vvEPA
April 2024
EPA Document No. 815R24006
FINAL
Human Health Toxicity Assessment for Perfluorooctanoic
Acid (PFOA) and Related Salts
-------
APRIL 2024
FINAL
Human Health Toxicity Assessment for Perfluorooctanoic Acid (PFOA) and
Related Salts
Prepared by:
U.S. Environmental Protection Agency
Office of Water (4304T)
Health and Ecological Criteria Division
Washington, DC 20460
EPA Document Number: 815R24006
April 2024
-------
APRIL 2024
Disclaimer
This document has been reviewed in accordance with U.S. Environmental Protection Agency
(EPA) policy and approved for publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
i
-------
APRIL 2024
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; and Ruth Etzel. 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 Czarina Cooper; Joyce Donohue (retired); Adrienne Keel;
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 (retired); Colleen Flaherty (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; Samantha
Hall; Anthony Hannani; 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; Kate
Weinberger; 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; Shanell Folger; Caroline Foster; Jeremy Frye; Angelina Guiducci; Tara Hamilton;
Pamela Hartman; Cara Henning; Audrey Ichida; Caroline Jasperse; Kaedra Jones; Michele
Justice; Afroditi Katsigiannakis; Gillian Laidlaw; Yi Lu; Mary Lundin; Elizabeth Martin;
Denyse Marquez Sanchez; Alicia Murphy; Emily Pak; Joei Robertson; Lucas Rocha Melogno;
Andrea Santa-Rios; Alessandria Schumacher; Swati Sriram; Nkoli Ukpabi; Harry Whately; and
Wanchen Xiong.
ii
-------
APRIL 2024
Contents
Disclaimer i
Acknowledgments ii
Contents iii
Figures vi
Tables xi
Acronyms and Abbreviations xiv
Executive Summary xx
1 Background 1-1
1.1 Purpose of This Document 1-1
1.2 Background on Per-and Polyfluoroalkyl Substances 1-2
1.3 Chemical Identity 1-3
1.4 Occurrence Summary 1-5
1.4.1 Biomonitoring 1-5
1.4.2 Ambient Water 1-5
1.4.3 Drinking Water 1-6
1.5 History of EPA's Human Health Assessment of PFOA 1-7
2 Summary of Assessment Methods 2-1
2.1 Introduction to the Systematic Review Assessment Methods 2-1
2.1.1 Literature Database 2-2
2.1.2 Literature Screening 2-3
2.1.3 Study Quality Evaluation for Epidemiological Studies and Animal
Toxicological Studies 2-4
2.1.4 Data Extraction 2-5
2.1.5 Evidence Synthesis and Integration 2-6
2.2 Dose-Response Assessment 2-7
2.2.1 Approach to POD and Candidate RfD Derivation for Noncancer Health
Outcomes 2-8
2.2.2 Cancer Assessment 2-10
2.2.3 Selecting Health Outcome-Specific and Overall Toxicity Values 2-12
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 3-5
3.2 Data Extraction Results 3-5
3.3 Toxicokinetic Synthesis 3-6
3.3.1 ADYli: 3-6
3.3.2 Pharmacokinetic Models 3-18
-------
APRIL 2024
3.4 Noncancer Health Effects Evidence Synthesis and Integration 3-25
3.4.1 Hepatic 3-25
3.4.2 Immune 3-102
3.4.3 Cardiovascular 3-154
3.4.4 Developmental 3-207
3.4.5 Evidence Synthesis and Integration for Other Noncancer Health Outcomes3-285
3.5 Cancer Evidence Study Quality Evaluation, Synthesis, Mode of Action Analysis
and Weight of Evidence 3-285
3.5.1 Human Evidence Study Quality Evaluation and Synthesis 3-285
3.5.2 Animal Evidence Study Quality Evaluation and Synthesis 3-293
3.5.3 Mechanistic Evidence 3-298
3.5.4 Weight of Evidence for Carcinogenicity 3-315
3.5.5 Cancer Classification 3-348
4 Dose-Response Assessment 4-1
4.1 Noncancer 4-2
4.1.1 Study and Endpoint Selection 4-2
4.1.2 Estimation or Selection of Points of Departure (PODs) for RfD Derivation .4-21
4.1.3 Pharmacokinetic Modeling Approaches to Convert Administered Dose to
Internal Dose in Animals and Humans 4-27
4.1.4 Application of Pharmacokinetic Modeling for Animal-Human
Extrapolation of PFOA Toxicological Endpoints and Dosimetric
Interpretation of Epidemiological Endpoints 4-39
4.1.5 Derivation of Candidate Chronic Oral Noncancer Reference Doses (RfDs). 4-55
4.1.6 RfD Selection 4-64
4.2 Cancer 4-69
4.2.1 Study and Endpoint Selection 4-69
4.2.2 Candidate CSF Derivation 4-72
4.2.3 Overall CSF Selection 4-76
4.2.4 Application of Age-Dependent Adjustment Factors 4-76
5 Effects Characterization 5-1
5.1 Addressing Uncertainties in the Use of Epidemiological Studies for Quantitative
Dose-Response Analyses 5-1
5.1.1 Uncertainty due to Potential Confounding by Co-Occurring PFAS 5-4
5.2 Comparisons Between Toxicity Values Derived from Animal Toxicological
Studies and Epidemiological Studies 5-10
5.3 Updated Approach to Animal Toxicological RfD Derivation Compared with the
2016 PFOA HESD 5-11
5.4 Consideration of Alternative Conclusions Regarding the Weight of Evidence of
PFOA Carcinogenicity 5-13
5.5 Health Outcomes with Evidence Integration Judgments of Evidence Suggests
Bordering on Evidence Indicates 5-16
5.6 Challenges and Uncertainty in Modeling 5-18
iv
-------
APRIL 2024
5.6.1 Modeling of Animal Internal Dosimetry 5-18
5.6.2 Modeling of Human Dosimetry 5-19
5.6.3 Approach of Estimating a Benchmark Dose from a Regression Coefficient. 5-21
5.7 Human Dosimetry Models: Consideration of Alternate Modeling Approaches 5-22
5.8 Sensitive Populations 5-25
5.8.1 Fetuses, Infants, Children 5-26
5.8.2 Sex Differences 5-26
5.8.3 Other Susceptible Populations 5-27
6 References 6-1
v
-------
APRIL 2024
Figures
Figure ES-1. Schematic Depicting Candidate RfDs Derived From Epidemiological and
Animal Toxicological Studies of PFOA xxiii
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 Design3 3-4
Figure 3-3. Summary of Animal Toxicological Studies of PFOA Exposure by Health
System, Study Design, and Speciesab 3-5
Figure 3-4. Schematic for a Physiologically Motivated Renal-Resorption
PK Model for PFOA 3-23
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 PFOA HESD) 3-27
Figure 3-6. Overall ALT Levels from 2016 PFOA HESD Epidemiology Studies Following
Exposure to PFOA 3-29
Figure 3-7. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects3 3-33
Figure 3-8. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects (Continued)3 3-34
Figure 3-9. Odds of Elevated ALT Levels from Epidemiology Studies Following Exposure
to PFOA 3-36
Figure 3-10. ALT Levels from Medium Confidence Epidemiology Studies Following
Exposure to PFOA 3-37
Figure 3-11. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects 3-39
Figure 3-12. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Hepatic Effects (Continued) 3-40
Figure 3-13. Relative Liver Weight in Rodents Following Exposure to PFOA (logarithmic
scale) 3-43
Figure 3-14. Relative Liver Weight in Rodents Following Exposure to PFOA (Continued,
logarithmic scale) 3-44
Figure 3-15. Percent Change in Serum Enzyme Levels Relative to Controls in Male Mice
Following Exposure to PFOAa b 3-46
Figure 3-16. Percent Change in Serum Enzyme Levels Relative to Controls in Male Rats
Following Exposure to PFOA3 3-47
vi
-------
APRIL 2024
Figure 3-17. Percent Change in Enzyme Levels Relative to Controls in Female Rodents
Following Exposure to PFOAa 3-49
Figure 3-18. Summary of Mechanistic Studies of PFOA and Hepatic Effects 3-55
Figure 3-19. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immune Effects Published Before 2016 (References in
2016 PFOA HESD) 3-103
Figure 3-20. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immunosuppression Effects 3-107
Figure 3-21. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immunosuppression Effects (Continued) 3-108
Figure 3-22. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA 3-111
Figure 3-23. Overall Diphtheria Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA 3-112
Figure 3-24. Overall Rubella Antibody Levels in Children and Adolescents from
Epidemiology Studies Following Exposure to PFOA 3-115
Figure 3-25. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immune Hypersensitivity Effects 3-120
Figure 3-26. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Autoimmune Effects 3-125
Figure 3-27. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Immune Effects 3-127
Figure 3-28. Globulin Levels in Rodents Following Exposure to PFOA (logarithmic scale). 3-133
Figure 3-29. Summary of Mechanistic Studies of PFOA and Immune Effects 3-133
Figure 3-30. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cardiovascular Effects Published Before 2016
(References from 2016 PFOA HESD) 3-156
Figure 3-31. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cardiovascular Effects 3-160
Figure 3-32. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cardiovascular Effects (Continued) 3-161
Figure 3-33. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued) 3-162
Figure 3-34. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Serum Lipids Published Before 2016 (References from
2016 PFOA HESD) 3-170
vii
-------
APRIL 2024
Figure 3-35. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids 3-175
Figure 3-36. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued) 3-176
Figure 3-37. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued) 3-177
Figure 3-38. Odds of High Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA 3-183
Figure 3-39. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA 3-184
Figure 3-40. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-185
Figure 3-41. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-186
Figure 3-42. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued) 3-187
Figure 3-43. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Cardiovascular Effects 3-191
Figure 3-44. Serum Lipid Levels in Rodents Following Exposure to PFOA (logarithmic
scale) 3-194
Figure 3-45. Summary of Mechanistic Studies of PFOA and Cardiovascular Effects 3-195
Figure 3-46. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Developmental Effects Published before 2016 (References
from 2016 PFOA HESD) 3-208
Figure 3-47. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Weight Effects 3-218
Figure 3-48. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued) 3-219
Figure 3-49. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Birth Weight Effects (Continued) 3-220
Figure 3-50. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA 3-222
Figure 3-51. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-223
Figure 3-52. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-224
viii
-------
APRIL 2024
Figure 3-53. Overall Mean Birth Weight from Epidemiology Studies Following Exposure
to PFOA (Continued) 3-224
Figure 3-54. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Small for Gestational Age and Low Birth Weight Effects21 3-228
Figure 3-55. Odds of Small for Gestational Age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA 3-229
Figure 3-56. Odds of Small for Gestational Age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA (Continued) 3-230
Figure 3-57. Odds of Small for Gestational Age in Children from Medium Confidence
Epidemiology Studies Following Exposure to PFOA 3-231
Figure 3-58. Odds of Low Birthweight in Children from Epidemiology Studies Following
Exposure to PFOA 3-232
Figure 3-59. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Length Effects 3-234
Figure 3-60. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Length Effects (Continued) 3-235
Figure 3-61. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Head Circumference Effects 3-238
Figure 3-62. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Postnatal Growth 3-244
Figure 3-63. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Gestational Age 3-247
Figure 3-64. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Preterm Birth Effects 3-250
Figure 3-65. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Fetal Loss 3-252
Figure 3-66. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Defects 3-253
Figure 3-67. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Developmental Effects 3-254
Figure 3-68. Maternal Body Weight in Rodents Following Exposure to PFOA (logarithmic
scale) 3-257
Figure 3-69. Placental Weights in Mice Following Exposure to PFOA 3-258
Figure 3-70. Offspring Mortality in Rodents Following Exposure to PFOAa 3-261
Figure 3-71. Offspring Body Weight in Rodents Following Exposure to PFOA (logarithmic
scale)a 3-264
ix
-------
APRIL 2024
Figure 3-72. Summary of Mechanistic Studies of PFOA and Developmental Effects 3-268
Figure 3-73. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cancer Effects Published Before 2016 (References from
2016 PFOA HESD) 3-287
Figure 3-74. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cancer Effects 3-290
Figure 3-75. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure and Cancer Effects 3-294
Figure 3-76. Summary of Mechanistic Studies of PFOA and Cancer Effects 3-298
Figure 4-1. Model Structure for Lifestage Modeling 4-31
Figure 4-2. Gestation and Lactation Predictions of PFOA in the Rat 4-33
Figure 4-3. Gestation and Lactation Predictions of PFOA in the Mouse in a Cross-Fostering
Study 4-34
Figure 4-4. Comparison of Candidate RfDs Resulting from the Application of Uncertainty
Factors to PODheds Derived from Epidemiological and Animal Toxicological
Studies 4-65
Figure 4-5. Schematic Depicting Selection of the Overall RfD for PFOA 4-68
x
-------
APRIL 2024
Tables
Table ES-1. Final Toxicity Values for PFOA XXV
Table 1-1. Chemical and Physical Properties of PFOA 1-4
Table 3-1. Database Literature Search Results 3-1
Table 3-2. Associations Between Elevated Exposure to PFOA and Hepatic Outcomes from
Studies Identified in the 2016 PFOA HESD 3-30
Table 3-3. Associations Between PFOA Exposure and Cell Death or Necrosis in Rodents 3-52
Table 3-4. Evidence Profile Table for PFOA Exposure and Hepatic Effects 3-92
Table 3-5. Associations Between Elevated Exposure to PFOA and Immune Outcomes from
Studies Identified in the 2016 PFOA HESD 3-105
Table 3-6. Associations between PFOA Exposure and Vaccine Response in Faroe Islands
Studies 3-113
Table 3-7. Effects of PFOA Exposure on Cytokines Impacting Adaptive Immune
Responses 3-139
Table 3-8. Effects of PFOA Exposure on Pro-Inflammatory Cytokines and Markers of
Inflammation 3-143
Table 3-9. Evidence Profile Table for PFOA Exposure and Immune Effects 3-147
Table 3-10. Associations Between Elevated Exposure to PFOA and Cardiovascular
Outcomes from Studies Identified in the 2016 PFOA HESD 3-157
Table 3-11. Associations Between Elevated Exposure to PFOA and Serum Lipids from
Studies Identified in the 2016 PFOA HESD 3-171
Table 3-12. Evidence Profile Table for PFOA Exposure and Cardiovascular Effects 3-200
Table 3-13. Associations Between Elevated Exposure to PFOA and Developmental
Outcomes in Children from Studies Identified in the 2016 PFOA HESD 3-211
Table 3-14. Study Design for Perinatal and Postweaning Exposure Levels for Fi Male and
Female Rats for the NTP {, 2020, 7330145} Study 3-256
Table 3-15. Evidence Profile Table for PFOA Exposure and Developmental Effects 3-276
Table 3-16. Incidences of Liver Tumors in Male Sprague-Dawley Rats as Reported by NTP
{, 2020, 7330145} 3-295
Table 3-17. Incidences of Pancreatic Acinar Cell Tumors in Male Sprague-Dawley Rats as
Reported by NTP {, 2020, 7330145} 3-296
Table 3-18. Incidences of Pancreatic Acinar Cell Tumors in Female Sprague-Dawley Rats
as Reported by NTP {, 2020, 7330145} 3-297
xi
-------
APRIL 2024
Table 3-19. Incidences of Uterine Adenocarcinomas in Female Sprague-Dawley Rats from
the Standard and Extended Evaluations (Combined) as Reported by NTP {,
2020, 7330145} 3-297
Table 3-20. Mutagenicity Data from In Vitro Studies 3-301
Table 3-21. DNA Damage Data from In Vivo Studies 3-302
Table 3-22. DNA Damage Data from In Vitro Studies 3-303
Table 3-23. Evidence of Key Events Associated with the Aromatase Inhibition Mode of
Action for Testicular Tumorsa in Male Rats and Mice Exposed to PFOA 3-321
Table 3-24. Evidence of Key Events Associated with the Estrogen Agonism Mode of
Action for Testicular Tumorsa in Male Rats and Mice Exposed to PFOA 3-323
Table 3-25. Evidence of Key Events Associated with the Testosterone Biosynthesis
Inhibition Mode of Action for Testicular Tumorsa in Male Rats and Mice
Exposed to PFOA 3-324
Table 3-26. Evidence of Key Events Associated with PPARa Agonist-Induced Mode of
Action for Testicular Tumorsa in Male Rats and Mice Exposed to PFOA 3-326
Table 3-27. Evidence of Key Events Associated with the Gastric Bile Alterations Mode of
Action for Pancreatic Tumorsa in Male and Female Rats and Mice 3-329
Table 3-28. Evidence of Key Events Associated with a Proposed Oxidative Stress Mode of
Action Involving the UPR Pathway for Pancreatic Tumorsain Male and Female
Rats and Mice 3-332
Table 3-29. Evidence of Key Events Associated with the PPARa Mode of Action for
Hepatic Tumorsa in Male Rats and Mice Exposed to PFOA 3-335
Table 3-30. Evidence of Key Events Associated with the PPARa Mode of Action for
Hepatic Tumorsa in Female Rats and Mice Exposed to PFOA 3-336
Table 3-31. Evidence of Key Events Associated with the CAR Mode of Action for Hepatic
Tumorsa in Male Rats and Mice Exposed to PFOA 3-339
Table 3-32. Evidence of Key Events Associated with the CAR Mode of Action for Hepatic
Tumorsa in Female Rats and Mice Exposed to PFOA 3-340
Table 3-33. Evidence of Key Events Associated with the Cytotoxicity Mode of Action for
Hepatic Tumorsa in Male Rats and Mice Exposed to PFOA 3-342
Table 3-34. Evidence of Key Events Associated with the Cytotoxicity Mode of Action for
Hepatic Tumorsa in Female Rats and Mice Exposed to PFOA 3-343
Table 3-35. 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} 3-350
Table 4-1. Summary of Observed Endpoints in Humans and Rodent Studies Considered for
Dose-Response Modeling and Derivation of Points of Departure 4-17
xii
-------
APRIL 2024
Table 4-2. Benchmark Response Levels Selected for BMD Modeling of Health Outcomes.... 4-26
Table 4-3. PK Parameters From Wambaugh et al. {, 2013, 2850932} Meta-Analysis of
Literature Data for PFOA 4-29
Table 4-4. Model Predicted and Literature PK Parameter Comparisons for PFOA 4-30
Table 4-5. Additional PK Parameters for Gestation/Lactation for PFOA 4-32
Table 4-6. Updated and Original Chemical-Specific Parameters for PFOA in Humans 4-37
Table 4-7. Summary of Studies Reporting the Ratio of PFOA Levels in Breastmilk and
Maternal Serum or Plasma 4-38
Table 4-8. PODheds Considered for the Derivation of Candidate RfD Values 4-41
Table 4-9. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
From Epidemiological Studies {U.S. EPA, 2002, 88824} 4-58
Table 4-10. Uncertainty Factors for the Development of the Candidate Chronic RfD Values
From Animal Toxicological Studies {U.S. EPA, 2002, 88824} 4-59
Table 4-11. Candidate Reference Doses (RfDs) 4-62
Table 4-12. Candidate Cancer Slope Factors Based on Epidemiological Data 4-73
Table 4-13. Candidate Cancer Slope Factors Based on Animal Toxicological Data from 2-
year Cancer Bioassays 4-75
Table 5-1. Correlation Coefficients Between PFOA and Other PFAS in Mutually Adjusted
Studies 5-5
Table 5-2. Impact of Co-Exposure Adjustment on Estimated Change in Mean Birth Weight
(grams) per Unit Change (ng/mL) in PFOA Levels 5-8
Table 5-3. Comparison of Candidate RfDs Derived from Animal Toxicological Studies for
Priority Health Outcomes21 5-12
Table 5-4. Comparison of the PFOA Carcinogenicity Database with Cancer Descriptors as
Described in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329} 5-15
xiii
-------
APRIL 2024
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
xiv
-------
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
APRIL 2024
CSM
Cholestyramine
CVD
Cardiovascular disease
DBP
Diastolic blood pressure
DCF
27 '-dichlorofluorescein
DCF-DA
Dichlorodihydro-
fluorescein diacetate
DDE
Di chl orodipheny 1
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
xv
-------
APRIL 2024
GBCA
Genetic and Biomarker
HOME
Health Outcomes and
Study for Childhood
Measures of the
Asthma
Environment
GCL
Glutamate-cysteine
HPA
Hypothalamic-pituitary-
ligase
adrenal
GD
Gestation day
HR
Hazard ratio
GFR
Glomerular filtration rate
HRL
Health reference level
GGT
y-glutamyltransferase
HSA
Human serum albumin
GM
Geometric mean
IARC
International Agency for
GO
Gene Ontology
Research on Cancer
GSH
Glutathione
IDL
Intermediate-density
GSPE
Grape seed
lipoprotein
proanthocyandidn extract
IFN
Interferon
GSSG
Glutathione disulfide
Ig
Immunoglobulin
GST
Glutathione S-
IGF-1
Insulin-like growth factor 1
transferases
IHD
Ischemic heart diseases
HAT
Hi stone acetylase
IHIC
Hepatic immune cell
HAWC
Health Assessment
IL
Inflammatory cytokine
Workplace Collaborative
INMA
Spanish Environment
HDAC
Hi stone deacetylase
and Childhood (Infancia
HDL
High-density-lipoprotein
y Medio Ambiente)
HED
Human equivalent dose
IP
Intraperitoneal
HEK-293
Human embryonic
IPCS
International Programme
kidney
on Chemical Safety
HERO
Health and
IQR
Interquartile range
Environmental Research
IRIS
Integrated Risk
Online
Information System
HESD
Health Effects Support
IV
Intravenous
Document
ki2
Intercompartment
HFC
7-hy droxytrifluoro-
transfer rate
methylcoumarin
ka
Absorption rate
HFD
High-fat diet
Kd
Disassociation constant
HFMD
Hand, foot, and mouth
Kh
Henry's Law Constant
disease
KK
Kallikrein-kinn system
HFPO
Hexafluoropropy 1 ene
KLH
Keyhole limpet
oxide
hemocyanin
Hib
Haemophilus influenza
Kmem/w
Membrane/water
type b
partition coefficients
HK
High-molecular-weight
Koc
Organic carbon-water
kininogen
partitioning coefficient
hL-FABP
Human liver fatty acid
Kow
Octanol-water partition
binding protein
coefficient
HMOX
Heme oxygenase
LBW
Low birth weight
HNF
Hepatocyte nuclear
factor
xvi
-------
APRIL 2024
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
xvii
-------
APRIL 2024
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
P milk
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
^milk
r2milk
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
xviii
-------
APRIL 2024
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
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
xix
-------
APRIL 2024
Executive Summary
The U.S. Environmental Protection Agency (EPA) is issuing final toxicity values for
perfltiorooctanoic acid (PFOA), including all isomers and nonmetal salts. The toxicity
assessment for PFOA is a scientific report that describes the evaluation of the available animal
toxicity and human epidemiology data in order to characterize noncancer and cancer human
health hazards. This assessment also includes final toxicity values associated with noncancer
health effects (i.e., oral reference doses, or RfDs) and cancer effects (i.e., cancer slope factors, or
CSFs) following oral PFOA exposure. It is not a risk assessment, as it does not include an
exposure assessment or an overall risk characterization nor does it address the legal, policy,
social, economic, or technical considerations involved in risk management. The PFOA toxicity
assessment can be used by EPA, states, Tribes, and local communities, along with specific
exposure and other relevant information, to determine, under the appropriate regulations and
statutes, the potential risk associated with human exposures to PFOA, its isomers, and its
nonmetal salts.
This final toxicity assessment was peer reviewed by the EPA Science Advisory Board (SAB)
per- and polyfluoroalkyl substances (PFAS) Review Panel in November 2021 and underwent
public comment in March 2023. It incorporated expert scientific recommendations received from
the SAB in 2022 {U.S. EPA, 2022, 10476098} as well as feedback from the public comment
period {U.S. EPA, 2024, 11414326}. This final assessment builds upon the literature review
presented in the 2016 Health Effects Support Document for Perflaorooctanoic Acid (PFOA)
(hereafter referred to as the 2016 PFOA HESD) {U.S. EPA, 2016, 3603279} and is an update of
the SAB review draft, Proposed Approaches to the Derivation of a Draft Maximum Contaminant
Level Goal for Perflaorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking Water {U.S.
EPA, 2021, 10428559}, and the subsequent Public Comment Draft Toxicity Assessment and
Proposed Maximum Contaminant Level Goal for Perflaorooctanoic Acid (PFOA) in Drinking
Water {U.S. EPA, 2022, 10476098}.
PFOA and its related salts are members of the PFAS group. These manufactured chemicals have
a history of industrial and consumer use in the United States and are considered persistent
chemicals based on their physicochemical properties. Some of the human health concerns about
exposure to PFOA and other PFAS stem from their resistance to hydrolysis, photolysis,
metabolism, and microbial degradation in the environment and in the human body. PFAS are not
naturally occurring; they are man-made compounds that have been used widely over the past
several decades in industrial applications and consumer products since many PFAS have
repellant and surfactant properties. Frequently used as emulsifiers and as stain-, oil-, or water-
repellents, PFAS are found in a variety of environmental media and in tissues of organisms,
including humans.
Under the EPA's PFOA Stewardship Program, the eight major companies of the
perfluoropolymer/fluorotelomer industry agreed to voluntarily reduce facility emissions and
product content of PFOA, precursor chemicals that can break down to PFOA, and related higher
homologue chemicals, longer-chain perfluoroalkyl carboxylic acids (PFCAs) by 95% on a global
basis by no later than 2010 and to eliminate these substances in products by 2015 {U.S. EPA,
2021, 6569670}. However, PFOA remains persistent in environmental media because it is
resistant to environmental degradation processes.
xx
-------
APRIL 2024
The purpose of this human health toxicity assessment is to derive toxicity values pertaining to
oral exposure for PFOA. The development of this toxicity assessment relied on a robust
systematic review process, based on the EPA peer-reviewed human health risk assessment
methodology outlined in the EPA ORD Staff Handbook for Developing IRIS Assessments {U.S.
EPA, 2022, 10367891}, to identify human epidemiological, animal toxicological, mechanistic,
and toxicokinetic data relevant to oral exposure. The PFOA systematic review protocol (see
Appendix A, {U.S. EPA, 2024, 11414343}) was developed prior to the initiation of this
assessment and largely mirrors the Systematic Review Protocol for the PFBA, PFHxA, PFHxS,
PFNA, and PFDA (Anionic and Acid Forms) IRIS Assessments {U.S. EPA, 2020, 8642427}.
The protocol outlines the scoping and problem-formulation efforts and describes the systematic
review, including study quality evaluation, and the dose-response methods used to conduct this
assessment. The final assessment incorporates peer-reviewed studies captured from: EPA's 2016
PFOAHESD {U.S. EPA, 2016, 3603279}, literature searches of scientific databases and gray
literature from 2013 through February 2023, the SAB PFAS Review Panel recommendations,
and public comment. Consistent with the analysis provided in the peer-reviewed draft assessment
{U.S. EPA, 2021, 10428559} and with recommendations from external peer review (i.e., the
SAB PFAS Review Panel; {U.S. EPA, 2022, 10476098}), this final assessment focused on
qualitative and quantitative assessment of five "priority" health outcome categories based on
those with the strongest weight of evidence. These five priority health outcomes are cancer,
hepatic, developmental, cardiovascular, and immune. The results of the systematic literature
reviews and qualitative assessments for the remaining "nonpriority" health outcomes are
presented in the Appendix accompanying this final assessment {U.S. EPA, 2024, 11414343}.
Qualitative Assessment of Noncancer Effects
Overall, the available evidence indicates that PFOA exposure is likely to cause hepatic,
immunological, cardiovascular, and developmental effects in humans, given sufficient exposure
conditions (e.g., at measured levels in humans as low as 1.1 to 5.2 ng/mL and at administered
doses in animals as low as 0.3 to 1.0 mg/kg/day). These judgments are based on data from
epidemiological studies of infants, children, adolescents, pregnant individuals, and nonpregnant
adults, as well as short-term (28-day), subchronic (90-day), developmental (gestational), and
chronic (2-year) oral-exposure studies in rodents. For hepatic effects, the primary support is
evidence of increased serum liver enzyme levels (i.e., alanine transaminase (ALT)) in humans
and coherent evidence of hepatotoxicity in animals, including increased liver weights and
hepatocellular hypertrophy accompanied by necrosis, inflammation, or increased liver enzyme
levels that indicate liver injury. For immunological effects, the primary support is evidence of
developmental immunosuppression in humans, specifically decreased antibody response to
vaccination against tetanus and diphtheria in children, and evidence of immunosuppression and
other types of immunotoxicity in studies of adult animals, including decreased IgM response to
sheep red blood cells, reduced spleen and thymus weights, changes in immune cell populations,
and decreased splenic and thymic cellularity. For cardiovascular effects, the primary support is
evidence of increased serum lipid levels in humans and alterations to lipid homeostasis in
animals. For developmental effects, the primary evidence is decreased birth weight in human
infants and decreased offspring survival, decreased fetal and pup weight, delayed time to eye
opening, and related pre- and postnatal effects in animal studies. According to the protocol
described in Appendix A {U.S. EPA, 2024, 11414343} and aligned with EPA peer-reviewed
human health risk assessment methodology {U.S. EPA, 2022, 10367891}, selected quantitative
xxi
-------
APRIL 2024
data in medium and high confidence studies from these identified hazards were used to derive
toxicity values (see Table ES-1). Specific criteria for data and study selection are provided in
Appendix A {U.S. EPA, 2024, 11414343} and Section 4.1.
Quantitative Assessment of Noncancer Effects and Oral RfD
Derivation
EPA followed agency guidelines and methodologies for risk assessment in determining points of
departure (PODs) for the derivation of the RfDs for PFOA {U.S. EPA, 2022, 10367891; U.S.
EPA, 2002, 88824; U.S. EPA, 2012, 1239433; U.S. EPA, 2011, 786546; U.S. EPA, 2014,
2520260} and performed modeling following EPA's Benchmark Dose Technical Guidance
Document {U.S. EPA, 2012, 1239433}. For data from epidemiological studies, the dose-
response modeling approach was selected based on the health outcome and available data. A
hybrid modeling approach, which estimated the probability of responses at specified exposure
levels above the control, was conducted when clinically adverse outcome levels could be defined
(i.e., for developmental, hepatic, and cardiovascular effects) following EPA's Benchmark Dose
Technical Guidance Document {U.S. EPA, 2012, 1239433}. For other outcomes (i.e., immune
effects), study results from multivariate models were used to define a benchmark response
(BMR). For data from animal toxicological studies, EPA conducted benchmark dose modeling,
when possible, to empirically model the dose-response relationship in the range of observed data.
When BMDLs could not be derived, EPA used a no-observed-adverse-effect level/lowest-
observed-adverse-effect level (NOAEL/LOAEL) approach.
PODs were converted to external POD human equivalent doses (PODheds) using
pharmacokinetic modeling (see Section 4.1.3). Consistent with the recommendations presented
in EPA's A Review of the Reference Dose and Reference Concentration Processes {U.S. EPA,
2002, 88824}, EPA considered the database of information to inform the application of
uncertainty factors (UFs) to PODheds to address intraspecies variability, interspecies variability,
extrapolation from a LOAEL to NOAEL, extrapolation from a subchronic to a chronic exposure
duration, and database deficiencies. EPA derived and considered multiple candidate RfDs from
both human epidemiological and animal toxicological studies across the four priority noncancer
health outcomes that EPA determined had the strongest weight of evidence (i.e., immune,
cardiovascular, hepatic, and developmental) (see Figure ES-1 for candidate RfD values).
Additional details on candidate RfD derivation for PFOA are available in Section 4.1.
xxii
-------
APRIL 2024
Human Animal
Decreased
serum
anti-tetanus
antibody
concentration
in children
Decreased
serum
anti-diptheria
antibody
concentration
in children
Decreased
IgM response
to SRBC
Timmerman, 2021, 9416315;
Medium confidence
Budtz-Jorgensen, 2018, 5083631;
Medium confidence
Timmerman, 2021, 9416315;
Medium confidence
Budtz-Jargensen, 2018, 5083631;
Medium confidence
Dewitt, 2008, 1290826;
Medium confidence
o
-o
-o
-o
RfD
UF
PODHED
o
¦o
Decreased
Birth Weight
Delayed Time
to Eye
Opening
Decreased
Offspring
Survival
Sagiv, 2018, 4238410;
High confidence
Wikstrom, 2020, 6311677;
High confidence
Lau, 2006, 1276159;
Medium confidence
Song, 2018, 5079725;
Medium confidence
-o
-o
-o
-o
Increased
Serum Total
Cholesterol
Dong, 2019, 5080195;
Medium confidence
Steenland, 2009, 1291109;
Medium confidence
-o
-o
Gallo, 2012, 1276142;
Medium confidence
o
Increased
Serum ALT
Necrosis
Darrow, 2016, 3749173;
Medium confidence
Nian, 2019, 5080307;
Medium confidence
NTP, 2020, 7330145;
High confidence
-o
-o
—o
10-3 10-2
10-s 10"7 10-6 10-5 10-4
PFOA Concentration (mg/kg-d)
Figure ES-1. Schematic Depicting Candidate RfDs Derived From Epidemiological and
Animal Toxicological Studies of PFOA
See text and Figure 4-4 in Section 4.1 for additional detail on dose-response modeling for PFOA studies.
xxiii
-------
APRIL 2024
The co-critical effects for the oral RfD of 3 x 10 8 mg/kg/day were decreased serum anti-tetanus
and anti-diphtheria antibody concentrations in children {Budtz-Jorgensen and Grandjean, 2018,
5083631}, decreased infant birth weight {Wikstrom et al., 2020, 6311677}, and increased total
cholesterol in adults {Dong et al., 2019, 5080195} (see Table ES-1). These co-critical effects
were selected based on the procedures outlined in the protocol (see Appendix A, {U.S. EPA,
2024, 11414343}) and consistent with EPA peer-reviewed human health risk assessment
methodology {U.S. EPA, 2022, 10367891}. The RfD was derived by using a total UF of 10 to
account for intraspecies variability (UFh). Notably, the RfD is protective of effects that may
occur in sensitive populations (i.e., embryo and fetus, infants, and young children), 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 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.
Qualitative Carcinogenicity Assessment
Consistent with EPA's Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329},
EPA reviewed the available data and conducted a weight of the evidence evaluation across the
human epidemiological, animal toxicological, and mechanistic studies and concluded that PFOA
is Likely to Be Carcinogenic to Ramans via the oral route of exposure (see Section 3.5).
Epidemiological studies provided evidence of kidney and testicular cancer in humans and some
evidence of breast cancer in a study of one susceptible subpopulation. Animal toxicological
studies in Sprague-Dawley rats reported Leydig cell tumors (LCT), pancreatic acinar cell tumors
(PACT), and hepatocellular tumors after chronic oral exposure. Available mechanistic data
suggest that multiple modes of action (MOAs) play a role in the renal, testicular, pancreatic, and
hepatic tumorigenesis associated with PFOA exposure in humans and animal models. A full
MOA analysis, including in-depth discussions on the potential MO As for kidney and testicular
tumors, as well as discussions on the potential MO As and human relevance for pancreatic and
liver tumors observed in rats, is presented in Section 3.5.4.2.
Quantitative Cancer Assessment and Cancer Slope Factor
Derivation
EPA followed agency guidelines for risk assessment in deriving CSFs for PFOA {U.S. EPA,
2012, 1239433; U.S. EPA, 2022, 10367891; U.S. EPA, 2005, 6324329}. EPA selected medium
and high confidence studies for derivation that met criteria outlined in the protocol (see
Appendix A, {U.S. EPA, 2024, 11414343}) and Section 4.1.1, conducted benchmark dose
modeling {U.S. EPA, 2012, 1239433}, and used the same pharmacokinetic modeling approach
as described for the derivation of noncancer RfDs above (see Section 4.2.2). From the studies
that met the criteria, EPA derived and considered multiple candidate CSFs from both
epidemiological and animal toxicological studies across multiple tissue and organ types (i.e.,
kidney, liver, pancreas, testes). Candidate CSFs were derived for epidemiological data on renal
cell carcinoma (RCC) and kidney cancer using weighted linear regressions to calculate quartile-
specific relative kidney cancer risks. Relative risks were then converted to the absolute risk scale,
yielding an internal CSF, which represents the excess cancer risk associated with each ng/mL
increase in serum PFOA. The internal serum CSF was then divided by the selected clearance
value and converted to an external dose CSF. For animal toxicological studies, multistage cancer
xxiv
-------
APRIL 2024
models were used to predict the doses at which the selected BMR for tumor incidence would
occur. BMDLs for each tumor type (LCTs, hepatocellular adenoma or carcinoma, and pancreatic
acinar cell adenoma or adenocarcinoma) served as the PODs, which were then converted to
PODheds by applying the human clearance value. CSFs were then calculated by dividing the
selected BMR by the PODheds for each tumor type.
The oral slope factor of 0.0293 (ng/kg/day) 1 for RCC in human males from Shearer et al. {,
2021, 7161466} was selected as the basis of the overall CSF for PFOA (see Table ES-1;
rationale in Section 4.2). Per EPA's Guidelines for Carcinogen Risk Assessment and
Supplemental Guidance for Assessing Susceptibility fi'om Early-Life Exposure to Carcinogens
{U.S. EPA, 2005, 6324329; U.S. EPA 2005, 88823}, age-dependent adjustment factors were not
applied during CSF derivation because there was a lack of information to support a mutagenic
MO A for PFOA, and the available evidence did not report an increased susceptibility to cancer
following PFOA exposure during early life. Additional detail on candidate CSF derivation and
CSF selection is provided in Table 4-12 and Table 4-13 in Section 4.2.
Final Toxicity Values for PFOA
Table ES-1. Final Toxicity Values for PFOA
Toxicity
Value Type
Critical Effect(s) Study, Confidence Species, Sex, Age Toxicity Valuea b
Reference
Dose
Co-critical effects:
decreased serum anti-
tetanus and anti-
diphtheria antibody
concentration in
children;
decreased birth weight
in infants;
Increased serum total
cholesterol in adults
Cancer Slope Renal cell carcinoma
Factor
Budtz-Jorgensen {,
2018, 5083631},
Medium',
Wikstrom et al. {,
2020, 6311677},
High',
Dong et al. {, 2019,
5080195}, Medium
Shearer et al. {,
2021, 7161466},
Medium
Human, male and female,
PFOA concentrations at
age five years and anti-
tetanus antibody serum
concentrations at age
seven years;
human male and female,
PFOA serum
concentrations in first and
second trimesters;
human male and female,
20-80 years
Human, male and female,
55-74 years
3xl0-8 (mg/kg/d)
0.0293 (ng/kg/d)-l
Notes:
a Reference doses were rounded to one significant figure.
b Increase in cancer risk per 1 ng/(kg*d) increase in dose.
XXV
-------
APRIL 2024
1 Background
1.1 Purpose of This Document
The primary purpose of this toxicity assessment for perfluorooctanoic acid (PFOA) is to describe
the best available science on the human health effects associated with PFOA exposure and the
derivation of toxicity values (i.e., noncancer reference doses (RfDs) and cancer slope factors
(CSFs)). The latest health science on PFOA was identified, evaluated using systematic review
methods, and described, and subsequently, a cancer classification was assigned and toxicity
values were developed. The final cancer classification and cancer and noncancer toxicity values
in this assessment build on the work described in the Public Comment Draft Toxicity Assessment
and Proposed Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) in
Drinking Water {U.S. EPA, 2023, 10841009}, Proposed Approaches to the Derivation of a
Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) (CASRN 335-67-1)
in Drinking Water {U.S. EPA, 2021, 10428559}, and the Health Effects Support Document for
Perfluorooctanoic Acid (PFOA) {U.S. EPA, 2016, 3603279}. This final toxicity assessment for
PFOA reflects expert scientific recommendations from the U.S. Environmental Protection
Agency (EPA) Science Advisory Board (SAB) {U.S. EPA, 2022, 10476098} and public
comments received on the draft assessment (https://www.resulations.gov/docket/EPA-HQ-OW-
2022-0114; U.S. EPA {, 2024, 11414326}).
In addition to documenting EPA's basis for the cancer classification and toxicity values, this
document serves to:
• Describe and document transparently the literature searches conducted and systematic
review methods used to identify health effects information (epidemiological and animal
toxicological studies and physiologically based pharmacokinetic models) in the literature
(Sections 2 and 3; Appendices A andB, {U.S. EPA, 2024, 11414343}).
• 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 (Section 2; Appendix A, {U.S. EPA, 2024,
11414343}).
• Identify epidemiological and animal toxicological literature that reports health effects
after exposure to PFOA (and its related salts) as outlined in the PECO criteria (Section 3).
• Describe and document the study quality evaluations conducted on epidemiological and
animal toxicological studies considered potentially useful for point-of-departure (POD)
derivation (Section 3).
• Describe and document the data from all epidemiological studies and animal toxicological
studies that were considered for POD derivation (Section 3).
• Synthesize and document the adverse health effects evidence across studies. The
assessment focuses on synthesizing the available evidence for five main health outcomes
that were found to have the strongest weight of evidence, as recommended by the SAB -
developmental, hepatic, immune, and cardiovascular effects, and cancer (Section 3) -and
also provides supplemental syntheses of evidence for dermal, endocrine, gastrointestinal,
hematologic, metabolic, musculoskeletal, nervous, ocular, renal, and respiratory effects,
1-1
-------
APRIL 2024
reproductive effects in males or females, and general toxicity (Appendix C, {U.S. EPA,
2024, 11414343}).
• 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 a focus on five main
health outcomes (developmental, hepatic, immune, and cardiovascular effects, and cancer)
(Section 3).
• Develop and document strength of evidence judgments across studies (or subsets of
studies) separately for epidemiological, animal toxicological, and mechanistic lines of
evidence for the five main health outcomes (Section 3).
• Develop and document integrated expert judgments across evidence streams (i.e.,
epidemiological, animal toxicological, and mechanistic streams) as to whether and to what
extent the evidence supports that exposure to PFOA has the potential to be hazardous to
humans (Section 3).
• Determine the cancer classification for PFOA using a weight-of-evidence approach
(Section 3.5.5).
• Describe and document the attributes used to evaluate and select studies for derivation of
toxicity values. These attributes are considered in addition to the study confidence
evaluation domains and enable extrapolation to relevant exposure levels (e.g., studies with
exposure levels near the range of typical environmental human exposures, broad exposure
range, or multiple exposure levels) (Section 4).
• Describe and document the dose-response analyses conducted on the studies identified for
POD derivation (Section 4).
• Derive candidate RfDs (Section 4.1) and CSFs (Section 4.2), select the final RfD (Section
4.1.6) and CSF (Section 4.2.3) for PFOA, and describe the rationale.
• Characterize hazards (e.g., uncertainties, data gaps) (Sections 3, 4, and 5).
1.2 Background on Per-and Polyfluoroalkyl Substances
Per-and polyfluoroalkyl substances (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. There
is no consensus definition of PFAS as a class of chemicals {OSTP, 2023, 11396268}. Consistent
with three related structural definitions associated with EPA's identification of PFAS included in
the fifth Contaminant Candidate List1 (CCL 5), the universe of environmentally relevant PFAS -
including parent chemicals, metabolites, and degradants - is approximately 15,000 compounds.2
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 1950's. PFAS have strong, stable carbon-fluorine (C-F)
1 The CCL is a list, published 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, and might require regulation under SDWA.
2 See the EPA List of PFAS Structures available at: https://comptox.epa.gov/dashboard/chemical-lists/PFASSTRUCT.
1-2
-------
APRIL 2024
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 to 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}.
Because of 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.
With regard to 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 belong to the PFAA
family of the non-polymer PFAS category and is among the most researched PFAS in terms of
human health toxicity and biomonitoring studies (for review, see {Podder, 2021,
9640865@@author-year}).
1.3 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 K0c) 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, oceangoing 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 K0c {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.
1-3
-------
APRIL 2024
For a more detailed discussion of the chemical and physical properties and environmental fate of
PFOA, please see the PFAS Occurrence and Contaminant Background Support Document for
the Final PFAS National Primary Drinking Water Regulation {U.S. EPA, 2024, 11414328}, the
2016 Health Effects Support Document for Perfluorooctanoic Acid (PFOA) {U.S. EPA, 2016,
3603279}, and the Draft Aquatic Life Ambient Water Quality Criteria for Perfluorooctanoic
Acid (PFOA) {U.S. EPA, 2022, 10671186}.
Table 1-1. Chemical and Physical Properties of PFOA
Property
Perfluorooctanoic Acid;
Experimental Average
Source
Chemical Abstracts Service Registry
335-67-1
{NLM, 2022,
Number (CASRN)3
10369702 (7 a a u t ho r-v e a r j
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 ComoTox Chemicals
acid; pentadecafluoro-n-octanoic acid;
Dashboard
octanoic acid, pentadecafluoro-;
perfluorocaprylic acid;
pentadecafluorooctanoic acid;
perfluoroheptanecarboxylic acid
Chemical Formula
CgHFisQ.
{NLM, 2022,
10369702 Vv, a a u t ho r-v e a r j
Molecular Weight
414.069 g/mol
{NLM, 2022,
10369702 Vv, a a u t ho r-v e a r j
Color/Physical State
White to off-white powder (ammonium
{NLM, 2022,
salt)
10369702 Vv, a a u t ho r-v e a r j
Boiling Point
192°C
{NLM, 2022,
10369702 (7 a a u t ho r-v e a r j
Melting Point
54.3°C
{NLM, 2022,
10369702V/, a a u t ho r-v e a r j
Vapor Pressure
0.0316 mm Hg at 19°C
{NLM, 2022,
0.017 mm Hg at 20°C
10369702(®(®author-year};
{ATSDR, 2021,
9642134(®(®author-year}
(extrapolated)
Henry's Law Constant (KH)
0.362 Pa-m3/mol (converts to
{ATSDR, 2021,
3.57E-06 atm-m3/mol)
9642134 a a au t ho r-v e a r j
pKa
1.30, 2.80, -0.5-4.2, 0.5, 0.5
{NLM, 2022,
10369702(®(®author-year};
{ATSDR, 2021,
9642134(®(®author-year}
Koc
631 ± 7.9 L/kg (mean ± 1 standard
{Zareitalabad, 2013,
deviation of selected values)
5080561 (®(®author-year}
(converted from log Koc to Koc)
Solubility in Water
2,290 mg/L at 24°C (estimated);
{NLM, 2022,
3,300 mg/L at 25°C; 4,340 mg/L at
10369700 (7 a a u t ho r-y c a r |
24.1°C
{ATSDR, 2021,
9,500 mg/L at 25°C; 3,300 mg/L at
9642134(®(®author-year}
25°C
Notes: CASRN = Chemical Abstracts Service Registry Number; Koc = organic carbon-water partitioning coefficient;
Kow = octanol-water partition coefficient; pKa: negative base-10 logarithm of acid dissociation constant.
1-4
-------
APRIL 2024
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.
1.4 Occurrence Summary
1.4.1 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 has 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 declined by >70% between 1999 and 2018, presumably due to restrictions on its
commercial usage in the United States {CDC, 2017, 4296146}. However, studies of residents in
locations of suspected PFAS contamination show higher serum levels of PFAS, including PFOA,
compared with the general U.S. population as reported by NHANES {Kotlarz, 2020, 6833715;
Yu, 2020, 6315796; Table 17-6 in \ITRC, 2023, 9959768; AT SDR, 2022, 10519308}.
Under EPA's PFOA Stewardship Program, the eight major companies of the
perfluoropolymer/fluorotelomer industry agreed to voluntarily reduce facility emissions and
product content of PFOA, precursor chemicals that can break down to PFOA, and related higher
homologue chemicals, including perfluorononanoic acid (PFNA) and longer-chain
perfluorocarboxylic acids (PFCAs), by 95% on a global basis by no later than 2010 and to
eliminate these substances in products by 2015 {USEPA, 2021, 6569670}. 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), PFNA, and 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.
1.4.2 Ambient Water
Among the PFAS with established analytical methods for detection, PFOA is one of the
dominant PFAS compounds detected in ambient water both in the United States 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 United States, 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}.
PFOA concentrations in surface water tend to increase with increasing levels of urbanization.
Across the Great Lakes region, PFOA was higher in the downstream lakes (Lake Erie and Lake
1-5
-------
APRIL 2024
Ontario), which are more heavily impacted by urbanization, and lower in the upstream lakes
(Lakes Superior, Michigan, and Huron), which are located in a relatively rural and forested area
{Remucal, 2019, 5413103}. Similarly, Zhang et al. {, 2016, 3470830} found measured surface
water PFOA concentrations in urban areas (urban average PFOA concentration = 10.17 ng/L;
n = 20) to be more than three times greater than concentrations in rural areas (rural average
PFOA concentration = 2.95 ng/L; n = 17) within New Jersey, New York, and Rhode Island.
Seasonal variations in PFOA levels in U.S. surface waters remain largely unknown because of a
lack of experimental evidence examining alterations in PFOA concentrations across time.
1.4.3 Drinking Water
Ingestion of drinking water is a potentially significant source of exposure to PFOA. Serum
PFOA concentrations are known to be elevated among individuals living in communities with
drinking water contaminated from environmental discharges.
EPA uses the Unregulated Contaminant Monitoring Rule (UCMR) to collect data for
contaminants that are suspected to be present in drinking water and do not have health-based
standards set under the Safe Drinking Water Act (SDWA). Under the UCMR, drinking water is
monitored from public water systems (PWSs), specifically community water systems and non-
transient, non-community water systems. The UCMR improves EPA's understanding of the
frequency and concentrations of contaminants of concern occurring in the nation's drinking
water systems. The first four UCMRs collected data from a census of large water systems
(serving more than 10,000 people) and from a statistically representative sample of small water
systems (serving 10,000 or fewer people). UCMR 3 monitoring occurred between 2013 and 2015
and is currently the most comprehensive nationally representative finished water dataset for
PFOA {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. Under UCMR 3, 36,972 samples
from 4,920 PWSs were analyzed. PFOA was found above the UCMR 3 minimum reporting level
(20 ng/L) in 379 samples at 117 systems serving a population of approximately 7.6 million
people located in 28 states, Tribes, or U.S. territories {USEPA, 2024, 11414345; USEPA, 2024,
11414328}.
More recent state data were collected using newer EPA-approved analytical methods and some
state results reflect lower reporting limits than those in the UCMR 3. State data are available
from 32 states: Alabama, Arizona, California, Colorado, Delaware, Georgia, Idaho, Illinois,
Indiana, Iowa, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, Missouri,
New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio,
Oregon, Pennsylvania, South Carolina, Tennessee, Vermont, Virginia, West Virginia, and
Wisconsin {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. State results show continued
occurrence of PFOA in multiple geographic locations. These data also show PFOA occurrence at
lower concentrations and significantly greater frequencies than were measured under the UCMR
3, likely because the more recent monitoring was able to rely on more sensitive analytical
methods {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. More than one-third of states
that conducted nontargeted monitoring detected PFOA and/or PFOS at more than 25% of
systems {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. Among the detections, PFOA
concentrations ranged from 0.21 to 650 ng/L with a range of median concentrations from 1.27 to
5.61 ng/L {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. Monitoring data for PFOA and
PFOS from states that conducted targeted monitoring efforts, including 15 states, demonstrate
results consistent with the nontargeted state monitoring. Within the 20 states that conducted
1-6
-------
APRIL 2024
nontargeted monitoring, there are 1,260 systems with results above 4.0 ng/L and 1,577 systems
with results above 4.0 ng/L {USEPA, 2024, 11414345; USEPA, 2024, 11414328}. These
systems serve populations of 12.5 and 14.4 million people, respectively. Monitoring data for
PFOA from states that conducted targeted sampling efforts showed additional systems exceeding
4 ng/L {USEPA, 2024, 11414345; USEPA, 2024, 11414328}.
Finally, the fifth UCMR (UCMR 5) was published in December 2021 and requires sample
collection and analysis for 29 PFAS, including PFOA, between January 2023 and December
2025 using drinking water analytical methods developed by EPA {USEPA, 2021, 11374428}.
The UCMR 5 defined the minimum reporting level at 4 ng/L for PFOA using EPA Method 533,
which is lower than the 20 ng/L used in the UCMR 3 with EPA Method 537 {USEPA, 2021,
11374428}. Therefore, UCMR 5 will be able to provide nationally representative occurrence
data for PFOA at lower detection concentrations. While the complete UCMR 5 dataset is not
currently available, the small subset of data released (7% of the total results that EPA expects to
receive) as of July 2023 is consistent with the results of UCMR 3 and the state data described
above {USEPA, 2024, 11414345; USEPA, 2024, 11414328}.
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.
1.5 History of EPA's Human Health Assessment of PFOA
EPA developed an HESD for PFOA after it was listed on the third CCL (CCL 3) in 2009 {U.S.
EPA, 2009, 1508321}. An HESD is synonymous with a toxicity assessment in that they both
describe the assessment of cancer and noncancer health effects and derive toxicity values. The
2016 PFOA HESD was peer reviewed in 2014 and revised based on consideration of peer
reviewers' comments, public comments, and additional studies published through December
2015. The resulting Health Effects Support Document for Perflaorooctanoic Acid (PFOA){U. S.
EPA, 2016, 3603279} was published in 2016 and described the assessment of cancer and
noncancer health effects and the derivation of a CSF and noncancer RfD for PFOA.
EPA initiated an update to the 2016 PFOA HESD in 2021 when the agency made a
determination to regulate PFOA with a national primary drinking water regulation (NPDWR)
{U.S. EPA, 2021, 9640861}. The initial update of the 2016 PFOA HESD was the Proposed
Approaches to the Derivation of a Draft Maximum Contaminant Level Goal for
Perflaorooctanoic Acid (PFOA) (CASRN 335-67-1) in Drinking Water {U.S. EPA, 2021,
10428559}. This assessment described the systematic review of cancer and noncancer health
effects, the derivation of candidate oral cancer and noncancer toxicity values, a relative source
contribution (RSC), and cancer classification, which would subsequently be used to prepare draft
and final toxicity assessments for PFOA. The agency sought peer review from the EPA SAB
PFAS Review Panel on key scientific issues, including the systematic review approach for
1-7
-------
APRIL 2024
evaluating health effects studies, the derivation of oral toxicity values, the RSC, and the cancer
classification for PFOA.
The SAB provided draft recommendations on June 3, 2022, and final recommendations on
August 23, 2022 {U.S. EPA, 2022, 10476098}. To be responsive to the SAB recommendations,
EPA developed a detailed response to comments document {USEPA OW, 2023, 11396269} and
addressed every recommendation from the SAB in the development of the Public Comment
Draft Toxicity Assessment and Proposed Maximum Contaminant Level Goal for
Perfluorooctanoic Acid (PFOA) in Drinking Water {U.S. EPA, 2023, 10841009}. Briefly, EPA:
• 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 across health outcomes;
• 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 discussion for cancer effects and rationale for the
cancer classification;
• strengthened the rationales for selection of PODs for the noncancer health outcomes; and
• clarified language related to the RSC determination, including the relevance of drinking
water exposures and the relationship between the RfD and the RSC.
EPA then released the Public Comment Draft Toxicity Assessment and Proposed Maximum
Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) in Drinking Water for a 60-day
public comment period. This assessment described the systematic review of cancer and
noncancer health effects, the derivation of candidate oral cancer and noncancer toxicity values,
an RSC, and cancer classification for PFOA.
EPA incorporated feedback from public comment into this final assessment and developed a
detailed response to public comment document {USEPA, 2024, 11414326}. Briefly, EPA has
improved descriptions of rationale and added clarifications related to the systematic review
protocol used for this assessment, study and endpoint selection for POD derivation, and the
modeling choices related to toxicity value derivation. Therefore, this Final Human Health
Toxicity Assessment for Perfluorooctanoic Acid (PFOA) and Related Salts incorporates feedback
from external peer review and public comment and supersedes all other health effects documents
produced by the EPA Office of Water for PFOA.
1-8
-------
APRIL 2024
2 Summary of Assessment Methods
This section summarizes the methods used for the systematic review of the health effects
literature for all isomers of perfluorooctanoic acid (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). The purposes of this systematic review were to identify the best available and
most relevant health effects literature, to evaluate studies for quality, and to subsequently
identify health effects and studies for dose-response assessment. A detailed description of these
methods is provided as a protocol in Appendix A {U.S. EPA, 2024, 11414343}.
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 {U.S. EPA, 2022, 11181062; U.S. EPA, 2023,
11133619}. Although the IRIS Handbook was finalized concurrently with the development of
this assessment, the revisions in the final IRIS Handbook compared with the draft version do not
conflict with the methods used in this assessment. The assessment team concluded that
implementing minor changes in study quality evaluation between the draft and final IRIS
Handbook versions 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. Additionally, the methods used to conduct the systematic review are also consistent
with and largely mirror the Systematic Review Protocol for the PFBA, PFHxA, PFHxS, PFNA,
and PFDA (anionic and acidforms) IRIS Assessments {U.S. EPA, 2020, 8642427}.
For this updated PFOA toxicity assessment, systematic review methods were consistent with
those in the IRIS Handbook {U.S. EPA, 2022, 10367891} and the Systematic Review Protocol
for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (anionic and acid forms) IRIS Assessments
{U.S. EPA, 2020, 8642427}. for the steps of literature search; screening; study quality
evaluation; data extraction; display of study evaluation results; synthesis of human and
experimental animal data; and evidence integration for all health outcomes through the 2020
literature searches, as presented in the preliminary analyses of the 2021 Proposed Approaches to
the Derivation of a Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA)
(CASRN 335-67-1) in Drinking Water draft document that was reviewed by the Science
Advisory Board (SAB) {U.S. EPA, 2021, 10428559; U.S. EPA, 2022, 10476098}. The EPA
then focused the remaining steps of the systematic review process (synthesis and integration of
mechanistic data; derivation of toxicity values) on health outcomes with the strongest weight of
evidence based on the conclusions presented in the 2021 draft documents, and consistent with
the recommendations of the SAB {U.S. EPA, 2022, 10476098}. These five "priority" health
outcomes are developmental, hepatic, immune, cardiovascular, and cancer. The updated
2-1
-------
APRIL 2024
systematic review focused on the priority health outcomes was published in 2023 as the Public
Comment Draft Toxicity Assessment and Proposed Maximum Contaminant Level Goal for
Perflaorooctanoic Acid (PFOA) in Drinking Water {U.S. EPA, 2023, 10841009}.
The following subsections provide a summary of methods used to search for and screen
identified literature, evaluate the identified studies to characterize study quality, extract data, and
select studies for dose-response analysis. Extracted data are available in interactive visual
formats (see Section 3) and can be downloaded in open access, interactive formats. The full
systematic review protocol (see Appendix A, {U.S. EPA, 2024, 11414343}) provides a detailed
description of the systematic review methods that were used. The protocol also includes the
description of the problem formulation and key science issues guiding this assessment.
2.1.1 Literature Database
The EPA assembled a database of epidemiological, animal toxicological, mechanistic, and
toxicokinetic studies for this PFOA toxicity assessment based on three main data streams: 1)
literature published from 2013 through February 6, 2023 identified via literature searches
conducted in 2019, 2020, 2022 and 2023 of a variety of publicly available scientific literature
databases, 2) literature identified via other sources (e.g., searches of the gray literature, studies
shared with EPA by the SAB, studies submitted through public comment), and 3) literature
identified in EPA's 2016 Health Effects Support Document for Perflaorooctanoic Acid (PFOA)
{U.S. EPA, 2016, 3603279}. All of these streams are described in detail below.
For the literature searches, the search strings focused on the chemical name (PFOA and its
related salts) with no limitations on lines of evidence (i.e., human/epidemiological, animal, in
vitro, in silico) or health outcomes. The EPA conducted a 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 prior to SAB review of the draft assessment (2020
literature search), a third search covering September 2020 through February 3, 2022 prior to
release of the draft assessment for public comment (2022 literature search), and a final
supplemental search covering February 4, 2022 through February 6, 2023.
The publicly available databases listed below were searched for literature containing the
chemical search terms outlined in Appendix A {U.S. EPA, 2024, 11414343}:
• 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).
The search strings and literature sources searched are described in Appendix A {U.S. EPA, 2024,
11414343}).
For the second data stream, other review efforts and searches of publicly available sources were
used to identify relevant studies (see Appendix A, {U.S. EPA, 2024, 11414343}), 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 {ATSDR, 2021, 9642134} and
2-2
-------
APRIL 2024
California Environmental Protection Agency {CalEPA, 2021, 9416932}),
• Studies identified during mechanistic or toxicokinetic evidence 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),
• Studies identified by the SAB in their final report dated August 23, 2022 {U.S. EPA,
2022, 10476098}, and
• Studies submitted through public comment by May 2023
(https://www.regulations.gov/docket/EPA-HQ-OW-2022-0114).
For the third data stream, EPA relied on epidemiological and animal toxicological literature
synthesized in the 2016 PFOA HESD to identify studies relevant to the five priority health
outcomes, as recommended by SAB and consistent with preliminary conclusions from EPA's
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 {U.S. EPA,
2021, 10428559}. The 2016 PFOA HESD contained a summary of all relevant literature
identified in searches conducted through 2013. EPA's 2016 PFOA 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 epidemiological
studies that were identified and presented in the 2016 PFOA HESD for the five priority health
outcomes. It also includes "key" animal toxicological studies from the 2016 PFOA HESD, which
includes studies that were selected in 2016 for dose-response modeling. The details of the studies
included from the 2016 PFOA HESD are described in Appendix A {U.S. EPA, 2024,
11414343}.
All studies identified through the data streams outlined above were uploaded into the publicly
available Health and Environmental Research Online (HERO) database
(https://hero.epa.gov/hero/index.cfm/proiect/page/proiect id/2608).
EPA has continued to monitor the literature published since February 2023 for other potentially
relevant studies. Potentially relevant studies identified after February 2023 that were not
recommended by the SAB in their final report or via public comment 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 Appendix A, {U.S. EPA,
2024, 11414343}).
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, the EPA used
populations, exposures, comparators, and outcomes (PECO) criteria to screen the literature
identified from the literature sources outlined above in order to prioritize studies for dose-
response assessment and to identify studies containing supplemental information such as
mechanistic studies that could inform the mode of action analyses. The PECO criteria used for
screening the health effects, toxicokinetic, and mechanistic literature are provided in Appendix A
{U.S. EPA, 2024, 11414343}.
Consistent with the IRIS Handbook {U.S. EPA, 2022, 10367891} and the Systematic Review
Protocol for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (anionic and acidforms) IRIS
2-3
-------
APRIL 2024
Assessments {U.S. EPA, 2020, 8642427}, 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 to human health risk assessment. Studies
captured then underwent title and abstract screening by at least two independent reviewers using
screening tools consistent with the IRIS Handbook ({U.S. EPA, 2022, 10367891}; DistillerSR or
SWIFT ActiveScreener software), and studies that passed this initial screening underwent full-
text review by at least two independent reviewers. Health effects studies that met PECO
inclusion criteria following both title and abstract screening and full-text review underwent study
quality evaluation as described below (Section 2.1.3). Studies that were tagged as containing
relevant PBPK models were sent to the modeling technical experts for scientific and technical
review. 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 Appendix A, {U.S. EPA, 2024, 11414343}). Supplemental studies that were
identified as mechanistic or ADME during screening did not undergo study quality evaluation.
For the supplemental literature search conducted in 2023 and literature received through public
comment, studies were screened for relevancy and considered for potential impact on the toxicity
assessments for PFOA. Consistent with the IRIS Handbook {U.S. EPA, 2022, 10367891}, the
studies identified after February 3, 2022, including studies recommended via public comment,
were "considered for inclusion only if they [were] directly relevant to the assessment PECO
criteria and [were] expected to potentially impact assessment conclusions or address key
uncertainties" {U.S. EPA, 2022, 10367891}. For the purposes of this assessment, the EPA
defined impacts on the assessment conclusions as data from a study (or studies) that, if
incorporated into the assessment, have the potential to significantly affect (i.e., by an order of
magnitude or more) the final toxicity values (i.e., RfDs and CSFs) or alter the cancer
classification for PFOA (see Appendix A, {U.S. EPA, 2024, 11414343}).
2.1.3 Study Quality Evaluation for Epidemiological Studies and
Animal Toxicological Studies
Study quality evaluations were performed consistent with the IRIS Handbook {U.S. EPA, 2022,
10367891} and the Systematic Review Protocol for the PFBA, PFHxA, PFHxS, PFNA, and
PFDA (anionic and acidforms) IRIS Assessments {U.S. EPA, 2020, 8642427}. For study quality
evaluation of the PECO-relevant human epidemiological and animal toxicological studies
(i.e., studies identified in the four literature searches (all health outcomes for the 2019 and 2020
searches; the five priority health outcomes for the 2022 search; studies impacting assessment
conclusions within the five priority health outcomes for the 2023 search (see Appendix A, {U.S.
EPA, 2024, 11414343})), studies recommended by the SAB, studies recommended via public
comment that reported potentially significant results on one or more of the five priority health
outcomes, epidemiological studies from the 2016 PFOA HESD that reported results on one or
more of the five priority health outcomes, and key animal toxicological studies from the 2016
PFOA HESD), two independent primary reviewers followed by a quality assurance (QA)
reviewer assigned ratings about the reliability of study results {good, adequate, deficient (or "not
2-4
-------
APRIL 2024
reported"), or critically deficient) for different evaluation domains as described in the IRIS
Handbook {U.S. EPA, 2022, 10367891} (see Appendix A, {U.S. EPA, 2024, 11414343}). These
study quality evaluation domains are listed below and details about the domains, including
prompting questions and suggested considerations, are described in Appendix A {U.S. EPA,
2024, 11414343}.
• 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 quality; 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.oru/). 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 aninformative for each PECO-relevant endpoint
evaluated in the study consistent with the IRIS Handbook {U.S. EPA, 2022, 10367891}. A study
can be given an overall mixed confidence rating if different PECO-relevant endpoints within the
study receive different confidence ratings (e.g., medium and low confidence ratings).
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. Because
of the abundance of medium and high confidence studies in this database, data were only
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, 10367891}. Studies evaluated as being iminformative for an endpoint were not considered
further when characterizing that endpoint 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 summary 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 a second
reviewer conducting an independent QA 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
2-5
-------
APRIL 2024
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 Appendix A {U.S. EPA,
2024, 11414343}.
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 outcome (i.e., an organ- or organ system-
level category of related health effects and endpoints). 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. Consistent
with the IRIS Handbook, groups of related outcomes within a health outcome category were
considered together as a unit of analysis during evidence synthesis and evidence integration
{U.S. EPA, 2022, 10367891}. For example, birth weight, birth length, and head circumference
were all considered under the unit of analysis of the fetal growth restriction.
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}. Consistent with the IRIS Handbook
{U.S. EPA, 2022, 10367891}, 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. As described above, iminformative studies were not extracted or
included in the evidence syntheses. Results from epidemiological studies were discussed within
sections organized by population type, including children, general population adults, pregnant
women, and occupational populations. Childhood was defined as the effect of environmental
exposure during early life: from conception, infancy, early childhood and through adolescence
until 21 years of age {U.S. EPA, 2021, 9641727}. 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 on the
same health outcome as another study were considered overlapping evidence, and to avoid
duplication or overrepresentation of results from the same group of participants, these additional
studies were not discussed in the evidence synthesis narrative. In cases of overlapping studies,
the study with the largest number of participants and/or the most accurate outcome measures was
given preference. For the five priority health outcomes, EPA also developed mechanistic
syntheses.
For evidence integration, conclusions regarding the strength of evidence were drawn for each
health outcome across human and animal evidence streams. For the five priority health
2-6
-------
APRIL 2024
outcomes, this included consideration of epidemiological studies identified in the 2016 PFOA
HESD, as well as mechanistic 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 judgments
were 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}.
The evidence integration was conducted according to guidance outlined in the IRIS Handbook
{U.S. EPA, 2022, 10367891} 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. Considerations for each evidence
integration judgment are summarized within corresponding evidence integration sections in an
evidence profile table (EPT). EPTs were organized by evidence stream (i.e., human, animal, and
mechanistic, respectively), and, within evidence streams, units of analysis with the strongest
evidence were presented first.
Additional details about evidence synthesis and integration are summarized in Appendix A {U.S.
EPA, 2024, 11414343}.
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 within those
health outcomes. Dose-response modeling was performed for endpoints within health outcomes
with data warranting evidence integration conclusions of evidence demonstrates and evidence
indicates (likely) for noncancer endpoints and carcinogenicity descriptors of Carcinogenic to
Ramans and Likely to be Carcinogenic to Ramans. EPA identified specific studies for dose-
response modeling and POD derivation following attributes described in Table 7-2 of the IRIS
Handbook {U.S. EPA, 2022, 10367891}. Examples of study attributes evaluated included study
design characteristics, study confidence, and data availability, among others (see Appendix A,
{U.S. EPA, 2024, 11414343}). Human epidemiological and animal toxicological studies that
were consistent with the overall weight of evidence for a specific endpoint were considered for
dose-response. Additionally, for human evidence, all high or medium confidence studies
pertaining to a specific endpoint were considered; for animal evidence, only animal toxicological
studies with at least two PFOA exposure groups that were of high or medium confidence were
considered. Relevance of the endpoint or species reported by animal toxicological studies to
2-7
-------
APRIL 2024
human health effects was also considered. Additional information on study selection is provided
in Appendix A {U.S. EPA, 2024, 11414343}.
2.2.1 Approach to POD and Candidate RfD Derivation for
Noncancer Health Outcomes
The current recommended EPA human health risk assessment approach for noncancer POD
derivation described in EPA's A Review of the Reference Dose and Reference Concentration
Processes includes selection of a benchmark response (BMR), analysis of dose and response
within the observed dose range, followed by extrapolation to lower exposure levels {U.S. EPA,
2002, 88824}. For noncancer health outcomes, EPA performed dose-response assessments to
define PODs, including low-dose extrapolation, when feasible, and applied uncertainty factors
(UFs) to those PODs to derive candidate RfDs. An RfD is an estimate, with uncertainty spanning
perhaps an order of magnitude, of an exposure to the human population (including susceptible
subgroups) that is likely to be without an appreciable risk of deleterious health effects over a
lifetime {U.S. EPA, 2002, 88824}. For PFOA, multiple candidate RfDs were derived within a
health outcome as described in Section 4.
For PFOA animal toxicological studies, EPA attempted benchmark dose (BMD) modeling on all
studies considered for dose response to refine the POD. BMD modeling was performed after
converting the administered dose reported by the study to an internal dose using a
pharmacokinetic model (see Section 4.1.3 for additional details). This approach resulted in dose
levels corresponding to specific response levels near the low end of the observable range of the
data and identified the lower limits of the BMDs (BMDLs) which serve as potential PODs {U.S.
EPA, 2012, 1239433}. EPA used the publicly available Benchmark Dose Software (BMDS)
program developed and maintained by EPA (https://www.epa.gov/bmds). BMDS fits
mathematical models to the data and determines the dose (i.e., BMD) that corresponds to a
predetermined level of response (i.e., 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. However, a POD derived
using a BMD approach typically provides a higher level of confidence in the conclusions for any
individual case, as the BMDL takes into account all the data from the dose-response curve,
incorporates the evaluation of the uncertainty in the BMD, and is related to a known and
predefined potential effect size (i.e., the BMR) {U.S. EPA, 2012, 1239433; U.S. EPA, 2022,
10367891}. For noncancer endpoints, there were several factors considered when selecting the
final model and BMD/BMDL, including the type of measured response variable
(i.e., dichotomous or continuous), experimental design, and covariates {U.S. EPA, 2012,
1239433}. However, as there is currently no prescriptive hierarchy, selection of model types was
often based on the goodness-of-fit and judged based on the x2 goodness-of-fit p-value (p > 0.1),
magnitude of the scaled residuals in the vicinity of the BMR, and visual inspection of the model
fit. The Benchmark Dose Technical Guidance provides a "BMD Decision Tree" to assist in
model selection {U.S. EPA, 2012, 1239433}. See Appendix E {U.S. EPA, 2024, 11414343} for
additional details on the study-specific modeling.
