IRIS Toxicological Review of Perfluorodecanoic Acid [PFDA, CASRN 335-76-2] and Related Salts, April 2023


II	I	

EPA/635/R-23/056a
External Review Draft

www.epa.gov/iris

IRIS Toxicological Review of Perfluorodecanoic Acid [PFDA, CASRN 335-

76-2] and Related Salts

April 2023

Integrated Risk Information System
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

DISCLAIMER

This document is an external review draft for review purposes only. This information is
distributed solely for the purpose of predissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should
not be construed to represent any Agency determination or policy. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.

This document is a draft for review purposes only and does not constitute Agency policy.

ii	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

CONTENTS

AUTHORS | CONTRIBUTORS | REVIEWERS	xii

EXECUTIVE SUMMARY	xv

1.	OVERVIEW OF BACKGROUND INFORMATION AND ASSESSEMENT METHODS	1-1

1.1.	BACKGROUND INFORMATION ON PERFLUORODECANOIC ACID (PFDA)	1-1

1.1.1.	Physical and Chemical Properties	1-1

1.1.2.	Sources, Production, and Use	1-2

1.1.3.	Environmental Fate and Transport	1-2

1.1.4.	Potential for Human Exposure, including Populations and Lifestages with

Potentially Greater Exposure	1-3

1.2.	SUMMARY OF ASSESSMENT METHODS	1-7

1.2.1.	Literature Search and Screening	1-7

1.2.2.	Evaluation of Individual Studies	1-10

1.2.3.	Additional Epidemiology Considerations	1-11

1.2.4.	Data Extraction	1-12

1.2.5.	Evidence Synthesis and Integration	1-13

1.2.6.	Dose-Response Analysis	1-14

2.	LITERATURE SEARCH RESULTS	2-1

2.1.	LITERATURE SEARCH AND SCREENING RESULTS	2-1

2.2.	SUMMARY OF STUDIES MEETING PECO CRITERIA	2-2

3.	PHARMACOKINETICS, EVIDENCE SYNTHESIS, AND INTEGRATION	3-1

3.1.	PHARMACOKINETICS	3-1

3.1.1.	Absorption	3-2

3.1.2.	Distribution	3-3

3.1.3.	Metabolism	3-12

3.1.4.	Excretion	3-12

3.1.5.	Summary of pharmacokinetic parameters	3-18

3.1.6.	Evaluation of PBPK and PK Modeling	3-19

3.1.7.	Approach for pharmacokinetic extrapolation of PFDA among rats, mice, and

humans	3-22

3.2.	NONCANCER HEALTH EFFECTS	3-25

This document is a draft for review purposes only and does not constitute Agency policy.

iii	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.2.1.	HEPATIC EFFECTS	3-25

