroalkyl
substances (PFASs) on the Korean general population: 10-year trend and health effects. Environ
Int 113: 149-161. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238334
Shah, D. (2009). Healthy worker effect phenomenon. Indian J Occup Environ Med 13: 77-79.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9570930
Shah-Kulkarni, S; Kim, BM; Hong, YC; Kim, HS; Kwon, EJ; Park, H; Kim, YJ; Ha, EH. (2016). Prenatal
exposure to perfluorinated compounds affects thyroid hormone levels in newborn girls. Environ
Int 94: 607-613. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859821
Shan, G; Wang, Z, hi; Zhou, L; Du, P, in; Luo, X; Wu, Q; Zhu, L. (2016). Impacts of daily intakes on the
isomeric profiles of perfluoroalkyl substances (PFASs) in human serum. Environ Int 89-90: 62-
70. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3360127
Shan, G; Ye, M; Zhu, B; Zhu, L. (2013). Enhanced cytotoxicity of pentachlorophenol by perfluorooctane
sulfonate or perfluorooctanoic acid in HepG2 cells. Chemosphere 93: 2101-2107.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850950
Shane, HL; Baur, R; Lukomska, E; Weatherly, L; Anderson, SE. (2020). Immunotoxicity and allergenic
potential induced by topical application of perfluorooctanoic acid (PFOA) in a murine model.
Food Chem Toxicol 136: 111114.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316911
Shankar, A; Xiao, J; Ducatman, A. (2011). Perfluoroalkyl chemicals and chronic kidney disease in US
adults. Am J Epidemiol 174: 893-900.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919232
Shankar, A; Xiao, J; Ducatman, A. (2012). Perfluorooctanoic acid and cardiovascular disease in US
adults. Arch Intern Med 172: 1397-1403.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/2919176
Shao, X; Ji, F; Wang, Y; Zhu, L; Zhang, Z; Du, X; Chung, ACK; Hong, Y; Zhao, Q; Cai, Z. (2018).
Integrative Chemical Proteomics-Metabolomics Approach Reveals Acaca/Acacb as Direct
Molecular Targets of PFOA. Anal Chem 90: 11092-11098.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079651
Shapiro, GD; Dodds, L; Arbuckle, TE; Ashley-Martin, J; Ettinger, AS; Fisher, M; Taback, S; Bouchard,
MF; Monnier, P; Dallaire, R; Morisset, AS; Fraser, W. (2016). Exposure to organophosphorus
and organochlorine pesticides, perfluoroalkyl substances, and polychlorinated biphenyls in
pregnancy and the association with impaired glucose tolerance and gestational diabetes mellitus:
The MIREC Study. Environ Res 147: 71-81.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3201206
Shearer, JJ; Callahan, CL; Calafat, AM; Huang, WY; Jones, RR; Sabbisetti, VS; Freedman, ND;
Sampson, JN; Silverman, DT; Purdue, MP; Hofmann, JN. (2021). Serum concentrations of per-
and polyfluoroalkyl substances and risk of renal cell carcinoma. J Natl Cancer Inst 113: 580-587.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7161466
Shen, M; Xiao, Y; Huang, Y; Jing, D; Su, J; Luo, D; Duan, Y; Xiao, S; Li, J; Chen, X. (2022).
7-66
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Perfluoroalkyl substances are linked to incident chronic spontaneous urticaria: A nested case-
control study. Chemosphere 287: 132358.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/10176753
Sheng, N; Cui, R; Wang, J; Guo, Y; Wang, J; Dai, J. (2018). Cytotoxicity of novel fluorinated alternatives
to long-chain perfluoroalkyl substances to human liver cell line and their binding capacity to
human liver fatty acid binding protein. Arch Toxicol 92: 359-369.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4199441
Shi, LC; Zheng, JJ; Yan, SK; Li, YX; Wang, YJ; Liu, XB; Xiao, CX. (2020). Exposure to
Perfluorooctanoic Acid Induces Cognitive Deficits via Altering Gut Microbiota Composition,
Impairing Intestinal Barrier Integrity, and Causing Inflammation in Gut and Brain. J Agric Food
Chem 68: 13916-13928.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/7161650
Shih, YH; Blomberg, AJ; Bind, MA; Holm, D; Nielsen, F; Heilmann, C; Weihe, P; Grandjean, P. (2021).
Serum vaccine antibody concentrations in adults exposed to per- and polyfluoroalkyl substances:
A birth cohort in the Faroe Islands. J Immunotoxicol 18: 85-92.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959487
Shin, HM; Bennett, DH; Calafat, AM; Tancredi, D; Hertz-Picciotto, I. (2020). Modeled prenatal exposure
to per- and polyfluoroalkyl substances in association with child autism spectrum disorder: A case-
control study. Environ Res 186: 109514.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6507470
Shin, HM; Vieira, VM; Ryan, PB; Steenland, K; Bartell, SM. (2011). Retrospective exposure estimation
and predicted versus observed serum perfluorooctanoic acid concentrations for participants in the
C8 Health Project. Environ Health Perspect 119: 1760-1765.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2572313
Shin, HM, oo; Vieira, VM; Ryan, PB; Steenland, K; Bartell, SM. (2013). Retrospective Exposure
Estimation and Predicted versus Observed Serum Perfluorooctanoic Acid Concentrations for
Participants in the C8 Health Project (vol 119, pg 1760, 2011). Environ Health Perspect 121:
A113-A113. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5082426
Shoeib, M; Harner, T; M Webster, G; Lee, SC. (2011). Indoor sources of poly- and perfluorinated
compounds (PFCS) in Vancouver, Canada: implications for human exposure. Environ Sci
Technol 45: 7999-8005.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1082300
Shrestha, S; Bloom, MS; Yucel, R; Seegal, RF; Rej, R; Mccaffrey, RJ; Wu, Q; Kannan, K; Fitzgerald,
EF. (2017). Perfluoroalkyl substances, thyroid hormones, and neuropsychological status in older
adults. Int J Hyg Environ Health 220: 679-685.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981382
Shrestha, S; Bloom, MS; Yucel, R; Seegal, RF; Wu, Q; Kannan, K; Rej, R; Fitzgerald, EF. (2015).
Perfluoroalkyl substances and thyroid function in older adults. Environ Int 75: 206-214.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851052
Sibai, BM; Frangieh, A. (1995). Maternal adaptation to pregnancy. Curr Opin Obstet Gynecol 7: 420-426.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1101373
Silverstein, RL; Febbraio, M. (2009). CD36, a scavenger receptor involved in immunity, metabolism,
angiogenesis, and behavior [Review]. Science Signaling 2: re3.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365842
Singer, AB; Whitworth, KW; Haug, LS; Sabaredzovic, A; Impinen, A; Papadopoulou, E; Longnecker,
MP. (2018). Menstrual cycle characteristics as determinants of plasma concentrations of
perfluoroalkyl substances (PFASs) in the Norwegian Mother and Child Cohort (MoBa study).
Environ Res 166: 78-85.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079732
Sinisalu, L; Sen, P; Salihovic, S; Virtanen, SM; Hyoty, H; Ilonen, J; Toppari, J; Veijola, R; Oresic, M;
Knip, M; Hyotylainen, T. (2020). Early-life exposure to perfluorinated alkyl substances
7-67
-------
DRAFT FOR PUBLIC COMMENT
March 2023
modulates lipid metabolism in progression to celiac disease. Environ Res 188: 109864.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7211554
Sinisalu, L; Yeung, LWY; Wang, J; Pan, Y; Dai, J; Hyotylainen, T. (2021). Prenatal exposure to poly-
/per-fluoroalkyl substances is associated with alteration of lipid profiles in cord-blood.
Metabolomics 17: 103.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/9959547
Skogheim, TS; Villanger, GD; Weyde, KVF; Engel, SM; Suren, P; 0ie, MG; Skogan, AH; Biele, G;
Zeiner, P; 0vergaard, KR; Haug, LS; Sabaredzovic, A; Aase, H. (2019). Prenatal exposure to
perfluoroalkyl substances and associations with symptoms of attention-deficit/hyperactivity
disorder and cognitive functions in preschool children. Int J Hyg Environ Health 223: 80-92.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/5918847
Skuladottir, M; Ramel, A; Rytter, D; Haug, LS; Sabaredzovic, A; Bech, BH; Henriksen, TB; Olsen, SF;
Halldorsson, TI. (2015). Examining confounding by diet in the association between
perfluoroalkyl acids and serum cholesterol in pregnancy. Environ Res 143: 33-38.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/3749113
Smit, LA; Lenters, V; Hover. BB; Lindh, CH; Pedersen, HS; Liermontova, I; Jonsson, BA; Piersma, AH;
Bonde, JP; Toft, G; Vermeulen, R; Heederik, D. (2015). Prenatal exposure to environmental
chemical contaminants and asthma and eczema in school-age children. Allergy 70: 653-660.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/2823268
Smith, E; Weber, J; Rofe, A; Gancarz, D; Naidu, R; Juhasz, AL. (2012). Assessment of DDT Relative
Bioavailability and Bioaccessibility in Historically Contaminated Soils Using an in Vivo Mouse
Model and Fed and Unfed Batch in Vitro Assays. Environ Sci Technol2928-2934.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/6702349
Smith, MT; Guyton, KZ; Gibbons, CF; Fritz, JM; Portier, CJ; Rusyn, I; DeMarini, DM; Caldwell, JC;
Kavlock, RJ; Lambert, PF; Hecht, SS; Bucher, JR; Stewart, BW; Baan, RA; Cogliano, VJ; Straif,
K. (2016). Key characteristics of carcinogens as a basis for organizing data on mechanisms of
carcinogenesis [Review]. Environ Health Perspect 124: 713-721.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/3160486
Smith, MT; Guyton, KZ; Kleinstreuer, N; Borrel, A; Cardenas, A; Chiu, WA; Felsher, DW; Gibbons, CF;
Goodson, WH; Houck, KA; Kane, AB; La Merrill, MA; Lebrec, H; Lowe, L; Mchale, CM;
Minocherhomji, S; Rieswijk, L; Sandy, MS; Sone, H; Wang, A; Zhang, L; Zeise, L; Fielden, M.
(2020). The Key Characteristics of Carcinogens: Relationship to the Hallmarks of Cancer,
Relevant Biomarkers, and Assays to Measure Them [Editorial]. Cancer Epidemiol Biomarkers
Prev. https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/6956443
Smith, PJ; Humiston, SG; Marcuse, EK; Zhao, Z; Dorell, CG; Howes, C; Hibbs, B. (2011). Parental delay
or refusal of vaccine doses, childhood vaccination coverage at 24 months of age, and the Health
Belief Model. 126: 135-146.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/9642143
Smithwick, M; Norstrom, RJ; Mabury, SA; Solomon, K; Evans, TJ; Stirling, I; Taylor, MK; Muir, DC.
(2006). Temporal trends of perfluoroalkyl contaminants in polar bears (Ursus maritimus) from
two locations in the North American Arctic, 1972-2002. Environ Sci Technol 40: 1139-1143.
https://hero.epa.gov/hero/index.cfrn/reference/details/reference_id/1424802
Sobolewski, M; Conrad, K; Allen, JL; Weston, H; Martin, K; Lawrence, BP; Cory-Slechta, DA. (2014).
Sex-specific enhanced behavioral toxicity induced by maternal exposure to a mixture of low dose
endocrine-disrupting chemicals. Neurotoxicology 45: 121-130.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/2851072
Son, HY; Kim, SH; Shin, HI; Bae, HI; Yang, JH. (2008). Perfluorooctanoic acid-induced hepatic toxicity
following 21-day oral exposure in mice. Arch Toxicol 82: 239-246.
https://hero .epa.gov/hero/index.cfim/reference/details/reference_id/1276157
Son, HY; Lee, S; Tak, EN; Cho, HS; Shin, HI; Kim, SH; Yang, JH. (2009). Perfluorooctanoic acid alters
T lymphocyte phenotypes and cytokine expression in mice. Environ Toxicol 24: 580-588.