2-8
-------
APRIL 2024
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} because the EPA was
able to define a level considered clinically adverse for these outcomes (see Appendix E, {U.S.
EPA, 2024, 11414343}). As sensitivity analyses for comparison purposes, EPA also performed
BMD modeling and provided study LOAELs/NOAELs as PODs for the epidemiological hepatic
and serum lipid dose-response studies. For the immune studies, for which a clinically defined
adverse level is not well established, 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 Appendix E {U.S. EPA, 2024, 11414343} for additional details on the
study-specific modeling.
After POD derivation, EPA used a pharmacokinetic model for human dosimetry to estimate
human equivalent doses (HEDs) from both animal and epidemiological studies. A
pharmacokinetic model for human dosimetry is used to simulate the HED from the animal PODs
and is also used to simulate selected epidemiological studies to obtain a chronic dose that would
result in the internal dose 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. Next, reference values are estimated by
applying relevant adjustments to the point-of-departure human equivalent doses (PODheds) to
account for five possible areas of uncertainty and variability: human variation, extrapolation
from animals to humans, extrapolation to chronic exposure duration, the type of POD being used
for reference value derivation, and extrapolation to a minimal level of risk (if not observed in the
data set). UFs used in this assessment were applied according to methods described in EPA's
Review of the Reference Dose and Reference Concentration Processes {U.S. EPA, 2002,
88824}. For additional detail on UFs see Appendix A {U.S. EPA, 2024, 11414343}. The
PODhed for a particular candidate RfD is divided by the composite UFs.
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 confidence, 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, magnitude of response, adversity of effect, and
consistency across studies.
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) that would correspond to the administered dose used in the study
(see 4.1.3 for additional detail). A number of dose-metrics across lifestages are selected for
simulation in a mouse, rat, or monkey. Concentrations of PFOA in blood are considered for all
the internal dose-metrics.
2-9
-------
APRIL 2024
Step 2: Conduct dose-response modeling. See above and Appendix E {U.S. EPA, 2024,
11414343} 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 (e.g., BMDL, NOAEL) is converted to an HED following
the method described in Section 4.1.3.
Step 4: Select appropriate UFs and provide rationale for UF selection. UFs are applied in
accordance with EPA methodology considering variations in sensitivity among humans,
differences between animals and humans (if applicable), the duration of exposure in the critical
study compared with the lifetime of the species studied, and the completeness of the
epidemiological or animal toxicological database {U.S. EPA, 2002, 88824}.
Step 5: Calculate the chronic RfD. The RfD is calculated by dividing the PODhed by the
composite (total) UF (UFC) specific to that PODhed.
PODhed = calculated from the internal dose POD 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 with the lifetime of the species studied, and completeness of the
toxicology database, in accordance with EPA methodology {U.S. EPA, 2002, 88824}.
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
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
where:
2.2.2 Cancer Assessment
2.2.2.1 Approach for Cancer Classification
2-10
-------
APRIL 2024
More than one carcinogenicity descriptor can be applied if a chemical's carcinogenic effects
differ by dose, exposure route, or mode of action (MOA)3. 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. The MOA analysis must be conducted separately for each target organ/tissue type {U.S.
EPA, 2005, 9638795}.
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 as a CSF. A CSF is a plausible upper bound lifetime cancer risk from
chronic ingestion of a chemical per unit of mass consumed per unit body weight per day (mg/kg-
day) {U.S. EPA, 2005, 9638795}. This process varied slightly depending on whether the CSF
was based on an animal toxicological or epidemiological study, as described below.
The first step in the process is using a model to fit a dose-response curve to the data, based on the
doses and associated tumors observed {U.S. EPA, 2005, 9638795}. 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
{U.S. EPA, 2005, 9638795}.
For animal toxicological studies, EPA used the publicly available Benchmark Dose Software
(BMDS) program developed and maintained by EPA (https://www.epa.gov/bmds). First, a PK
model converted the administered dose reported by the study to an internal dose (see Section
4.1.3 for additional details). Then, BMDS fits multistage models, the preferred model type {U.S.
EPA, 2012, 1239433}, to the data 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 for that dose) are also reported and are
often used as the POD. Similar to noncancer PODs, selection of model types is often based on
the goodness-of-fit {U.S. EPA, 2012, 1239433}. For PFOA, after a POD was determined, a PK
model was used to calculate the HED for animal oral exposures (PODhed). The CSF is derived
by dividing the BMR by the PODhed. See Appendix E {U.S. EPA, 2024, 11414343} for
additional details on the study-specific modeling.
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
3MOA 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.
2-11
-------
APRIL 2024
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. 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. 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 Appendix E {U.S. EPA, 2024, 11414343} for additional details on the study-
specific modeling.
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
an early-life exposure compared with the same 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 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" {U.S. EPA, 2005, 9638795}. In these cases, an
RfD-like value is calculated based on the key event4 for carcinogenesis or the tumor response.
2.2.3 Selecting Health Outcome-Specific and Overall Toxicity
Values
Once all of the candidate toxicity values were derived, EPA then selected a health outcome-
specific toxicity value for each hazard (cancer and noncancer) identified in the assessment. This
selection can be based on the study confidence considerations, the most sensitive outcome, a
clustering of values, or a combination of such factors; the rationale for the selection is presented
in the assessment. Key considerations for candidate value selection are described in the IRIS
Handbook {U.S. EPA, 2022, 10476098} and include: 1) the weight of evidence for the specific
effect or health outcome; 2) study confidence; 3) sensitivity and basis of the POD; and 4)
uncertainties in modeling or extrapolations. The value selected as the organ/system-specific
toxicity value is discussed in the assessment.
4The 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.
2-12
-------
APRIL 2024
The selection of overall toxicity values for noncancer and cancer effects involves the study
preferences described above, consideration of overall toxicity, study confidence, and confidence
in each value, including the strength of various dose-response analyses and the possibility of
basing a more robust result on multiple data sets. The values selected as the overall RfD and CSF
are discussed in Section 4.
2-13
-------
APRIL 2024
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.gov/.
The three database searches yielded 7,160 unique records (combined for PFOA and PFOS) prior
to running SWIFT Review. Table 3-1 shows the results from database searches conducted in
April 2019, September 2020, February 2022, and February 2023.
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
2/6/2023: 966 results
PubMed
4/10/2019: 2,191 results
9/3/2020: 811 results
2/2/2022: 1,728 results
2/6/2023: 719 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
2/2023: 1,153 results
Total number of references after running SWIFT Review3
4/2019: 1,977 results
9/2020: 867 results
2/2022: 1,370 results
2/2023: 881 results
Total number of unique references moved to screeningb
4,802
Notes:
a Hie 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 epidemiological, mechanistic, or toxicokinetic syntheses,
studies identified by the Science Advisory Board (SAB), and EPA's 2016 Health Effects Support
Documents (HESDs) for perfluorooctanoic acid (PFOA) {U.S. EPA, 2016, 3603279} and
perfluorooctane sulfonate (PFOS) {U.S. EPA, 2016, 3603365}) yielded 238 unique records
(combined for PFOA and PFOS).
3-1
-------
APRIL 2024
The 4,802 studies captured with the SWIFT Review evidence streams filters and the 238 records
identified from additional sources yielded a total of 5,011 unique studies. These 5,011 studies
were moved to the next stage of screening (title and abstract screening using either DistillerSR or
SWIFT ActiveScreener). Of the 5,011 unique studies, 1,062 moved on to full-text level review,
1,697 were excluded during title and abstract screening, and 2,252 were tagged as containing
potentially relevant supplemental material. Of the 1,062 screened at the full-text level, 784 were
considered to meet population, exposure, comparison, outcome (PECO) eligibility criteria (see
Appendix A, {U.S. EPA, 2024, 11414343}) and included relevant information on PFOA. The
784 studies that were determined to meet PECO criteria after full-text level screening included
451 epidemiological (human) studies, 40 animal toxicological studies, 15 physiologically based
pharmacokinetic (PBPK) studies (2 of which were also relevant epidemiological studies), and
280 studies that were not extracted (e.g., low confidence studies, meta-analyses, studies from the
2022 and 2023 searches 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 35
PBPK studies. Details of the literature search and screening process are shown in Figure 3-1.
The 451 epidemiological studies and 40 animal toxicological studies relevant to PFOA
underwent study quality evaluation and were subsequently considered for data extraction as
outlined in Sections 2.1.3 and 2.1.4 (see Appendix A, {U.S. EPA, 2024, 11414343}). The results
of the health outcome-specific study quality evaluations and data extractions are described in
Sections 3.4 and 3.5.
Additionally, the 35 studies tagged as containing relevant PBPK models relevant to PFOA were
reviewed by pharmacokinetic (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 129 toxicokinetic and 273 mechanistic studies identified as relevant for PFOA
moved on to a limited data extraction as described in the Appendix {U.S. EPA, 2024,
11414343}. The toxicokinetic studies pertaining to ADME are synthesized in Section 3.3.1. The
mechanistic studies relevant to the five priority health outcomes are synthesized in Sections 3.4
and 3.5 and were considered as part of the evidence integration.
In addition to the studies identified through database searches and the other sources outlined
above, public comments submitted in response to the Public Comment Draft Toxicity Assessment
and Proposed Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) in
Drinking Water {U.S. EPA, 2023, 10841009} included 944 references relevant to PFOA and/or
PFOS, which were reviewed for relevance to the toxicity assessment. Of the 944 studies, 297
were duplicates of studies included in the toxicity assessment and 31 were duplicates of studies
included in the 2016 PFOA or PFOS HESD assessment. The 599 studies that were not identified
in the HESDs and were not included in the toxicity assessments underwent additional review to
identify studies with that could impact assessment conclusions as outlined in Appendix A.3
{U.S. EPA, 2024, 11414343}. Ultimately, none of the 599 studies were incorporated in the
toxicity assessments upon further screening. The submitted references were either deemed not
relevant after secondary review, were supplemental studies (e.g., PFOA or PFOS assessments
published by other scientific bodies, mechanistic, ADME, etc), or were already included in the
PFOA or PFOS toxicity assessments. Additionally, several references reported information on
3-2
-------
APRIL 2024
PFOA or PFOS and non- priority health outcomes and were therefore not included. The results
of this screening can be found in the docket ("Review of Public Comment References Related to
PFOA and PFOS Ffealth Effects;" https://www.regulations.gov/docket/EPA-HQ-OW-2Q22-
0114).
References identified through
database search
2013 to 2019: 3,382
2019 to 2020:
1,153
2020 to 2022:
1,858
2022 to 2023:
1,153
References after
SWIFT filters applied
2013 to 2019:
1,977
2019 to 2020:
867
2020 to 2022:
1,370
2022 to 2023:
881
References screened for
PFOA and PFOSb
5,011
References assessed for eligibility
for PFOA and PFOS
1,062
Relevant references for PFOAc
Key references identified from
2016 PFOA & PFOS HESDs
References identified through
other sources3
151
References excluded
1,697
^ Not PECO Relevant
810
Excluded by SWIFT-Active
887
Supplemental Tag
Mechanistic
Toxicokinetic
Other Supplemental
SWIFT Supplemental
Relevant to PFOS only
2,252
525
244
1,309
629
83
References excluded 11
Not PECO Relevant 11
Supplemental Tag
Mechanistic
Toxicokinetic
Other Supplemental
184
78
21
149
Toxicokinetic/Mechanistic g46
references assessed for eligibility01
Mechanistic 657
Toxicokinetic 294
I Til
1
Did Not Extract*
Human
Animal
PBPK9
Toxicokinetic
Mechanistic
280
451
40
35
129
273
Figure 3-1. Summary of Literature Search and Screening Process for PFOA
Interactive figure and additional study details available on HAWC.
Interactive figure based on work by Magnuson et al. {, 2022, 10442900}.
3-3
-------
APRIL 2024
"Other sources" include assessments published by other agencies, studies identified during epidemiological, mechanistic, or
toxicokinetic syntheses, and studies identified by the SAB.
a References identified by SAB and through database searches were counted as identified through database search only.
b Includes number of unique references after deduplication of studies captured with the SWIFT Review evidence streams filters
and records identified from additional sources.
c Includes number of unique references considered to meet PECO eligibility criteria at the full-text level and include relevant
information on PFOA.
d Includes number of unique references identified during title/abstract screening, full-text screening, and data extraction assessed
for toxicokinetic and/or mechanistic eligibility.
e Only includes references with relevant information on PFOA.
f References tagged to 'Not a priority human health system' include those identified in the 2019 search that overlap with 2016
PFOA HESD references or those identified in 2022 and 2023 searches.
g Includes 15 PBPK references (2 of which were also relevant epidemiological references) determined to meet PECO criteria plus
an additional 20 PBPK references identified during the toxicokinetic screening.
3.1.1 Results for Epidemiology Studies of PFOA by Health
Outcome
Of the 451 epidemiological studies that met the inclusion criteria and underwent extraction, 193
had a cohort study design, 177 had a cross-sectional design, 42 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 = 96), developmental
(n = 92), metabolic (n = 78), or immune systems (n = 68). 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
8
6
3
5
22
Cardiovascular
5
24
60
7
96
Dermal
0
1
0
0
1
Developmental
4
61
20
7
92
Endocrine
1
8
18
8
35
Gastrointestinal
1
6
0
0
7
Hematologic
0
0
7
1
8
Hepatic
1
7
20
4
32
Immune
5
35
19
9
68
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
42
193
177
39
451
Figure 3-2. Summary of Epidemiology Studies of PFOA Exposure by Health System and
Study Design3
Interactive figure and additional study details available on HAWC.
3-4
-------
APRIL 2024
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 40 animal toxicological studies that met the inclusion criteria and underwent extraction,
most studies had either short-term (n = 16) or developmental (n = 16) study designs and most
were conducted in mice (n = 33). The mouse studies had developmental (n = 16), short-term
(n = 15), 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 health outcomes. Most studies reported results for the hepatic (n = 30),
whole-body (n = 25; i.e., systemic effects such as bodyweight), reproductive (n = 19), or
developmental (n = 15) 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 1
0
4
Cardiovascular
2
2
0
0
0
2
3
0
8
Developmental
0
0
0
0
0
1
13
1
15
Endocrine
3
2
0
0
0
3
3
1
11
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
11
1
30
Immune
5
2
2
0
0
2
2
1
13
Metabolic
0
1
0
0
0
2
3
6
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
9
1
19
Respiratory
0
1
0
0
0
1
0
2
Whole Body
2
2
1
0
3
7
1
25
Grand Total
15
2
2
1
1 3
16
1
40
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 HAWC.
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
All data from this project are available in the public HAWC site
(https://hawc.epa.gov/assessment/10050Q248/) displayed as exposure-response arrays, forest
plots, and evidence maps. Data extracted from the 451 epidemiological studies are available
here. Data extracted from the 40 animal toxicological studies are available here. 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 are also available in HAWC.
3-5
-------
APRIL 2024
3.3 Toxicokinetic Synthesis
As described in Section 3.1, EPA identified 129 and 35 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 {U.S. EPA, 2024, 11414343}.
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
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 modes of action (MOAs) 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}.
3-6
-------
APRIL 2024
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 Appendix, {U.S. EPA, 2024,
11414343}). In vitro absorption data indicate that uptake is influenced by pH, temperature, and
concentration as well as OATP activity (see Appendix B, {U.S. EPA, 2024, 11414343}).
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 the role of transporters. Cellular transfection of cells
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.
3-7
-------
APRIL 2024
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 versus 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
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
{Hinderliter, 2006, 135732} and dermal routes {Fasano, 2005, 3749187; O'Malley, 1981,
4471529; Kennedy, 1985, 3797585}.
3.3.1.2 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 Appendix B {U.S. EPA, 2024, 11414343}. 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 et al. {,
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-
3-8
-------
APRIL 2024
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 versus cord blood, lower cord blood albumin levels compared with
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.
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.
3-9
-------
APRIL 2024
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 nine 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,
2816710; Wang, 2018, 5080654; Maestri, 2006, 1048866}, major organs including lung and
kidney {Maestri, 2006, 1048866}, endocrine tissues including the thyroid gland, pituitary gland,
and pancreas {Pirali, 2009, 757881; Maestri, 2006, 1048866}, and gonads and follicular fluid
{Kang, 2020, 6356899; Maestri, 2006, 1048866}. 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. Maestri et al. {,
2006, 1048866} measured pooled post-mortem tissue samples and found the highest levels in
lung (3.8 ng/g), kidney (3.5 ng/g), and liver (3.1 ng/g). 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}. Several studies examined a limited number of tissues in primates
and observed higher levels in serum compared with liver {Butenhoff, 2002, 1276161; Butenhoff,
2004, 3749227; Griffith, 1980, 1424955}.
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 with 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.
3-10
-------
APRIL 2024
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 [j,mol/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}.
Fujii and colleagues {, 2015, 2816710} observed that perfluoroalkyl carboxylic acids (PFCAs)
(C6 and C7) were excreted relatively 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
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}.
3-11
-------
APRIL 2024
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 Appendix, {U.S. EPA, 2024,
11414343}) 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 with 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
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.
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 versus 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 with other lipophilic persistent organic pollutants such
as polychlorinated biphenyls.
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.
3-12
-------
APRIL 2024
{, 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.
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 Appendix B, {U.S. EPA, 2024, 11414343}). Vd values may be influenced by
differences in distribution between males and females, between pregnant and nonpregnant
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 5.6.2.
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 literature reviewed for this assessment
do not provide evidence that PFOA is 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 Appendix B, {U.S. EPA, 2024,
11414343}). 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 with females and in rodents
compared with humans. Nevertheless, a comprehensive set of principles governing resorption by
3-13
-------
APRIL 2024
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 |iinol/kg) or gavage
(3.13 (j,mol/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.
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 with
males and compared with 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
3-14
-------
APRIL 2024
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.
Zhao et al. {,2017, 3856461} demonstrated that PFOA was a substrate for human OATP IB 1,
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 (C7). 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 for PFOS {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 lifestage 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
nonpregnant 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}.
3-15
-------
APRIL 2024
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.
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 lifestage.
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, a reanalysis 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
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
3-16
-------
APRIL 2024
weeks (see Appendix B, {U.S. EPA, 2024, 11414343}). 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
with 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
Appendix B, {U.S. EPA, 2024, 11414343}). 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-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 Appendix B, {U.S. EPA, 2024, 11414343}). 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 with
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 nonurinary 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}. The multiple routes of
exposure to PFOA and the need to understand historical exposure levels to estimate PFOA intake
is an ongoing challenge for many studies that examine PFOA elimination. 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
3-17
-------
APRIL 2024
estimates were shorter in adult female rats compared with 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 with linear forms.
3.3.2 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
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
3-18
-------
APRIL 2024
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
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 concentrations 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. Because of 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.
3-19
-------
APRIL 2024
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 toxicity 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 PFOA
HESD. 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 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 |iinol/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
3-20
-------
APRIL 2024
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
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 with
adults, under the assumption that PFAS distribution into tissues, quantified by the Vd, will
increase proportionally to intracellular water content. This lifestage difference in intracellular
water content may explain why the ratio of PFOA in cord blood versus 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).
Pharmacokinetic models that can accommodate longer half-life values than would be predicted
based on standard ADME concepts and allow for dose-dependent changes in excretion rate
compared with the classic one- or two-compartment approaches have been published as tools to
estimate internal doses for humans, monkeys, mice, and rats {Andersen, 2006, 818501;
Wambaugh, 2013, 2850932; Loccisano, 2011, 787186; Loccisano, 2012, 1289830; Loccisano,
3-21
-------
APRIL 2024
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.
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}.
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. On the basis of 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.
3-22
-------
APRIL 2024
Second
Compartment
(Vf Ctjssue)
Oral Dose
(Agut)
Filtrate
Compartment
(V«, Cm)
~
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
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 versus
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.
3-23
-------
APRIL 2024
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 no-observed-adverse-effect level
(NOAEL) and lowest-observed-adverse-effect level (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 the amount and concentration of the substance can be predicted in
specific tissues, in addition to blood. This can be valuable for certain endpoints for which it is
expected that a tissue concentration would better reflect the relevant dosimetry compared with
blood concentration.
The first PBPK model developed for PFOA 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 two-
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
3-24
-------
APRIL 2024
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 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 lifestages 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
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 PFOA HESD {U.S. EPA, 2016, 3603279} for a more in-
depth review of PFOA PBPK models.
3.4 Noncancer Health Effects Evidence Synthesis and
Integration
3.4.1 Hepatic
EPA identified 33 epidemiological studies (reported in 39 publications)5 6 and 31 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
5 Multiple publications of the same data: Jain and Ducatman {, 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 used NHANES data from overlapping years.
0 Olsen {,2003,1290020} is the peer-review paper of Olsen {,2001, 10228462} and Olsen {,2001, 10240629}; however, data
for PFOA and hepatic outcomes is reported in Olsen {,2001, 10228462}.
3-25
-------
APRIL 2024
mixed {medium low) confidence, and 9 were considered iminformative (Section 3.4.1.1). Of the
31 animal toxicological studies, 5 were classified as high confidence, 22 as medium confidence,
2 as low confidence, and 2 were considered mixed {medium iminformative and
medium low iminformative) (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 13 epidemiological studies (14 publications)6 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. Emmett et al. {, 2006, 1290905} and Jain
et al. {, 2014, 2969807} were rated as iminformative and will not be further discussed. Nine out
of the 12 remaining studies were rated as medium quality and all investigated changes in serum
liver enzymes. Results from studies summarized in the 2016 PFOA HESD are described in
Table 3-2 and below.
3-26
-------
APRIL 2024
0* e e^^C0
^ .vP.c^ .>c\Se «*&
Costa etal.. 2009, 1429922-
Emmett et al.. 2006, 1290905
Galloetal., 2012, 1276142
Jain, 2014, 2969807-
Lin et al., 2010, 1291111 -
Olsen and Zobel, 2007, 1290836-
Olsen et al., 2000, 1424954 -
Olsen et al2001, 10228462 -
Olsen etal., 2003, 1290020
Sakretal., 2007, 1291103 -
Sakretal., 2007, 1430761 -
Steenland and Woskie, 2012, 2919168-
Steenland et al., 2015, 2851015 -
Yamaguchi et al., 2013, 2850970
+
+
+
+
+
+
-
~
-
-
+
B
-
*P
+
+
++
+
+
+
* ~
+
+
+
-
-
+
* M
+
+
+
+
+
+
+
+
-
+
+
+
-
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
+
-
+
+
+
+
+*
+
+
-
+
+
+
+*
-
+
-
+
++
+
+
-
-
+
++
+
-
+
+
+
Legend
Q
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
p
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
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 PFOA
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.
3-27
-------
APRIL 2024
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 In-ALT in linear
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}.
3-28
-------
APRIL 2024
Confidence ^ Matrix^ Design Exposure Levels Sub-population Comparison EE
Effect Estimate
0 12 3 4
Gallo et al. (2012. Serum Cross- Median=28.0 ng/mL (25th^ - Regression coefficient (per
1276142), sectional 75th percentile: 13.5-70.8 Hn ng/mL increase in
Medium ng/mL) PFOA) 002
1
l
1
1
Gteason et al. Serum Cross - Median=3.7 ng/mL - Regression coefficient (per
(2015,2966740), sectional (25th-75th percentile: 2.5 - 1-In ng/mL increase in
Medium 5.2 ng/mL) PFOA) 0 04
1
k
1
l
Lin et al. (2010, Serum Cross- Mean: 4.06 ng/mL (standard Women Regression coefficient (per
1291111), sectional error: 1.04 ng/mL) 1-log ng/mL increase in
Medium PFOA) 187
l
I •
l
1
Mean: 4.22 ng/mL (standard Ages >=60 years Regression coefficient (per
error 1.04 ng/mL) 1-log ng/mL increase in
PFOA) 193
1
1
< •
1
l
Mean: 4.48 ng/mL (standard Ages 18-39 years Regression coefficient (per
error 1.03 ng/mL) 1-log ng/mL increase in
PFOA) 102
i
< •
l
1
Mean: 4.71 ng/mL (standard Ages 40-59 years Regression coefficient (per
error 1.04 ng/mL) Hog ng/mL increase in
PFOA) 183
1
1
•
1
1
Mean: 5.05 ng/mL (standard Men Regression coefficient (per
error 1.03 ng/mL) 1-log ng/mL increase in
PFOA) 155
1
l
•
l
1
Median: 4.20 ng/mL 11 Regression coefficient (per
(25th-75th percentile: 1-log ng/mL increase in
2.90-5.95 ng/mL) PFOA) 4 05
1
1
•
1
1
Olsen et al. (2000, Serum Cross- 1993 Median (min-max): 1.1 - Regression coefficient (per
1424954), sectional ppm (0.0-80.0 ppm); 1995: 1 ppm increase in serum
Medium 1.2 ppm (0.0-114.1 ppm); PFOA) 4 47
1997: 1.3 ppm (0.1-81.3 pp..
l
¦ •
l
1
Sakr et al. (2007a, Serum Cohort Mean: 1.13 ppm (standard - Regression coefficient (per
1430761), deviation: 2.1 ppm) ppm increase in PFOA)
Medium ®-®4
1
1
1 m
i
i
Sakr et al. (2007b, Serum Cross - Mean (SD): 0.428 ppm - Regression coefficient (per
1291103), sectional (0.86); Min-max: 0.005-9.550 ppm increase PFOA)
Medium ppm 002
I
i
i
I |
Workers not on Regression coefficient (per
lipid-lowering ppm increase PFOA)
medications 003
i
b
i
i
0 12 3 4
Figure 3-6. Overall ALT Levels from 2016 PFOA HESD Epidemiology Studies Following
Exposure to PFOA
Interactive figure and additional study details available on HAWC.
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
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 nonmalignant 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} or elevated ALT {Gallo, 2012,
1276142} after PFOA exposure. Increases were also observed for AST and GGT, though less
consistently across the available studies.
3-29
-------
APRIL 2024
Table 3-2. Associations Between Elevated Exposure to PFOA and Hepatic Outcomes from Studies Identified in the 2016 PFOA
HESD
Reference, confidence
Study
Design
Population
ALT3
AST3
GGTa
ALPa
Liver Diseaseb
Serum Protein3
Albumin3
Costa, 2009, 1429922
Medium
Cross-
sectional
Occupational
tt
t
tt
tt
NA
1
1
Gallo, 2012, 1276142
Medium
Cross-
sectional
Adults
tt
NA
tt
NA
NA
NA
NA
Lin, 2010, 1291111
Medium
Cohort
Adults
tt
NA
tt
NA
NA
NA
NA
Olsen and Zobel, 2007,
1290836
Low
Cross-
sectional
Occupational
tt
4
tt
tt
NA
NA
NA
Olsen, 2003, 1290020
Medium
Cross-
sectional
Occupational
tt
-
t
NA
NA
NA
NA
Olsen, 2001, 10228462
Medium
Cohort
Occupational
t
t
4
t
NA
NA
NA
Olsen, 2000, 1424954
Medium
Cross-
sectional
Occupational
tt
NA
NA
NA
NA
NA
NA
Sakr, 2007, 1291103
Medium
Cross-
sectional
Occupational
t
t
tt
NA
NA
NA
NA
Sakr, 2007, 1430761
Medium
Cohort
Occupational
t
tt
t
NA
NA
NA
NA
Steenland and Woskie, 2012, Cohort
2919168
Mixed c
Occupational
NA
NA
NA
NA
NA
NA
Steenland, 2015, 2851015
Low
Cohort
Occupational
NA
NA
NA
NA
t
NA
NA
Yamaguchi, 2013, 2850970
Medium
Cross-
sectional
Adults and
adolescents
tt
tt
t
NA
NA
NA
NA
Notes'. ALP = alkaline phosphatase; ALT = alanine transferase; AST = aspartate transaminase; GGT = gamma-glutamyl transferase; NA = no analysis was for this outcome was
performed; | = nonsignificant positive association; ft = significant positive association; j = nonsignificant inverse association; jj = significant inverse association; - = no (null)
association.
Emmett et al., 2006,1290905 was not included in the table due to their uninformative overall study confidence ratings.
Jain et al., 2014,2969807 was not included in the table due to their uninformative overall study confidence ratings.
a Arrows indicate the direction in the change of the mean response of the outcome (e.g., j indicates decreased mean birth weight).
b Arrows indicate the change in risk of the outcome (e.g., | indicates an increased risk of the outcome).
3-30
-------
APRIL 2024
cSteenland and Woskie, 2012, 2919168 was rated medium confidence for comparisons with the DuPont referent group and low confidence for comparisons with the U.S.
population.
3-31
-------
APRIL 2024
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)7 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 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}, use of PFAS as the dependent
variable {Jain, 2020, 6833623}, or in cases for which 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 (see Appendix D, {U.S. EPA, 2024, 11414343}). For endpoints with
fewer studies (e.g., AST serum levels, functional assays), the evidence synthesis below included
details on any low confidence studies available in addition to high and medium confidence
studies. Studies considered uninformative were not considered further in the evidence synthesis.
7 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}; and Gleason et al. {,
2015, 2966740} all used NHANES data from overlapping years.
3-32
-------
APRIL 2024
///////
\G©
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-
+
+
++
-
+
+
-
~
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 3-7. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Hepatic Effects"
Interactive figure and additional study details available on IiAWC.
^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 use NHANES data from overlapping years.
3-33
-------
APRIL 2024
Ne^0°
9^
^&0^
*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 etal. {, 2020, 6988477}; Liu et al. {, 2018, 4238514}; Gleason et al. {, 2015,
2966740} all use NHANES data from overlapping years.
3-34
-------
APRIL 2024
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 {U.S. EPA, 2024,
11414343}. 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 versus
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 presented in Appendix D {U.S. EPA, 2024, 11414343} 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 or identified in the updated literature search.
3-35
-------
APRIL 2024
Reference
SeSlag C°^" "ST Sn Exposure Le,els Compos™ EE
Effect Estimate
1.0 1.5 2.0
Pre-2016 Gallo et al. Serum Cross - Median=28.0 ng/mL OR (per 1-ln ng/mL increase in PFOA) 77
Literature (2012, sectional (25th-75th percentile:
-0-
Search 1276142), 13.5-70.8 ng/mL) 0R ({or deCj|e 2 Vs. decile 1 of PFOA) "IT
Medium 1.09
OR (for decile 3 vs. decile 1 of PFOA) ~~~~
OR (for decile 4 vs. decile 1 of PFOA) ~]2&
OR (for decile 5 vs. decile 1 of PFOA) 1 4
OR (for decile 6 vs. decile 1 of PFOA) 1 3g
OR (for decile 7 vs. decile 1 of PFOA) ^ ^
OR (for decile 8 vs. decile 1 of PFOA) ^ ^
OR (for decile 9 vs. decile 1 of PFOA)
1.4
OR (for decile 10 vs. decile 1 of PFOA)
Gieason et al. Serum Cross- Median=3.7 ng/mL OR (for Q2 vs. Q1) .
(2015, sectional (25th-75th percentile:
2966740), 2.5 - 5.2 ng/mL) 0R «or Q3 vs Q-n
Medium 1.56
OR (for Q4 vs. Q1) 1 K
Updated Darrow et al. Modeled Cohort and 20th - 80th percentile: OR (per Hn y-ng/mL increase in modeled cumulative ..
Literature (2016, (serum) cross- 191.2 -3997.6y-ng/mL PFOA) 104
Search 3749173), sectional OR [for quintile 2 (191.2-<311.3 y-ng/mL) vs. quintile 1
Medium (50 3 _ <191.2 y-ng/mL) modeled cumulative PFOA)
-0-
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
(50.3 - <191.2 y-ng/mL) modeled cumulative PFOA) ' 2
OR [for quintile 5 (3997.6 - 205667.3 y-ng/mL) vs. quintile
1 (50.3 - <191.2 y-ng/mL) modeled cumulative PFOA)
Median = 16 5 ng/mL, OR (per 1-In y-ng/mL increase in 2005/2006 estimated
(range: 2.6 - 3559 serum PFOA) 104
n9/mL> OR [for quintile 2 (5.8- <11.4 ng/mL) vs. quintile 1 (2.6 -
<5.8 ng/mL) 2005/2006 estimated serum PFOA] 0 94
OR [for quintile 3 (11.4-< 26.7 ng/mL) vs. quintile 1 (2.6 - ..
<5.8 ng/mL) 2005/2006 estimated serum PFOA] 1 ut>
OR [for quintile 4 (26.7-< 81.5 ng/mL) vs. quintile 1 (2.6 -
<5.8 ng/mL) 2005/2006 estimated serum PFOA] 1 16
OR [for quintile 5 (81.5 - 3558.8 ng/mL) vs. quintile 1 (2.6 - . .
<5.8 ng/mL) 2005/2006 estimated serum PFOA] 11
1.0 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 HAWC
3-36
-------
APRIL 2024
Reference,
Confidence
Rating
Exposure
Matrix
Study
Design
Exposure Levels
Sub-population
Comparison
EE
0.00 0.02
Effect Estimate
0.04 0.06 0.08
Darrow et al.
(2016, 3749173),
Medium
Modeled
(serum)
Cohort and
cross -
sectional
20th - 80th percentile: 191.2 - 3997.6
y-ng/mL
Regression coefficient (per
Hn 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
>= 50 years old
Regression coefficient (per
1-ta y-ng/mL increase in
modeled cumulative PFOA)
0.01
1
1
1
Female
Regression coefficient (per
Hn y-ng/mL increase in
modeled cumulative PFOA)
0.01
1
1
1
Male
Regression coefficient (per
1-ln y-ng/mL increase in
modeled cumulative PFOA)
0.01
1
1
1
Median = 16.5 ng/mL, (range. 2.6 - 3559
ng/mL)
Regression coefficient (per
1-ln ng/mL increase in
2005/2006 estimated seru..
0.01
r
i
i
< 50 years old
Regression coefficient (per
1-ln ng/mL increase in
2005/2006 estimated seru.
0.01
I
i
i
>= 50 years old
Regression coefficient (per
1-ln ng/mL increase in
2005/2006 estimated seru..
0.01
i
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
1
i
i
Jain et al. (2019.
5080621),
Medium
Serum
Cohort
Geometric mean (95% CI) = 2.0 ng/mL (1.8
2.1)
Obese
Regression coefficient (per
1-Iogl0 ng/mL increase in
PFOA)
0.07
i
1
I •
1
Nian etal. (2019,
5080307),
Medium
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
1
I
i •
I
Serum
Cross -
sectional
Median=6.19 ng/mL (25th-75th percentile:
4.08-9.31 ng/mL)
Excluding
medicine takers
Regression coefficient (per
1-ln ng/mL increase in
PFOA)
0.05
I
1 •
I
Salihovic et ai.
(2018, 5083555),
Medium
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 ng/mL increase in
PFOA)
0.04
1
I
I •
1
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 HAWC.
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 Appendix D {U.S. EPA, 2024, 11414343}. 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
3-37
-------
APRIL 2024
associations for bilirubin were observed in some studies {Salihovic, 2018, 5083555; Darrow,
2016, 3749173}.
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 nonmonotonic and not statistically significant. Jin et al. {, 2020,
6315720} additionally reported lower odds of ballooning and portal inflammation, but higher
odds of steatosis (association nonmonotonic) 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 with regional mortality statistics and a control group of nonchemical
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 12 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 31 studies are shown in Figure
3-11 and Figure 3-12.
3-38
-------
APRIL 2024
O*®
Abbott et al., 2007, 1335452 -
l
+
l
+
1
+
—I—
+
++ ++
++
~
Biegel etal., 2001, 673581 -
++
++
NR
++
++
+
++ ++
++
++
Blake et al., 2020, 6305864-
+
+
++
+
+
+
g
++
~
Butenhoffetal., 2004, 1291063-
++
NR
NR
++ ++
+
++ ++
++
++
Butenhoff et al., 2012, 2919192-
+
++
NR
+
D
++
+
Cope etal., 2021, 10176465-
+
+
+
+
-09
+
Crebelli et al.,2019, 5381564-
+
+
NR
+
+
BB
B
+
De Guise et al., 2021, 9959746-
+
+
NR
+
+
+
++ ++ ++
+
Filgo et al., 2015, 2851085-
+
+*
+
+
BB
+
Guo et al.,2019, 5080372-
+
+
NR
+
+
1 +
+
+
Guo et al., 2021, 7542749-
+
NR
+
+
+
++ ++aje
+
Guo et al., 2021, 9960713-
-
+
+
-
++
0
-
Guo et al., 2021, 9963377-
+
+
NR
+
-*
++
++ ++*
++
Hu etal., 2010, 1332421 -
++
NR
NR
++
-
+
B
++
+
Lau et al., 2006, 1276159-
+
+
NR
+
+
++
++ ++
~
+
Legend
Q Good (metric) or High confidence (overall)
+ | Adequate (metric) or Medium confidence (overall)
- Deficient (metric) or Low confidence (overall)
9 Critically deficient (metric) or Uninformative (overall)
NR 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.
3-39
-------
APRIL 2024
Li etal., 2017, 4238518-
+
NR
NR
+
+
NR
+
+
NR
Macon et al., 2011, 1276151
NTP, 2019, 5400977
NTP, 2020, 7330145
Perkins etal., 2004, 1291118-
Quist et al., 2015, 6570066 -
Shi et al., 2020, 7161650
Tucker et al., 2015, 2851046 -
Wolf etal., 2007, 1332672-
Yan et al., 2014, 2850901 -
Yan et al„ 2017, 3981501 -
Yuet al., 2016, 3981487-
Zhang et al., 2020, 6505878 -
Zhang etal., 2021, 10176453
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.
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 exposures 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, lifestage, and exposure paradigm (Figure
3-13 and Figure 3-14). 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,
3-40
-------
APRIL 2024
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 to 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 to 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,
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 to GD 11.5 or GD 1.5 to GD
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. Quist et al. {, 2015, 6570066} exposed pregnant CD-I mice to PFOA from GD 1
to GD 17. At PND 21, significantly increased relative liver weights in offspring were observed
as low as 0.3 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
3-41
-------
APRIL 2024
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 to GD 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 to
GD 17 {Hu, 2010, 1332421}.
3-42
-------
APRIL 2024
PFOA Hepatic Effects - Relative Liver Weight
Study Name
Study Design
Observation Time
Animal Description
9 No significant change A Significant increase ~ Significant decrease
Blake et al.. 2020, 6305864
developmental (GD1.5-11.5)
GD11.5
P0 Mouse, CD-1 (2, N=11)
« A
A
developmental (GD1.5-17.5)
GD17.5
PO Mouse, CD-1 ( -, N=11)
M A—
A
Lietal., 2018, 5084746
developmental (GD1-17)
GD18
P0 Mouse, Kunming (2, N=10)
« A—
Quistetal., 2015, 6570066
developmental (GD1-17)
RND21
F1 Mouse, CD-1 U,N=10)
Tucker et al., 2015, 2851046
developmental (GD1-17)
PND21
F1 Mouse, CD-1 (2, N=19-22)
F1 Mouse, C57BI/6 (9, N=2-6)
m • m i
Wolf eta I., 2007, 1332672
developmental (GD1-17)
PND22
P0 Mouse, CD-1 {", N=25-39)
m
F1 Mouse, CD-1 0, N=11-14)
—A-A
F1 Mouse, CD-1 (2, N=11-14)
m
developmental (GD15-17)
PND22
P0 Mouse, CD-1 (2, N=3-10)
F1 Mouse, CD-1 (2, N=3-10)
A—•
developmental (GD1-PND22)
PND22
F1 Mouse, CD-1 (d', N=12-14)
«
F1 Mouse, CD-1 (2, N=12-14)
m
developmental (PND1-22)
PND22
F1 Mouse, CD-1 (o, N=11-14)
«
—A-A
F1 Mouse, CD-1 (O, N=11-14)
Abbott et al.. 2007, 1335452
developmental (GD1-17)
PND22
PO Mouse, 129S1/SvlmJ (?, N=11-36)
m • • • A—
F1 Mouse, 129S1/SvlmJ ($$, N=4-9)
Hu et al., 2010,1332421
developmental (GD6-17)
PND48
F1 Mouse, C57BL/6n (¥, N=8)
m • ~
Macon et al., 2011, 1276151
developmental (GD10-17)
PND1
F1 Mouse, CD-1 (2, N=3-5)
+ • ~
PND7
F1 Mouse, CD-1 (2, N=3-5)
m • A
PND14
F1 Mouse, CD-1 ( ", N=2-5)
PND21
F1 Mouse, CD-1 (°, N=2-5)
m • i
developmental (GD1-17)
PND7
F1 Mouse, CD-1 N=3-6)
m A A—
F1 Mouse, CD-1 (2, N=4-5)
m A A—
-A
PND14
F1 Mouse, CD-1 (c, N=4-6)
< » •
—A
F1 Mouse, CD-1 (^, N=4-6)
* • A—
-A
PND21
F1 Mouse, CD-1 (J', N=4)
* • •
—A
F1 Mouse, CD-1 (2, N=3-6)
m • •
—A
PND84
F1 Mouse, CD-1 ( N=3-5)
m • •
F1 Mouse, CD-1 (2, N=2-5)
N • •
0.01 0.1 1
10 100
Concentration (mg/kg/day)
Figure 3-13. Relative Liver 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; PNW = postnatal week; LD = lactational day; Po = parental generation; Pi = first
generation; d = day; wk = week; y = year.
3-43
-------
APRIL 2024
PFOA Hepatic Effects - Relative Liver Weight
Study Name
Study Design
Observation Time
Animal Description
0 No significant changed Significant increase V Significant decrease
Cope etal.,2021, 10176465
developmental (GD1.5-17.5)
PND22
F1 Mouse, CD-1 (3, N=8)
A
F1 Mouse, CD-1 (J, N=9)
«
PND128
F1 Mouse, CD-1 (9, N=7)
«
F1 Mouse, CD-1 (¦?, N=8)
*—
Yan etal., 2014, 2850901
short-term (28d)
28d
Mouse, BALB/c (;5, N=16)
—•—A A A A
Yuet al, 2016, 3981487
short-term (28d)
28d
Mouse. BALB/c(o, N=5)
A
Guo etal., 2021, 9963377
short-term (28d)
28d
Mouse, BALB/c (d, N=12)
[ >
[ >
!~
~e Guise et al., 2021, 9959746
short-term (4wk)
4wk
Mouse, B6C3F1 (9, N=12-16)
A A
Guo etal., 2019, 5080372
short-term (4wk)
4wk
Mouse, BALB/c (o, N=12)
+
A A A
Loveless et al., 2008, 988599
short-term (29d)
29d
Mouse, Crl;CD-1(ICR)BR ( \ N=20)
m
¦ A A—A
Shi etal., 2020, 7161650
subchronic (5wk)
5wk
Mouse, C57BL/6J (J', N=8)
m
A A—A
Butenhoff et al., 2004. 1291063
reproductive (84d)
LD22
P0 Rat, Crl:CD(SD)IGS BR ( V, N=26-29)
• V V ~
reproductive (64d)
106d
P0 Rat, Crl:CD(SD)IGS BR (-\ N=29-30)
| >
[ >
| >
1 ~
reproductive (GD1-PND106)
LD22
F1 Rat, Crl:CD(SD)IGS BR (N=28-29)
m
reproductive (GD0-PND120)
PND120
F1 Rat, Crl:CD(SD)IGS BR(•:. N=29-30)
M
Loveless et al., 2008, 988599
short-term (29d)
29d
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
m
NTP, 2019, 5400977
short-term (28d)
29d
Rat, Sprague-Dawley , N=10)
m
Rat, Sprague-Dawley N=9-10)
• •AAA
Perkins et al., 2004, 1291118
subchronic (13wk)
13wk
Rat, Sprague-Dawley Crl:Cd Br N=15)
m
NTP, 2020, 7330145
chronic (GD6-PNW21)
16wk
F1 Rat, Sprague-Dawley ( f, N=10)
m
chronic (GD6-PNW107)
16wk
F1 Rat, Sprague-Dawley (;?, N=10)
m
F1 Rat, Sprague-Dawley (_, N=10)
i
A
chronic (PND21-PNW21)
16wk
F1 Rat, Sprague-Dawley N=10)
m
A A
chronic (PND21-PNW107)
16wk
F1 Rat, Sprague-Dawley ( :'. N=10)
«4
A—A—A
F1 Rat, Sprague-Dawley (_. N=10)
*
A
Biegel etal., 2001, 673581
chronic (2y)
15mo
Rat, Crl:Cd Br(.;', N=6)
W
A
Butenhoff et al., 2012,. 2919192
chronic (2y)
2y
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) (c, N=15)
«
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) (_ , N=15)
m i t
0.01
0.1 1 10 100
Concentration (mg/kg/day)
Figure 3-14. Relative Liver Weight in Rodents Following Exposure to PFOA (Continued,
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; 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,
3-44
-------
APRIL 2024
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 3-15 (male mice), Figure 3-16
(male rats), Figure 3-17 (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 >twofold compared with 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 to 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 to GD 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}.
3-45
-------
APRIL 2024
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)
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
Aspartate Aminotransferase (AST) Cope etal.. 2021,10176465 developmental (GD1.5-17.5) PNW18 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 3-15. 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 here and here.
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 six 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
3-46
-------
APRIL 2024
ranging from 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.
Endpoint Study Name
Alanine Aminotransferase (ALT) NTP, 2019, 5400977
Study Design Observation Time Animal Description
(28d) 29d Rat. Sprague-Dawley (5, N"10)
Dose (mgfkg/day)
PFOA Hepatic Effects - Serum Enzymes in Male Rats
| Q Statistically significant 0 Not statistically significant I | 95% CI |
NTP, 2020, 7330145 chronic (GD6-PNW21)
chronic (GD6-PNW107) 16wk
chronic (PND21-PNW21) 16wk
Butenhoff etal., 2012, 2919192 chronic (2y>
Alkaline Phosphatase (ALP) NTP, 2019, 5400977
<28d) 29d
chronic (PND21-PNW21) 16wk
chronic (PND21-PNW107) '
Butenhoff etal., 2012, 2919192 chronic (2y)
Aspartate Aminotransferase (AST) NTP, 2019, 5400977
n (28d) 29d
Butenhoff et al., 2012. 2919192 chronic (2y)
F1 Rat, Sprague-Dawley (_C, N=10)
F1 Rat, Sprague-Dawley (J, N=10)
F1 Rat. Sprague-Dawley ( N=10)
chronic (PND21-PNW107) 16wk F1 Rat. Sprague-Dawley (•-?, N=10)
Rat. Sprague-Dawley Crl:Cd(Sd)(Br) . N=14-15)
Rat, Sprague-Dawley N=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)
F1 Rat, Sprague-Dawley (-*, N=10)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) (c, N=14-15)
Rat, Sprague-Dawley (•;', N=10)
Rat. Sprague-Dawley Cri:Cd(Sd)(Br) N=14-15)
-20 0 20 40 60 80 100 120 140 160 180 200
Percent control response (%)
Figure 3-16. Percent Change in Serum Enzyme Levels Relative to Controls in Male Rats
Following Exposure to PFOAa
Interactive figure and additional study details available on HAWC here and here.
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.
3-47
-------
APRIL 2024
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 to GD 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 to GD 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.
3-48
-------
APRIL 2024
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
Rat, Sprague-Dawley (-, N=9-10) 0
6.25
H*
HH
NTP 2020,7330145 chronic (GD6-PNW107) 16wk
1 Rat, Sprague-Dawley (", N=10) 0
chronic (PND21-PNW107) 16wk
Alkaline Phosphatase (ALP) Blake et al., 2020, 6305864 developmental (GD1.5-11.5) GD11.5
63.5
F1 Rat, Sprague-Dawley («, N=10) 0
18.2
63.4
P0 Mouse, CD-1 (". N=5) 0
1
Cope el al., 2021, 10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse. CD-1 {'>. N=7)
NTP, 2019, 5400977 short-term (28d)
Rat, Spraguo-Dawley (-, N=9-10) 0
6.25
12.5
25
Cope et al.. 2021,10176465 developmental (GD1.5-17.5) PNW18 F1 Mouse, CD-1 N=7)
developmental (GD1.5-17.5) GD17.5
PO Mouse. CD-1 (k, N=4-6) 0
h*~l
NTP, 2020,7330145 chronic (GD6-PNW107) 16wk
chronic (PND21-PNW107) 16wk
Aspartate Aminotransferase (AST) Blake et al.. 2020. 6305864 developmental (GD1.5-11.5) GD11.5
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
jH
developmental (GD1.5-17.5) GD17.5 PO Mouse. CD-1 ( -, N=4-6) 0
m
NTP, 2019, 5400977
i (28d) 29d
Rat, Sprague-Dawley (", N=9-10) 0
~
-i 1 l—
-150 -100 -50 0 50 100 150 200 250 300
Percent control response (%)
Figure 3-17. Percent Change in Enzyme Levels Relative to Controls in Female Rodents
Following Exposure to PFGAa
Interactive figure and additional study details available on HAWC here and here.
ALT = alanine aminotransferase; ALP = alkaline phosphatase; AST = aspartate aminotransferase; GD = gestation day;
PND = postnatal day; PNW = postnatal week; Po = parental generation; Ei = 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.
3-49
-------
APRIL 2024
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 with
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 with 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 to GD 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-
3-50
-------
APRIL 2024
month-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 with 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 with 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 with 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 with 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 death8 and degeneration was the final category of hepatic histological effects
observed across multiple studies, species, and sexes (Table 3-3). 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 3-3 (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 to GD
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 3-3) {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.
8 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.
3-51
-------
APRIL 2024
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
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 3-3 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 3-3. 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-d Sprague-Dawley rat
oral gavage dosing; 0, 0.625,
1.25,2.5, 5, 10 mg/kg/d
29-d Crl:CD(SD)IGS BR rat
oral gavage dosing; 0, 0.3, 1,
10, 30 mg/kg/d
29-d Crl: CD-1 (ICR)BR
mouse oral gavage dosing; 0,
0.3, 1, 10, 30 mg/kg/d
29-d Crl: CD-1 (ICR)BR
mouse oral gavage dosing; 0,
0.3, 1, 10, 30 mg/kg/d
4-wk Crl:CD®BR rat feeding
study; 0, 0.06, 0.64, 1.94,
6.5 mg/kg/d
7-wk Crl:CD®BR rat feeding
study; 0, 0.06, 0.64, 1.94,
6.5 mg/kg/d
13-wk Crl:CD®BR rat
feeding study; 0, 0.06, 0.64,
1.94, 6.5 mg/kg/d
2-yr Crl:COBS® CD(SD)BR
rat feeding study; 0, 1.3,
14.2 mg/kg/d
Gestational CD-I mouse
gavage dosing from GD 1.5
to GD 17.5 (offspring); 0,
0.1, 1 mg/kg/d
16-wk 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
3-52
-------
APRIL 2024
Reference
Study Design
Endpoint Name
Incidence
16-wk 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-yr 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-d Hsd:Sprague-Dawley
SD rat oral gavage dosing; 0,
6.25, 12.5,25, 50,
100 mg/kg/d
2-yr Crl:COBS® CD(SD)BR
rat feeding study; 0, 1.6,
16.1 mg/kg/d
Gestational CD-I mouse
gavage dosing from GD 1.5
to GD 11.5 (dams); 0, 1,
5 mg/kg/d
Gestational CD-I mouse
gavage dosing from GD 1.5
to GD 17.5 (dams); 0, 1,
5 mg/kg/d
Gestational CD-I mouse
gavage dosing from GD 1.5
to GD 17.5 (offspring); 0,
0.1, 1 mg/kg/d
16-wk 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-yr 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}.
3-53
-------
APRIL 2024
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 as organized by mechanistic data category (see Appendix A, {U.S.
EPA, 2024, 11414343}) and source is shown in Figure 3-18.
3-54
-------
APRIL 2024
Mechanistic Pathway Animal Human In Vitro Grand Total
Angiogenic, Antiangiogenic, Vascular Tissue Remodeling
Atherogenesis And Clot Formation
0
0
1
1
Big Data, Non-Targeted Analysis
9
0
11
19
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
21
0
19
37
Hormone Function
6
1
1
8
Inflammation And Immune Response
5
1
3
9
Oxidative Stress
8
0
14
21
Renal Dysfunction
Xenobiotic Metabolism
8
1
12
20
Other
0
0
3
3
Not Applicable/Not Specified/Review Article
Grand Total 42 2 47 83
Figure 3-18. Summary of Mechanistic Studies of PFOA and Hepatic Effects
Interactive figure and additional study details available on HAWC.
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; vanEsterik,
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.
3-55
-------
APRIL 2024
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 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 |iM
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 [xM 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 [xM 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
3-56
-------
APRIL 2024
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, lg, 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, lg, 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.
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.
3-57
-------
APRIL 2024
{, 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 with
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 with 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 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 with 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-
3-58
-------
APRIL 2024
related genes Cyp2bl whereas males exhibited the greatest fold increase for Cyp4al and Cyp2bl
compared with 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 with 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
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
3-59
-------
APRIL 2024
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
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}. Given 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
nonalcoholic 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 with 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.
3-60
-------
APRIL 2024
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
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 with
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 with 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 to 100 and from 0 to 200 [xM, respectively, or the PPARa activator
Wyl4,643. In mouse cells, many fewer genes were altered by PFOA treatment compared with
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, CYP3A4, 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
3-61
-------
APRIL 2024
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
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 very4ow-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
nonalcoholic 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-, lifestage-, 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}.
3-62
-------
APRIL 2024
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 with 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 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 eightfold and a 15-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 lifestages, 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
3-63
-------
APRIL 2024
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
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
3-64
-------
APRIL 2024
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 with 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
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 day) 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. RNA-Seq 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. On the basis of 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 ofPFOA, it may
preferentially bind to proteins harboring reactive cysteine residues. To test this hypothesis,
3-65
-------
APRIL 2024
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,
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 toxicological 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 with 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. On the basis of the sexually dimorphic lipid levels, as well as on extrahepatic
changes, females appear more sensitive to PFOA-mediated alterations in metabolic
programming.
3-66
-------
APRIL 2024
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 lifestages 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
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). On the basis of 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 [xM) in drinking water for
6 weeks that led to serum PFOA levels of 48 |ig/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-
3-67
-------
APRIL 2024
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
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 with 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.
3-68
-------
APRIL 2024
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
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 [xM 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 with 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
3-69
-------
APRIL 2024
authors linked PFOA-mediated gene expression changes to diminished global methylation,
implicating epigenetic factors in PFOA-mediated 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 [xM 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, UGT1A1). The gene profiles after
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 knockout 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
lifestage. 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
3-70
-------
APRIL 2024
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
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.
3-71
-------
APRIL 2024
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 Appendix, {U.S. EPA, 2024, 11414343}), 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 nontumor 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 regarding 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.
3.4.1.3.4 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
3-72
-------
APRIL 2024
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 with 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 [xM PFOA
dissolved in water from 3 to 120 hours post-fertilization (hpf) and evaluated 180 days post-
fertilization (dpf) at adult lifestage 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-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 [xM) 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 [xM. 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
3-73
-------
APRIL 2024
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 fivefold reduction in the expression.
Conversely, Behr et al. {, 2020, 6505973} and Louisse et al. {, 2020, 6833626} reported
upregulation of CYP3 A4 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 (xM) 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.
Franco et al. {, 2020, 6315712} also assessed gene expression of phase II enzymes, glutathione-
s-transferase mul (GST-MI) and UDP glucuronosyltransferase-lAl (UGT-1A1), 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 [xM) of PFOA for 48 hours. As a result, 1,973 genes expressed >1.5-fold changes in the
exposed groups compared with 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.
3-74
-------
APRIL 2024
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; Wielsoe, 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 [xM PFOA {Xu,
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 [xM 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 [xM, suggesting a concentration-dependent reduction in cell
viability {Florentin, 2011, 6333436}. In contrast, cell viability dropped below 80% in HepaRG
cells exposed to 100 [xM 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
[xM 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 [xM PFOA after 72 hours
{Buhrke et al., 2013, 2325346}, 102-145 pM after 24 hours {Ojo, 2020, 6333436; Franco, 2020,
6315712}, to 305 [xM after 48 hours of exposure in HepG2 cells {Song, 2016, 9959776}. In a
fetal liver cell line (HL-7702), IC50 values were 647 [xM after 24 hours exposure and 111 [xM
after 48 hours exposure {Hu, 2014, 2325340; Sheng, 2018, 4199441}. One study in zebrafish
liver cells reported IC50 values of 84.76 [j,g/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 [xM
3-75
-------
APRIL 2024
and 1,000 [xM PFOA for 24 hours {Xu, 2019, 5381556}. In zebrafish liver cells exposed to the
IC50 (84.76 [xg/mL) and IC80 (150.97 [xg/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 with 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 [xM but increased by more than 1,000% at 1,000 [xM {Xu, 2020, 6316207}. In a
spheroid model of mouse liver cells (AML12), increased activity of caspase 3/7 was detected
from 14 to 28 days of >100 [xM PFOA exposure {Sun, 2019, 5024252}. In contrast, 100 [xM
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 with 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 [xg/mL PFOA for 3 hours, BAX mRNA
was significantly increased while B cell lymphoma 2 (Bcl-2) decreased compared with control
{Lv, 2019, 5080368}. Proteomic analysis of 28 proteins differentially expressed in PFOA-
exposed human nontumor 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
3-76
-------
APRIL 2024
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.
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 to 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 with 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 [xM for 48
or 96 hours {Zhang, 2016, 3748826}. All except the highest dose (400 [xM) group showed an
increase in cell proliferation compared with control at 48 hours. Other studies have reported a
similar pattern for which 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 [xM for 24 hours compared with
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 [xM at the same duration, adding to the
evidence that PFOA-induced proliferation is dose-dependent and may vary by cell type.
3-77
-------
APRIL 2024
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
(nondividing) significantly decreased while cells in the S-phase increased after 48 hours of
exposure to 50 and 100 |xM PFOA. However, at the 200 |xM and 400 [xM 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 with 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 nontumor 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 with 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
third study in a zebrafish liver cell line also used flow cytometry to identify changes in the cell
cycle after 85 and 151 ng/mL PFOA exposure for 48 hours. In corroboration with the study in L-
02 cells, PFOA concentration of 151 ng/mL 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 [xM
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 [xM). However, cyclin A2, cyclin B1 and p21
proteins were significantly inhibited at high dose (400 [xM) at the same duration (48 hours)
{Zhang, 2016, 3748826}. In primary human hepatocytes treated with 10 [xM 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 to
305 [xM 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
3-78
-------
APRIL 2024
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
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).
3-79
-------
APRIL 2024
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.
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 in 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 with mice fed the control diet. Taken together with the
histopathological findings, these gene expression changes suggest 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.
3-80
-------
APRIL 2024
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.
3.4.1.3.6.2 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 multicellular 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.
3-81
-------
APRIL 2024
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
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,
3-82
-------
APRIL 2024
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 with 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 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 with 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
with 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
3-83
-------
APRIL 2024
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 indicate 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 twofold 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 threefold and 1.3-fold, respectively. The same doses of PFOA also led to a nearly
twofold 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
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 to 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 [xM PFOA exposure in
multicellular 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 [xM 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 [xM PFOA displayed significantly increased levels of
3-84
-------
APRIL 2024
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
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 GSSGto 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 [xM 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.
3-85
-------
APRIL 2024
In human HL-7702 liver cells, 24 hours of PFOA exposure at 1, 2.5, or 7.5 [xg/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 [xM {Florentin,
2011, 2919235} or following 3-hour PFOA exposures up to 400 [xM {Shan, 2013, 2850950}.
Wiels0e et al. {, 2015, 2533367} incubated HepG2 cells with up to 2 x io~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 with 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 |xM 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 |xM PFOA, respectively.
Additionally, superoxide anion levels were detectable in 43% and 71% of cells exposed to 200
and 400 |xM 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 [xM 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 [xM
PFOA. Activity levels of SOD and CAT were not altered in exposed cells compared with
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 [xM 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 [xM PFOA for 24 hours, Florentin et al. {, 2011, 2919235}
observed significant increases in the percentage of DNA tails, an indicator of DNA damage
3-86
-------
APRIL 2024
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 |iM 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 with 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 [xM 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.
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 [xM PFOA for 24 and 72 hours. The spheroid cell model was
exposed to 50, 100, and 200 [xM PFOA for up to 28 days. In monolayer cells exposed to 200 [xM
PFOA for 72 hours, ROS levels, measured via an ROS-Glo assay kit, increased 1.6-fold
compared with controls. In the spheroid cell models, however, ROS levels decreased in cells
exposed to 100 and 200 [xM 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
3-87
-------
APRIL 2024
this updated systematic review, there is consistent evidence of a positive association between
exposure to PFOA and ALT in adults {Gleason, 2015, 2966740; Salihovic, 2018, 5083555;
Yamaguchi, 2013, 2850970; Jain, 2019, 5381541; Jain, 2019, 5080621; Gallo, 2012, 1276142;
Darrow, 2016, 3749173; Lin, 2010, 1291111; Nian, 2019, 5080307}. 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 from 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 from 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 multipollutant 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 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 27 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 to GD 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
3-88
-------
APRIL 2024
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 theNTP {, 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
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 with 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 regarding hepatotoxicity, specifically steatosis.
PPARa has a vastly different expression in rodents compared with 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
3-89
-------
APRIL 2024
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
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,
which can occur via nuclear receptor-mediated pathways.
3-90
-------
APRIL 2024
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 3-4). 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.
3-91
-------
APRIL 2024
Table 3-4. Evidence Profile Table for PFOA Exposure and Hepatic 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 3.4.1.1)
Serum
biomarkers of
hepatic injury
17 Medium
confidence
studies
5 Low confidence
studies
Studies in adults consistently
reported significant increases in ALT
(9/11). Findings in adults were
generally positive for AST (5/7) and
GGT (7/10). Some studies reported
conflicting or nonsignificant
associations, however, these were
mostly of low confidence.
Occupational studies generally
reported significant increases in ALT
(4/7), but there were some
nonsignificant associations based on
type of analysis, location, or years
analyzed. In occupational studies,
findings for liver enzymes other than
ALT were mixed, varying at times
by time, location, or sex. Findings in
studies of children were limited, but
one study observed
• Medium
confidence
studies that
reported an
effect
• Consistent
direction of
effect for ALT
• Coherence of
findings across
biomarkers
• Low confidence studies
0©O
Moderate
Evidence for
hepatic effects is
based on increases
in ALT in adults,
including
increases in ALT
in occupational
populations.
Supporting
evidence includes
increases in other
liver enzymes
such as AST and
GGT, and
increased
incidence of liver
disease mortality
in occupational
settings. Minor
uncertainties
remain regarding
mixed liver
enzyme findings
in children and
coherence across
biomarkers and
limited
availability of
0©O
1 Evidence Lndicates (likely)
Primary basis and cross-
stream coherence:
Human data indicated
consistent evidence of
hepatoxicity as noted by
increased serum biomarkers
of hepatic injury (primarily
ALT) with coherent results
for increased incidence of
hepatic nonneoplastic
lesions, increased liver
weight, and elevated serum
biomarkers of hepatic injury
in animal models. Although
associations between PFOA
exposure and other serum
biomarkers of hepatic injury
were identified in medium
confidence epidemiological
studies, there is considerable
uncertainty in the results due
to inconsistency across
studies.
Human relevance and other
inferences:
The available mechanistic
information overall provide
support for the biological
3-92
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key Findings
Factors that
Increase
Certainty
Factors that Decrease Certainty
Evidence Stream
Judgment
high-quality
studies on liver
disease.
significant positive associations for
ALT, AST, and GGT in girls (1/3).
Evidence Integration
Summary Judgment
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.
Liver disease or
injury
4 Medium
confidence
studies
3 Low confidence
studies
1 Mixed
confidence study
A limited number
of studies
examined liver
disease or injury
in general
population adults
and occupational
populations. One
occupational
study reported
significantly
higher mortality
from cirrhosis of
the liver
compared with a
group of similar,
non-exposed
workers (1/1).
Two occupational
and one general
population study
reported no
significant
• No factors
noted
• Association only observed in Low confidence studies
• Lack of coherence of across measures of liver
inflammation
3-93
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and Summary and Key Findings Factors that Factors that Decrease Certainty Evidence Stream Evidence Integration
Interpretation Increase Judgment Summary Judgment
Certainty
association with
any form of liver
disease (0/3).
Other measures of
inflammation in
the liver were
mixed and lacked
coherence.
Serum protein Significant increases in albumin
4 Medium were consistently observed in adults
confidence (4/5), while findings from the single
studies occupational study were
2 Low confidence nonsignificant. Findings for total
studies serum protein
• Medium
confidence
studies
• Consistent
direction of
effect for
albumin
• Low confidence studies
• Lmprecision of findings for
fibrinogen and other serum
proteins
and fibrinogen were mixed or
imprecise.
3-94
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key Findings
Factors that
Increase
Certainty
Factors that Decrease Certainty
Evidence Stream
Judgment
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 from In Vivo Animal Toxicological Studies (Section 3.4.1.2)
Histopathology Histopathological
3 High confidence alterations in liver
studies were observed in
11 Medium male and female
confidence rodents exposed
studies to PFOA for
various durations
(14/14). Increased
hepatocellular
hypertrophy
(10/14) and
necrosis (5/12)
were the most
common lesions.
Other lesions
included
inflammation or
cellular
infiltration (5/14),
cytoplasmic
alteration or
vacuolation
(3/12), mitosis or
mitotic figures
(3/12), bile duct
hyperplasia
(2/13),
cystic/cystoid
degeneration
• High and
medium
confidence
studies
Consistent
direction of
effects across
study design,
sex, and
species
• Dose-
dependent
response
• Coherence of
findings across
other endpoints
indicating liver
damage
(i.e., increased
serum
biomarkers and
liver weight)
• Large
magnitude of
effect, with
some responses
reaching 100%
• No factors noted
©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., hepatocellu
lar hypertrophy
alone) and those
Evidence Integration
Summary Judgment
3-95
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and
Interpretation
Summary and Key Findings
Factors that
Increase
Certainty
Factors that Decrease Certainty
(2/12), fatty
change (2/13),
and/or pigment
(1/12).
incidence in
some dose
groups
(i.e., hypertrop
hy,
vacuolation,
single-cell
death) or are
considered
severe
Evidence Stream
Judgment
that indicate
functional
impairment or
lesions. EPA
considers
responses such as
increased
Evidence Integration
Summary Judgment
(i.e., cell or
tissue
death/necrosis
and cystoid
degeneration)
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
discussed in this
section reported
dose-dependent
increases in liver
weight and
3-96
-------
APRIL 2024
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
Liver weight Absolute (17/21)
5 High confidence and relative
studies (18/22) liver
21 Medium weights were
confidence increased in male
studies and female
rodents exposed
to PFOA for
various durations.
Several studies
that included both
males and females
suggested that
males may be
more sensitive
than females
(4/7).
• High and
medium
confidence
studies
• Consistent
direction of
effects across
study design,
sex, and
species
• Dose-
dependent
response
• Coherence of
effects with
other responses
indicating
¦ Confounding variables such as
decreases in body weights
hepatocellular
hypertrophy in
rodents of both
sexes. Although a
limited number of
these studies
additionally
examined
functional or
histopathological
hepatic
impairment,
several provide
evidence of
adverse hepatic
responses.
3-97
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and Summary and Key Findings Factors that Factors that Decrease Certainty Evidence Stream Evidence Integration
Interpretation Increase Judgment Summary Judgment
Certainty
increased liver
size
(e.g., hepatocel
lular
hypertrophy)
Serum Increases were
biomarkers of observed in ALT
hepatic injury (6/9), AST (6/7),
3 High confidence ALP in (4/6), and
studies
7 Medium
confidence
studies
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
• 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
• Limited number of studies
examining specific outcomes
3-98
-------
APRIL 2024
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
unchanged in
females (3/3).
Inconsistent
changes in
hepatobiliary
damage
(i.e., histopatho
logical lesions)
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).
• Large
magnitude of
effect, with
evidence of
biologically
significant
increases
(i.e., >100%
control
responses) in
serum liver
enzymes
indicating
adversity
3-99
-------
APRIL 2024
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
Mechanistic Evidence and Supplemental Information (Section 3.4.1.3)
Biological Events Summary of Key Findings, Interpretation, and Evidence Stream
or Pathways
Limitations
Judgment
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 Key findings and interpretation:
cellular initiating . increased apoptosis is a high dose effect
events - other demonstrated in vivo, as well as in vitro, occurring
pathways 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
3-100
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and Summary and Key Findings Factors that Factors that Decrease Certainty Evidence Stream Evidence Integration
Interpretation Increase Judgment Summary Judgment
Certainty
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.
• 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.
3-101
-------
APRIL 2024
3.4.2 Immune
EPA identified 50 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, 29 as medium confidence, 12 as low confidence, 6 as mixed (6 medium low)
confidence, and 2 were considered iminformative (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 13 epidemiological studies (14 publications9) from the 2016 PFOAHESD {U.S. EPA,
2016, 3603279} that investigated the association between PFOA and immunosuppresive effects.
Study quality evaluations for these 14 studies are shown in Figure 3-19. Results from studies
summarized in the 2016 PFOA HESD are described in Table 3-5 and below.
9 Okada, 2012, 1332477 reports overlapping eczema results with Okada, 2014, 2850407
3-102
-------
APRIL 2024
><*•
,0®
Costa et al., 2009, 1429922 -
1—
+
1—
+
1—
+
—I—
+
—i—
+
—i—
+
i
+
Dong et al., 2013, 1937230-
+
++
+
+
++
+
+
+
Emmett et al., 2006, 1290905-
-
-
+
D
-
-
Fei et al., 2010, 1290805-
+
++
+
+
++
+
+
+
Grandjean et al., 2012, 1248827 -
+
+
+
+
+
+
+
Granum etal., 2013, 1937228-
+*
+
+
-
+*
Humblet et al., 2014, 2851240-
+
+
+
++
+
+
+
Looker et al., 2014, 2850913 -
+
+
++
+
+
+
+
+
Okada et al., 2012, 1332477-
+
+
+
+
+
+
+
+
Okada et al., 2014, 2850407 -
+
+
+
+
+
+
+
+
Osuna et al., 2014, 2851190 -
-
-
++
-
-
+
-
-
Steenland et al., 2015, 2851015 -
-
+
-
+
++
+
+
-
Stein et al., 2016, 3108691 -
++ ++
+
+
+
+
+
+
Wang etal., 2011, 1424977-
-
+
+
+
+
+
+
+
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
J|J Critically deficient (metric) or Uninformative (overall)
* Multiple judgments exist
Figure 3-19. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immune Effects Published Before 2016 (References in 2016 PFOA
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,
3-103
-------
APRIL 2024
1248827}. Prenatal and age five serum PFOA concentrations were inversely associated with
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. Another study on Faroese children conducted a pilot investigation on
the association between elevated PFOA exposure and autoantibodies to antigens indicating tissue
damage, but the results were unclear {Osuna, 2014, 2851190}. 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. Stein et al. {, 2016,
3108691} reported significant inverse associations between PFOA exposure and mumps and
rubella antibody concentrations in adolescents (12-19 years old) from multiple NHANES cycles
(1999-2000, 2003-2004), but no association was observed for measles. 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 observed 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 Appendix D, {U.S. EPA, 2024,
11414343}), 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.