3.2.2.	IMMUNE EFFECTS	3-54

3.2.3.	DEVELOPMENTAL EFFECTS	3-98

3.2.4.	MALE REPRODUCTIVE EFFECTS	3-166

3.2.5.	FEMALE REPRODUCTIVE EFFECTS	3-190

3.2.6.	CARDIOMETABOLIC EFFECTS	3-209

3.2.7.	NEURODEVELOPMENTAL EFFECTS	3-240

3.2.8.	ENDOCRINE EFFECTS	3-247

3.2.9.	URINARY EFFECTS	3-269

3.2.10.	GENERAL TOXICITY	3-282

3.2.11.OTHER HEALTH EFFECTS	3-287

3.3. CARCINOGENICITY	3-288

3.3.1. CANCER	3-288

4.	SUMMARY OF HAZARD IDENTIFICATION CONCLUSIONS	4-1

4.1.	SUMMARY OF CONCLUSIONS FOR NONCANCER HEALTH EFFECTS	4-1

4.2.	SUMMARY OF CONCLUSIONS FOR CARCINOGENICITY	4-5

4.3.	CONCLUSIONS REGARDING SUSCEPTIBLE POPULATIONS AND LIFE STAGES	4-5

5.	DERIVATION OF TOXICITY VALUES	5-1

5.1.	NONCANCER AND CANCER HEALTH EFFECT CATEGORIES CONSIDERED	5-1

5.2.	NONCANCER TOXICITY VALUES	5-1

5.2.1.	Oral Reference Dose (RfD) Derivation	5-2

5.2.2.	Selection of Lifetime Toxicity Value(s)	5-25

5.2.3.	Subchronic Toxicity Values for Oral Exposure (Subchronic Oral Reference Dose

[RfD]) Derivation	5-29

5.2.4.	Inhalation Reference Concentration (RfC) Derivation	5-41

5.3.	CANCER TOXICITY VALUES	5-42

REFERENCES	R-l

SUPPLEMENTAL INFORMATION (see Volume 2)

This document is a draft for review purposes only and does not constitute Agency policy.

iv	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

TABLES

Table 1-1. Physical-chemical properties of PFDA and related salts 1-1

Table 1-2. Serum PFDA concentrations based on NHANES 2013-2014 data (ng/L) 1-4

Table 1-3. PFDA levels at 10 military installations 1-6

Table 1-4. Populations, exposures, comparators, and outcomes (PECO) criteria 1-9

Table 2-1. Animal toxicity studies examining health effects after PFDA administration 2-3

Table 3-1. Volume of distribution values reported for animal studies 3-6

Table 3-2. PFDA total clearance in rats and mice 3-13

Table 3-3. Rat, mouse, and human pharmacokinetic parameters 3-18

Table 3-4. DDEF calculations 3-23

Table 3-5. Associations between PFDA and serum biomarkers of hepatic function in medium

confidence epidemiology studies 3-29

Table 3-6. Incidence and severity of hepatocyte lesions in S-D rats exposed to PFDA in 28-day

gavage studies 3-34

Table 3-7. Percent change relative to controls in hepatocellular serum markers in short-term

animal studies after PFDA exposure 3-37

Table 3-8. Percent change relative to controls in hepatobiliary serum markers in a 28-day rat

study after PFDA exposure (NTP, 2018) 3-37

Table 3-9. Percent change relative to controls in serum proteins in a 28-day rat study after PFDA

exposure (NTP, 2018) 3-39

Table 3-10. Percent change relative to controls in liver weight (relative to body weight) due to

PFDA exposure in short-term oral toxicity studies 3-42

Table 3-11. Evidence profile table for PFDA exposure and liver effects 3-51

Table 3-12. Summary of PFDA exposure and selected data on antibody response in humans 3-58

Table 3-13. Studies on PFDA and infectious disease in humans 3-62

Table 3-14. Studies on PFDA and hypersensitivity-related outcomes in humans 3-69

Table 3-15. Summary of PFDA and selected data on hypersensitivity in humans 3-70

Table 3-16. Percent change relative to controls in absolute spleen cell population counts in

female B6C3F1/N mice exposed to PFDA exposure for 28-days (Frawley et al.,

2018) 3-78

Table 3-17. Percent change relative to controls in blood leukocyte counts in female S-D rats

exposed to PFDA exposure for 28-days (NTP, 2018) 3-81

Table 3-18. Percent change relative to controls in immune organ weights in short-term animal

studies after exposure to PFDA 3-86

Table 3-19. Evidence profile table for PFDA exposure and immune effects 3-94

Table 3-20. Summary of 33 studies (from 35 publications) of PFDA exposure in relation to fetal

and postnatal growth restriction measures sorted by overall confidence 3-123

Table 3-21.Summary of 12 studies of PFDA exposure and gestational duration measures 3-146

Table 3-22. Associations between PFDA and spontaneous abortion in epidemiology studies 3-150

Table 3-23. Percent changes relative to controls in fetal body weight in a developmental mouse

study after PFDA exposure (Harris and Birnbaum, 1989) 3-153

Table 3-24. Evidence profile table for PFDA exposure and developmental effects 3-159

Table 3-25. Associations between serum PFDA and semen parameters in epidemiology studies 3-167

Table 3-26. Evidence profile table for PFDA exposure and male reproductive effects 3-186