7-68
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290821
Song, M, i-Kyung; Cho, Y, oon; Jeong, S, eung-Chan; Ryu, J, ae-Chun. (2016). Analysis of gene
expression changes in relation to hepatotoxicity induced by perfluorinated chemicals in a human
hepatoma cell line. Toxicol Environ Health Sci 8: 114-127.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9959776
Song, P; Li, D; Wang, X; Zhong, X. (2018). Effects of perfluorooctanoic acid exposure during pregnancy
on the reproduction and development of male offspring mice. Andrologia 50: el3059.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079725
Song, P; Li, D; Wang, X; Zhong, X. (2019). Lycopene protects from perfluorooctanoic acid induced liver
damage and uterine apoptosis in pregnant mice. International Journal of Clinical and
Experimental Medicine 12: 212-219.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079965
Song, X; Tang, S; Zhu, H; Chen, Z; Zang, Z; Zhang, Y; Niu, X; Wang, X; Yin, H; Zeng, F; He, C.
(2018). Biomonitoring PFAAs in blood and semen samples: Investigation of a potential link
between PFAAs exposure and semen mobility in China. Environ Int 113: 50-54.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4220306
Sorli. JB; Lag, M; Ekeren, L; Perez-Gil, J; Haug, LS; Da Silva, E; Matrod, MN; Giitzkow, KB;
Lindeman, B. (2020). Per- and polyfluoroalkyl substances (PFASs) modify lung surfactant
function and pro-inflammatory responses in human bronchial epithelial cells. Toxicol In Vitro 62:
104656. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5918817
Spliethoff, HM; Tao, L; Shaver, SM; Aldous, KM; Pass, KA; Kannan, K; Eadon, GA. (2008). Use of
newborn screening program blood spots for exposure assessment: declining levels of
perluorinated compounds in New York State infants. Environ Sci Technol 42: 5361-5367.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919368
Spratlen, MJ; Perera, FP; Lederman, SA; Rauh, VA; Robinson, M; Kannan, K; Trasande, L; Herbstman,
J. (2020). The association between prenatal exposure to perfluoroalkyl substances and childhood
neurodevelopment. Environ Pollut 263: 114444.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6364693
Spratlen, MJ; Perera, FP; Lederman, SA; Robinson, M; Kannan, K; Herbstman, J; Trasande, L. (2020).
The association between perfluoroalkyl substances and lipids in cord blood. J Clin Endocrinol
Metab 105: 43-54. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5915332
Staples, RE; Burgess, BA; Kerns, WD. (1984). The embryo-fetal toxicity and teratogenic potential of
ammonium perfluorooctanoate (APFO) in the rat. Fundam Appl Toxicol 4: 429-440.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332669
Starling, AP; Adgate, JL; Hamman, RF; Kechris, K; Calafat, AM; Ye, X; Dabelea, D. (2017).
Perfluoroalkyl substances during pregnancy and offspring weight and adiposity at birth:
Examining mediation by maternal fasting glucose in the healthy start study. Environ Health
Perspect 125: 067016.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858473
Starling, AP; Engel, SM; Richardson, DB; Baird, DD; Haug, LS; Stuebe, AM; Klungsoyr, K; Harmon, Q;
Becher, G; Thomsen, C; Sabaredzovic, A; Eggesbo, M; Hoppin, JA; Travlos, GS; Wilson, RE;
Trogstad, LI; Magnus, P, er; Longnecker, MP. (2014). Perfluoroalkyl Substances During
Pregnancy and Validated Preeclampsia Among Nulliparous Women in the Norwegian Mother
and Child Cohort Study. Am J Epidemiol 179: 824-833.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2446669
Starling, AP; Engel, SM; Whitworth, KW; Richardson, DB; Stuebe, AM; Daniels, JL; Haug, LS;
Eggesbo, M; Becher, G; Sabaredzovic, A; Thomsen, C; Wilson, RE; Travlos, GS; Hoppin, JA;
Baird, DD; Longnecker, MP. (2014). Perfluoroalkyl substances and lipid concentrations in
plasma during pregnancy among women in the Norwegian Mother and Child Cohort Study.
Environ Int 62: 104-112.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850928
7-69
-------
DRAFT FOR PUBLIC COMMENT
March 2023
StataCorp. (2021). Stata Statistical Software: Release 17 [Computer Program]. College Station, TX:
StataCorp LLC. Retrieved from
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10406419
Steenland, K; Barry, V; Savitz, D. (2018). Serum perfluorooctanoic acid and birthweight: an updated
meta-analysis with bias analysis. Epidemiology 29: 765-776.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079861
Steenland, K; Fletcher, T; Stein, CR; Bartell, SM; Darrow, L; Lopez-Espinosa, MJ; Barry Ryan, P;
Savitz, DA. (2020). Review: Evolution of evidence on PFOA and health following the
assessments of the C8 Science Panel [Review]. Environ Int 145: 106125.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/7161469
Steenland, K; Kugathasan, S; Barr, DB. (2018). PFOA and ulcerative colitis. Environ Res 165: 317-321.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079806
Steenland, K; Tinker, S; Frisbee, S; Ducatman, A; Vaccarino, V. (2009). Association of Perfluorooctanoic
Acid and Perfluorooctane Sulfonate With Serum Lipids Among Adults Living Near a Chemical
Plant. Am J Epidemiol 170: 1268-1278.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1291109
Steenland, K; Tinker, S; Shankar, A; Ducatman, A. (2010). Association of perfluorooctanoic acid (PFOA)
and perfluorooctane sulfonate (PFOS) with uric acid among adults with elevated community
exposure to PFOA. Environ Health Perspect 118: 229-233.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290810
Steenland, K; Winquist, A. (2021). PFAS and cancer, a scoping review of the epidemiologic evidence
[Review]. Environ Res 194: 110690.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7491705
Steenland, K; Woskie, S. (2012). Cohort mortality study of workers exposed to perfluorooctanoic acid.
Am J Epidemiol 176: 909-917.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919168
Steenland, K; Zhao, L; Winquist, A. (2015). A cohort incidence study of workers exposed to
perfluorooctanoic acid (PFOA). Occup Environ Med 72: 373-380.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851015
Steenland, K; Zhao, L; Winquist, A; Parks, C. (2013). Ulcerative Colitis and Perfluorooctanoic Acid
(PFOA) in a Highly Exposed Population of Community Residents and Workers in the Mid-Ohio
Valley. Environ Health Perspect 121: 900-905.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937218
Stein, CR; Ge, Y; Wolff, MS; Ye, X; Calafat, AM; Kraus, T; Moran, TM. (2016). Perfluoroalkyl
substance serum concentrations and immune response to FluMist vaccination among healthy
adults. Environ Res 149: 171-178.
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3 860 111
Stein, CR; Savitz, DA; Bellinger, DC. (2013). Perfluorooctanoate and neuropsychological outcomes in
children. Epidemiology 24: 590-599.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850964
Stein, CR; Savitz, DA; Bellinger, DC. (2014). Perfluorooctanoate exposure in a highly exposed
community and parent and teacher reports of behaviour in 6-12-year-old children. Paediatr
Perinat Epidemiol 28: 146-156.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2721873
Stein, CR; Savitz, DA; Dougan, M. (2009). Serum levels of perfluorooctanoic acid and perfluorooctane
sulfonate and pregnancy outcome. Am J Epidemiol 170: 837-846.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290816
Stengel, D; Wahby, S; Braunbeck, T. (2018). In search of a comprehensible set of endpoints for the
routine monitoring of neurotoxicity in vertebrates: sensory perception and nerve transmission in
zebrafish (Danio rerio) embryos. Environ Sci Pollut Res Int 25: 4066-4084.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238489
7-70
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Stock, NL; Furdui, VI; Muir, DC; Mabury, SA. (2007). Perfluoroalkyl contaminants in the Canadian
Arctic: evidence of atmospheric transport and local contamination. Environ Sci Technol 41:
3529-3536. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1289794
Stratakis, N; Rock, S; La Merrill, MA; Saez, M; Robinson, O; Fecht, D; Vrijheid, M; Valvi, D; Conti,
DV; McConnell, R; Chatzi, VL. (2022). Prenatal exposure to persistent organic pollutants and
childhood obesity: A systematic review and meta-analysis of human studies. Obes Rev 23(S1):
e 13383. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10176437
Strom, M; Hansen, S; Olsen, SF; Haug, LS; Rantakokko, P; Kiviranta, H; Halldorsson, TI. (2014).
Persistent organic pollutants measured in maternal serum and offspring neurodevelopmental
outcomes—a prospective study with long-term follow-up. Environ Int 68: 41-48.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2922190
Su, TC; Kuo, CC; Hwang, JJ; Lien, GW; Chen, MF; Chen, PC. (2016). Serum perfluorinated chemicals,
glucose homeostasis and the risk of diabetes in working-aged Taiwanese adults. Environ Int 88:
15-22. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860116
Suh, CH; Cho, NK; Lee, CK; Lee, CH; Kim, DH; Kim, JH; Son, BC; Lee, JT. (2011). Perfluorooctanoic
acid-induced inhibition of placental prolactin-family hormone and fetal growth retardation in
mice. Mol Cell Endocrinol 337: 7-15.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1402560
Sun, Q; Zong, G; Valvi, D; Nielsen, F; Coull, B; Grandjean, P. (2018). Plasma concentrations of
perfluoroalkyl substances and risk of Type 2 diabetes: A prospective investigation among U.S.