3-104
-------
APRIL 2024
Table 3-5. Associations Between Elevated Exposure to PFOA and Immune Outcomes from Studies Identified in the 2016
PFOA HESD
Reference,
Confidence
Study
Design
Population
Tetanus
Aba
Diphtheria Aba
Rubella
Aba
Influenza
Aba
Infectious
Diseaseb
Asthmab
Eczemab
Autoimmune
Diseaseb
White Blood
Cell Count3
Costa 2009, 1429922 Cohort
Medium
Occupational
NA
NA
NA
NA
NA
NA
NA
NA
t
Dong, 2013,
1937230
Medium
Case-
control
Children
NA
NA
NA
NA
NA
tt
NA
NA
NA
Fei, 2010, 1290805
Medium
Cohort
Children
NA
NA
NA
NA
-
NA
NA
NA
NA
Grandjean, 2012,
1248827
Medium
Cohort
Children
44
44
NA
NA
NA
NA
NA
NA
NA
Granum, 2013,
1937228
Mixedc
Cohort
Children
NA
14
NA
tt
NA
NA
Humblet, 2014,
2851240
Medium
Cross-
sectional
Adolescents
NA
NA
NA
NA
NA
tt
NA
NA
NA
Looker, 2014,
2850913
Medium
Cohort
Adults
NA
NA
NA
44
NA
NA
NA
NA
Okada, 2014,
2850407
Medium
Cohort
Children
NA
NA
NA
NA
t
t
NA
NA
Steenland, 2015,
2851015
Low
Cohort
Adults
NA
NA
NA
NA
NA
NA
NA
tt
NA
Stein, 2016,3108691 Cross-
Medium sectional
Adolescents
NA
NA
44
NA
NA
t
NA
NA
NA
Wang, 2011,
1424977
Medium
Cohort
Children
NA
NA
NA
NA
NA
NA
4
NA
NA
Notes'. Ab = antibody; NA = no analysis was for this outcome was performed; | = nonsignificant positive association; ft = significant positive association; j = nonsignificant
inverse association; jj = significant inverse association; - = no (null) association.
Emmett et al., 2006,1290905 was not included in the table due to their uninformative overall study confidence ratings.
Osuna, 2014, 2851190 analyzed autoantibody response to indicators of tissue damage and was not included in the table.
Okada, 2012,1332477 reports overlapping eczema results with Okada, 2014, 2850407, which was considered the most updated data.
a Arrows indicate the direction in the change of the mean response of the outcome (e.g., j indicates decreased mean birth weight).
b Arrows indicate the change in risk of the outcome (e.g., | indicates an increased risk of the outcome).
c Granum, 2013, 1937228 was rated medium confidence for antibody response, common cold, and gastroenteritis, and low confidence for all other outcomes.
3-105
-------
APRIL 2024
3.4.2.1.2 Immunosuppression Study Quality Evaluation and Synthesis from the
Updated Literature Review
There are 27 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 27 studies are shown in Figure 3-20 and Figure 3-21.
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}.
3-106
-------
APRIL 2024
.gSjf"
&
,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 3-20. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Immunosuppression Effects
Interactive figure and additional study details available on HAWC.
3-107
-------
APRIL 2024
>0®
Kielsen etal., 2016, 4241223-
I
1
1
+
1
1
+
1
+
~
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 etal., 2016, 3860111 -
-
++
++
++
+
"
Timmermann et al., 2020, 6833710 -
+
+
+ *
+
++
+
+
Timmermann et al., 2021, 9416315-
+
+
+
+
+ +
+
+
+
Wang et al., 2022, 10176501 -
+
++
+
+
++
+
+
+
Zeng etal., 2019, 5081554-
-
+
*
++
+
+
Zeng etal., 2020, 6315718-
+
"
+
+
++
+
+
Zhang et al. 2023, 10699594 -
+
+
+
++
++
+
+
+
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 3-21. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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 (see Appendix D, {U.S. EPA, 2024, 11414343}). 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.
3-108
-------
APRIL 2024
3.4.2.1.2.1 Vaccine Response
Ten studies (11 publications1011) 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}, one study measured rubella antibodies in both adolescents
(aged 12 and older) and adults {Pilkerton, 2018, 5080265}, and one study measured rubella,
measles, and mumps antibodies in adolescents {Zhang, 2023, 10699594}. 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, eight 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; Zhang, 2023, 10699594}, 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 and adolescents, four
studies were cohorts {Timmermann, 2020, 6833710; Grandjean, 2017, 3858518; Grandjean,
2017; 4239492; Mogensen, 2015, 3981889}, and four were cross-sectional {Abraham, 2020,
6506041; Timmermann, 2021, 9416315; Pilkerton, 2018, 5080265; Zhang, 2023, 10699594}
(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; two studies measured
antibody response to measles vaccination {Timmermann, 2020, 6833710; Zhang, 2023,
10699594}, two studies measured antibody response to rubella vaccination {Pilkerton, 2018,
5080265; Zhang, 2023, 10699594} one study measured antibody response to mumps vaccination
{Zhang, 2023, 10699594}, 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 3-6 and Appendix D {U.S. EPA,
2024, 11414343}. 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
10 Multiple publications of the same study: the study populations are the same in Grandjean et al. {, 2017, 3858518} and
Mogensen et al. {, 2015, 3981889}.
11 Zhang {, 2023, 10699594} analyzes NHANES cycles 2003-2004 and 2009-2010 partially overlapping with Pilkerton {, 2018,
5080265} and Stein {,2016, 3108691} which both analyze cycles 1999-2000 and 2003-2004.
3-109
-------
APRIL 2024
(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
3-6). 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. Changes in tetanus and diphtheria antibody concentrations in children from
all high and medium confidence studies are provided in Figure 3-22 and Figure 3-23.
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-j0rgensen, 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.
3-110
-------
APRIL 2024
Reference, . - Sub- Exposure Outcome
Confidence Rating Exposure Levels Comparison population Age Age E
Effect Estimate
-60 -40 -20 0 20 40 60
Grandjean et al. PFOA at 7 years median percent change (per doubling of - Age 7 Age 13
(2017a. 3858518), High (25th-75th percentile^ 4.4 ng/mL PFOA) 9 4
(3.5-5.7 ng/mL)
I
1 _
1 •
1
PFOA at 13 years median percent change (per doubling of - Age 13 Age 13
(25th-75th percentile)=2.0 ng/mL PFOA) 3.3
(1.6-2.5 ng/mL)
1
l_
I*
1
Grandjean et al. (2012, Age 5 PFOA: Geometric Percent difference (per doubling in - Age 5 Age 7
1248827), Medium mean=4.06 ng/mL (25th-75th age 5 PFOA) -35.8
percentile=3.33-4.96 ng/mL)
1
1
1
Adj for Age 5 Age 5 Age 7
Ab -28.2
1
1
• 1
1
Pre-booster Age 5 Age 5
-13.3
1
• 1
1
Post-booster Age 5 Age 5
-9.7
_ |
• 1
1
Maternal PFOA: Geometric Percent difference (per doubling in - Prenatal Age 7
mean=3.20 ng/mL (25th-75th maternal PFOA) 7.4
percentile=2.56-4.01 ng/mL)
l
1 _
1 •
1
Adj for Age 5 Prenatal Age 7
Ab 12.3
1
1 "
1
Pre-booster Prenatal Age 5
-10.5
_ 1
• 1
1
Post-booster Prenatal Age 5
14.5
1
1 *
1
Grandjean et al. - Percent change (per doubling of Cohort 3 Age 1.5 Age 5
(2017b: 4239492), PFOA) -19.24
Medium
_ 1
• 1
1
Age 5 Age 5
-13.28
m 1
• 1
I
Cord blood Age 5
-10.46
_ |
• 1
1
Cohort 3 and Age 1.5 Age 5
5 -16.47
1
• 1
1
Age 5 Age 5
-18.75
1
• 1
1
Cord blood Age 5
-17.59
I
_ 1
¦ 1
1
Cohort 5 Cord blood Age 5
-22.25
l
1
¦
1
Median = 2.2 ng/mL (25th - 75th Percent change (per doubling of Cohort 5 Age 5 Age 5
percentile: 1.8-2.8 ng/mL) PFOA) -25.26
1
¦ 1
1
Median = 2.8 ng/mL (25th - 75th Percent change (per doubling of Cohort 5 Age 1.5 Age 5
percentile: 2.0 - 4.5 ng/mL) PFOA) -16.31
1
• 1
1
Mogensen et al. (2015. median=4.4 ng/ml (25th-75th Percent change per doubling of Age 7 Age 7 Age 7
3981889), Medium percentile=3.5-5.7 ng/ml) PFOA -20.5
1
• 1
1
Timmermann et al. median=2.13 ng/mL (25th - 75th percent difference (per unit increase Ages 7-12 Prenatal Age 7-12
(2022,9416315), percentiles: 1.68 - 2.54 ng/mL) in maternal PFOA concentration) -7
Medium
_ I
• 1
1
median=2.28 ng/mL (25th - 75th percent difference (per unit increase Ages 7-12 Age 7-12 Age 7-12
percentiles: 1.89-2.88 ng/mL) in child PFOA concentration) -8
. 1
• 1
1
-60 -40 -20 0 20 40 60
Figure 3-22. Overall Tetanus Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on HAWC.
Grandjean et al. {, 2012, 1248827} was reviewed as a part of the 2016 PFOA HESD.
3-111
-------
APRIL 2024
Reference,
Confidence Rating
Exposure Levels
Comparison
Sub-
population
Exposure
Age
Outcome
Age
EE
Effect Estimate
-50 0 50 100
150
Grandjean et al.
(2017a. 3858518),
High
PFOA at 7 years median (25th-75th percent change (per doubling of
percentile,^ 4.4 ng/mL (3.5-5.7 ng/mL) PFOA)
Age 7
Age 13
-4.1
i
1
PFOA at 13 years median (25th-75th
percentile)=2.0 ng/mL (1.6-2.5 ng/mL)
percent change (per doubling of
PFOA)
Age 13
Age13
-17.5
1
—«—u
1
Grandjean et al. (2012
1248827), Medium
Age 5 PFOA: Geometric mean=4.06
ng/mL (25th-75th percentile=3.33-4.96
ng/mL)
Percent difference (per doubling in
age 5 PFOA)
Age 5
Age 7
-25.2
1
• 1
1
Adj for Age 5
Ab
Age 5
Age 7
-23.4
1
1
Pre-booster
Age 5
Age 5
-6.8
r
—#-i—
i
Post-booster
Parental
Age 5
-6.1
i
«-l—
1
Maternal PFOA; Geometric
mean=3.20 ng/mL (25th-75th
percentile=2.56-4.01 ng/mL)
Percent difference (per doubling in
maternal PFOA)
Prenatal
Age 7
-22.8
1
1
Adj for Age 5
Ab
Prenatal
Age 7
-16.8
1
1
Pre-booster
Prenatal
Age 5
-16.2
1
•—\-
1
Post-booster
Prenatal
Age 5
-6.2
(
1
Grandjean et al.
-
Percent change (per doubling of
Cohort 3
Age 1.5
Age 5
30.49
1
(2017b. 4239492),
Medium
PFOA)
1
Age 5
Age 5
-6.84
r
#4
1
Cord blood
Age 5
-35.17
1
• ^
1
Cohort 3 and
5
Age 1.5
Age 5
5.44
1
1
Age 5
Age 5
3.38
1
1
Cord blood
Age 5
-17.82
1
1
Median = 2.2 ng/mL (25th - 75th
percentile: 1.8-2.8 ng/mL)
Percent change (per doubling of
PFOS)
Cohort 5
Age 5
Age 5
18.31
I
1 •
i
Median = 2.8 ng/mL (25th - 75th
percentile: 2.0 - 4.5 ng/mL)
Percent change (per doubling of
PFOA)
Cohort 5
Age 1.5
Age 5
4.19
i
i
Mogensen et al. (2015, median=4.4 ng/ml (25th-75th
3981889), Medium percentile=3.5-5.7 ng/ml)
Percent change per doubling of
PFOA
Age 7
Age 7
Age 7
-25.4
i
r
Timmermann et al.
(2022, 9416315).
Medium
median=2.13 ng/mL (25th - 75th
percentiles. 1.68 - 2.54 ng/mL)
percent difference (per unit increase
Ages 7-12
Prenatal
Age 7-12
44
i
in maternal PFOA concentration)
i
median=2.28 ng/mL (25th - 75th
percentiles. 1.89 - 2 88 ng/mL)
percent difference (per unit increase Ages 7-12
in child PFOA concentration)
Age 7-12
Age 7-12
-22
i
•—i—
i
-50 0
50 100
150
Figure 3-23. Overall Diphtheria Antibody Levels in Children from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on HAWC.
Grandjean et al. {, 2012, 1248827} was reviewed as a part of the 2016 PFOA HESD.
3-112
-------
APRIL 2024
Table 3-6. Associations between PFOA Exposure and Vaccine Response in Faroe Islands Studies
Diphtheria Antibody Associations with PFOA by Age at Tetanus Antibody Associations with PFOA by Age at
Exposure Assessment Assessment
measurement
timing, PFOA
levels (ng/mL)a
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: GM: 3.20
(2.56-4.01)
| (C3; age, sex)b
BMD/BMDL (C3
and 5; sex, birth
cohort, log-PFOA)°
| (C3; age, sex,
booster type, and the
child's specific
antibody
concentration at age
5 yr)b
| (C3; age, sex)b
BMD/BMDL (C3
and 5; sex, birth
cohort, log-PFOA)°
t (C3; age, sex,
booster type, and the
child's specific
antibody
concentration at age
5 yr)b
Birth (modeled) | (C3; age, sex)d
|| (C3 and 5; age,
sex)d
|| (C5; age, sex)d
| (C3; age, sex)d
|| (C3 and 5; age,
sex)d
|| (C5; age, sex)d
18 mo
C3:NR
C5: 2.8 (2.0-
4.5)
t (C3; age, sex)d
t (C3 and 5; age,
sex)d
t (C5; age, sex)d
| (C3; age, sex)d
|| (C3 and 5; age,
sex)d
|| (C5; age, sex)d
5 yr
C3: GM: 4.06
(3.33-4.96)
C5: 2.2 (1.8-
2.8)
| (C3; age, sex)b
| (C3; age, sex)d
t (C3 and 5; age,
sex)d
t (C5; age, sex)d
(C3; age, sex,
booster type, and the
child's specific
antibody
concentration at age
5 yr)b
BMD/BMDL (C3;
sex, age, and booster
type at age 5 yr)e
BMD/BMDL (C3;
sex, booster type at
age 5 yr, log-PFOA)°
| (C3; age, sex)b
| (C3; age, sex)d
| (C3 and 5; age,
sex)d
|| (C5; age, sex)d
|| (C3; age, sex,
booster type, and the
child's specific
antibody
concentration at age
5 yr)b
BMD/BMDL (C3;
sex, age, and booster
type at age 5 yr)e
BMD/BMDL (C3;
sex, booster type at
age 5 yr, log-PFOA)°
3-113
-------
APRIL 2024
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 yr
C3: 4.4 (3.5
5.7)
(C3; age, sex, | (C3; sex, age at
booster type)f antibody assessment,
| (C3; sex, age at booster type at age
antibody assessment, 5 yr)g
booster type at age
5yr)g
| (C3; age, sex, t CC3: sex. age at
booster type)f antibody assessment,
t (C3; sex, age at booster type at age
antibody assessment, 5 yr)g
booster type at age
5yr)g
13 yr
C3: 2.0 (1.6
2.5)
| (C3; sex, age at
antibody assessment,
booster type at age
5yr)g
t (C3; sex, age at
antibody assessment,
booster type at age
5yr)g
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
cBudtz-Jergensen and Grandjean {, 2018, 5083631}; medium confidence
dGrandjean et al. {, 2017, 4239492}; medium confidence
e Grandjean and Budtz-Jergensen {, 2013, 1937222}; medium confidence
fMogensen et al. {, 2015, 3981889}; medium confidence
g Grandjean et al. {, 2017, 3858518}; high confidence
3-114
-------
APRIL 2024
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
nonsignificant 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.
Two medium cross-sectional studies of adolescents examined associations between elevated
levels of PFOA and vaccine response {Pilkerton, 2018, 5080265; Zhang, 2023, 10699594}.
Inverse associations were observed in cross-sectional analyses in adolescents from NHANES
(2003-2004; 2009-2010) for rubella, mumps, and measles {Zhang, 2023, 10699594}, including
a significant reduction in the antibody response to mumps per 2.7-fold increase in serum (Figure
3-24). No association was observed for rubella vaccine response in the other cross-sectional
study of adolescents {Pilkerton, 2018, 5080265}, however, an overlapping study {Stein, 2016,
3108691} reporting on adolescents from the same NHANES cycles (i.e., 1999-2000 and 2003-
2004) observed a significant inverse association in adolescents seropositive for rubella.
Reference,
Confidence
Rating
Study Design
Exposure Levels
Comparison
EE
-45
-40
-35
-30
Effect Estimate
-25 -20
-15
-10
-5
0
5
Granum et al.
(2013, 3135),
Medium
Cohort
Median (25th-75th
percentiles): 1.1
(0.8-1.4) ng/mL
Percent change per ng/mL
increase PFOA
-33
1
1
1
1
1
I
Zhang et al.
(2023,
10699594).
Cross-sectional
geometric
mean=3.33 ng/mL
(25th-75th
Percent change (per
2.7-fold increase in serum
PFOA)
-4.36
1
1
1
Medium
percentile:
2.50-4.70 ng/mL)
1
1
1
-45
-40
-35
-30
-25 -20
-15
-10
-5
0
5
Figure 3-24. Overall Rubella Antibody Levels in Children and Adolescents from
Epidemiology Studies Following Exposure to PFOA
Interactive figure and additional study details available on HAWC.
Adolescent regression coefficients from Pilkerton, 2018, 5080265 were not reported quantitatively.
Regression coefficients from Granum, 2013 were re-expressed as percent change.
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, two were cohorts {Stein, 2016,
3860111; Shih, 2021, 9959487} and one was across-sectional analysis {Pilkerton, 2018,
3-115
-------
APRIL 2024
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} was considered low confidence as the analysis
suffered from potential exposure misclassification due to concurrent exposure and outcome
measurements, considering the amount of time since rubella vaccination in childhood. This was
less of a concern for the study of adolescent participants, which 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. 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
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.
3-116
-------
APRIL 2024
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, 10 studies (11 publications)12 measured associations between PFOA exposure and
infectious diseases (or disease symptoms) in children with follow-up ranging 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 {Ait Bamai, 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
{Ait Bamai, 2020, 6833636; Dalsager, 2016, 3858505; Manzano-Salgado, 2019, 5412076;
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; Ait Bamai, 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 Appendix D, {U.S. EPA, 2024,
11414343 }). 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
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.
3-117
-------
APRIL 2024
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 {AitBamai, 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 {Ait Bamai,
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 l-to-4 years
and found a positive association with fever and nasal 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 Appendix D, {U.S.
EPA, 2024, 11414343}). 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.
3-118
-------
APRIL 2024
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 PFOA HESD, two medium confidence 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 onNHANES (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}. 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 23 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 23 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 (see Appendix D, {U.S. EPA, 2024, 11414343}). 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.
3-119
-------
APRIL 2024
ed»0
-------
APRIL 2024
Thirteen studies (15 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, 10 publications were rated
medium confidence and five publications were rated low confidence (Figure 3-25). 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 Appendix D, {U.S. EPA, 2024, 11414343}),
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 and 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 nonsignificant positive associations in
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.
3-121
-------
APRIL 2024
youths of 13-22 years in age. The low 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 with
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 nonsignificant 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 Appendix D, {U.S. EPA, 2024, 11414343}). 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"
versus "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 Appendix, {U.S. EPA, 2024, 11414343}).
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. {,
3-122
-------
APRIL 2024
2020, 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}. Nonsignificant 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 Appendix D, {U.S. EPA, 2024,
11414343}).
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 nonmonotonic across
categories of exposure. Impinen et al. {, 2018, 4238440} also observed a nonsignificant
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.
3-123
-------
APRIL 2024
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 PFOA HESD {U.S. EPA, 2016, 3603279} identified one low confidence 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 six 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 3-26. 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 (see Appendix, {U.S. EPA,
2024, 11414343}). 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.
3-124
-------
APRIL 2024
r\N0
-------
APRIL 2024
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 Ammitzboll {, 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 compared with 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 nonsignificantly 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 nonsignificant and inconsistent in the
direction of effect {Steenland, 2013, 1937218}.
Evidence from a case-control study suggested lower PFOA concentrations among healthy
controls compared with those with MS {Ammitzboll, 2019, 5080379}. Serum PFOA
concentrations were 12% lower (95% CI: -24%, 2%; p = 0.099) in healthy controls compared
with 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 with cases, but this effect was not seen in women. Steenland et al. {, 2018, 5079806}
detected significantly increased levels of PFOA in ulcerative colitis cases versus 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 four studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and nine 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 3-27.
3-126
-------
APRIL 2024
I I
l i
Butenhoffetal., 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
++
++
n
B
B
Guo et al., 2021, 7542749-
+
+
NR
B
++
Huetal., 2010, 1332421 -
++
NR
NR
++
¦
++
B
++
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-
+
+
+
m
w\
3]
++
J\
++
B
Legend
0
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)
F
Not reported
*
Multiple judgments exist
Figure 3-27. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure 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
thymus weights, immune system cellular composition, and the ability to generate an immune
response following PFOA doses ranging from approximately 1 to 40 mg/kg/day.
3-127
-------
APRIL 2024
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 with 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 to GD 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 was reduced following 0/150 and 0/300 ppm {NTP,
2020, 7330145}. No changes in spleen or thymus weights were reported in females.
3-128
-------
APRIL 2024
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 with 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 with 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}.
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
3-129
-------
APRIL 2024
hypocellularity of minimal severity was reported in male rats exposed to 10 mg/kg/day (6/10
compared with 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 quantitative 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}.
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
3-130
-------
APRIL 2024
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)BRmice {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 72% 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
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
3-131
-------
APRIL 2024
response to horse red blood cells {Yang, 2002, 1332454}. PFOA-exposed mice had no increase
in plaque-forming cells in response to the immune challenge, compared with unimmunized
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 to GD 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 with males (Figure 3-28) {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 3-28) {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 to GD 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 to
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 3-28).
3-132
-------
APRIL 2024
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 ( j1, 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 3-28. 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 PFOA HESD {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 by mechanistic data category (see Appendix A, {U.S. EPA,
2024, 11414343}) and source is shown in Figure 3-29.
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 3-29. Summary of Mechanistic Studies of PFOA and Immune Effects
Interactive figure and additional study details available on HAWC.
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.
3-133
-------
APRIL 2024
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 by
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 {Sorli, 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
3-134
-------
APRIL 2024
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. Given these diet-related
results along with gene expression data showing that PPARa and PPARy were also upregulated
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.
3-135
-------
APRIL 2024
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 nonmonotonic
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
3-136
-------
APRIL 2024
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 knockout mice suggested the immunosuppressive effects of
PFOA are independent of PPARa {DeWitt, 2016, 2851016}. In this study, female C57BL/6Tac
PPARa knockout 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 knockout 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 with 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.
3-137
-------
APRIL 2024
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 3-7).
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 toward 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 {Sorli, 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-
3-138
-------
APRIL 2024
kB pathway {Lee, 2017, 3981419}. Zhang et al. {, 2014, 2851150} reported that PFOA exposure
for 21 days can disrupt the NF-kB pathway to mediate inflammatory cytokines in zebrafish. The
authors reported a nonmonotonic 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 3-7. 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.
3-139
-------
APRIL 2024
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 with 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.
3-140
-------
APRIL 2024
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 3-8). 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 Sorli 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 interspecies 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
3-141
-------
APRIL 2024
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 Sorli et al. {, 2020, 5918817} and Bassler et al. {, 2019, 5080624}. IL6
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-I8C0).
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 10 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 [xM and
significantly elevated at 5 |iM, 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
3-142
-------
APRIL 2024
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 3-8. 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
2020,6833625}
3 years post-
partum
CRP
blood protein
1
(immunoturbidimetric high-
sensitivity assay)
Bassleretal. {,
Human males
In vivo
IL-6
serum protein (Multispot
None
2019,5080624}
and females, C8
Immunoassay)
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
2020,5918817} bronchial
(ELISA)
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}
3-143
-------
APRIL 2024
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 {Timmerman, 2021, 9416315; Abraham, 2020, 6506041;
Grandjean, 2012, 1248827; Budtz-Jorgensen, 2018, 5083631}. 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
3-144
-------
APRIL 2024
(though overlapping) birth cohorts in the Faroe Islands, supported by a low confidence study in
adults.
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 with 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 with 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
3-145
-------
APRIL 2024
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
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 3-9). 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. Given 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 lifestages represents a source of uncertainty in the immunotoxicity database of
PFOA.
3-146
-------
APRIL 2024
Table 3-9. Evidence Profile Table for PFOA Exposure and 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
Studies conducted in the
• High and medium
• Low confidence studies ©©O
1 High confidence study
Faroe Islands examined
confidence studies
• Imprecision of findings Moderate
19 Medium confidence
antibody levels among
the reported effects
studies
children at various
• Consistent direction
Evidence for immune
8 Low confidence
timepoints compared with
of effect
effects is based on
studies
exposure measured
• Coherence of
decreases in childhood
3 Mixeda confidence
prenatally and throughout
findings across
antibody responses to
study
childhood. Lower
antibody response
pathogens such as
antibody levels against
and increased
diphtheria and tetanus.
tetanus and diphtheria
infectious disease
Reductions in antibody
were observed in children
response were observed
at birth, 18 mo, age 5 yr
at multiple timepoints in
(pre-and post-booster),
childhood, using both
and at age 7 yr. Similarly,
prenatal and childhood
antibody levels against
exposure levels. Similar
rubella (2/2) were
decreases in antibody
significantly decreased in
response to other
medium confidence
pathogens, such as
studies of children.
rubella, were observed,
Findings in the five
although the number of
studies examining adults
studies analyzing these
and adolescents were less
antibody responses to
consistent than in
these pathogens was
children. Three studies
limited. An increased risk
reported inverse
of upper and lower
associations, one for
respiratory tract
rubella, one for hepatitis
infections was observed
B antibodies and one for
among children, coherent
influenza A/H3N2, but
with findings of reduced
other antibody responses
antibody response. There
were inconsistent across
was also supporting
®©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.
3-147
-------
APRIL 2024
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
all exposure windows.
Infectious disease was
examined in 14 studies of
children. Studies
examining infections of
the respiratory system
observed some positive
associations (5/14),
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
20 Medium confidence
studies
6 Low confidence
studies
2 Mixeda 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 (8/12), 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. Eight studies
• 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 of increased risk Human relevance and
of asthma, and other inferences:
autoimmune disease, Given the antibody
however, the number of response data in humans,
studies examining the children and young
same type of autoimmune individuals exposed during
disease was limited. 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.
3-148
-------
APRIL 2024
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
examined allergies,
rhinitis, or
rhinoconjunctivitis. Some
positive associations (3/8)
were observed, although
this varied by outcome
timing and were at times
inconsistent. Significantly
increased odds of eczema
or atopic dermatitis were
observed in several
studies (4/13), although
these associations were
sometimes limited to
subgroups (2/4).
Increased risk of • Medium confidence • Low confidence studies
autoimmune disease was studies • Limited number of
reported in several studies examining
studies (4/6). One study outcome
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
in men. One study (1/2)
observed increased risk
of celiac disease among
female children and
young adults. Findings
Autoimmune disease
2 Medium confidence
studies
4 Low confidence
studies
3-149
-------
APRIL 2024
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
for Crohn's disease were
less consistent.
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.2.2)
Organ weights
3 High confidence
studies
7 Medium confidence
studies
Decreases in absolute
(6/8) and relative (4/8)
spleen weights and in
absolute (5/5) and
relative (3/5) thymus
weights were observed
across studies regardless
of study design. Overall,
decreases in spleen and
thymus weights were
more frequently observed
in males than females and
tended to coincide with
reductions in body
weight.
• High and medium
confidence studies
• Dose-response
relationship seen
within multiple
studies
• Coherence of
findings of other
immunological
endpoints
• Inconsistent direction
of effects across sex
• Confounding variables
such as decreases in
body weights
Immune cellularity
1 High confidence study
4 Medium 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
• 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
0©O
Moderate
Evidence is based on 13
high or moderate
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 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.
3-150
-------
APRIL 2024
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
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.
Globulins and
immunoglobulins
2 High confidence
studies
2 Medium confidence
studies
Mixed results were
reported for
concentrations of
globulins and
immunoglobulins.
Decreased globulin levels
(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.
• High and medium
confidence studies
• Dose-response
relationship
• Inconsistent direction
of effects between
species
• 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).
• 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
3-151
-------
APRIL 2024
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
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).
Histopathology
3 High confidence
studies
2 Medium confidence
studies
A short-term study in
male mice and rats
reported increased
incidence of granulocytic
hyperplasia of the bone
marrow and 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
• High and medium
confidence studies
• Coherence of
findings
• Limited number of
studies examining
specific outcomes
3-152
-------
APRIL 2024
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
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
Judgment
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.
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 PPARa.
Findings support
plausibility that PFOA
exposure can lead to
dysregulation of
signaling pathways
related to immune
response; however, data
have inconsistencies.
Notes: HFMD = hand, foot, and mouth disease; A/H3N2 = influenza A virus subtype H3N2; 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.
a Studies may be of mixed confidence due to differences in how individual outcomes within the same study were assessed (e.g., clinical test versus self-reported data).
3-153
-------
APRIL 2024
3.4.3 Cardiovascular
EPA identified 112 epidemiological and 10 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 iminformative (Section 3.4.3.1). Of the 89 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 iminformative (Section 3.4.3.1). Of the animal toxicological studies, three were
classified as high confidence, five as medium confidence, and two 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. Cardiovascular outcomes were synthesized
separately by population (i.e., adults, children, occupational populations), and definitions of
certain conditions may vary by age. For example, high blood pressure and/or hypertension is
generally defined as SBP >140 mmHg and DBP >90 mmHg in adults and SBP >130 mmHg and
DBP >80 mmHg in children and adolescents, although consistent blood pressure measurements
in youth can be challenging {Falkner, 2023, 11279612}.
There are seven epidemiological studies from the 2016 PFOA HESD {U.S. EPA, 2016,
3603279} that investigated the association between PFOA and cardiovascular effects. Study
quality evaluations for these seven studies are shown in Figure 3-30. Results from studies
summarized in the 2016 PFOA HESD are described in Table 3-10 and below.
The 2016 PFOA 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
3-154
-------
APRIL 2024
across four quartiles of cumulative exposure {Steenland, 2012, 2919168}. On the basis of 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. A later
study of community residents from the C8 Health Project reported an elevated risk of stroke in
quintiles 2 through 4 of PFOA concentrations compared with quintile 1 (HRs ranging 1.36 to
1.45); however, the association was null in continuous analyses (HR, linear = 1.00, 95% CI:
0.99, 1.01) {Simpson, 2013, 2850927}. Study authors reported a significant increased risk (HR,
linear = 1.10, 95% CI: 1.02, 1.18) after excluding the highest quintile of exposure. 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, 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 Perfluoroalkyls
{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}.
3-155
-------
APRIL 2024
xec*>0
-------
APRIL 2024
Table 3-10. Associations Between Elevated Exposure to PFOA and Cardiovascular Outcomes from Studies Identified in the
2016 PFOA HESD
Reference, Confidence
Study Design
Population
SBPa
DBPa
Hypertensionb
Strokeb
CHD, IHD, CADb
Geiger et al., 2014,
2851286
Medium
Cross-sectional
Children
4
NA
NA
Min, 2012, 2919181
Cross-sectional
Adults
NA
NA
tt
NA
NA
Raleigh et al., 2014,
2850270°
Cohort
Occupational
NA
NA
NA
NA
-
Steenland and Woskie,
2012, 2919168d
Mixecf
Cohort
Occupational
NA
NA
NA
Simpson, 2013, 2850927
Medium
Cohort
Adults and
Occupational
NA
NA
NA
t
NA
Steenland, 2015, 2851015 Cohort
Low
Occupational
NA
NA
-
t
-
Winquist and Steenland,
2014,2851142
Mixecf
Cohort
Occupational
NA
NA
t
NA
Notes'. SBP = systolic blood pressure; DBP = diastolic blood pressure; CHD = coronary heart disease; IHD = ischemic heart disease; CAD = coronary heart disease;
t = nonsignificant positive association; ft = significant positive association; j = nonsignificant inverse association; jj = significant inverse association; - = no (null) association;
NA = no analysis was for this outcome was performed.
a Arrows indicate the direction in the change of the mean response of the outcome (e.g., j indicates decreased mean birth weight).
b Arrows indicate the change in risk of the outcome (e.g., | indicates an increased risk of the outcome).
c Gilliland, 1993, 1290858 and Lundin, 2009, 1291108 report overlapping data with Raleigh, 2014,2850270, which was considered the most updated data.
dLeonard, 2008, 1291100 and Sakr, 2009,2593135 report overlapping data with Steenland and Woskie, 2012,2919168, which was considered the most updated data.
eSteenland and Woskie, 2012, 2919168 was rated medium confidence for comparisons with the DuPont referent population and low confidence for comparisons with the U.S.
population.
fWinquist and Steenland, 2014,2851142 was rated medium confidence for hypertension and low confidence for coronary heart disease.
3-157
-------
APRIL 2024
Since publication of the 2016 PFOAHESD {U.S. EPA, 2016, 3603279}, 48 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 {U.S. EPA, 2024, 11414343}. Two of the
publications {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,
3-158
-------
APRIL 2024
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;
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 3-31, Figure 3-32, and Figure 3-33.
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 iminformative {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 iminformative {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 iminformative 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 (see Appendix, {U.S. EPA, 2024, 11414343}). 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.
3-159
-------
APRIL 2024
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 3-31. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cardiovascular Effects
Interactive figure and additional study details available on HAWC.
3-160
-------
APRIL 2024
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 3-32. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Cardiovascular Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-161
-------
APRIL 2024
i\e° ^
9S>
^ ,,C
&
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 -
+
+
-*
+
+
+
+
+
Legend
a
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
ta
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-33. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Cardiovascular Effects (Continued)
Interactive figure and additional study details available on HAWC.
3-162
-------
APRIL 2024
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 Appendix, {U.S. EPA,
2024, 11414343}). 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 Appendix, {U.S. EPA, 2024, 11414343}). 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;
3-163
-------
APRIL 2024
Mitro, 2020, 6833625; Liao, 2020, 6356903; Lin, 2020, 6311641; Liu, 2018, 4238514; Liu,
2018, 4238396; Mi, 2020, 6833736; Pitter, 2020, 6988479; Yang, 2018, 4238462}.
Of the 10 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 (P 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 nonsignificant association with SBP (P
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 (P: 2.18; 95% CI: 1.38, 2.98) and SBP (P: 1.69; 95% CI: 0.25, 3.13). After
stratification by sex, a positive association was observed in men only for DBP (P: 1.48; 95% CI:
0.58, 2.37) and in women only for SBP (P: 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 (P: 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 (P: 0.34;
95% CI: 0.21, 0.47) and SBP (P: 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 (P: 0.46; 95% CI: 0.19, 0.73)
and larger increases in DBP among women (P: 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 (P: 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}.
3-164
-------
APRIL 2024
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}
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 11 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) versus 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 with 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
3-165
-------
APRIL 2024
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
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 with 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 with 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.
3-166
-------
APRIL 2024
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 with the first quartile. No associations were observed with CHF and angina pectoris.
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
nonsignificant 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 (P: -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 (P 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}. Onq 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 provide consistent evidence for an association between
PFOA and blood pressure. 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 Appendix, {U.S. EPA, 2024, 11414343}). 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 versus the 1st
quartile (Rate Ratio (RR): 2.63; 95% CI: 1.06, 6.56). No association was observed with 10-year
3-167
-------
APRIL 2024
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 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. Evidence for changes in serum lipids was synthesized by population (i.e., children,
pregnant women, adults, occupational populations), and there may be differences in the
interpretation of an effect depending on age. For example, while elevated levels of TC, LDL, and
triglycerides are associated with increased cardiovascular risks in adults, serum lipid changes in
children are age-dependent and fluctuate during puberty {Daniels, 2008, 6815477}. There are 22
epidemiological studies (24 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 23 studies are shown in Figure 3-34. Results from studies summarized in
the 2016 PFOA HESD are described in Table 3-11 and below.
In the 2016 Health Assessment {U.S. EPA, 2016, 3603279} for PFOA, there was relatively
consistent and strong 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
14 Olsen {,2003, 1290020} is the peer-review paper of Olsen {,2001, 10228462} and Olsen {,2001, 10240629}.
3-168
-------
APRIL 2024
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}.
3-169
-------
APRIL 2024
Legend
| 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
Costa et al., 2009, 1429922-
-
——I
+
~.
i
+
i
+
1—
+
~
R
Emmett et al., 2006, 1290905 -
-
+
-
-
-
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 -
+
++ ++
+
+
+
+
+
Jain, 2014, 2969807-
+
+
+
-
+
+
~
Lin et al., 2009, 1290820-
+
+
+
+
+
+
+
+
Maisonet et al., 2015, 3981585 -
+
+
+*
+•
+
+
+
+*
Nelson etal., 2010, 1291110-
++
+
+
+
++
+
+
+
Olsen and Zobel, 2007, 1290836 -
+
+
+*
-
+
+
+
Olsen et al., 2000, 1424954 -
-
+
+
+
+
Olsen etal., 2001, 10228462-
+
-*
+
+
+
+
-*
Olsen et al., 2003, 1290020-
+
+
+
+
+
Sakretal., 2007, 1291103 -
+
+
+
+
+
+
+
+
Sakr et al., 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 3-34. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Serum Lipids Published Before 2016 (References from 2016 PFOA
HESD)
Interactive figure and additional study details available on HAWC.
3-170
-------
APRIL 2024
Table 3-11. Associations Between Elevated Exposure to PFOA and Serum Lipids from Studies Identified in the 2016 PFOA
HESD
Reference, Confidence
Study Design
Population
TCa
HDLa
LDLa
TGa
High Cholesterolb
Costa, 2009, 1429922
Mixed0
Cohort
Occupational
tt
4
NA
t
NA
Eriksen, 2013, 2919150
Medium
Cross-sectional
Adults
tt
NA
NA
NA
NA
Fisher, 2013, 2919156
Medium
Cross-sectional
Adults
-
-
-
-
NA
Fitz-Simon, 2013, 2850962
Mixed0
Cohort
Adults
t
1
t
-
NA
Frisbee, 2010, 1430763
Mixed0
Cross-sectional
Children
tt
-
tt
tt
NA
Fu, 2014, 3749193
Low
Cross-sectional
Adults and children
tt
-
t
t
NA
Geiger, 2014, 2850925
Medium
Cross-sectional
Adolescents
tt
II
tt
-
NA
Lin, 2009, 1290820
Medium
Cross-sectional
Adults
NA
t
NA
-
NA
Maisonet, 2015, 3981585
Mixed0
Cohort
Children
4
-
-
1
NA
Nelson, 2010, 1291110
Medium
Cross-sectional
Adults
tt
1
t
NA
NA
Olsen, 2000, 1424954
Low
Cross-sectional
Occupational
t
II
-
NA
NA
Olsen, 2003, 1290020
Low0
Cohort
Occupational
tt
NA
NA
tt
NA
Olsen and Zobel, 2007,
1290836
Low
Cross-sectional
Occupational
t
II
t
tt
NA
3-171
-------
APRIL 2024
Reference, Confidence
Study Design
Population
TCa
HDLa
LDLa
TGa
High Cholesterolb
Sakr, 2007, 1291103
Medium
Cross-sectional
Occupational
tt
4
tt
t
NA
Sakr, 2007, 1430761
Mixed0
Cohort
Occupational
tt
44
t
-
NA
Starling, 2014, 2850928
Mixed0
Cohort
Children
t
t
t
-
NA
Steenland, 2009, 1291109
Mixed0
Cross-sectional
Adults
tt
t
t
t
NA
Steenland, 2015, 2851015
Low
Cohort
Occupational
NA
NA
NA
NA
-
Timmerman, 2014, 2850370
Medium
Cohort
Children
NA
NA
NA
t
NA
Winquist and Steenland, 2014, Cohort
2851142
Mixed0
Occupational
NA
NA
NA
NA
tt
Notes'. HDL = high-density lipoprotein cholesterol; LDL = low-density lipoprotein; NA = no analysis was for this outcome was performed; TC = total cholesterol;
TG = triglycerides; | = nonsignificant positive association; ft = significant positive association; j = nonsignificant inverse association; jj = significant inverse association;
- = no (null) association.
a Arrows indicate the direction in the change of the mean response of the outcome (e.g., j indicates decreased mean birth weight).
b Arrows indicate the change in risk of the outcome (e.g., | indicates an increased risk of the outcome).
c Olsen {, 2001, 10228462} and Olsen {, 2001, 10240629} report data overlapping with Olsen {, 2003, 1290020}, which was considered the most updated information.
Jain et al., 2014,2969807 was not included in the table due to their uninformative overall study confidence ratings.
3-172
-------
APRIL 2024
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 PFOA HE SD {U.S. EPA, 2016, 3603279}, 64 new
epidemiological studies (65 publications)15 report on the association between PFOA exposure
and serum lipids. Except for 10 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 3-35,
Figure 3-36, Figure 3-37.
On the basis of 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, 10 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 et al. {, 2019, 5080195} counted as two studies, one in adolescents and one in adults.
3-173
-------
APRIL 2024
10176518; Dalla Zuanna, 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
(adolescent portion) {Dong, 2019, 5080195}.
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 (see Appendix, {U.S. EPA, 2024, 11414343}). 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.
3-174
-------
APRIL 2024
^e<
-------
APRIL 2024
,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 3-36. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued)
Interactive figure and additional study details available on HAWC.
3-176
-------
APRIL 2024
Vvsi © c,e^ \\ o°
**%# a""*
,o©
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-
+
+
+*
+
+
+
+
+*
Legend
Q
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 3-37. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA and Serum Lipids (Continued)
Interactive figure and additional study details available on HAWC.
3-177
-------
APRIL 2024
3.4.3.1.2.3 Findings From Children
Results for the studies that examined TC in children are presented in the Appendix {U.S. EPA,
2024, 11414343}. 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 10 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 Appendix, {U.S. EPA,
2024, 11414343}) {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.
3-178
-------
APRIL 2024
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 Appendix, {U.S. EPA, 2024,
11414343}). 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 Appendix, {U.S. EPA, 2024, 11414343}). 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
nonsignificant 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 Appendix, {U.S. EPA, 2024, 11414343}).
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.
3-179
-------
APRIL 2024
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
One high confidence study {Gardener, 2021, 7021199} and 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 Appendix, {U.S.
EPA, 2024, 11414343}) {Matilla-Santander, 2017, 4238432; Skuladottir, 2015, 3749113}. One
medium confidence study in a high-exposure community in Italy {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
Appendix, {U.S. EPA, 2024, 11414343}). No association between PFOA and TC was observed
in a cohort of pregnant women in the United States {Gardener, 2021, 7021199} or 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, nonsignificant 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 Appendix, {U.S. EPA, 2024,
11414343}).
Three medium confidence studies examined PFOA and HDL in pregnant women (Starling, 2017,
3858473; Dalla Zuanna, 2021, 7277682; Yang, 2020, 7021246;) and two observed positive
statistically significant associations (see Appendix, {U.S. EPA, 2024, 11414343}) {Starling,
2017, 3858473; DallaZuanna, 2021, 7277682}. Starling et al. {, 2017, 3858473} reported a
positive association between maternal PFOA serum concentrations (collected during 20 to
34 weeks of pregnancy with a median of 27 weeks) and HDL in a United States cohort. Dalla
Zuanna {, 2021, 7277682} observed significant positive associations when considering blood
samples across all trimesters of pregnancy in a high-exposure community in Italy. The
association was consistent, but no longer significant, when trimesters were modeled individually.
3-180
-------
APRIL 2024
{Yang, 2020, 7021246} observed a null association between PFOA exposures and HDL levels
measured in early pregnancy.
One high confidence, one medium confidence, and three low confidence studies examined the
association between PFOA and triglycerides in pregnant women. The high confidence study
reported a significant increasing trend for triglycerides with increasing PFOA exposure quartile
in a cohort of pregnant women from the United States {Gardener, 2021, 7021199}. 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 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 3-35, Figure 3-36, and Figure 3-37). 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 10 medium confidence studies, eight reported positive associations (Figure 3-39, Figure
3-40, Figure 3-41, Figure 3-42). 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, including an increased risk of
high cholesterol (Figure 3-38). 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 Appendix,
3-181
-------
APRIL 2024
{U.S. EPA, 2024, 11414343}). 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-Vargas et al. {, 2019, 5080588} were free of diabetes for at least
10 years of follow-up. It is unclear whether 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 13 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 with 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 whether
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.
3-182
-------
APRIL 2024
Reference, FxDosure
Confidence , - Study Design Exposure Levels Sub-population Comparison EE
n„»: MaulX
Rating
Effect Estimate
0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
CanovaetaT serum Cross-sectional median=35.8 ng/mL - OR [for02 vs. Q1)
(2020, (25th-75th percentile:
7021512), 13.6-78.8 ng/mL) 12
Medium
OR [for Q3 vs. Q1]
1.25
OR [for Q4 vs. Q1)
1.46
females OR [for 02 vs. 01]
1.12
OR [for 03 vs. 01]
1.19
OR [for 04 vs. 01]
138
males OR [for 02 vs. 01]
1.27
OR [for 03 vs. 01]
1.22
OR [for 04 vs. 01]
1.48
0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
Figure 3-38. Odds of High Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on HAWC
3-183
-------
APRIL 2024
Reference, FXDOSUre
Confidence ,, L Study Design Exposure Levels Sub-population Comparison EE
r»_»: Matrix
Rating
Effect Estimate
-10 -5 0 5 10 15 20 25 30 35 40
Canova et al. serum Cross-sectional median=35.8 ng/mL - regression
(2020, (25th-75th percentile: coefficient [per
7021512), 13.6-78.8 ng/mL) Hn(PFOA) mg/mL 194
Medium increase PFOA]
•
females regression
coefficient [per
l-ln(PFOA) mg/mL 164
increase PFOA]
~
regression
coefficient [for Q2
vs. Q1 PFOA] 111
regression
coefficient [for Q3
vs. Q1 PFOA] 2.8
—
regression
coefficient [for 04
vs. Q1 PFOA] 5.75
—
males regression
coefficient [per
Hn(PFOA) mg/mL 1.77
increase PFOA]
regression
coefficient [for 02
vs. 01 PFOA] 3.51
regression
coefficient [for 03
vs. 01 PFOA] 2.41
regression
coefficient [for 04
vs. Q1 PFOA] 5.4
—
Costa et al. serum Cohort production workers - Regression
(2009, (2007): median=3.89 coefficient
1429922), ug/mL (25th percentile: (exposed vs all 21.7
Medium 2.18-18.66 ug/mL) workers)
Regression
coefficient (per unit
increase in PFOA) 0.03
-10 -5 0 5 10 15 20 25 30 35 40
Figure 3-39. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA
Interactive figure and additional study details available on HAWC
3-184
-------
APRIL 2024
Reference, FXDOsure
Confidence . Study Design Exposure Levels Sub-population Comparison EE
n.,: Matrix
Rating
Effect Estimate
-10 0 10 20 30 40
Donat-Vargas plasma Cohort Baseline median: 2.9 - prospective regression coefficient (per
etal. (2019, (ng/ml) (25th-75th 1-SD 1.19 ng/mL PFOA) -0.04
5080588), percentile:2.2-4.2 ng/ml)
i
1
i
f
Medium Follow-up median: 2.7 prospective regression coefficient (mean
(ng/ml) (25th-75th change) for tertile 2 vs tertile 1 PFOA 0
percentile. 1.9-3.6 ng/ml) prospective regression coefficient (mean
change) for tertile 3 vs tertile 1 PFOA 0.16
i
i
i
!
I
h
regression coefficient [per 1-SD 1.38
ng/mL PFOA -0-12
i
I
f
regression coefficient (mean change) for
tertile 2 vs tertile 1 PFOA 0.03
i
1
i
f
regression coefficient (mean change) for
tertile 3 vs tertile 1 PFOA -0.11
i
1
i
I
median: 2.9 (ng/ml) baseline regression coefficient (per 1-SD 145
(25th-75th ng/mL PFOA) -0-19
percentile:2.2-4.2 ng/ml)
i
1
i
!
regression coefficient (mean change) for
tertile 2 vs tertile 1 PFOA 0.01
i
1
I
\
regression coefficient (mean change) for
tertile 3 vs tertile 1 PFOA -0.29
i
1
\
I
median: 2.7 (ng/ml) follow-up regression coefficient (per 1-SD 1.27
(25th-75th percentile: ng/mL PFOA) -0.03
1.9-3.6 ng/ml)
I
t
regression coefficient (mean change) for
tertile 2 vs tertile 1 PFOA 0.1
1
t
regression coefficient (mean change) for
tertile 3 vs tertile 1 PFOA 0.1
—i—
t
Dong et al. serum Cross-sectional Not reported Adults (20-80 years) in Regression Coefficient (change in TC
(2019, NHANES 2003-2004 per unit increase in serum PFOA) 1-6
5080195),
i
i
Medium Adults (20-80 years) in Regression Coefficient (change in TC
NHANES 2005-2006 per unit increase in serum PFOA) 0.7
i
V
I
Adults (20-80 years) in Regression Coefficient (change in TC
NHANES 2007-2008 per unit increase in serum PFOA) 1 -5
i
i
Adults (20-80 years) in Regression Coefficient (change in TC
NHANES 2009-2010 per unit increase in serum PFOA) 0.9
i
*~-
i
Adults (20-80 years) in Regression Coefficient (change in TC
NHANES 2011-2012 per unit increase in serum PFOA) 4.1
t
i
Adults (20-80 years) in Regression Coefficient (change in TC
NHANES 2013-2014 per unit increase in serum PFOA) 2.9
i
i
Median: 3.0 ng/ml: Mean Adults (20-80 years) Regression Coefficient (change in TC
± SD: 3.7 ± 3.4 ng/ml per unit increase in serum PFOA) 1.5
I
1
-10 0 10 20 30 40
Figure 3-40. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on HAWC
3-185
-------
APRIL 2024
Reference, Exoosure
Confidence J?. . Study Design Exposure Levels Sub-population Comparison EE
Rating Mainx
Effect Estimate
-10 0 10 20 30 40
Fan et al. (2020, serum Cross-sectional median=2.05 ug/L (25th-75th percentile: 1.31-3.10 - regression coefficient [per
7102734), ug/L) Hog(10) increase in PFOA] 6.74
Medium
!
Fisher etal. plasma Cross-sectional geometric mean (SD) = 2.46 ug/L (1.83) - regression coefficient (per In
(2013,2919156), unit increase PFOA) 0.03
Medium
\
i
Han et al. (2021. serum Case-control Cases: median=10.05 ng/mL (25th - 75th percentile: - regression coefficient (per
7762348), 6.75 -17.05 ng/mL); Controls: median=11.40 ng/mL Iog10 ng/mL Increase in 0.01
Medium (25th - 75th percentile: 9.20-17.40 ng/mL) PFOA)
+
i
Jain et al. (2019. serum Cross-sectional Geometric mean=2.5 ng/ml; 25th - 75th percentiles: Non-obese regression coefficient (per
5080642), 1.5-4.1; SD 2.1 females Hog10 unit change in PFOA) o.01
Medium
\
i
Obese females regression coefficient (per
1-log10 unit change in PFOA) o
i
+
1
Geometric mean=3.4 ng/ml; 25th - 75th percentiles: Non-obese males regression coefficient (per
2.3 - 5.3; SD 1.9 1-log10 unit change in PFOA) -0.01
+
i
Obese males regression coefficient (per
1-log10 unit change in PFOA) 0.05
+
i
Lin et al. (2019, plasma Cohort and median=4.9 ng/ml (25th-75th percentile: 3.5 - 6.7) participants who Regression coefficient (mean
5187597), cross-sectional were not on difference) (per doubling of 6.09
Medium lipid-lowering baseline plasma PFOA)
1
i
i
medication Regression coefficient (mean
difference) (for Q2 vs 01) 2
i
i _
i •
i
Regression coefficient (mean
difference) (for 03 vs 01) 10.13
i
i
i ~
i
Regression coefficient (mean
difference) (for Q4 vs Q1) 13.36
i
i
i
i
Lin et al. (2020, serum Cross-sectional median=8.6 ng/mL (25th-75th percentile: 6.2-11.6) Participants not Regression Coefficient (for
6988476), taking lipid Q2 vs Q1 ) 2.48
Medium lowering
i.
i •
i
medication Regression Coefficient (for
03 vs 01 ) 2.88
i
i _
i •
Regression Coefficient (for
04 vs Q1 ) 4.04
i _
i •
i
Nelson et al. serum Cross-sectional median: 3.9 ug/L (range: 0.1-37.3 ug/L) 20- to regression coefficient (per
(2010,1291110), 80-year-olds ug/L increase in PFOA) 1.22
Medium
!•
i
regression coefficient [for 04
(5.5-37.3 ug/L) vs. Q1 9.8
(0.1-2.7 ug/L) PFOA]
1
i
i •
i
-10 0 10 20 30 40
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 HAWC
3-186
-------
APRIL 2024
Reference,
Confidence
Rating
Exposure
Matrix
Study Design
Exposure Levels
Sub-population
Comparison
EE
-10 -5
0
5
Effect Estimate
10 15 20
25
30
35 40
Olsen et al.
(2003,
1290020),
Medium
serum
Cohort
Antwerp Mean (SD) =
1.03 ppm (1.09);
Decatur=1.90 ppm (1.59)
Males
Regression
coefficient (per unit
increase in PFOA)
0.03
1
1
•
1
i
Sakr et al.
(2007,
1291103),
Medium
serum
Cross-sectional
Mean (SD): 0.428 ppm
(0.86); Min-max:
0.005-9.550 ppm
Regression
coefficient (per
ppm increase
PFOA)
4.04
i
i
i
i
•
Workers not on
lipid-lowering
medications
Regression
coefficient (per
ppm increase
PFOA)
5.52
i
i
i
I
•
Steenland et
al. (2009,
1291109),
Medium
serum
Cross-sectional
median=26.6 ng/mL (min
- max: 0.25-17556.6
ng/mL)
regression
coefficient (for
decile 2 vs. decile
1)
0.01
i
•
i
regression
coefficient (for
decile 3 vs. decile
1)
0.02
i
•
i
i
regression
coefficient (for
decile 4 vs. decile
1)
0.03
i
i
•
i
regression
coefficient (for
decile 5 vs. decile
1)
0.04
i
•
i
i
regression
coefficient (for
decile 6 vs. decile
1)
0.03
i
i
•
i
regression
coefficient (for
decile 7 vs. decile
1)
0.04
i
•
i
i
regression
coefficient (for
decile 8 vs. decile
1)
0.04
i
i
•
i
regression
coefficient (for
decile 9 vs. decile
1)
0.04
i
•
i
i
regression
coefficient (per Hn
ng/mL increase in
PFOA)
0.01
i
i
~
i
Regression
coefficient for
PFOA Decile 10 vs
Decile 1
0.05
i
•
i
-10 -5
0
5
10 15 20
25
30
35 40
Figure 3-42. Overall Levels of Total Cholesterol in Adults from Epidemiology Studies
Following Exposure to PFOA (Continued)
Interactive figure and additional study details available on HAWC.
Six medium confidence studies examined PFOA and LDL in adults, and all reported positive
associations (Figure 3-35, Figure 3-36, and Figure 3-37). Higher PFOA was significantly
associated with higher LDL at baseline in the DPPOS {Lin, 2019, 5187597} (see Appendix,
{U.S. EPA, 2024, 11414343}). 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-III-containing fractions of IDL and LDL were consistent with
the positive associations reported for LDL.
3-187
-------
APRIL 2024
Consistent with these findings, nine of the 13 low confidence studies report positive associations
between PFOAand 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 13 low confidence studies examined PFOA and HDL or
clinically defined low HDL in adults (). 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 Appendix, {U.S. EPA, 2024,
11414343}). 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 16 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; Zare Jeddi, 2021,
3-188
-------
APRIL 2024
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 Appendix, {U.S. EPA, 2024, 11414343}).
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-Ill-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
whether 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.
3-189
-------
APRIL 2024
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} {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 from those in the general adult population. It is unclear whether
well-known biases in occupational studies such as "healthy worker effect" may have attenuated
the association between PFOA and an unfavorable serum lipid profile. Additional higher-quality
occupational studies are needed to improve hazard identification among workers.
3-190
-------
APRIL 2024
3.4.3.2 Animal Evidence Study Quality Evaluation and Synthesis
There are three studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and seven
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 10 studies are shown in Figure
3-43.
Blake etal., 2020, 6305864
Butenhoff etal., 2012, 2919192-
Cope et al., 2021, 10176465
Guo etal., 2021, 9960713-
Loveless et al., 2008, 988599 -
NTP, 2019, 5400977
NTP, 2020, 7330145
Quist et al., 2015, 6570066-
NR
NR
Yan et al., 2014, 2850901 -
-
+
NR
van Esterik et al., 2015, 2850288 -
~~
NR
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)
NRl Not reported
Figure 3-43. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure 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,
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 3-44).
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,
3-191
-------
APRIL 2024
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 to GD 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 nonsignificant 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 with HFD control. Males fed the LFD exhibited nonsignificant 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 Appendix, {U.S. EPA, 2024,
11414343}). 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 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 be 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
3-192
-------
APRIL 2024
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 with 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.
3-193
-------
APRIL 2024
PFOA Cardiovascular Effects - Serur
Lipid,
Endpoint
Study Name
Study Design
Observation Time
Animal Description
0 No significant change^ Significant increase ~ Significant decrease |
High Density Lipoprotein (HDL)
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 etal.. 2021,10176465
developmental (GD1.5-17.5)
PND22
F1 Mouse. CD-1 (o. N=7)
—~
F1 Mouse, CD-1 (2. N=7)
—~
PNW18
F1 Mouse, CD-1 (o- N=8)
F1 Mouse. CD-1 (: , N=7)
—~
Quistetal.. 2015, 6570066
developmental (GD1-17)
PND91
F1 Mouse. CD-1 (2, N=7)
—~
Yan etal., 2014. 2850901
short-term (28d)
23d
Mouse, BALB/c (;?, N=6)
V V-
V
Loveless et al.. 2008, 988599
short-term (29d)
29d
Mouse. Crl:CD-1(ICR)BR 0. N=20)
—*
_5Z v
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
—V-
—V
-V .
Non-HDL Cholesterol
Loveless et al„ 2008, 988599
short-term (29d)
29d
Mouse, Cr1:CD-1(lCR)BR (o, N=20)
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
—V
Low Density Lipoprotein (LDL)
Blake et al.. 2020. 6305864
developmental (GD1.5-11.5)
GD11.5
P0 Mouse. CD-1 (2, 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)
PND22
F1 Mouse, CD-1 N=7)
—A
F1 Mouse. CD-1 (2, N=7)
—•
PNW18
F1 Mouse, CD-1 (o. N=8)
—~
F1 Mouse, CD-1 (2. N=7)
—4
Quistetal., 2015. 6570066
developmental (GD1-17)
PND91
F1 Mouse. CD-1 (: , N=7)
Yan etal., 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 (2, N=5)
developmental (GD1.5-17.5)
GD17.5
P0 Mouse. CD-1 (2, N=4-6)
Cope el al , 2021; 10176465
developmental (GD1.5-17.5)
PND22
F1 Mouse, CD-1 (,j\ N=7)
—•
F1 Mouse. CD-1 (•' , N=7)
—~
PNW18
F1 Mouse. CD-1 N=8)
—«
F1 Mouse. CD-1 (2. N=7)
—~
Quistetal., 2015. 6570066
developmental (GD1-17)
PND91
F1 Mouse, CD-1 (2. N=7)
—*
Yan et al„ 2014. 2850901
short-term (28d)
23d
Mouse. BALB/'c N=6)
V ¥-
—w
Loveless et al., 2008, 988599
short-term (29d)
29d
Mouse, Crl:CD-1(ICR)BR (N=20)
-V—w
Rat, Sprague-Dawley Cri:Cd(Sd)(Br) N=10)
—SA
—v
NTP. 2019, 5400977
short-term (28d)
29d
Rat, Sprague-Dawley N=10)
V ~ V V
Rat, Sprague-Dawley (' , N=9-10)
NTP. 2020, 7330145
chronic {GD6-PNW21)
16wk
F1 Rat, Sprague-Dawley {v. N=10)
chronic (GD6-PNW107)
16wk
F1 Rat, Sprague-Dawley N=10)
—V . V
F1 Rat, Sprague-Dawley (2, N=10)
V .
chronic (PND21-PNW21)
16wk
F1 Rat, Sprague-Dawley 0, N=10)
chronic {PND21-PNW107)
16wk
F1 Rat, Sprague-Dawley 0, N=10)
F1 Rat, Sprague-Dawley N=10)
Triglycerides
Blake et al.. 2020.6305864
developmental (GD1.5-11.5)
GD11.5
P0 Mouse. CD-1 (2, N=5)
—5? V
developmental (GD1.5-17.5)
GD17.5
P0 Mouse. CD-1 (2, N=4-6)
—. w
Quist et al., 2015. 6570066
developmental (GD1-17)
PND91
F1 Mouse, CD-1 (r . N=7-10)
Cope etal., 2021, 10176465
developmental (GD1.5-17.5)
PND22
F1 Mouse. CD-1 0, N=7)
—V
F1 Mouse. CD-1 (•-, N=7)
V-
—V
PNW18
F1 Mouse. CD-1 (o. N=8)
F1 Mouse. CD-1 ('\ N=7)
Yan etal., 2014, 2850901
short-term (28d)
28d
Mouse, BALB/c 0, N=6)
A W-
—w
Loveless et al., 2008, 988599
short-term (29d)
29d
Mouse, Crl:CD-1(ICR)BR (o, N=20)
Sf—V
Rat, Sprague-Dawley Cri;Cd(Sd)(Br) (<>, N=10)
-V—V
NTP. 2019, 5400977
short-term (28d)
29d
Rat, Sprague-Dawley (;?, N=10)
V V V V
-V
Rat, Sprague-Dawley (' . N=9-10)
NTP, 2020, 7330145
chronic (GD6-PNW21)
16wk
F1 Rat, Sprague-Dawley 0, N=10)
V V
chronic (GD6-PNW107)
16wk
F1 Rat, Sprague-Dawley 0, N=10)
—V V V
F1 Rat, Sprague-Dawley {•}. N=10)
chronic (PND21-PNW21)
16wk
F1 Rat, Sprague-Dawley 0, N=10)
V V
chronic (PND21-PNW107)
16wk
F1 Rat. Sprague-Dawley N=10)
V V »
F1 Rat, Sprague-Dawley (y. N=10)
0.01
0.1
1
10
100
Concentration (mg/kg/day)
Figure 3-44. Serum Lipid 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; Po = parental generation; PNW = postnatal week; Fi = first generation; PND = postnatal day; d = day;
wk = week.
3-194
-------
APRIL 2024
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 eight
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
cardiovascular effects. A summary of these studies by mechanistic data category (see Appendix
A, {U.S. EPA, 2024, 11414343}) and source is shown in Figure 3-45.
Mechanistic Pathway Animal Human In Vitro Grand Total
Angiogenic, Antiangiogenic, Vascular Tissue Remodeling
0
1
0
1
Atherogenesis 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
8
Figure 3-45. Summary of Mechanistic Studies of PFOA and Cardiovascular Effects
Interactive figure and additional study details available on HAWC.
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.
3-195
-------
APRIL 2024
One study summarized in the 2016 PFOA HESD {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
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 with 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
3-196
-------
APRIL 2024
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 [xM PFOA.
Bassler et al. {, 2019, 5080624} focused on several disease biomarkers, including plasminogen
activator ihhibitor-1 (PAI-1), an indicator of clot formation 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.
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 versus 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.
3.4.3.4 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,
and TC {Nelson, 2010, 1291110; Liu, 2018, 4238514; Dong, 2019, 5080195; Jain, 2019,
5080642; Fan, 2020, 7102734; Steenland, 2009, 1291109; Eriksen, 2013, 2919150; Fitz-Simon,
2013, 2850962; Donat-Vargas, 2019, 5080588; Lin, 2019, 5187597; Canova, 2020, 7021512;
Lin, 2020, 6988476; Winquist, 2014, 2851142}. 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.
On the basis of 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 nonsignificant 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 PFOA HESD. Differences in
findings from occupational studies between the 2016 PFOA HESD and this review may be
attributable to limitations of occupational studies in this review. Similar to the 2016 PFOA
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
3-197
-------
APRIL 2024
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
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
3-198
-------
APRIL 2024
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
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 3-12).
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. The consistent findings for
serum lipids are also supported by evidence of associations with blood pressure in adult
populations in high and medium confidence studies.
3-199
-------
APRIL 2024
Table 3-12. Evidence Profile Table for PFOA Exposure and 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
2 High confidence
studies
27 Medium confidence
studies
22 Low confidence
studies
19 Mixed* confidence
studies
Examination of serum
lipids included measures
of TC, LDL, HDL, TG,
and VLDL. In studies of
serum lipids in adults
from the general
population (29), there is
evidence of positive
associations with TC
(13/15) in medium
confidence studies.
Positive associations
were also observed for
LDL (6/8) in medium
confidence studies, and
mostly null, but some
positive associations with
TG (4/11) in medium
confidence studies.
Evidence from studies of
children (19) was mixed,
and observed associations
often failed to reach
significance, but findings
were mostly positive for
TC (10/19). In studies of
pregnant women (6),
evidence indicated
positive associations with
TC (3/4) and HDL (2/4)
but no other serum lipid
• High and medium
confidence studies
• Consistent findings
of positive
associations with
serum lipid
measures in adults
from the general
population
• Coherence of
findings across
serum lipids serum
lipid effects
• Low confidence studies
• Lnconsistent findings in
children, likely due to
variations in measured
exposure windows
• Lnconsistent findings by
sex or health status
0©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
occupational studies were
generally consistent with
studies of general
population adults,
particularly for increases
in TC. Results from some
studies of children and
pregnant women also
observed positive effects
for TC and TG, however,
interpretations of changes
in serum lipids for
children are less clear.
High and medium
confidence studies of
adults reported positive
associations with blood
pressure and risk of
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. The consistent
findings for serum lipids
are also supported by
evidence of associations
with blood pressure in
adult populations in high
and medium confidence
studies
Human relevance and
other inferences:
No specific factors are
noted.
3-200
-------
APRIL 2024
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
measures. In occupational
studies (10), positive
associations or increased
risks were observed for
TC and high cholesterol
(8/10), LDL (3/5), and
TG (4/8). Findings on
HDL in occupational
studies were mixed.
Blood pressure and
hypertension
2 High confidence
studies
18 Medium confidence
studies
7 Low confidence
studies
Studies examining
changes in blood
pressure, including DBP
and SBP, and risk for
hypertension in general
population adults (15),
showed consistent
positive associations for
SBP (5/6), DBP (6/6),
combined BP (2/2), and
hypertension (9/10) in
high and medium
confidence studies. In
studies of children (9),
mixed results were
observed for SBP (7),
DBP (5), and general BP
(3). The only study
examining hypertension
in children reported a
positive, dose-dependent
association. In
occupational studies, one
study reported a positive
association for
hypertension (1/3). In the
• High and medium
confidence studies
• Consistent findings
of 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
• Lnconsistent findings in
children, likely due to
variation in measured
exposure windows
hypertension, though
other medium and low
confidence studies
reported nonsignificant
associations. Observed
effects were inconsistent
for CVD and imprecise
for atherosclerotic
changes across all study
populations.
3-201
-------
APRIL 2024
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
only study of pregnant
women (1), a positive
association was reported
with hypertension.
Hypertension analyses
provided evidence of
modification by sex, with
males having higher risk
in some studies.
Cardiovascular disease
1 High confidence study
6 Medium confidence
studies
6 Low confidence
studies
CVD measures included
CHD, stroke, angina,
heart attack, MVD, IHD,
PAD, and arrhythmia.
Studies of general
population adults (9)
reported mixed results.
The most commonly
investigated endpoints
were CHD (5), general
CVD (5), and stroke (3);
in all cases, positive and
inverse associations were
observed. A significant
positive association for
risk of heart attack was
observed in a medium
confidence study (1/1).
Observations for other
outcomes were limited to
nonsignificant, imprecise
findings by singular
studies. In occupational
studies (4), consistent
inverse associations were
observed for IHD (3/3),
1 High and medium
confidence studies
• Low confidence studies
• Lnconsistent findings for
CVD-related outcomes
• Lmprecision of findings
3-202
-------
APRIL 2024
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
but results remained
mixed for stroke (1/2).
Atherosclerotic
changes
1 High confidence study
3 Medium confidence
studies
3 Low confidence
studies
In studies of children (2),
one study reported
significantly increased
associations in brachial
artery distensibility (1/1).
No significant
associations were
observed for CIMT
among Taiwanese
children (1/1) or pulse
wave velocity among
American children (1/1).
Studies of adults (4)
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
4 Medium confidence
studies
Significant decreases in
serum TC were observed
in 4/7 studies that
examined this endpoint,
regardless of species, sex,
or study design. In three
developmental studies, no
changes were observed in
mice. Similar decreases
• High and medium
confidence studies
• Consistency of
findings across
species, sex, or study
design
• Dose-response
relationship
• Lncoherence of findings
in other cardiovascular
outcomes
• Biological significance
of the magnitude of
effect is unclear
0©O
Moderate
Evidence based on six
high or medium
confidence studies
observed that PFOA
affects serum lipids in
3-203
-------
APRIL 2024
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
were observed in serum observed within
animal models. The most
TG (6/7). In a multiple studies
consistent results are for
developmental study.
total cholesterol and
decreased serum TG were
triglycerides, although
observed in mice at PND
direction of effect can
22 but not during
vary by dose. The
adulthood. In a short-
biological significance of
term exposure study.
the decrease in various
female rats were given
serum lipid levels
10-fold higher doses of
observed in these animal
PFOA than males due to
models regardless of
sex differences in
species, sex, or exposure
excretion, and it was
paradigm is unclear;
found that serum TC and
however, these effects
TG were decreased in
indicate a disruption in
males but increased in
lipid metabolism. No
females. Fewer studies
effects or minimal
examined HDL and LDL,
alterations were noted for
with decreases found in
heart weight and
HDL (2/5). Three studies
histopathology in the
found no changes in
heart and aorta. However,
LDL, but one
many of the studies
developmental study in
identified may not be
mice observed increased
adequate in exposure
LDL in males at PND 22
duration to assess
but no changes during
potential toxicity to the
adulthood.
cardiovascular system.
Histopathology
2 High confidence
studies
1 Medium confidence
study
myocarditis in female rats
in the mid-dose group
No changes in heart
histopathology were
reported in two studies.
One chronic study
reported decreased
incidence of chronic
• High and medium
confidence studies
• Limited number of
studies examining
outcome
3-204
-------
APRIL 2024
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
only. No changes in aorta
histopathology were
noted in two studies.
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 of TG 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.
3-205
-------
APRIL 2024
aMixed confidence studies had split confidence determinations for different serum lipid measures with some measures rated medium confidence and others rated low confidence.
bMixed confidence studies had split confidence determinations for different subgroups of participants with some measures rated medium confidence and others rated low
confidence.