women. Environ Health Perspect 126: 037001.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4241053
Sun, S; Guo, H, ua; Wang, J; Dai, J. (2019). Hepatotoxicity of perfluorooctanoic acid and two emerging
alternatives based on a 3D spheroid model. Environ Pollut 246: 955-962.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5024252
Sun, S; Wang, J; Lu, Y; Dai, J. (2018). Corticosteroid-binding globulin, induced in testicular Leydig cells
by perfluorooctanoic acid, promotes steroid hormone synthesis. Arch Toxicol 92: 2013-2025.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079802
Takagi, A; Sai, K; Umemura, T; Hasegawa, R; Kurokawa, Y. (1991). Short-term exposure to the
peroxisome proliferators, perfluorooctanoic acid and perfluorodecanoic acid, causes significant
increase of 8-hydroxydeoxyguanosine in liver DNA of rats. Cancer Lett 57: 55-60.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2325496
Tan, YM; Clewell, HJ; Andersen, ME. (2008). Time dependencies in perfluorooctylacids disposition in
rat and monkeys: a kinetic analysis. Toxicol Lett 177: 38-47.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919374
Tao, L; Kannan, K; Wong, CM; Arcaro, KF; Butenhoff, JL. (2008). Perfluorinated compounds in human
milk from Massachusetts, USA. Environ Sci Technol 42: 3096-3101.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290895
Taylor, KW; Hoffman, K; Thayer, KA; Daniels, JL. (2014). Polyfluoroalkyl chemicals and menopause
among women 20-65 years of age (NHANES). Environ Health Perspect 122: 145-150.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850915
TCEQ. (2016). Perfluoro compounds (PFCs).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5975349
Thayer, KA; Angrish, M; Arzuaga, X; Carlson, LM; Davis, A; Dishaw, L; Druwe, I; Gibbons, C; Glenn,
B; Jones, R; Kaiser, JP; Keshava, C; Keshava, N; Kraft, A; Lizarraga, L; Persad, A; Radke, EG;
Rice, G; Schulz, B; Shaffer, R; T, S; Shapiro, A; Thacker, S; Vulimiri, S; Williams, AJ; Woodall,
G; Yost, E; Blain, R; Duke, K; Goldstone, AE; Hartman, P; Hobbie, K; Ingle, B; Lemeris, C; Lin,
C; Lindahl, A; McKinley, K; Soleymani, P; Vetter, N. (2022). Systematic evidence map (SEM)
template: report format and methods used for the US EPA integrated risk information system
(iris) program, provisional peer reviewed toxicity value (PPRTV) program, and other "fit for
purpose" literature-based human health analyses (manuscript-in-progress) (pp. 1-69). Thayer,
7-71
-------
DRAFT FOR PUBLIC COMMENT
March 2023
KA; Angrish, M; Arzuaga, X; Carlson, LM; Davis, A; Dishaw, L; Druwe, I; Gibbons, C; Glenn,
B; Jones, R; Kaiser, JP; Keshava, C; Keshava, N; Kraft, A; Lizarraga, L; Persad, A; Radke, EG;
Rice, G; Schulz, B; Shaffer, R; Shannon. T; Shapiro, A; Thacker, S; Vulimiri, S; Williams, AJ;
Woodall, G; Yost, E; Blain, R; Duke, K; Goldstone, AE; Hartman, P; Hobbie, K; Ingle, B;
Lemeris, C; Lin, C; Lindahl, A; McKinley, K; Soleymani, P; Vetter, N.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10259560
Thomford, PJ. (2001). 4-Week capsule toxicity study with ammonium perfluorooctanoate (APFO) in
cynomolgus monkeys. 159.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5432382
Thompson, J; Lorber, M; Toms, LM; Kato, K; Calafat, AM; Mueller, JF. (2010). Use of simple
pharmacokinetic modeling to characterize exposure of Australians to perfluorooctanoic acid and
perfluorooctane sulfonic acid. Environ Int 36: 390-397.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919278
Thompson, J; Lorber, M; Toms, LML; Kato, K; Calafat, AM; Mueller, JF. (2010). Use of simple
pharmacokinetic modeling to characterize exposure of Australians to perfluorooctanoic acid and
perfluorooctane sulfonic acid (vol 36, pg 390, 2010). Environ Int 36: 647-648.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5082271
Thomsen, C; Haug, LS; Stigum, H; Froshaug. M; Broadwell, SL; Becher, G. (2010). Changes in
concentrations of perfluorinated compounds, polybrominated diphenyl ethers, and
polychlorinated biphenyls in Norwegian breast-milk during twelve months of lactation. Environ
Sci Technol 44: 9550-9556.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/759807
Thorsdottir, I; Birgisdottir, BE. (1998). Different weight gain in women of normal weight before
pregnancy: postpartum weight and birth weight. Obstet Gynecol 92: 377-383.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4940407
Tian, M; Reichetzeder, C; Li, J; Hocher, B. (2019). Low birth weight, a risk factor for diseases in later
life, is a surrogate of insulin resistance at birth. J Hypertens 37: 2123-2134.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8632212
Tian, Y; Liang, H; Miao, M; Yang, F; Ji, H; Cao, W; Liu, X; Zhang, X; Chen, A; Xiao, H; Hu, H; Yuan,
W. (2019). Maternal plasma concentrations of perfluoroalkyl and polyfluoroalkyl substances
during pregnancy and anogenital distance in male infants. Hum Reprod 34: 1356-1368.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5390052
Tian, Y; Miao, M; Ji, H; Zhang, X; Chen, A; Wang, Z; Yuan, W; Liang, H. (2020). Prenatal exposure to
perfluoroalkyl substances and cord plasma lipid concentrations. Environ Pollut 268: 115426.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7026251
Tian, YP; Zeng, XW; Bloom, MS; Lin, S; Wang, SQ; Yim, SHL; Yang, M; Chu, C; Gurram, N; Hu, LW;
Liu, KK; Yang, BY; Feng, D; Liu, RQ; Nian, M; Dong, GH. (2019). Isomers of perfluoroalkyl
substances and overweight status among Chinese by sex status: Isomers of C8 Health Project in
China. Environ Int 124: 130-138.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080586
Tilston, EL; Gibson, GR; Collins, CD. (2011). Colon extended physiologically based extraction test (CE-
PBET) increases bioaccessibility of soil-bound PAH. Environ Sci Technol 45: 5301-5308.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5120687
Timmermann, CA; Budtz-Jorgensen. E; Jensen, TK; Osuna, CE; Petersen, MS; Steuerwald, U; Nielsen,
F; Poulsen, LK; Weihe, P; Grandjean, P. (2017). Association between perfluoroalkyl substance
exposure and asthma and allergic disease in children as modified by MMR vaccination. J
Immunotoxicol 14: 39-49.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858497
Timmermann, CA; Budtz-Jorgensen, E; Petersen, MS; Weihe, P; Steuerwald, U; Nielsen, F; Jensen, TK;
Grandjean, P. (2017). Shorter duration of breastfeeding at elevated exposures to perfluoroalkyl
substances. Reprod Toxicol 68: 164-170.
7-72
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981439
Timmermann, CAG; Jensen, KJ; Nielsen, F; Budtz-Jorgensen. E; van Der Klis, F; Benn, CS; Grandjean,
P; Fisker, AB. (2020). Serum Perfluoroalkyl Substances, Vaccine Responses, and Morbidity in a
Cohort of Guinea-Bissau Children. Environ Health Perspect 128: 87002.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833710
Timmermann, CAG; Pedersen, HS; Weihe, P; Bjerregaard, P; Nielsen, F; Heilmann, C; Grandjean, P.
(2021). Concentrations of tetanus and diphtheria antibodies in vaccinated Greenlandic children
aged 7-12 years exposed to marine pollutants, a cross sectional study. Environ Res 203: 111712.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416315
Trefts, E; Gannon, M; Wasserman, DH. (2017). The liver. Curr Biol 27: R1147-R1151.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10284972
Trudel, D; Horowitz, L; Wormuth, M; Scheringer, M; Cousins, IT; Hungerbuhler, K. (2008). Estimating
consumer exposure to PFOS and PFOA.[erratum appears in Risk Anal. 2008 Jun;28(3):807], Risk
Anal 28: 251-269. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/214241
Tsai, MS; Chang, SH; Kuo, WH; Kuo, CH; Li, SY; Wang, MY; Chang, DY; Lu, YS; Huang, CS; Cheng,
AL; Lin, CH; Chen, PC. (2020). A case-control study of perfluoroalkyl substances and the risk of
breast cancer in Taiwanese women. Environ Int 142: 105850.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833693
Tsai, MS; Lin, CC; Chen, MH; Hsieh, WS; Chen, PC. (2017). Perfluoroalkyl substances and thyroid
hormones in cord blood. Environ Pollut 222: 543-548.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860107
Tsai, MS; Lin, CY; Lin, CC; Chen, MH; Hsu, SH; Chien, KL; Sung, FC; Chen, PC; Su, TC. (2015).
Association between perfluoroalkyl substances and reproductive hormones in adolescents and
young adults. Int J Hyg Environ Health 218: 437-443.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850160
Tucker, DK; Macon, MB; Strynar, MJ; Dagnino, S; Andersen, E; Fenton, SE. (2014). The mammary
gland is a sensitive pubertal target in CD-I and C57B1/6 mice following perinatal
perfluorooctanoic acid (PFOA) exposure. Reprod Toxicol 54: 26-36.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851046
U.S. EPA. (1985). National primary drinking water regulations; synthetic organic chemicals, inorganic
chemicals and microorganisms. Fed Reg 50: 46936-47025.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9207
U.S. EPA. (1986). Guidelines for carcinogen risk assessment [EPA Report] (pp. 33993-34003).
(EPA/630/R-00/004). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
Forum. https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/199530
U.S. EPA. (1991). Guidelines for developmental toxicity risk assessment. Fed Reg 56: 63798-63826.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/732120
U.S. EPA. (1991). National primary drinking water regulations—synthetic organic chemicals and
inorganic chemicals; monitoring for unregulated contaminants; national primary drinking water
regulations implementation; national secondary drinking water regulations. Fed Reg 56: 3526-
3597. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5499
U.S. EPA. (1996). Guidelines for reproductive toxicity risk assessment (pp. 1-143). (EPA/630/R-96/009).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/30019
U.S. EPA. (1998). Health effects test guidelines OPPTS 870.3800 reproduction and fertility effects [EPA
Report]. (EPA 712-C-98-208). Washington D.C.: U.S. Environmental Protection Agency, Office
of Prevention, Pesticides and Toxic Substances.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2229410
U.S. EPA. (1998). National Primary Drinking Water Regulations: Disinfectants and Disinfection
Byproducts, Federal Register: 63 FR 69390. 63: 69390-69476.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10442462
7-73
-------
DRAFT FOR PUBLIC COMMENT
March 2023
U.S. EPA. (1999). Guidelines for carcinogen risk assessment [review draft] [EPA Report]. (NCEA-F-
0644). Washington, DC: U.S. Environmental Protection Agency, Office of the Science Advisor.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/41631
U.S. EPA. (2000). Methodology for deriving ambient water quality criteria for the protection of human
health (2000). (EPA/822/B-00/004). Washington, DC: U.S. Environmental Protection Agency,
Office of Water, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/19428
U.S. EPA. (2000). National Primary Drinking Water Regulations; Radionuclides; Final Rule. Federal
Register: 65 FR 76708. 65: 76708-76753.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10442463
U.S. EPA. (2001). National Primary Drinking Water Regulations; Arsenic and Clarifications to
Compliance and New Source Contaminants Monitoring: Delay of Effective Date. Federal
Register: 66 FR 28342. 66: 28342-28350.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10442464
U.S. EPA. (2002). Hepatocellular hypertrophy. HED guidance document #G2002.01 [EPA Report].
Washington, DC. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/625713
U.S. EPA. (2002). A review of the reference dose and reference concentration processes.
(EPA630P02002F). Washington, DC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/88824
U.S. EPA. (2005). Guidelines for carcinogen risk assessment [EPA Report]. (EPA630P03001F).
Washington, DC. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6324329
U.S. EPA (U.S. Environmental Protection Agency). (2005). Supplemental guidance for assessing
susceptibility from early-life exposure to carcinogens [EPA Report]. (EPA/630/R-03/003F).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/88823
U.S. EPA. (2006). 2010/2015 PFOA stewardship program. Available online at
http://www2.epa.gov/assessing-and-managing-chemicals-under-tsca/20102015-pfoa-stewardship-
program; https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/pfoa-stewardship-
program-baseline-year-summary-report 3005012
U.S. EPA. (2009). Final Contaminant Candidate List 3 Chemicals: Screening to a PCCL. (815R09007).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1508321
U.S. EPA. (2009). Perfluorocarboxylic acid content in 116 articles of commerce. (EPA/600/R-09/033).
Research Triangle Park, NC: National Risk Management Research Laboratory, Office of
Research and Development,U.S. Environmental Protection Agency.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290922
U.S. EPA. (2010). 2008-2009 National Rivers and Streams Assessment Fish Tissue Study. Washington,
DC. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369692
U.S. EPA. (2011). 2010 Great Lakes Human Health Fish Tissue Study. Washington, DC: U.S.
Environmental Protection Agency, National Coastal Condition Assessment.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369695
U.S. EPA. (2011). Age Dependent Adjustment Factor (ADAF) application [EPA Report]. Washington,
DC. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/783747
U.S. EPA. (2011). Exposure factors handbook: 2011 edition [EPA Report]. (EPA/600/R-090/052F).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development,
National Center for Environmental Assessment.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/786546
U.S. EPA (U.S. Environmental Protection Agency). (2011). Toxicological Review of Trichloroethylene
(CASRN 79-01-6) in support of summary information on the Integrated Risk Information System
(IRIS). Washington, DC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642147
U.S. EPA. (2012). Benchmark dose technical guidance. (EPA/100/R-12/001). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum.
7-74
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/123 943 3
U.S. EPA. (2014). Guidance for applying quantitative data to develop data-derived extrapolation factors
for interspecies and intraspecies extrapolation [EPA Report]. (EPA/100/R-14/002F). Washington,
DC: Risk Assessment Forum, Office of the Science Advisor.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2520260
U.S. EPA. (2015). 2013-2014 National Rivers and Streams Assessment Fish Tissue Study. Washington,
DC. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369694
U.S. EPA (U.S. Environmental Protection Agency). (2016). Drinking Water Contaminant Candidate List
4-Final, Federal Registrar: 81 FR 81099. 81: 81099.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6307617
U.S. EPA. (2016). Drinking water health advisory for perfluorooctane sulfonate (PFOS) [EPA Report].
(EPA 822-R-16-004). Washington, DC: U.S. Environmental Protection Agency, Office of Water.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3982043
U.S. EPA. (2016). Drinking water health advisory for perfluorooctanoic acid (PFOA) [EPA Report].
(EPA 822-R-16-005). Washington, DC: U.S. Environmental Protection Agency, Office of Water.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3982042
U.S. EPA. (2016). Health effects support document for perfluorooctane sulfonate (PFOS) [EPA Report].
(EPA 822-R-16-002). Washington, DC: U.S. Environmental Protection Agency, Office of Water,
Health and Ecological Criteria Division.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/3 603 3 65
U.S. EPA. (2016). Health effects support document for perfluorooctanoic acid (PFOA) [EPA Report].
(EPA 822-R-16-003). Washington, DC: U.S. Environmental Protection Agency, Office of Water,
Health and Ecological Criteria Division.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3603279
U.S. EPA. (2016). National Coastal Condition Assessment: 2015 Results. Washington, DC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369696
U.S. EPA. (2017). Occurrence Data for the Unregulated Contaminant Monitoring Rule: UCMR 3 (2013-
2015). Washington, D.C.: U.S. Environmental Protection Agency, Office of Water. Retrieved
from https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9419085
U.S. EPA. (2018). An umbrella Quality Assurance Project Plan (QAPP) for PBPK models [EPA Report].
(ORD QAPP ID No: B-0030740-QP-1-1). Research Triangle Park, NC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4326432
U.S. EPA. (2019). Exposure factors handbook chapter 3 (update): Ingestion of water and other select
liquids [EPA Report], (EPA/600/R-18/259F). Washington, DC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7267482
U.S. EPA. (2019). Systematic review protocol for the PFBA, PFHxA, PFHxS, PFNA, and PFDA IRIS
assessments [EPA Report]. (EPA635R19050). Integrated Risk Information System. Center for
Public Health and Environmental Assessment. Office of Research and Development. U.S.
Environmental Protection Agency.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6572089
U.S. EPA. (2020). 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://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7006986
U.S. EPA. (2020). Systematic review protocol for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (anionic
and acid forms) IRIS assessments: Supplemental information appendix A [EPA Report].
(EPA/635/R-20/131). Washington, DC: US EPA, ORD, CPHEA, Integrated Risk Information
System, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8642427
U.S. EPA. Announcement of final regulatory determinations for contaminants on the Fourth Drinking
Water Contaminant Candidate List, 86 FR 12272 (2021).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7487276
7-75
-------
DRAFT FOR PUBLIC COMMENT
March 2023
U.S. EPA. (2021). EXTERNAL PEER REVIEW DRAFT: Proposed Approaches to the Derivation of a
Draft Maximum Contaminant Level Goal for Perfluorooctane Sulfonic Acid (PFOS) (CASRN
1763-23-1) in Drinking Water [EPA Report]. Washington, DC: U.S. Environmental Protection
Agency (EPA). https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10428576
U.S. EPA. (2021). EXTERNAL PEER REVIEW DRAFT: Proposed Approaches to the Derivation of a
Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-
1) in Drinking Water [EPA Report]. Washington, DC: U.S. Environmental Protection Agency
(EPA), https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10428559
U.S. EPA. (2021). Final Regulatory Determination 4 Support Document [EPA Report]. (EPA
815R21001). U.S. Environmental Protection Agency (EPA).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9640861
U.S. EPA. (2021). 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.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9960186
U.S. EPA. (2021). 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://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7310530
U.S. EPA. (2021). 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://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10064222
U.S. EPA. (2022). Draft Aquatic Life Ambient Water Quality Criteria for Perfluorooctanoic Acid
(PFOA). (EPA-842-D-22-001). Washington, DC: U.S. Environmental Protection Agency, Office
of Water, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10671186
U.S. EPA. (2022). Draft Economic Analysis for the Proposed PFAS Rule. Washington, DC.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369698
U.S. EPA. (2022). INTERIM Drinking Water Health Advisory: Perfluorooctanoic Acid (PFOA) CASRN
335-67-1. (EPA/822/R-22/003). Washington, DC: U.S. Environmental Protection Agency, Office
of Water, Office of Science and Technology.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10671184
U.S. EPA. (2022). Review of EPA's Analysis to Support EPA's National Primary Drinking Water
Rulemaking for PFAS. (EPA-SAB-22-008). U.S. Environmental Protection Agency, Science
Advisory Board, https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10476098
U.S. EPA. (2023). Economic analysis for the proposed PFAS national primary drinking water regulation
[EPA Report], (EPA-822-P-22-001).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10692765
U.S. EPA. (2023). Technical support document - per- and polyfluoroalkyl substances (PFAS) occurrence
& contaminant background [EPA Report]. (EPA-822-P-22-007).
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10692764
Uhl, SA; James-Todd, T; Bell, ML. (2013). Association of Osteoarthritis with Perfluorooctanoate and
Perfluorooctane Sulfonate inNHANES 2003-2008. Environ Health Perspect 121: 447-452,
452e441-443. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1937226
Vagi, SJ; Azziz-Baumgartner, E; Sjodin, A; Calafat, AM; Dumesic, D; Gonzalez, L; Kato, K; Silva, MJ;
Ye, X; Azziz, R. (2014). Exploring the potential association between brominated diphenyl ethers,
polychlorinated biphenyls, organochlorine pesticides, perfluorinated compounds, phthalates, and
bisphenol a in polycystic ovary syndrome: a case-control study. BMC Endocrine Disorders 14:
7-76
-------
DRAFT FOR PUBLIC COMMENT
March 2023
86. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2718073
Valenti, L; Pelusi, S; Bianco, C; Ceriotti, F; Berzuini, A; Iogna Prat, L; Trotti, R; Malvestiti, F;
D'Ambrosio, R; Lampertico, P; Colli, A; Colombo, M; Tsochatzis, E; Fraquelli, M; Prati, D.
(2021). Definition of Healthy Ranges for Alanine Aminotransferase Levels: A 2021 Update.
Hepatology Communications 5: 1824-1832.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10369689
Valvi, D; Oulhote, Y; Weihe, P; Dalgard, C; Bjerve, KS; Steuerwald, U; Grandjean, P. (2017).
Gestational diabetes and offspring birth size at elevated environmental pollutant exposures.
Environ Int 107: 205-215.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3983872
van den Dungen, MW; Murk, AJ; Kampman, E; Steegenga, WT; Kok, DE. (2017). Association between
DNA methylation profiles in leukocytes and serum levels of persistent organic pollutants in
Dutch men. Environ Epigenet 3: dvxOOl.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080340
van der Veen, I; Hanning, AC; Stare, A; Leonards, PEG; de Boer, J; Weiss, JM. (2020). The effect of
weathering on per- and polyfluoroalkyl substances (PFASs) from durable water repellent (DWR)
clothing. Chemosphere 249: 126100.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316195
van Esterik, JC; Sales, LB; Dolle, ME; Hakansson, H; Herlin, M; Legler, J; van der Ven, LT. (2015).
Programming of metabolic effects in C57BL/6JxFVB mice by in utero and lactational exposure
to perfluorooctanoic acid. Arch Toxicol 90: 701-715.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850288
Varshavsky, JR; Robinson, JF; Zhou, Y; Puckett, KA; Kwan, E; Buarpung, S; Aburajab, R; Gaw, SL;
Sen, S; Gao, S; Smith, SC; Park, JS; Zakharevich, I; Gerona, RR; Fisher, SJ; Woodruff, TJ.
(2021). Organophosphate Flame Retardants, Highly Fluorinated Chemicals, and Biomarkers of
Placental Development and Disease during Mid-Gestation. Toxicol Sci 181: 215-228.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7410195
Velarde, MC; Chan, AFO; Sajo, MEJ, V; Zakharevich, I; Melamed, J; Uy, GLB; Teves, JMY; Corachea,
AJM; Valparaiso, AP; Macalindong, SS; Cabaluna, ND; Dofitas, RB; Giudice, LC; Gerona, RR.
(2022). Elevated levels of perfluoroalkyl substances in breast cancer patients within the Greater
Manila Area. Chemosphere 286 Pt 1: 131545.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9956482
Velez, MP; Arbuckle, TE; Fraser, WD. (2015). Maternal exposure to perfluorinated chemicals and
reduced fecundity: the MIREC study. Hum Reprod 30: 701-709.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851037
Verner, MA; Loccisano, AE; Morken, NH; Yoon, M; Wu, H; Mcdougall, R; Maisonet, M; Marcus, M;
Kishi, R; Miyashita, C; Chen, MH; Hsieh, WS; Andersen, ME; Clewell, HJ; Longnecker, MP.
(2015). Associations of Perfluoroalkyl Substances (PFAS) with Lower Birth Weight: An
Evaluation of Potential Confounding by Glomerular Filtration Rate Using a Physiologically
Based Pharmacokinetic Model (PBPK). Environ Health Perspect 123: 1317-1324.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3150627
Verner, MA; Ngueta, G; Jensen, ET; Fromme, H; Voelkel, W; Nygaard, UC; Granum, B; Longnecker,
MP. (2016). A simple pharmacokinetic model of prenatal and postnatal exposure to
perfluoroalkyl substances (PFASs). Environ Sci Technol 50: 978-986.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3299692
Vested, A; Ramlau-Hansen, CH; Olsen, SF; Bonde, JP; Kristensen, SL; Halldorsson, TI; Becher, G;
Haug, LS; Ernst, EH; Toft, G. (2013). Associations of in utero exposure to perfluorinated alkyl
acids with human semen quality and reproductive hormones in adult men. Environ Health
Perspect 121: 453-458, 458e451-455.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2317339
Vestergren, R; Cousins, I; Trudel, D; Wormuth, M; Scheringer, M. (2008). Estimating the contribution of
7-77
-------
DRAFT FOR PUBLIC COMMENT
March 2023
precursor compounds in consumer exposure to PFOS and PFOA. Chemosphere 73: 1617-1624.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2558842
Vestergren, R; Cousins, IT. (2009). Tracking the pathways of human exposure to perfluorocarboxylates
[Review]. Environ Sci Technol 43: 5565-5575.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290815
Vieira, VM; Hoffman, K; Shin, HM; Weinberg, JM; Webster, TF; Fletcher, T. (2013). Perfluorooctanoic
acid exposure and cancer outcomes in a contaminated community: a geographic analysis. Environ
Health Perspect 121: 318-323.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919154
Volkel, W; Genzel-Boroviczeny, O; Demmelmair, H; Gebauer, C; Koletzko, B; Twardella, D; Raab, U;
Fromme, H. (2008). Perfluorooctane sulphonate (PFOS) and perfluorooctanoic acid (PFOA) in
human breast milk: results of a pilot study. Int J Hyg Environ Health 211: 440-446.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3103448
von Hoist, H; Nayak, P; Dembek, Z; Buehler, S; Echeverria, D; Fallacara, D; John, L. (2021).