3-206
-------
APRIL 2024
3.4.4 Developmental
EPA identified 100 epidemiological and 19 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 fiminformative) confidence, and 7 were considered
uninformative (Section 3.4.4.1). Of the animal toxicological studies, 2 were classified as high
confidence, 12 as medium confidence, and 4 as low confidence, and 1 was considered mixed
{medium 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), birth defects, and fetal loss (i.e., spontaneous abortion/miscarriage and stillbirths), 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 22 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 22 studies are shown in
Figure 3-46. 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 from studies summarized in the 2016 PFOA HESD are described in
Table 3-13 and below.
3-207
-------
APRIL 2024
Andersen et al., 2010, 1429893
Apelberg et al., 2007, 1290833
Chen et al., 2012, 1332466 -
Darrow et al., 2013, 2850966
Darrow et al., 2014, 2850274
Fei et al., 2007, 1005775
Fei et al., 2008, 1290822-
Fei et al., 2008, 2349574
Fei et al., 2010, 1430760
Hamm etal.,2010, 1290814
Maisonet et al., 2012, 1332465
Monroy et al., 2008, 2349575
Nolan et al., 2009, 2349576
Nolan et al.,2010, 1290813-
Savitz et al., 2012, 1276141
Savitz et al., 2012, 1424946
Stein et al., 2009, 1290816
Stein et al., 2014, 2850277
Washino et al., 2009, 1291133
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
+
+
+
+
+
+
+
~
+
+
+
++
+
+
~
- +
+
+
+
+
~
+
++
+
+
+
+
+
+
+
+
+
+
+
+
+
•
+
+
++
+
+
++
+
+ 1
+
+
+
+
+
+
+
++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
++
11II
+
+
+
+
-I +
+
+
+
+
+
+
+
+
+
+
+
+
- -
++
++
+
+
-
+
n
-
+
+
+
¦\
+
++
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
•
+
++
++ ++
Figure 3-46. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Developmental Effects Published before 2016 (References from 2016
PFOA HESD)
Interactive figure and additional study details available on HAWC.
3-208
-------
APRIL 2024
As noted in the 2016 PFOA 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 (P per log unit increase: -104 g; 95% CI: -213, -5) in a
study of 293 infants born in Maryland in 2004-2005 (mean PFOA concentration of
0.0016 (j,g/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 (j,g/mL) was inversely associated with birth weight (P per log unit increase: -34.2 g; 95%
CI: -54.8, -13). A study of 252 pregnant women in Alberta, Canada found no statistically
significant association between PFOA concentration measured in maternal blood during the
second trimester (mean concentration of 0.0021 (j,g/mL) and birth weight {Hamm, 2010,
1290814}. In a Japanese prospective cohort of 428 infants in the Hokkaido Study on
Environment and Children's Health (2002-2005), Washino et al. {, 2009, 1291133} observed a
large nonsignificant association between PFOA concentration in maternal blood during
pregnancy (mean PFOA concentration of 0.0014 (j,g/mL) and birth weight (P per each logio
increase: -75.1 g; 95% CI: -191.8 to 41.6). Chen et al. {, 2012, 1332466} examined 429 mother-
infant pairs from the Taiwan Birth Panel Study and found no statistically significant association
between umbilical cord blood PFOA concentration (geometric mean (GM) of 0.0018 (j,g/mL) and
birth weight (P per each ln-unit increase: -19.2 g; 95% CI: -63.5, 25.1).
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 a small, nonsignificant inverse association between maternal PFOA
concentration (blood samples taken in the first and second trimester) and birth weight (P 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
measures of standardized birth weight, birth length, and infant body mass index (BMI) and body
weight at 5 and 12 months of age in DNBC participants. Andersen et al. {, 2010, 1429893} also
reported an inverse association with birth weight, but the study population overlapped with
participants reported in Fei et al. {, 2007, 1005775}. Regarding post-natal growth, they observed
a positive association between adiposity and maternal PFOA concentration based on 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
high PFAS 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 PFOA HESD, none of the
studies reported statistically significant or large magnitude associations between PFOA and
either birth weight or the risk of low birth weight. Darrow et al. {, 2013, 2850966} reported a
non-statistically significant increased risk (ORs ranging 1.3 to 1.49) for participants in the upper
three quintiles of PFOA exposure (PFOA concentrations >11.1 ng/mL) compared with the
3-209
-------
APRIL 2024
lowest (PFOA concentration >8.6 ng/mL), but results from other C8 studies reported null
associations for preterm birth. In the low confidence study {Stein, 2014, 2850277} on the C8
Health Project community population, modeled maternal serum PFOA was associated with brain
birth defects (albeit with only 13 cases), but no associations were observed for other birth
defects. 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 for most endpoints and uninformative for congenital anomalies in
Nolan etal. {,2010, 1290813}.
3-210
-------
APRIL 2024
Table 3-13. Associations Between Elevated Exposure to PFOA and Developmental Outcomes in Children from Studies
Identified in the 2016 PFOA HESD
Reference, Confidence
Study Design
Birth
Weight3
LBWb
SGAb
Gestational
Duration3
Preterm
Birthb
Birth
Defectsb
Pregnancy Lossb
PNG3
Andersen, 2010, 1429893°
Medium
Cohort
44
NA
NA
NA
NA
NA
NA
4
Apelberg, 2007, 1290833
Medium
Cross-sectional
44
NA
NA
t
NA
NA
NA
NA
Chen, 2012, 1332466d
Medium
Cohort
4
4
t
-
4
NA
NA
NA
Darrow, 2014, 2850274
Medium
Cohort
NA
NA
NA
NA
NA
NA
-
NA
Darrow, 2013,2850966
High
Cohort
-
-
NA
-
t
NA
NA
NA
Fei, 2007, 1005775°
Medium
Cohort
4
t
-
NA
tt
NA
NA
NA
Hamm, 2010, 1290814
Medium
Cohort
t
NA
4
4
t
NA
NA
NA
Maisonet, 2012, 1332465
Medium
Cohort
44
NA
NA
4
NA
NA
NA
-
Nolan, 2009, 2349576
Low
Cross-sectional
-
NA
NA
-
NA
NA
NA
NA
Nolan, 2010, 1290813
Mixed1
Cross-sectional
NA
NA
NA
-
NA
-
NA
NA
Savitz, 2012, 1276141
Medium
Cohort
NA
-
NA
NA
-
-
-
NA
Savitz, 2012, 1424946
Medium
Cohort
4
-
4
NA
t
NA
-
NA
Stein, 2009, 1290816
Cohort
NA
4
NA
NA
-
t
-
NA
3-211
-------
APRIL 2024
Reference, Confidence
Study Design
Birth
Weight3
LBWb
SGAb
Gestational
Duration3
Preterm
Birthb
Birth
Defectsb
Pregnancy Lossb
PNG3
Medium
Stein, 2014, 2850277
Low
Cohort
NA
NA
NA
NA
NA
-
NA
NA
Washino, 2009, 1291133f
Medium
Cohort
4
NA
NA
NA
NA
NA
NA
NA
Whitworth, 2012,
2349577
High
Cohort
4
NA
NA
II
NA
NA
NA
Notes'. LBW = low birth weight; NA = no analysis was for this outcome was performed; PNG = post-natal growth; SGA = small-for-gestational age; | = nonsignificant positive
association; ft = significant positive association; j = nonsignificant inverse association; jj = significant inverse association; - = no (null) association.
Apelberg et al. {, 2007, 1290900} and Monroy et al. {, 2008, 2349575} were not included in the table due to their uninformative overall study confidence ratings. Fei et al. {, 2008,
1290822}, Fei et al. {, 2008, 2349574}, and Fei et al. {, 2010, 1430760} were not included in the table because the studies only analyzed other developmental outcomes that were
more prone to measurement error (see Study Evaluation Considerations in Section 3.4.4.1.2) or were not as heavily studied (i.e., other measures of fetal growth restriction such as
birth length and head circumference and breastfeeding duration or developmental milestones, respectively).
aArrows indicate the direction in the change of the mean response of the outcome (e.g., j indicates decreased mean birth weight).
•Arrows indicate the change in risk of the outcome (e.g., | indicates an increased risk of the outcome).
cFei {,2007, 1005775} reports results from a population overlapping with Meng et al. {,2018, 4829851}, which was considered the most updated data.
dChen {, 2012, 1332466} reports results from a population overlapping with Chen et al. {, 2017, 3981292}, which was considered the most updated data.
eNolan {, 2010, 1290813} was rated uninformative for congenital abnormalities and low confidence for all other outcomes.
fWashino et al. {, 2009, 1291133} reports results from a population overlapping with Kashino et al. {, 2020, 6311632}, which was considered the most updated data.
3-212
-------
APRIL 2024
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. Further discussion of considerations for potential confounding
by co-occurring PFAS can be found in Section 5.1.
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 toward the null if ascertainment of fetal loss is not
associated with PFOA exposures (i.e., non-differential). In some situations, differential loss is
3-213
-------
APRIL 2024
possible and bias away from the null can manifest as an apparent protective effect. Fetal and
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 for Updated Literature Search
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
3-214
-------
APRIL 2024
Hokkaido Study on Environment and Children's Health birth cohort as Kashino et al. {, 2020,
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 (see Appendix, {U.S. EPA, 2024, 11414343}). 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.
3-215
-------
APRIL 2024
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 11 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;
3-216
-------
APRIL 2024
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 3-47, Figure 3-48, and Figure 3-49.
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 18 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}.
3-217
-------
APRIL 2024
^ .«
\\-j8
,c.®
•<9V
Alkhalawi et al„ 2016, 3859818 -
Ashley-Martin et al., 2017, 3981371 -
Bach et al., 2016, 3981534-
Belletal.,2018, 5041287
Bjerregaard-Olesen et al., 2019, 5083648
Buck Louis et al., 2018, 5016992
Callan et al., 2016, 3858524
Cao et al., 2018, 5080197
Chang et al., 2022, 9959688
Chen et al., 2017, 3981292
Chen et al., 2021, 7263985
Chu etal., 2020, 6315711 -) +
Eick etal., 2020, 7102797^
Espindola Santos et al., 2021, 8442216-
Gao etal., 2019, 5387135-
Gennings et al., 2020, 7643497 -
Govarts et al., 2016, 3230364 -
Gross etal., 2020, 7014743-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 3-47. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Weight Effects
Interactive figure and additional study details available on HAWC.
3-218
-------
APRIL 2024
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 3-48. 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.
3-219
-------
APRIL 2024
V6c^
C0* ^
&
Meng et a!., 2018, 4829851 -
Minatoya et al„ 2017, 3981691 -
Robledo et al„ 2015, 2851197 -
Sagiv et al., 2018, 4238410 -
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 3-49. 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.
3-220
-------
APRIL 2024
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 Appendix,
{U.S. EPA, 2024, 11414343}). 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 10 studies presenting results based on categorical data, two
studies {Meng, 2018, 4829851; Starling, 2017, 3858473} showed inverse monotonic exposure-
response relationships (Figure 3-50, Figure 3-51, Figure 3-52, and Figure 3-53).
Among the 21 studies showing some inverse 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. 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).
However, five {Chang, 2022, 9959688; Hjermitslev, 2020, 5880849; Meng, 2018, 4829851;
Sagiv, 2018, 4238410; Wikstrom, 2020, 6311677} of nine medium and high confidence studies
still reported some evidence of reductions in mean BWT based on early pregnancy biomarker
samples.
3-221
-------
APRIL 2024
Reference, _
Period0 Confidence study des'9n Matrix Sub-population Exposure levels Comparison EE
Rating
Effect Estimate
-100 -50 0 50 100 150 200
Early Bachetal. Cohort maternal Term births (GA>= median=2.0 ng/mL Regression
pregnancy (2016, serum 37 weeks) (25th-75th percentile: coefficient for 02
3981534), 1.5-2.6 ng/mL) (1.54-2.02 ng/mL) -36
High vs. Q1 (<1.54
ng/mL)
r
i
i
• i
i
i
Regression
coefficient for Q3
(2.03-2.64 ng/mL) -7
vs. Q1 (<1.54
ng/mL)
i
i
•i
i
i
Regression
coefficient for Q4
(2.65-15.10 ng/mL) 4
vs. Q1 (<1.54
ng/mL)
1
1
1*
1
1
Regression
coefficient per IQR
(1.1 ng/mL) 13
increase
1
1
1
1
1
Buck Louis et Cohort maternal - Median (25th-75th Regression
al. (2018, blood percentiles): 1.985 coefficient (change
5016992), ng/mL (1.297-3.001 in birth weight per -5,9
High ng/mL) SD increase in
log-PFOA)
1
1
1
•l
1
1
Later Bell et al. Cross-sectional blood Singleton median=1.10 ng/mL Regression
pregnancy (2018, (25th-75th percentile: coefficient (per
5041287), 0.69-1.63 ng/mL) log(PFOA+1) unit -n.6
High increase)
1
1
1
• 1
1
1
Chuetal. Cohort m^emai — median=1.538 ng/mL Regression
(2020, serum (25th percentile=0.957 coefficient (per 11n
6315711), ng/mL. 75th change in PFOA) -73.6
High percentile=2.635 ng/mL)
1
1
1
• 1
1
1
Eick at al. Cohort serum full-term births median= 0.76 ng/mL Regression
(2020, (25th-75th percentile= Coefficient [for T2
7102797), 0.46-1.12 ng/mL) (1.40-1.96 ng/ml) 62.9
High vs. T1 (<1.40
ng/ml)]
1
1
T '•
1
1
Regression
Coefficient [for T3
(>1.96 ng/ml) vs. 86 1
T1 (<1.40 ng/ml)]
f
1
1
1 *
1
1
Govartsetal. Cohort cord blood - geometric mean = 1.52 Regression
(2016, ug/L (25th-75th coefficient (per
3230364), percentile = 1.10-2.10 IQR change in -34.5
High ug/L) PFOA z-score)
1
1
1
• 1
1
1
-100 -50 0 50 100 150 200
Figure 3-50. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA
Interactive figure and additional study details available on HAWC
3-222
-------
APRIL 2024
Reference, P
Period9 Confidence Study design Matrix^ Sub-population Exposure levels Comparison EE
Rating
Effect Estimate
-200 -150 -100 -50 0 50
Early Manzano- Cohort plasma, — Mean (SD): 2.35 ng/mL (1.25 Regression coefficient
pregnancy Salgado et al. maternal ng/mL) (change in birth Q -
(2017, blood weight per doubling of
4238465) PFOA)
1
1
1
High Regression coefficient
for birth weight (02 vs 0Q
Q1)
1
• 1
1
1
Regression coefficient
for birth weight (Q3 vs
Q1)
1
~ 1
1
1
Regression coefficient
for birth weight (Q4 vs ,9
Q1)
1
• 1
1
1
Sagiv et al. Cohort maternal - median=25.7 ng/mL (IQR: Regression coefficient
(2018, blood 16.0 ng/mL) per IQR increase
4238410), "18:>
High
1
• h-
1
1
Regression coefficient
(for Q2 [4.2 -5.8 fi .
ng/mL] vs Q1 [0.3 -
4.1 ng/mL])
1
1
Regression coefficient
(forQ3vsQ1)
1
1
• 1
1
1
Regression coefficient
(forQ4vsQ1) _367
1
• 1
1
1
Later Lauritzenet Cohort maternal - Norway: median=1.62 ng/mL Regression coefficient
pregnancy al. (2017, serum (range: 0.31-7.97 ng/mL); per unit increase in fli 7
3981410) Sweden: median=2.33 ng/mL InPFOA ~81'
High (range: 0.60-6.70 ng/mL)
1
• 1
1
1
Luo et al. Cohort maternal - median {25th-75th Regression coefficient
(2021, blood, cord percentile): 3.51 ng/mL (perln-ng/mL ,
9959610), blood (2.23-4.80) increase PFOA)
High
I
• 1
i
i
Starling et al. Cohort maternal - median=1.1 ng/mL (25th Regression coefficient
(2017, serum percentile=0.7,75th (per 11n increase in .
3858473), percentile^ .6) PFOA)
High
I
• I
I
I
Regression coefficient
for fertile 2 (0.9-1.4
ng/mL) vs tertile 1
(0.1-0.8 ng/mL)
i
•—1
i
i
Regression coefficient
for tertile 3 (1.4-17.0 Q7 .
ng/mL) vs. tertile 1
(0.1-0.8 ng/mL)
i
• i
i
i
-200 -150 -100 -50 0 50
Figure 3-51. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
Interactive figure and additional study details available on HAWC
3-223
-------
APRIL 2024
Reference
Period9 Confidence Study design Sub-population ^evels^ Comparison EE
Rating
Effect Estimate
-150 -100 -50 0 50 100
Earty Wikstrom et Cohort maternal — Median-1.61 Regression
pregnancy al. (2020, serum ng/ml (25th-75th coefficient (for Q2
6311677), percentiles: vsQ1) ?7
High 1.11-2.30 ng/mL)
T
I
I
1 9
I
I
I
Regression
coefficient (for Q3
vsQ1) _41
I
I
I
¦ I
I
I
1
Regression
coefficient (for Q4
vs Q1) _90
T
1
1
• 1
1
1
j
Regression
coefficient (per Hn
ng/mL change in fio
PFOA)
T
1
1
• 1
1
1
1
Later Valvi et al. Cohort maternal - median=3.31 Regression
pregnancy (2017, serum ng/mL (25th-75th coefficient [per
3983872), percentile: doubling of serum 11
High 2.54-3.99 ng/mL) PFOA]
I
1
1
•—1
1
1
1
Yao et al Cross-sectional other. maternal exposure median: 42.83 Regression
(2021, maternal ng/mL (range: coefficient (per 1-In
9960202), serum 1.16-602.79 ng/mL increase in 252
High ng/mL) maternal serum
PFOA)
1
1
1
• 1
1
1
1
-150 -100 -50 0 50 100
Figure 3-52. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
Interactive figure and additional study details available on HAWC.
Wikstrom et al. {, 2020, 6311677} has a manuscript error in the regression coefficient for Q4 vs. Ql.
Reference _
Sp"eriod9 Confidence Study design ^^trix^ Sub-population E*^"re Comparison EE
Rating
Effect Estimate
-300 -200 -100 0 100 200 300 400 500
Early Chang et al. Cohort maternal term births median: 0.71 Regression
pregnancy (2022,9959688), serum ng/mL coefficient (per
Medium (25th-75th doubling in PFOA) ^
percentile:
0.45-1.07
ng/mL)
1
1
1
1
1
1
Regression
coefficient [for Q2
(0.45-0.71 ng/mL) 1?fi
vs. 01 (999 ng/L) 191
vs. Tertile 1 (<664
ng/L)
1
1
1
1 •
1
1
1
Gyllenhammar et Cohort and maternal - - Regression
al. (2018, cross-sectional serum coefficient per
4238300), unit-log increase in 177
Medium PFOA
1
1
1
• 1
1
1
l
-300 -200 -100 0 100 200 300 400 500
Figure 3-53. Overall Mean Birth Weight from Epidemiology Studies Following Exposure to
PFOA (Continued)
3-224
-------
APRIL 2024
Interactive figure and additional study details available on HAWC.
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 (out of 25 in total), 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 medium and high confidence 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, 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 five 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 among four studies in girls {Ashley-Martin, 2017, 3981371; Wang,
2016, 3858502; Wang, 2019, 5080598; Wikstrom, 2020, 6311677} including three high
confidence studies based on analyses of continuous PFOA measurements (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
3-225
-------
APRIL 2024
{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 changes in
standardized BWT 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}.
Eight 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-score measures
by sex.
Among the 14 studies examining continuous standardized BWT 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 (5/8), compared with 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, four out of six studies in boys {Chen, 2017, 3981292; Gross, 2020, 7014743; Wikstrom,
2020, 6311677; Xiao, 2019, 5918609} and 3 of 5 in girls {Gross, 2020, 7014743; Wikstrom,
2020, 6311677; Xiao, 2019, 5918609} 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 in both sexes (males P: -0.17; SE = 0.29; p-value = 0.57; females P: -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 standardized
BWT measures by sex in their analysis.
3.4.4.1.4.1.5 Standardized BWT summary
Eight out of 15 studies with standardized BWT scores in the overall population showed some
inverse associations with PFOA exposures. Seven of these 8 studies were either medium or high
confidence studies (of 17 in total), 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. Many
studies (5 of 8) showing inverse associations were based on later{Chen, 2017, 3981292; Gross,
2020, 7014743; Shoaff, 2018, 4619944; Wang, 2019, 5080598; Xiao, 2019, 5918609} versus
3-226
-------
APRIL 2024
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 4 out of 5 studies of male neonates
and 3 of 5 studies in females.
3.4.4.1.4.2 Small for Gestational Age/Low Birth Weight
Eleven informative and two iminformative 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}) (Figure 3-54). Five studies were rated
high confidence {Chu, 2020, 6315711; Lauritzen, 2017, 3981410; Manzano-Salgado, 2017,
4238465; Wang, 2016, 3858502; Wikstrom, 2020, 6311677}, three were rated 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 two were iminformative {Arbuckle, 2013, 2152344; Gundacker, 2021,
10176483}. Of the informative studies, four studies had good study sensitivity {Lauritzen, 2017,
3981410; Manzano-Salgado, 2017, 4238465; Meng, 2018, 4829851; Wang et al. 2016,
3858502}, four were considered adequate {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}.
3-227
-------
APRIL 2024
p,^v
Arbuckle etal., 2013, 2152344-
1
+
l_
+
I
+
—J
i
+
l_
+
+
~
Chang et al., 2022, 9959688 -
+
+
+
+
+* I
Chu etal., 2020, 6315711 -
+
++
+
+
++
Govarts et al., 2018, 4567442 -
+
+
++
B
+
-
J
Gundackeret al., 2021, 10176483-
+
+
-
-
+
-
n
Hjermitslev et al., 2020, 5880849 -
+
+
++
B
+
-
+
Lauritzen etal., 2017, 3981410-
++
++
E
+
++
+
++
++
Manzano-Salgado et al., 2017, 4238465 -
++
++
3*
+
++
+
++
++
Meng et al., 2018, 4829851 -
+
B
++
+
++
+
++
B
Souza et al., 2020, 6833697 -
+
+
+
-
-
+
-
J
Wang et al., 2016, 3858502 -
+
++
++
+
++
+
Wikstrom et al., 2020, 6311677-
++
++
++
+
++
+
+
Xu et al., 2019, 5381338-
-
++
++
+
+
+
-
J
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 3-54. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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 forLBW.
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 increased risk, while two studies were entirely null {Manzano-Salgado, 2017, 4238465;
Xu, 2019, 5381338} (Figure 3-55, Figure 3-56, Figure 3-57). Although they were not always
statistically significant, the relative risks reported in the five studies examining the overall
3-228
-------
APRIL 2024
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 nonmonotonic 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 quartile (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 twofold increased risk per each ln-unit increase risk (OR = 1.96; 95% CI: 1.18, 3.28)
and nonmonotonic 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. SGA findings from low confidence studies are not included in figures.
Reference
SPeriod9 Confidence ^M^trix^ Study Design Exposure Levels Sub-population Comparison EE
Rating
Effect Estimate
0 5 10 15 20 25 30 35 40
Early Manzano- plasma, Cohort Mean (SD): 2.35 ng/mL Boys OR (per doubling
pregnancy Salgadoetal. maternal (1.25 ng/mL) in maternal plasma
(2017, blood PFOA) 1 18
4238465)
High
"1
1
V
1
1
Girls OR (per doubling
in maternal plasma
PFOA) 0.72
1
1
«i
i
i
~ OR (per doubling
in maternal plasma
PFOA) 0.92
n
i
i
i
i
Later Lauritzen et maternal Cohort median=1.62 ng/mL Norway OR (per In unit
pregnancy al. (2017, serum (range: 0.31-7.97 ng/mL) increase in PFOA)
3981410) 0.66
High
~i
i
«r
i
i
median=2.33 ng/mL Sweden OR (per In unit
(range: 0.60-6.70 ng/mL) increase in PFOA)
5.25
~i
i
i
i •
i
i
Sweden; Boys OR (per In unit
increase in PFOA)
6.55
n
i
i
i •
i
i
Sweden; Girls OR (per In unit
increase in PFOA)
4.73
i
i
i
•
i
i
Norway: median=1.62 - OR (per In unit
ng/mL (range: 0.31-7.97 increase in PFOA)
ng/mL); Sweden: 1.21
median=2.33 ng/mL
(range: 0.60-6.70 ng/mL)
n
i
i
f-
i
i
Wang et al. maternal Cohort median=2.37 ng/mL Male OR (per 1 ln-unit
(2016, serum (25th-75th percentile: increase of PFOA)
3858502), 1.35-3.47 ng/mL) 0.63
High
~i
i
•i
i
i
median=2.34 ng/mL Female OR (per 1 ln-unit
(25th-75th percentile: increase of PFOA)
1.57-3.43 ng/mL) 1.48
i
i
i
i
0 5 10 15 20 25 30 35 40
Figure 3-55. Odds of Small for Gestational Age in Children from High Confidence
Epidemiology Studies Following Exposure to PFOA
3-229
-------
APRIL 2024
Interactive figure and additional study details available on HAWC.
Small-for-gestational-age defined as birthweight below the 10th percentile for the reference population.
Reference
SPeriod9 Confidence ^^trix^ study Design Exposure Levels Sub-population Comparison EE
Rating
Effect Estimate
0 5 10 15 20 25 30 35 40
Early Wikstrom et maternal Cohort Median-1.61 ng/mL Boys OR (per 1-ln ng/mL
pregnancy al. (2020, serum (25th-75th percentiles: change in PFOA)
6311677), 1.11-2.30 ng/mL) 116
High
1
fc-
1
1
OR (for Q2vsQ1)
0.67
1
1
#
1
1
OR (for 03 vs 01)
0.66
1
1
#
1
1
OR (for Q4vsQ1)
1.04
i
l
¦*-
1
1
Girls OR (per Hn ng/mL
change in PFOA) ^ ^
"I
l
1
I
OR (for Q2vsQ1)
1
l
I
l
OR (for 03 vs 01)
1.64
1
l
l
OR(for Q4 vs Q1)
2.33
1
I
l
I
- OR (per 1-tn ng/mL
change in PFOA) _
1.43
l
l
I
I
OR (for Q2 vs Q1)
0.77
"l
l
*
I
l
OR (for Q3 vs Q1)
0.96
"I
l
4-
1
I
OR (forQ4 vs Q1)
1.44
l
I
l
0 5 10 15 20 25 30 35 40
Figure 3-56. 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 HAWC.
Small-for-gestational-age defined as birthweight below the 10th percentile for the reference population.
3-230
-------
APRIL 2024
Sampling
Period
Early
pregnancy
Reference,
Confidence
Rating
Chang et al. maternal
(2022,
9959688),
Medium
Exposure
Matrix
serum
Exposure Levels Sub-population Comparison
Effect Estimate
median: 0.71 ng/mL
(25th-75th percentile:
0.45-1.07 ng/mL)
OR (per doubling
in PFOA)
OR [for Q2
(0.45-0.71 ng/mL)
vs. Q1 (
-------
APRIL 2024
Co... Reference, Measured c»..h„
p^P'"9 Confidence Effect/ MPtri n'"jnyn Sub-population Comparison EE
Penod Rating Endpoints Matnx Desifln
Effect Estimate
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Early Manzano-Salgado Low birth plasma, Cohort - OR (per doubling in
pregnancy etal. (2017, weight maternal maternal plasma PFOA) 0 9
4238465) High blood
• 1
l
Boys OR (per doubling in
maternal plasma PFOA) 112
l
1 •
l
Girls OR (per doubling in
maternal plasma PFOA) o.76
1
• 1
l
Low birth plasma, Cohort - OR (per doubling in
weight at maternal maternal plasma PFOA) o.85
term blood
1
• 1
l
Boys OR (per doubling in
maternal plasma PFOA) 1 67
l
1 •
l
Girls OR (per doubling in
maternal plasma PFOA) o.62
l
• 1
l
Hjermitslev et al. Low birth maternal Cohort - OR (per 11n-ng/mL
(2020,5880849), weight serum change in PFOA) o,44
Medium
l
• 1
l
Meng et al. (2018. Low birth maternal Cohort - OR (per doubling of
4829851), weight serum PFOA) 1
Medium
1
1
OR (for 02 vs. 01)
1.5
l
1 •
1
OR (for 03 vs. 01)
1.2
l
1 •
1
OR (for 04 vs. 01)
1.5
1
1 •
1
Later Chu et al. (2020, low birth maternal Cohort - OR (per 11n ng/mL
pregnancy 6315711), High weight serum increase in PFOA) 116
1
1 •
1
OR for 02 (> 0.96 to 1.54
ng/mL PFOA) vs. Q1 0 61
(<=0.96 ng/mL PFOA)
1
• 1
l
OR for 03 (>1.54 to 2.63
ng/mL PFOA) vs. Q1 0 27
(<=0.96 ng/mL PFOA)
1
• 1
l
OR for Q4 (> 2.63 ng/mL
PFOA) vs. Q1 (<=0.96 1
ng/mL PFOA)
1
1
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Figure 3-58. Odds of Low Birthweight in Children from Epidemiology Studies Following
Exposure to PFOA
Interactive figure and additional study details available on HAWC.
Low birthweight defined as birthweight <2,500 g.
Overall, eight of the 11 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 {Chang, 2022, 9959688; Manzano-Salgado,
2017, 4238465; Meng, 2018, 4829851; Wikstrom, 2020, 6311677} of the eight studies showing
increased odds of either SGA or LBW were based on early sampling biomarkers, suggesting the
results were not overly influenced by pregnancy hemodynamics. Collectively, the majority (8 of
11) of epidemiological studies were supportive of an increased risk of either SGA or LBW with
increasing PFOA exposures.
3.4.4.1.4.3 Birth Length
As shown in Figure 3-59 and Figure 3-60, 34 birth length studies were considered as part of the
study evaluation. Four studies were considered uninformative {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;
3-232
-------
APRIL 2024
Kobayashi, 2022, 10176408} were not further considered for multiple publications from the
same cohort studies. Among the 26 non-overlapping informative studies examined birth length in
relation to PFOA, including five studies with standardized birth length measures {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. 13 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.
3-233
-------
APRIL 2024
lSgN® ,«\e»s
,oe
Alkhalawi et al., 2016, 3859818-
+
+
+
+
Bach et al., 2016, 3981534-
+
+
+
+
++
+
++
Belletal., 2018, 5041287-
D
+
++
+
B
++
Bjerregaard-Olesen et al., 2019, 5083648 -
++
+
++
+
++
++
Buck Louis et al., 2018, 5016992 -
B
++
+
+
++
+
B
++
Callan et al., 2016, 3858524-
+
+
+
-
+
+
+
-
Cao et al., 2018, 5080197-
-
+
+
-
+
+
+
-
Chen et al., 2017, 3981292-
+
D
++
+
++
+
+
+
Chen et al., 2021, 7263985-
+
++
+
++
+
++
+
Espindola Santos etal., 2021, 8442216-
+
++
B
+
-
+
-
-
Gao et al., 2019, 5387135-
+
++
+
-
-
+
+
-
Gundacker et al., 2021, 10176483-
+
+
+
B
-
+
-
Ell
Gyllenhammar et al., 2018, 4238300 -
+
+
+
+
++
+
+
+
Hjermitslev et al., 2020, 5880849 -
+
+
+
+
+
-
+
+
Jin et al., 2020, 6316202-
-
+
+
B
-
+
++
-
Kashino et al., 2020, 6311632 -
+
++
+
+
+
+
+
+
Kishi et al., 2015, 2850268-
+
+
+
+
+
-
+
+
Legend
Q
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 3-59. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Length Effects
Interactive figure and additional study details available on HAWC.
3-234
-------
APRIL 2024
>$¦
,0®
Kobayashi etal., 2017, 3981430-
+
++
D
Kobayashi etal., 2022, 10176408-
+
++ ++
Lauritzen et al., 2017, 3981410-
++ ++
¦
Lee etal., 2013, 3859850
Luo et al., 2021, 9959610-
Manzano-Salgado et al., 2017, 4238465
Marks etal., 2019, 5081319-
Robledo et al., 2015, 2851197^
Shiet al., 2017, 3827535-
Shoaff etal., 2018, 4619944
Valvi et al., 2017, 3983872
Wang et al., 2016, 3858502
Wang et al., 2019, 5080598
Workman et al., 2019, 5387046 -
Wu etal., 2012, 2919186
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)
Figure 3-60. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Length Effects (Continued)
Interactive figure and additional study details available on IiAWC.
3-235
-------
APRIL 2024
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 (P per log2 increase in
PFOA: -0.14; 95% CI: -0.40, 0.13) in the overall population that appeared to be driven by male
neonates (P: -0.27; 95% CI: -0.65, 0.10). The low confidence Workman et al. {2019, 5387046}
study reported a nonsignificant deficit similar in magnitude (P: -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 (P: -0.32; 95% CI: -0.72, 0.07) compared with
tertile 1. In contrast, the low confidence study by Espindola-Santos et al. {, 2021, 8442216}
reported a larger birth length z-score (P per logio PFOA increase: 0.26; 95% CI: -0.21, 0.73).
Among the 21 studies examining mean birth length differences, eight different studies showed
inverse 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 (P: -0.39 cm; 95% CI: -1.80,
1.02) and 3 (P: -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 (P per ln-unit
PFOA increase: -0.27 cm; 95% CI: -0.61, 0.07), males (P: -0.21; 95% CI: -0.73, 0.32) and
females (P: -0.21; 95% CI: -0.74, 0.33). 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 (P: -0.49 cm; 95% CI: -0.99,
0.02), but detected the strongest association when restricted to the Swedish population (P:
-1.2 cm; 95% CI: -2.1, -0.3) and especially Swedish boys (P: -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 inverse 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 (P per loglO increase: -1.91 cm; 95% CI: -3.31,
-0.52). The low confidence study by Cao et al. {, 2018, 5080197} showed consistent results
across their overall population (P: -0.45 cm; 95% CI: -0.79, -0.10 per each ln-unit PFOA
increase), male (P: -0.36 cm; 95% CI: -0.80, 0.09), and female neonates (P: -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, 9 different studies out of 26 studies examining birth length reported deficits in relation
to PFOA exposures, including 6 medium or high confidence studies. There was no apparent
relationship between studies showing inverse associations and study confidence ratings.
However, seven of these studies sampled PFOA biomarkers later in pregnancy {Cao, 2018,
3-236
-------
APRIL 2024
5080197; Lauritzen, 2017, 3981410; Shoaff, 2018, 4619944; Wang, 2016, 3858502; Workman,
2019, 5387046; Wu, 2012, 2919186; Xiao, 2019, 5918609} and may be more prone to potential
bias from 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 (P per loglO PFOA increase = -1.91 cm; 95% CI: -3.21, -0.52) was
reported in an earlier study {Wu, 2012, 2919186} that reported the largest exposure range. There
was a preponderance of inverse associations among females (6 of 12 studies) compared with
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 3-61, 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 seven studies 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 12 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}.
3-237
-------
APRIL 2024
Bell et al.
2018,5041287
2019,5083648
2018, 5016992
2016, 3858524-
2018, 5080197-
2021,7263985 -
2021, 8442216-
Bjerregaard-Olesen et al.
Buck Louis et al.
Callan et al.
Cao et al.
Chen et al.
Espindola Santos et al.
Gundackeretal., 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 et al., 2017, 3983872-
Wang et al.
Wang et al.
Workman et al.
Xiao et al.
Xu et al.
2016, 3858502
2019, 5080598-
2019, 5387046-
2020, 5918609-
2019, 5381338
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
Figure 3-61. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Birth Head Circumference Effects
Interactive figure and additional study details available on IiAWC.
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
3-238
-------
APRIL 2024
the Xiao et al. {, 2019, 5918609} z-score data, 13 of these 21 studies reported sex-specific 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, 10 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 (P per ln-unit PFOA increase: -0.4 cm; 95% CI: -1.0,
0.1); this was largely due to an association seen in male neonates (P: -0.6 cm; 95% CI: -1.3,
0.1). The high confidence study by Buck Louis et al. {, 2018, 5016992}, reported nonsignificant
head circumference differences (P: -0.14 cm; 95% CI: -0.29, 0.02) among Black neonates but
no main effect association in the overall population. Six out of 17 studies based on the overall
population reported some inverse 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 (P: -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 (P: -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 (P per loglO PFOA increase: 0.62; 95% CI: -0.06,
1.29). The medium confidence study by Gyllenhammar et al. {, 2018, 4238300} was null based
on their standardized head circumference measure.
Among the 14 studies that examined mean head circumference at birth in the overall population,
four of them reported inverse associations. Nine studies were largely null, and one study showed
larger mean head circumference in the overall population with increasing PFOA exposures. Of
the 11 different studies examining sex-specific results associations were observed 5 of 10 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 11 studies in male
neonates. The medium confidence study by Wang et al. {, 2019, 5080598} reported an
association in the overall population (P: -0.37 cm; 95% CI: -0.70, -0.40) with larger deficits
noted in female (P: -0.57 cm; 95% CI: -1.07, -0.08) than in male neonates (P: -0.35 cm; 95%
CI: -0.79, -0.10). The medium confidence study by Hjermitslev et al. {, 2019, 5880849} showed
a significant reduction in head circumference for the term births in the overall population (P per
ng/mL PFOA increase: -0.30 cm; 95% CI: -0.56, -0.04) which seemed to be driven by results in
females (P: -0.25 cm; 95% CI: -0.65, 0.14). The high confidence study by Manzano-Salgado et
al. {, 2017, 4238465} reported a nonsignificant decrease only in quartile 4 (P: -0.16 cm; 95%
CI: -0.38, 0.06) compared with quartile 1 from the overall population and a deficit among male
neonates only (P per log2 PFOA increase: -0.13 cm; 95% CI: -0.27, 0.0). 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 (P per 1 SD PFOA increase: -0.18 cm; 95%
CI: -0.59, 0.23). In their low confidence study, Cao et al. {, 2018, 5080197} reported an overall
null association, while divergent and large changes were seen for male (P per ln-unit PFOA
increase: 0.72 cm; 95% CI: -0.51, 1.94) and female neonates (P: -1.46 cm; 95% CI: -2.96,
3-239
-------
APRIL 2024
0.05). The low confidence study by Callan et al. {, 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 mean differences and standardized
measures of head circumference, 10 different studies reported some evidence of inverse
associations in the overall population or across sexes or race. This included 4 of 15 studies in the
overall population and 5 of 12 sex-specific studies in either or both sexes. No definitive patterns
across sex were observed 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 were seen based on 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 {Bjerregaard-Olesen, 2019, 5083648; Buck Louis,
2018, 5016992; Hjermitslev, 2020, 5880849; Manzano-Salgado, 2017, 4238465; Robledo, 2015,
2851197}. This suggests that pregnancy hemodynamics is not fully explaining the inverse
asociations detected here.
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 inverse
associations. For example, 10 (out of 21) different epidemiological studies of PFOA examining
head circumference reported some evidence of inverse associations in either the overall
population or across the sexes, which included 8 of 15 medium or high confidence studies. Nine
different studies out of 26 studies reported some birth length deficits in relation to PFOA
exposures with limited evidence of exposure-response relationships. This included 6 of 17
medium or high confidence studies of birth length. 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 statistical
power to detect sex-specific differences that vary considerably in magnitude. There was minimal
evidence of exposure-response relationships reported among 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 SGA 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
3-240
-------
APRIL 2024
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. Interestingly, 11 of the 12
largest mean BWT deficits (-48 grams or larger per unit change) in the overall population were
detected among studies with later biomarker sampling. However, five {Chang, 2022, 9959688;
Hjermitslev, 2020, 5880849; Meng, 2018, 4829851; Sagiv, 2018, 4238410; Wikstrom, 2020,
6311677} of nine medium and high confidence studies still reported some evidence of reductions
in mean BWT based on early pregnancy biomarker samples.
The current database (since the 2016 PFOA HESD) is fairly strong 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. 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 al. {, 2018, 5079861} have
shown that some measures like mean BWT may be prone to bias from pregnancy hemodynamics
especially in studies with later biomarker sampling. 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 Jensen et al. {, 2020, 6833719} depicted in Figure 3-62),
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
3-241
-------
APRIL 2024
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
de Cock 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 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 ln-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,
nonsignificant increases small in magnitude for boys (0.11; 95% CI: -0.04, 0.27) and decreases
in greater height per each ln-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 ln-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
relationship was detected when examined across PFOA categories with the highest exposure
associated with smaller statistically significant height increases at age 2 compared with 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 IQR PFOA 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 nonsignificant 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
3-242
-------
APRIL 2024
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
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 with 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 inverse associations with PFOA. Study quality ratings, including study sensitivity
and overall confidence, did not appear to be explanatory factors for heterogeneous results across
studies.
3-243
-------
APRIL 2024
¦ ' I 1 ¦ '
ffs':S."
Cao et al., 2018, 5080197-
l_i
—I—
+
++
I
i
+
•
+
i
+
Chen et al., 2017, 3981292-
+
+
++
+
++
+
+
+
Gross et al., 2020, 7014743 -
+
+
+
+
-
+
-
-
Gyllenhammar et al., 2018, 4238300 -
+
+
++
+
++
+
+
+
Jensen et al., 2020, 6833719 -
+
+
+
+
+
+
+
++
Lee et al., 2018, 4238394-
D
++
a
+
++
B
Manzano-Salgado et al., 2017, 4238509 -
I
++
++
B
++
+
++
B
Shoaff et al., 2018, 4619944-
++
++
++
++
++
+
++
++
Starling et al., 2019, 5412449-
++
++
++
++
+
a
++
Tanner et al., 2020, 6322293 -
++
D
++
B
++
+
++
++
Wang etal., 2016, 3858502-
B
++
++
++
+
++
++
Yeung etal., 2019, 5080619-
++
D
++
++
+
B
++
de Cock etal., 2014, 2713590-
D
B
++
+
-
j
Legend
H
Good (metric) or High confidence (overall)
+
Adequate (metric) or Medium confidence (overall)
-
Deficient (metric) or Low confidence (overall)
D
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-62. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Postnatal Growth
Interactive figure and additional study details available on HAWC.
3.4.4.1.5.1 Adiposity/BM!
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.
3-244
-------
APRIL 2024
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 with 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-based z-scores 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
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
3-245
-------
APRIL 2024
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 or females, but did detect increased infant weight measures among males. Three of
the seven studies reporting quantitative results showed some evidence of inverse 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 2 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 iminformative 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) (Figure 3-63). 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.
3-246
-------
APRIL 2024
>c®
Bach etal.,2016, 3981534-
Bell et al., 2018, 5041287 -J
Buck Louis et al., 2018, 5016992 -
Chu et al., 2020, 6315711 -
Eick et al., 2020, 7102797
Gao et al., 2019, 5387135
Gundacker et al., 2021, 10176483
Gyllenhammar et al., 2018, 4238300
Hjermitslev et al., 2020, 5880849
Huo et al., 2020, 6835452
Lauritzen et al., 2017, 3981410
Lee et al., 2013, 3859850
Lind et al., 2017, 3858512 -|
Manzano-Salgado et al., 2017, 4238465-j
Meng etal.,2018, 4829851 ^
Sagiv etal.,2018, 4238410 -J
Workman et al., 2019, 5387046-
Wu et al., 2012, 2919186-
Xu et al., 2019, 5381338-
Yang et al., 2022, 10176806-
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 3-63. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Gestational Age
Interactive figure and additional study details available on HAWC.
3-247
-------
APRIL 2024
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,
3858512} examined only sex-specific data and reported larger deficits in female (-0.21 cm; 95%
CI: -0.61, 0.19 per each ln-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 inverse 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 3-64, 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 11 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
3-248
-------
APRIL 2024
{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
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/mL {Huo, 2020, 6835452}.
3-249
-------
APRIL 2024
lSe<^
Bach et al., 2016, 3981534-
I
+
—i—
+
++
i
+
++
—i—
+
i
+
++
Chu et al., 2020, 6315711 -
B
++
D
+
++
+
+
++
Eick et al., 2020, 7102797-
++
++
I
+
++
+
+
++
Gardener et al., 2021, 7021199 -
++
++
++
+
++
+
+
++
Hjermitslev et al., 2020, 5880849 -
+
+
+
+
+
-
+
*
Huo et al., 2020, 6835452-
++
++
++
+
++
+
++
++
Liu et al., 2020, 6833609-
++
D
B
-
++
+
-
*
Manzano-Salgado et al., 2017, 4238465 -
++
++
++
+
++
+
++
++
Meng et al., 2018, 4829851 -
+
+
++
+
++
+
++
D
Sagiv et al., 2018, 4238410-
++
+
+
+
++
+
++
++
Yang et al., 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 3-64. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure and Preterm Birth Effects
Interactive figure and additional study details available on HAWC.
Six of the 11 studies reported an increased risk of PTE with elevated exposure to PFOA. 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 nonmonotonic ORs for PTB in the upper three PFOA quartiles
(OR range: 1.7-3.2), but little evidence was observed per each doubling of PFOA exposures
(OR = 1.1; 95% CI: 0.8, 1.5). Although they were not statistically significant, the high
3-250
-------
APRIL 2024
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
with 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} reported increased
nonsignificant 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.
Overall, 6 of the 11 studies showed increased risk of PTB with PFOA exposures with limited
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
five studies with later PFOA biomarker sampling showed some increased odds of preterm birth
compared with two of six studies with earlier sampling.
3.4.4.1.6.3 Gestational Duration Summary
Overall, there was mixed evidence of exposure to PFOA and both inverse associations with
gestational age and increased risk of 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 (two high, two medium and one 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 3-65).
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.
3-251
-------
APRIL 2024
{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) =0.93; 95%
CI: 0.75, 1.16 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.
<\ rtv©^
Buck Louis et al., 2016, 3858527 -
+
+
+
+
+
+
+
+
Jensen et al., 2015, 2850253 -
+
++
+
+
+
+
+
-
Liew et al., 2020, 6387285-
B
B
++
+
++
+
+
+
Wang et al., 2021, 10176703-
++
++
++
+
++
+
++
++
Wikstrom et al., 2021, 7413606 -
++
++
++
+
++
+
++
++
Figure 3-65. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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 3-66. This included a medium
confidence study by Vesterholm Jensen et al. {, 2014, 2850926} that reported no increased risk
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
3-252
-------
APRIL 2024
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 nonspecific 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 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.
Cao etal.,2018, 5080197
Nolan etal., 2010, 1290813
Ou etal., 2021, 7493134
Vesterholm Jensen et al., 2014, 2850926
Legend
p
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 3-66. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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 6 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
3-253
-------
APRIL 2024
PFOA HESD that investigated the association between PFOA and developmental effects in
animal models. Study quality evaluations for these 19 studies are shown in Figure 3-67.
Abbott etal., 2007, 1335452
Blake et al., 2020, 6305864 -
Butenhoff et al., 2004, 1291063-
Chen etal., 2017, 3981369-
Cope et al., 2021, 10176465-
Hu etal., 2010, 1332421
Hu etal., 2012, 1937235-
Jiang et al., 2020, 6320192 -
Lau etal., 2006, 1276159
Li etal., 2018, 5084746-
Li etal., 2019, 5387402-
Macon et al., 2011, 1276151 -
NTP, 2020, 7330145
Salimi etal., 2019, 5381528
Song etal., 2018, 5079725
Tucker et al., 2015, 2851046 -
Wolf etal., 2007, 1332672
Zhang et al., 2021, 10176453
van Esterik et al., 2015, 2850288 ~KIS 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)
NRl Not reported
* Multiple judgments exist
Figure 3-67. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure 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
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
3-254
-------
APRIL 2024
relate to other health outcomes may be discussed in the corresponding health outcome section
(e.g., neurodevelopmental effects are discussed in the Appendix {U.S. EPA, 2024, 11414343}).
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 to GD 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 3-68).
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).
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
3-255
-------
APRIL 2024
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 3-14. Dose
groups for this study are referred to as "[perinatal exposure level]/[postweaning exposure level]"
(e.g., 300/100).
Table 3-14. 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 3-68). 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}.
3-256
-------
APRIL 2024
PFOA Developmental Effects - Maternal Body Weight
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)
LauetaL, 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)
) No significant change A Significant
Y Significant
V V V
A
Concentration (mg/kg/day)
Figure 3-68. 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 3-69).
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
3-257
-------
APRIL 2024
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)
7
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 3-69. 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
3-258
-------
APRIL 2024
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 (Figure 3-70).
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 with
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 multigenerational 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
3-259
-------
APRIL 2024
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 with 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 to GD 17.5 did not significantly differ
from control groups {Cope, 2021, 10176465} (Figure 3-70).
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 with 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 3-70).
3-260
-------
APRIL 2024
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,o-i/.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
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
7
HI II ffni 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 (;f~, 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, Cri: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)JGS 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 multigeneration 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.
3-261
-------
APRIL 2024
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
3-262
-------
APRIL 2024
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 with
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 with 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 "net" body weights (i.e., absolute body weight minus absolute liver
weight) at PND 21 and PND 35 but not at PND 56 {Tucker, 2015, 2851046}; the absolute body
weights of female offspring were not altered due to gestational PFOA treatment. Macon et al. {,
2011, 1276151} found no effects on offspring body weights following exposure of pregnant CD-
1 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 3-71).
3-263
-------
APRIL 2024
PFOA Developmental Effects - Offspring Body Weight
Endpoint
Study Name
Study Design
Observation Time
Animal Description
0 No significant change^ Significant increase V Significant decrease
Body Weight Change
Wolf el al., 2007; 1332672
developmental (GD1-17)
PND1-22
F1 Mouse, CD-1 N=11-14)
Copeetal., 2021. 10176465
developmental (GD1.5-17.5)
PND22-PNW18
F1 Mouse. CD-1 N=B)
F1 Mouse. CD-1 U. N=9)
PND22-128
F1 Mouse, CD-1 ( ;')
¦ ~
F1 Mouse. CD-1 ('')
Fetal Body Weight
Blake et al.. 2020. 6305864
developmental (GD1.5-11.5)
GD11.5
F1 Mouse. CD-1 N=62)
developmental (GD1.5-17.5)
GD17.5
F1 Mouse. CD-1 N=62)
V
Lau etal.. 2006. 1276159
developmental (GD1-17)
GD18
F1 Mouse. CD-1 N=5-42)
¦ . . 1 ~
Lietal., 2018, 5084746
developmental (GD1-17)
GD18
F1 Mouse. Kunming ( N=10)
¦ y y y y
Litter Weight
Cope et al., 2021, 10176465
developmental {GD1.5-17.5)
PND0.5
Fl Mouse, CD-1 (;*\ N=9)
PND5.5
F1 Mouse. CD-1 N=9)
¦ ~
Huetal.. 2010,1332421
developmental (GD6-17)
PND2
F1 Mouse, C57BL/6n ( ;"N=10)
¥-¥
PND7
F1 Mouso, C57BL/'6n (J -', N=10)
—¥
PND14
F1 Mouse, C57BL,'6n ( N=10)
—w
Pup Body Weight
Copeetal., 2021, 10176465
developmental (GD1.5-17.5)
PND0.5
PO Mouse, CD-1 (y, N=9)
• ~
Abbott et al.. 2007,1335452
developmental (GD1-17)
PND1
F1 Mouse, 129S1.'SvlmJ (5, N=0-18)
F1 Mouse. 129S1.'SvlmJ N=0-17)
F1 Mouse, 129S4/SvJae PPARa null {", N=0-18)
F1 Mouse. 12954/SvJae PPARa null (. \ N=0-17)
Macon etal., 2011, 1276151
developmental (GD10-17)
PND1
F1 Mouse, CD-1 ('. N=7-9)
PND21
F1 Mouso. CD-1 N=7-11)
* ~
developmental {GD1-17)
PND7
F1 Mouse, CD-1 N=3-6)
F1 Mouse. CD-1 ( ', N=4-5)
PND21
F1 Mouse, CD-1 N=4)
F1 Mouse. CD-1 (J, N=3-6)
Songet al.,2018. 5079725
developmental {GD1-17)
PND21
F1 Mouse, Kunming ( J, N=9-10)
Tucker et al., 2015,2851046
developmental (GD1-17)
PND21
F1 Mouse, C57BI/6 (i. N=2-6)
F1 Mouse. CD-1 ("\ N=20-22)
»
Pup Weight Rotative to Litter
Cope et al., 2021, 10176465
dovolopmontal (GD1.5-17.5)
PND 0.5
F1 Mouse, CD-1 (J" , N=9)
« 4
PND5.5
F1 Mouse, CD-1 N=9)
• ~
Hu etal., 2012, 1937235
developmental (14d mating-PND2l)
PND1
F1 Mouse. C57BI6'N ( -.N=11-14)
PND21
F1 Mouse, C578I6/N ( .; i', N=7-11)
. v
Lau etal., 2006,1276159
developmental (GD1-18)
PND22
F1 Mouse. CD-1 (-"1. N=7-23)
. WW W V
Wolf et al., 2007, 1332672
developmental (GD1-17)
PND1
F1 Mouse, CD-1 N=11-14)
* ~
PO Mouse. CD-1 (v. N=25-39)
¦ w
F1 Mouse. CD-1 (N=11-14)
PND22
F1 Mouse, CD-1 (''. N=11-14)
developmental (GD7-17)
PND22
F1 Mouse. CD-1 N=7-13)
V
developmental (GD1-PND22)
PND1
F1 Mouse, CD-1 N=12-14)
• V
F1 Mouse, CO-1 (O, N=12-14)
w
PND22
F1 Mouse. CD-1 ( -'. N=12-14)
w
F1 Mouse. CD-1 (?, N=12-14)
w
developmental (PND1-22)
PND1
F1 Mouse. CD-1 N=11-14)
F1 Mouse. CD-1 (•', N=11-14)
PND22
F1 Mouse, CD-1 N=11-14)
F1 Mouse. CD-1 (2. N=11-14)
Butenhoff et al., 2004, 1291063
reproductive (84d)
PND1
F1 Rat, Crl:CD(SD)IGS BR (; '. N=27-29)
V ¥
PND5
F1 Rat, Crt:CD(SD)lGS BR -, N=27-29)
. , , S7
PND8
F1 Rat, Crl:CD(SD)IGS BR (" Li. N=27-29)
PND15
F1 Rat, Crl:CD(SD)IGS BR (; ~, N=27-29)
—.—.—V
PND22
F1 Rat, Crl:CD(SD)IGS BR N=27-29)
reproductive (GD1-PND106)
PND1
F2 Rat, Crl:CD(SD)IGS BR (- N=29-30)
PND22
F2 Rat, Crl:CD(SD)IGS BR ( / " , N=29-30)
NTP, 2020, 7330145
chronic (GD6-PND21)
PND1
F1 Rat, Sprague-Dawley (
-------
APRIL 2024
3.4.4.2.5 Skeletal and Visceral Alterations
Following exposure of pregnant CD-I mice to 1, 3, 5, 10, 20, or 40 mg/kg/day PFOA during
gestation, Lau et al. {, 2006, 1276159} reported decreases in ossification of the forelimb
proximal 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 mice. Lau et al. {,
2006, 1276159} and Wolf et al. {, 2007, 1332672} reported delayed eye opening in CD-I 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 with controls at 0.6 and
1 mg/kg/day (significant at 1 mg/kg/day; however, extensive litter loss was observed 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}.
3-265
-------
APRIL 2024
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
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 with 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 with 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).
3-266
-------
APRIL 2024
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,
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 was also 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 toward 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).
3-267
-------
APRIL 2024
3.4.4.3 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 PFOA HESD {U.S. EPA, 2016,
3603279}. There are 19 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 developmental effects. A summary of these studies by mechanistic data
category (see Appendix A, {U.S. EPA, 2024, 11414343}) and source is shown in Figure 3-72.
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
8
Cell Signaling Or Signal Transduction
2
1
0
3
Fatty Acid 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
Xen obi otic 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 3-72. Summary of Mechanistic Studies of PFOA and Developmental Effects
Interactive figure and additional study details available on HAWC.
Mechanistic data available from in vitro, in vivo, and epidemiological studies were evaluated to
inform the mode of action of developmental effects of PFOA. The mechanistic data are
organized by the following outcomes: 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 to GD 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
3-268
-------
APRIL 2024
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 to GD 11.5 or GD 1.5 to 17.5 via oral gavage
{Blake, 2020, 6305864}. However, 5 and 10 mg/kg exposure via oral gavage from GD 1 GD 17
decreased survival rate in 5-day old pups, possibly due to hepatotoxicity; the authors observed
significantly 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 (tgfblaj 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 (ytgl). Vasa is expressed in the germline and is
3-269
-------
APRIL 2024
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
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). 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
Fetal 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 to 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 with 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 [xM PFOA, including interocular distance, total
body length, and yolk sac area. The size of all three parameters was reduced in groups exposed
3-270
-------
APRIL 2024
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), tfc3ci
(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}.
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-P, PDGF, PPAR, NFKB, MAPK, Endothelin 1, TNRF2, tight junctions, cytokines
including IFNY and IFNa, and possible FOS signaling {Midic, 2018, 4241048}. A result from
the Kingsley et al {, 2017, 3981315} study in human cord blood mentioned above was
methylation changes to the RASAS gene associated with exposure to PFOA (high exposure
group, which could result in impaired cell growth and differentiation, contributing to reduced
fetal growth and birth weight.
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
3-3,000 (J,g/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 Acotl and A cox I (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}.
3-271
-------
APRIL 2024
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
mediated through the PPARa pathway. Salimi et al. {, 2019, 5381528} exposed pregnant mice to
PFOA from GD 5 to 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
versus 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 |ig/mL,
3-272
-------
APRIL 2024
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 to 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.
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} (see Appendix A, {U.S. EPA, 2024, 11414343}).
3-273
-------
APRIL 2024
Overall, there was mixed evidence of inverse 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 inverse
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
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., as many of studies did not
show inverse associations for gestational age measures or for an increased risk of preterm birth.
However, as noted previously, considerable uncertainty remains as to what degree the evidence
may be impacted by pregnancy hemodynamics factors related to sample timing may result in
either confounding or reverse causality and explain some of the observed birth weight deficits
{Steenland, 2018, 5079861}. Additional uncertainty exists due to the potential for confounding
by other PFAS, and considerations for potential confounding by co-occurring PFAS are
described in Section 5.1. 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 with 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.
3-274
-------
APRIL 2024
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
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 3-15). 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.
3-275
-------
APRIL 2024
Table 3-15. Evidence Profile Table for PFOA Exposure and Developmental 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 3.4.4.1)
Fetal growth
restriction
26 High confidence
studies
25 Medium
confidence studies
13 Low confidence
studies
3 Mixed confidence
studies
Some deficits in
mean birth weight
were observed in
most studies (30/42)
in the overall
population. The
majority of studies
on changes in
standardized birth
weight measures
reported inverse
associations (10/18),
with most (7/10) of
these being high and
medium confidence.
Similarly, most
studies (12/17)
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 decreased birth
length and head
circumference in the
overall population or
across sexes.
• High and medium confidence
studies
• Consistent direction of effects
for most outcomes
• Coherence of findings across
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
0©O
Moderate
Epidemiological
evidence for
developmental
effects is based on
consistent adverse
effects for FGR and
post-natal growth.
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
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
support the human relevance of the
animal findings.
3-276
-------
APRIL 2024
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
Gestational
duration
13 High confidence
studies
13 Medium
confidence studies
7 Low confidence
studies
In medium and high
confidence studies,
inverse effects were
observed on
gestational age
(10/20). An
increased risk of
preterm birth was
also observed
• High and medium confidence
studies
• Potential bias due s'laPc l'1c
to hemodynamic exposure-response
difference noted
in studies using
samples from
later pregnancy
relationship, and the
potential impact of
hemodynamics in
later pregnancy due
to use of
biomonitoring
samples from the
second and third
trimester or post-
partum^
in medium
and high confidence
studies (9/18).
Fetal Loss
2 High confidence
studies
6 Medium
confidence studies
1 Low confidence
study
A significantly
increased risk of
fetal loss was
reported in one high
(1/2) and one
medium (1/6)
confidence study.
The response in the
high confidence
study was monotonic
across exposure
quartiles. Other
medium confidence
studies (5/6) reported
mixed results,
differing by the
exposure
comparison. One
study reported a
• High and medium confidence
studies
• Good or adequate sensitivity
• Consistent magnitude of
effect
• Exposure-response
relationship
• No factors noted
3-277
-------
APRIL 2024
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
decreased risk of
fetal loss, but the
study was considered
low confidence.
Post-natal growth
6 High confidence
studies
7 Medium
confidence studies
3 Low confidence
studies
Five medium and
high confidence
studies (5/11)
reported inverse
associations with
infant weight and
two studies (2/11)
reported positive
associations, while
the remaining studies
were mixed by sex or
timepoint. Similarly,
inverse associations
with BMI were
observed in five
medium and high
confidence studies
(5/8),
• High and medium confidence
studies
• Good or adequate sensitivity
for most studies
• Inconsistent
timing of follow-
up evaluation
and increased risk of
rapid growth rate
was observed in only
one study (1/5). Two
medium and high
confidence studies
(2/8) observed
increased infant
length or height and
one study reported
an inverse
association, while
3-278
-------
APRIL 2024
Evidence Stream Summary and Interpretation
Studies and Summary and Key Factors that Increase Factors that Evidence Stream Evidence Integration Summary
Interpretation Findings Certainty Decrease Judgment Judgment
Certainty
other studies were
null or mixed by sex.
Birth Defects
Two low confidence . n0 factors noted
• Low confidence
4 Medium
studies and two
studies
confidence studies
medium confidence
2 Low confidence
studies reported
• Limited number
studies
mixed results for
of studies
total or combined
examining
birth defects. No
individual
association with
defects
cryptorchidism was
reported in one
study; one study
reported decreased
odds of septal
defects, conotruncal
defects, and total
congenital heart
defects.
Evidence from In Vivo Animal Toxicological Studies (Section 3.4.4.2)
Maternal body
weight
2 High confidence
studies
6 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
• High and medium confidence
studies
• Inconsistent
direction of
effects
©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
3-279
-------
APRIL 2024
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
observing an
increase (2/5).
Offspring body
weight
2 High confidence
studies
10 Medium
confidence studies
Many rodent studies
observed changes in
fetal or pup body
weight following
PFOA exposure
(9/12). Most of these
show a decrease in
offspring weight
(8/9). One study
observed an increase
in offspring body
weight, but only in
male mice. Three
mouse studies
showed no change in
offspring body
weight (3/12).
• High and medium confidence
studies
• Consistent direction of
effects
• No factors noted
Offspring
mortality
2 High confidence
studies
7 Medium
confidence studies
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
• High and medium confidence
studies
• Consistent direction of
effects
• No factors noted
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
maternal effects in
dams. Notably, mice
appear to be more
sensitive to
developmental
toxicity as a result of
"gestational exposure
compared with 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.
3-280
-------
APRIL 2024
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
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).
Placenta effects
2 Medium
confidence studies
Two mouse studies
noted a decrease in
relative placenta
weight following
gestational PFOA
exposure. In these
studies, lesions on
the placenta and
• Medium confidence studies
• Limited number
of studies
examining
outcomes
3-281
-------
APRIL 2024
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
other
histopathologic^
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).
Offspring liver
weight
3 Medium
confidence studies
Increases in
offspring relative
liver weight were
noted in three mouse
studies following
gestational PFOA
exposure (3/3).
• Medium confidence studies
• Limited number
of studies
examining
outcomes
Developmental
timing
2 Medium
confidence studies
Delayed eye opening . Medium confidence studies
(2/2) and delayed
body hair
development (1/1)
were observed in
both sexes of mice.
• Limited number
of studies
examining
outcomes
Structural
abnormalities
1 Medium
confidence study
One mouse study
found structural
abnormalities
(e.g., reduced
skeletal ossification)
after developmental
exposure to PFOA.
> Medium confidence study
• Limited number
of studies
examining
outcomes
3-282
-------
APRIL 2024
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
Mammary gland
development
2 Medium
confidence studies
Two mouse studies
(2/2) 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 study
• Limited number
of studies
examining
outcomes
Lactation index
2 High confidence
studies
Of the two rat studies . High confidence studies
that evaluated
lactation index, one
noted a decrease
following PFOA
(1/2).
• Limited number
of studies
examining
outcomes
Mechanistic Evidence and Supplemental Information (Section 3.4.4.3)
Summary of Key Findings, Interpretation, and Limitations
Evidence Stream
Judgment
Key findings and interpretation:
• Decreased survival in mice offspring exposed to PFOA in utero related to PPARa-related
hepatotoxicity.
• 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,
3-283
-------
APRIL 2024
Studies and
Interpretation
Evidence Stream Summary and Interpretation
Summary and Key
Findings
Factors that Increase
Certainty
Factors that
Decrease
Certainty
Evidence Stream
Judgment
Evidence Integration Summary
Judgment
although the
relationship has not
been directly
measured.
Notes: DNA = deoxyribonucleic acid; FGR = fetal growth restriction; mRNA = messenger ribonucleic acid; PPARa = peroxisome proliferator-activated receptor alpha;
SGA = small-for-gestational-age.
3-284
-------
APRIL 2024
3.4.5 Evidence Synthesis and Integration for Other Noncancer
Health Outcomes
Consistent with the SAB's recommendation {U.S. EPA, 2022, 10476098}, EPA concluded that
the noncancer 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 from the
preliminary scoping considered in the Proposed Approaches to the Derivation of a Draft
Maximum Contaminant Level Goal for Perflaorooctanoic 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 {U.S. EPA, 2021, 10428559}. Based on this
analysis, these outcomes were not prioritized for the subsequent literature search update efforts;
the evidence synthesis and integration for these outcomes are presented in Appendix C {U.S.
EPA, 2024, 11414343}. In addition, Section 5.5 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 28 (29 publications16) epidemiological and 5 animal toxicological studies that
investigated the association between PFOA and cancer. Of the epidemiological studies, 12 were
classified as medium confidence, 12 as low confidence, 2 were considered aninformative, and 2
were mixed confidence (1 medium/low and 1 low fiminformative confidence) (Section 3.5.1). Of
the animal toxicological studies, 2 were classified as high confidence, 1 as medium confidence,
and 2 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 10 epidemiological studies (11 publications17) from the 2016 PFOAHESD {U.S. EPA,
2016, 3603279} that investigated the association between PFOA and cancer effects. Study
quality evaluations for these 10 studies are shown in Figure 3-73.
The 2016 PFOA HESD {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 two occupational cohorts (Figure 3-73). 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
10 Ghisari, 2014,2920449 analyzes interactions between gene polymorphisms and PFOA exposure on breast cancer risk in the
same population analyzed in Bonefeld-Jorgensen, 2011, 2150988.
17 Ghisari, 2014, 2920449 analyzes interactions between gene polymorphisms and PFOA exposure on breast cancer risk in the
same population analyzed in Bonefeld-Jorgensen, 2011, 2150988.
3-285
-------
APRIL 2024
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 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, 16 incident kidney cancer cases, and
five incidence testicular cancer cases in Raleigh et al. {, 2014, 2850270}; 12 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 HESD 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-J0rgensen, 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.
3-286
-------
APRIL 2024
Barry et al., 2013, 2850946 -
—I—
+
1—
+
1—
+
1
+
++
1—|
+
i
+
+
Bonefeld-Jorgensen etal., 2011, 2150988-
+
+
+
+
+
+
+
+
Bonefeld-Jorgensen et al., 2014, 2851186 -
+
+
-
+
-
-
+
-
Eriksen et al., 2009, 2919344 -
++
+
+
+
+
+
+
+
Ghisari et al., 2014, 2920449 -
+
+
+
+
+
+
+
+
Hardell et al., 2014, 2968084 -
++
¦
++
-
+
+
+
-
Innes et al., 2014, 2850898 -
+
-
+
+
+
+
+
-
Raleigh et al., 2014, 2850270 -
+
-
+
-
+
+
-
-
Steenland and Woskie, 2012, 2919168-
+*
+
+
-
+
+
+
+*
Steenland et al., 2015, 2851015 -
-
+
-
+
++
+
+
-
Vieira et al., 2013, 2919154-
+
+
+
+
++
+
+
+
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 3-73. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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 PFOA HESD {U.S. EPA, 2016, 3603279}, 18 epidemiological
studies have been published that investigated the association between PFOA and cancer (see
Appendix, {U.S. EPA, 2024, 11414343}). One of the publications {Girardi, 2019, 6315730} was
an occupational study 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}, six case-control studies {Wielsoe, 2017, 3858479; Tsai, 2020, 6833693; Lin,
2020, 6835434; Itoh, 2021, 9959632; Liu, 2021, 10176563; Cao, 2022, 10412870}, six nested
case-control studies {Mancini, 2020, 5381529; Ghisari, 2017, 3860243; Shearer, 2021, 7161466;
Hurley, 2018, 5080646; Cohn, 2020, 5412451; Goodrich, 2022, 10369722}, and three cross-
3-287
-------
APRIL 2024
sectional studies {Christensen, 2016, 3858533; Ducatman, 2015, 3859843; Omoike, 2021,
7021502}. The studies were conducted in different study populations including populations from
China {Lin, 2020, 6835434; Liu, 2021, 10176563; Cao, 2022, 10412870}, Denmark {Ghisari,
2017, 3860243}, France {Mancini, 2020, 5381529}, Greenland {Wielsoe, 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; Goodrich, 2022, 10369722}. 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; Wielsoe, 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; Cao, 2022,
10412870; Goodrich, 2022, 10369722}, 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 18 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 18 studies are
shown in Figure 3-74.
Of the 18 studies identified since the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}, eight were
considered medium confidence, and eight were low confidence {Christensen, 2016, 3858533;
Girardi, 2019, 6315730; Itoh, 2021, 9959632; Lin, 2020, 6835434; Liu, 2021, 10176563;
Omoike, 2021, 7021502; Tsai, 2020,6833693; Cao, 2022, 10412870}. One study conducted in
the high exposure to PFAS Ronneby Register Cohort in Sweden was aninformative {Li, 2022,
9961926} because of concerns about exposure assessment and lack of data on important
covariates. One study conducted in Greenland was aninformative {Wielsoe, 2017, 3858479}
because of concerns about selection bias and exposure assessment. One study included a liver
cancer biomarker analysis which was aninformative due to lack of information on biomarker
measurement methods {Cao, 2022, 10412870}. As a result, these two studies and the biomarker
analysis 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
3-288
-------
APRIL 2024
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. 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. Limited details or
reporting issues were also a concern for some low confidence studies which resulted in difficulty
in quantitatively interpreting analysis results {Cao, 2022, 10412870}.
3-289
-------
APRIL 2024
,
,0®
Christensen et al.
Cohn et al.
Ducatman et al.
Fry et al.
Ghisari et al.
Girardi et al.
2016, 3858533
2020, 5412451 -
2015, 3859843-I
2017. 4181820-
2017, 3860243
2019, 6315730-
Goodrich et al., 2022, 10369722
Hurley et al., 2018, 5080646 -
Itoh et al., 2021, 9959632 -
Li et al., 2022, 9961926-
Lin et al., 2020, 6835434
Liu et al., 2021, 10176563
Mancini et al., 2019, 5381529
Omoike et al.
Shearer et al.
Tsai et al.
Wielsoe et al.