Perfluoroalkyl substances exposure and immunity, allergic response, infection, and asthma in
children: review of epidemiologic studies [Review]. Heliyon 7: e08160.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9960586
vonderEmbse, AN; DeWitt, JC. (2018). Developmental immunotoxicity (DIT) testing: Current
recommendations and the future of DIT testing. In JC DeWitt; CE Rockwell; CC Bowman (Eds.),
Immunotoxicity testing: Methods and protocols (2nd ed., pp. 47-56). Totowa, NJ: Humana Press.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6741321
Vuong, A; Yolton, K; Webster, GM; Sjodin, A; Calafat, AM; Braun, JM; Dietrich, K; Lanphear, BP;
Chen, A. (2016). Prenatal polybrominated diphenyl ether and perfluoroalkyl substance exposures
and executive function in school-age children. Environ Res 147: 556-564.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3352166
Vuong, AM; Braun, JM; Yolton, K; Wang, Z; Xie, C; Webster, GM; Ye, X; Calafat, AM; Dietrich, KN;
Lanphear, BP; Chen, A. (2018). Prenatal and childhood exposure to perfluoroalkyl substances
(PFAS) and measures of attention, impulse control, and visual spatial abilities. Environ Int 119:
413-420. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079693
Vuong, AM; Xie, C; Jandarov, R; Dietrich, KN; Zhang, H; Sjodin, A; Calafat, AM; Lanphear, BP;
Mccandless, L; Braun, JM; Yolton, K; Chen, A. (2020). Prenatal exposure to a mixture of
persistent organic pollutants (POPs) and child reading skills at school age. Int J Hyg Environ
Health 228: 113527. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833684
Vuong, AM; Yolton, K; Braun, JM; Sjodin, A; Calafat, AM; Xu, Y; Dietrich, KN; Lanphear, BP; Chen,
A. (2020). Polybrominated diphenyl ether (PBDE) and poly- and perfluoroalkyl substance
(PFAS) exposures during pregnancy and maternal depression. Environ Int 139: 105694.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6356876
Vuong, AM; Yolton, K; Wang, Z; Xie, C; Webster, GM; Ye, X; Calafat, AM; Braun, JM; Dietrich, KN;
Lanphear, BP; Chen, A. (2018). Childhood perfluoroalkyl substance exposure and executive
function in children at 8 years. Environ Int 119: 212-219.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 079675
Vuong, AM; Yolton, K; Xie, C; Dietrich, KN; Braun, JM; Webster, GM; Calafat, AM; Lanphear, BP;
Chen, A. (2019). Prenatal and childhood exposure to poly- and perfluoroalkyl substances (PFAS)
and cognitive development in children at age 8 years. Environ Res 172: 242-248.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080218
WA DOH. (2020). DOH Approach to Developing PFAS State Action Levels.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9418278
Wambaugh, JF; Setzer, RW; Pitruzzello, AM; Liu, J; Reif, DM; Kleinstreuer, NC; Wang, NC; Sipes, N;
Martin, M; Das, K; Dewitt, JC; Strynar, M; Judson, R; Houck, KA; Lau, C. (2013). Dosimetric
anchoring of in vivo and in vitro studies for perfluorooctanoate and perfluorooctanesulfonate.
Toxicol Sci 136: 308-327.
7-78
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850932
Wan, C; Han, R; Liu, L; Zhang, F; Li, F; Xiang, M; Ding, W. (2016). Role of miR-155 in fluorooctane
sulfonate-induced oxidative hepatic damage via the Nrf2-dependent pathway. Toxicol Appl
Pharmacol 295: 85-93.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981504
Wang, B; Zhang, R; Jin, F; Lou, H; Mao, Y; Zhu, W; Zhou, W; Zhang, P; Zhang, J. (2017).
Perfluoroalkyl substances and endometriosis-related infertility in Chinese women. Environ Int
102: 207-212. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3856459
Wang, H; Du, H; Yang, J; Jiang, H; O, K; Xu, L; Liu, S; Yi, J; Qian, X; Chen, Y; Jiang, Q; He, G. (2019).
PFOS, PFOA, estrogen homeostasis, and birth size in Chinese infants. Chemosphere 221: 349-
355. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080598
Wang, H; Yang, J; Du, H; Xu, L; Liu, S; Yi, J; Qian, X; Chen, Y; Jiang, Q; He, G. (2018). Perfluoroalkyl
substances, glucose homeostasis, and gestational diabetes mellitus in Chinese pregnant women: A
repeat measurement-based prospective study. Environ Int 114: 12-20.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080352
Wang, IJ; Hsieh, WS; Chen, CY; Fletcher, T; Lien, GW; Chiang, HL; Chiang, CF; Wu, TN; Chen, PC.
(2011). The effect of prenatal perfluorinated chemicals exposures on pediatric atopy. Environ Res
111: 785-791. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1424977
Wang, J; Pan, Y; Cui, Q; Yao, B; Wang, J; Dai, J. (2018). Penetration of PFASs across the blood
cerebrospinal fluid barrier and its determinants in humans. Environ Sci Technol 52: 13553-13561.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080654
Wang, J; Zeng, XW; Bloom, MS; Qian, Z; Hinyard, LJ; Belue, R; Lin, S; Wang, SQ; Tian, YP; Yang, M;
Chu, C; Gurram, N; Hu, LW; Liu, KK; Yang, BY; Feng, D; Liu, RQ; Dong, GH. (2019). Renal
function and isomers of perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS):
Isomers of C8 Health Project in China. Chemosphere 218: 1042-1049.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080583
Wang, J; Zhang, Y; Zhang, W; Jin, Y; Dai, J. (2012). Association of perfluorooctanoic acid with HDL
cholesterol and circulating miR-26b and miR-199-3p in workers of a fluorochemical plant and
nearby residents. Environ Sci Technol 46: 9274-9281.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919184
Wang, L; Wang, Y; Liang, Y; Li, J; Liu, Y; Zhang, J; Zhang, A; Fu, J; Jiang, G. (2013). Specific
accumulation of lipid droplets in hepatocyte nuclei of PFOA-exposed BALB/c mice. Sci Rep 3:
2174. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850952
Wang, W; Zhou, W; Wu, S; Liang, F; Li, Y; Zhang, J; Cui, L; Feng, Y; Wang, Y. (2019). Perfluoroalkyl
substances exposure and risk of polycystic ovarian syndrome related infertility in Chinese
women. Environ Pollut247: 824-831.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080500
Wang, Y; Han, W; Wang, C; Zhou, Y; Shi, R; Bonefcld-Jorgenscn. EC; Yao, Q; Yuan, T; Gao, Y; Zhang,
J; Tian, Y. (2019). Efficiency of maternal-fetal transfer of perfluoroalkyl and polyfluoroalkyl
substances. Environ Sci Pollut Res Int 26: 2691-2698.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5083694
Wang, Y; Miao, Y; Mir, AZ; Cheng, L; Wang, L; Zhao, L; Cui, Q; Zhao, W; Wang, H. (2016). Inhibition
of beta-amyloid-induced neurotoxicity by pinocembrin through Nrf2/HO-l pathway in SH-SY5Y
cells. J Neurol Sci 368: 223-230.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3983465
Wang, Y; Rogan, WJ; Chen, HY; Chen, PC; Su, PH; Chen, HY; Wang, SL. (2015). Prenatal exposure to
perfluroalkyl substances and children's IQ: The Taiwan maternal and infant cohort study. Int J
Hyg Environ Health 218: 639-644.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3860120
Wang, Y; Rogan, WJ; Chen, PC; Lien, GW; Chen, HY; Tseng, YC; Longnecker, MP; Wang, SL. (2014).
Association between maternal serum perfluoroalkyl substances during pregnancy and maternal
7-79
-------
DRAFT FOR PUBLIC COMMENT
March 2023
and cord thyroid hormones: Taiwan maternal and infant cohort study. Environ Health Perspect
122: 529-534. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850394
Wang, Y; Wang, L; Li, J; Liang, Y; Ji, H; Zhang, J; Zhou, Q; Jiang, G. (2014). The mechanism of
immunosuppression by perfluorooctanoic acid in BALB/c mice. Toxicology Research 3: 205-
213. https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/3860153
Wang, Y; Zhang, C. (2019). The Roles of Liver-Resident Lymphocytes in Liver Diseases [Review]. Front
Immunol 10: 1582. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365737
Wang, Y; Zhang, L; Teng, Y; Zhang, J; Yang, L; Li, J; Lai, J; Zhao, Y; Wu, Y. (2018). Association of
serum levels of perfluoroalkyl substances with gestational diabetes mellitus and postpartum blood
glucose. J Environ Sci 69: 5-11.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079666
Wang, Z; Shi, R; Ding, G; Yao, Q; Pan, C; Gao, Y; Tian, Y. (2022). Association between maternal serum
concentration of perfluoroalkyl substances (PFASs) at delivery and acute infectious diseases in
infancy. Chemosphere 289: 133235.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/10176501
Wang, Z; Zhang, T; Wu, J; Wei, X; Xu, A; Wang, S; Wang, Z. (2021). Male reproductive toxicity of
perfluorooctanoate (PFOA): Rodent studies [Review]. Chemosphere 270: 128608.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7152781
Warembourg, C; Maitre, L, ea; Tamayo-Uria, I; Fossati, S; Roumeliotaki, T; Aasvang, GM; Andrusaityte,
S; Casas, M; Cequier, E; Chatzi, L; Dedele, A; Gonzalez, J. R.; Grazuleviciene, R; Haug, LS;
Hernandez-Ferrer, C; Heude, B; Karachaliou, M; Krog, NH; Mceachan, R; Nieuwenhuijsen, M;
Petraviciene, I; Quentin, J; Robinson, O; Sakhi, AK; Slama, R; Thomsen, C; Urquiza, J; Vafeiadi,
M; West, J; Wright, J; Vrijheid, M; Basagana, X. (2019). Early-life environmental exposures and
blood pressure in children. J Am Coll Cardiol 74: 1317-1328.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5881345
Watkins, DJ; Josson, J; Elston, B; Bartell, SM; Shin, HM; Vieira, VM; Savitz, DA; Fletcher, T;
Wellenius, GA. (2013). Exposure to perfluoroalkyl acids and markers of kidney function among
children and adolescents living near a chemical plant. Environ Health Perspect 121: 625-630.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850974
Weaver, YM; Ehresman, DJ; Butenhoff, JL; Hagenbuch, B. (2010). Roles of rat renal organic anion
transporters in transporting perfluorinated carboxylates with different chain lengths. Toxicol Sci
113: 305-314. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2010072
Webster, GM; Venners, SA; Mattman, A; Martin, JW. (2014). Associations between perfluoroalkyl acids
(PFASs) and maternal thyroid hormones in early pregnancy: a population-based cohort study.
Environ Res 133: 338-347.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850208
Weingand, K; Bloom, J; Carakostas, M; Hall, R; Helfrich, M; Latimer, K; Levine, B; Neptun, D; Rebar,
A; Stitzel, K; Troup, C. (1992). Clinical pathology testing recommendations for nonclinical
toxicity and safety studies. Toxicol Pathol 20: 539-543.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/670731
Weiss, JM; Andersson, PL; Lamoree, MH; Leonards, PEG; van Leeuwen, SPJ; Hamers, T. (2009).