2020, 7021502-
2021, 7161466
2020, 6833693 -
2017, 3858479-
- -
+
-
-
+
+
2
- -
+
•
-
+
+
-
-
+
+
+
+
+
-
+
+
+
~
-
+
+
+
+
+
+
+
+
+
+
+
+
+
++
+
+
+
- -
+
-
-
-
+
-
-
++
+
++
+
+
+
-
+
+
-
++
+
++
+
+
+
- +
-
++
+
+
+
+
-
'
*
+
+
+
~
- -
+
+
+
+
+
-
-
+
-
++
-
+
+
+
-
++
++
+
+
+
++
+
+
+
-
-
+
+
+
+
-
1"
++ ++
-
++
+
+
+
H
-
+
-
+
+
-
-
-
++
-
+
+
~
Legend
E3
Good (metric) or High confidence (overall)
F
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
b
Critically deficient (metric) or Uninformative (overall)
*
Multiple judgments exist
Figure 3-74. Summary of Study Quality Evaluation Results for Epidemiology Studies of
PFOA Exposure 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
compared with 42 controls in blood samples collected from the children one week after
3-290
-------
APRIL 2024
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 Appendix D, {U.S. EPA, 2024,
11414343}).
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 with 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
3-291
-------
APRIL 2024
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, lifestyle
factors and other PFAS. One low confidence study {Tsai, 2020, 6833693} conducted in Taiwan
observed an increased 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
Appendix D, {U.S. EPA, 2024, 11414343}). 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).
Two general population studies published since the 2016 assessment examined liver cancer
{Cao, 2022, 10412870; Goodrich, 2022, 10369722} and observed mixed results. One study was
considered medium confidence {Goodrich, 2022, 10369722} and one study was considered low
confidence {Cao, 2022, 10412870}. The medium confidence nested case-control study of U.S.
adults observed a nonsignificant increase in risk of liver cancer comparing participants with
PFOA exposure concentrations above the 85th percentile (8.6 ng/mL) compared with those at or
below (OR = 1.20; 95% CI: 0.52, 2.80). There was no association in analyses of continuous
PFOA exposure. However, the sample size was small (n = 50 cases and controls each) which
likely limited study sensitivity {Goodrich, 2022, 10369722}. Elevated risk of liver cancer was
also observed in a low confidence Chinese case-control study in adults and children (OR per log-
ng/mL increase in PFOA exposure = 1.036; 95% CI: 1.002, 1.070) {Cao, 2022, 10412870}.
However, the confidence in the study results was considered low due to limited information
regarding selection of controls, diagnosis method for liver cancer, adjustment for potential
confounding, and details on the statistical analysis.
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-years-old) or older (i.e., 50-69-years-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
versus 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 Appendix D, {U.S.
EPA, 2024, 11414343}). 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}.
3-292
-------
APRIL 2024
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
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 with the general population (median serum levels in workers was
113 ng/mL in 2005 compared with 4 ng/mL in the general population). There was modest
evidence of a positive nonsignificant 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 three studies from the 2016 PFOA HESD {U.S. EPA, 2016, 3603279} and two 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 five studies are shown in Figure 3-75.
3-293
-------
APRIL 2024
'pf* 'VsVV6* ^ V^V^ 0^
Abdellatif et al., 1991, 1290862-
+
NR
NR
~
~
-
Biegel et al., 2001, 673581 -
NR
++ ++
+
Butenhoff et al., 2012, 2919192 -\
Filgo et al.,2015, 2851085-
NTR2020, 7330145 A
+*
~Legend
Good (metric) or High confidence (overall)
+ Adequate (metric) or Medium confidence (overall)
- Deficient (metric) or Low confidence (overall)
S Critically deficient (metric) or Uninformative (overall)
Not reported
* Multiple judgments exist
Figure 3-75. Summary of Study Quality Evaluation Results for Animal Toxicological
Studies of PFOA Exposure 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}, Ley dig cell tumors (LCTs)
{Biegel, 2001, 673581; Butenhoff, 2012, 2919192}, and pancreatic acinar cell tumors (PACTs;
adenomas or adenocarcinomas) {Biegel, 2001, 673581; NTP, 2020, 7330145} in male Sprague-
Dawley rats. Neoplastic lesions were also observed in female Sprague-Dawley rats, but the
incidences were not as high as the incidences observed in the males and often did not achieve
statistical significance, though there were reported incidences of rare and/or malignant
neoplasms of the liver, pancreas, and uterus {Butenhoff, 2012, 2919192; NTP, 2020, 7330145}.
Another study {Filgo, 2015, 2851085} assessed hepatic tumor development in three strains of
female mice after perinatal exposures to PFOA. This study 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
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).
3-294
-------
APRIL 2024
Liver adenomas in male rats were observed in the Biegel et al. {, 2001, 673581} study at an
incidence of 10/76 (13%) at 13.6 mg/kg/day, compared with 2/80 (3%) in controls. Liver
adenomas in male rats were also significantly increased in the NTP {, 2020, 7330145} in the
0/40, 0/80, and 300/80 ppm groups (Table 3-16). Both the 0/0 and 300/0 ppm control groups had
no observed liver adenomas. NTP {, 2020, 7330145} reported increases in the incidence of
hepatocellular carcinomas in the male 300/80 ppm group only and a statistically significant trend
of increased incidence with dose in the groups exposed during both perinatal and postnatal
periods. Although no liver adenomas were observed in Butenhoff et al. {, 2012, 2919192},
carcinomas were identified in the male controls (3/49), males in the low-dose group
(1.3 mg/kg/day; 1/50), and male (5/50) and female (1/50) rats in the high-dose group (14.2 and
16.1 mg/kg/day, respectively). The differences in carcinoma incidences from controls were not
statistically significant in the Butenhoff et al. {, 2012, 2919192} study.
Table 3-16. Incidences of Liver Tumors in Male Sprague-Dawley Rats as Reported by NTP
{, 2020, 7330145}
Perinatal Dose
Postweaning Dose
0 ppm 20 ppm
40 ppm
80 ppm
Hepatocellular Adenomas
0 ppm
0/50 (0%)*** 0/50 (0%)
7/50 (14%)*
11/50 (22%)**
300 ppm
0/50 (0%)*** 1/50 (2%)
Hepatocellular Carcinomas
5/50 (10%)
10/50 (20%)**
0 ppm
0/50 (0%) 0/50 (0%)
0/50 (0%)
0/50 (0%)
300 ppm
0/50 (0%)* 0/50 (0%)
0/50 (0%)
4/50 (8%)
Notes:
* Statistically significant compared with the respective control group (0/0 or 300/0 ppm) at p < 0.05.
**Statistically significant compared with the respective control group (0/0 or 300/0 ppm) at p < 0.01.
* "Statistically significant trend of response at p < 0.001.
Nonneoplastic/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 {, 2020, 7330145} similarly reported a variety of nonneoplastic
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 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. {, 2001, 673581} study included one dose group
(13.6 mg/kg/day); the tumor incidence was 8/76 (11%) compared with 0/80 (0%) in the control
group. LCT incidence at similar dose levels was comparable between the two studies (11% and
3-295
-------
APRIL 2024
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 etal. {, 2001, 673581}
studies. NTP {, 2020, 7330145} reported increased incidences of pancreatic acinar cell
adenomas in males in all treatment groups compared with their respective controls (Table 3-17).
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 (Table 3-17 and Table 3-18). 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 with 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 (21%), 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 3-17. Incidences of Pancreatic Acinar Cell Tumors in Male Sprague-Dawley Rats as
Reported by NTP {, 2020, 7330145}
Postweaning Dose
Perinatal Dose
0 ppm 20 ppm 40 ppm 80 ppm
Pancreatic Acinar Cell Adenomas
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%)**
Pancreatic Acinar Cell Adenocarcinomas
0 ppm 0/50(0%) 3/50(6%) 1/50(2%) 3/50(6%)
300 ppm 0/50 (0%) 2/50 (4%) 1/50 (2%) 3/50 (6%)
Notes:
* Statistically significant compared with the respective control group (0/0 or 300/0 ppm) at p < 0.05.
**Statistically significant compared with 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.
3-296
-------
APRIL 2024
Table 3-18. Incidences of Pancreatic Acinar Cell Tumors in Female Sprague-Dawley Rats
as Reported by NTP {, 2020, 7330145}
Perinatal Dose
Postweaning Dose
0 ppm 300 ppm
1,000 ppm
Pancreatic Acinar Cell Adenomas
0 ppm
0/50 (0%) 0/50 (0%)
1/49 (2%)
150 ppm
0/50 (0%)
-
300 ppm
-
3/50 (6%)
Pancreatic Acinar Cell Adenocarcinomas
0 ppm
0/50 (0%) 0/50 (0%)
1/49 (2%)
150 ppm
0/50 (0%)
-
300 ppm
-
2/50 (4%)
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 3-19.
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 3-19. 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 with 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 {NTP, 2020, 7330145}.
3-297
-------
APRIL 2024
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 HE SD {U.S. EPA,
2016, 3603279}. There are 42 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 3-76.
Evidence Stream
Animal Human In Vitro Grand Total
Figure 3-76. Summary of Mechanistic Studies of PFOA and Cancer Effects
Interactive figure and additional study details available on HAWC.
In 2016, 10 key characteristics of carcinogens were selected by a multidisciplinary 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
(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
10 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 10 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' (key characteristic #1) and 'Alters DNA Repair and Causes Genomic Instability'
(key characteristic #3).
3.5.3.1 Key Characteristic #2: Is Genotoxic
Genotoxicity is a well-characterized 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. Overall, the evidence suggests that PFOA does not induce mutations or
operate through a genotoxic mechanism, with the majority of the study data demonstrating a lack
of genotoxic effect of PFOA in both in vitro and in vivo assays. A notable exception is
aneuploidy and DNA fragmentation of sperm significantly associated with PFOA exposure in
humans. The genotoxicity evidence is detailed below.
3-298
-------
APRIL 2024
3.5.3.1.1 Gene Mutation
All of the studies identified in this assessment that investigated 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 3-20). 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 (xM). 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 3-20). 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.
3.5.3.1.2 DNA Damage
3.5.3.1.2.1 In Vivo Evidence
3.5.3.1.2.1.1 Human Studies
Two studies reported on the genotoxic potential of PFOA exposure in humans (Table 3-21).
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 with 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 with PFAS-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 PFAS-positive group compared with
the PFAS-negative group.
3-299
-------
APRIL 2024
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 3-21). 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.
Quantifying micronuclei formation in rats via optimal and reliable methods has been previously
described {Witt, 2000, 783839; WHO, 2020, 11347555; WHO, 2009, 10455024}. With the
exception of one micronucleus assay, there was no evidence for PFOA-induced genotoxic effects
after acute or subchronic exposures (Figure 3-16).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} reported using flow cytometry to analyze micronuclei formation in immature
polychromatic erythrocytes from the peripheral blood of male and female Sprague-Dawley rats.
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}.
3.5.3.1.2.2 In Vitro Evidence
3.5.3.1.2.2.1 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 3-22). 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.
3.5.3.1.2.2.2 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
3-22). 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
3-300
-------
APRIL 2024
cells exposed to >100 [xM PFOA for 24 hours in one study {Yao and Zhong, 2005, 5081563},
>10 [xM PFOA for 24 hours in another study {Wielsoe, 2015, 2533367}, and at 10 and 200 [xM
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 [xM 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 with 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}.
3.5.3.1.2.2.3 Micronuclei Formation
Three studies measured micronucleus formation in cells exposed to PFOA (Table 3-22). Buhrke
et al. {, 2013, 2325346} demonstrated that PFOA exposure (10 [xM, 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 [xM 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 [xM for 24 hours compared with controls {Florentin et al., 2011, 2919235}. The
micronucleus assays were performed according to the same method {Natarajan, 1991, 5143588}.
Table 3-20. Mutagenicity Data from In Vitro Studies
Reference
Cell Line or
Results
Concentration
Bacterial Strain
S9-Activated
Non-Activated
(Duration of exposure)
NTP {,
2019,
5400977}
Salmonella
tvphimurium (TA98,
TA100)
Escherichia coli
(WP2«vrA/pkM101)
Equivocal3
(Not reproducible)
Negative
Equivocal3
(Not reproducible)
Negative
100-5,000 |xg/plate
100-10,000 |xg/plate
Zhao et al.
{,2011,
847496}
Human-hamster
hybrid (Al) cells
Mitochondrial DNA-
deficient human-
hamster hybrid
(P°Al) cells
N/A
N/A
Positiveb
Negative
1-200 pM
(1-16 d)
1-200 pM
(1-16 d)
Sadhu {,
2002,
10270882}
CHOK-1
Negative
Negative
<39 |xg/mL
(5 or 17 hr)
Butenhoff et Salmonella
al. {,2014, tvphimurium (TA98,
5079860} TA100, TA1535,
TA1537)
Positive0
Negative
20-1,000 |xg/plate
Buhrke et
al. {,2015,
2850235}
Salmonella
txphimurium (TA98,
TA100, TA1535,
TA1537, TA1538)
Negative
Negative
5 pM
3-301
-------
APRIL 2024
Cell Line or
Results
Concentration
Reference
Bacterial Strain
S9-Activated
Non-Activated
(Duration of exposure)
Fernandez
Salmonella
Negative
Negative
100 or 500 pM
Friere et al.
typhimurium (TA98,
{, 2008,
TA100, TA102,
2919390}
TA104)
Lawlor {,
Salmonella
Negative
Negative
100-5,000 |xg/plate
1995,
typhimurium (TA98,
10228128}
TA100, TA1535,
TA1537)
Salmonella
Negative
Negative
100-5,000 |xg/plate
typhimurium (TA98,
TA100, TA1535,
TA1537)
Escherichia coli
Negative
Negative
100-5,000 |xg/plate
(WP2uvrA)
Escherichia coli
Negative
Negative
6.67-5,000 |xg/plate
(WP2uvrA)
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 |xM for 16 days.
c Mutagenicity found at cytotoxic concentrations only.
Table 3-21. DNA Damage Data from In Vivo Studies
Reference
Species, Strain
(Sex)
Tissue
Results
PFOA Concentration
(Dosing Regimen)
DNA Strand Breakage
Frankenetal. {,
Human
Peripheral Blood
Positive
Average Blood Concentration of
2017,3789256}
(Male and
Female)
Cells
2.55 (ig/L
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 wk)
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 wk)
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 d)
Murli {, 1995,
Mouse
Bone Marrow
Negative
1,250-5,000 mg/kg
10228120}
Mouse
Bone Marrow
Negative
(Single dose delivered via gavage)
498-1,990 mg/kg
(Single dose delivered via gavage)
Notes: f.w. = formula weight.
1 A slight increase in micronuclei in the male 10 mg/kg/day group was within the historical control range. No change in females.
3-302
-------
APRIL 2024
Table 3-22. 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
al. {, 2014, Lymphocytes
5079860} Chinese Hamster Positive
Negative
N/A
12.4-600 (ig/mL
(3-46 hr)
50-1,500 (ig/mL
Ovarian Cells
(3 hr)
Chinese Hamster N/A
Positive
25-1,000 (ig/mL
Ovarian Cells
(Not reproducible)
(3-41.8 hr)
NOTOX {, Human
2000, Lymphocytes
10270878}
Negative
Negative
-------
APRIL 2024
Reference In Vitro Model
S9 Activated
Results
Non-Activated
Concentration
(Duration of exposure)
Yao and Human HepG2 N/A
Positive"
50-400 \M
Zhong {, 2005, Cells
(24 hr)
5081563}
Notes: N/A = not applicable.
aFindings based on the 2016 EPA's Health Effects Support Document {U.S. EPA, 2016, 3603279}, concentration(s) unknown.
b Slight increase was observed at 10 and 200 |xM in a non-dose-dependent manner after 24-hour exposure only.
cMicronuclei were present in cells that were exposed only to >100 |xM 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 {Smith, 2016, 3160486}. Overall, the evidence
demonstrates that PFOA exposure can lead to cancer-relevant changes in DNA methylation at
both the global and gene-specific level, across human, animal, and in vitro studies. The evidence
related to epigenetic alterations is detailed below.
3.5.3.2.1.1 In Vivo Evidence
3.5.3.2.1.2 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
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.
3-304
-------
APRIL 2024
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 HAWC visual for
additional supporting information and study details).
3.5.3.2.1.3 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
translocation (Tet) 1/2/3 methylcytosine dioxygenases was decreased at low doses of PFOA
exposure compared with 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 [xM, 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 [xM PFOA, 48 hours), and Liu and
Irudayaraj {, 2020, 6512127} in a study of MCF7 cells (20-400 [xM 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 [xM, 48 hours) in HepG2 cells. Further analysis
found PFOA (200 (xM) 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
3-305
-------
APRIL 2024
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 {Smith, 2016,
3160486}. Overall, the evidence supports that oxidative stress can result from PFOA exposure,
based on animal and in vitro studies. The evidence related to oxidative stress is detailed below
and in the referenced sections.
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 damage 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 with 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 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-mx\\
(129S4/SvJae-PparatmlGonz/J) mice to PFOA (<50 (j,mol/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, Ppara-null mice demonstrated a dose-dependent increase in 8-OHdG
levels, with a significant increase at 50 |imol/kg/day when compared with 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-
3-306
-------
APRIL 2024
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 HAWC visual for additional supporting information and study details).
Evidence that PFOA induces oxidative stress in the immune system has 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 HAWC
visual 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 HAWC for additional supporting
information and study details).
3.5.3.3.2 In Vitro Evidence
The ability of PFOA to induce oxidative stress has been assessed in vitro in several human,
nonhuman 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 [xM (3 hours) {Yao, 2005, 5081563}. Yao andZhong {, 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 nonsignificantly 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%.
3-307
-------
APRIL 2024
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 [xM Both
concentrations significantly increased hydrogen peroxide and superoxide anions. Wielsoe et al.
{, 2014, 2533367} noted nonsignificant 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 [xM 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 [xM)
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 in 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 [j,g/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 [xM; 1-
16 days) and found significantly increased intracellular ROS, NO, and O2" levels at all
timepoints exposed to >100 [xM These increases correlated with cytotoxicity, which was
significant at all timepoints at 100 and 200 [xM 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 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 {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. Overall, the evidence
demonstrates that PFOA exposure is related to increased markers of inflammation in animal and
in vitro studies. The evidence related to chronic inflammation is detailed below.
3-308
-------
APRIL 2024
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 with 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 with 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,
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
3-309
-------
APRIL 2024
(Section 3.4.1.3), and cardiovascular (Section 3.4.3.3) mechanistic sections (refer to the
interactive HAWC visual 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 {Smith, 2016, 3160486}. 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). Overall,
the evidence demonstrates that PFOA exposure can alter and impair immune and inflammatory
response and function in both humans and animals, as detailed briefly in the following paragraph
and in further detail in the referenced section.
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 HAWC visual 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) {Smith,
2016, 3160486}. Overall, the animal and in vitro evidence demonstrates that PFOA activates
several nuclear receptors: PPARa, CAR/PXR, ERa, and HNF4a, as detailed briefly in the
following paragraphs and in detail in the referenced sections.
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
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
3-310
-------
APRIL 2024
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 HAWC visual 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 [xM; 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 regarding 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 HAWC visual 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}. Overall, the limited evidence demonstrates that PFOA does not alter cell
transformation or cause immortalization, as detailed in the following paragraph.
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 3.7 x 10 4 to 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 [j,g/mL PFOA in a 14-day colony assay
for transformation nor in a 38-day foci transformation assay {Garry and Nelson, 1981,
10228130}.
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
3-311
-------
APRIL 2024
energetics and signaling pathways related to cellular replication or cell cycle control {Smith,
2016, 3160486}. Overall, the evidence demonstrates that PFOA exposure can increase cell
proliferation in animals and in cell models, and results are conflicting on the ability of PFOA to
induce or inhibit apoptosis. The evidence related to cell proliferation, cell death, and migration
(cancer cell invasiveness) is detailed below.
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 HAWC visual 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 (xM) 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
concentrations (>100 [xM PFOA) {Burhke, 2015, 2850235; Burhke, 2013, 2325346; Wen, 2020,
6302274}. Similar results were observed in the breast epithelial (nontumorigenic) cell line MCF-
3-312
-------
APRIL 2024
10A, in which PFOA exposure (50 and 100 pM; 24-72 hours) increased cell proliferation,
whereas proliferation rates decreased as the PFOA concentration was increased to a cytotoxic
level (250 pM) {Pierozan, 2018, 4241050}. A subsequent study by Pierozan et al. {, 2020,
6833637} reported that PFOA-induced (100 pM, 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 pg/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 pg/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
(j,M) 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 (j,M 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 HAWC
visual 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-
dependent manner {Wen, 2020, 6302274}. Apoptosis was also found to increase in HepG2 cells
after PFOA exposure (200 or 400 pM; <24 hours) and was associated with an increase in
3-313
-------
APRIL 2024
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 [xM) for 24 hours that reached significance at >500 [xM {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 [xM) 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 FTC 133 cells were exposed to 1 nM
(72 hours) {Saejia2019, 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, 72 hours) {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}, and DLD-1 (1 nM-1 [M, 72 hours) {Miao, 2015, 3981523} after PFOA
exposure.
Pierozan et al. {, 2020, 6833637} exposed MCF-10A cells to PFOA (100 [xM, 72 hours) and
found that invasion and migration of daughter cell passages was elevated when compared with
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}.
3-314
-------
APRIL 2024
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 HAWC visual 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 from medium confidence epidemiological studies is primarily
based on the incidence of kidney and testicular cancer, as well as some evidence of increased
breast cancer incidence in 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 high or medium confidence
chronic oral animal bioassays in Sprague-Dawley rats which together identified neoplastic
lesions of the liver, pancreas, and testes. The available mechanistic data suggest that multiple
MO As could play a role in the renal, testicular, pancreatic, and hepatic tumorigenesis associated
with PFOA exposure in human populations as well as animal models.
3.5.4.1.1 Evidence From Epidemiological Studies
The strongest evidence of an association between PFOA exposure and cancer in human
populations is from studies of kidney cancer. Two medium confidence studies of the C8 Health
Project population reported positive associations between PFOA levels (mean at enrollment
0.024 (j,g/mL) and kidney cancer among the residents living near the DuPont plant in
Parkersburg, West Virginia {Vieira, 2013, 2919154; Barry, 2013, 2850946}. Vieira et al. {,
2013, 2919154} reported elevated risk of kidney cancer in residents of the Little Hocking water
district of Ohio (OR: 1.7, 95% CI: 0.4, 3.3; n = 10) and the Tuppers Plains water district of Ohio
(OR: 2.0, 95% CI: 1.3, 3.1; n = 23). Barry et al. {, 2013, 2850946} extended this work, and
found increased risk of kidney cancer (HR: 1.10, 95% CI: 0.98, 1.24; n = 105), though the levels
did not reach statistical significance. 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. As part
of the C8 Health Project, the C8 Science Panel {, 2012, 9642155} concluded a probable link
between PFOA exposure and kidney cancer {Steenland, 2020, 7161469}.
The findings of another recently published medium confidence study add support to the previous
evidence of an association between PFOA and kidney cancer {Shearer, 2021, 7161466}. Shearer
et al. {, 2021, 7161466} is a multicenter case-control study nested within the National Cancer
Institute (NCI) Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial (n = 326).
The authors reported a statistically significant increase in risk of renal cell carcinoma (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). The
association remained significant in analyses on a per doubling increase in PFOA after adjusting
for other PFAS. 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
PFAS), but it was no longer statistically significant. Statistically significant increased odds of
3-315
-------
APRIL 2024
RCC were observed in a subgroup of participants ages 55-59 years, and in men and in women,
analyzed separately. A recent critical review and meta-analysis of the epidemiological literature
concluded that there was an increased risk for kidney tumors (16%) for every 10 ng/mL increase
in serum PFOA {Bartell, 2021, 7643457}. Although the authors concluded that the associations
were likely causal, they noted the limited number of studies and therefore, additional studies with
larger cohorts would strengthen the conclusion. Taken together, the recent pooled analysis of the
NCI nested case-control study {Shearer, 2021, 7161466} of 324 cases and controls and the C8
Science Panel Study {Barry, 2013, 2850946} of 103 cases and 511 controls provide evidence of
concordance in kidney cancer findings from studies of the general population and studies of
high-exposure communities {Steenland, 2022, 10607676}. CalEPA {, 2021, 9416932} similarly
concluded, "[t]here is evidence from epidemiologic studies that exposure to PFOA increases the
risk of kidney cancer."
There is also evidence of associations between PFOA serum concentrations and testicular cancer
in humans, though no new epidemiological studies reporting these associations have been
published since the studies described in the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}.
Similar to their results for kidney cancer, Vieira et al. {, 2013, 2919154} reported an increased
adjusted OR for testicular cancer (OR: 5.1, 95% CI: 1.6, 15.6; n = 8) in residents of the Little
Hocking water district of Ohio. Barry et al. {, 2013, 2850946} also found significantly increased
testicular cancer risk with an increase in estimated cumulative PFOA serum levels (HR: 1.34,
95% CI: 1.00, 1.79; n = 17). The C8 Science Panel {, 2012, 9642155} concluded that a probable
link also exists between PFOA exposure and testicular cancer {Steenland, 2020, 7161469}. A
recent critical review and meta-analysis of the epidemiological literature concluded that there
was an increased risk for testicular tumors (3%) for every 10 ng/mL increase in serum PFOA
{Bartell, 2021, 7643457} (see Appendix A, {U.S. EPA, 2024, 11414343}). In their review of the
available epidemiological data, 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, CalEPA {, 2021, 9416932}
concluded, "[ojverall, the epidemiologic literature to date suggests that PFOA is associated with
testicular cancer."
The majority of epidemiological studies examining the carcinogenicity after PFOA exposure
reported on breast cancer risk. Two nested case-control studies found associations between
PFOA exposure and breast cancer, but only in participants with known genetic susceptibility
(e.g., specific genotype or tumor estrogen receptor (ER) type) {Ghisari, 2017, 3860243; Mancini,
2020, 5381529}. In Taiwan, Tsai et al. {, 2020, 6833693} observed an increased risk of breast
cancer only in all women 50 years old or younger (including ER+ and ER- participants), 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 odds of breast cancer were also
observed in an NHANES population across serum PFOA quartiles with a significant dose-
response trend {Omoike, 2021, 7021502}. 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 serum after case
diagnosis and breast cancer {Hurley, 2018, 5080646}. One nested case-cohort study did not
report an association between breast cancer and PFOA concentrations measured in a group of
predominantly premenopausal women {Bonefeld-J0rgensen, 2014, 2851186}. In the C8 Health
Project cohort, Barry et al. {, 2013, 2850946} observed a significant inverse association with
3-316
-------
APRIL 2024
breast cancer for both unlagged (i.e., concurrent) and 10-year lagged (i.e., cumulative exposures
occurring 10 years in the past) estimated cumulative PFOA serum concentrations. Similarly, a
recent study in a Japanese population reported an inverse association across serum PFOA
quartiles with a significant dose-response trend {Itoh, 2021, 9959632}. Overall, study design
differences, lack of replication of the results, and a lack of mechanistic understanding of specific
breast cancer subtypes or susceptibilities of specific populations limit firm conclusions regarding
PFOA and breast cancer. However, there is suggestive evidence that PFOA exposure may be
associated with an increased breast cancer risk based on studies in populations with specific
genetic polymorphisms conferring increased susceptibility and for specific types of breast
tumors.
3.5.4.1.2 Evidence From Animal Bioassays
In addition to the available epidemiological data, two multidose 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 Leydig cell tumors (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%). Pancreatic acinar cell tumors (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 and
adenocarcinomas in males in all treatment groups compared with their respective controls (Table
3-17). These pancreatic tumor types were also observed in female rats in the highest dose group,
a rare occurrence compared with historical controls (0/340), 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 and carcinomas in male rats (Table 3-16) and Biegel et al. {, 2001, 673581}
also observed increased incidence of adenomas in male rats. Overall, NTP concluded that in their
2-year feeding studies, 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 oral exposure
accompanied by perinatal exposure (i.e., exposure beginning at gestation day 5 through lactation)
to PFOA does not increase cancer risk when compared with chronic exposure scenarios
beginning during the postnatal (i.e., exposure initiated after weaning) stage. The incidences of all
tumor types examined did not differ significantly between 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). Lifestage-dependent
sensitivity to the carcinogenic effects of PFOA exposure was previously assessed in the study by
Filgo et al. {, 2015, 2851085} which exposed two mouse strains during gestation only
(i.e., prenatal exposure with no comparisons to mice exposed through adulthood). Filgo et al. {,
2015, 2851085} observed a nonmonotonic increase in hepatocellular adenomas in the female
offspring of one strain (CD-I) and hepatocellular adenoma incidence in approximately 13% of
all PFOA-exposed peroxisome proliferator-activated receptor (PPAR) a-knockout mice.
3-317
-------
APRIL 2024
However, these results are not conclusive due to the study's limited sample size and study
design.
3.5.4.2 Mode of Action Analysis
In the 2016 PFOA HESD {U.S. EPA, 2016b, 3603279}, the EPA concluded that the induction of
tumors was likely due to multiple MO As, specifically noting 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,
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. The few available mechanistic studies focusing on PFOA-
induced renal toxicity highlight several potential underlying mechanisms of PFOA exposure-
induced renal tumorigenesis, including altered cell proliferation and apoptosis, epigenetic
alterations, and oxidative stress. However, due to data limitations, it is difficult to distinguish
which mechanism(s) are operative for PFOA-induced kidney cancer. Similarly for testicular
cancer, the available literature highlights several potential MOAs by which PFOA exposure may
result in increased incidence of LCTs in animals, though it is unclear whether these MOAs are
relevant to testicular cancers associated with PFOA exposure in humans.
As described in the following subsections, the available mechanistic data continue to suggest that
multiple MOAs 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 Mechanistic Evidence for Renal Tumors
As discussed in Section 3.5.13.4.5, 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
underlying mechanisms of PFOA exposure-induced 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 mechanisms involved in
tumorigenesis is described in the subsections below, which are all key characteristics of
carcinogens and may be related to 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 Appendix C, {U.S. EPA, 2024, 11414343}) {NTP, 2019, 5400977; Butenhoff,
3-318
-------
APRIL 2024
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 relatively high dose levels along with the enhanced efficiency of
excretion and increased urinary concentrations of PFOA in female rats (compared with 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 [xM 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 were
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 [xM.
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
3-319
-------
APRIL 2024
(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 versus control. PFOA exposure also altered mRNA expression of several proteins
that regulate DNA methylation, including DNA methyl transferases and 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.
3.5.4.2.2 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 MOAs 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 nongenotoxic hormonal MOAs,
(i.e., androgen receptor antagonism; testosterone biosynthesis inhibition; 5a-reductase inhibition;
aromatase inhibition; estrogen agonism; GnRH agonism; and dopamine agonism), five of which
were considered relevant to humans, and the majority of which involved downstream increases
in luteinizing hormone (LH) levels and subsequent Leydig cell hyperplasia/tumorigenesis. The
working group noted that sensitivity for the initiating events in these MOAs varies across
species, with rodents being more sensitive relative to humans. It has also 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}. However, CalEPA noted that "PFOA appears to act through multiple MOAs,
and the PPARa MOA does not adequately explain the incidences of pancreatic and testicular
tumors reported" {CalEPA, 2021, 9416932}.
The testes-specific evidence for the six human-relevant 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 that may also be relevant to the MOA for testicular cancers
associated with increased serum PFOA concentrations in humans.
3.5.4.2.2.1 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
3-320
-------
APRIL 2024
metabolism (5a-reductase inhibition, androgen receptor antagonism, aromatase inhibition,
estrogen agonism, testosterone biosynthesis inhibition). In addition, some compounds have been
shown to influence Ley dig cell function, including steroidogenesis, via hormone-mediated
MO As 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. Evidence for
the key events involved in the human-relevant MOAs for testicular tumors in rodents exposed to
PFOA is summarized in the paragraphs below and in Table 3-23, Table 3-24, Table 3-25, and
Table 3-26. There was no evidence of PFOA treatment resulting in 5a-reductase inhibition in the
identified literature, and the majority of the limited available in vitro studies for PFOA report
that PFOA does not act as an androgen receptor antagonist {McComb, 2019, 6304412; Kang,
2016, 3749062; Du, 2013, 2850983; Rosenmei, 2013, 2919164}. Thus, these two MOAs are not
summarized herein.
3.5.4.2.2.1.1 Aromatase Inhibition MOA
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 expression {Li, 2011, 1294081; Biegel, 1995,
1307447; Liu, 1996, 1307751}. A reduction in serum testosterone is also opposite of the
expected key event following aromatase inhibition (increased serum testosterone), further
supporting that PFOA does not operate through this MOA. The hepatic aromatase activity
provides some support for the MOA that is secondary to liver PPARa induction {Klaunig, 2003,
5772415}. Evidence demonstrating the lack of activity for the key events involved in the
aromatase inhibition MOA for testicular tumors, as presented in Clegg et al. {, 1997, 224277}, in
rodents exposed to PFOA is summarized in Table 3-23.
Table 3-23. Evidence of Key Events Associated with the Aromatase Inhibition Mode of
Action for Testicular Tumors3 in Male Rats and Mice Exposed to PFOA
„ . , Key Event 1: Key Event 2: Key Event 3: Key Event 4: Key Event 5: Outcome:
C lnnmr'il
MOA CYP19A1 Increased Decreased Increased Leydig Cell Testicular
Inhibition Serum T Serum E2 Serum LH Hyperplasia Tumor
Dose
(mg/kg/day)
CYP19A1
Activity in
Liver
Serum T
Serum E2
Serum LH
Leydig Cell
Hyperplasia
Testicular
Tumor
0.06
NR
- (4, 7, 13 wk)
- (4, 7, 13 wk)
- (4, 7, 13 wk)
NR
NR
0.2
- (14 d)
NR
- (14 d)
NR
NR
NR
0.31
NR
- (28 d)
NR
NR
NR
NR
0.64
NR
- (4, 7, 13 wk)
- (4, 7, 13 wk)
- (4, 7, 13 wk)
NR
NR
| (6 wk)
1
- (6 wk)
- (GD1-17)
- (14 d)
- (14 d)
- (6 wk)
- (6 wk)
- (14 d)
l.lb
t (16 wk)
NR
NR
NR
NR
- (16 wk)
1.25
NR
4 (28 d)
NR
NR
NR
NR
1.3
NR
NR
NR
NR
NR
-(105 wk)
3-321
-------
APRIL 2024
Canonical
MOA
Key Event 1:
CYP19A1
Inhibition
Key Event 2:
Increased
Serum T
Key Event 3:
Decreased
Serum E2
Key Event 4:
Increased
Serum LH
Key Event 5:
Leydig Cell
Hyperplasia
Outcome:
Testicular
Tumor
1.94
NR
- (4, 7, 13 wk)
- (4, 7, 13 wk)
- (4, 7, 13 wk)
NR
NR
2
t (14 d)
NR
t (14 d)
NR
NR
NR
2.2b
t (16 wk)
NR
NR
NR
NR
- (16 wk)
2.5
NR
| (GD1-17)
NR
NR
NR
NR
4.6b
t (16 wk)
4 (28 d)
NR
NR
NR
- (16 wk)
| (GD1-17)
5
- (6 wk)
i (6 wk)
NR
NR
- (6 wk)
- (6 wk)
6.5
NR
- (4, 7, 13 wk)
- (4, 7, 13 wk)
- (4, 7, 13 wk)
NR
NR
10
NR
- (14 d)
t (14 d)
- (14 d)
NR
NR
13.6
NR
t (26 wk)
t (4, 12, 26, 39,
| (78 wk)
| (104 wk)
t (104 wk)
-(4, 12,39, 52,
52 wk)
- (4, 12, 26, 39,
65, 78, 91 wk)°
- (65, 78,
52, 65, 91 wk)°
91 wk)°
14.2
NR
NR
NR
NR
NR
t (105 wk)
20
t (14 d)
1 (28 d)
t (14 d)
NR
NR
NR
I (1, 3, 5 d)
25
t (14 d)
- (14 d)
t (14 d)
- (14 d)
NR
NR
40
t (14 d)
NR
t (14 d)
NR
NR
NR
50
NR
- (14 d)
t (14 d)
- (14 d)
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls; MOA = mode of action; CYP19A1 = cytochrome P-450 19A1
(aromatase); T = testosterone; E2 = P-estradiol; LH = luteinizing hormone; NR = not reported; wk = week(s); d = day(s);
GD = gestational day.
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from Biegel et al. {, 1995,1307447}; Biegel et al. {, 2001, 673581}; Butenhoff et al. {, 2012,
2919192}; Cook et al. {, 1992, 1306123}; Li et al. {, 2011, 1294081}; Liu et al. {, 1996, 1307751}; Martin et al. {,2007,
758419}; NTP {, 2020, 7330145}; Perkins et al. {, 2004, 1291118}; Song et al. {, 2019, 5079725}; and Zhang et al. {, 2014,
2850230}. Data from Biegel et al. {, 2001, 673581} represent significant differences from pair-fed controls and/or from ad
libitum controls. Data from Li et al. {, 2011, 1294081} are in a hPPARa model.
3Reviewed in Clegg et al. {, 1997, 224277} andKlaunig et al. {, 2003, 5772415}.
bNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
c Biegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9,12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
3.5.4.2.2.1.2 Estrogen Agonism MOA
Although increased aromatase activity was observed, indicating potential increases in the
conversion of androgens to estrogens, 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
3-322
-------
APRIL 2024
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. Evidence for the key events involved in the estrogen
agonism MOA for testicular tumors, as presented in Clegg et al. {, 1997, 224277}, in rodents
exposed to PFOA is summarized in Table 3-24.
Table 3-24. Evidence of Key Events Associated with the Estrogen Agonism Mode of Action
for Testicular Tumors3 in Male Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1:
PPARa
Activation in
Liver
Key Event 2:
Increased
CYP19A1
Activity in
Liver
Key Event 3:
Increased
Serum E2
Key Event 4:
Increased
TGFa in
Testis
Key Event 5:
Increased
Serum LH
Key Event 6:
Leydig Cell
Hyperplasia
Outcome:
Testicular
Tumor
Dose
(mg/kg/day)
PPARa
Activation in
Liverb
CYP19A1
Activity in
Liver
Serum E2
TGFa in
Testis
Serum LH
Leydig Cell
Hyperplasia
Testicular
Tumor
0.06
NR
NR
- (4, 7, 13 wk)
NR
- (4, 7,
13 wk)
NR
NR
0.2
NR
- (14 d)
- (14 d)
NR
NR
NR
NR
0.64
NR
NR
- (4, 7, 13 wk)
NR
- (4, 7,
13 wk)
NR
NR
1
NR
- (6 wk)
- (14 d)
NR
- (14 d)
- (6 wk)
- (6 wk)
1.1°
t (16 wk)
t (16 wk)
NR
NR
NR
NR
- (16 wk)
1.3
NR
NR
NR
NR
NR
NR
-(105 wk)
1.94
NR
NR
- (4, 7, 13 wk)
NR
- (4, 7,
13 wk)
NR
NR
2
NR
t (14 d)
t (14 d)
NR
NR
NR
NR
2.2°
t (16 wk)
t (16 wk)
NR
NR
NR
NR
- (16 wk)
4.6°
t (16 wk)
t (16 wk)
NR
NR
NR
NR
- (16 wk)
5
NR
- (6 wk)
NR
NR
NR
- (6 wk)
- (6 wk)
6.5
NR
NR
- (4, 7, 13 wk)
NR
- (4, 7,
13 wk)
NR
NR
10
NR
NR
t (14 d)
NR
- (14 d)
NR
NR
13.6
t (4, 12, 26,
39, 52, 65,
78, 91 wk)d
NR
t (4, 12, 26, 39,
52 wk)
- (65, 78,
91 wk)d
NR
4 (78 wk)
- (4, 12, 26,
39, 52, 65,
91 wk)d
| (104 wk)
t (104 wk)
14.2
NR
NR
NR
NR
NR
NR
t (105 wk)
19
t (1 J. 28 d)
NR
NR
NR
NR
NR
NR
20
-(1,3, 5 d)
t (14 d)
t (14 d)
NR
NR
NR
NR
23
t (1. 7, 28 d)
NR
NR
NR
NR
NR
NR
25
NR
t (14 d)
t (14 d)
t (14 d)
- (14 d)
NR
NR
40
NR
t (14 d)
t (14 d)
NR
NR
NR
NR
50
NR
NR
t (14 d)
NR
- (14 d)
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J, = statistically
significant decrease in response compared with controls; MOA = mode of action; PPARa = peroxisome proliferator-activated
receptor a; CYP19A1 = cytochrome P-450 19A1 (aromatase); E2 = P-estradiol; TGFa = transforming growth factor a;
LH = luteinizing hormone; NR = not reported; w = week(s); d = day(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
3-323
-------
APRIL 2024
Data represented in table extracted from Biegel et al. {, 1995,1307447}; Biegel et al. {, 2001, 673581}; Butenhoff et al. {, 2012,
2919192}; Cook et al. {, 1992, 1306123}; Elcombe et al. {, 2010, 2850034}; Li et al. {,2011, 1294081}; Liu et al. {, 1996,
1307751}; Martin etal. {, 2007, 758419}; NTP {, 2020, 7330145}; and Perkins et al. {,2004, 1291118}. Data from Biegel etal.
{, 2001, 673581} represent significant differences from pair-fed controls and/or from ad libitum controls. Data from Li et al. {,
2011, 1294081} are in a hPPARa model.
3Reviewed in Clegg et al. {, 1997, 224277} andKlaunig et al. {, 2003, 5772415}.
b Indirect measurement ofPPARa induction provided as hepatic acyl-CoA oxidase activity in NTP {, 2020, 7330145}, as hepatic
P-oxidation activity in Biegel et al. {, 2001, 673581}, as CYP4A1 protein expression and hepatic P-oxidation activity in
Elcombe et al. {, 2010, 2850034}, and as Cvp4al4, C\p7al, Cvp7bl, Cvp8bl, and Cvpl7al gene expression in Martin et al. {,
2007,758419}.
CNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
dBiegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9, 12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
3.5.4.2.2.1.3 Testosterone Biosynthesis Inhibition MOA
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 to 21 months
{Perkins, 2004, 1291118; Biegel, 2001, 673581} (Table 3-25). 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}.
Evidence for the key events involved in the testosterone biosynthesis inhibition MOA for
testicular tumors, as presented in Clegg et al. {, 1997, 224277}, in rodents exposed to PFOA is
summarized in Table 3-25.
Table 3-25. Evidence of Key Events Associated with the Testosterone Biosynthesis
Inhibition Mode of Action for Testicular Tumors3 in Male Rats and Mice Exposed to
PFOA
Canonical MOA
Key Event
1: PPARa
Activation
Key Event 2:
Decreased
Testosterone
Biosynthesis
Key Event
3:
Decreased
Serum T
Key Event
4:
Increased
Serum LH
Key Event 5:
Leydig Cell
Hyperplasia
Outcome:
Testicular
Tumor
Dose (mg/kg/day)
PPARa
Activation
in Liverb
Testosterone
Biosynthesis0
Serum T
Serum LH
Leydig Cell
Hyperplasia
Testicular
Tumor
0.06
NR
NR
- (4, 7,
13 wk)
- (4, 7,
13 wk)
NR
NR
0.31
NR
- (28 d)
- (28 d)
NR
NR
NR
0.64
NR
NR
- (4, 7,
13 wk)
- (4, 7,
13 wk)
NR
NR
1
NR
| (6 wk)
| (6 wk)
- (14 d)
- (6 wk)
- (6 wk)
- (14 d)
- (GD1-17)
3-324
-------
APRIL 2024
Canonical MOA
Key Event
1: PPARa
Activation
Key Event 2:
Decreased
Testosterone
Biosynthesis
Key Event
3:
Decreased
Serum T
Key Event
4:
Increased
Serum LH
Key Event 5:
Leydig Cell
Hyperplasia
Outcome:
Testicular
Tumor
l.ld
t (16 wk)
NR
NR
NR
NR
- (16 wk)
1.25
NR
4 (28 d)
1 (28 d)
NR
NR
NR
1.3
NR
NR
NR
NR
NR
-(105 wk)
1.94
NR
NR
- (4, 7,
- (4, 7,
NR
NR
13 wk)
13 wk)
2.2d
t (16 wk)
NR
NR
NR
NR
- (16 wk)
2.5
NR
NR
| (GD1-17)
NR
NR
NR
4.6d
t (16 wk)
4 (28 d)
1 (28 d)
NR
NR
- (16 wk)
| (GD1-17)
5
NR
| (6 wk)
i (6 wk)
NR
- (6 wk)
- (6 wk)
6.5
NR
NR
- (4, 7,
- (4, 7,
NR
NR
13 wk)
13 wk)
10
NR
NR
- (14 d)
- (14 d)
NR
NR
13.6
t (4, 12, 26,
NR
t (26 wk)
| (78 wk)
| (104 wk)
t (104 wk)
39, 52, 65,
-(4, 12,39,
- (4, 12,
78, 91 wk)e
52, 65, 78,
26, 39, 52,
91 wk)e
65, 91 wk)e
14.2
NR
NR
NR
NR
NR
t (105 wk)
19
td. 7.
NR
NR
NR
NR
NR
28 d)
20
-(1, 3, 5d)
1 (28 d)
1 (28 d)
NR
NR
NR
1(1,3, 5 d)
23
td. 7.
NR
NR
NR
NR
NR
28 d)
25
NR
NR
- (14 d)
- (14 d)
NR
NR
50
NR
NR
- (14 d)
- (14 d)
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls; MOA = mode of action; PPARa = peroxisome proliferator-activated
receptor a; T = testosterone; LH = luteinizing hormone; wk = week(s); d = day(s); GD = gestational day.
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from Biegel et al. {, 1995,1307447}; Biegel et al. {, 2001, 673581}; Butenhoff et al. {,2012,
2919192}; Cook et al. {, 1992, 1306123}; Elcombe et al. {, 2010, 2850034}; Li et al. {,2011, 1294081}; Liu et al. {, 1996,
1307751}; Martin et al. {, 2007, 758419}; NTP {, 2020, 7330145}; Perkins et al. {, 2004, 1291118}; Song et al. {, 2019,
5079725}; and Zhang et al. {, 2014, 2850230}. Data from Biegel et al. {, 2001, 673581} represent significant differences from
pair-fed controls and/or from ad libitum controls. Data from Li et al. {, 2011, 1294081} are in a hPPARa model.
3Reviewed in Clegg et al. {, 1997, 224277} andKlaunig et al. {, 2003, 5772415}.
b Indirect measurement of PPARa induction provided as hepatic acyl-CoA oxidase activity in NTP {, 2020, 7330145}, as hepatic
P-oxidation activity in Biegel et al. {, 2001, 673581}, as CYP4A1 protein expression and hepatic P-oxidation activity in
Elcombe et al. {, 2010, 2850034}, and as Cvp4al4, Cvp7al, C\p7bl, C\p8bl, and Cvpl7al gene expression in Martin et al. {,
2007,758419}.
c Testosterone biosynthesis provided as gene expression of 3P-HSD, 17-P-HSD, and/or CYP17A1 in Zhang et al. {, 2014,
2850230} and as gene expression of 3P-HSD, 17-P-HSD, and/or CYP17A1 inLietal. {,2011, 1294081}.
dNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
e Biegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9,12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
3-325
-------
APRIL 2024
3.5.4.2.2.1.4 PPARa activation MOA
Support for at least partial PPARa mediation of testosterone production inhibition due to PFOA
administration is available from one study in mice {Li, 2011, 1294081}. Significantly reduced
plasma testosterone concentrations were observed in male wild-type PPARa mice and
humanized PPARa transgenic mice. These decreases were evident but not 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, 1294081}. 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. Additional
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 downstream increases in serum
E2. There was also no indication of increased Leydig 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. Evidence for the key events involved in the PPARa agonist-induced
MOA for testicular tumors in rodents exposed to PFOA is summarized in Table 3-26.
Table 3-26. Evidence of Key Events Associated with PPARa Agonist-Induced Mode of
Action for Testicular Tumors3 in Male Rats and Mice Exposed to PFOA
Canonical MOA
Key Event
" 1:
PPARa
Activation in
Liver
Key Event 2:
Increased
CYP19A1
Activity in
Liver
Key Event 3:
Increased Serum
E2
Key Event 4:
Increased
TGFa in
Testis
Key Event 5:
Leydig Cell
Hyperplasia
Outcome:
Testicular Tumor
PPARa
Dose (mg/kg/day) Activation in
Liverb
CYP19A1
Activity in
Liver
Serum E2
TGFa in
Testis
Leydig Cell
Hyperplasia
Testicular Tumor
0.06
NR
NR
- (4, 7, 13 wk)
NR
NR
NR
0.2
NR
- (14 d)
- (14 d)
NR
NR
NR
0.64
NR
NR
- (4, 7, 13 wk)
NR
NR
NR
1
NR
- (6 wk)
- (14 d)
NR
- (6 wk)
- (6 wk)
1.1°
t (16 wk)
t (16 wk)
NR
NR
NR
- (16 wk)
1.3
NR
NR
NR
NR
NR
-(105 wk)
1.94
NR
NR
- (4, 7, 13 wk)
NR
NR
NR
2
NR
t (14 d)
t (14 d)
NR
NR
NR
2.2°
t (16 wk)
t (16 wk)
NR
NR
NR
- (16 wk)
4.6°
t (16 wk)
t (16 wk)
NR
NR
NR
- (16 wk)
5
NR
- (6 wk)
NR
NR
- (6 wk)
- (6 wk)
6.5
NR
NR
- (4, 7, 13 wk)
NR
NR
NR
10
NR
NR
t (14 d)
NR
NR
NR
13.6
t (4, 12, 26,
39, 52, 65,
78, 91 wk)d
NR
t (4, 12, 26, 39,
52 wk)
- (65, 78,
91 wk)d
NR
| (104 wk)
t (104 wk)
14.2
NR
NR
NR
NR
NR
t (105 wk)
3-326
-------
APRIL 2024
Canonical MOA
Key Event
" 1:
PPARa
Activation in
Liver
Key Event 2:
Increased
CYP19A1
Activity in
Liver
Key Event 3:
Increased Serum
E2
Key Event 4:
Increased
TGFa in
Testis
Key Event 5:
Leydig Cell
Hyperplasia
Outcome:
Testicular Tumor
19
td. 7.
28 d)
NR
NR
NR
NR
NR
20
-(1, 3, 5d)
t (14d)
t (14 d)
NR
NR
NR
23
td. 7.
28 d)
NR
NR
NR
NR
NR
25
NR
t (14 d)
t (14 d)
t (14 d)
NR
NR
40
NR
t (14 d)
t (14 d)
NR
NR
NR
50
NR
NR
t (14 d)
NR
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls; MOA = mode of action; PPARa = peroxisome proliferator-activated
receptor a; CYP19A1 = cytochrome P-450 19A1 (aromatase); E2 = P-estradiol; TGFa = transforming growth factor a; NR = not
reported; wk = week(s); d = day(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in the table were extracted from Biegel et al. {, 1995, 1307447}; Biegel et al. {, 2001, 673581}; Butenhoffet al.
{,2012, 2919192}; Cook etal. {, 1992, 1306123}; Elcombe et al. {, 2010, 2850034}; Li et al. {, 2011, 1294081}; Liu et al. {,
1996, 1307751}; Martin et al. {, 2007, 758419}; NTP {, 2020, 7330145}; and Perkins et al. {, 2004,1291118}. Data from
Biegel et al. {, 2001, 673581} represent significant differences from pair-fed controls and/or from ad libitum controls. Data from
Li etal. {,2011,1294081} are in a hPPARa model.
3Reviewed in Clegg et al. {, 1997,224277} and Klaunig et al. {,2003,5772415}.
b Indirect measurement of PPARa induction provided as hepatic acyl-CoA oxidase activity in NTP {, 2020, 7330145}, as hepatic
P-oxidation activity in Biegel et al. {, 2001, 673581}, as CYP4A1 protein expression and hepatic P-oxidation activity in
Elcombe et al. {, 2010, 2850034}, and as Cvp4al4, C\p7al, Cvp7bl, Cvp8bl, and Cvpl7al gene expression in Martin et al. {,
2007,758419}.
CNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
dBiegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9, 12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
3.5.4.2.3 Mode of Action for Pancreatic Tumors
As discussed in Section 3.5.2, pancreatic acinar cell tumors (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 increased incidences of pancreatic acinar cell adenomas in
males in all treatment groups, as well as increased incidence, though nonsignificant, 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}.
Two proposed MOAs for PFOA-induced pancreatic tumors in animal models were identified in
the literature, including one study that utilizes a transgenic mouse model to mimic the histologic
progression of pancreatic cancer in humans {Kamendulis, 2022, 10176439; Klaunig, 2003,
5772415; Klaunig, 2012, 1289837}. The proposed MOAs are: 1) changes in bile acids,
potentially linked to activation of hepatic PPARa, leading to cholestasis, a positive
cholecystokinin (CCK) feedback loop, and acinar cell proliferation; and 2) oxidative stress.
3-327
-------
APRIL 2024
However, the existing database is limited in its ability to determine the relationship between
PFOA exposure and these MO As, particularly for the PACTs observed in chronic rat studies.
Evidence for the key events involved in the relevant MO As for pancreatic tumors in rodents
exposed to PFOA is summarized in Table 3-27 and Table 3-28.
3.5.4.2.3.1 Gastric Bile Alterations
Gastric bile compositional changes or flow alterations can lead to cholestasis, which is the
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 evidence suggesting 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 the liver via bile. Minata et al. {,2010, 1937251} found that PFOA levels in bile
were much higher in wild-type male mice versus 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 with those that
resulted in PACTs in male rats following 2 years of PFOA exposure {NTP, 2020, 7330145}. In
the NTP study, bile acid concentrations were increased greater than twofold in male rats exposed
to PFOA in the diet at doses of 15.6 and 31.7 mg/kg/day for 4 weeks compared with the control
group. In the same study, serum ALP levels were mildly increased (less than twofold). While
these increases may be due to cholestasis, mild increases in ALP (and ALT) activity are also
associated with the administration of hepatic microsomal enzyme inducer compounds, including
PPARa agonists {NTP, 2020, 7330145}. There was no further evidence of cholestasis reported
in the literature. Additionally, CalEPA noted that "PFOA appears to act through multiple MO As,
and the PPARa MOA does not adequately explain the incidences of pancreatic and testicular
tumors reported" {CalEPA, 2021, 9416932}.
Additionally, there is no evidence of alterations in CCK associated with PFOA exposure 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, due to limited evidence for altered bile flow in
animals that developed tumors and an overall lack of evidence for alterations in CCK levels in
PFOA-exposed animals, there is not sufficient evidence to determine whether bile acid
alterations contribute to the MOA for PACTs observed in rodents chronically exposed to PFOA.
3-328
-------
APRIL 2024
Evidence for the key events involved in the gastric bile acid alteration MOA for pancreatic
tumors in rodents exposed to PFOA is summarized in Table 3-27.
Table 3-27. Evidence of Key Events Associated with the Gastric Bile Alterations Mode of
Action for Pancreatic Tumors3 in Male and Female Rats and Mice
Canonical
MOA
Key Event 1:
PPARa
Activation in
Key Event 2:
Altered Bile Flow
and/or Bile Acid
Key Event 3:
Cholestasis
Key Event 4:
Increase in
CCK Levels
Key Event 5:
Acinar Cell
Proliferation or
Outcome:
Pancreatic
Tumors
Liver
Composition
Hyperplasia
Dose
PPARa
Altered Bile Flow
Cholestasis0
CCK Levels
Acinar Cell
Pancreatic
(mg/kg/day)
Activation in
Liverb
and/or Bile Acid
Composition
Proliferation or
Hyperplasia
Tumors'1
l.le
t (16 wk)
NR
t (16 wk) for
ALT, ALP,
SDH
- (16 wk) for
bile acids
NR
t (104 wk)
t (104 wk)
1.3 (males)/
NR
NR
NR
NR
-(105 wk)
-(105 wk)
1.6 (females/
2.2e
t (16 wk)
NR
t (16 wk) for
ALT, ALP,
SDH
- (16 wk) for
bile acids
NR
t (104 wk)
t (104 wk)
4.6e
t (16 wk)
NR
t (16 wk) for
ALT, ALP,
SDH
- (16 wk) for
bile acids
NR
t (104 wk)
t (104 wk)
5.4
NR
- (4w)
t (4 wk) for
ALT
i (4 wk) for
bilirubin
- (4 wk) for
AST, bile acid
NR
NR
NR
10.8
NR
- (4w)
t (4 wk) for
AST, ALT
- (4 wk) for
bile acid,
bilirubin
NR
NR
NR
13.6
t (4, 12, 26, 39,
52, 65, 78,
91 wk)g
NR
NR
NR
t (104 wk)
t (104 wk)
14.2 (males)/
NR
NR
NR
NR
-(105 wk)
-(105 wk)
16.1 (females)
15.6
t (16 wk)
NR
t (16 wk) for
ALP, ALT,
SDH, bile acid
NR
NR
NR
18.2
NR
- (16 wk) for
NR
-(104 wk)
-(104 wk)
(females)6
t (16 wk)
ALP, ALP,
SDH
19
t (1. 7, 28 d)
NR
NR
NR
NR
NR
20
-(1, 3, 5 d)
NR
NR
NR
NR
NR
3-329
-------
APRIL 2024
Canonical
Key Event 1:
PPARa
Key Event 2:
Altered Bile Flow
Key Event 3:
Key Event 4:
Increase in
CCK Levels
Key Event 5:
Acinar Cell
Outcome:
Pancreatic
Tumors
MOA
Activation in
and/or Bile Acid
Cholestasis
Proliferation or
Liver
Composition
Hyperplasia
NR
- (4 wk)
t (4 wk) for
AST, ALT,
NR
NR
NR
21.6
bilirubin
- (4 wk) for
bile acid
23
t (1. 7, 28 d)
NR
NR
NR
NR
NR
31.7
t (16 wk)
NR
t (16 wk) for
NR
NR
bile acid.
NR
ALP, ALT,
SDH
40
NR
t(2 d)
NR
NR
NR
NR
63.4
(females)6
t (16 wk)
NR
t (16 wk) for
ALT, ALP
NR
-(104 wk)
-(104 wk)
80
NR
t(2 d)
NR
NR
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls; MOA = mode of action; PPARa = peroxisome proliferator-activated
receptor a; CCK = cholecystokinin; wk = week(s); NR = not reported; ALT = alanine transaminase; ALP = alkaline
phosphatase; SDH = sorbitol dehydrogenase; AST = aspartate transferase; d = day(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in the table were extracted from: Biegel et al. {, 2001, 673581}; Butenhoff et al. {, 2012, 2919192}; Cheng et
al. {, 2008, 758807}; Elcombe et al. {, 2010, 2850034}; Kamendulis et al. {, 2022,10176439}; Martin et al. {, 2007, 758419};
NTP {, 2020, 7330145}; and from wild-type animals in Minata et al. {,2010, 1937251}. Doses in mg/kg/day for Minata et al. {,
2010, 1937251} were converted from 12.5, 25, and 50 |xmol/kg/d as reported in the primary study. Data from Biegel et al. {,
2001, 673581} represent significant differences from pair-fed controls and/or from ad libitum controls.
aReviewed in Klaunig, 2003, 5772415 and Klaunig, 2012, 1289837.
bIndirect measurement of PPARa induction provided as hepatic acyl-CoA oxidase activity {NTP, 2020, 7330145}, as hepatic P-
oxidation activity {Biegel, 2001, 673581}, and as CYP4A1 protein expression and hepatic P-oxidation activity {Elcombe, 2010,
2850034}.
c Observations consistent with cholestasis include significant increases in serum bile acid concentrations and increased serum
liver enzyme activities (e.g., ALP, ALT) in NTP {, 2020, 7330145}, and increased total bilirubin and ALT in Minata et al. {,
2010, 1937251}.
d Pancreatic tumors reflect increased incidence of acinar cell adenoma and/or adenocarcinoma (combined) in male rats {NTP,
2020, 7330145; Biegel, 2001, 673581}.
eNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures in male (20, 40, and 80 ppm, or 1.1, 2.2, and 4.6 mg/kg/day) and female (300 and 1,000 ppm,
or 18.2 and 63.4 mg/kg/day) rats in order to provide a representative set of the available mechanistic data involved in this MOA
from bioassays, and because the treatment effects were very similar in the perinatal and postweaning exposure groups. Further
study design details are in Section 3.4.4.2.1.2 and study results are in Section 3.5.2.
f All data are from male rats with the exception of Butenhoff et al. {, 2012, 2919192} and NTP {, 2020, 7330145}, which include
both males and females, as indicated.
gBiegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9, 12, 15,18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
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. Evidence for the key events involved in the
proposed oxidative stress MOA for pancreatic tumors in rodents exposed to PFOA is
summarized in Table 3-28. 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
3-330
-------
APRIL 2024
the oxidative stress response, and protein kinase-like endoplasmic reticulum kinase (PERK), a
signaler of endoplasmic reticulum (ER) stress, and subsequent upregulation of 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 oxidative stress 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 [j,g/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 |ig/mL 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 Hocevar et al. {, 2020, 6833720} 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 LSL-KRasG12D;Pdx-l Cre (KC) mouse model of pancreatic cancer, which has
a mutation in the KRas gene, a mutations that is present in over 90% of human pancreatic
cancers. This gene mutation 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 increased PanIN was observed after 4 and 7 months of treatment compared with untreated
KC mice.
Oxidative stress was also apparent in the PFOA-treated KC mice {Kamendulis, 2022,
10176439}. The authors reported increases in Sod enzyme activity at 4 and 7 months, along with
threefold 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 significantly increased at 7 months of exposure relative to
untreated mice, but not at 4 months, indicating a potential accumulation of oxidative damage
overtime. Altogether, the results of Kamendulis et al. {, 2022, 10176439} demonstrated that
PFOA increased PanIN area and number at 4 months, indicating early lesion formation. The
increased desmoplasia and inflammation (MDA levels) after 7 months of exposure suggest
PFOA increased disease severity over time, potentially through prolonged oxidative stress,
resulting in pancreatic cancer progression.
3-331
-------
APRIL 2024
Overall, although plausible, there is not sufficient evidence for key events related to an oxidative
stress MOA to conclude that the pancreatic tumors in rodents chronically exposed to PFOA are
the result of oxidative stress and related molecular events.
Table 3-28. Evidence of Key Events Associated with a Proposed Oxidative Stress Mode of
Action Involving the UPR Pathway for Pancreatic Tumors3 in Male and Female Rats and
Mice.
Canonical
MOA
Key Event
" 1:
Activation
of UPR
Pathway
Key Event
2a:
Activation
of Nrf2 and
PERK
Key Event
2b: ROS
Production
Key Event 3:
Upregulation
of
Antioxidant
Responses
Key Event
" 4:
Increased
Production
Key Event
5a:
Increased
Cell
of Hydrogen „ Apoptosis
_ J . p Proliferation 1 1
Peroxide
Key
Event 5b:
Decreased
Outcome:
Pancreatic
Tumors
Dose
(mg/kg/day)
UPR
Pathway
Nrf2 and
PERK
ROS
Production
Antioxidant
Response
Hydrogen
Peroxide
Production
Cell
Proliferation
Apoptosis
Pancreatic
Tumorsb
l.lc
NR
NR
NR
NR
NR
t (104 wk)
NR
t (104 wk)
1.28d
NR
NR
t (28 wk)
t (16 wk)
NR
NR
NR
t (16 wk)
1.3 (males)/
1.6
(females)6
NR
NR
NR
NR
NR
-(105 wk)
NR
-(105 wk)
2.2°
NR
NR
NR
NR
NR
t (104 wk)
NR
t (104 wk)
4.6°
NR
NR
NR
NR
NR
t (104 wk)
NR
t (104 wk)
13.6
NR
NR
NR
NR
NR
t (104 wk)
NR
t (104 wk)
14.2 (males)/
16.1
(females)
18.2°
(females)
63.4°
(females)
NR
NR
NR
NR
NR
-(105 wk)
NR
-(105 wk)
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
- (104 wk)
- (104 wk)
NR
NR
- (104 wk)
- (104 wk)
50 |ig/mLr
t (in vitro)
t (in vitro)
NR
NR
NR
NR
NR
NR
Notes', t = statistically significant increase in response compared with controls; - = no significant response; UPR = unfolded
protein response; MOA = mode of action; ROS = reactive oxygen species; Nrf2 = nuclear factor erythroid 2-related factor 2;
PERK = protein kinase-like endoplasmic reticulum kinase; NR = not reported; wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in the table were extracted from: Biegel et al. {, 2001, 673581}; Butenhoff et al. {, 2012, 2919192};
Kamendulis et al. {,2022, 10176439}; andNTP {, 2020, 7330145}. Data from Biegel et al. {,2001,673581} represent
significant differences from pair-fed controls and/or from ad libitum controls.
aReviewed in Hocevaretal. {,2020,6833720} and Kamendulis et al. {,2022, 10176439}.
b Pancreatic tumors reflect increased incidence of acinar cell adenoma and/or adenocarcinoma (combined) in male rats {NTP,
2020, 7330145; Biegel, 2001, 673581}.
CNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures in male (20, 40, and 80 ppm, or 1.1, 2.2, and 4.6 mg/kg/day) and female (300 and 1,000 ppm,
or 18.2 and 63.4 mg/kg/day) rats in order to provide a representative set of the available mechanistic data involved in this MOA
from bioassays, and because the treatment effects were very similar in the perinatal and postweaning exposure groups. Further
study design details are in Section 3.4.4.2.1.2 and study results are in Section 3.5.2.
dDose from Kamendulis et al. {, 2022, 10176439} converted from 5 ppm by summary authors using default assumptions for food
consumption, water consumption, and body weight, in the absence of such data in the primary study, which used a Kras
mutation model of mouse pancreatic cancer.
eAll data are from male rats with the exception of Butenhoff et al. {, 2012, 2919192} and NTP {, 2020, 7330145}, which include
both males and females, as indicated.
3-332
-------
APRIL 2024
fIndicates in vitro evidence from Hocevar et al. {, 2020, 6833720}, which used mouse pancreatic acinar cells (266-6 cells); data
are included here owing to the only available demonstration of two of the key events in the proposed MOA.
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, cytotoxicity, and an oxidative stress-mediated MOA. Evidence
for the key events involved in the relevant MO As for hepatic tumors in rodents exposed to PFOA
is summarized in Table 3-29,Table 3-30,Table 3-31,Table 3-32, Table 3-33, and Table 3-34.
Evidence related to genotoxicity and other plausible modes of action are also detailed in
subsequent sections.
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.
Evidence exists suggesting that although PPARa activators cause liver tumors in rodents, they
may be 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, some have argued 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 plausible MOAs
in addition to PPARa activation suggests 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 has 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
3-333
-------
APRIL 2024
tumorigenesis 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 MOA 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 MOA 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, nonneoplastic 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 tumorigenesis, 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 [xM. 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}.
Nonneoplastic (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 nonneoplastic 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 chronic
carcinogenicity 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 Ppara-mi\\ (129S4/SvJae-PparatmlGonz/J) mice
treated with PFOA (<50 (j,mol/kg/day) for four weeks, while there was no effect of PFOA on
wild-type (129S4/SvlmJ) mice in the same study. TNFa is a pro-inflammatory cytokine that can
be released upon activation of Kupffer cells {Corton, 2018, 4862049}. Further study is needed to
3-334
-------
APRIL 2024
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.
Studies in both rats and mice have demonstrated (either directly or indirectly) that PFOA induces
peroxisome proliferation in the liver, an indication of PPARa activation {Elcombe, 2010,
2850034; Martin, 2007, 758419; Minata, 2010, 1937251; Pastoor, 1987, 3748971; Wolf, 2008,
1290827; Yang, 2001, 1014748}. Gene expression profiling of HepG2 cells exposed to low
PFOA concentrations (0.1 and 1 |iM) revealed 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
tumors in rats, Abdellatif et al. {, 1990, 2328171} reported that PFOA had promoting activity
and increased the incidence of hepatocellular carcinomas following tumor initiation with
diethylnitrosamine (DEN). Jacquet et al. {, 2012, 2124683} exposed SHE cells to PFOA at
concentrations ranging from 3.7 x 10 4 to 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.
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 in Section
3.5.4.2.4.5.2). Evidence for the key events involved in the PPARa activation MOA for hepatic
tumors in male and female rodents exposed to PFOA is summarized in Table 3-29 and Table
3-30, respectively.
Table 3-29. Evidence of Key Events Associated with the PPARa Mode of Action for
Hepatic Tumors3 in Male Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1:
PPARa
Activation
Key Event 2:
Altered Cell
Growth
Signaling
Key Event 3a:
Increased
Hepatic Cell
Proliferation
Key Event 3b:
Inhibition of
Apoptosis
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic
Tumors
Dose
(mg/kg/day)
PPARa
Activationb
Altered Cell
Growth
Signaling
Hepatic Cell
Proliferation
Apoptosis
Preneoplastic
Clonal
Expansion
Hepatic
Tumors0
1
NR
NR
- (7 d)
NR
NR
NR
3-335
-------
APRIL 2024
anonical
MOA
Key Event 1:
PPARa
Activation
Key Event 2:
Altered Cell
Growth
Signaling
Key Event 3a:
Increased
Hepatic Cell
Proliferation
Key Event 3b:
Inhibition of
Apoptosis
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic
Tumors
l.ld
t (16, 104 wk)
NR
t (16, 104 wk)
NR
NR
- (104 wk)
1.3
NR
NR
-(104 wk)
NR
NR
- (104 wk)
2.2d
t (16, 104 wk)
NR
t (16, 104 wk)
NR
NR
t (104 wk)
3
NR
NR
-(7 d)
NR
NR
NR
4.6d
t (16, 104 wk)
NR
t (16, 104 wk)
NR
NR
t (104 wk)
5.4
NR
- (4 wk)
NR
- (4 wk)
NR
NR
10
NR
NR
T (7 d)
NR
NR
NR
10.8
NR
- (4 wk)
NR
t (4 wk)
NR
NR
13.6
t (4, 12, 26, 39,
52, 65, 78,
91 wk)e
NR
-(4, 12, 26, 39,
52, 65, 78,
91 wk)e
NR
NR
t (104 wk)
14.2
NR
NR
-(104 wk)
NR
NR
- (104 wk)
19
t (1.7, 28 d)
NR
t (1. 7, 28 d)
NR
NR
NR
20
-(1, 3, 5 d)
NR
NR
NR
NR
NR
21.6
NR
- (4 wk)
NR
t (4 wk)
NR
NR
23
t (1.7, 28 d)
NR
t (1. 7, 28 d)
-(1, 7, 28 d)
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; MOA = mode of
action; PPARa = peroxisome proliferator-activated receptor a; NR = not reported; d = day(s); wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: Biegeletal. {, 2001, 673581}; NTP {, 2020, 7330145}; Elcombe et al. {,2010,
2850034}; Minata et al. {,2010,1937251} (wild-type); Wolf et al. {,2008, 1290827} (sex of mice not stated); Martin et al. {,
2007, 758419}; and Butenhoff et al. {, 2012, 2919192}.
aReviewed in Klaunig et al. {, 2003, 5772415}; Corton et al. {, 2014, 2215399}; and Corton et al. {, 2018, 4862049}.
b Indirect measurement of PPARa induction provided as CYP4A1 protein expression and hepatic P-oxidation activity {Elcombe,
2010, 2850034}, as hepatic acyl-CoA oxidase activity in NTP {, 2020, 7330145}, as hepatic P-oxidation activity inBiegel et al.
{, 2001, 673581}, as Cyp4al4, Cyp7al, Cyp7bl, Cyp8bl, and Cypl7al gene expression in Martin et al. {, 2007, 758419}.
c Hepatic tumors reflect increased incidence of adenoma in Biegel {, 2001, 673581}, and carcinoma and/or adenoma in NTP {,
2020,7330145} and Butenhoff et al. {,2012,2919192}.
dNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
e Biegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9,12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
Table 3-30. Evidence of Key Events Associated with the PPARa Mode of Action for
Hepatic Tumors3 in Female Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1- Kcv EvCnt 2:
PPARa Altered Cell
. Growth
Activation ..
Signaling
Key Event 3a:
Increased Hepatic
Cell Proliferation
Key Event 3b:
Inhibition of
Apoptosis
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic
Tumors
Dose
(mg/kg/day)
PPARa Altered Cell
Activationb Growth
Signaling
Hepatic Cell
Proliferation0
Apoptosisd
Preneoplastic
Clonal
Expansion
Hepatic
Tumorse
1 NR NR | (P,;,GD 1.5-17.5)f t (P,;,GD 1.5-17.5)f NR NR
3-336
-------
APRIL 2024
Canonical
MOA
Key Event 1:
PPARa
Activation
Key Event 2:
Altered Cell
Growth
Signaling
Key Event 3a:
Increased Hepatic
Cell Proliferation
Key Event 3b:
Inhibition of
Apoptosis
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic
Tumors
- (Po GD 1.5-11.5)
- (P0 GD 1.5-11.5)
1.6
NR
NR
- (104 wk)
NR
NR
-(104 wk)
5
NR
NR
t (Po GD 1.5-11.5)f
| (Po GD 1.5-17.5)
t (Po GD 1.5-11.5,
Po GD 1.5—17.5)f
NR
NR
16.1
NR
NR
- (104 wk)
NR
NR
-(104 wk)
18.2 g
t (16 wk)
NR
- (104 wk)
NR
NR
-(104 wk)
63.4 g
t (16 wk)
NR
- (104 wk)
NR
NR
-(104 wk)
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls unless otherwise noted; MOA = mode of action; PPARa = peroxisome
proliferator-activated receptor a; NR = not reported; Po = parental generation; GD = gestational day; wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: NTP {, 2020, 7330145}; Blake et al. {, 2020, 6305864} (dams); and Butenhoff et al. {,
2012,2919192}.
aReviewed in Klaunig et al. {, 2003, 5772415}; Corton et al. {, 2014, 2215399}; and Corton et al. {, 2018, 4862049}.
b Indirect measurement of PPARa induction provided as hepatic acyl-CoA oxidase activity in NTP {, 2020, 7330145}.
c Increased hepatic cell proliferation as provided by number of increased mitoses in NTP {, 2020, 7330145}.
d Apoptosis as both apoptosis and single-cell necrosis in Blake et al. {, 2020, 6305864}.
e Hepatic tumors reflect increased incidence of carcinoma and/or adenoma in NTP {, 2020, 7330145} and Butenhoff et al. {,
2012,2919192}.
fNo statistics were reported for hepatic cell proliferation or for apoptosis in Blake et al. {, 2020, 6305864}; thus, the arrows
indicate direction of increased incidence relative to the control group per the authors' results narrative.
gNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (300 and 1,000 ppm in female rats, or 18.2 and 63.4 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
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}. Nonneoplastic
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, many of which are also altered by PPARa
activation, in rats following PFOA exposure, and Wen et al. {, 2019, 5080582} observed
increased CAR activation in PFOA-exposed PPARa knockout mice compared with PFOA-
exposed wild-type mice. Other studies have shown altered gene expression of transcriptional
targets of CAR in both wild-type and PPARa knockout mice exposed to PFOA {Rosen, 2008,
1290828; Rosen, 2008, 1290832; Rosen, 2017, 3859803}. As with PPARa-mediated
3-337
-------
APRIL 2024
turn oogenesis, there are claims that CAR-mediated tumorigenesis in animals is not relevant to
human risk assessment due to differences in CAR-mediated alterations between species. For
example, 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 knockout mice compared with 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}.
There is also evidence that PFOA can suppress hepatocyte nuclear factor alpha (HNF4a) protein,
a master regulator of hepatic differentiation. Beggs et al. {, 2016, 3981474} observed a decrease
in HNF4a 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. Beggs et al. {, 2016, 3981474} also exposed human primary hepatocytes to 0.01-10
|iM PFOA for 48 or 96 hours to determine pathways affected by PFOA exposure; after 96 hours
of 10 |iM PFOA, HNF4a protein expression was significantly decreased. In primary human
hepatocytes exposed to 1, 25, or 100 |iM PFOA for 24 hours, the number of differentially
regulated genes was measured using a human genome gene chip; these microarray data
demonstrated that PFOA exposure at 25 and 100 |iM inhibited HNF4a function, as evidenced by
changes in gene targets of HNF4a using upstream regulator analysis {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 10 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. Evidence for the key events involved in
the CAR activation MOA for hepatic tumors in male and female rodents exposed to PFOA is
summarized in Table 3-31 and Table 3-32.
3-338
-------
APRIL 2024
Table 3-31. Evidence of Key Events Associated with the CAR Mode of Action for Hepatic
Tumors3 in Male Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1:
CAR Activation
Key Event 2:
Altered Gene
Expression
Key Event 3:
Increased Hepatic
Cell Proliferation
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic
Tumors
Dose
(mg/kg/day)
CAR Activationb
Altered Gene
Expression0
Hepatic Cell
Proliferation
Preneoplastic
Clonal
Expansion
Hepatic
Tumorsd
1
-(7 d)
T (7 d)
NR
NR
NR
l.le
NR
NR
t (16, 104 wk)
NR
- (104 wk)
1.3
NR
NR
- (104 wk)
NR
- (104 wk)
2.2e
NR
NR
t (16, 104 wk)
NR
t (104 wk)
3
t(7 d)
T (7 d)
NR
NR
NR
4.6e
NR
NR
t (16, 104 wk)
NR
t (104 wk)
5.4
NR
- (4 wk)
NR
NR
NR
10
t(7 d)
T (7 d)
NR
NR
NR
10.8
NR
- (4 wk)
NR
NR
NR
13.6
NR
NR
- (4, 12, 26, 39, 52,
65, 78, 91 wk)f
NR
t (104 wk)
14.2
NR
NR
- (104 wk)
NR
- (104 wk)
19
t (1.7, 28 d)
NR
t (1.7, 28 d)
NR
NR
20
-(1, 3, 5d)
NR
NR
NR
NR
21.6
NR
- (4 wk)
NR
NR
NR
23
t (1.7, 28 d)
NR
t (1.7, 28 d)
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; MOA = mode of
action; CAR = constitutive androstane receptor; d = day(s); NR = not reported; wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: Biegel et al. {, 2001, 673581}; NTP {, 2020, 7330145}; Elcombe et al. {, 2010,
2850034}; Martin et al. {, 2007, 758419}; Minata et al. {, 2010, 1937251}; Wen et al. {, 2019, 5080582} (wild-type); Rosen et
al. {,2008,1290828}; Rosen etal. {,2008, 1290832}; Rosen et al. {, 2017, 3859803}; and Butenhoffet al. {,2012,2919192}.
aReviewed in Felter,etal. {,2018, 9642149}.
bDirect and indirect measurement of CAR induction provided CAR gene expression in Wen et al. {, 2019, 5080582}, as Cyp3al,
Cyp3a3, and Cyp3a9 gene expression in Martin et al. {, 2007, 758419}, as Cyp2bl/2, Cyp3al, and Cyp4cil gene expression in
Elcombe et al. {, 2010, 2850034}, and as CAR gene biomarker set expression in Rosen et al. {, 2017, 3859803}.
c Gene expression as measured by differential expression of CAR target genes by microarray analysis {Rosen, 2017, 3859803} or
RT-PCR {Rosen, 2008, 1290832; Wen, 2019, 5080582},
dHepatic tumors reflect increased incidence of adenoma {Biegel, 2001, 673581}, and carcinoma and/or adenoma in NTP {, 2020,
7330145} andButenhoffetal. {,2012,2919192},
eNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, and 80 ppm in male rats, or 1.1, 2.2, and 4.6 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
f Biegel et al. {, 2001, 673581} included timepoints at 1, 3, 6, 9,12, 15, 18, and 21 months, which are represented in the table as
4, 12, 26, 39, 52, 65, 78, and 91 weeks, respectively.
3-339
-------
APRIL 2024
Table 3-32. Evidence of Key Events Associated with the CAR Mode of Action for Hepatic
Tumors3 in Female Rats and Mice Exposed to PFOA
Canonical MOA
Key Event 1:
CAR
Activation
Key Event 2:
Altered Gene
Expression
Key Event 3:
Increased Hepatic
Cell Proliferation
Key Event 4:
Preneoplastic
Clonal
Expansion
Outcome:
Hepatic Tumors
Dose (mg/kg/day)
CAR
Activation
Altered Gene
Expression
Hepatic Cell
Proliferationb
Preneoplastic
Clonal
Expansion
Hepatic Tumors
1
1.6
NR
NR
NR
NR
| (Po GD 1.5-17.5)°
- (P0 GD 1.5-11.5)
-(104 wk)
NR
NR
NR
-(104 wk)
5
NR
NR
t (Po GD 1.5-11.5)°
| (Po GD 1.5-17.5)°
NR
NR
16.1
18.2d
NR
NR
NR
NR
-(104 wk)
-(104 wk)
NR
NR
-(104 wk)
-(104 wk)
63.4d
NR
NR
-(104 wk)
NR
-(104 wk)
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls unless otherwise noted; MOA = mode of action; CAR = constitutive
androstane receptor; NR = not reported; Po=parental generation; GD = gestational day; wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: NTP {, 2020, 7330145}; Blake et al. {, 2020, 6305864} (dams); and Butenhoff et al. {,
2012,2919192}.
aReviewed in Felter, et al. {, 2018, 9642149}.
b Proliferation as provided by number of increased mitoses in Blake et al. {, 2020, 6305864}, and liver cell proliferation or
hyperplasia (no change) in NTP {, 2020, 7330145}.
cNo statistics were reported for hepatic cell proliferation for Blake et al. {, 2020, 6305864}; thus, the arrows indicate direction of
increased incidence relative to the control group per the authors' results narrative.
dNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (300 and 1,000 ppm in female rats, or 18.2 and 63.4 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects were
very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and study
results are in Section 3.5.2.
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 (Section
3.5.3.1), there is little experimental evidence that PFOA can induce DNA damage, supporting the
first key event of the cytotoxicity MOA. Quantitative liver histopathology is available in two
studies {NTP, 2020, 7330145; Butenhoff, 2012, 2919192}. Significantly increased single-cell
(hepatocyte) death and necrosis in male and female was reported in Sprague-Dawley rats, with a
3-340
-------
APRIL 2024
significant dose-response trend. Evidence for the key events involved in the cytotoxicity MOA
for hepatic tumors in male and female rodents exposed to PFOA is summarized in Table 3-33
and Table 3-34.
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 in vitro. Buhrke et
al. {, 2013, 2325346} determined that PFOA exposures of 10 [xM and 25 [xM 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 [xM). The higher PFOA concentration of 100 [xM was associated
with increased expression of p53, pl6, and p21 regulators (a nonsignificant increase was
observed at 25 (xM).
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 (xM), 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 [xM, and decreased at 100
and 200 [xM. 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 (xM) 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 [xM. Lipid metabolism and transport genes were also altered in the study:
increased expression of lipid anabolism gene ACSIJ, 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. In addition to the in
vitro evidence for the key events in the cytotoxicity MOA for hepatic tumors, data from rodent
studies are also available for PFOA. Histopathological and flow cytometric analyses are
available for rodent studies, demonstrating hepatocyte cell death {NTP, 2019, 5400977; NTP,
2020, 7330145; Crebelli, 2019, 5381564; Cope, 2021, 10176465}, increased proliferation in the
presence of cell death {NTP, 2020, 7330145; Loveless, 2008, 988599}, and hyperplasia {NTP,
2019, 5400977; NTP, 2020, 7330145}. Data are also available for increased serum enzymes
related to hepatotoxicity in rodents exposed to PFOA {NTP, 2019, 5400977; NTP, 2020,
7330145; Elcombe, 2010, 2850034; Minata, 2010, 1937251; Yan, 2014, 2850901; Loveless,
2008, 988599; Guo, 2019, 5080372; Butenhoff, 2012, 2919192; Cope, 2021, 10176465}.
Evidence for the key events involved in the cytotoxicity MOA for hepatic tumors in male and
female rodents exposed to PFOA is summarized in Table 3-33 and Table 3-34.
3-341
-------
APRIL 2024
Table 3-33. Evidence of Key Events Associated with the Cytotoxicity Mode of Action for
Hepatic Tumors3 in Male Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1:
Cytotoxicity
Key Event 2:
Increased Serum
Enzymes
Key Event 3:
Regenerative
Proliferation
Key Event 4:
Hyperplasia and/or
Preneoplastic Lesions
Outcome:
Hepatic
Tumors
Dose
(mg/kg/day)
Cytotoxicity1"
Serum Enzymes0
Regenerative
Proliferationd
Hyperplasia and/or
Preneoplastic Lesionse
Hepatic
Tumors'
0.08
NR
- (4 wk)
NR
NR
NR
0.10
0.30
-(Fi GD 1.5-
17.5)
- (5 wk)
- (29 d)g
-(Fi GD 1.5-
17.5)
- (5 wk)
NR
NR
- (29 d)g
NR
- (29 d)g
NR
NR
0.31
NR
- (4 wk)
NR
NR
NR
0.40
NR
- (4 wk)
NR
NR
NR
0.625
NR
t (4 wk)
NR
NR
NR
1.0
t (29 d)g
-(Fi GD 1.5-
17.5)
- (5 wk)
-(Fi GD 1.5-
17.5)
- (5 wk)
- (29 d)g
t (29 d)g
NR
l.lh
t (16 wk)
- (104 wk)
t (16 wk)
NR
4 (104 wk)
- (104 wk)
1.25
NR
t (4 wk)
NR
NR
NR
1.3
2.0
- (104 wk)
NR
t (12, 24, 52,
78 wk)
- (104 wk)
t (4 wk)
NR
NR
- (104 wk)
NR
- (104 wk)
NR
2.2h
t (16, 104 wk)
- (16 wk)
NR
4 (104 wk)
t (104 wk)
2.5
NR
t (4 wk)
NR
NR
NR
4.6h
t (16, 104 wk)
- (16 wk)
NR
| (104 wk)
t (104 wk)
5.0
t (5 wk)
t (4 wk)
t (5 wk)
NR
NR
NR
5.4
NR
t (4 wk)
NR
NR
NR
10
t (29 d)g
t (4 wk)
t (29 d)g
t (29 d)g
NR
10.8
NR
t (4 wk)
NR
NR
NR
14.2
- (104 wk)
t (12, 24, 52, 78,
104 wk)
NR
- (104 wk)
- (104 wk)
15.6h
t (16 wk)
t (16 wk)
NR
NR
NR
19
NR
-(1,7, 28 d)
td.7d)
t (28 d)
- (1- 7 d)
NR
20
NR
t (4 wk)
NR
NR
NR
21.6
NR
t (4 wk)
NR
NR
NR
23
NR
NR
t (1.7, 28 d)
t (28 d)
- (1- 7 d)
NR
30
t (29 d)g
NR
t (29 d)g
t (29 d)g
NR
31.7h
t (16 wk)
t (16 wk)
NR
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; J. = statistically
significant decrease in response compared with controls unless otherwise noted; MOA = mode of action; NR = not reported;
wk = week(s); Fi =first generation of offspring; GD = gestational day; d = day(s).
3-342
-------
APRIL 2024
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: NTP {, 2019, 5400977}; NTP {, 2020, 7330145}; Elcombe et al. {, 2010, 2850034};
Minata et al. {, 2010,1937251} (wild-type); Yan et al. {, 2014, 2850901}; Loveless et al. {, 2008, 988599}; Crebelli et al. {,
2019, 5381564}; Guo et al. {, 2019, 5080372}; Butenlioff et al. {, 2012, 2919192}; and Cope et al. {, 2021, 10176465} (low-fat
diet only; Fi pups exposed from GD 1.5 to 17.5, and evaluated at postnatal day (PND) 126).
aReviewed in Felter et al. {, 2018, 9642149}.
b Cytotoxicity provided as increased incidence of late apoptosis/necrosis in Crebelli et al. {, 2019, 5381564}, necrosis in
Butenlioff et al. {, 2012, 2919192}, and as necrosis and/or single-cell necrosis in NTP {, 2020, 7330145} and Cope et al. {,
2021, 10176465}.
c Serum enzyme changes provided as changes in alkaline phosphatase (ALP), alanine transaminase (ALT), and/or aspartate
transaminase (AST) in Butenlioff et al. {, 2012, 2919192}, NTP {, 2020, 7330145}, NTP {, 2019, 5400977}, and Cope et al. {,
2021, 10176465}, and as changes inALT and/or AST in Elcombe et al. {, 2010, 2850034}, Minata et al. {,2010,1937251},
Guo et al. {, 2019, 5080372}, and Yan et al. {, 2014, 2850901},
dRegenerative proliferation provided as increased hepatic S-phase labeling indices (%) and/or increased number of hepatocytes
in Elcombe et al. {, 2010, 2850034} and as liver proliferation in NTP {, 2020, 7330145}.
hyperplasia and/or preneoplastic lesions provided as hepatocellular hyperplasia (qualitative results) in Elcombe et al. {, 2010,
2850034}; as bile duct hyperplasia in NTP {, 2020, 7330145}; as hyperplastic nodules in Butenlioff et al. {, 2012,2919192};
and as bile duct hyperplasia in rats and mice in Loveless et al. {, 2008, 988599}.
fHepatic tumors reflect increased incidence of carcinoma and/or adenoma in NTP {, 2020, 7330145} and Butenlioff et al. {,
2012,2919192},
gNo statistics were reported for histopatliology results for Loveless et al. {, 2008, 988599}; thus, the arrows indicate direction of
increased incidence of individual cell necrosis for Key Event (KE)1, mitotic figures for KE3, and bile duct hyperplasia for KE4
relative to the control group.
hNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (20, 40, 80, 150, and 300 ppm in male rats, or 1.1, 2.2., 4.6, 15.6, and 31.7 mg/kg/day) in order
to provide a representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment
effects were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section
3.4.4.2.1.2 and study results are in Section 3.5.2.
Table 3-34. Evidence of Key Events Associated with the Cytotoxicity Mode of Action for
Hepatic Tumors3 in Female Rats and Mice Exposed to PFOA
Canonical
MOA
Key Event 1:
Cytotoxicity
Key Event 2:
Increased Serum
Enzymes
Key Event 3:
Regenerative
Proliferation
Key Event 4:
Hyperplasia and/or
Preneoplastic Lesions
Outcome:
Hepatic
Tumors
Dose
(mg/kg/day)
Cytotoxicity1"
Serum Enzymes0
Regenerative
Proliferationd
Hyperplasia and/or
Preneoplastic Lesionse
Hepatic
Tumors'
0.1
-(Fi GD 1.5-17.5)
-(Fi GD 1.5-17.5)
NR
NR
NR
1.0
-(Fi GD 1.5-17.5,
PoGD 1.5-11.5,
PoGD 1.5-17.5)
- (Fi GD 1.5-17.5,
PoGD 1.5-11.5,
PoGD 1.5-17.5)
NR
NR
NR
1.6
- (104 wk)
4 (78 wk)
-(12, 24, 52,
104 wk)
NR
-(104 wk)
-(104 wk)
5.0
-(PoGD 1.5-11.5,
PoGD 1.5-17.5)
t (PoGD 1.5-17.5)
-(PoGD 1.5-11.5)
NR
NR
NR
6.25
NR
t (4 wk)
NR
NR
NR
12.5
NR
t (4 wk)
NR
NR
NR
16.1
- (104 wk)
-(12, 24, 52, 78,
104 wk)
NR
-(104 wk)
-(104 wk)
18.2 g
- (16 wk, 104 wk)
- (16 wk)
-(104 wk)
-(16, 104 wk)
-(104 wk)
25
NR
t (4 wk)
NR
NR
NR
50
NR
t (4 wk)
NR
NR
NR
63.4 g
t (104 wk)
t (16 wk)
-(104 wk)
-(104 wk)
-(104 wk)
3-343
-------
APRIL 2024
Canonical
MOA
Key Event 1:
Cytotoxicity
Key Event 2:
Increased Serum
Enzymes
Key Event 3:
Regenerative
Proliferation
Key Event 4:
Hyperplasia and/or
Preneoplastic Lesions
Outcome:
Hepatic
Tumors
- (16 wk)
- (16 wk)
100
NR
t (4 wk)
NR
NR
NR
Notes: f = statistically significant increase in response compared with controls; - = no significant response; MOA = mode of
action; Fi = first generation of offspring; GD = gestational day; NR = not reported; Po=parental generation; wk = week(s).
Cells in bolded text and blue shading indicate that the response direction is concordant with the key event in the published MOA.
Cells with NR (not reported) indicate that no data were measured for that particular key event at that dose in the studies
reviewed.
Data represented in table extracted from: NTP {, 2019, 5400977}; NTP {, 2020, 7330145}; Butenhoff et al. {, 2012, 2919192};
Blake et al. {, 2020, 6305864} (dams); and Cope et al. {, 2021, 10176465} (low-fat diet only; Fi pups exposed from GD 1.5 to
17.5 and evaluated at postnatal day (PND) 126).
aReviewed in Felter et al. {, 2018, 9642149}.
b Cytotoxicity provided as increased incidence of hepatic necrosis in Butenhoff et al. {, 2012, 2919192}, focal necrosis in Blake
et al. {,2020, 6305864}, and as single-cell necrosis in NTP {,2020,7330145} and Cope et al. {,2021, 10176465}.
c Serum enzyme changes provided as changes in alkaline phosphatase (ALP), alanine transaminase (ALT), and/or aspartate
transaminase (AST) in Butenhoff et al. {, 2012,2919192}, Blake et al. {, 2020, 6305864}, Cope et al. {, 2021,10176465}, NTP
{, 2020, 7330145}, and NTP {, 2019, 5400977}, For Butenhoff et al. {, 2012,2919192}, only ALP was significantly decreased
at 18 months (78 weeks).
dRegenerative proliferation provided as liver proliferation in NTP {, 2020, 7330145}.
hyperplasia and/or preneoplastic lesions provided as bile duct hyperplasia NTP {, 2020, 7330145} and as hyperplastic nodules
in Butenhoff et al. {, 2012, 2919192 },
fHepatic tumors reflect increased incidence of carcinoma and/or adenoma in NTP {, 2020, 7330145} and Butenhoff et al. {,
2012,2919192},
gNTP {, 2020, 7330145} included perinatal (gestation and lactation) and postweaning exposures. This table reports only data
from the postweaning exposures (300 and 1,000 ppm in female rats, or 18.2 and 63.4 mg/kg/day) in order to provide a
representative set of the available mechanistic data involved in this MOA from bioassays, and because the treatment effects
were very similar in the perinatal and postweaning exposure groups. Further study design details are in Section 3.4.4.2.1.2 and
study results are in Section 3.5.2.
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 3-22). 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
nonsmokers (PFOA detected in 75% of the samples) compared with PF AS-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 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.
3-344
-------
APRIL 2024
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 3-22).
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 direct
interaction between PFOA and genetic material. The available in vivo evidence suggests that
exposure to PFOA at levels resulting in cytotoxicity (e.g., hepatotoxicity, bone marrow toxicity)
may lead to secondary genotoxicity in target tissues. Although unlikely, genotoxicity cannot be
ruled out as a potential key event for PFOA-inducted hepatic tumor formation.
3.5.4.2.4.5 Consideration of Other Plausible Modes of Action
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.
3.5.4.2.4.5.1 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 (J,g/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.
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
3-345
-------
APRIL 2024
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}.
3.5.4.2.4.5.2 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 to GD 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-mA\ (129S4/SvJae-PparatmlGonz/J) mice with PFOA
(<50 |iinol/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-null mice, with a significant increase at 50 |iinol/kg/day when compared with 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).
PFOA exposure caused increases in 8-OHdG, a biomarker of oxidative stress, in human
lymphoblast cells (TK6) and HepG2 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
3-346
-------
APRIL 2024
following cessation of PFOA exposure. Several other in vitro studies reported increases in ROS
in PFOA-exposed cells, including HepG2, nonhuman primate kidney, and human-hamster hybrid
(AL) cells {Panaretakis, 2001, 5081525; Wielsoe, 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 [xM PFOA for 24-hours, despite increased
cytotoxicity at 200 [xM 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 [xM 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 [xM 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.
Wielsoe 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.
3.5.4.2.4.6 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 tumorigenesis, 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}. For example, CalEPA concluded that PFOA "can induce
biological activity and hepatotoxicity that is independent of PPARa activation. This indicates
that the toxicity observed in rodent studies may not act entirely through the PPARa activation
pathway. As such, OEHHA cannot conclude that all hepatotoxic endpoints of PFOA and PFOS
in rodents are the result of PPARa activation" {CalEPA, 2021, 9416932}. Similarly, NJDWQI
agreed that "effects of PFOA clearly occur through both PPAR-alpha independent and PPAR-
alpha dependent processes" {NJDWQI, 2017, 5024840}. The existence of multiple MOAs in
addition to PPARa activation suggest that PFOA-induced liver cancer in rats may be more
3-347
-------
APRIL 2024
relevant to humans than previously thought. Additional research is warranted to better
characterize the MO As for PFOA-induced hepatic tumorigenesis.
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 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. The few available mechanistic studies focusing on PFOA-
induced renal toxicity highlight several potential underlying mechanisms of PFOA exposure-
induced renal tumorigenesis, including altered cell proliferation and apoptosis, epigenetic
alterations, and oxidative stress. However, due to data limitations, it is difficult to distinguish
which mechanism(s) are operative for PFOA-induced kidney cancer. Similarly for testicular
cancer, 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. Combined, the
epidemiological and animal toxicological literature indicate that the testes are a common site of
PFOA-induced tumorigenesis. Overall, the EPA concluded that the available mechanistic data
suggest that multiple MO As could play role in the renal, testicular, pancreatic, and hepatic
tumorigenesis associated with PFOA exposure in studies of human populations and animal
models. IARC {2016, 3982387; andZahm, 2023, 11347256}, CalEPA {2021, 9416932} and
NJDWQI {2017, 5024840} similarly concluded that there is evidence for many potential
mechanisms for PFOA-induced carcinogenicity. For example, IARC concluded there is strong
mechanistic evidence of carcinogenicity in exposed humans and that PFOA is
immunosuppressive, induces epigenetic alterations, induces oxidative stress, modulates receptor-
mediated effects (via (PPAR) a, constitutive androstane receptor/pregnane X receptor
[CAR/PXR], and PPARy), and alters cell proliferation, cell death, and nutrient and energy supply
{Zahm, 2023, 11347256}.
3.5.5 Cancer Classification
Under the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}, the 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 and carcinomas in rats.
3-348
-------
APRIL 2024
The Guidelines {U.S. EPA, 2005, 6324329} provide examples 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";
• "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}. The PFOA cancer database would benefit from additional
large high confidence cohort studies in independent populations.
The evidence of carcinogenicity in animals is based on three studies that used the same strain of
rat. Taken together, these results provide evidence of increased incidence of three different tumor
types (LCTs, PACTs, and hepatocellular tumors) in males administered diets contaminated with
PFOA. Additionally, pancreatic acinar cell adenocarcinomas are a rare tumor type {NTP, 2020,
7330145}, and their occurrence in PFOA-treated animals in this study increases the confidence
that this incidence is treatment-related since these tumors are unlikely to be observed in the
absence of a carcinogenic agent {U.S. EPA, 2005, 6324329}. The historical control incidence for
pancreatic acinar cell adenocarcinomas in the female rats is 0/340 and in the male rats is 2/340,
highlighting the rarity of this particular 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 {U.S. EPA,
2005, 6324329}. While site concordance was present between human studies of testicular cancer
and animal studies reporting increased incidence of LCTs, evidence of carcinogenicity of PFOA
from other cancer sites where concordance between humans and animals is not present is still
relevant to the carcinogenicity determination for PFOA. See Table 3-35 below for specific
rationale on how PFOA aligns with examples supporting the Likely to Be Carcinogenic to
3-349
-------
APRIL 2024
Humans cancer descriptor in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329}.
Table 3-35. 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" {U.S. EPA, 2005, 6324329}.
PFOA data are consistent with this description.
Epidemiological evidence supports a plausible
association between exposure and cancer, though there
are uncertainties regarding the MO As for tumor types
observed in humans. There is supporting experimental
evidence, including 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" {U.S. EPA, 2005,
6324329}.
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.
"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" {U.S. EPA,
2005, 6324329}.
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.
"A rare animal tumor response in a single experiment
that is assumed to be relevant to humans" {U.S. EPA,
2005, 6324329}.
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}.
"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}.
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.
Notes: DNA = deoxyribonucleic acid; MOA = mode of action.
EPA recognizes that other state and international health agencies have recently classified PFOA
as carcinogenic to humans {CalEPA, 2021, 9416932; IARC as reported in Zahm, 2023,
11347256}. As the SAB PFAS Review Panel {U.S. EPA, 2022, 10476098} noted, "the criteria
used by California EPA, for determination that a chemical is a carcinogen, are not identical to the
criteria in the U.S. EPA's Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329}" and, similarly, IARC's classification criteria are not identical to the EPA's guidelines
{IARC, 2019, 11320796}. Rationale for why PFOA does not meet the Carcinogenic to Humans
descriptor according to the EPA's Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005,
6324329} is detailed in Section 5.4.
3-350
-------
APRIL 2024
4 Dose-Response Assessment
Considerations in Selecting Studies and Endpoints for Dose-Response Analysis
There is evidence from both human epidemiological and animal toxicological studies that oral
perfluorooctanoic acid (PFOA) exposure can result in adverse health effects across a range of
health outcomes. In response to recommendations made by the SAB and the conclusions from
EPA's systematic review of the available health effects evidence, presented in the EPA's
preliminary analysis, the 2021 SAB review draft Proposed Approaches to the Derivation of a
Draft Maximum Contaminant Level Goal for Perfluorooctanoic Acid (PFOA) in Drinking Water
{U.S. EPA, 2021, 10428559}, EPA focused its final toxicity value derivation efforts herein "on
those health outcomes that have been concluded to have the strongest evidence" {U.S. EPA,
2022, 10476098}. Therefore, EPA prioritized health outcomes and endpoints with the strongest
overall weight of evidence, which were the outcomes with evidence demonstrates or evidence
indicates integration judgments, based on the synthesis of the available human, animal, and
mechanistic evidence (Section 3.4 and 3.5) for points of departure (POD) derivation using the
systematic review methods described in Section 2 and Appendix A {U.S. EPA, 2024,
11414343}. For PFOA, the health outcomes with the strongest weight of evidence are cancer
(described in Section 4.2) and the noncancer health outcomes of immunological, developmental,
cardiovascular (serum lipids), and hepatic effects (described in Section 4.1). For all other health
outcomes (e.g., reproductive, endocrine, nervous, hematological, musculoskeletal), the evidence
integration summary judgment for the human epidemiological and animal toxicological evidence
was suggestive or inadequate and these outcomes were not assessed quantitatively. Health
outcomes for which the results were suggestive are discussed in the evidence profile tables
provided in Appendix C {U.S. EPA, 2024, 11414343}, as well as Section 5.5.
In the previous section describing the hazard judgment decisions (Section 3.4 and 3.5), EPA
qualitatively considered high, medium, and sometimes low confidence studies of PFOA exposure
to characterize the weight of evidence for each health outcome. For the quantitative analyses
described in the following subsections, EPA focused exclusively on high or medium confidence
human epidemiological and animal toxicological studies for POD derivation, as recommended in
Chapter 7.2 of the IRIS Handbook {U.S. EPA, 2022, 10367891}. While the IRIS Handbook also
includes consideration of low confidence studies for dose-response analysis under certain
circumstances, this EPA assessment did not consider low confidence studies because of the
relatively large and robust database for PFOA. At this stage, EPA considered additional study
attributes to enable extrapolation to relevant exposure levels in humans. These attributes are
described in Table 7-2 of the IRIS Handbook and include relevance of the test species, relevance
of the studied exposure to human environmental exposures, quality of measurements of exposure
and outcomes, and other aspects of study design including specific reconsideration of the
potential for bias in the reported association between exposure and outcomes {U.S. EPA, 2022,
10367891}.
Consideration of these attributes facilitates the transparent selection of studies and data for dose-
response modeling and potential RfD or CSF derivation. Studies exhibiting these attributes are
expected to provide more accurate human equivalent toxicity values and are therefore preferred
4-1
-------
APRIL 2024
in the selection process. Consideration of these attributes in the study selection process are
described below for noncancer and cancer endpoints.
4.1 Noncancer
4.1.1 Study and End point Selection
For study and endpoint selection for noncancer health outcomes, the human studies that provided
all necessary analytical information (e.g., exposure distribution or variance, dose-response data,
etc.) for POD derivation, analyzed the outcome of interest in the general population or
susceptible population, and demonstrated a larger number of the study attributes outlined above
were preferred. If available, high and medium confidence studies with exposures levels within or
near the range of typical environmental human exposures, especially exposure levels comparable
to human exposure levels in the general United States population, were preferred over studies
reporting considerably higher exposure levels (e.g., occupational exposure levels). Exposure
levels near the typical range of environmental human exposure can facilitate extrapolation to the
lower dose range of exposure levels that are relevant to the overall population. When available
for a given health outcome, studies with analyses that addressed potential confounding factors
affecting exposure concentrations (e.g., addressing temporal variations of PFOA concentrations
during pregnancy due to hemodynamics) were also preferred. Additionally, when studies
presented overlapping data on the same cohort or study population, various factors were
considered to facilitate selection of one study for POD derivation. These factors included the
duration of exposure, the length of observation of the study cohort, and the comprehensiveness
of the analysis of the cohort in order to capture the most relevant results for dose-response
analysis.
The preferred animal toxicological studies consisted of medium and high confidence studies with
exposure durations appropriate for the endpoint of interest (e.g., chronic or subchronic studies vs.
short-term studies for chronic effects) or with exposure during sensitive windows of
development and with exposure levels near the lower dose range of doses tested across the
evidence base. 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 5.3 for a discussion of animal toxicological studies and endpoints
selected for POD derivation for this updated assessment compared with those selected for the
2016 PFOA HESD {U.S. EPA, 2016, 3603279}.
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 3-4, the majority of
epidemiological studies assessed endpoints related to serum biomarkers of hepatic injury (18
medium confidence studies), while fewer studies reported on liver disease or injury (5 medium
confidence studies) and other serum markers of liver function (4 medium confidence studies).
EPA prioritized studies that evaluated endpoints related to serum biomarkers of injury for
quantitative analyses because the reported effects on these endpoints were well-represented
within the database and were generally consistent across the available medium confidence
studies. Additionally, serum levels of alanine aminotransferase (ALT) and aspartate
aminotransferase (AST) are considered reliable markers of hepatocellular function/injury, with
4-2
-------
APRIL 2024
ALT considered more specific and sensitive {Boone, 2005, 782862}. Specifically, all five
medium confidence studies of general population adults from the updated literature searches
reported positive associations between PFOA serum concentrations and ALT, three of which
reported statistically significant responses {Nian, 2019, 5080307; Darrow, 2016, 3749173;
Salihovic, 2018, 5083555; Gleason, 2015, 2966740; Jain, 2019, 5381541}. These more recently
published studies provided additional evidence for increased ALT in adults beyond the three
medium confidence studies reporting positive associations for ALT from the 2016 PFOA HESD
{Lin, 2010, 1291111; Yamaguchi, 2013, 2850970; Gallo, 2012, 1276142}. Findings from studies
of other liver enzymes, AST and GGT, in adults generally reported a positive association, though
less consistently than studies of ALT; therefore, studies of AST and GGT are supportive of the
selection of ALT as an endpoint for POD derivation because these results demonstrate coherence
across the different liver serum enzyme endpoints.
As mentioned above, 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 when extrapolated to the overall population {Gilbert, 2006, 174259}. For
example, a Scandinavian study in people without any symptoms of liver disease but with
relatively small increased serum ALT levels were later diagnosed 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}.
Numerous studies have also 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}. Taken together, results of human epidemiological and animal
toxicological studies and the positions of the AASLD and the ACG demonstrate the clinical
significance of increased serum ALT. It is also important to note that while evaluation of direct
liver damage is possible in animal toxicological studies, it is difficult to obtain biopsy-confirmed
histological data in humans. Therefore, liver injury in humans is typically assessed using serum
biomarkers of hepatotoxicity {Costello, 2022, 10285082}.
4-3
-------
APRIL 2024
Among the available medium confidence epidemiological studies reporting alterations in serum
ALT in humans, studies of adults in the general population were prioritized over studies in other
populations (e.g., occupational) or life stages (e.g. children), as the adult study findings provided
the most consistent evidence of increases in ALT (see Section 3.4.1.1). Several of these medium
confidence studies reporting increases in ALT in adults were excluded from POD derivation for
reasons such as combined adolescent and adult populations {Gleason, 2015, 2966740},
populations consisting of only elderly adults {Salihovic, 2018, 5083555}, use of correlation
analyses only {Yamaguchi, 2013, 2850970}, and reporting analyses stratified by glomerular
filtration status without stratifying by exposure level, which were not amenable to modeling
{Jain, 2019, 5381541}.
Exclusions of these studies resulted in the consideration of four medium confidence studies for
POD derivation {Gallo, 2012, 1276142; Darrow, 2016, 3749173; Lin, 2010, 1291111; Nian,
2019, 5080307} (Table 4-1). These studies exhibited many of the study attributes outlined in
Section 4 above and in Appendix A {U.S. EPA, 2024, 11414343}. For example, the two largest
studies of PFOA and ALT are Gallo et al. {, 2012, 1276142} and Darrow et al. {, 2016,
3749173}, both conducted in over 30,000 individuals from the general population, aged 18-years
and older, as part of the C8 Health Project in the United States. Further, Gallo et al. {,2012,
1276142} demonstrated a statistically significant trend of increased ALT across deciles of PFOA
exposure and Darrow et al. {, 2016, 3749173} provided an exposure-response gradient for
PFOA. Two additional studies {Lin, 2010, 1291111; Nian, 2019, 5080307} were considered for
POD derivation because they reported associations in general populations in the United States
and a Chinese population located near a PFAS manufacturing facility, respectively. Nian et al. {,
2019, 5080307} examined a large population of adults (1,605) in Shenyang (one of the largest
fluoropolymer manufacturing centers in China) as part of the Isomers of C8 Health Project and
reported significantly increased level of ALT associated with PFOA. Lin et al. {, 2010,
1291111} was also considered for POD derivation since it is a large (2,216 men and 1,063
women) nationally representative study in an NHANES adult population and observed increased
ALT levels per log-unit increase in PFOA and these associations remained after accounting for
other PFAS in the regression models. However, several methodological limitations precluded its
use for POD derivation. Limitations include lack of clarity about the base of logarithmic
transformation applied to PFOA concentrations in regression models, and the choice to model
ALT as an untransformed variable, which is a departure from the lognormality assumed in most
of the ALT literature. Therefore, three medium confidence epidemiological studies were
prioritized for POD derivation {Gallo, 2012, 1276142; Darrow, 2016, 3749173; Nian, 2019,
5080307} (Table 4-1).
Liver toxicity results reported in animal toxicological studies after PFOA exposure are
concordant with the observed increased ALT indicative of hepatic damage observed 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 increased ALT observed in human
populations. Mechanistic studies in mammals and evidence from in vitro studies and
4-4
-------
APRIL 2024
nonmammalian animal models provide additional support for the biological plausibility and
human relevance of the PFOA-induced hepatic effects observed in animals. These studies
suggest multiple potential MO As for the observed liver toxicity, including PPARa-dependent
and -independent MOAs. The observed increases in liver enzymes (e.g., ALT) in rodents are
supportive of the hepatic damage confirmed during histopathological examinations in several
studies. Taken together, the study results suggest that at least some mechanisms for PFOA-
induced hepatic effects are relevant to humans.
For animal toxicological hepatic endpoints, EPA preferred studies reporting quantitative
biologically or statistically significant measures of severe toxicity (i.e., histopathological lesions
related to cell or tissue death or necrosis) with study designs best suited for quantitative analysis
(e.g., large sample size, reported effects in the lower dose range). Of the seven studies that
quantitatively reported incidences of hepatic cell or tissue death or necrosis, five were excluded;
two studies were excluded because they did not report statistically or biologically significant
responses {Perkins, 2004, 1291118; Butenhoff, 2012, 2919192} and three were excluded
because they had relatively small sample sizes (i.e., n < 10) {NTP, 2019, 5400977; Blake, 2020,
6305864; Cope, 2021, 10176465}. After these exclusions, EPA identified two studies reporting
adverse liver effects in male rodents due to PFOA exposure, 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 (Table 4-1). 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 that presented a large sample size (n = 50), numerous and relatively low dose
levels, and assessment of a suite of hepatic 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} reported a number
of hepatotoxic effects, a low dose range, relatively large sample sizes (n = 19-20), and dose-
dependent increases in histopathological outcomes indicating adverse effects in male mice
gavaged with PFOA for 29 days. In addition, Loveless et al., (2008, 988599) was the only study
in mice to report quantitative histopathological examinations of liver tissue in non-pregnant
adults and had the longest exposure duration of the available mouse studies. Therefore, the
incidences of two endpoints, 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 3-9, the majority of
epidemiological studies assessed endpoints related to immunosuppression (1 high and 20
medium confidence studies) and immune hypersensitivity (1 high and 20 medium confidence
studies), while 2 medium confidence studies also reported on endpoints related to autoimmune
disease. Studies that reported on specific autoimmune diseases were excluded from POD
derivation because there were a limited number of studies that assessed the same diseases (e.g.,
rheumatoid arthritis, celiac disease). Studies that evaluated endpoints related to immune
hypersensitivity (e.g., asthma) were also not considered for POD derivation because there were
inconsistencies in the direction and precision of effects across gender or age subgroups in the
4-5
-------
APRIL 2024
available 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, reported differing results across medium and high
confidence studies and were therefore excluded from further consideration, though they provide
qualitative support of an association between PFOA exposure and altered immune function.
Evidence of immunosuppression in children associated with exposure to PFOA reported by
epidemiological studies was consistent across studies and endpoints. Specifically,
epidemiological studies reported associations between PFOA exposure and reduced humoral
immune response to routine childhood immunizations, including lower levels of tetanus and
diphtheria {Timmerman, 2021, 9416315; Abraham, 2020, 6506041; Grandjean, 2012, 1248827;
Budtz-j0rgensen, 2018, 5083631}, HiB {Abraham, 2020, 6506041}, and rubella {Granum,
2013, 1937228; Stein, 2016, 3108691; Zhang, 2023, 10699594} antibody titers. Reductions in
antibody response were observed at multiple timepoints during childhood (specifically ages
between 3-19 years in these studies), for either prenatal or postnatal childhood PFOA exposure
levels, and were consistent across studies in children populations from medium confidence
studies. Therefore, reduced antibody response in children was selected as an endpoint for POD
derivation.
Measurement of antigen-specific antibodies following vaccination(s) is a measure of the overall
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, which is a type of
immunity that develops when a person's immune system responds to a foreign substance or
microorganism, and it has been used as a comprehensive 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 specific challenge and is considered an adverse
immunological health outcome" {U.S. EPA, 2022, 10476098}. This is consistent with a review
article 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 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. {, 2016, 10293267;, 2017, 5926400;, 2019, 5080663} and in comments
from other subject matter experts on the SAB's PFAS review panel {U.S. EPA, 2022,
10476098}, the clinical manifestation of a disease after chemical exposure is not required 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 in response to chemical exposure 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}. Decreases in antibody response, even at
smaller magnitudes in individuals, are clinically relevant when extrapolated to the overall
4-6
-------
APRIL 2024
population {Gilbert, 2006, 174259}. This response also translates across multiple species,
including rodents, and extensive historical data indicate that suppression of antigen-specific
antibody responses by exogenous agents is predictive of immunotoxicity.
Studies of developmental exposure to environmental toxicants demonstrate an association with
immune suppression {Selgrade, 2007, 736210}. When immunosuppression occurs during
immune system development, the risks of developing infectious diseases and other
immunosuppression-linked diseases may increase {Dietert, 2010, 644213}. The immune system
continues developing postnatally; because of this, exposures to PFAS and other immunotoxic
agents during development may have serious, long-lasting, and irreversible health consequences
{DeWitt, 2019, 5080663; MacGillivray, 2014, 6749084; Selgrade, 2007, 736210}. Indeed,
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 that were
assessed. Developmental immunotoxicity as a result of chemical exposure is generally observed
at doses lower than required to elicit immunotoxicity in adults {vonderEmbse, 2018, 6741321}.
Therefore, developmental immunotoxicity is generally a highly sensitive health outcome, both
when considering other types of developmental toxicity and when comparing it to
immunotoxicity observed in exposed adults. Luster et al. {, 2005, 2174509} similarly noted that
the specific immunotoxic endpoint of 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.
One high and 10 medium confidence studies {Grandjean, 2012, 1248827; Granum, 2013,
1937228; Mogensen, 2015, 3981889; Grandjean, 2017, 3858518; Grandjean, 2017, 4239492;
Budtz-j0rgensen, 2018, 5083631; Timmerman, 2021, 9416315; Pilkerton, 2018, 5080265; Shih,
2021, 9959487; Stein, 2016, 3108691; Zhang, 2023, 10699594} reported findings on antibody
response to tetanus, diphtheria, or rubella in children or adolescents. Only one low confidence
study reported findings on antibody response to Hib {Abraham, 2020, 6506041}, which was
excluded from POD derivation because of the limited evidence and the low confidence rating.
Three studies {Pilkerton, 2018, 5080265; Stein, 2016, 3108691; Zhang, 2023, 10699594}
reported on antibody response to rubella in adolescents and one study reported on antibody
response in young children {Granum, 2013, 1937228}. From the adolescent studies, one study
observed decreased rubella antibody response in a specific subpopulation of only seropositive
adolescents {Stein, 2016, 3108691} and the other two studies did not report statistically
significant associations between PFOA and rubella antibody response in the overall population
{Zhang, 2023, 10699594; Pilkerton, 2018, 5080265}. Granum et al. {, 2013, 1937228} reported
a statistically significant association between PFOA exposure and rubella antibody response in
young children. Because studies reporting rubella antibody response were mixed (2/4
demonstrating associations), rubella studies were not further considered for POD derivation.
Overall, EPA prioritized studies reporting responses to tetanus and diphtheria because the
responses were consistently observed across a large number of studies {medium and low
confidence) in children from multiple populations for these two vaccine types.
Five separate studies {Grandjean, 2012, 1248827; Mogensen, 2015, 3981889; Grandjean, 2017,
3858518; Grandjean, 2017, 4239492; Shih, 2021, 9959487} reported on diphtheria and tetanus
antibody responses and PFOA exposure in the same cohort (i.e., same individuals) of Faroese
4-7
-------
APRIL 2024
children. One study reported on the same Faroese children cohort in a more recent medium
confidence publication {Budtz-j0rgensen, 2018, 5083631}. Because this most recent medium
confidence study is the only one of the five studies that provided dose-response data with
untransformed PFOA concentrations which are more amenable to BMD modeling, only results
from Budtz-j0rgensen and Grandjean {, 2018, 5083631} were prioritized for POD derivation
and the four other studies conducted in the Faroe Island population were excluded. One medium
confidence cross-sectional study {Timmerman, 2021, 9416315} reported on tetanus and
diphtheria antibody response and PFOA exposure in Greenlandic children. This study was also
prioritized for POD derivation. The results from the Faroe Island and Greenlandic populations
are qualitatively supported by a low confidence cross-sectional study of associations between
diphtheria and tetanus antibody responses and PFOA in German children {Abraham, 2020,
6506041}.
In total, two medium confidence epidemiologic studies that reported decreased antibody
responses in children exposed to PFOA {Budtz-j0rgensen, 2018, 5083631; Timmerman, 2021,
9416315} were considered for POD derivation (Table 4-1). These two epidemiological studies
report 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 PFOA exposure measured at age 5 or prenatally; Timmerman et al. {, 2021,
9416315} investigated anti-tetanus and anti-diphtheria responses and PFOA exposure in
Greenlandic children aged 7-12. Both studies examined antibody responses associated with
PFOA exposure in well-characterized cohorts, and in the case of Budtz-j0rgensen and Grandjean
{, 2018, 5083631}, multiple prior publications supported the finding of an inverse relationship
between PFOA exposure concentrations and antibody response in the same study cohorts.
Immunotoxicity results reported in animal toxicological studies in adult rodents are concordant
with the immunosuppression observed 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) (four medium confidence studies) and altered
immune cell populations (e.g., altered white blood cell counts, altered splenic and thymic
cellularity) (one high and four medium confidence studies). Immunosuppression evidenced by
functional assessments of the immune responses, such as analyses of globulin and
immunoglobulin levels after challenges, are comparable and thus, supportive of the
immunosuppression reported as decreased antibody responses seen in human populations and
were therefore prioritized for quantitative assessment. Additionally, EPA identified
immunosuppressive effects in multiple species and both sexes of animal toxicological studies,
further supporting the selection of these endpoints for dose-response analyses. Animal
toxicological studies assessing alterations in immune cell populations were not considered
further as there were a limited number of studies assessing specific endpoints of interest.
Although the other reported immune effects, such as altered organ weights and histopathology,
are consistent with evidence indicating alterations in immune function and response from animal
toxicological studies, they were not considered for POD derivation as these effects may be
confounded by changes in body weight, effects were not consistent across studies, and/or a
limited number of studies assessed specific outcomes. Of the four medium confidence studies
reporting impaired IgM response in mice, EPA selected Dewitt et al. {, 2008, 1290826}, a 15-
day drinking water exposure study in female mice, and Loveless et al. {, 2008, 988599}, a 29-
4-8
-------
APRIL 2024
day study in male mice, for POD derivation as these two studies presented data for a larger
number of dose groups spanning a broader dose range than either Dewitt et al. {, 2016, 2851016}
orDe Guise et al. {, 2021, 995974}.
4.1.1.3 Cardiovascular Effects
As reviewed in Section 3.4.3.4, evidence indicates that exposure to PFOA are associated with
cardiovascular effects in humans. As described in Table 3-12, the majority of epidemiological
studies assessed endpoints related to serum lipids (2 high, 27 medium, and 19 mixed18 confidence
studies) and blood pressure and hypertension (2 high and 18 medium confidence studies), while
some studies also reported on cardiovascular disease (1 high and 6 medium confidence studies)
and atherosclerosis (1 high and 3 medium confidence studies). Endpoints related to
cardiovascular disease and atherosclerosis were excluded from consideration for POD derivation
because there were limited high and medium confidence studies and they reported mixed (i.e.,
positive and inverse associations) or mostly null results. Endpoints related to blood pressure and
hypertension were also excluded from quantitative analyses because there was higher confidence
in analytically determined serum lipid levels compared with blood pressure measurements and
there was a larger evidence base for serum lipids as compared to blood pressure. However, there
was consistent evidence of associations between PFOA exposure and continuous measures of
blood pressure and risk of hypertension in adults from the general population, including adults
living in high-exposure communities located near PFAS manufacturing plants, which
qualitatively support an association between PFOA and cardiovascular effects in humans.
The majority of studies in adults in the general population, including high-exposure
communities, reported positive associations between PFOA serum concentrations and serum
lipids. Studies in adults were prioritized based on reported age-dependent fluctuations in serum
lipids as a result of puberty {Daniels, 2008, 6815477}, which may impact the consistency of
results from studies in children. Specifically, medium confidence epidemiological studies in
adults reported positive associations between PFOA exposure and total cholesterol (TC) (15/18
studies) and low-density lipoprotein (LDL) (12/17 studies). Of these two endpoints, EPA
selected TC for quantitative assessments because the association was the most consistently
observed in adults and the studies for TC were of higher confidence for outcome measurements
compared with LDL. Additionally, the positive associations with TC in these studies were
further supported by a recent meta-analysis that included 14 general population studies in adults
{U.S. EPA, 2024, 11414059} and reported an association between increased cholesterol and
increased PFOA exposure.
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; Gofif, 2014, 3121148; Lloyd-Jones, 2017, 10694407}. Additionally, disturbances in
cholesterol homeostasis contribute to the pathology of nonalcoholic 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
18 Mixed confidence studies on serum lipids were primarily of medium confidence for total cholesterol and HDL cholesterol, and
Low confidence for LDL cholesterol and triglycerides.
4-9
-------
APRIL 2024
hepatoxicity. Increases in serum cholesterol, even at smaller magnitudes at the individual level,
are clinically relevant when extrapolated to the overall population {Gilbert, 2006, 174259}. This
is because, at the population level, even small magnitude increases in serum cholesterol could
shift the distribution of serum cholesterol in the overall population relative to the clinical cut-off,
leading to an increased number of individuals at risk for cardiovascular disease. The SAB PFAS
Panel agreed with this interpretation, stating that "an increase in the number of subjects with a
clinically abnormal value is also expected from the overall change (shift in the distribution
curve) in the abnormal direction. While the clinical relevance of exposure to PFOA.. .cannot be
predicted on an individual basis, the increased number of individuals within a population with
clinically defined abnormal values is of public health concern" {U.S. EPA, 2022, 10476098}.
A total of 15 medium confidence studies {Costa, 2009, 1429922; Nelson, 2010, 1291110; Liu,
2018, 4238514; Dong, 2019, 5080195; Jain, 2019, 5080642; Fan, 2020, 7102734; Steenland,
2009, 1291109; Eriksen, 2013, 2919150; Fitz-Simon, 2013, 2850962; Lin, 2019, 5187597;
Canova, 2020, 7021512; Lin, 2020, 6988476; Olsen, 2003, 1290020; Sakr, 2007, 1291103;
Winquist, 2014, 2851142} reported positive associations between exposure to PFOA and total
cholesterol in adults from the general population. One study {Winquist, 2014, 2851142} was
excluded from POD derivation because the study estimated the risk of levels above clinical
thresholds for TC and these data were not amenable to modeling continuous changes in TC.
Three studies were excluded from POD derivation because they were in occupationally exposed
adult populations only and would not represent typical exposure scenarios for human
environmental exposure {Costa, 2009, 1429922; Olsen, 2003, 1290020; Sakr, 2007, 1291103}.
Three studies {Eriksen, 2013, 2919150; Lin, 2020, 6988476; Canova, 2020, 7021512} were
excluded from POD derivation due to narrow age ranges (i.e., 50-65 years of age, 55-75 years of
age, 40-60 years of age, and 20-39 years of age, respectively) of the study populations that were
less comprehensive than the age groups included by other studies and therefore, may not apply
across the general adult population. One study {Jain, 2019, 5080642} was excluded from POD
derivation because the study reported findings stratified by BMI status without stratification by
exposure.
Although the positive associations between PFOA and TC were supported by the findings of a
recent meta-analysis that included 14 general population studies of adults {U.S. EPA, 2024,
11414059}, EPA did not use the pooled effect from this meta-analysis for POD derivation. This
meta-analysis was not comprehensive of the entire database of studies on PFOA and TC because
it was performed specifically with the purpose of informing aspects of the Pooled Cohort
Atherosclerotic Cardiovascular Disease (ASCVD) model which relies on CVD risk reduction
analysis for those ages 40-89 {U.S. EPA, 2024, 11414059}. The results of another recent meta-
analysis on PFOA and serum lipids {Abdullah Soheimi, 2021, 9959584} was excluded from
POD derivation because the pooled effects reported combined 11 studies with TC, triglycerides
and LDL in multiple populations (i.e., children, adolescents, pregnant women, and adults). As
previously noted, serum lipids rise in early childhood and fluctuate in puberty {Daniels, 2008,
6815477}, and combining study populations at different lifestages would likely result in
unaddressed confounding by age.
Four studies presented overlapping data from NHANES {Nelson, 2010, 1291110; Liu, 2018,
4238514; Fan, 2020, 7102734; Dong, 2019, 5080195}. Of these four, Dong et al. {, 2019,
5080195} was selected for POD derivation because this larger study included data from all
4-10
-------
APRIL 2024
NHANES cycles between 2003 and 2014, while the other three studies reported results for only
one or two cycles within the 2003-2014 range and were therefore not further considered.
Similarly, two studies {Fitz-Simon, 2013, 2850962; Steenland, 2009, 1291109} presented data
on the C8 Health Project population. Fitz-Simon et al. {, 2013, 2850962} was not selected for
POD derivation because it was a part of a short-term follow-up and was not as comprehensive as
the population examined by Steenland et al. {, 2009, 1291109}. Therefore, Steenland et al. {,
2009, 1291109} was also selected for POD derivation. Finally, Lin et al. {, 2019, 5187597} was
also selected for POD derivation because it provided data for a large number of adults (n = 940)
in the general U.S. population from the Diabetes Prevention Program (DPP) population, with
PFOA levels at baseline comparable to those from NHANES 1999-2000.
In summary, 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 describe a variety of PFOA exposure measures across various adult
populations and exhibited many of the study attributes outlined in Section 4 above and in
Appendix A {U.S. EPA, 2024, 11414343}. Dong et al. {, 2019, 5080195} investigated the
NHANES population (2003-2014), Steenland et al. {, 2009, 1291109} investigated effects in a
high-exposure community (the C8 Health Project study population), and Lin et al. {, 2019,
5187597} collected data from prediabetic adults from the DPP and DPPOS study (1996-1999).
Though results reported in animal toxicological studies support the alterations in lipid
metabolism associated with PFOA exposure observed in epidemiological studies, there are
species differences in the direction of effect with increasing dose. As a result of these
differences, there is some uncertainty about the human relevance of these observed responses in
rodents. Additionally, the available mechanistic data do not provide an increased understanding
of the observed non-monotonicity of serum lipid levels and decreased serum lipid levels at
higher dose levels in rodents (Section 3.4.3.2). Due to the uncertainties regarding human
relevance of the animal toxicology studies, EPA did not derive PODs for animal toxicological
studies reporting cardiovascular effects, such as altered serum lipid levels.
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 3-15, the majority of
epidemiological studies assessed endpoints related to fetal growth restriction (26 high and 25
medium confidence studies) and gestational duration (13 high and 13 medium confidence
studies), while fewer studies reported on endpoints related to fetal loss (2 high and 6 medium
confidence studies) and birth defects (4 medium confidence studies). Evidence for birth defects
was limited in that there are only 4 medium confidence studies and those studies provided mixed
findings. Therefore, birth defects not prioritized for POD derivation. 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 POD derivation as there were a relatively limited number
of studies compared with endpoints related to gestational duration and fetal growth restriction
and results from the high confidence studies were mixed. The impacts observed on fetal loss are
supportive of an association between PFOA exposure and adverse developmental effects.
Approximately half of the available studies reporting metrics of gestational duration observed
increased risk associated with PFOA exposure, including among high confidence studies. Six of
4-11
-------
APRIL 2024
the 14 medium or high confidence studies reported inverse associations for gestational age at
birth and 5 of the 11 medium or high confidence studies reported an increased risk of preterm
birth. Gestational age was not prioritized for quantitative analyses because the majority of studies
did not report inverse associations and this endpoint is more prone to measurement error (see
Section 3.4.4.1.2). There were generally more consistent findings showing positive associations
between PFOA exposure and 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
some associations between PFOA exposure and gestational duration, the inconsistencies in the
database and lack of studies sampling in the first trimester of pregnancy resulted in this effect not
being considered for POD derivation. Additionally, the database for fetal growth restriction was
both larger and consisted of more medium and high confidence studies. Therefore, studies
demonstrating fetal growth restriction were prioritized for POD derivation.
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 reported some inverse associations in
high and medium confidence studies. Endpoints characterizing fetal growth restriction were
included for POD derivation because multiple studies reported effects on these endpoints,
particularly decreased birth weight, and reported generally consistent findings across high and
medium confidence studies. As noted in the Developmental Human Evidence Study Evaluation
Considerations (Section 3.4.4.1.2), measures of birth weight were considered higher confidence
outcomes compared with other measures of fetal growth restriction such as birth length, head
circumference, or ponderal index because birth weight measures are less prone to measurement
error {Shinwell, 2003, 6937192}. Studies reporting changes in mean birth weight were more
amenable to modeling compared with those reporting changes in standardized (e.g., z-score)
birth weight measurements. Standardized measurements depend on sources of standardization
and are harder to interpret and compare across studies. As a result, measures of mean changes in
birth weight were considered for quantitative analysis.
Low birth weight (LBW) 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}. LBW is widely considered a useful population level public health measure {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}. Decreases in birthweight, even at smaller
magnitudes at the individual level, are clinically relevant when extrapolated to the overall
population {Gilbert, 2006, 174259}. This is because, at the population level, even small
magnitude decreases in birthweight could shift the distribution of birthweight in the overall
population relative to the clinical cut-off, leading to an increased number of individuals at risk
for decreased birthweight and subsequent effects related to decreased birthweight. The SAB
PFAS Panel agreed with this interpretation, stating that "an increase in the number of subjects
with a clinically abnormal value is also expected from the overall change (shift in the distribution
curve) in the abnormal direction. While the clinical relevance of exposure to PFOA.. .cannot be
4-12
-------
APRIL 2024
predicted on an individual basis, the increased number of individuals within a population with
clinically defined abnormal values is of public health concern" {U.S. EPA, 2022, 10476098}.
Substantial evidence links LBW to a variety of irreversible adverse health outcomes at various
later life stages. 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}.
LBW 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 LBW. These associations between LBW and infant mortality, childhood disease, and
adult disease establish LBW as an adverse effect. Considering the established consequences of
LBW, as well as the consistency of the database and large number of medium and high
confidence studies reporting mean birth weight and other binary birth weight-related measures,
the endpoint of decreased birth weight in infants was selected for POD derivation.
Given the abundance of high confidence epidemiological studies that evaluated decreases in birth
weight, low and medium confidence studies were excluded from POD derivation. Thus, 15 high
confidence studies reporting inverse associations between exposure to PFOA and mean birth
weight {Ashley-Martin, 2017, 3981371; Bell, 2018, 5041287; Chu, 2020, 6315711; Gardener,
2021, 7021199; 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; Wikstrom, 2020, 6311677; Whitworth, 2012, 2349577; Yao,
2021, 9960202} were considered for POD derivation. Three studies {Ashley-Martin, 2017,
3981371; Gardener, 2021, 7021199; Whitworth, 2012, 2349577} were excluded because they
reported sex-stratified results rather than results in both sexes or results for the overall population
in terms of standardized measurements (e.g., z-score) only. Analyses utilizing standardized
measurements as the dependent variable are internally valid, but this type of analysis estimates a
change in birthweight relative to the study population, which would not be generalizable to other
populations. Two studies {Bell, 2018, 5041287; Luo, 2021, 9959610} were not considered
because they used non-preferred exposure measures such as infant whole blood samples from a
heel stick and postpartum maternal exposure samples, which are prone to exposure
misclassification. Four studies {Lauritzen, 2017, 3981410; Lind, 2017, 3858512; Manzano-
Salgado, 2017, 4238465; Valvi, 2017, 3983872} were not considered for POD derivation
because of inconsistencies in associations by sex or study location with no clear biological
explanation for the inconsistency.
As a result of these exclusions, the six remaining high confidence epidemiologic studies {Chu,
2020, 6315711; Govarts, 2016, 3230364; Sagiv, 2018, 4238410; Starling, 2017, 3858473;
4-13
-------
APRIL 2024
Wikstrom, 2020, 6311677; Yao, 2021, 9960202} were considered for POD derivation (Table
4-1). The candidate epidemiological studies described a variety of PFOA exposure measures
across both fetal and neonatal developmental windows. 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. Yao et al. {, 2021, 9960202} was
not further considered because the PFOA exposure concentrations in this cohort were
considerably higher than typical human environmental exposure levels and the exposure median
in this study was at least 10 times higher than the other studies considered. Two of the six studies
examined PFOA levels primarily during trimester one {Sagiv, 2018, 4238410, Wikstrom, 2020,
6311677} and one during trimesters two and three {Starling, 2017, 3858473}. Two studies
examined PFOA collected within days of birth {Chu, 2020, 6315711; Govarts, 2016, 3230364}.
Wikstrom et al. {, 2020, 6311677} reported associations between PFOA levels and decreased
birth weight in the large Swedish Environmental Longitudinal, Mother and child, Asthma and
allergy (SELMA) study cohort with samples collected between 2007 and 2010. Sagiv et al. {,
2018, 4238410} reported on first trimester PFOA samples collected between 1999-2002 in a
Project Viva birth cohort in the U.S. Chu et al. {, 2020, 6315711} reported inverse associations
between maternal PFOA collected within three days of delivery and birth weight in the Chinese
Guangzhou Birth Cohort Study (2013). Starling et al. {, 2017, 3858473} reported associations
between PFOA collected in later pregnancy (range: 20 to 34 weeks gestational age) and
decreased birth weight in the Healthy Start prospective cohort in Colorado (2009-2014). Govarts
et al. {, 2016, 3230364} reported a modest inverse association between PFOA in cord blood and
birth weight in the Flemish Human Environment Health Survey II (FLEHS II) cohort (2008-
2009).
Developmental toxicity results reported in animal toxicological studies are concordant with the
observed developmental effects in epidemiological studies. Specifically, studies in rodents found
that gestational PFOA exposure 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),
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). Some of the developmental effects seen in the offspring of rodents treated with
PFOA (e.g., reduced offspring weight) are consistent with the decreases in birth weight and
adverse effects associated with LBW observed in human populations.
Given the large number of adverse effects identified in the animal toxicological database for the
developmental health outcome, EPA prioritized only the most sensitive effects (i.e., those
observed at lower dose levels and/or higher magnitude) in offspring that were 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 decreased birth weight observed in human infants.
Though several studies measured pregnant dam weight or dam weight at birth, there were
inconsistencies in results across the database, with some studies reporting decreased maternal
weight, some reporting no effect, and some reporting increased maternal weight as a result of
PFOA exposure. 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 results from multiple animal toxicological studies corroborated the observed effect,
thereby increasing the confidence in that effect. EPA additionally focused on studies with
4-14
-------
APRIL 2024
exposure durations lasting through the majority of gestation and/or lactation (i.e., from GD 1
through early postnatal development) rather than those that targeted a specific period of gestation
or postnatal development as they were more likely to detect developmental effects. Multiple
animal toxicological studies observed effects at low dose levels and demonstrated a dose-related
response in decreased offspring weight, decreased pup survival, and developmental delays.
Therefore, these endpoints were prioritized for dose-response analysis.
Numerous studies in both rats and mice reported decreased offspring body weight after
gestational PFOA exposure. Reduced fetal body weight was consistently observed, with 5/5
studies in mice reporting this effect {Blake, 2020, 6305864; Lau, 2006, 1276159; Li, 2018,
5084746; Suh, 2011, 1402560; Yahia, 2010, 1332451}. Reduced pup body weight was also
consistently observed; the majority of the available studies (10/13) reported this effect, two of
which were high confidence studies in rats {NTP, 2020, 7330145; Butenhoff, 2004, 1291063},
indicating consistency across species. EPA selected both reduced pup and fetal weights because
the timing is concordant with the endpoint of decreased infant birth weight prioritized for POD
derivation from the human epidemiological studies and also represents two different
developmental stages (i.e., fetus and pup) across the sensitive perinatal period of development.
Of the five studies reporting decreased fetal body weight in mice, results from Li et al. {, 2018,
5084746} were selected for POD derivation because the exposure duration encompassed the
majority of gestation, the study used a relatively large number of dose groups, and the effect was
observed in multiple dose groups. The two high confidence rat studies reporting reduced pup
weight were not selected for POD derivation due to study design limitations, including the use of
relatively high dose levels, and non-monotonic responses, although they provide qualitative
support for this effect in mice. Of the eight studies reporting reduced pup body weight in mice
{Wolf, 2007, 1332672; Hu, 2010, 1332421; Yahia, 2010, 1332451; Lau, 2006, 1276159; Hu,
2012, 1937235; Song, 2018, 5079725; Abbott, 2007, 1335452; White, 2011, 1276150},
decreased pup weight relative to litter at PND 22 as reported by Lau et al. {, 2006, 1276159} was
ultimately selected for POD derivation because this study reported results as pup weight
averaged by litter rather than individual pups, used an exposure duration that spanned the
majority of gestation, used a larger number of dose groups than the other studies, and reported
the effect in multiple dose groups.
In addition to effects on offspring weight, six studies in mice {Song, 2018, 5079725; Lau, 2006,
1276159; Abbott, 2007, 1335452; Wolf, 2007, 1332672; White, 2011, 1276150; Yahia, 2010,
1332451} reported alterations in pup survival after gestational exposure to PFOA. Pup survival
was selected over fetal survival because the metrics used to determine fetal mortality varied (e.g.,
reported as prenatal loss, litter loss, resorption, reduced fetal survival) and difficult to directly
compare, Additionally, pup survival provides concordance with the timing of the effect of
decreased infant birth weight in humans. Among the six available studies examining pup
survival, Abbott et al. {, 2007, 1335452} was determined to be low confidence for this endpoint
and was therefore excluded for quantitative assessment. EPA selected results from Song et al. {,
2018, 5079725} (PND 21) and Lau et al. {, 2006, 1276159} (PND 0 and 23) for POD derivation
because this study presented data for a larger number of treatment groups spanning broader or
lower dose ranges as compared with Wolf et al. {, 2007, 1332672}, White et al. {, 2011,
1276150}, and Yahia etal. {,2010, 1332451}.
4-15
-------
APRIL 2024
Three studies in mice {Lau, 2006, 1276159; Abbott, 2007, 1335452; Wolf, 2007, 1332672}
reported developmental delays, specifically delayed eye opening, as a result of gestational PFOA
exposure. Abbott et al. {, 2007, 1335452} was not further considered for POD derivation due to
the extensive litter loss in dose groups greater than 1 mg/kg/day and the effect was only observed
in that dose group, limiting the available dose-response range as compared to Lau et al. {, 2006,
1276159} and Wolf et al. {, 2007, 1332672}. EPA selected results from Lau et al. {, 2006,
1276159} over Wolf et al. {,2007, 1332672} for POD derivation because Lau et al. {,2006,
1276159} presented data for a larger number of dose groups spanning a greater dose range.
Additionally, Lau et al. {, 2006, 1276159} reported the effect in multiple dose groups.
Table 4-1 summarizes the studies and endpoints considered for POD derivation.
4-16
-------
APRIL 2024
Table 4-1. Summary of Observed Endpoints in Humans and Rodent Studies Considered for Dose-Response Modeling and
Derivation of Points of Departure
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Justification
Immune Effects
Reduced Antibody
Concentrations for
Diphtheria and
Tetanus
Reduced
immunoglobulin M
(IgM) Response
Budtz-Jorgensen and
Grandjean {, 2018,
508363 l}a
Medium
Timmerman et al. {,
2021,9416315}
Medium
Human (male Yes Decreases in antibody responses to pathogens such as diphtheria and tetanus
and female were observed at multiple ages during childhood, associated with both prenatal
children) and childhood PFOA exposure levels. Effect was large in magnitude and
generally coherent with epidemiological evidence for other antibody effects.
Effects were observed in multiple populations and are supported by studies of
other vaccine types (e.g., rubella {Granum, 2013, 1937228}).
Loveless et al. {,
2008,988599}
Medium
DeWitt et al. {,
2008,1290826}
Medium
C57BL/6N mice
(adult females),
Crl:CD-
1(ICR)BR mice
(adult males)
Yes Functional assessment indicative of immunosuppression. Immune effects were
consistently observed across multiple studies including reduced spleen and
thymus weights, altered immune cell populations, and decreased splenic and
thymic cellularity. Reduced IgM response is coherent with epidemiological
evidence of reduced immune response to vaccinations.