Competitive Binding of Poly- and Perfluorinated Compounds to the Thyroid Hormone Transport
Protein Transthyretin. Toxicol Sci 109: 206-216.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/534503
Weisskopf, MG; Seals, RM; Webster, TF. (2018). Bias amplification in epidemiologic analysis of
exposure to mixtures. Environ Health Perspect 126.
https://hero.epa.gov/hero/index.cfim/reference/details/reference_id/7325521
Wen, HJ; Wang, SL; Chen, PC; Guo, YL. (2019). Prenatal perfluorooctanoic acid exposure and
glutathione s-transferase Tl/Ml genotypes and their association with atopic dermatitis at 2 years
of age. PLoS ONE 14: e0210708.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081172
7-80
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Wen, HJ; Wang, SL; Chuang, YC; Chen, PC; Guo, YL. (2019). Prenatal perfluorooctanoic acid exposure
is associated with early onset atopic dermatitis in 5-year-old children. Chemosphere 231: 25-31.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387152
Wen, LL; Lin, LY; Su, TC; Chen, PC; Lin, CY. (2013). Association between serum perfluorinated
chemicals and thyroid function in U.S. adults: the National Health and Nutrition Examination
Survey 2007-2010. J Clin Endocrinol Metab 98: E1456-E1464.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850943
Wen, X; Baker, AA; Klaassen, CD; Corton, JC; Richardson, JR; Aleksunes, LM. (2019). Hepatic
carboxylesterases are differentially regulated in PPARa-null mice treated with perfluorooctanoic
acid. Toxicology 416: 15-22.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080582
Wen, X; Wang, M; Xu, X; Li, T. (2022). Exposure to Per- and Polyfluoroalkyl Substances and Mortality
in U.S. Adults: A Population-Based Cohort Study. Environ Health Perspect 130: 67007.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10328873
Wen, Y; Miiji, N; Irudayaraj, J. (2020). Epigenetic toxicity of PFOA and GenX in HepG2 cells and their
roles in lipid metabolism. Toxicol In Vitro 65: 104797.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6302274
Weng, J; Hong, C; Tasi, J; Shen, CY, u; Su, P; Wang, S. (2020). The association between prenatal
endocrine-disrupting chemical exposure and altered resting-state brain fMRI in teenagers. Brain
Struct Funct 225: 1669-1684.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6718530
White, SS; Kato, K; Jia, LT; Basden, BJ; Calafat, AM; Hines, EP; Stanko, JP; Wolf, CJ; Abbott, BD;
Fenton, SE. (2009). Effects of perfluorooctanoic acid on mouse mammary gland development
and differentiation resulting from cross-foster and restricted gestational exposures. Reprod
Toxicol 27: 289-298. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/194811
White, SS; Stanko, JP; Kato, K; Calafat, AM; Hines, EP; Fenton, SE. (2011). Gestational and chronic
low-dose PFOA exposures and mammary gland growth and differentiation in three generations of
CD-I mice. Environ Health Perspect 119: 1070-1076.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/127615 0
Whitehead, HD; Venier, M; Wu, Y; Eastman, E; U, r, S.; Diamond, ML; al., e. (2021). Fluorinated
Compounds in North American Cosmetics. Environ Sci Technol Lett 8: 538-544.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9416542
WHO. (2012). Guidance for immunotoxicity risk assessment for chemicals. (Harmonization Project
Document No. 10). Geneva, Switzerland.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9522548
WHO. (2012). Guidance for immunotoxicity risk assessment for chemicals. (Harmonization Project
Document No. 10). Geneva, Switzerland.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/10633091
WHO. (2017). Diphtheria Vaccine: Review of Evidence on Vaccine Effectiveness and Immunogenicity to
Assess the Duration of Protection >10 Years After the Last Booster Dose.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642150
WHO. (2017). Tetanus vaccines: WHO position paper - February 2017. 92: 53-76.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642138
Wielsoe. M; Kern, P; Bonefcld-Jorgenscn. EC. (2017). Serum levels of environmental pollutants is a risk
factor for breast cancer in Inuit: a case control study. Environ Health 16: 56.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858479
Wielsoe, M; Long, M; Ghisari, M; Bonefcld-Jorgenscn. EC. (2015). Perfluoroalkylated substances
(PFAS) affect oxidative stress biomarkers in vitro. Chemosphere 129: 239-245.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/25 3 3367
Wiener, RC; Waters, C. (2019). Perfluoroalkyls/polyfluoroalkyl substances and dental caries experience
in children, ages 3-11 years, National Health and Nutrition Examination Survey, 2013-2014. J
7-81
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Public Health Dent 79: 307-319.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5386081
Wikstrom, S; Lin, PI; Lindh, CH; Shu, H; Bornehag, CG. (2019). Maternal serum levels of perfluoroalkyl
substances in early pregnancy and offspring birth weight. Pediatr Res 87: 1093-1099.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6311677
Wikstrom, S; Lindh, CH; Shu, H; Bornehag, CG. (2019). Early pregnancy serum levels of perfluoroalkyl
substances and risk of preeclampsia in Swedish women. Sci Rep 9: 9179.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387145
Winquist, A; Steenland, K. (2014). Modeled PFOA exposure and coronary artery disease, hypertension,
and high cholesterol in community and worker cohorts. Environ Health Perspect 122: 1299-1305.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851142
Winquist, A; Steenland, K. (2014). Perfluorooctanoic acid exposure and thyroid disease in community
and worker cohorts. Epidemiology 25: 255-264.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/23 3 7818
Wolf, CJ; Fenton, SE; Schmid, JE; Calafat, AM; Kuklenyik, Z; Bryant, XA; Thibodeaux, J; Das, KP;
White, SS; Lau, CS; Abbott, BD. (2007). Developmental toxicity of perfluorooctanoic acid in the
CD-I mouse after cross-foster and restricted gestational exposures. Toxicol Sci 95: 462-473.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332672
Wolf, CJ; Rider, CV; Lau, C; Abbott, BD. (2014). Evaluating the additivity of perfluoroalkyl acids in
binary combinations on peroxisome proliferator-activated receptor-a activation. Toxicology 316:
43-54. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850908
Wolf, DC; Moore, T; Abbott, BD; Rosen, MB; Das, KP; Zehr, RD; Lindstrom, AB; Strynar, MJ; Lau, C.
(2008). Comparative hepatic effects of perfluorooctanoic acid and WY 14,643 in PPAR-alpha
knockout and wild-type mice. Toxicol Pathol 36: 632-639.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1290827
Woodcraft, MW; Ellis, DA; Rafferty, SP; Burns, DC; March, RE; Stock, NL; Trumpour, KS; Yee, J;
Munro, K. (2010). Experimental characterization of the mechanism of perfluorocarboxylic acids'
liver protein bioaccumulation: the key role of the neutral species. Environ Toxicol Chem 29:
1669-1677. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919284
Workman, CE; Becker, AB; Azad, MB; Moraes, TJ; Mandhane, PJ; Turvey, SE; Subbarao, P; Brook, JR;
Sears, MR; Wong, CS. (2019). Associations between concentrations of perfluoroalkyl substances
in human plasma and maternal, infant, and home characteristics in Winnipeg, Canada. Environ
Pollut 249: 758-766. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387046
Worley, RR; Moore, SM; Tierney, BC; Ye, X; Calafat, AM; Campbell, S; Woudneh, MB; Fisher, J.
(2017). Per- and polyfluoroalkyl substances in human serum and urine samples from a
residentially exposed community. Environ Int 106: 135-143.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859800
Worley, RR; Yang, X; Fisher, J. (2017). Physiologically based pharmacokinetic modeling of human
exposure to perfluorooctanoic acid suggests historical non drinking-water exposures are
important for predicting current serum concentrations. Toxicol Appl Pharmacol 330: 9-21.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981311
Wu, H; Yoon, M; Verner, MA; Xue, J; Luo, M, an; Andersen, ME; Longnecker, MP; Clewell, HJ, III.
(2015). Can the observed association between serum perfluoroalkyl substances and delayed
menarche be explained on the basis of puberty-related changes in physiology and
pharmacokinetics? Environ Int 82: 61-68.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3223290
Wu, K; Xu, X; Peng, L; Liu, J; Guo, Y; Huo, X. (2012). Association between maternal exposure to
perfluorooctanoic acid (PFOA) from electronic waste recycling and neonatal health outcomes.
Environ Int 48: 1-8. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919186
Wu, LL; Gao, HW; Gao, NY; Chen, FF; Chen, L. (2009). Interaction of perfluorooctanoic acid with
human serum albumin. BMC Struct Biol 9: 31.
7-82
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/5 36376
Wu, X; Xie, G; Xu, X; Wu, W; Yang, B. (2018). Adverse bioeffect of perfluorooctanoic acid on liver
metabolic function in mice. Environ Sci Pollut Res Int 25: 4787-4793.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238318
Wu, XM; Bennett, DH; Calafat, AM; Kato, K; Strynar, M; Andersen, E; Moran, RE; Tancredi, DJ; Tulve,
NS; Hertz-Picciotto, I. (2014). Serum concentrations of perfluorinated compounds (PFC) among
selected populations of children and Adults in California. Environ Res 136C: 264-273.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2533322
Xiao, C; Grandjean, P; Valvi, D; Nielsen, F; Jensen, TK; Weihe, P; Oulhote, Y. (2019). Associations of
exposure to perfluoroalkyl substances with thyroid hormone concentrations and birth size. J Clin
Endocrinol Metab 105: 735-745.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/5 918609
Xu, H; Zhou, Q; Zhang, J; Chen, X; Zhao, H; Lu, H; Ma, B; Wang, Z; Wu, C; Ying, C; Xiong, Y; Zhou,
Z; Li, X. (2020). Exposure to elevated per- and polyfluoroalkyl substances in early pregnancy is
related to increased risk of gestational diabetes mellitus: A nested case-control study in Shanghai,
China. Environ Int 143: 105952.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833677
Xu, J; Nave, R; Lahu, G; Derom, E; Derendorf, H. (2010). Population pharmacokinetics and
pharmacodynamics of inhaled ciclesonide and fluticasone propionate in patients with persistent
asthma. J Clin Pharmacol 50: 1118-1127.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1295081
Xu, L; Liu, W; Bai, F; Xu, Y; Liang, X; Ma, C; Gao, L. (2021). Hepatic Macrophage as a Key Player in
Fatty Liver Disease [Review]. Front Immunol 12: 708978.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365843
Xu, M; Liu, G; Li, M; Huo, M; Zong, W; Liu, R. (2020). Probing the Cell Apoptosis Pathway Induced by
Perfluorooctanoic Acid and Perfluorooctane Sulfonate at the Subcellular and Molecular Levels. J
Agric Food Chem 68: 633-641.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316207
Xu, M; Wan, J; Niu, Q; Liu, R. (2019). PFOA and PFOS interact with superoxide dismutase and induce
cytotoxicity in mouse primary hepatocytes: A combined cellular and molecular methods. Environ
Res 175: 63-70. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5381556
Xu, Y; Fletcher, T; Pineda, D; Lindh, CH; Nilsson, C; Glynn, A; Vogs, C; Norstrom, K; Lilja, K;
Jakobsson, K; Li, Y. (2020). Serum Half-Lives for Short- and Long-Chain Perfluoroalkyl Acids
after Ceasing Exposure from Drinking Water Contaminated by Firefighting Foam. Environ
Health Perspect 128: 77004.
https://hero .epa.gov/hero/index.cfm/reference/details/reference_id/6781357
Xu, Y; Li, Y; Scott, K; Lindh, CH; Jakobsson, K; Fletcher, T; Ohlsson, B; Andersson, EM. (2020).
Inflammatory bowel disease and biomarkers of gut inflammation and permeability in a
community with high exposure to perfluoroalkyl substances through drinking water. Environ Res
181: 108923. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315709
Yahia, D; El-Nasser, MA; Abedel-Latif, M; Tsukuba, C; Yoshida, M; Sato, I; Tsuda, S. (2010). Effects of
perfluorooctanoic acid (PFOA) exposure to pregnant mice on reproduction. J Toxicol Sci 35:
527-533. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1332451
Yahia, D; Haruka, I; Kagashi, Y; Tsuda, S. (2016). 8-Hydroxy-2'-deoxyguanosine as a biomarker of
oxidative DNA damage induced by perfluorinated compounds in TK6 cells. Environ Toxicol 31:
192-200. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851192
Yamaguchi, M; Arisawa, K; Uemura, H; Katsuura-Kamano, S; Takami, H; Sawachika, F; Nakamoto, M;
Juta, T; Toda, E; Mori, K; Hasegawa, M; Tanto, M; Shima, M; Sumiyoshi, Y; Morinaga, K;
Kodama, K; Suzuki, T; Nagai, M; Satoh, H. (2013). Consumption of seafood, serum liver
enzymes, and blood levels of PFOS and PFOA in the Japanese population. J Occup Health 55:
184-194. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850970
7-83
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Yan, S; Wang, J; Dai, J. (2015). Activation of sterol regulatory element-binding proteins in mice exposed
to perfluorooctanoic acid for 28 days. Arch Toxicol 89: 1569-1578.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851199
Yan, S; Wang, J; Zhang, W; Dai, J. (2014). Circulating microRNA profiles altered in mice after 28 d
exposure to perfluorooctanoic acid. Toxicol Lett 224: 24-31.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850901
Yan, S; Zhang, H; Guo, X; Wang, J; Dai, J. (2017). High perfluorooctanoic acid exposure induces
autophagy blockage and disturbs intracellular vesicle fusion in the liver. Arch Toxicol 91: 247-
258. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3981501
Yan, S; Zhang, H; Wang, J; Zheng, F; Dai, J. (2015). Perfluorooctanoic acid exposure induces
endoplasmic reticulum stress in the liver and its effects are ameliorated by 4-phenylbutyrate. Free
Radic Biol Med 87: 300-311.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3981567
Yang, B; Zou, W; Hu, Z; Liu, F; Zhou, L; Yang, S; Kuang, H; Wu, L; Wei, J; Wang, J; Zou, T; Zhang, D.
(2014). Involvement of oxidative stress and inflammation in liver injury caused by
perfluorooctanoic acid exposure in mice. BioMed Res Int 2014: 409837.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2850321
Yang, CH; Glover, KP; Han, X. (2009). Organic anion transporting polypeptide (Oatp) lal-mediated
perfluorooctanoate transport and evidence for a renal reabsorption mechanism of Oatplal in renal
elimination of perfluorocarboxylates in rats. Toxicol Lett 190: 163-171.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2919328
Yang, CH; Glover, KP; Han, X. (2010). Characterization of cellular uptake of perfluorooctanoate via
organic anion-transporting polypeptide 1A2, organic anion transporter 4, and urate transporter 1
for their potential roles in mediating human renal reabsorption of perfluorocarboxylates. Toxicol
Sci 117: 294-302. https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/2919288
Yang, D; Han, J; Hall, DR; Sun, J; Fu, J; Kutarna, S; Houck, KA; Lalone, CA; Doering, JA; Ng, CA;
Peng, H. (2020). Nontarget Screening of Per- and Polyfluoroalkyl Substances Binding to Human
Liver Fatty Acid Binding Protein. Environ Sci Technol 54: 5676-5686.
https://hero .epa.gov/hero/index.cfin/reference/details/reference_id/63 5 63 70
Yang, J; Wang, H; Du, H; Fang, H; Han, M; Xu, L; Liu, S; Yi, J; Chen, Y; Jiang, Q; He, G. (2020).
Serum perfluoroalkyl substances in relation to lipid metabolism in Chinese pregnant women.
Chemosphere 273: 128566.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/7021246
Yang, L; Li, J; Lai, J; Luan, H; Cai, Z; Wang, Y; Zhao, Y; Wu, Y. (2016). Placental transfer of
perfluoroalkyl substances and associations with thyroid hormones: Beijing prenatal exposure
study. Sci Rep 6: 21699.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/3858535
Yang, Q; Abedi-Valugerdi, M; Xie, Y; Zhao, XY; Moller, G; Nelson, BD; Depierre, JW. (2002). Potent
suppression of the adaptive immune response in mice upon dietary exposure to the potent
peroxisome proliferator, perfluorooctanoic acid. Int Immunopharmacol 2: 389-397.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1332454
Yang, Q; Guo, X; Sun, P; Chen, Y; Zhang, W; Gao, A. (2018). Association of serum levels of
perfluoroalkyl substances (PFASs) with the metabolic syndrome (MetS) in Chinese male adults:
A cross-sectional study. Sci Total Environ 621: 1542-1549.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/4238462
Yang, Q; Xie, Y; Alexson, SE; Nelson, BD; Depierre, JW. (2002). Involvement of the peroxisome
proliferator-activated receptor alpha in the immunomodulation caused by peroxisome
proliferators in mice. Biochem Pharmacol 63: 1893-1900.
https://hero.epa.gov/hero/index.cfin/reference/details/reference_id/1332453
Yang, Q; Xie, Y; Depierre, JW. (2000). Effects of peroxisome proliferators on the thymus and spleen of
mice. Clin Exp Immunol 122: 219-226.
7-84
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/699394
Yang, Q; Xie, Y; Eriksson, AM; Nelson, BD; Depierre, JW. (2001). Further evidence for the involvement
of inhibition of cell proliferation and development in thymic and splenic atrophy induced by the
peroxisome proliferator perfluoroctanoic acid in mice. Biochem Pharmacol 62: 1133-1140.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1014748
Yang, Y; Lv, QY; Guo, LH; Wan, B; Ren, XM; Shi, YL; Cai, YQ. (2017). Identification of protein
tyrosine phosphatase SHP-2 as a new target of perfluoroalkyl acids in HepG2 cells. Arch Toxicol
91: 1697-1707. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981427
Yang, Z; Liu, HY; Yang, QY; Chen, X; Li, W; Leng, J; Tang, NJ. (2022). Associations between exposure
to perfluoroalkyl substances and birth outcomes: A meta-analysis. Chemosphere 291: 132909.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10176603
Yao, Q; Gao, Y; Zhang, Y; Qin, K; Liew, Z; Tian, Y. (2021). Associations of paternal and maternal per-
and polyfluoroalkyl substances exposure with cord serum reproductive hormones, placental
steroidogenic enzyme and birth weight. Chemosphere 285: 131521.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9960202
Yao, Q; Shi, R; Wang, C; Han, W; Gao, Y; Zhang, Y; Zhou, Y; Ding, G; Tian, Y. (2019). Cord blood
per- and polyfluoroalkyl substances, placental steroidogenic enzyme, and cord blood reproductive
hormone. Environ Int 129: 573-582.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5187556
Yao, X; Zhong, L. (2005). Genotoxic risk and oxidative DNA damage in HepG2 cells exposed to
perfluorooctanoic acid. Mutat Res 587: 38-44.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081563
Yarahalli Jayaram, V; Baggavalli, S; Reddy, D; Sistla, S; Malempati, R. (2018). Effect of endosulfan and
bisphenol A on the expression of SUMO and UBC9. Drug Chem Toxicol 43: 1-8.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080662
Ye, WL; Chen, ZX; Xie, YQ; Kong, ML; Li, QQ; Yu, S; Chu, C; Dong, GH; Zeng, XW. (2021).
Associations between serum isomers of perfluoroalkyl acids and metabolic syndrome in adults:
Isomers of C8 Health Project in China. Environ Res 196: 110430.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6988486
Ylinen, M; Kojo, A; Hanhijarvi, H; Peura, P. (1990). Disposition of perfluorooctanoic acid in the rat after
single and subchronic administration. Bull Environ Contam Toxicol 44: 46-53.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5085631
York, RG; Kennedy, GL; Olsen, GW; Butenhoff, JL. (2010). Male reproductive system parameters in a
two-generation reproduction study of ammonium perfluorooctanoate in rats and human relevance.
Toxicology 271: 64-72.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2919279
Young, W; Wiggins, S; Limm, W; Fisher, CM; Dejager, L; Genualdi, S. (2022). Analysis of Per- and
Poly(fluoroalkyl) Substances (PFASs) in Highly Consumed Seafood Products from U.S. Markets.
J Agric Food Chem 70: 13545-13553.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10601281
Yu, N; Wei, S; Li, M; Yang, J; Li, K; Jin, L; Xie, Y; Giesy, JP; Zhang, X; Yu, H. (2016). Effects of
Perfluorooctanoic Acid on Metabolic Profiles in Brain and Liver of Mouse Revealed by a High-
throughput Targeted Metabolomics Approach. Sci Rep 6: 23963.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981487
Yu, S; Feng, WR; Liang, ZM; Zeng, XY; Bloom, MS; Hu, GC; Zhou, Y; Ou, YQ; Chu, C; Li, QQ; Yu,
Y; Zeng, XW; Dong, GH. (2021). Perfluorooctane sulfonate alternatives and metabolic syndrome
in adults: New evidence from the Isomers of C8 Health Project in China. Environ Pollut 283:
117078. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8453076
Yuan, G; Peng, H; Huang, C; Hu, J. (2016). Ubiquitous occurrence of fluorotelomer alcohols in eco-
friendly paper-made food-contact materials and their implication for human exposure. Environ
Sci Technol 50: 942-950.
7-85
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859226
Yue, Y; Sun, Y; Yan, X; Liu, J; Zhao, S; Zhang, J. (2016). Evaluation of the binding of perfluorinated
compound to pepsin: Spectroscopic analysis and molecular docking. Chemosphere 161: 475-481.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3479514
Yusoff, AF; Mohd Sharani, ZZ; Kee, CC; Md Iderus, NH; Md Zamri, ASS; Nagalingam, T; Mohamad
Bashaabidin, MS; Wan Ibadullah, WAH; Ghazali, SM; Yusof, AY; Ching, YM; Mohamed Nor,
N; Kamarudin, B; Ahmad, N; Arip, M. (2021). Seroprevalence of diphtheria toxoid IgG
antibodies in the Malaysian population. BMC Infect Dis 21: 581.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642157
Zabaleta, I; Blanco-Zubiaguirre, L; Baharli, EN; Olivares, M; Prieto, A; Zuloaga, 0; Elizalde, MP.
(2020). Occurrence of per- and polyfluorinated compounds in paper and board packaging
materials and migration to food simulants and foodstuffs. Food Chem 321: 126746.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505866
Zabaleta, I; Negreira, N; Bizkarguenaga, E; Prieto, A; Covaci, A; Zuloaga, 0. (2017). Screening and
identification of per- and polyfluoroalkyl substances in microwave popcorn bags. Food Chem
230: 497-506. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981827
Zafeiraki, E; Gebbink, WA; Hoogenboom, R; Kotterman, M; Kwadijk, C; Dassenakis, E; van Leeuwen,
SPJ. (2019). Occurrence of perfluoroalkyl substances (PFASs) in a large number of wild and
farmed aquatic animals collected in the Netherlands. Chemosphere 232: 415-423.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5387058
Zai'r, ZM; Eloranta, JJ; Stieger, B; Kullak-Ublick, GA. (2008). Pharmacogenetics of OATP
(SLC21/SLCO), OAT and OCT (SLC22) and PEPT (SLC15) transporters in the intestine, liver
and kidney. Pharmacogenomics 9: 597-624.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9641805
Zare Jeddi, M; Dalla Zuanna, T; Barbieri, G; Fabricio, ASC; Dapra, F; Fletcher, T; Russo, F; Pitter, G;
Canova, C. (2021). Associations of Perfluoroalkyl Substances with Prevalence of Metabolic
Syndrome in Highly Exposed Young Adult Community Residents-A Cross-Sectional Study in
Veneto Region, Italy. Int J Environ Res Public Health 18: 1194.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/7404065
Zare Jeddi, M; Soltanmohammadi, R; Barbieri, G; Fabricio, ASC; Pitter, G; Dalla Zuanna, T; Canova, C.
(2021). To which extent are per-and poly-fluorinated substances associated to metabolic
syndrome? [Review]. Rev Environ Health.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/8347183
Zareitalabad, P; Siemens, J; Hamer, M; Amelung, W. (2013). Perfluorooctanoic acid (PFOA) and
perfluorooctanesulfonic acid (PFOS) in surface waters, sediments, soils and wastewater - A
review on concentrations and distribution coefficients [Review]. Chemosphere 91: 725-732.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080561
Zasada, AA; Rastawicki, W; Rokosz, N; Jagielski, M. (2013). Seroprevalence of diphtheria toxoid IgG
antibodies in children, adolescents and adults in Poland. BMC Infect Dis 13: 551.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3194760
Zeeshan, M; Yang, Y; Zhou, Y; Huang, W; Wang, Z; Zeng, XY; Liu, RQ; Yang, BY; Hu, LW; Zeng,
XW; Sun, X; Yu, Y; Dong, GH. (2020). Incidence of ocular conditions associated with
perfluoroalkyl substances exposure: Isomers of C8 Health Project in China. Environ Int 137:
105555. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315698
Zenewicz, LA; Yancopoulos, GD; Valenzuela, DM; Murphy, AJ; Karow, M; Flavell, RA. (2007).
Interleukin-22 but not interleukin-17 provides protection to hepatocytes during acute liver
inflammation. Immunity 27: 647-659.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10365732
Zeng, X; Chen, Q; Zhang, X; Li, H; Liu, Q; Li, C; Ma, M; Zhang, J; Zhang, W; Zhang, J; Huang, L.