Developmental Effects
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 Yes Evidence for developmental effects is based on consistent inverse effects for
and female FGR including birth weight measures, which are the most accurate endpoint
infants) examined. 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.
4-17
-------
APRIL 2024
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Justification
Decreased Birth
Weight
Decreased Pup
Survival
Decreased Fetal
Body Weight
Yaoetal. {,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 10 times higher than the other studies
considered (see Appendix D, {U.S. EPA, 2024, 11414343}).
Song et al. {, 2018,
5079725}
Medium
Lau et al. {, 2006,
1276159}
Medium
Li et al. {,2018,
5084746}
Medium
Decreased Pup Body
Weight (relative to
litter)
Lau et al. {, 2006,
1276159}
Medium
Delayed Time to Eye Lau et al. {, 2006,
Opening 1276159}
Medium
Kunming mice Yes Effects on pre- and postnatal offspring survival were consistently observed
(Fi males and across multiple studies and species. Decreased pup survival was reported in six
females, studies and three strains of mice {Song, 2018, 5079725; Lau, 2006, 1276159;
PND 21) Wolf, 2007, 1332672; White, 2011, 1276150; Abbott, 2007, 1335452; Yahia,
CD-I mice (Fi 2010, 1332451} and is coherent timing of the critical effect selected in humans
males and (i.e., decreased birth weight in infants). This critical effect is supported by
females, PND 0 observations of prenatal loss, litter loss/resorption, reduced fetal survival, and
and PND 23) increased postweaning mortality observed in mice and rats. Song et al. {, 2018,
5079725} and Lau et al. {, 2006, 1276159} were selected for POD derivation
because they reported data for a larger number of dose groups and tested lower
or broader dose ranges than the other four studies reporting this effect.
Kunming mice
(Fi males and
females, GD 18)
Yes Effects on pre- and postnatal offspring weight were consistently observed
across multiple studies and species. Decreased fetal weight was observed in 5/5
studies in mice and is supported by reduced pup weight observed in studies of
mice and rats. Li et al. {, 2018, 5084746} was selected for POD derivation
because the study tested a relatively large number of dose groups and had
decreased variability compared with the other four 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.
CD-I mice (Fi Yes Effects on pre- and postnatal offspring weight were consistently observed
males and across multiple studies and species. Decreased pup weight was observed in nine
females, studies across two species and is supported by reduced fetal weight reported by
PND 22) five studies in mice. Reduced pup weight at PND 22 reported by Lau et al. {,
2006, 1276159} was selected for POD derivation because the study reported
pup weight relative to litter, tested a relatively large number of dose groups
compared with the other six studies in mice, and reported the effect in multiple
dose groups.
CD-I mice (Fi Yes Effect also observed in Wolf et al. {, 2007, 1332672} and Abbott et al. {, 2007,
males and 1335452}. Lau et al. {, 2006, 1276159} was selected for dose-response because
females, PND this study reported dose response information across a larger number of dose
14 - PND 18) groups (5) and a relatively low dose range (1, 3, 5, 10 and 20 mg/kg/day).
4-18
-------
APRIL 2024
Endpoint
Reference,
Confidence
Strain/Species/
Sex
POD Derived?
Justification
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 adult populations including
and female general population adults in NHANES {Dong, 2019, 5080195}; from
adults) prediabetic adults from the DPP and DPPOS cohort {Lin, 2019, 5187597} 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}. Endpointis supported by
associations between PFOA and blood pressure.
Hepatic Effects
Increased ALT
Increased ALT
Gallo et al. {,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}.
Lin et al. {, 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,
individual cell, both)
in the Liver
Loveless et al. {, Crl:CD- Yes Effect was accompanied in both studies by other liver lesions including
2008, 988599} 1(ICR)BR mice cytoplasmic alteration and apoptosis. Necrotic liver cells were also observed in
Medium (adult males), male mice in Crebelli et al. {, 2019, 5381564} and pregnant dams in Blake et
NTP {, 2020, Sprague- al. {, 2020, 6305864}. Effect is further supported by changes in serum ALT
7330145} Dawley rats levels in animals and humans. Data from females were not considered for POD
High (adult males) 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 etal. {,2012, 1248827}, Grandjean et al. {, 2017, 3858518}, and Grandjean et al. {,2017,4239492}.
4-19
-------
APRIL 2024
b See Section 5.6.3 for discussion on the approach to estimating BMDs from regression coefficients.
4-20
-------
APRIL 2024
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. The 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 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 1 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 serious effects that can result in irreversible structural or functional changes to the
organism, and thq 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. Thq 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 1 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, 10367891} 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 (e.g., mortality, infertility). 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
noncancer health outcomes are described in the subsections below and alongside the modeling
methods and results provided in Appendix E {U.S. EPA, 2024, 11414343}. Considerations for
BMR selection for cancer endpoints are described in Section 4.2 and Appendix E {U.S. EPA,
2024, 11414343}.
4-21
-------
APRIL 2024
4.1.2.1 Hepatic Effects
For the hepatic endpoint of increased serum ALT in adults associated with PFOA exposure, the
BMD and the BMDL were estimated using a BMR of 5% extra risk from the biologically
significant adverse serum ALT level (see Table 4-2). As described in detail in Appendix E {U.S.
EPA, 2024, 11414343}, EPA reviewed the available information regarding potential clinical
definitions of adversity for the endpoint of elevated ALT. Specifically, EPA modeled elevated
human ALT using cutoff levels of 42 IU/L for males and 30 IU/L for females {Valenti, 2021,
10369689}. These are the most updated clinical consensus cutoffs which post-date the American
Association for the Study of Liver Diseases (AASLD) journal of Clinical Liver Disease
recommended values of 30 IU/L for males, and 19 IU/L for females {Kasarala, 2016, 11350060;
Ducatman, 2023, 11412135}. Valenti et al. (2021, 1036989) determined the values using the
same approach at the same center, but using an updated standardized method, a large cohort of
apparently healthy blood donors (ages 18-65 years) and showed that the updated cutoffs were
able to better predict liver disease.
Because EPA identified a biological basis for BMR selection, EPA used the hybrid approach
(see Section 2.3.3.1 of USEPA {, 2012, 1239433}) to estimate the probability of an individual
with an adverse serum ALT level as a function of PFOA exposure. This approach effectively
dichotomizes the data; therefore, EPA considered BMRs of 1%, 5%, and 10% extra risk for this
endpoint. As described in the Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433}, a 10% BMR is often used to describe quantal data, however, "for epidemiological
data, response rates of 10% extra risk would often involve upward extrapolation, in which case it
is desirable to use lower levels, and 1% extra risk is often used as a BMR." EPA considered
BMRs of 5% and 10% extra risk. EPA did not select a 1% BMR because it is often used for
frank effects and cancer incidence {U.S. EPA, 2012, 1239433}, neither of which apply to the
endpoint of elevated serum ALT.
EPA selected a BMR of 5% extra risk because studies have demonstrated that ALT levels at or
slightly above the selected cutoff levels can be associated with more severe liver diseases
{Mathiesen, 1999, 10293242; Wedmeyer, 2010, 11374673}, increased risk of liver-related
mortality {Park, 2019, 10293238; Ruhl, 2009, 3405056; Kim, 2004, 10473876}, and mortality
{Lee, 2008, 10293233}. Based on the severity of the health effects associated with increased
ALT, EPA determined that a BMR of 5% extra risk is warranted {U.S. EPA, 2012, 1239433}; a
10% extra risk would result in a greater number of individuals, especially those in sensitive
subpopulations, experiencing more severe liver diseases such as advanced fibrosis, chronic liver
disease, and even liver-related death. Since there is currently a relatively high prevalence of
elevated ALT in the general population (14% and 13% of U.S. male and female adults,
respectively, aged 20 and older {Valenti, 2021, 10369689}), a small increase in the prevalence
of elevated ALT associated with PFOA exposure would likely increase the number of
individuals with severe liver-related health effects. This also supports using a more health
protective BMR of 5% extra risk (rather than 10%) for POD derivation. EPA presents PODs with
a 10% extra risk BMR for comparison to the selected 5% BMR in Appendix E {U.S. EPA, 2024,
11414343}, as recommended by agency guidance {U.S. EPA, 2012, 1239433}.
For the adverse effects of single cell and focal liver necrosis observed in adult rats following
PFOA exposure, there is currently inadequate available biological or toxicological information to
permit determination of an effect-specific minimal biologically significant response level.
Therefore, in accordance with EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012,
4-22
-------
APRIL 2024
1239433}, a BMR of 10% extra risk was used because it is considered the standard reporting
level for quantal (dichotomous) data and a minimally biologically significant response level (see
Table 4-2).
4.1.2.2 Immune Effects
For the developmental immune endpoint of decreased diphtheria and tetanus antibody response
in children 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 SD benchmark response (BMR) when performing dose-response modeling of data from an
endpoint resulting from developmental exposure. Because Budtz-Jorgensen and Grandjean {,
2018, 5083631} and Timmerman et al. {, 2021, 9416315} measured antibody concentrations in
childhood and PFOA exposure during gestation or childhood, these are considered
developmental studies based on EPA's Guidelines for Developmental Toxicity Risk Assessment
{U.S. EPA, 1991, 732120}, 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 cutoff for the health outcome {U.S. EPA,
2012, 1239433}. As described in detail in Appendix E {U.S. EPA, 2024, 11414343}, EPA
reviewed the available information regarding potential clinical definitions of adversity for this
effect. A blood concentration for tetanus and diphtheria antibodies of 0.1 IU/mL has been cited
in the literature as a "protective level" {Grandjean, 2017, 4239492; Galazka, 1989, 9642152}.
However, in the Immunological Basis for Immunization Series of modules {WHO, 2018,
10406857}, the WHO argued that assay-specific "protective levels" of tetanus antitoxin may not
actually guarantee immunity. Galazka et al. {, 1993, 10228565} similarly argued that several
factors give rise to variability in the vulnerability of individuals to diphtheria and there is no
consensus on what level offers full protection. As such, EPA determined that there is no clear
definition of an adverse effect threshold for the endpoints of reduced tetanus or diphtheria
antibody concentrations in children.
With these two factors in mind, a 0.5 SD was selected as the BMR because: 1) the health
outcome is developmental, and 2) there is no accepted definition of an adverse level of change or
clinical cutoff 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. EPA also presented BMDL modeling using a BMR equivalent to a 1 SD change, as
recommended by agency guidance {U.S. EPA, 2012, 1239433}.
For the effect of reduced IgM response observed in animal toxicological studies, there is
currently inadequate available biological or toxicological information to permit determination of
a minimal biologically significant response level. In accordance with recommendations in EPA's
Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} for continuous data in adult
4-23
-------
APRIL 2024
animal models with no known biologically significant response level, a BMR of 1 SD change
from the control mean was employed (see Table 4-2).
4.1.2.3 Cardiovascular Effects
For the cardiovascular endpoint of increased serum TC in adults associated with PFOA exposure,
the BMD and the BMDL were estimated using a BMR of 5% extra risk from the biologically
significant adverse serum TC concentration {Dong, 2019, 5080195; Steenland, 2009, 1291109}
or a BMR of 0.5 SD {Lin, 2019, 5187597}, depending on the data provided by the study (see
Table 4-2). As described in detail in Appendix E {U.S. EPA, 2024, 11414343}, EPA reviewed
the available information regarding potential clinical definitions of adversity for this effect and
identified the definition of hypercholesterolemia from the American Heart Association {NCHS,
2019, 10369680} as providing the most recent upper reference limit for clinically adverse serum
TC. Specifically, when possible, EPA modeled human cholesterol using a cutoff level of
240 mg/dL for elevated serum total cholesterol {NCHS, 2019, 10369680}.
Because EPA identified a biological basis for BMR selection, EPA used the hybrid approach
(see Section 2.3.3.1 of USEPA {, 2012, 1239433}) to estimate the probability of an individual
with an adverse TC level as a function of PFOA exposure. This approach effectively
dichotomizes the data; therefore, EPA considered BMRs of 1%, 5%, and 10% extra risk for this
endpoint. As described in the Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433}, a 10% BMR is often used to describe quantal data, however, "for epidemiological
data, response rates of 10% extra risk would often involve upward extrapolation, in which case it
is desirable to use lower levels, and 1% extra risk is often used as a BMR." EPA considered
BMRs of 5% and 10% extra risk. EPA did not select a 1% BMR because it is often used for
frank effects and cancer incidence {U.S. EPA, 2012, 1239433}, neither of which apply to the
effect of elevated serum TC. For Lin {, 2019, 5187597), EPA relied on the mean serum TC
concentrations reported across PFOA quartiles (i.e., continuous data) provided by the study, and
therefore considered a BMR of a change in the mean equal to 0.5 SD or 1 SD from the control
mean.
Increased serum cholesterol is associated with changes in incidence of cardiovascular disease
events such as myocardial infarction (MI, i.e., heart attack), IS, and cardiovascular mortality
occurring in populations without prior CVD events {D'Agostino, 2008, 10694408; Goff, 2014,
3121148; Lloyd-Jones, 2017, 10694407}. Based on the severity of the cardiovascular-related
health effects associated with increased cholesterol, EPA determined that selection of a BMR of
5% extra risk or 0.5 SD is warranted {U.S. EPA, 2012, 1239433}; a 10% extra risk or 1SD
would result in a greater number of individuals, especially those in sensitive subpopulations,
experiencing increased incidence of cardiovascular disease events. Since there is currently a
relatively high prevalence of elevated TC in the general population (11.5% of U.S. adults aged
20 and older {NCHS, 2019, 10369680}), a small increase in the prevalence of elevated TC
associated with PFOA exposure would likely increase risk of severe health outcomes, such as
cardiovascular-related events. Thus, this supports using a more conservative BMR of 5% extra
risk or 0.5 SD for POD derivation. EPA presents PODs with a BMR of 10% extra risk {Dong,
2019, 5080195; Steenland, 2009, 1291109} or 1 SD {Lin, 2019, 5187597} for comparison
4-24
-------
APRIL 2024
purposes in Appendix E {U.S. EPA, 2024, 11414343}, as recommended by agency guidance
{U.S. EPA, 2012, 1239433}.
4.1.2.4 Developmental Effects
For the developmental endpoint of decreased birth weight associated with PFOA exposure, the
BMD and the BMDL were estimated using a BMR of 5% extra risk from the biologically
significant birth weight deficit (see Table 4-2). As described in Appendix E {U.S. EPA, 2024,
11414343}, LBW is clinically defined as birth weight less than 2,500 g (approximately 5.8 lbs)
and can include but is not exclusive to 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}.
Consistent with EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433}, EPA
typically selects a 5% or 0.5 SD benchmark response (BMR) when performing dose-response
modeling of data from an endpoint resulting from developmental exposure. Low birthweight is
associated with increased risk for adverse health effects throughout life {Hack, 1995, 8632216;
Reyes, 2005, 1065677; Tian, 2019, 8632212} and therefore, supports a more stringent BMR
below 10% (for dichotomous data) or 1 SD (for continuous data). Because EPA identified a
biological basis for BMR selection, EPA used the hybrid approach (see Section 2.3.3.1 of
USEPA {, 2012, 1239433}) to estimate the probability of an individual with a birth weight
deficit as a function of PFOS exposure. This approach effectively dichotomized the data,
resulting in a BMR defined as a 5% increase in the number of infants with birth weights below
2,500 g.
For delayed time to eye opening and decreased pup survival observed in animal studies, a BMR
of 0.5 SD from the control was employed (see Table 4-2). For decreased fetal and pup weights
observed in animal studies, a BMR of 5% relative deviation was employed. These BMR
selections are consistent with EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433} and the IRIS Handbook {U.S. EPA, 2022, 10367891}, 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
(i.e., fetal and pup weight) 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 SD, and found they were statistically similar {Kavlock, 1995, 75837}. EPA presented
modeling results using a BMR of 0.5 SD for decreased fetal or pup body weight, a BMR of
0.1 SD for the frank effects of decreased fetal or pup survival, and a BMR of 1 SD for delayed
time to eye opening for comparison purposes, based on severity of the endpoints, as
recommended by agency guidance {U.S. EPA, 2012, 1239433} (see Appendix E, {U.S. EPA,
2024, 11414343}).
4-25
-------
APRIL 2024
Table 4-2. Benchmark Response Levels Selected for BMD Modeling of Health Outcomes
Endpoint
BMR
Rationale
Immune Effects
Reduced antibody
concentrations for diphtheria and
tetanus in children
Reduced immunoglobulin M
(IgM) response
0.5 SD Consistent with EPA guidance. EPA typically selects a 5%
or 0.5 SD benchmark response (BMR) when performing
dose-response modeling of data from an endpoint resulting
from developmental exposure in consideration of the
severity of the effect and selects a 1 or 0.5 SD change in
cases where there is no accepted definition of an adverse
level of change or clinical cutoff 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
1 SD was used as per EPA guidance {U.S. EPA, 2012,
1239433}
Developmental Effects
Decreased Birth Weight in
Infants
Decreased Fetal or Pup Weight
Decreased Pup Survival
Delayed Time to Eye Opening
5% extra risk of
exceeding adversity
cutoff (hybrid
approach)
5%
0.5 SD
0.5 SD
Consistent with EPA guidance. EPA typically selects a 5%
or 0.5 SD benchmark response (BMR) when performing
dose-response modeling of data from an endpoint resulting
from developmental exposure in consideration of the
severity of the effect {U.S. EPA, 2012, 1239433}. The use
of the hybrid approach results in dichotomization of the
data and therefore a 5% BMR was selected {U.S. EPA,
2012,1239433}
Consistent with EPA guidance. EPA typically selects a 5%
or 0.5 SD benchmark response (BMR) when performing
dose-response modeling of data from an endpoint resulting
from developmental exposure in consideration of the
severity of the effect {U.S. EPA, 2012, 1239433}
Consistent with EPA guidance. EPA typically selects a 5%
or 0.5 SD benchmark response (BMR) when performing
dose-response modeling of data from an endpoint resulting
from developmental exposure in consideration of the
severity of the effect {U.S. EPA, 2012, 1239433}
Consistent with EPA guidance. EPA typically selects a 5%
or 0.5 SD benchmark response (BMR) when performing
dose-response modeling of data from an endpoint resulting
from developmental exposure in consideration of the
severity of the effect {U.S. EPA, 2012, 1239433}
Cardiovascular Effects
Increased Cholesterol
5% extra risk of
exceeding adversity
cutoff (hybrid
approach)
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, "for epidemiological data, response rates of 10%
extra risk would often involve upward extrapolation, in
which case it is desirable to use lower levels" {U.S. EPA,
2012, 1239433}. Because increased TC is not a frank
4-26
-------
APRIL 2024
Endpoint BMR Rationale
effect but is associated with increased incidence of severe
cardiovascular-related effects a 5% extra risk was used as
the BMR. The response rate of 5% extra risk limits further
increases in the prevalence of this effect.
0.5 SD Because increased TC is not a frank effect but is associated
with increased incidence of severe cardiovascular-related
effects, a 0.5 SD was used as the BMR. A change from the
mean of 0.5 SD limits further increases in the prevalence of
this effect
Hepatic Effects
Increased ALT 5% extra risk of Although EPA's Benchmark Dose Technical Guidance
exceeding adversity {U.S. EPA, 2012, 1239433 } recommends a BMR based on
cutoff (hybrid a 10% extra risk for dichotomous endpoints when
approach) biological information is not sufficient to identify the
BMR, "for epidemiological data, response rates of 10%
extra risk would often involve upward extrapolation, in
which case it is desirable to use lower levels" {U.S. EPA,
2012, 1239433}. Because increased ALT is not a frank
effect but is associated with increased incidence of severe
liver-related effects a 5% extra risk was used as the BMR.
The response rate of 5% extra risk limits further increases
in the prevalence of this effect
Single Cell and Focal Liver 10% Insufficient information available to determine minimal
Necrosis 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
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 Appendix F, {U.S. EPA, 2024, 11414343}),
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. The species used in the
toxicological studies (i.e., species of interest) were rats, mice, and nonhuman primates; model
parameters for these species of interest were available. Good agreement with training and test
4-27
-------
APRIL 2024
(out-of-sample) datasets shows that the model performance is good compared with both the data
used to identify model parameters and to external data. This was assessed using the mean square
log error (MSLE) to compare model predicted concentration values to observed PFOA serum
concentrations following single dose exposure to animals. Training set data demonstrated an
MSLE of 0.40 for PFOA. For test set data, the MSLE was 1.4 for PFOA. As evidenced in the
supplementary code, the discrepancy in model predictions for test set data is driven by higher
animal PFOA doses that were outside the scope of the original model calibration. The general
agreement between test and training datasets increases confidence that the model can be used to
make accurate predictions of internal dose metrics for the dose magnitudes used in the available
toxicology studies. 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 lifestages.
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 lifestage 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 5.6.1.
4-28
-------
APRIL 2024
4.1.3.1.1 Animal Model Parameters
Pharmacokinetic parameters for different species and strains 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/li/kg1174
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 (ki:)
(3.1 x e"10- 38,000)
(0.058-52)
(0.039-0.27)
(0.0096-0.08)
(2.4 xe10- 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
|.uno 1/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).
aDatasets 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}.
4-29
-------
APRIL 2024
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 PFOAHESD (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 in Section
3.3.2.1, a 24 hour 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 with 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
tl/2,P
Vd,a Vd,p
CL
tl/2,a
tl/2,P
Vd,a
vd,p
CL
(days)
(days)
(L/kg) (L/kg)
(L/d/kg)
(days)
(days)
(L/kg)
(L/kg)
(L/d/kg)
Rat
Model
5.8
16.5
0.12 0.35
0.0147
0.16
2.84
0.16
2.81
0.686
Literature
1.64a,
10.25b
0.11ac. 0.15b c
0.0473,
0.19a,
0.22b
0.17a'°, 0.12b'°
0.6133,
2.8b
0.013b
0.028b
0.81b
Mouse
Model
-
-
-
-
17.8
18.9
0.18
0.19
0.007
Literature
-
-
-
-
-
-
-
-
-
Notes: CL = clearance; PK = pharmacokinetic; Xm,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.
aInformation obtained from Kim et al. {,2016, 3749289}.
b Information 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 postweaning 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 postweaning phase utilized the body weight-scaled Wambaugh model to
4-30
-------
APRIL 2024
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 Lifestage 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 lactational 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 postweaning 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
lifestage-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
4-31
-------
APRIL 2024
(-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 postpartum 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, postweaning, 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°
18.5 g
Volume of Distribution (Vd)
L/kg
0.18d
0.2 g
Starting Milk Consumption Rate (r°m,ii
kgmiik/dav
0.001h
0.000P
Week 1 Milk Consumption Rate (rVik)
kgmiik/dav
0.003h
0.00031
Week 2 Milk Consumption Rate (r2min:)
kgmiik/dav
0.0054h
0.000541
Week 3 Milk Consumption Rate (r\nin:)
kgmiik/day
0.0059h
0.000591
Notes: PK = pharmacokinetic.
aInformation obtained from Loccisano etal. {,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 etal. {,2020, 6512379}.
information obtained from Blake et al. {, 2020, 6305864}.
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.
4-32
-------
APRIL 2024
where Rmiik used in the model is a piecewise linear function comprising each rVik 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 and lactation study design in rat
dams exposed to 30 mg/kg/day PFOA from GD 1-LD 22 that gave birth to pups who are exposed
through gestation and lactation until weaning {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
either in atero or during lactation only.
PFOA: 30 mg/kg/day
£ 101 1
u
8 io0 -
<
o
£ 10"2 -
0 5 10 IB 20 25 30 35
time [days]
cn
— 101
CL i_i
ss
< 10*
o
Li-
ft.
0 5 10 15 20 25 30 35
Time [days]
Figure 4-2. Gestation and 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.
4-33
-------
APRIL 2024
PFOA: 5 mg/kg/day
20
Time [days]
Figure 4-3. Gestation and 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 utero 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 laboratory
animal species 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 lifestages 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 lifestages: during gestation (Cavg dam gest), during
lactation (Cavg damjact), or combined gestation and lactation (Cavg dam gestjact). In pups for Cmax,
two different lifestages were calculated either during gestation or lactation (Cmax_puP gest,
Cmax_puPjact). In pups for time averaged metrics, a Cavg was calculated during gestation, lactation,
or combined gestation and lactation (Cavg_pup gest, Cavg_pup lact, and Cavg_pup gest lact).
EPA selected the metric of Ciast7 for studies examining noncancer effects using
nondevelopmental exposure paradigms. This metric provides a consistent internal dose for use
across disparate chronic and subchronic study designs where steady state may or may not have
been reached in the animal following continuous dosing. When the animal has reached steady
state, Ciast7 is equal to the steady-state concentration and for non-steady-state study designs, this
metric averages the concentration variability over a week's worth of dosing rather than using a
4-34
-------
APRIL 2024
single, maximal concentration. This allows for extrapolation to the human model where a steady-
state assumption is implemented for adult dose metric calculations.
For developmental endpoints, the metric of Cmax is typically used when there is a known
mechanism of action (MOA) related to a threshold effect during a specific window of
susceptibility. From the Guidance for applying quantitative data to develop data-derived
extrapolation factors for interspecies and intraspecies extrapolation {U.S. EPA, 2014,
2520260}, the choice of this metric "depends on whether toxicity is best ascribed to a transient
tissue exposure or a cumulative dose to the target tissue." Furthermore, the guidance clarifies that
"for chronic effects, in the absence of MO A information to the contrary, it is generally assumed
that some integrated cumulative measure of tissue exposure to the active toxicant is the most
appropriate dose metric (e.g., AUC)" {U.S. EPA, 2014, 2520260}. Repeat dosing coupled with a
lack of a defined MOA for the apical endpoints used for dose-response modeling resulted in EPA
excluding Cmax as an internal dose metric for animal toxicological endpoints. However, EPA
provides modeling results using Cmax for comparison purposes in Appendix E {U.S. EPA, 2024,
11414343}.
EPA selected the metric of Cavg for studies with reproductive or developmental exposure designs
encompassing gestation and/or lactation. One factor considered for this selection pertains to the
long half-life of PFOA and the degree of accumulation throughout pregnancy and lactation.
Because PFOA is not cleared within 24 hours, daily dosing throughout pregnancy/lactation will
result in a Cmax that falls on the final day of pregnancy or lactation or a Ciast7 only representative
of the final days of gestation or lactation, even if dosing ceases after birth, due to ongoing
lactational exposure. The endpoints in this assessment (decreased fetal or pup weight, decreased
pup survival, delayed time to eye opening) do not have established MO As or known windows of
susceptibility and instead are expected to result from sustained internal dose from repeated
exposures. If, as anticipated, this window of susceptibly for a given endpoint is not on the final
day or the last week of exposure, the Cmax or Ciast7 will not capture the exposure at the time
associated with the adverse effect. A Cavg metric is more representative of the exposure
throughout the potential window of susceptibility. This selection is also supported by the
Guidelines for Developmental Toxicity Risk Assessment {U.S. EPA, 1991, 11346204}, which
state that when pharmacokinetic data are available, as is the case for PFOA, "adjustments may be
made to provide an estimate of equal average concentration at the site of action for the human
exposure scenario of concern." The selection of Cavg for developmental animal studies is
therefore consistent with the guidance for humans.
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
(postweaning) and then divides by 2 years.
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 body weight during
growth, and that required minimal new development. Previous modeling efforts suggest that
4-35
-------
APRIL 2024
limiting model complexity helps to prevent errors and facilitates rapid implementation
{Bernstein, 2021, 9639956}. For the human epidemiological and animal toxicological 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,
3299692} model as the basis for the pharmacokinetic modeling approach used for PFOA is
available in Sections 5.6.2 and 5.7.
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
nonpregnant 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 body
weight {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 Appendix,
{U.S. EPA, 2024, 11414343}). 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.
The use of the Verner model in humans presents a substantial advancement in approach for
endpoints in children compared with the previous EPA assessment of PFOA {U.S. EPA, 2016,
3603279}. The 2016 PFOA HESD 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 in children. The current approach explicitly models PFOA exposure to
4-36
-------
APRIL 2024
infants during nursing who are undergoing rapid development, including growth, through
childhood and 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 Appendix, {U.S.
EPA, 2024, 11414343}). 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 Appendix, {U.S. EPA,
2024, 11414343}).
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}.
bThompsonetal. {,2010,2919278}.
cLi etal. {,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 Appendix, {U.S. EPA, 2024, 11414343}). 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
after a long period of exposure to drinking water contaminated with PFOA. The second study
involved 106 participants with a median number of six samples per person but with only a 2-year
observation period Li et al. {, 2017, 9641333}. The good agreement between the second study
4-37
-------
APRIL 2024
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 {U.S. EPA, 2024, 11414343}.
Uncertainties related to EPA's selected half-life are discussed in Section 5.6.2.
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 nonhuman 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 {U.S. EPA, 2024, 11414343}.
Uncertainties related to EPA's selected Vd are discussed in Section 5.6.2.
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
{U.S. EPA, 2024, 11414343}. 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.
Table 4-7. Summary of Studies Reporting the Ratio of PFOA Levels in Breastmilk and
Maternal Serum or Plasma
Source
Milk: Maternal
HERO ID
Plasma Ratio
Included in Verner et al.
{, 2016,3299692} Analysis
Haug et al. {,2011,2577501}
2577501 0.038
No
4-38
-------
APRIL 2024
Source
HERO ID
Milk: Maternal
Included in Verner et al.
Plasma Ratio
{, 2016,3299692} Analysis
Seung-Kyu Kim et al. {,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 milkmaternal plasma ratio used by Verner et al. {, 2016, 3299692}.
d Average value of milk:matemal 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 Appendix, {U.S. EPA, 2024,
11414343}). 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 Appendix, {U.S. EPA, 2024, 11414343}). 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 (https://github.eom/USEPA/OW-PFOS-PFO A-MCLG-support-
PK-models). 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 Endpoints and
Dosimetric Interpretation of Epidemiological Endpoints
Different approaches were taken to estimate PODheds depending on the species (i.e., human
versus animal model) and lifestage (e.g., developmental, adult). The PODs from epidemiological
studies (immune, developmental, hepatic, and serum lipid endpoints) were derived using hybrid
or benchmark dose modeling (see Appendix E.l, {U.S. EPA, 2024, 11414343}) which provided
an internal serum concentration in ng/L. The 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
4-39
-------
APRIL 2024
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 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 E.2 {U.S. EPA, 2024,
11414343}. 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 nonlinear 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 Appendix E, {U.S. EPA, 2024, 11414343}). If BMD modeling did
not produce a viable model, a NOAEL or LOAEL approach was used consistent with EPA
guidance {U.S. EPA, 2012, 1239433}. 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,avg), 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 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.
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.
4-40
-------
APRIL 2024
Table 4-8. PODheds Considered for the Derivation of Candidate RfD Values
Endpoint
Reference,
Confidence
POD T rie POD Internal
Strain/Species/Sex/Age M , | ' Dose/Internal Dose
Metric3
PODhed
(mg/kg/day)
Notes on Modeling
Immunological Effects
Decreased serum
anti-tetanus
antibody
concentration in
children
Decreased serum
anti-diphtheria
antibody
concentration in
children
Budtz-Jorgensen
and Grandjean {,
2018, 508363 l}b
Medium
Budtz-Jorgensen
and Grandjean {,
2018, 508363 l}b
Medium
Timmerman et al.
{,2021,9416315}
Medium
Budtz-Jorgensen
and Grandjean {,
2018, 508363 l}b
Medium
Budtz-Jorgensen
and Grandjean {,
2018, 508363 l}b
Human, male and female; BMDL0
PFOA concentrations at
age 5 and anti-tetanus
antibody serum
concentrations at age 7
Human, male and female; BMDL0
PFOA concentrations in
the mother0 and anti-
tetanus antibody serum
concentrations at age 5
Human, male and female; BMDLo
PFOA concentrations and
anti-tetanus antibody
concentrations at ages 7-
12
Human, male and female; BMDLo
PFOA concentrations at
age five and anti-
diphtheria antibody serum
concentrations at age 7
Human, male and female;
PFOA concentrations in
the mother0 and anti-
BMDLo
3.47 ng/mL 3.05 x 10 BMRof0.5 SD provided
reasonably good estimate of
5% extra risk; single- and
multi-PFAS models resulted in
same BMDL; selected BMDL
was based on significant
regression parameter
3.31 ng/mL 5.21 x 10 7 PFOA concentrations may be
influenced by pregnancy
hemodynamics; single- and
multi-PFAS models resulted in
similar BMDLs; selected
BMDL was based on
significant regression
parameter
2.26 ng/mL 3.34 x 10~7 BMR of 0.5 SD may not be a
reasonably good estimate of
5% extra risk; BMDL was
based on nonsignificant
regression parameter; no multi-
PFAS modeling was conducted
3.32 ng/mL 2.92 x 10 7 Single-and multi-PFAS
models resulted in comparable
BMDLs though there was a
30% change in the effect size
when controlling for PFOS;
selected BMDL was based on
significant regression
parameter
1.24 ng/mL 1.95 / 10 PFOA concentrations may be
influenced by pregnancy
hemodynamics; single- and
4-41
-------
APRIL 2024
Reference,
POD Type,
Model
POD Internal
PODhed
Endpoint
Confidence
Strain/Species/Sex/Age
Dose/Internal Dose
Metric3
(mg/kg/day)
Notes on Modeling
Medium
diphtheria antibody serum
concentrations at age 5
multi-PFAS models resulted in
similar BMDLs though there
was a 30% change in the effect
size when controlling for
PFOS
Timmerman et al.
Human, male and female;
BMDLo.5sd
1.49 ng/mL
2.20 >
< 10~7
BMR of 0.5 SD may not be a
{,2021,9416315}
PFOA concentrations and
reasonably good estimate of
Medium
anti-diphtheria antibody
concentrations at ages 7-
12
5% extra risk; BMDL was
based on nonsignificant
regression parameter
Decreased IgM
Dewitt et al. {,
C57BL/6N Mice, females,
BMDLisd,
18.2 mg/L
2.18 >
< 10~3
Selected model showed
response to SRBC
2008,1290826}
Medium
adults, Study 1
Polynomial
Degree 4
Clast7,avg
adequate fit (p > 0.1) and
lowest AIC
Dewitt et al. {,
C57BL/6N Mice, females,
NOAELd
45.3 mg/L
5.43 >
< 10~3
Test for constant variance and
2008,1290826}
adults, Study 2
(1.88 mg/kg/da
Clast7,avg
test for nonconstant variance
Medium
y)
failed therefore a NOAEL
approach was taken
Loveless et al. {,
Crl: CD-1 (ICR)BR Mice,
BMDLisd,
57.6 mg/L
6.91 >
< 10~3
Selected model showed
2008,988599}
males, adults
Exponential 3
Clast7,avg
adequate fit (p > 0.1) and
Medium
lowest AIC
Developmental Effects
Decreased Birth
Chu et al. {,2020,
Human, male and female;
BMDL5RD,
2.0 ng/mL
3.15 >
< 10~7
PFOA concentrations may be
Weight
6315711}
High
PFOA serum
concentrations in third
trimester
Hybrid
influenced by pregnancy
hemodynamics; selected
BMDL was based on
significant regression
parameter
Govarts et al. {,
Human, male and female;
BMDL5RD,
1.2 ng/mL
2.28 >
< 10~7
PFOA concentrations may be
2016, 3230364}
PFOA concentrations in
Hybrid
influenced by pregnancy
High
umbilical cord
hemodynamics; selected
BMDL was based on
nonsignificant regression
parameter
4-42
-------
APRIL 2024
Reference,
POD Type,
Model
POD Internal
PODhed
Endpoint
Confidence
Strain/Species/Sex/Age
Dose/Internal Dose
Metric3
(mg/kg/day)
Notes on Modeling
Sagiv et al. {,
Human, male and female;
BMDL5RD,
9.1 ng/mL
1.21 :
< 10~6
Selected BMDL was based on
2018,4238410}
PFOA serum
Hybrid
nonsignificant regression
High
concentrations in first and
second trimesters
parameter
Starling et al. {,
Human, male and female;
BMDL5RD,
1.8 ng/mL
2.65 :
< 10~7
PFOA concentrations may be
2017,3858473}
PFOA serum
Hybrid
influenced by pregnancy
High
concentrations in second
and third trimesters
hemodynamics; selected
BMDL was based on
significant regression
parameter
Wikstrom et al. {,
Human, male and female;
BMDL5RD,
2.2 ng/mL
2.92 :
< 10~7
Selected BMDL was based on
2020,6311677}
PFOA serum
Hybrid
significant regression
High
concentrations in first and
second trimesters
parameter
Decreased Pup
Song et al. {, 2018,
Kunming Mice, Fi males
BMDL0.5SD,
12.3 mg/L
6.40 ;
< 10~4
Selected model showed
Survival
5079725}
Medium
and females (PND 21)
Polynomial
Degree 3
Cavg_pup gest lact
adequate fit (p > 0.1) and
lowest AIC
Lau et al. {, 2006,
CD-I Mice, Fi males and
NOAELd
19.1 mg/L
3.23 :
< 10~3
No models had adequate fit.
1276159}
females (PND 0)
(3 mg/kg/day)
Cavg_pup gest
Test for constant variance
Medium
failed, and test for nonconstant
variance failed. NOAEL
approach taken
Lau et al. {, 2006,
CD-I Mice, Fi males and
NOAELd
26.6 mg/L
1.38 ;
< 10~3
Test for constant variance
1276159}
females (PND 23)
(3 mg/kg/day)
Cavg_pup gest lact
failed. For nonconstant
Medium
variance models, goodness of
fit for nonconstant models was
poor. NOAEL approach taken
Decreased Fetal
Li et al. {,2018,
Kunming Mice, Fi males
NOAELd
8.5 mg/L
1.44 ;
< 10~3
No models had adequate fit.
Body Weight
5084746}
Medium
and females (GD 18)
(1 mg/kg/day)
Cavg_pup gest
Test for constant variance
failed, and test for nonconstant
variance failed. NOAEL
approach taken
Decreased Pup
Lau et al. {, 2006,
CD-I Mice, Fi males and
NOAELd
15.8 mg/L
8.2 x
10'
No models had adequate fit.
Body Weight
1276159}
Medium
females (PND 23)
(1 mg/kg/day)
Cavg_pup gest lact
Test for constant variance
failed. For nonconstant
4-43
-------
APRIL 2024
Endpoint
Reference,
Confidence
Strain/Species/Sex/Age
POT> T ni> POD Internal
.. , ' Dose/Internal Dose
Model A/r . . a
Metric3
PODhed
(mg/kg/day)
Notes on Modeling
variance models, goodness of
fit for nonconstant models was
poor. NOAEL approach taken
Delayed Time to
Lau et al. {, 2006,
CD-I Mice, Fi males and
BMDLo.5sd,
8.0 mg/L
4.17 >
< 10~4
Selected model showed
Eye Opening
1276159}
Medium
females (PND 14 - PND
18)
Polynomial
Degree 2
Cavg_pup gest lact
adequate fit (p > 0.1) and
lowest AIC
Cardiovascular Effects (Serum Lipids)
Increased Total
Dong et al. {, 2019, Human, male and female,
BMDL5RD,
2.29 ng/mL
2.75 >
< 10~7
BMDL based on analyses
Cholesterol
5080195}
Medium
age 20-80
Hybrid
excluding individuals
prescribed cholesterol
medication and significant
regression parameter
Steenland et al. {,
Human, male and female,
BMDL5RD,
4.25 ng/mL
5.10 >
< 10~7
BMDL based on analyses
2009,1291109}
Medium
age 18 and older
Hybrid
excluding individuals
prescribed cholesterol
medication and significant
regression parameter
Lin et al. {,2019,
Human, male and female,
BMDL0.5SD,
5.28 ng/mL
6.34 >
< 10~7
Analyses include individuals
5187597}
Medium
age 25 and older
Linear
prescribed cholesterol
medication and significant
regression parameter
Hepatic Effects
Increased ALT
Gallo et al. {,2012,
1276142}
Medium
Human, female, age 18
and older
BMDL5RD,
Hybrid
17.9 ng/mL
2.15 >
< 10~6
BMDL based on significant
regression parameter
Darrow et al. {,
Human, female, age 18
BMDL5RD,
66.0 ng/mL
7.92 >
< 10~6
BMDL based on modeled
2016,3749173}
and older
Hybrid
serum PFOA concentrations
Medium
and significant regression
parameter
Nian et al. {, 2019,
Human, female, age 22
BMDL5RD,
3.76 ng/mL
4.51 >
< 10~7
BMDL based on significant
5080307}
Medium
and older
Hybrid
regression parameter
4-44
-------
APRIL 2024
Endpoint
Reference,
Confidence
Strain/Species/Sex/Age
POD Type,
Model
POD Internal
Dose/Internal Dose
Metric3
PODhed
(mg/kg/day)
Notes on Modeling
Increased Focal
Loveless et al. {,
Crl: CD-1 (ICR)BR Mice,
BMDLiord,
10.0 mg/L
1.20 >
< 10~3
Selected model showed
Necrosis
2008,988599}
adult male
Dichotomous
Clast7,avg
adequate fit (p > 0.1) and
Medium
Hill
presented most protective
BMDL in consideration of the
adversity of effect
Increased Individual
Loveless et al. {,
Crl: CD-1 (ICR)BR Mice,
BMDLiord,
36.0 mg/L
4.32 >
< 10~3
Selected model showed
Cell Necrosis
2008,988599}
adult male
Probit
Clast7,avg
adequate fit (p > 0.1) and
Medium
lowest AIC
Increased
NTP {, 2020,
Sprague-Dawley Rats,
BMDLiord,
100 mg/L
1.20 >
< 10"2
Selected model showed
Hepatocyte Single
7330145}
males; perinatal and
Gamma
Cavg_pup total
adequate fit (p > 0.1) and
Cell Death
High
postweaning
lowest AIC
Increased Necrosis
NTP {, 2020,
Sprague-Dawley Rats,
BMDLiord,
26.9 mg/L
3.23 >
< 10~3
Selected model showed
7330145}
males; perinatal and
Multistage
Cavg_pup total
adequate fit (p > 0.1) and
High
postweaning
Degree 1
lowest AIC
Notes: AIC = Akaike information criterion; 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 SD 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;
Cavg_pup_gest = average blood concentration normalized per day during gestation; Cavg_pup total = average blood concentration in pup; Ciast7,avg = average blood concentration over 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 Appendix {U.S. EPA, 2024, 11414343} for additional details on BMD modeling.
b Supported by Grandjean et al. {,2012, 1248827}, Grandjeanetal. {, 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, a NOAEL/LOAEL approach was selected.
4-45
-------
APRIL 2024
4.1.4.1 Hepatic Effects
Increased ALT in individuals aged 18 and older {Gallo, 2012,1276142; Darrow, 2016,
3749173} or 22 and older {Nian, 2019, 5080307}
The POD for increased ALT in adults was derived by quantifying a benchmark dose using a
hybrid modeling approach (see Appendix E.l, {U.S. EPA, 2024, 11414343}) 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. A BMR of 5% extra risk was chosen per EPA's Benchmark
Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2).The internal serum POD
was converted to an external dose (PODhed), in mg/kg/day (see Section 4.1.3.2). 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 was 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,avg {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 per EPA's
Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2). The Ciast7.avg
was selected for all non-developmental studies (i.e., studies with exposure during adulthood
only) rather than alternate metrics such as Cmax to provide a consistent internal dose for use
across chronic and subchronic study designs where steady state may or may not have been
reached and to allow extrapolation to the human PK model (Section 4.1.3.1.3). 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 (Section
4.1.3.2). This calculation was 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,avg {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
per EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2).
The Ciast7,avg was selected for all non-developmental studies (i.e., studies with exposure during
adulthood only) than alternate metrics such as Cmax to provide a consistent internal dose for use
across chronic and subchronic study designs where steady state may or may not have been
reached and to allow extrapolation to the human PK model (Section 4.1.3.1.3). 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 (Section
4.1.3.2). This calculation was 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)).
Necrosis, Sprague-Dawley rats, males, perinatal and postweaning, Cavg_pupjotai {NTP, 2020,
7330145}
4-46
-------
APRIL 2024
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
perEPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2).
For endpoints derived from NTP {, 2020, 7330145}, an additional dose metric was developed
which averages the concentration in the offspring from conception to the end of the 2-year
postnatal exposure period (Cavg_puPjotai; see Section 4.1.3.1.3). 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 (Section 4.1.3.2). This
calculation was 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)).
Hepatocyte Single Cell Death, Sprague-Dawley rats, males, perinatal and postweaning,
Cavg pup total {NTP, 2020, 7330145}
Increased incidence of single cell death 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 perEPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section
4.1.2). For endpoints derived from NTP {, 2020, 7330145}, an additional dose metric was
developed which averages the concentration in the offspring from conception to the end of the 2-
year postnatal exposure period (Cavg_puP totai; see Section 4.1.3.1.3). 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 (Section
4.1.3.2). This calculation was 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)).
4.1.4.2 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 Appendix E.l, {U.S. EPA, 2024, 11414343}) on the measured PFOA serum
concentrations at age 5, which provided an internal serum concentration POD in mg/L. A BMR
of 0.5 SD was chosen per EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012,
1239433} (Section 4.1.2). The internal serum POD was converted to an external dose (PODhed),
in mg/kg/day, using the updated Verner model (described in Section 4.1.3.2). For this, the model
was run starting at the birth of the mother, with constant exposure relative to body weight.
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, where
the exposure to the child was only through lactation, which was much greater than the exposure
to the mother. After 1 year, the exposure to the child, relative to body weight, was set to the same
value as the mother. The model provided predictions for a child aged 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 different growth
curves specific to male and female children were used in the model, the model predicted slightly
different (less than 5%) serum concentrations for each. The slightly lower HED in males was
then selected as it was the most health protective.
4-47
-------
APRIL 2024
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 Appendix E, {U.S. EPA, 2024, 11414343}) 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. A BMR of 0.5 SD was selected was chosen per EPA's Benchmark Dose Technical
Guidance (U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was converted
to an external dose (PODhed), in mg/kg/day, using the updated Verner model (described in
Section 4.1.3.2). 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
was 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 Appendix E, (U.S. EPA, 2024, 11414343}) on the measured PFOA serum
concentrations at ages 7-12, which provided an internal serum concentration POD in mg/L. A
BMR of 0.5 SD was chosen per EP A's Benchmark Dose Technical Guidance (U.S. EPA, 2012,
1239433} (Section 4.1.2). The internal serum POD was converted to an external dose (PODhed),
in mg/kg/day, using the updated Verner model (described in Section 4.1.3.2). For this, the model
was run similarly to the endpoint based on antibodies at age 7 {Budtz-j0rgensen, 2018,
5083631}, but the model was run until the median age of this cohort at blood collection,
9.9 years. Reverse dosimetry was used to calculate the PODhed that resulted in a serum level
equal to the POD at that age. Because different growth curves specific to male and female
children were used in the model, the model predicted slightly different (less than 5%) serum
concentrations for each sex. The lower HED was then selected as it was the most health
protective.
Decreased IgM response to SRBC, C57BL/6N mice, Female, Studies 1 and 2, Ciast7,avg
{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 (Section 4.1.3.1). 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} (Section 4.1.2). The Ciast7,avg was selected
for all non-developmental studies (i.e., studies with exposure during adulthood only) rather than
alternate metrics such as Cmax to provide a consistent internal dose for use across chronic and
subchronic study designs where steady state may or may not have been reached and to allow
extrapolation to the human PK model (Section 4.1.3.1.3). For Study 1, the BMDS produced a
BMDL in mg/L. For Study 2, the tests for constant and nonconstant variance failed therefore a
4-48
-------
APRIL 2024
NOAEL approach was taken. 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
(Section 4.1.3.2). This calculation was 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, Crl:CD-l(ICR)BR mice, Male, Ciast7,avg {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. 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} (Section
4.1.2). The Ciastzavg was selected for all non-developmental studies (i.e., studies with exposure
during adulthood only) rather than alternate metrics such as Cmax to provide a consistent internal
dose for use across chronic and subchronic study designs where steady state may or may not
have been reached and to allow extrapolation to the human PK model (Section 4.1.3.1.3). 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
(Section 4.1.3.2). This calculation was 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 Appendix E, {U.S. EPA, 2024, 11414343}) 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. A BMR of 5% extra risk was chosen per EPA's Benchmark Dose Technical Guidance
{U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was converted to an
external dose (PODhed), in mg/kg/day (Section 4.1.3.2). 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 was 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 Appendix E, {U.S. EPA, 2024, 11414343}) 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. A BMR of 5% extra risk was chosen per EPA's Benchmark Dose
4-49
-------
APRIL 2024
Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day (Section 4.1.3.2). 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 was 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 BMDS
(see Appendix E, {U.S. EPA, 2024, 11414343}) from 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 DPP and Outcomes Study (DPPOS), which provided an internal serum
concentration POD in mg/L. A BMR of 0.5 SD was selected per EPA's Benchmark Dose
Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day (Section 4.1.3.2). 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 was 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 Appendix E, {U.S. EPA, 2024, 11414343}) 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. A BMR of 5% extra risk was chosen per EPA's Benchmark Dose Technical Guidance
{U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was converted to an
external dose (PODhed), in mg/kg/day, using the updated Verner model (described in Section
4.1.3.2). 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 body weight.
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, 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 was
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 Appendix E, {U.S. EPA, 2024, 11414343}) on the
measured PFOA serum concentrations collected from an umbilical cord sample, which provided
4-50
-------
APRIL 2024
an internal serum concentration POD in mg/L. A BMR of 5% extra risk was chosen per EPA's
Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal
serum POD was converted to an external dose (PODhed), in mg/kg/day, using the updated
Verner model (described in Section 4.1.3.2). 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 was independent of the sex of the child in the model.
Decreased birthweight using the mother's serum PFOA concentration collected in the first
and second trimesters {Sagiv, 2018, 4238410}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see Appendix E, (U.S. EPA, 2024, 11414343}) on the
measured PFOA serum concentrations collected from the mother primarily in the first trimester
(median gestational age: 9 weeks; range: 5-19 weeks), which provided an internal serum
concentration POD in mg/L. A BMR of 5% extra risk was chosen per EPA's Benchmark Dose
Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was
converted to an external dose (PODhed), in mg/kg/day, using the updated Verner model
(described in Section 4.1.3.2). 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 was
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 Appendix E, {U.S. EPA, 2024, 11414343}) on the
measured PFOA serum concentrations collected from the mother in trimesters 2 and 3 (median
gestational age of 27 weeks), which provided an internal serum concentration POD in mg/L. A
BMR of 5% extra risk was chosen per EPA's Benchmark Dose Technical Guidance {U.S. EPA,
2012, 1239433} (Section 4.1.2). The internal serum POD was converted to an external dose
(PODhed), in mg/kg/day, using the updated Verner model (described in Section 4.1.3.2). 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 was 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}
The POD for decreased birth weight in infants was derived by quantifying a benchmark dose
using a hybrid modeling approach (see Appendix E, {U.S. EPA, 2024, 11414343}) on the
measured PFOA serum concentrations collected from the mother in the trimesters 1 and 2
4-51
-------
APRIL 2024
(median gestational age of 10 weeks), which provided an internal serum concentration POD in
mg/L. A BMR of 5% extra risk was chosen per EPA's Benchmark Dose Technical Guidance
{U.S. EPA, 2012, 1239433} (Section 4.1.2). The internal serum POD was converted to an
external dose (PODhed), in mg/kg/day, using the updated Verner model (described in Section
4.1.3.2). This was performed as described for the Chuetal. {,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 Pup Survival, Kunming Mice, Fi males and females (PND 21), CaVg_pup_gest_iact
{Song, 2018, 5079725}
Decreased mean response of number of offspring survival at weaning on PND 21 was observed
in Fl male and female Kunming mice. Continuous models were used to fit dose-response data. A
BMR of a change in the mean equal to 0.5 standard deviations from the control mean was
selected for POD derivation was chosen per EPA's Benchmark Dose Technical Guidance {U.S.
EPA, 2012, 1239433} (Section 4.1.2) and a BMR of a change in the mean equal to 0.1 standard
deviations from the control mean was provided for comparison purposes because decreased pup
survival is a severe, frank effect {U.S. EPA, 2012, 1239433}(see Appendix E.2, {U.S. EPA,
2024, 11414343}). The Cavg,PuP,gest,iact internal dose metric was selected for this model since an
average concentration metric is expected to better correlate with this developmental effect that
may have resulted from exposure during gestation or lactation (Section 4.1.3.1.3). 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 4.1.3.2). For this, the model
was run starting at the birth of the mother, with constant exposure relative to body weight.
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 males and females. The lower HED was
selected to be more health protective.
Decreased Pup Survival, CD-I Mice, Fi males and females (PND 0), Cavg_pup_gest {Lau, 2006,
1276159}
Decreased mean response of pup survival was observed in Fl male and female CD-I mice at
PND 0. Continuous models were used to fit dose-response data. A BMR of a change in the mean
equal to 0.5 standard deviations from the control mean was chosen per EPA's Benchmark Dose
Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2) and a BMR of a change in the
mean equal to 0.1 standard deviations from the control mean was provided for comparison
purposes because decreased pup survival is a severe, frank effect {U.S. EPA, 2012, 1239433}
(see Appendix E.2, {U.S. EPA, 2024, 11414343}). The C avg,pup,gest internal dose metric was
selected for this model since an average concentration metric is expected to better correlate with
4-52
-------
APRIL 2024
this developmental effect that may have resulted from exposure any time during gestation
(Section 4.1.3.1.3). The tests for constant and nonconstant variance failed therefore 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 4.1.3.2). For this, the model
was run starting at the birth of the mother, with constant exposure relative to body weight.
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. This metric was independent of the sex of the
child in the model.
Decreased Pup Survival, CD-I Mice, Fi males and females (PND 23), Cavg_pup_gestiact{Lau,
2006,1276159}
Decreased mean response of pup survival was observed in Fl male and female CD-I mice at
PND 23. Continuous models were used to fit dose-response data. A BMR of a change in the
mean equal to 0.5 standard deviations from the control mean was chosen per EPA's Benchmark
Dose Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2) and a BMR of a change in
the mean equal to 0.1 standard deviations from the control mean was provided for comparison
purposes because decreased pup survival is a severe, frank effect (U.S. EPA, 2012, 1239433}
(see Appendix E.2, (U.S. EPA, 2024, 11414343}). The C avg,pup,gest_lact internal dose metric was
selected for this model since an average concentration metric is expected to better correlate with
this developmental effect that may have resulted from exposure during gestation or lactation
(Section 4.1.3.1.3). The tests for constant and nonconstant variance failed therefore 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 4.1.3.2). For this, the model
was run starting at the birth of the mother, with constant exposure relative to body weight.
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 males and females. The lower HED was
selected to be more health protective.
Decreased Fetal Body Weight, Kunming Mice, Fi males and females (GD 18), Cavg_pup_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
chosen per EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (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 Appendix E.2, {U.S. EPA, 2024, 11414343}). The
Cavg,pup,gest internal dose metric was selected for this model since an average concentration metric
is expected to better correlate with this developmental effect that may have resulted from
4-53
-------
APRIL 2024
exposure any time during gestation (Section 4.1.3.1.3). The tests for constant and nonconstant
variance failed therefore 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 4.1.2). For
this, the model was run starting at the birth of the mother, with constant exposure relative to
body weight. 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. This metric was independent of
the sex of the child in the model.
Decreased Pup Body Weight (relative to litter), CD-I Mice, Fi males and females (PND 23),
Cavg pup gest lact {Lau,2006,1276159}
Decreased mean response of pup body weight was observed in Fi male and female CD-I mice at
PND 23. Continuous models were used to fit dose-response data. A BMR of 5% extra risk was
chosen per EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} (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 Appendix E.2, (U.S. EPA, 2024, 11414343}). The
Cavg,pup,gestjact internal dose metric was selected for this model since an average concentration
metric is expected to better correlate with this developmental effect that may have resulted from
exposure during gestation or lactation (Section 4.1.3.1.3). 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 4.1.3.2).
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 males
and females. The lower HED was selected to be more health protective.
Delayed Time to Eye Opening, CD-I Mice, Fi males and females (PND 14 - PND 18),
Cavg pup gest lact {Lau,2006,1276159}
Decreased mean response of time to eye opening was observed in FI male and female CD-I
mice. Continuous models were used to fit dose-response data. BMR of a change in the mean
equal to 0.5 standard deviations from the control mean was chosen per EPA's Benchmark Dose
Technical Guidance {U.S. EPA, 2012, 1239433} (Section 4.1.2), and a BMR of a change in the
mean equal to 1 standard deviations from the control mean was provided for comparison
purposes (see Appendix E.2, {U.S. EPA, 2024, 11414343}). The C avg,pup,gest_lact internal dose
metric was selected for this model since an average concentration metric is expected to better
correlate with this developmental effect that may have resulted from exposure during gestation
or lactation (Section 4.1.3.1.3). 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
4-54
-------
APRIL 2024
lactation, was converted to an external dose (PODhed), in mg/kg/day, using the updated Verner
model (described in Section 4.1.3.2). For this, the model was run starting at the birth of the
mother, with constant exposure relative to body weight. 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 entire 1-year breastfeeding period because the lactational duration in
humans that equates to time to eye opening in rodents is unknown. Additionally, there is
currently no mechanistic information to identify a specific window of susceptibility in lactation
for this endpoint. 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 different growth curves specific to male and
female children were used in the model, the model predicted slightly (less than 5%) different
serum concentrations for each sex. The lower HED was selected to be more health protective.
4.1.5 Derivation of Candidate Chronic Oral Noncancer
Reference Doses (RfDs)
Though multiple candidate 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. For epidemiological studies, similar to
the discussion of study selection factors in Sections 4 and 4.1.1, EPA critically considered
attributes for each PODhed including timing of endpoint collection or measurement,
uncertainties associated with modeling (see Appendix E {U.S. EPA, 2024, 11414343} and Table
4-8), and consideration of confounding. For animal toxicological studies, attributes considered
included study confidence (i.e., high confidence studies were prioritized over medium confidence
studies), amenability to benchmark dose modeling, study design, sensitive lifestages, and health
effects observed after exposure in the lower dose range among the animal toxicological studies.
As described in the subsections below, this examination of epidemiological and toxicological
studies led to the exclusion of a number of studies from consideration for candidate RfD
derivation. Health outcome- and study-specific considerations are discussed in Sections 4.1.5.1
(Hepatic), 4.1.5.2 (Immune), 4.1.5.3 (Cardiovascular), and 4.1.5.4 (Developmental).
Once studies and their corresponding PODheds were prioritized for candidate RfD derivation,
EPA applied uncertainty factors (UFs) according to methods described in EPA's Review of the
Reference Dose and Reference Concentration Processes {U.S. EPA, 2002, 88824}.
Considerations for individual UFs differed between epidemiological and animal toxicological
studies and are further described in Section 4.1.5.5. Presentation of the candidate RfDs for each
health outcome is provided in Section 4.1.5.6.
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.
Additionally, these studies had many study strengths including sufficient study sensitivity and
sound methodological approaches, analysis, and design, as well as no evidence of bias. The three
4-55
-------
APRIL 2024
studies reported analyses examining different forms of confounding factors and consideration of
cumulative PFOA exposure {Darrow, 2016, 3749173}, sensitivity analyses excluding
participants with lifestyle characteristics (e.g., excluding smokers, drinkers, medicine takers)
impacting outcome assessment {Nian, 2019, 5080307}, and nonlinear exposure-response
relationships {Gallo, 2012, 1276142}. All three of these studies provided the necessary data for
modeling.
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}. Increased liver necrosis from NTP {, 2020, 7330145} was selected
for candidate RfD derivation over the effect of increased hepatocyte single cell death due to the
increased biological severity of the former endpoint. Increased liver necrosis additionally
resulted in a more protective PODhed.
4.1.5.2 Immune Effects
Two medium confidence epidemiological studies were carried forward for candidate RfD
determination {Budtz-Jorgensen, 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-Jorgensen 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 potential increased risk of bias due to pregnancy-related
hemodynamic effects. EPA also derived candidate RfDs for both tetanus and diphtheria vaccine
responses from Timmerman et al. {, 2021, 9416315} for comparison to a second population of
children. In total, four immunological PODheds from two epidemiological studies were carried
forward for candidate RfD derivation.
One medium confidence animal toxicological study was carried forward for candidate RfD
determination {Dewitt, 2008, 1290826}. The PODhed from Study 1 was selected over Study 2
because the former was amenable to benchmark dose modeling and had a PODhed based on a
BMDL, the preferred POD for animal toxicological studies {U.S. EPA, 2012, 1239433; U.S.
EPA, 2022, 10367891}. 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
PODhed based on Dewitt et al. {, 2008, 1290826} for candidate RfD derivation. The candidate
RfD derived from Dewitt et al. {, 2008, 1290826} is expected to be protective of the immune
effects observed in Loveless et al. {, 2008, 988599}.
4-56
-------
APRIL 2024
4.1.5.3 Cordiovosculor 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}
excluded individuals who were prescribed cholesterol medication, minimizing concerns
surrounding confounding due to the medical intervention altering serum total cholesterol levels.
This is in contrast to Lin et al. {, 2019, 5187597} which did not control for individuals
prescribed cholesterol medication and was therefore excluded from further consideration.
Modeling of both Dong et al. {, 2019, 5080195} and Steenland et al. {, 2009, 1291109} resulted
in PODheds with minimal risk of bias, representing both the general population and a high-
exposure community, respectively and thus, were both 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 decreased birth weight {Sagiv, 2018, 4238410; Wikstrom,
2020, 6311677}. Of the five epidemiological studies for which PODheds were derived, Sagiv et
al. {,2018,4238410} and Wikstrom et al. {,2020, 6311677} assessed maternal PFOA serum
concentrations primarily in the first trimester, minimizing concerns surrounding bias due to
pregnancy-related hemodynamic effects. Although Wikstrom et al. {, 2020, 6311677} collected
approximately 4% of samples during early weeks of the second trimester, sensitivity analyses
showed no differences when trimester two samples were excluded. Additionally, these two
studies had many study strengths including sufficient study sensitivity and sound methodological
approaches, analysis, and design, as well as no evidence of bias and reflected two different study
populations. Therefore, both studies were considered further for candidate RfD derivation. The
other three studies assessed PFOA concentrations in either umbilical cord blood or primarily
during the second or third trimesters, increasing the uncertainty associated with the derived
PODheds due to potential pregnancy-related hemodynamic effects, and as a result, were
excluded from consideration for candidate RfD derivation {Chu, 2020, 6315711; Govarts, 2016,
3230364; Starling, 2017, 3858473}.
Two medium confidence animal toxicological studies representing two endpoints, decreased pup
survival and delayed time to eye opening, were carried forward for candidate RfD determination
{Lau, 2006, 1276159; Song, 2018, 5079725}. These two datasets were amenable to benchmark
dose modeling and had PODheds based on BMDLs, the preferred POD for animal toxicological
studies {U.S. EPA, 2012, 1239433; U.S. EPA, 2022, 10367891}. In contrast, the endpoints of
decreased fetal body weight derived from data published by Li et al. {, 2018, 5084746} and
decreased pup survival and decreased pup weight derived from data published by Lau et al. {,
2006, 1276159} were not amenable to BMD modeling and had NOAELs as the basis of the
PODheds. Therefore, these PODheds were excluded from further consideration for candidate
RfD derivation. As the delayed time to eye opening and decreased pup survival endpoints
reported by Lau et al. {, 2006, 1276159} and Song et al. {, 2018, 5079725}, respectively,
encompassed sensitive populations (i.e., fetuses and pups) and different effects in two different
strains of mice, these two PODheds were considered further for candidate RfD derivation. These
two endpoints appear to be more sensitive (i.e., have lower PODheds) than the effects reported
by Li {, 2018, 5084746} and Lau {, 2006, 1276159}.
4-57
-------
APRIL 2024
4.1.5.5 Application of Uncertainty Factors
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 lifestage 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 lifestages 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; 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 = 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, lifestage, 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. This
was the case for all epidemiological endpoints and studies advanced for candidate RfD
derivation.
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
4-58
-------
APRIL 2024
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
lifestage 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}. AUFs of 1 was also
applied to the immune endpoints observed in children and adolescents because exposure is
assumed to occur from gestation through childhood, when the response variable was measured.
There is uncertainty regarding the critical window of exposure that results in these immune
effects in children and adolescents. Therefore, EPA expects that any exposure during this period
of development has the potential to impact this response {U.S. EPA, 1991, 732120}. According
to the WHO/International Programme on Chemical Safety (IPCS) Immunotoxicity Guidance for
Risk Assessment, developmental immunotoxicity is assessed during the prenatal, neonatal,
juvenile and adolescent life stages because immune system development occurs throughout these
life stages and should be viewed differently in part due to increased susceptibility compared with
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 large number of medium and high confidence
epidemiological studies 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.
The composite 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 composite UF 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
lor 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
4-59
-------
APRIL 2024
UF
Value
Justification
recognized as a susceptible lifestage 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 lifestages 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; LTFa= interspecies uncertainty factor; UFd = database uncertainty factor; LTFh= 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 threefold 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, lifestage, and health status) and extrinsic (lifestyle) factors can 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
was used as the basis for the PODhed derivation. BMDLs were available for all animal
toxicological endpoints and studies advanced for candidate RfD derivation.
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 lifestage 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 large number of medium and high confidence
epidemiological studies which was used quantitatively in this assessment. Typically, the specific
4-60
-------
APRIL 2024
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.
In summary, the composite UF that was applied to candidate RfDs derived from all of the
epidemiological studies were the same value (UFc = 10) (Table 4-9). The composite UF that was
applied to candidate RfDs derived from animal toxicological studies was either UFc = 30 or 300
(Table 4-10). In all of these cases, the total uncertainty is well below the maximum
recommended UFc = 3,000 {U.S. EPA, 2002, 88824}.
4.1.5.6 Candidate RfDs
Table 4-11 shows the UFs applied to each candidate study to subsequently derive the candidate
RfDs.
4-61
-------
APRIL 2024
Table 4-11. Candidate Reference Doses (RfDs)
Endpoint
Study,
Confidence
Strain/Species/
Sex/Age
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 >
< 10~7
1
10
1
1
1
10
3.05 x 10 8 = 3 x 10~8
tetanus antibody
Grandjean {, 2018,
female, PFOA
concentration in
508363 l}b
concentrations at age 5
children
Medium
and antibody
concentrations at age 7
Timmerman et al. {,
Human, male and
3.34 >
< 10"
1
10
1
1
1
10
3.34 x 10 8 = 3 x 10~8
2021,9416315}
female, PFOA and
Medium
antibody concentrations
at ages 7-12
Decreased serum anti-
Budtz-Jorgensen and
Human, male and
2.92 >
< 10~7
1
10
1
1
1
10
2.92 x 10 8 = 3 x 10 8
diphtheria antibody
concentration in
Grandjean {, 2018,
508363 l}b
female, PFOA
concentrations at age 5
children
Medium
and antibody
concentrations at age 7
Timmerman et al. {,
Human, male and
2.20 >
< 10 "
1
10
1
1
1
10
2.20 x 10~8 = 2 x 10~8
2021,9416315}
female, PFOA and
Medium
antibody concentrations
at ages 7-12
Decreased IgM
Dewitt et al. {, 2008,
Mouse, female, adults,
2.18 >
< 10~3
3
10
10
1
1
300
7.27 x 10~6 = 7 x 10~6
response to SRBC
1290826}
Medium
study 1
Developmental Effects
Decreased Birth
Sagiv et al. {,2018,
Human, male and
1.21 >
< 10~6
1
10
1
1
1
10
1.21 x 10 7 = 1 x 10 7
Weight
4238410}
High
female, PFOA
concentrations in first
and second trimesters
Wikstrom et al. {,
Human, male and
2.92 >
< 10~7
1
10
1
1
1
10
2.92 x 10~8 = 3 x 10~8
2020,6311677}
female, PFOA
High
concentrations in first
and second trimesters
Decreased Offspring
Song et al. {, 2018,
Kunming Mice, Fi
6.40 >
< 10~4
3
10
1
1
1
30
2.13 x 10~5 = 2 x 10~5
Survival
5079725}
males and females
4-62
-------
APRIL 2024
Endpoint
Study,
Confidence
Strain/Species/
Sex/Age
PODhed
(mg/kg/day)
UFa
UFh
UFs
UFl
UFd
UFc
Candidate RfDa
(mg/kg/day)
Medium
Delayed Time to Eye
Opening
Lau et al. {, 2006,
1276159}
Medium
CD-I Mice, Fi males
and females (PND 14 -
PND 18)
4.17 >
< 10~4
3
10
1
1
1
30
1.39 x 10 5 = 1 x 10~5
Cardiovascular Effects
Increased Serum Total
Cholesterol
Dong et al. {, 2019,
5080195}
Medium
Human, male and
female, age 20-80
2.75 >
< 10~7
1
10
1
1
1
10
2.75 x 10~8 = 3 x 10~8
Steenland et al. {, 2009, Human, male and
1291109} female, age 18 and
Medium older
5.10 >
< 10 "
1
10
1
1
1
10
5.10 x 10~8 = 5 x 10~8
Hepatic Effects
Increased Serum ALT
Gallo etal. {,2012,
1276142}
Medium
Human, female, age 18
and older
2.15 >
< 10~6
1
10
1
1
1
10
2.15 x 10~7 = 2 x m1
Darrow et al. {, 2016,
3749173}
Medium
Human, female, age 18
and older
7.92 >
< 10~6
1
10
1
1
1
10
7.92 x 10 7 = 8 x 10 7
Nian et al. {, 2019,
5080307}
Medium
Human, female, age 22
and older
4.51 >
< 10~7
1
10
1
1
1
10
4.51 x 10~8 = 5 x 10~8
Necrosis
NTP {, 2020, 7330145} Sprague-Dawley rats,
High perinatal and
postweaning (2-year),
male
3.23 >
< 10~3
3
10
1
1
1
30
1.08 x 10~4= 1 x io~4
Notes: ALT = alanine aminotransferase; NTP = National Toxicology Program; PODhed = point-of-departure human equivalence dose; RiD = reference dose; SRBC = sheep red
blood cells; UFa = interspecies uncertainty factor; UFh = intraspecies uncertainty factor; UFs = subchronic-to-chronic extrapolation uncertainty factor; UFl = extrapolation from a
LOAEL-to-NOAEL uncertainty factor; UFd = database uncertainty factor; UFc = composite uncertainty factor.
aRfDs were rounded to one significant figure.
b Supported by Grandjean et al. {,2012, 1248827}, Grandjeanetal. {, 2017, 3858518}, and Grandjean etal. {,2017,4239492}.
4-63
-------
APRIL 2024
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 noncancer 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
either the adverse effect(s) observed or 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 Sections 5.1, 5.6, and 5.7), 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, in part, the higher magnitude of candidate RfDs derived from animal toxicological
studies compared to the candidate RfDs derived from epidemiological studies. Moreover, the
human epidemiological studies also have greater relevance to human exposure than animal
toxicological studies 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, 10367891}. 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 5.2 for further comparisons between toxicity values derived
from epidemiological and animal toxicological studies.
4-64
-------
APRIL 2024
Decreased
serum
anti-tetanus
antibody
concentration
in children
Decreased
serum
anti-diptheria
antibody
concentration
in children
Decreased
IgM response
to SRBC
Timmerman, 2021, 9416315;
Medium confidence
Budtz-J0rgensen, 2018, 5083631;
Medium confidence
Timmerman, 2021, 9416315;
Medium confidence
Budtz-Jorgensen, 2018, 5083631;
Medium confidence
Dewitt, 2008, 1290826;
Medium confidence
-o
¦o
-o
-o
Human Animal
RfD PODHED
o
UF
-o
Decreased
Birth Weight
Delayed Time
to Eye
Opening
Decreased
Offspring
Survival
Sagiv, 2018, 4238410;
High confidence
Wikstrom, 2020, 6311677;
High confidence
Lau, 2006, 1276159;
Medium confidence
Song, 2018, 5079725;
Medium confidence
o
¦o
-o
-o
Increased
Serum Total
Cholesterol
Dong, 2019, 5080195;
Medium confidence
Steenland, 2009, 1291109;
Medium confidence
-o
-o
Gallo, 2012, 1276142;
Medium confidence
o
Increased
Serum ALT
Darrow, 2016, 3749173;
Medium confidence
-o
Nian, 2019, 5080307;
Medium confidence
¦o
Necrosis
NTP, 2020, 7330145;
High confidence
¦o
10'3
10-2
10-8 10-7 10-6 10-5 10-4
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
4-65
-------
APRIL 2024
As described in the subsections below, EPA selected amongst the candidate RfDs to identify an
RfD representative of each of the four priority 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
At least two candidate RfDs were derived from epidemiological studies for each of the four
prioritized noncancer health outcomes. EPA considered several factors when selecting health
outcome-specific RfDs, including relevance of exposure or population characteristics to the
general population, potential confounding factors, and characteristics of the modeled data. Health
outcome- and study-specific considerations are discussed in Sections 4.1.6.1.1 (Hepatic),
4.1.6.1.2 (Immune), 4.1.6.1.3 (Cardiovascular), and 4.1.6.1.4 (Developmental), below.
4.1.6.1.1 Hepatic Effects
Three medium confidence epidemiological studies were selected for candidate RfD derivation for
the endpoint of increased 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, 1276142} reported measured serum ALT levels, unlike Darrow et al.
{, 2016, 3749173} 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. Due to these factors, the
candidate RfD derived from Darrow et al. {, 2016, 3749173} was excluded from further
consideration as the health outcome-specific RfD for hepatic effects.
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. Both Nian et al. {,
2019, 5080307} and Gallo et al. {, 2012, 1276142} provided measured PFOA serum
concentrations and a measure of serum ALT levels. However, the Gallo et al. {, 2012, 1276142}
study was conducted in a community exposed predominately to PFOA, whereas Nian et al. {,
2019, 5080307} was conducted in a community exposed predominately to PFOS. The candidate
RfD derived from Gallo et al. {, 2012, 1276142} was ultimately selected as the health outcome-
specific RfD due to reduced risk of bias related to potential confounding from other PFAS in this
population. The resulting health outcome-specific RfD is 2 x 10 7 mg/kg/day (Figure 4-5).
4.1.6.1.2 Immune Effects
Candidate RfDs were derived from two medium confidence epidemiological studies for the
endpoint of decreased antibody production in response to various vaccinations in children
{Budtz-Jorgensen, 2018, 5083631; Timmerman, 2021, 9416315}. Candidate RfDs were derived
from Timmerman et al. {, 2021, 9416315} were considered lower confidence candidate RfDs
than those derived from Budtz-Jorgensen and Grandjean {, 2018, 5083631}. PODHEDs derived
from Timmerman et al. {, 2021, 9416315} were considered to have increased uncertainty
compared with Budtz-Jorgensen and Grandjean {, 2018, 5083631} due to two features of the
4-66
-------
APRIL 2024
latter study that strengthen the confidence in the PODHEDs: 1) the analyses considered co-
exposures of other PFAS (i.e., PFOS); and 2) the response reported by this study was more
precise in that it reached statistical significance. Therefore, the candidate RfDs from Timmerman
et al. {, 2021, 9416315} were not considered for selection as the health outcome-specific RfD.
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-Jorgensen and Grandjean {, 2018, 5083631} were
ultimately selected for the immune outcome as they are the same value and have no
distinguishing qualitative (e.g., strength of evidence) or quantitative (e.g., model fit)
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.
4.1.6.1.3 Cardiovascular Effects
Two medium confidence epidemiological studies were selected for candidate 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 candidate 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 selected for candidate RfD derivation for the
endpoint of decreased birth weight {Sagiv, 2018, 4238410; Wikstrom, 2020, 6311677}. These
candidate studies assessed maternal PFOA serum concentrations primarily 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. Between these two studies, PFOA exposure concentrations observed in
Wikstrom et al. {, 2020, 6311677} are more comparable to current exposure levels in the U.S.
general population and therefore may be more relevant to the general population than the
candidate RfD derived from Sagiv et al. {, 2018, 4238410}. Additionally, the BMDL derived
from Wikstrom et al. {, 2020, 6311677} was based on a statistically significant regression
parameter. For these reasons, the RfD for decreased birth weight from Wikstrom et al. {, 2020,
4-67
-------
APRIL 2024
6311677} was selected as the basis for the health outcome-specific RfD for developmental
effects. The resulting health outcome-specific RfD is 3 « 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.
Immune
Anti-tetanus
antibody
response
Anti-diphtheria
antibody
response
Budtz-Jorgensen
and Grandjean -
(2018, 5083631)
sTimmerman et al.
(2021, 9416315) "
Budtz-Jorgensen
and Grandjean -
(2018, 5083631)
^Timmerman et al.
(2021, 9416315) "
3 x 10e
3 x 106
3x 10 s
2 x 106
3 x 10"B
Developmental
Decreased
birth weight
Sagiv et al.
(2018, 4238410) '
Wikstrom et al.
(2020, 6311677) '
1 x 10"7
3x 10-8
3x 10-8
3x 108
Health Outcome
Endpoint
Study
3 x 10~e
5x 10e
2 x 107
8 x 107
5 x 10e
Candidate RfD
(mg/kg/day)
3 x 108
2 x 107
Health Outcome
Specific RfD
(mg/kg/day)
Overall RfD
(mg/kg/day)
Figure 4-5. Schematic Depicting Selection of the Overall RfD for PFOA
RID = 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 Nonconcer 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 se 10 x mg/kg/day). Therefore, EPA has selected an overall RfD for PFOA of
3 10 x mg/kg/day. The immune, developmental, and cardiovascular RfDs based on endpoints
4-68
-------
APRIL 2024
of decreased anti-tetanus and anti-diphtheria antibody concentrations in children, decreased 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 (e.g., infants,
children; see Section 5.8), 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 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 evidence indicates 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 therefore, the overall RfD is expected to be protective of all other
noncancer health effects in humans.
4.2 Cancer
As described in the introduction of Section 4, there is evidence from both epidemiological and
animal toxicological studies that oral PFOA exposure may result in adverse health effects across
many health outcomes, including cancer (Section 3.5). In Section 3.5.5, EPA concluded that
PFOA is Likely to be Carcinogenic to Ramans in accordance with the Guidelines for Carcinogen
Risk Assessment {U.S. EPA, 2005, 6324329}. Therefore, the quantification of cancer effects was
prioritized along with the four noncancer health outcomes that are described in Section 4.1. EPA
considered only high or medium confidence human and animal toxicological studies for CSF
derivation.
4.2.1 Study and End point Selection
Human studies selected for CSF derivation reported all necessary analytical information (e.g.,
exposure distribution or variance) for the outcome of interest (any cancer). If available, high and
medium confidence studies with exposures levels near the range of typical environmental human
exposures, especially exposure levels comparable to human exposure in the general population,
were preferred over studies reporting considerably higher exposure levels. Exposure levels near
the typical range of environmental human exposure can facilitate extrapolation to exposure levels
that may be more relevant to the U.S. general population. Additionally, the most recent and
comprehensive publication on a single study population was preferred over prior publications on
the same or portions of the same population (e.g., selection of Vieira et al. {, 2013, 2919154}
over other C8 Health Project studies (see Section 4.2.1.1)).
Preferred animal toxicological studies consisted of medium and high confidence studies with
chronic exposure durations to capture potential latency of cancer effects. Studies with exposure
durations during development (e.g., gestation) were also considered informative for assessing
potential early lifestage susceptibility to cancer (see Section 4.2.4). Studies encompassing lower
dose ranges were also preferred. These types of animal toxicological studies increase the
confidence in the CSF relative to other animal toxicological studies because they are based on
data with relatively low risk of bias, have sufficient study designs to observe the critical effects,
4-69
-------
APRIL 2024
and are associated with less uncertainty related to low-dose and exposure duration
extrapolations.
4.2.1.1 Epidemiological Studies
The available evidence indicates that there is an increase in risk for kidney or Renal cell
carcinoma (RCC) and testicular cancers with PFOA exposure {Shearer, 2021, 7161466; Chang,
2014, 2850282; Bartell, 2021, 7643457; Barry, 2013, 2850946; Vieira, 2013, 2919154;
Steenland, 2012, 2919168; Raleigh, 2014, 2850270}. Results are most consistent for kidney
cancer in adults based on a nested case-control study {Shearer, 2021, 7161466}, two C8 Health
Project studies {Barry, 2013, 2850946; Vieira, 2013, 2919154}, two occupational mortality
studies {Steenland, 2012, 2919168; Raleigh, 2014, 2850270}, and 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}. Therefore, the
endpoint of kidney cancer was selected for CSF derivation.
Testicular cancer was identified as supporting evidence for carcinogenicity in humans in the
2016 PFOAHESD {U.S. EPA, 2016, 3603279}. However, additional epidemiological studies
examining risk of testicular cancer were not identified in the updated literature search and only
two studies in the same high-exposure community (C8 Health Project) reported this association
{Barry, 2013, 2850946; Vieira, 2013, 2919154}. Therefore, the endpoint of testicular cancer in
humans was not selected for dose-response modeling. Evidence was mixed or limited for other
cancer sites (e.g., breast, liver cancers), which were not considered further.
Two studies reporting associations between kidney cancer and PFOA serum concentrations,
Shearer et al. {, 2021, 7161466} and Vieira et al. {, 2013, 2919154}, were selected for dose-
response modeling. Shearer et al. {, 2021, 7161466} was selected because it is a well-conducted,
U.S.-based multicenter case-control study in the general population reporting a relatively large
number of cases (N = 326). Median PFOA levels in controls was 5.0 ng/mL, comparable with
4.8 ng/mL in adults 60 and over from NHANES 1999-2000. Additionally, 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. There was also 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 versus lowest quartile (OR = 2.63, 95% CI:
1.33, 5.2) and 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 both men and women, separately.
EPA also selected the C8 Health Project study {Vieira, 2013, 2919154} for dose-response
modeling. The Vieira et al. {, 2013, 2919154} study was a cancer registry-based case-control
conducted in 13 counties in Ohio and West Virginia that surround the DuPont Washington
Works PFOA facility (C8 study area). Analyses were adjusted for several factors including age,
sex, diagnosis year, smoking status (current, past, unknown, or never), and insurance provider
(government-insured Medicaid, uninsured, unknown, or privately insured). There was a
statistically significant increase in the odds of kidney cancer when comparing both the high
(OR = 2.0; 95% CI: 1.3, 3.2) and the very high (OR = 2.0; 95% CI: 1.0, 3.9) exposure categories
to the unexposed reference population. Vieira et al. {, 2013, 2919154} was selected for modeling
over Barry et al. {, 2013, 2850946}, the populations of which likely overlapped, because Barry
4-70
-------
APRIL 2024
et al. {, 2013, 2850946} did not report the necessary exposure measurements for CSF
calculation. Specifically, exposure levels were reported separately for the community
participants and workers, but not for the overall study population and therefore, CSF calculations
were not feasible. Vieira (2013, 2919154) included the most complete and up-to-date data from
this population, including all information needed for CSF derivation.
The high-exposure occupational study by Steenland and Woskie {, 2012, 2919168} was not
selected for dose-response analysis because it was limited by the small number of observed
cancer cases (six kidney cancer deaths) and the exposure levels reported in the study population
(average annual serum concentration of 350 ng/mL) are less comparable to the U.S. general
population than the levels reported by Shearer et al. {, 2021, 7161466} and Vieira et al. {, 2013,
2919154}. The study by Raleigh et al. {, 2014, 2850270} was also not selected prioritized
because of the concerns of exposure assessment methods (i.e., estimated air PFOA
concentrations rather than biomonitoring data) and study quality (i.e., relatively small numbers
of cases and lack of information regarding adjustment of risk factors for kidney cancer such as
smoking status and BMI).
4.2.1.2 Animal Toxicologicol Studies
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}. Combined, at least two of the three studies report increased
incidences each of hepatic, testicular, and pancreatic neoplastic lesions. Increased incidences of
neoplastic lesions were primarily observed in male rats, though results in females, particularly
the reports of rare tumor types (i.e., pancreatic acinar cell adenomas and adenocarcinomas), are
supportive of potential carcinogenicity of PFOA. Additionally, 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). Uterine adenocarcinomas were not selected 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 selected for CSF derivation because the incidences were not
observed by NTP {, 2020, 7330145}. As these results are inconclusive and there was increased
magnitude of hepatic and pancreatic tumor incidences in males, likely due to the increased
sensitivity of male rats resulting from toxicokinetic differences between the sexes (see Section
3.3.1), quantitative analyses were focused on males rather than females.
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 and adenocarcinomas in males in all
treatment groups compared with their respective controls. These rare tumor types were also
observed in female rats in 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
4-71
-------
APRIL 2024
addition, bothNTP {, 2020, 7330145} and Biegel et al. {, 2001, 673581} reported dose-
dependent increases in the incidence of liver adenomas in male rats. NTP {, 2020, 7330145} also
reported several male rats with hepatocellular carcinomas in the highest dose group
(300/80 ppm). Butenhoff et al. {, 2012, 2919192} additionally reported incidences of
hepatocellular carcinomas in male rats from every treatment group, including controls, and
female rats in the highest dose group. Given the consistency across the three available studies,
the observation of malignant pancreatic and hepatic tumors, and the site concordance between
the testicular tumors in rats and humans, tumors from all three sites (i.e., liver, pancreas, testes)
were selected for CSF derivation.
In further evaluation of the studies, Biegel et al. {, 2001, 673581} was not considered for dose-
response modeling because it is a single-dose study. Therefore, NTP {, 2020, 7330145} was
selected for candidate CSF derivation for the PACTs and hepatocellular tumors and Butenhoff et
al. {, 2012, 2919192} was selected for candidate CSF derivation for LCTs.
4.2.2 Candidate CSF Derivation
4.2.2.1 Epidemiological Studies
EPA calculated CSFs for RCC from Shearer et al. {, 2021, 7161466} and for kidney cancer from
Vieira et al. {, 2013, 2919154} 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 {U.S. EPA, 2024, 11414343}. The underlying model involves a linear
regression between PFOA exposure and cancer relative risk used to estimate the dose-response
between PFOA and RCC or kidney cancer 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 (or models were adjusted for age in Vieira et al. {, 2013,
2919154}), 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. For
Shearer et al. {, 2021, 7161466}, 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. For Vieira et al. {, 2013, 2919154}, the lifetime kidney cancer of Ro of
0.0202 was used, and the model fit was better when the highest exposure level was excluded.
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).
4-72
-------
APRIL 2024
The results of the modeling and the candidate CSFs derived are presented in Table 4-12.
Table 4-12. Candidate Cancer Slope Factors Based on Epidemiological Data
Strain/
Species/Sex/
Age
Internal CSF -
CSF - Increase in
Tumor Type
Reference,
POD Type,
Increase in Cancer
Cancer Risk per
Confidence
Model
Risk per 1 ng/mL
1 ng/(kg*d) Increase
Serum Increase
in Dose
Renal cell
Shearer et al. {,
Human, male
CSF serum in
3.52 x 10-3
0.0293 (ng/kg/d) 1
carcinoma
2021,
and female 55-
adults (per
(ng/mL)"1
(RCC)
7161466}
Medium
74 yr
ng/mL of
serum PFOA);
upper limit of
the 95% CI
(see Appendix {U.S.
EPA, 2024,
11414343 }for
additional detail)
Kidney cancer
Vieira et al. {,
Human, male
CSF serum in
4.81 x 10-4
0.00401 (ng/kg/d)"1
2013,
and female.
adults (per
(ng/mL)"1
2919154}
median age 67
ng/mL of
(see Appendix {U.S.
Medium
years
serum PFOA);
upper limit of
the 95% CI,
highest
exposure group
excluded
EPA, 2024,
11414343 }for
additional detail)
Notes: CI = Confidence Interval; CSF = cancer slope factor; POD = point of departure.
EPA's Benchmark Dose Technical Guidance {U.S. EPA, 2012, 1239433} notes that approaches
for combining datasets in dose-response modeling may be used when datasets are statistically
and biologically compatible. This type of approach was utilized in the CalEPA analysis of
kidney cancer {CalEPA, 2021, 9416932}. EPA therefore considered this approach for candidate
CSF derivation and performed a sensitivity analysis to derive a CSFsemm based on the pooled data
from Shearer et al. {, 2021, 7161466} and Vieira et al. {, 2013, 2919154}. These analyses are
presented in Appendix E {U.S. EPA, 2024, 11414343}. However, EPA identified several
considerable differences between the two studies, including the outcome measured (RCC versus
any kidney cancer) and the exposure metric (measured vs. modeled serum PFOA), among others.
Additionally, the slope of the dose-response relationship was very different between the two
studies (0.0981, 95% CI: 0.0025, 0.1937 vs. 0.0122, 95% CI: 0.0006, 0.0238 from Shearer et al.
{, 2021, 7161466} and Vieira et al. {, 2013, 2919154}, respectively). Given these differences,
EPA determined that these two studies are not statistically or biologically comparable and
therefore, they were not pooled for dose-response modeling {U.S. EPA, 2012, 1239433}.
4.2.2.2 Animal Toxicologicol Studies
In the 2016 PFOA HESD {U.S. EPA, 2016, 3603279}, EPA derived a CSF based on LCTs
reported by Butenhoff et al. {, 2012, 2919192}. 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. EPA has
reanalyzed 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 and an updated version of
4-73
-------
APRIL 2024
BMDS (Version 3.2). These modeling results are described in Appendix E {U.S. EPA, 2024,
11414343}. A BMR of 10% was modeled because it is the recommended standard level for
comparison across chemicals {U.S. EPA, 2012, 1239433}. However, for this dataset, a BMR of
10% resulted in a BMDL value higher than the lowest dose tested (see Appendix E {U.S. EPA,
2024, 11414343}). Therefore, a BMR of 4% was ultimately selected because it was
representative of the low end of the observed response range within the study results {U.S. EPA,
2012, 1239433}.
EPA also derived candidate CSFs for the tumor types observed in the NTP study that provide
further evidence of carcinogenic activity of PFOA in male Hsd:Sprague-Dawley rats:
hepatocellular neoplasms (hepatocellular adenomas and carcinomas) and acinar cell neoplasms
(adenomas and adenocarcinomas) of the pancreas {NTP, 2020, 7330145} (Table 4-13). 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}, EPA selected the AUC averaged over
the study duration (AUCavg), equivalent to the mean serum concentration over the duration of the
study, as the dose metric for modeling cancer endpoints. This is consistent with the Guidelines
for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329} and the IRIS Handbook {U.S.
EPA, 2022, 10367891}, which recommend the cumulative dose received over a lifetime as the
measure of exposure to a carcinogen when modeling chronic cancer effects. 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. As discussed previously in Section 4.1.3.1.3, 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
(postweaning) and dividing by 2 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, which would result in serum concentration equal to the POD at steady state. The
candidate CSF is then calculated by dividing the BMR by the PODhed. These modeling results
are described further in the Appendix {U.S. EPA, 2024, 11414343}.
4-74
-------
APRIL 2024
Table 4-13. Candidate Cancer Slope Factors Based on Animal Toxicological Data from 2-year Cancer Bioassays
Tumor Type
Reference,
Confidence
Strain/
Species/Sex
POD Type, Model
POD Internal
Dose/Internal
Dose Metric"
PODhed
CSF
(BMR/PODhed)
Notes on Model
Selection
Leydig Cell
Adenomas in the
Testes
Butenhoff et al. {,
2012,2919192}
Medium
Male Sprague- BMDL4RD, 27,089.3 4.75 x 10 3 8.42
Dawley Rats Multistage Degree AUCavg (mg x d/L) mg/kg/day (mg/kg/day) 1
1
Model selected based
on lowest AIC as all
models had adequate
fit and BMDLs were
within sufficiently
close.
Hepatocellular
Adenomas or
Carcinoma
NTP {, 2020,
7330145}
High
Fi Male Sprague-
Dawley Rats,
Perinatal and
Postweaning
Exposure
BMDLiord, 88.7
Multistage Degree (Cavg_pupjotai in
2 mg/L)
1.06 x 10 2 mg/ 9.4
kg/day (mg/kg/day) 1
Model selected based
on lowest AIC as all
models had adequate
fit and BMDLs were
within sufficiently
close.
Hepatocellular
Adenomas
NTP {, 2020,
7330145}
High
Fi Male Sprague-
Dawley Rats,
Perinatal and
Postweaning
Exposure
BMDLiord, 93.0
Multistage Degree (CaVg_pup_totai in
2 mg/L)
1.12 x io-2 mg/ 9.0
kg/day (mg/kg/day) 1
Model selected based
on lowest AIC as all
models had adequate
fit and BMDLs were
within sufficiently
close.
Pancreatic Acinar NTP {, 2020,
Cell Adenoma or 7330145}
Adenocarcinoma High
Fi Male Sprague-
Dawley Rats,
Perinatal and
Postweaning
Exposure
BMDLiord, 15.2
Multistage Degree (Cavg_pupjotai in
3 mg/L)
1.83 x 10~3 54.7
(mg/kg/day) 1
Model selected based
on lowest AIC as all
models had adequate
fit and BMDLs were
within sufficiently
close.
Pancreatic Acinar NTP {, 2020,
Cell Adenoma 7330145}
High
Fi Male Sprague-
Dawley Rats,
Perinatal and
Postweaning
Exposure
BMDLiord, 15.7
Multistage Degree (Cavg_pUpjotai in
1 mg/L)
1.88 x 10~3 53.2
(mg/kg/day) 1
Model selected based
on lowest AIC as all
models had adequate
fit and BMDLs were
within sufficiently
close.
Notes: AUC = area under the curve; BMDL4RD = benchmark dose level corresponding to the 95% lower confidence limit of a 4% change; BMDLiord = lower bound on the dose
level corresponding to the 95% lower confidence limit for a 10% change; BMR = benchmark response; CSF = cancer slope factor; NTP = National Toxicology Program.
a See Appendix {U.S. EPA, 2024, 11414343} for additional details on benchmark dose modeling.
4-75
-------
APRIL 2024
4.2.3 Overall CSF Selection
Overall, recently published studies and the candidate CSFs indicate that PFOA is a more potent
carcinogen than previously understood and described in the 2016 PFOA HESD {U.S. EPA,
2016, 3603279}. To select an overall CSF, EPA focused on the CSFs derived from the
epidemiological data consistent with the IRIS 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, 10367891}. As with data underlying noncancer RfDs, the use of human data
eliminates 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. The use of human data also ensures that the values are based on human-relevant
exposure conditions and human-relevant tumor types/sites.
Therefore, EPA selected the critical effect of renal cell carcinomas in human males reported by
Shearer et al. {, 2021, 7161466} as the basis of the overall CSF for PFOA. Shearer et al. {, 2021,
7161455} is a well-conducted, multicenter case-control epidemiological study nested within
NCI's PLCO with median PFOA levels relevant to the general U.S. population. The CSF derived
from Shearer et al. {, 2021, 7161466} was selected as the overall CSF over the CSF derived
from Vieira et al. {, 2013, 2919154} due to multiple study design considerations. Specifically,
Shearer et al. {, 2021, 7161466} exhibited several preferred study attributes 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),
the source population (multicenter 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 overall CSF for PFOA based on RCC reported by Shearer et al. {, 2021, 7161466}
is 0.0293 (ng/kg/day) 1 (29,300 (mg/kg/day) ').
4.2.4 Application of Age-Dependent Adjustment Factors
EPA's Guidelines for Carcinogen Risk Assessment and Supplemental Guidance for Assessing
Susceptibility fi'om Early-Life Exposure to Carcinogens require the consideration of applying
age-dependent adjustment factors (ADAFs) to CSFs to address the potential for increased risk
for cancer due to early lifestage susceptibility to chemical exposure {U.S. EPA, 2005, 6324329;
U.S. EPA 2005, 88823}. Per EPA guidelines, 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}.
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,
4-76
-------
APRIL 2024
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 versus adult-only PFOA exposure. However, there are two
studies that assessed cancer risk after PFOA exposure during various developmental stages.
An NTP 2-year cancer bioassay in rats chronically exposed to PFOA both perinatally and
postweaning did not report an increased cancer risk compared with 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 with
postweaning-only exposure. The NTP cancer bioassay does not include dose groups that were
only exposed during early-lifestages (i.e., only during development) and therefore, 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 Fi female mice
perinatally treated with PFOA from GD 1-17, with potential residual exposure through lactation,
and necropsy at 18 months of age. This study is also limited in that there was no adult-only
exposure comparison group, the authors only assessed female mice, and the authors 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.
4-77
-------
APRIL 2024
5 Effects Characterization
5.1 Addressing Uncertainties in the Use of Epidemiological
Studies for Quantitative Dose-Response Analyses
In the 2016 Health Effects Support Document for Perflaorooctcmoic Acid (PFOA) 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:
• Unexplained 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 studies 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 of PFOA
for POD derivation in this assessment.
In this assessment EPA has assessed the strength of epidemiological and animal evidence
following current agency best practices for systematic review {U.S. EPA, 2022, 10367891}, 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, increased total cholesterol, increased 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 high and/or medium confidence epidemiological studies with low risk of bias in
different populations, including general and highly exposed populations. Many 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 approach
used in 2016. Though there are challenges in estimations of human dosimetry from measured or
5-1
-------
APRIL 2024
modeled serum concentrations (see Section 5.6.2), EPA has evaluated the available literature and
developed a pharmacokinetic model that estimates PFOA exposure concentrations from the
serum PFOA concentrations provided in epidemiological studies, which reduces uncertainties
related to exposure estimations in humans. This new approach is supplemented with the
uncertainty factor (UF) accounting for intraspecies variation of 10x 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 for
chronic, non-developmental effects, is the documented decline 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 Appendix E, {U.S. EPA,
2024, 11414343}), 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 United States, which occurred in the 1990's {Dong, 2019, 5080195}.
The modeling results show that the BMDL estimates for increased TC derived using the longest
range of exposure data (1999-2018) 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.
Given 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 consistently
selected 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 Appendix E, {U.S. EPA, 2024,
11414343}).
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, mixture analysis remains an
area of emerging research {Taylor, 2016, 11320539}, and there is no scientific consensus yet for
the best approach to account for exposure by co-occurring PFAS. Additionally, multipollutant
analyses from studies included in this assessment did not provide direct evidence that
associations between exposure to PFOA and health effects are confounded by or are fully
attributable to confounding by co-occurring PFAS. A detailed discussion of statical approaches
for accounting for co-occurring PFAS and results from studies performing multipollutant
analysis is provided in Section 5.1.1. For an extended review of the uncertainties associated with
PFAS co-exposures, see the Systematic Review Protocol for the PFBA, PFHxA, PFHxS, PFNA,
andPFDA (anionic and acidforms) 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;
Vestergren, 2008, 2558842; D'eon, 2011, 2903650}. The contributions of PFAS precursors to
5-2
-------
APRIL 2024
serum concentrations also varies between populations with differing PFAS exposure histories
(i.e., individuals living at or near sites with aqueous film-forming foam 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 2000's) 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 complexities related to 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 Appendix {U.S. EPA, 2024, 11414343} 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, the four selected critical effects (i.e.,
decreased antibody response to vaccination, increased serum ALT, increased TC, and decreased
birthweight) are biologically significant effects and/or precursors to disease (e.g., CVD), which,
according to agency guidance and methods, both warrant consideration as the basis of RfDs for
PFOA {U.S. EPA, 2002,88824 ; U.S. EPA, 2005, 6324329; U.S. EPA, 2022, 10367891}. EPA's
A Review of the Reference Dose and Reference Concentration Processes, states that a reference
dose (RfD) should be based on an adverse effect or a precursor to an adverse effect
(e.g., increased risk of an adverse effect occurring) {U.S. EPA, 2002, 88824}. Also, at the
individual level, the interpretation and impact of small magnitude changes in endpoints such as
increased TC, increased ALT, decreased birth weight, and decreased antibody response to
vaccination may be less clear. However, at the population level, even small magnitude changes
in these effects will shift the distribution in the overall population and increase the number of
individuals at risk for diseases, such as cardiovascular disease and liver disease{Gilbert, 2006,
174259}.
5-3
-------
APRIL 2024
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.
5.1.1 Uncertainty due to Potential Confounding by Co-Occurring
PFAS
5.1.1.1 PFAS Co-exposure Statistical Approaches and Confounding
Analysis
A potential source of uncertainty in epidemiologic studies examining associations between a
particular PFAS and health outcomes is confounding by other co-occurring PFAS. In studies of
PFOA, such confounding may occur if there are other PFAS that are moderately or highly
correlated with PFOA, associated with the outcome of interest, and not on the causal pathway
between PFOA and the outcome. If the association between co-occurring PFAS and the outcome
is in the same direction as the association between PFOA and that outcome, the anticipated
direction of bias resulting from not accounting for other PFAS would be away from the null. For
an extended review of the uncertainties associated with PFAS co-exposures, see the Systematic
Review Protocol for the PFBA, PFHxA, PFHxS, PFNA, and PFDA (anionic and acidforms)
IRIS Assessments {U.S. EPA, 2020, 8642427}.
Several statistical methods are used to estimate associations while accounting for potential
confounding by co-occurring PFAS and other pollutants. One common approach is to include co-
occurring PFAS as covariates in regression models. This approach allows for an estimation of
the association between PFOA and the outcome of interest, adjusted for other covariates and the
co-pollutants. Another approach is to screen large groups of exposures to identify which ones are
most strongly related to the outcome, using principal components analysis, elastic net regression,
and Bayesian kernel machine regression (BKMR). Each of these approaches has strengths and
limitations. For example, when PFOA and the co-pollutants are highly correlated, then
multipollutant models could be affected by multicollinearity or result in amplification bias, rather
than reduce confounding bias compared with single pollutant models {Weisskopf, 2018,
7325521}. Additionally, accounting for a variable in a multivariable regression model that is not
a significant predictor of the response variable reduces the degrees of freedom and effectively
dilutes the significance of the other exposure variables that are predictors of the response. The
use of screening approaches, while effective at accounting for co-pollutants, can result in
estimates that are not easily interpretable and make it difficult to differentiate the impact and
contribution of individual PFAS {Meng, 2018, 4829851}. Mixture analysis is an emerging
research area {Taylor, 2016, 11320539; Liu, 2022, 10606356}, and there is no scientific
consensus yet on the best approach for estimating independent effects of PFOA within complex
PFAS mixtures.
In this assessment, the risk of bias due to confounding by co-occurring PFAS was considered as
part of the study quality evaluation process. To further support the assessment, Section 5.1.1.2
below summarizes evidence from high and medium confidence studies that included at least one
of the approaches described above (hereafter referred to collectively as "multipollutant models")
5-4
-------
APRIL 2024
to account for co-pollutants, in order to assesses the extent to which there may be confounding
by other PFAS in studies reporting the associations between PFOA and birth weight.
5.1.1.2 Multipollutant Models of PFOA and Birth Weight
When assessing the associations between PFOA and a health effect of interest (e.g., decreased
birth weight), there is concern for potential confounding by other PFAS when there is a strong
correlation between the occurrence of PFOA and another PFAS and when the magnitude of the
association between the co-exposure and the health effect is large.
In order to identify the most co-occurring PFAS, Table 5-1 shows correlations between PFOA
and other PFAS exposures in the nine studies on PFOA and birth weight that included mutually
adjusted models. Four of these studies are medium confidence studies {Lenters, 2016, 5617416;
Meng, 2018, 4829851; Robledo, 2015, 2851197; Woods, 2017, 4183148} and five are high
confidence studies {Ashley-Martin, 2017, 3981371; Luo, 2021, 9959610; Manzano-Salgado,
2017, 4238465; Shoaff, 2018, 4619944; Starling, 2017, 3858473}. Moderately positive
correlations (around 0.6) between PFOA and PFOS were consistently observed in six of the
seven studies that reported such information. Correlations between PFOA with other commonly
examined PFAS, including PFNA (four studies), PFDA (four studies), and PFHxS (five studies),
were less consistent but generally weaker than correlations with PFOS, suggesting that other
PFAS may not consistently co-occur with PFOA.
Table 5-1. Correlation Coefficients Between PFOA and Other PFAS in Mutually Adjusted
Studies
Reference
Study Setting
Correlations with PFOA
PFOS
PFNA
PFDA
PFHxS
Asliley-Martin et al. {, 2017, 3981371 }a
High
Canada (10 cities)
0.59
_b
-
0.47
Luo et al. {, 2021, 9959610}3
High
Guangzhou, China
0.11
0.28
0.19
0.02
Manzano-Salgado et al. {, 2017, 4238465}°
High
Gipuzkoa, Sabadell, and
Valencia, Spain
NR
NR
NR
NR
Shoaff et al. {, 2018, 4619944}d
High
Cincinnati, Ohio, USA
0.60
- - -
Starling et al. {,2017, 3858473}e
High
Colorado, USA
0.68
0.76
0.56
0.61
Lenters et al. {, 2016, 5617416}6
Medium
Greenland; Kharkiv,
Ukraine; Warsaw, Poland
0.61
0.30
0.50
0.34
Meng et al. {, 2018, 4829851}d
Medium
Denmark
0.66
0.47
0.28
0.33
Robledo et al. {, 2015, 2851197}c
Medium
Michigan and Texas,
USA
NR
NR
NR
NR
Woods etal. {, 2017, 4183148}f
Medium
Cincinnati, Ohio, USA
+8
+
+
+
Notes: NR = not reported.
Shaded cells indicate analytes for which a correlation with PFOA was not measured or reported.
a Pearson correlation of loglo-transfonned {Ashley-Martin, 2017, 3981371} and ln-transformed {Luo, 2021, 9959610} PFAS
values.
5-5
-------
APRIL 2024
b Analyte not measured.
c Correlation coefficients not reported.
dPearson correlation of PFAS values, unclear if transformed prior to correlation analysis.
e Spearman rank correlation of PFAS values.
f Correlation type not specified.
g Correlations not reported numerically. Heat map of correlation coefficients (Figure S2, in Woods et al. {2017, 4183148}) shows
positive correlations between PFOA and PFOS, PFNA, PFHxS, and PFDA, ranging from about 0.6 to about 0.1, respectively.
Results from mutually adjusted models are summarized and compared in Table 5-2. The
statistical approaches for accounting for PFAS co-exposures varied across the studies. Six
studies included at least one additional PFAS as a predictor in ordinary least squares (OLS)
regression models {Ashley-Martin, 2017, 3981371; Manzano-Salgado, 2017, 4238465; Meng,
2018, 4829851; Robledo, 2015, 2851197; Shoaff, 2018, 4619944; Starling, 2017, 3858473}.
Woods et al. {2017, 4183148} included multiple PFAS as predictors in aBayesian hierarchical
linear model. Three studies {Lenters, 2016, 5617416; Starling, 2017, 3858473; Woods, 2017,
4183148} used elastic net regression and one study used BKMR {Luo, 2021, 9959610}. The
impact of other PFAS adjustment on the association between PFOA and birth weight is evaluated
by comparing the magnitude and direction of the effects from the single-PFOA model (when
available) to those from mutually adjusted models.
Six studies provided results from both single and multipollutant models {Lenters, 2016,
5617416; Luo, 2021, 9959610; Manzano-Salgado, 2017, 4238465; Meng, 2018, 4829851;
Shoaff, 2018, 4619944; Starling, 2017, 3858473}. Multipollutant models in these studies
included PFOS but varied with respect to other PFAS considered (Table 5-2). Lenters et al.
{2016, 5617416} also adjusted for other types of chemicals (such as phthalates and
organochlorides) in addition to several PFAS. Generally, the direction of effect estimates
remained the same following adjustment for other PFAS, but precision was reduced. None of the
studies that showed birth weight deficits in single-pollutant models reported greater or more
precise results following statistical adjustment for other PFAS.
Starling et al. {2017, 3858473} observed a statistically significant association between PFOA
and birth weight reductions in the single pollutant model. This association increased in
magnitude but precision was decreased in the multipollutant OLS model with four other PFAS.
PFOA was also retained in the elastic net regression model, showing an inverse relationship with
birth weight, but the association was attenuated. Lenters et al. {2016, 5617416} reported
associations between PFOA and reduced birth weight in single pollutant OLS and in elastic net
regression models with PFOA retained but the association attenuated. Luo et al. {2021,
9959610} observed non-significant inverse associations between PFOA and birth weight in
single pollutant and in BKMR models. Manzano-Salgado et al. {2017, 4238465} and Shoaff et
al. {2018, 4619944} reported null results in single and in multi-PFAS regression models. Meng
et al. {2018, 4829851} observed an association between PFOA and reduced birth weight in the
single pollutant model; this association was attenuated in a multipollutant model with PFOS.
Three studies provided results only from multipollutant models {Ashley-Martin, 2017, 3981371;
Robledo, 2015, 2851197; Woods, 2017, 4183148}, thus making assessment of impact of co-
pollutants difficult. Ashley-Martin et al. {2017,3981371} and Robledo etal. {2015,2851197}
reported non-significant inverse associations between PFOA and birth weight in girls in
multipollutant models. Woods et al. {2017, 4183148} reported on an overlapping population
from the same HOME cohort as Shoaff et al. {2018, 4619944} and observed non-significant
5-6
-------
APRIL 2024
inverse associations from a multipollutant Bayesian hierarchical linear model. PFOA was not
selected for inclusion in an elastic net regression model that included other endocrine-disrupting
chemicals in addition to PFAS.
In summary, in the six studies that included both single and multipollutant models, associations
were often attenuated following adjustment for other PFAS {Lenters, 2016, 5617416; Luo, 2021,
9959610; Manzano-Salgado, 2017, 4238465; Meng, 2018, 4829851; Shoaff, 2018, 4619944;
Starling, 2017, 3858473}. Three additional studies presented results from multipollutant models
only, making it difficult to determine the extent to which confounding by other PFAS may have
impacted the PFOA birth weight associations {Ashley-Martin, 2017, 3981371; Robledo, 2015,
2851197; Woods, 2017, 4183148}. Considering all nine studies together, it is challenging to
draw definitive conclusions about the extent of confounding by co-occurring PFAS, particularly
given differences in modeling approaches, PFAS adjustment sets, and exposure contrasts used
across studies. Additionally, these studies represented only a small fraction of the total number of
studies examining associations between PFOA and birth weight and it is unclear whether their
results are generalizable to the broader evidence base. Although it is an important source of
uncertainty, there is no evidence in the entirety of the large evidence base that the observed
associations between PFOA and birth weight deficits are fully attributable to confounding by co-
occurring PFAS.
Similar conclusions can be drawn for other health outcomes. Budtz-Jorgensen {2018, 5083631}
evaluated the possibility of confounding across PFAS in analyses of decreased antibody
response. The study reported significant decreases in the antibody response with elevated PFOA
exposure, and there was no notable attenuation of the observed effects after estimates were
adjusted for PFOS (see Section 3.4.2.1.2.1) {Budtz-Jorgensen, 2018, 5083631}. A limited
number of studies performed co-exposure analyses for increased ALT and increased TC in
adults. Lin et al. {2010, 1291111} performed multipollutant modeling for the effects on serum
ALT and observed that when PFOS, PFHxS, and PFNA were simultaneously added in the fully
adjusted regression models, the significant positive association between PFOA exposure and
ALT remained and was slightly larger. Fan et al. {2020, 7102734} examined cross-sectional
associations between exposure to PFOA and increased TC in single- and multipollutant models
in a sample of adults from NHANES (2012-2014). Exposure to PFOA was associated with
statistically significantly elevated TC in the single-pollutant model, but the association was no
longer significant in multipollutant analyses. A statistically significant positive association was
also observed for PFAS mixture and TC in WQS regression analyses {Fan, 2020, 7102734}.
Overall, there is no evidence that the consistently observed associations between exposures to
PFOA and the four priority noncancer health outcomes are confounded by or are fully
attributable to confounding by co-occurring PFAS.
5-7
-------
APRIL 2024
Table 5-2. Impact of Co-Exposure Adjustment on Estimated Change in Mean Birth Weight (grams) per Unit Change (ng/mL)
in PFOA Levels.
Reference
Single PFAS Model
Result (95% CI)ab
Multi-PFAS Model
Result (95% CI)a b
Elastic Net
Regression
Resultb
Exposure
Comparison
Effect of Other PFAS
Adjustment on PFOA
Birth Weight Results
PFAS
Adjustments
Ashley-Martin et al.
{2017,3981371}
High
NR
Girls: -89.51 (-263.40,
84.38)
Bovs: -35.51 (-198.99,
127.97)
c
logio-unit (ng/mL)
increase
PFOS, PFHxS
Luo et al. {2021,
9959610}
High
-62.37 (-149.08, 24.35)
-24 (-84, 36)d
Sinsle PFAS model: Attenuated
ln-unit (ng/mL)
increase
Multi-PFAS model:
75th vs. 25th
percentile
PFOS, PFBA,
PFNA, PFDA,
PFUnDA,
PFDoDA,
PFTrDA, PFBS,
PFHxS, 6:2 Cl-
PFESA, 8:2 Cl-
PFESA
Manzano-Salgado et al.
{2017,4238465}
High
-9.33 (-38.81,20.16)
1.02 (-42.73, 44.77)
log2-unit (ng/mL)
increase
Slightly attenuated
PFOS, PFNA,
PFHxS
Shoaff et al. {2018,
4619944}
High
-0.03 (-0.17, 0.10)e
0.00 (-0.16, 0.18)e
log2-unit (ng/mL)
increase
Slightly attenuated
PFOS, PFNA,
PFHxS
Starling et al. {2017,
3858473}
High
-51.4 (-97.2, -5.7)
-69.66 (-148.19, 8.87)
-14.47
ln-unit (ng/mL)
increase
Attenuated
PFOS, PFNA,
PFDA, PFHxS
Lenters et al. {2016,
5617416}
Medium
-78.52 (-137.01,
-20.03)
-38.82
2 SD ln-unit
(ng/mL) increase
Attenuated
PFOS, PFNA,
PFDA, PFHxS,
PFHpA,
PFUnDA,
PFDoDA
Meng et al. {2018,
482985 l}f
Medium
-35.6 (-66.3, -5.0)
-9.94 (-52.63, 32.75)
log2-unit (ng/mL)
increase
Attenuated
PFOS
5-8
-------
APRIL 2024
Reference
Single PFAS Model
Result (95% CI)ab
Multi-PFAS Model
Result (95% CI)a b
Elastic Net
Regression
Resultb
Exposure
Comparison
Effect of Other PFAS
Adjustment on PFOA
Birth Weight Results
PFAS
Adjustments
Robledo et al. {2015,
2851197 }g
Medium
NR
Girls: -61.64 (-159.15,
35.87)
Boys: 4.78 (-85.44,
95.01)
1 SD ln-unit
(ng/mL) increase
PFOS, PFDA,
PFNA, PFOSA,
Et-PFOSA-
AcOH, Me-
PFOSA-AcOH
Woods etal. {2017,
4183148}
Medium
NR
-13 (-54, 27)h
N/S logio-unit (ng/mL)
increase
PFOS, PFNA,
PFDA, PFHxS
Notes: N/S = PFOA not selected in elastic net regression model; SD = standard deviation.
aFrom ordinary least squares regression models unless otherwise specified.
b Outcome variable unit is grams unless otherwise specified.
cNot applicable.
d Results estimated from Luo et al. {2021, 9959610} Figure 3 using WebPlotDigitizer. Results are from a Bayesian kernel machine regression model comparing the PFOA at its
75th vs. 25th percentile, holding other PFAS constant at their 50th percentiles.
e Outcome variable unit in Shoaff et al. {2018, 4619944} models is birth weight z-score.
fMeng et al. {2018, 4829851} estimates associations between serum PFOA and birth weight in three samples of the Danish National Birth Cohort, two of which were analyzed by
the same laboratory for PFOA, PFOS, PFDA, PFNA, PFHxS, and PFHpS and one of which was analyzed by a separate laboratory for PFOA and PFOS only.
g Robledo et al. {2015, 2851197} estimated associations using both maternal and paternal PFAS concentrations; results shown here are from maternal PFAS models, also adjusted
for "the individual and partner sum of remaining chemical concentrations in each chemical's respective class."
hEffect estimates and posterior 95% credible intervals based on a Bayesian hierarchical linear model. Results estimated from Woods et al. {2017, 4183148} Figure 1 using
WebPlotDigitizer.
5-9
-------
APRIL 2024
5.2 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-, 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
were derived from studies conducted in rats or mice, adding a potential source of uncertainty
related to toxicokinetic differences in these species compared with humans. For PFOA, sex-
specific differences in the toxicokinetics of rats is an additional source of uncertainty. To address
toxicokinetic differences between species and sexes, EPA utilized a pharmacokinetic (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 that 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 with
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
5-10
-------
APRIL 2024
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.
5.3 Updated Approach to Animal Toxicological RfD Derivation
Compared with 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 with the 2016 PFOA 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.
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 relationship 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,
10367891} 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
5-11
-------
APRIL 2024
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 Lau et al. {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 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 additional or improved study attributes compared to the
older study. Specifically, the NTP {2020, 7330145} study was identified as a high confidence
study (rather than medium confidence) that used a chronic (rather than 14-week) exposure
duration, larger sample sizes (n = 50 rather than n = 15), 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 5-3).
Table 5-3. Comparison of Candidate RfDs Derived from Animal Toxicological Studies for
Priority Health Outcomes3
Studies and Effects Used in 2016 for Candidate RfD Studies and Effects Used in 2024 for Candidate RfD
Derivationb Derivation
Immune
Dewitt et al. {2008, 1290826}, medium confidence - Dewitt et al. {2008, 1290826}, medium confidence -
reduced immunoglobulin M (IgM) response reduced IgM response
5-12
-------
APRIL 2024
Studies and Effects Used in 2016 for Candidate RfD Studies and Effects Used in 2024 for Candidate RfD
Derivationb Derivation
Developmental
Lau et al. {2006, 1276159}, medium confidence - Lau et al. {2006, 1276159}, medium confidence -
reduced pup ossification (forelimb and hindlimb) and delayed time to eye opening
accelerated male puberty (preputial separation)
Wolf et al. {2007, 1332672}, medium confidence - Song et al. {2018, 5079725}, medium confidence -
decreased pup body weight decreased pup survival
Hepatic
Perkins et al. {2004, 1291118}, medium confidence - NTP {2020, 7330145}, high confidence - liver necrosis
increased liver weight and necrosis
Notes: RfD = reference dose; IgM = immimoglobulin M; NTP = National Toxicology Program.
a Note that candidate RfDs for the fourth priority noncancer 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
PFOA HESD or the current assessment.
b Candidate RfDs from the 2016 PFOA HESD that correspond to non-priority health outcomes (e.g., renal) are not presented here.
5.4 Consideration of Alternative Conclusions Regarding the
Weight of Evidence of PFOA Carcinogenicity
While reviewing the weight of evidence for PFOA, EPA also evaluated consistencies of the
carcinogenicity database with other cancer descriptors according to the Guidelines for
Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}. 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, the agency identified several new studies reporting on
cancer outcomes that strengthened the evidence. As a result of conducting a weight of evidence
evaluation of the available carcinogenicity database, EPA determined that PFOA is consistent
with the descriptions for Likely to Be Carcinogenic to Ramans according to the Guidelines for
Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}, as described above. 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 Suggestive
descriptions (outlined below) 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;
5-13
-------
APRIL 2024
• 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 {Shearer, 2021,
7161466; Vieira, 2013, 2919154} and testicular {Vieira, 2013, 2919154; Barry, 2013, 2850946})
and studies of equal confidence did not provide conflicting evidence for these cancer types.
Studies within the same species of rat consistently report multisite tumorigenesis (i.e., testicular,
pancreatic, and hepatic {NTP, 2020, 7330145; Biegel, 2001, 673581; Butenhoff, 2012,
2919192}) 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 PFOA 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}. See Table 5-4 below for specific details on how
PFOA exceeds the examples supporting the Suggestive Evidence of Carcinogenic Potential
cancer descriptor in the Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}.
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. The Guidelines 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
• 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}.
As discussed in Section 3.5.5, convincing epidemiological evidence supporting a causal
association between human exposure to PFOA and cancer is currently lacking. 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}.
Additionally, though the available evidence indicates that there are positive associations between
PFOA and multiple cancer types, there is uncertainty regarding the identification of carcinogenic
5-14
-------
APRIL 2024
MOA(s) for 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 multisite tumorigenesis. The animal and mechanistic databases do not provide clarity
to discern the MOA(s) of PFOA in humans, though there is some animal toxicological study
evidence supporting hormone-mediated MO As for testicular tumors and oxidative stress-
mediated MO As for pancreatic tumors. The full mode of action analysis, including in-depth
discussions on the potential MO As for kidney and testicular tumors, as well as discussions on the
potential MO As and human relevance for pancreatic and liver tumors observed in rats, is
presented in Section 3.5.4.2. See Table 5-4 below for specific details on how PFOA does not
align with the examples supporting the Carcinogenic to Humans cancer descriptor in the
Guidelines for Carcinogen Risk Assessment {U.S. EPA, 2005, 6324329}.
Table 5-4. 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 epidemiologic evidence of a causal
association between human exposure and cancer"
{U.S. EPA, 2005, 6324329}.
PFOA data are not consistent with this description.
There is evidence of a plausible association between
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.
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," {U.S. EPA, 2005, 6324329}. mechanistic information that could contribute to the
determination of a causal relationship.
"There is extensive evidence of carcinogenicity in
animals." {U.S. EPA, 2005, 6324329}.
PFOA data are not consistent with this description.
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
database would benefit from additional high confidence
chronic studies in other species and/or strains.
"The mode(s) of carcinogenic action and associated
key precursor events have been identified in animals
and" {U.S. EPA, 2005, 6324329}.
PFOA data are not consistent with this description. A
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
anticipated to occur in humans and progress to tumors,
based on available biological information" {U.S. EPA,
2005, 6324329}.
The animal database does not provide significant clarity
on the MOA(s) of PFOA in humans, though there is
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
PFOA data exceed this description. Statistically
significant increases in tumor incidence of multiple
5-15
-------
APRIL 2024
Comparison of Evidence for Carcinogenic and Suggestive Cancer Descriptors
of evidence for the descriptor "Likely to Be tumor types were observed across several human and
Carcinogenic to Humans." The study generally would animal toxicological studies,
not be contradicted by other studies of equal quality in
the same population group or experimental system"
{U.S. EPA, 2005, 6324329}.
"A small increase in a tumor with a high background This description is not applicable to the tumor types
rate in that sex and strain, when there is some but observed after PFOA exposure,
insufficient evidence that the observed tumors may be
due to intrinsic factors that cause background tumors
and not due to the agent being assessed" {U.S. EPA,
2005, 6324329}.
"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)" {U.S. EPA, 2005,
6324329}.
"A statistically significant increase at one dose only, PFOA data exceed this description. Increases in
but no significant response at the other doses and no kidney cancer in humans were statistically significant in
overall trend" {U.S. EPA, 2005, 6324329}. two exposure groups in one study {Vieira, 2013,
2919154}, and there was a statistically significant
increased odds for the highest exposure quartile and an
increasing trend across exposure quartiles in a second
study {Shearer, 2021, 7161466}. Statistically significant
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.
5.5 Health Outcomes with Evidence Integration Judgments of
Evidence Suggests Bordering on Evidence Indicates
EPA evaluated 16 noncancer 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, 10367891}, for both quantitative and qualitative analyses in the final 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
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 Perflaorooctanoic Acid (PFOA)
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.
5-16
-------
APRIL 2024
(CASRN 335-67-1) in Drinking Water {U.S. EPA, 2021, 10428559}. However, upon further
examination using the protocols for evidence integration outlined in Appendix A {U.S. EPA,
2024, 11414343} and Section 2.1.5, EPA concluded that the available epidemiological and
animal toxicological evidence did not meet the criteria recommended 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 published since the 2016 PFOA HESD 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
eight 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 {U.S.
EPA, 2024, 11414343} for a detailed description of endocrine evidence synthesis and
integration.
Epidemiological studies of reproductive effects in males published since the 2016 PFOA HESD
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 across studies and populations, and it was difficult
to assess the impacts of the alterations. Animal toxicological studies considered for evidence
integration included three high confidence studies {Biegel, 2001, 673581; NTP, 2019, 5400977;
NTP, 2020, 7330145} and five 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 {U.S. EPA, 2024, 11414343} for a detailed
description of male reproductive evidence synthesis and integration.
Female reproductive epidemiological studies published since the 2016 PFOA 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,
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
5-17
-------
APRIL 2024
inconsistent across reproductive hormone parameters, and it was difficult to assess the adversity
of these alterations. Animal toxicological studies considered for evidence integration included
one high confidence study {NTP, 2019, 5400977} and three 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 {U.S. EPA, 2024, 11414343} 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} and the final
IRIS Toxicological Reviews for both PFBA {U.S. EPA, 2022, 11181062} and PFHxA {U.S.
EPA, 2023, 11181061} concluded that the available evidence indicates that the observed thyroid
effects were likely due to PFBA and PFHxA exposure, respectively. 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
reviews conducted in February 2022 and 2023. 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 suggest the
potential for endocrine and reproductive effects after PFOA exposure. Studies on endocrine and
reproductive health outcomes represent two important research needs.
5.6 Challenges and Uncertainty in Modeling
5.6.1 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
Appendix F, {U.S. EPA, 2024, 11414343}), 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.
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
5-18
-------
APRIL 2024
sensitivity to serum concentrations (see Appendix F, {U.S. EPA, 2024, 11414343}).
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 Appendix F, {U.S. EPA, 2024,
11414343}). 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 lifestages, 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
lifestages. 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 lifestages 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}.
5.6.2 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 with blood in neonates and infants compared with adults
{Fernandez, 2011, 9641878}. This is well known for pharmaceuticals, but PFOA is unlike most
pharmaceuticals in that it undergoes extensive protein interaction, such that its distribution in the
5-19
-------
APRIL 2024
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
Appendix B, {U.S. EPA, 2024, 11414343}). 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 with 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
1,200 mg weekly for a period of 6 weeks. In this study, the average plasma level after 6 weeks
ranged from 34 |ag/m L at 0.1 mg/kg/day to 492.7 |ag/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 95th percentile serum PFOA concentration in NHANES 2007-08 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}
estimated a half-life of 4.1 years and Fujii et al. {2015, 2816710} estimated a renal clearance
5-20
-------
APRIL 2024
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 1 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 1 year. This is a relatively long duration of breastfeeding; only
27% of children in the United States are being breastfed at 1 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 1-year breastfeeding duration is a conservative approach and will
result in a lower PODhed than a scenario with earlier weaning. Children in the United States are
very unlikely to be exclusively breastfed for up to 1 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.
5.6.3 Approach of Estimating a Benchmark Dose from a
Regression Coefficient
EPA identified epidemiological studies that reported associations between PFOA exposure and
response variables as regression coefficients. Since such a regression coefficient is associated
5-21
-------
APRIL 2024
with a change in the biological response variable, it is biologically meaningful and can therefore
be used for POD derivation. EPA modeled these regression coefficients using the same approach
used to model studies that reported measured response variables. The SAB PFAS Review Panel
agreed with this approach, stating, "it would seem straightforward to apply the same
methodology to derive the beta-coefficients ("re-expressed," if necessary, in units of per ng/mL)
for antibody responses to vaccines and other health-effect-specific endpoints. Such a coefficient
could then be used for deriving PODs" {U.S. EPA, 2022, 10476098}. When modeling regression
coefficients that were reported per log-transformed units of exposure, EPA used the SAB's
recommended approach and re-expressed the reported P coefficients in units of per ng/mL (see
Appendix E, {U.S. EPA, 2024, 11414343}). Sensitivity analyses to evaluate the potential impact
of re-expression in a hybrid approach when modeling hepatic and serum lipid studies for PFOA
showed little impact on BMDLs (see Appendix E, {U.S. EPA, 2024, 11414343}).
To evaluate this potential uncertainty in BMDLs derived based on regression coefficients, 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 Appendix E {U.S. EPA, 2024, 11414343}.
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 threefold
difference. The RfD allows for an order of magnitude (10-fold or 1,000%) uncertainty in the
estimate. Therefore, EPA is confident in its use of regression coefficients, re-expressed or not as
the basis of PODheds.
5.7 Human Dosimetry Models: Consideration of Alternate
Modeling Approaches
Physiologically based pharmacokinetic (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, for which 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. Because of
5-22
-------
APRIL 2024
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.
Given 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. Following 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. The overall agreement in approach between
the two models supports its validity for the task of human health risk assessment for PFOA.
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 with 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 body weight. After testing these features within
5-23
-------
APRIL 2024
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.
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 with the adult value. There is no chemical-specific information to suggest that
Vd is larger in children compared with 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 Appendix F, {U.S.
EPA, 2024, 11414343}). This resulted in reduced predictions of serum concentrations, primarily
during their peak in early childhood. The model with variable Vd did not decrease the root mean
squared error compared with the model with constant Vd. Since the model with constant Vdhad
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 body weight in young children compared with
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 Appendix F, {U.S.
EPA, 2024, 11414343}). 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 a maximum contaminant level goal (MCLG) calculated
thereafter using lifestage 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 with
the model with constant exposure relative to body weight. Use of water ingestion rates to adjust
5-24
-------
APRIL 2024
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,
Father, and Child Cohort Study (MoBa) cohort with a single timepoint. There was a much
greater effect on PFOS model results relative to PFOA, but in both cases model performance, as
quantified by root mean squared error, was superior with constant exposure compared with
exposure based on drinking water consumption. 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 under SDWA. The
epidemiological endpoints can be placed into three categories based on the age of the individuals
at the time of exposure measurement: 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 based on noncancer effects or for
nonlinear carcinogens is an algebraic calculation that incorporates the RfD, RSC, and drinking
water intake. The drinking water intake used for this type of MCLG calculation would be chosen
based on the exposure interval used in the critical study and/or the target population relevant to
the timing of exposure measurement and the response variable 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 presented in Appendix F {U.S. EPA, 2024,
11414343}, EPA determined that the updated Verner model was the most appropriate available
model structure for PODhed calculation for PFOA. Specifically, EPA concluded that the
determination that assuming Vd in children equal to the adult values and calculating an RfD
assuming a constant dose (mg/kg/day) were appropriate for this assessment.
5.8 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 more severe response than others despite similar PFOA exposure due to increased
biological sensitivity, as well as populations exhibiting a more severe response due to higher
PFOA exposure and/or exposure to other chemicals or nonchemical stressors. Populations with
greater biological sensitivity may include pregnant women and their developing fetuses, lactating
women, the elderly, children, adolescents, and people with certain underlying medical conditions
(see Section 5.8.1). Additionally, some available data indicates that there may be sex-specific
differences in sensitivity to potential effects of PFOA (see Section 5.8.2). 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 5.8.3).
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
5-25
-------
APRIL 2024
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.
5.8.1 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 Sections
3.4.4.1 and 3.4.2.1, results of some epidemiological studies indicate an association between
PFOA exposure during pregnancy and/or early childhood and adverse outcomes such as
decreased birth weight and decreased antibody response to vaccination. 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 lifestages or exposure windows (e.g., prenatal or early postnatal exposure windows) may
be more consequential than others. 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 mount a relatively low antibody response compared to other children 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 that 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.
5.8.2 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}.
5-26
-------
APRIL 2024
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 with males {Han, 2005, 5081570}. Several studies have
demonstrated that female rats and rabbits eliminate PFOA from the body faster than males
{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 Appendix B {U.S.
EPA, 2024, 11414343}.
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
PFOAin malerats (see Section3.3.1 and Appendix B, {U.S. EPA, 2024, 11414343}).
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.
5.8.3 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
5-27
-------
APRIL 2024
racial/ethnic groups (see Section 5.1), 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 Final PFAS National Primary Drinking Water Regulation {U. S. EPA,
2024, 11414059}). 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.
5-28
-------
APRIL 2024
6 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
6-1
-------
APRIL 2024
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.cfm/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.cfrn/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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,
6-2
-------
APRIL 2024
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 [ATSDRTox Profile]. Atlanta, GA: U.S.
Department of Health and Human Services, Public Health Service.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/9642134
Attanasio, R. (2019). Sex differences in the association between perfluoroalkyl acids and liver function in
US adolescents: Analyses ofNHANES 2013-2016. Environ Pollut254: 113061.
https://hero.epa.gov/hero/index.cfm/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.cfm/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.cfm/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.cfim/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.cfm/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.cfm/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.cfm/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. cfm/reference/details/reference_id/3 981559
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.cfm/reference/details/reference_id/5918617
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.cfm/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
6-3
-------
APRIL 2024
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/3 79025
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.cfm/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 perfluorooctane sulfonic
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;
6-4
-------
APRIL 2024
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/127413 3
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/2713 5 74
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.cfm/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.cfim/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
6-5
-------
APRIL 2024
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/3 3 78942
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-
6-6
-------
APRIL 2024
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/5 381552
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.
PLoSONE 13: e0205388.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/5083631
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 Pollut275: 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
6-7
-------
APRIL 2024
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/5079678
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/5 3 813 04
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
6-8
-------
APRIL 2024
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. cfm/reference/details/reference_id/7021512
Canova, C; Di Nisio, A; Barbieri, G; Russo, F; Fletcher, T; Batzella, E; Dalla Zuanna, 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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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/5097914
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.cfm/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.
6-9
-------
APRIL 2024
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/1936457
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 Pathol 192623320911606.
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.
6-10
-------
APRIL 2024
(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/3 981369
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/3 9813 04
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/5 024207
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/3 858533
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/3 3 5 0721
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
6-11
-------
APRIL 2024
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
6-12
-------
APRIL 2024
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/5 3 815 64
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/29193 3 5
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/683 3 614
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-
6-13
-------
APRIL 2024
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.cfm/reference/details/reference_id/45 63 919
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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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. cfm/reference/details/reference_id/3 393032
Davies, B; Morris, T. (1993). Physiological parameters in laboratory animals and humans [Review].
PharmRes 10: 1093-1095.
https ://hero .epa.gov/hero/index.cfm/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.cfm/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.
6-14
-------
APRIL 2024
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/75 643 88
Del Campo, JA; Gallego, P; Grande, L. (2018). Role of inflammatory response in liver diseases:
Therapeutic strategies [Review]. World Journal of Hepatology 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/l 3 35605
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.cfm/reference/details/reference_id/2851016
6-15
-------
APRIL 2024
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/3 981435
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;
6-16
-------
APRIL 2024
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/683 3 676
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/5 918597
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.cfm/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 Pollut238: 389-396.
https://hero.epa.gov/hero/index.cfm/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 orgavage administration. Xenobiotica 50: 1-11.
https ://hero.epa.gov/hero/index. cfm/reference/details/reference_id/5 916078
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.cfim/reference/details/reference_id/68 3 3691
6-17
-------
APRIL 2024
Dzierlenga, MW; Allen, BC; Ward, PL; Clewell, HJ; Longnecker, MP. (2019). A model of functional
thyroid disease status over the 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/5381535
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.
http s: //hero. epa.gov/hero/index. cfm/reference/details/reference_id/5 085376
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/3056793
6-18
-------
APRIL 2024
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/541243 0
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 Pollut 266 Pt 2: 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/1023 5 84
6-19
-------
APRIL 2024
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/8 8170
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;
6-20
-------
APRIL 2024
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 ?~Analy sis 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/3858513
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/4003 913
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 aNorwegian 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
6-21
-------
APRIL 2024
reveals dysregulation of lipid profiles in human liver cells exposed to environmental obesogens.
Toxicol Appl Pharmacol 398: 115009.
https://hero.epa.gov/hero/index.cfin/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.cfim/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;
6-22
-------
APRIL 2024
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/3 810188
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.
6-23
-------
APRIL 2024
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/21495 3 0
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:
el 14295. 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; Roge. 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.
6-24
-------
APRIL 2024
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 noncancer 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.
6-25
-------
APRIL 2024
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/3 981536
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;
6-26
-------
APRIL 2024
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.cfm/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.cfm/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
6-27
-------
APRIL 2024
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/5 081570
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 ofPFOS 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.
6-28
-------
APRIL 2024
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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/reference/details/reference_id/1332671
Hines, EP; White, SS; Stanko, JP; Flournoy, EAG; Lau, C; Fenton, SE. (2009). Phenotypic dichotomy
6-29
-------
APRIL 2024
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: el3030.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/103 65 73 8
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/5 3 815 62
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
6-30
-------
APRIL 2024
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/5 024212
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/65 05 752
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
6-31
-------
APRIL 2024
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/1249755
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/5915990
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
6-32
-------
APRIL 2024
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.cfm/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.cfm/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/5079656
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. cfm/reference/details/reference_id/5 381566
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/5080621
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.cfm/reference/details/reference_id/6718854
6-33
-------
APRIL 2024
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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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
6-34
-------
APRIL 2024
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/75 7874
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.
6-35
-------
APRIL 2024
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4937567
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
6-36
-------
APRIL 2024
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. cfm/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.cfm/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.cfm/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.cfrn/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.cfm/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. cfm/reference/details/reference_id/3 351917
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.cfm/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.cfm/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.cfm/reference/details/reference_id/6833758
6-37
-------
APRIL 2024
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/3 981330
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.),
6-38
-------
APRIL 2024
(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.cfm/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.cfrn/reference/details/reference_id/683 3 715
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.cfrn/reference/details/reference_id/2850246
6-39
-------
APRIL 2024
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:
644-65 0. 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
6-40
-------
APRIL 2024
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.cfin/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.cfin/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.cfm/reference/details/reference_id/3 981419
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.cfin/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.cfin/reference/details/reference_id/3 983 5 76
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.cfin/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.cfin/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.cfin/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.cfin/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.cfin/reference/details/reference_id/5080366
Lenters, V; Portengen, L; Rignell-Hydbom, A; Jonsson, BA; Lindh, CH; Piersma, AH; Toft, G; Bonde,
6-41
-------
APRIL 2024
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
6-42
-------
APRIL 2024
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
6-43
-------
APRIL 2024
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; Bonefcld-Jorgenscn. 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.cfm/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.cfm/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.cfm/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
6-44
-------
APRIL 2024
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 Pollut 235: 446-452.
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/4239494
6-45
-------
APRIL 2024
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.cfm/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.cfm/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.cfrn/reference/details/reference_id/683 3 599
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.cfrn/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfrn/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.cfm/reference/details/reference_id/2324799
6-46
-------
APRIL 2024
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/5 881135
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-
6-47
-------
APRIL 2024
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/5381625
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/3858516
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
6-48
-------
APRIL 2024
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 Pollut 247: 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/3981426
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/4238291
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.
6-49
-------
APRIL 2024
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.cfm/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.cfm/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.cfm/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.cfm/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 mother to fetus in a Spanish birth cohort. Environ Res 142: 471-478.
https://hero.epa.gov/hero/index.cfm/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.cfm/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.cfrn/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.cfrn/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.cfm/reference/details/reference_id/8632225
Martin, MT; Brennan, RJ; Hu, W; Ayanoglu, E; Lau, C; Ren, H; Wood, CR; Corton, JC; Kavlock, RJ;
6-50
-------
APRIL 2024
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.
PLoSONE 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 IB 10 (AKR1B10) as abiomarker to 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/63 04412
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
6-51
-------
APRIL 2024
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
6-52
-------
APRIL 2024
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.cfm/reference/details/reference_id/9571900
MPCA. (2008). PFCs in Minnesota's Ambient Environment: 2008 Progress Report.
https://hero.epa.gov/hero/index.cfm/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.
6-53
-------
APRIL 2024
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/1041145 9
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
6-54
-------
APRIL 2024
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/73 3 0145
6-55
-------
APRIL 2024
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.cfm/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.
6-56
-------
APRIL 2024
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.
Neurotox Res 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: e0150284.
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.cfim/reference/details/reference_id/7542684
6-57
-------
APRIL 2024
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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/reference/details/reference_id/10176842
Pecquet, AM; Maier, A; Kasper, S; Sumanas, S; Yadav, J. (2020). Exposure to perfluorooctanoic acid
6-58
-------
APRIL 2024
(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.
Med Eng Phys 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 Faroese men with elevated exposure to
poly chlorinated 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/683 3 63 7
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
6-59
-------
APRIL 2024
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/5 387175
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.
6-60
-------
APRIL 2024
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/193725 9
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/193 723 6
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 Pollut212: 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
6-61
-------
APRIL 2024
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/3351439
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/3 981499
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,
6-62
-------
APRIL 2024
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.cfm/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.cfm/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.cfm/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.cfm/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.cfm/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).
6-63
-------
APRIL 2024
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/5 387114
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/423 8410
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/143 0761
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
6-64
-------
APRIL 2024
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/10365760
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. MutatRes 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
6-65
-------
APRIL 2024
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/33 60127
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).
6-66
-------
APRIL 2024
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
6-67
-------
APRIL 2024
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.cfin/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.cfin/reference/details/reference_id/5 918847
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.cfm/reference/details/reference_id/3 749113
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.cfin/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.cfin/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.cfin/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.cfin/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.cfin/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.cfin/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.cfin/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.cfin/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.
6-68
-------
APRIL 2024
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: el 3059.
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:
10465 6. https://hero.epa.gOv/hero/index.cfm/reference/details/reference_id/5 918817
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
6-69
-------
APRIL 2024
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/3860111
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
6-70
-------
APRIL 2024
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,
6-71
-------
APRIL 2024
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.
6-72
-------
APRIL 2024
https ://hero. epa.gov/hero/index. cfm/reference/details/reference_id/3 981439
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
6-73
-------
APRIL 2024
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/8 8 824
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 aPCCL. (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.
6-74
-------
APRIL 2024
https://hero.epa.gov/hero/index.cfm/reference/details/reference_id/1239433
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/3603365
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 forthe 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 forthe 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
6-75
-------
APRIL 2024
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 in NHANES 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:
6-76
-------
APRIL 2024
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
6-77
-------
APRIL 2024
precursor compounds in consumer exposure to PFOS and PFOA. Chemosphere 73: 1617-1624.
https ://hero .epa.gov/hero/index.cfm/reference/details/reference_id/25 5 8 842
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/5079675
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.
6-78
-------
APRIL 2024
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/3 9815 04
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 Pollut 247: 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
6-79
-------
APRIL 2024
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.cfm/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
6-80
-------
APRIL 2024
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/1276150
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 33 67
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
6-81
-------
APRIL 2024
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/2337818
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.
6-82
-------
APRIL 2024
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/5918609
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
6-83
-------
APRIL 2024
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.cftn/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.cfm/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. cfm/reference/details/reference_id/3 981567
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.cfm/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.cftn/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.cfm/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.cfm/reference/details/reference_id/6356370
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. cfm/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.cfm/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.cfm/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.cfm/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.cfm/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.
6-84
-------
APRIL 2024
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/3 981427
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/845 3 076
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.
6-85
-------
APRIL 2024
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, O. (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; DallaZuanna, 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.
6-86
-------
APRIL 2024
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/3748826
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/5 081488
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
6-87
-------
APRIL 2024
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/65 05 878
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/5 08 5130
Zhao, W; Zitzow, JD; Weaver, Y; Ehresman, DJ; Chang, SC; Butenhoff, JL; Hagenbuch, B. (2017).
6-88
-------
APRIL 2024
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/3 85 9799
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
6-89
------- |