(2019). Association between prenatal exposure to perfluoroalkyl substances and asthma-related
diseases in preschool children. Environ Sci Pollut Res Int 26: 29639-29648.
7-86
-------
DRAFT FOR PUBLIC COMMENT
March 2023
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5412431
Zeng, XW; Bloom, MS; Dharmage, SC; Lodge, CJ; Chen, D; Li, S; Guo, Y; Roponen, M; Jalava, P;
Hirvonen, MR; Ma, H; Hao, YT; Chen, W; Yang, M; Chu, C; Li, QQ; Hu, LW; Liu, KK; Yang,
BY; Liu, S; Fu, C; Dong, GH. (2019). Prenatal exposure to perfluoroalkyl substances is
associated with lower hand, foot and mouth disease viruses antibody response in infancy:
Findings from the Guangzhou Birth Cohort Study. Sci Total Environ 663: 60-67.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081554
Zeng, XW; Li, QQ; Chu, C; Ye, WL; Yu, S; Ma, H; Zeng, XY; Zhou, Y; Yu, HY; Hu, LW; Yang, BY;
Dong, GH. (2020). Alternatives of perfluoroalkyl acids and hepatitis B virus surface antibody in
adults: Isomers of C8 Health Project in China. Environ Pollut 259: 113857.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315718
Zeng, XW; Lodge, CJ; Dharmage, SC; Bloom, MS; Yu, Y; Yang, M; Chu, C; Li, QQ; Hu, LW; Liu, KK;
Yang, BY; Dong, GH. (2019). Isomers of per- and polyfluoroalkyl substances and uric acid in
adults: Isomers of C8 Health Project in China. Environ Int 133: 105160.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5918630
Zeng, XW; Qian, Z; Emo, B; Vaughn, M; Bao, J; Qin, XD; Zhu, Y; Li, J; Lee, YL; Dong, GH. (2015).
Association of polyfluoroalkyl chemical exposure with serum lipids in children. Sci Total
Environ 512-513: 364-370.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851005
Zhang, C; Sundaram, R; Maisog, J; Calafat, AM; Barr, DB; Buck Louis, GM. (2015). A prospective
study of prepregnancy serum concentrations of perfluorochemicals and the risk of gestational
diabetes. Fertil Steril 103: 184-189.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2857764
Zhang, H; Cui, R; Guo, X; Hu, J; Dai, J. (2016). Low dose perfluorooctanoate exposure promotes cell
proliferation in a human non-tumor liver cell line. J Hazard Mater 313: 18-28.
https ://hero .epa.gov/hero/index. cfm/reference/details/reference_id/3 748 826
Zhang, H; Fang, W; Wang, D; Gao, N; Ding, Y; Chen, C. (2014). The role of interleukin family in
perfluorooctanoic acid (PFOA)-induced immunotoxicity. J Hazard Mater 280: 552-560.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851150
Zhang, H; Lu, Y; Luo, B; Yan, S; Guo, X; Dai, J. (2014). Proteomic analysis of mouse testis reveals
perfluorooctanoic acid-induced reproductive dysfunction via direct disturbance of testicular
steroidogenic machinery. J Proteome Res 13: 3370-3385.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850230
Zhang, H; Yolton, K; Webster, GM; Ye, X; Calafat, AM; Dietrich, KN; Xu, Y; Xie, C; Braun, JM;
Lanphear, BP; Chen, A. (2018). Prenatal and childhood perfluoroalkyl substances exposures and
children's reading skills at ages 5 and 8 years. Environ Int 111: 224-231.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238294
Zhang, L; Ren, XM; Guo, LH. (2013). Structure-based investigation on the interaction of perfluorinated
compounds with human liver fatty acid binding protein. Environ Sci Technol 47: 11293-11301.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5081488
Zhang, R; Zhang, H; Chen, B; Luan, T. (2020). Fetal bovine serum attenuating perfluorooctanoic acid-
inducing toxicity to multiple human cell lines via albumin binding. J Hazard Mater 389: 122109.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6316915
Zhang, S; Kang, Q; Peng, H; Ding, M; Zhao, F; Zhou, Y; Dong, Z; Zhang, H; Yang, M; Tao, S; Hu, J.
(2019). Relationship between perfluorooctanoate and perfluorooctane sulfonate blood
concentrations in the general population and routine drinking water exposure. Environ Int 126:
54-60. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5080526
Zhang, S; Tan, R; Pan, R; Xiong, J; Tian, Y; Wu, J; Chen, L. (2018). Association of perfluoroalkyl and
polyfluoroalkyl substances with premature ovarian insufficiency in Chinese women. J Clin
Endocrinol Metab 103: 2543-2551.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5079665
7-87
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Zhang, T; Qin, X. (2014). Assessment of fetal exposure and maternal elimination of perfluoroalkyl
substances. Environ Sci Process Impacts 16: 1878-1881.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2850251
Zhang, T; Sun, H; Lin, Y; Qin, X; Zhang, Y; Geng, X; Kannan, K. (2013). Distribution of poly- and
perfluoroalkyl substances in matched samples from pregnant women and carbon chain length
related maternal transfer. Environ Sci Technol 47: 7974-7981.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859792
Zhang, T; Sun, H; Qin, X; Gan, Z; Kannan, K. (2015). PFOS and PFOA in paired urine and blood from
general adults and pregnant women: assessment of urinary elimination. Environ Sci Pollut Res Int
22: 5572-5579. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2851103
Zhang, X; Lohmann, R; Dassuncao, C; Hu, XC; Weber, AK; Vecitis, CD; Sunderland, EM. (2016).
Source attribution of poly- and perfluoroalkyl substances (PFASs) in surface waters from Rhode
Island and the New York metropolitan area. Environ Sci Technol Lett 3: 316-321.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3470830
Zhang, Y; Beesoon, S; Zhu, L; Martin, JW. (2013). Biomonitoring of perfluoroalkyl acids in human urine
and estimates of biological half-life. Environ Sci Technol 47: 10619-10627.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859849
Zhang, Y; Beesoon, S; Zhu, L; Martin, JW. (2013). Isomers of perfluorooctanesulfonate and
perfluorooctanoate and total perfluoroalkyl acids in human serum from two cities in North China.
Environ Int 53: 9-17. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/2639569
Zhang, Y; Cao, X; Chen, L; Qin, Y; Xu, Y; Tian, Y; Chen, L. (2020). Exposure of female mice to
perfluorooctanoic acid suppresses hypothalamic kisspeptin-reproductive endocrine system
through enhanced hepatic fibroblast growth factor 21 synthesis, leading to ovulation failure and
prolonged dioestrus. J Neuroendocrinol 32: el2848.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6505878
Zhang, Y; Le, Y; Bu, P; Cheng, X. (2020). Regulation of Hox and ParaHox genes by perfluorochemicals
in mouse liver. Toxicology 441: 152521.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6833704
Zhang, Y; Mustieles, V; Sun, Y; Oulhote, Y; Wang, YX; Messerlian, C. (2022). Association between
serum per- and polyfluoroalkyl substances concentrations and common cold among children and
adolescents in the United States. Environ Int 164: 107239.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/10410662
Zhang, Y; Pan, C; Ren, Y; Wang, Z; Luo, J; Ding, G; Vinturache, A; Wang, X; Shi, R; Ouyang, F;
Zhang, J; Li, J; Gao, Y; Tian, Y. (2022). Association of maternal exposure to perfluoroalkyl and
polyfluroalkyl substances with infant growth from birth to 12 months: A prospective cohort
study. Sci Total Environ 806: 151303.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9944433
Zhang, YM; Dong, XY; Fan, LJ; Zhang, ZL; Wang, Q; Jiang, N; Yang, XS. (2017). Poly- and
perfluorinated compounds activate human pregnane X receptor. Toxicology 380: 23-29.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3604013
Zhang, YM; Wang, T; Yang, XS. (2020). An in vitro and in silico investigation of human pregnane X
receptor agonistic activity of poly- and perfluorinated compounds using the heuristic method-best
subset and comparative similarity indices analysis. Chemosphere 240: 124789.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6324307
Zhao, G; Wang, J; Wang, X; Chen, S; Zhao, Y; Gu, F; Xu, A; Wu, L. (2011). Mutagenicity of PFOA in
mammalian cells: role of mitochondria-dependent reactive oxygen species. Environ Sci Technol
45: 1638-1644. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/847496
Zhao, L; Zhang, Y; Zhu, L; Ma, X; Wang, Y; Sun, H; Luo, Y. (2017). Isomer-Specific Transplacental
Efficiencies of Perfluoroalkyl Substances in Human Whole Blood. Environ Sci Technol Lett 4:
391-398. https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5085130
Zhao, W; Zitzow, JD; Weaver, Y; Ehresman, DJ; Chang, SC; Butenhoff, JL; Hagenbuch, B. (2017).
7-88
-------
DRAFT FOR PUBLIC COMMENT
March 2023
Organic anion transporting polypeptides contribute to the disposition of perfluoroalkyl acids in
humans and rats. Toxicol Sci 156: 84-95.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3856461
Zheng, F; Sheng, N; Zhang, H; Yan, S; Zhang, J; Wang, J. (2017). Perfluorooctanoic acid exposure
disturbs glucose metabolism in mouse liver. Toxicol Appl Pharmacol 335: 41-48.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4238507
Zhong, Y; Shen, L; Ye, X; Zhou, D; He, Y; Zhang, H. (2020). Mechanism of immunosuppression in
zebrafish (Danio rerio) spleen induced by environmentally relevant concentrations of
perfluorooctanoic acid. Chemosphere 249: 126200.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/6315790
Zhou, R; Cheng, W; Feng, Y; Wei, H; Liang, F; Wang, Y. (2017). Interactions between three typical
endocrine-disrupting chemicals (EDCs) in binary mixtures exposure on myocardial
differentiation of mouse embryonic stem cell. Chemosphere 178: 378-383.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981356
Zhou, W; Zhang, L; Tong, C; Fang, F; Zhao, S; Tian, Y; Tao, Y; Zhang, J. (2017). Plasma perfluoroalkyl
and polyfluoroalkyl substances concentration and menstrual cycle characteristics in
preconception women. Environ Health Perspect 125: 067012.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3859799
Zhou, Y; Bao, WW; Qian, ZM; Dee Geiger, S; Parrish, KL; Yang, BY; Lee, YL; Dong, GH. (2017).
Perfluoroalkyl substance exposure and urine CC16 levels among asthmatics: A case-control study
of children. Environ Res 159: 158-163.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3981296
Zhou, Y; Hu, LW; Qian, ZM; Chang, JJ; King, C; Paul, G; Lin, S; Chen, PC; Lee, YL; Dong, GH.
(2016). Association of perfluoroalkyl substances exposure with reproductive hormone levels in
adolescents: By sex status. Environ Int 94: 189-195.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3856472
Zhou, Y; Hu, LW; Qian, ZM; Geiger, SD; Parrish, KL; Dharmage, SC; Campbell, B; Roponen, M;
Jalava, P; Hirvonen, MR; Heinrich, J; Zeng, XW; Yang, BY; Qin, XD; Lee, YL; Dong, GH.
(2017). Interaction effects of polyfluoroalkyl substances and sex steroid hormones on asthma
among children. Sci Rep 7: 899.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3858488
Zhu, Y; Qin, XD; Zeng, XW; Paul, G; Morawska, L; Su, MW; Tsai, CH; Wang, SQ; Lee, YL; Dong, GH.
(2016). Associations of serum perfluoroalkyl acid levels with T-helper cell-specific cytokines in
children: By gender and asthma status. Sci Total Environ 559: 166-173.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3360105
Zong, G; Grandjean, P; Wang, X; Sun, Q. (2016). Lactation history, serum concentrations of persistent
organic pollutants, and maternal risk of diabetes. Environ Res 150: 282-288.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/3350666
7-89
------- |