Table 3-27. Associations between PFDA and time to pregnancy in epidemiology studies 3-194

This document is a draft for review purposes only and does not constitute Agency policy.

v DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-28. Percent changes relative to controls in time spent in each estrous stage (proestrus,
estrus, metestrus, diestrus) in female S-D rats exposed to PFDA exposure for 28

days (NTP, 2018) 3-199

Table 3-29. Evidence profile table for PFDA exposure and female reproductive effects 3-205

Table 3-30. Associations between PFDA and blood lipids in medium confidence epidemiology

studies 3-213

Table 3-31. Associations between PFDA and hypertensive disorders of pregnancy in

epidemiology studies 3-217

Table 3-32. Associations between PFDA and coronary heart disease in epidemiology studies 3-219

Table 3-33. Associations between PFDA and insulin resistance in epidemiology studies 3-223

Table 3-34. Associations between PFDA and adiposity in epidemiology studies 3-229

Table 3-35. Percent change relative to controls in serum lipids in a 28-day rat study after PFDA

exposure (NTP, 2018) 3-232

Table 3-36. Percent change relative to controls in heart weights in a 28-day rat study after PFDA

exposure (NTP, 2018) 3-233

Table 3-37. Evidence profile table for PFDA exposure and cardiometabolic effects 3-236

Table 3-38. Results for medium confidence epidemiology studies of PFDA exposure and

behavioral and attention effects 3-243

Table 3-39. Evidence profile table for PFDA neurodevelopmental effects 3-246

Table 3-40. Percent changes relative to controls in thyroid hormone levels in a 28-day rat study

after PFDA exposure (NTP, 2018) 3-253

Table 3-41. Evidence profile table for PFDA exposure and endocrine effects 3-265

Table 3-42. Associations between serum PFDA and urinary effects in low confidence

epidemiology studies 3-270

Table 3-43. Percent change relative to controls in serum biomarkers of kidney function in a 28-

day rat study after PFDA exposure (NTP, 2018) 3-275

Table 3-44. Percent change relative to controls in kidney weights (absolute and relative to body

weight) due to PFDA exposure in short-term oral toxicity studies 3-277

Table 3-45. Evidence profile table for PFDA urinary effects 3-280

Table 3-46. Test evaluating genotoxicity and mutagenicity 3-292

Table 4-1. Hazard conclusions across published EPA PFAS human health assessments 4-3

Table 5-1. Endpoints considered for dose-response modeling and derivation of points of

departure for liver effects in animals 5-3

Table 5-2. Endpoints considered for dose-response modeling and derivation of points of

departure for immune effects in humans 5-7

Table 5-3. Mean Birth Weight deficit studies considered for dose-response modeling and

derivation of points of departure for developmental effects in humans 5-9

Table 5-4. Endpoints considered for dose-response modeling and derivation of points of

departure for developmental effects in animals 5-10

Table 5-5. Endpoints considered for dose-response modeling and derivation of points of

departure for male reproductive effects in animals 5-11

Table 5-6. Endpoints considered for dose-response modeling and derivation of points of

departure for female reproductive effects in animals 5-13

Table 5-7. Benchmark response levels selected for BMD modeling of PFDA health outcomes 5-13

Table 5-8. PODs considered for the derivation of PFDA candidate values 5-16

Table 5-9. Uncertainty factors for the development of the candidate lifetime toxicity values for

PFDA 5-19

Table 5-10. Candidate values for PFDA 5-24

This document is a draft for review purposes only and does not constitute Agency policy.

vi DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 5-11. Confidence in the organ/system-specific (osRfDs) for PFDA 5-25

Table 5-12. Organ/System-specific and overall lifetime RfDsfor PFDA 5-27

Table 5-13. Uncertainty factors for the development of the candidate subchronic values for

PFDA 5-30

Table 5-14. Candidate values for deriving the subchronic RfD for PFDA 5-34

Table 5-15. Confidence in the subchronic organ/system specific RfDs (subchronic osRfDs) for

PFDA 5-36

Table 5-16. Organ/system-specific and overall subchronic RfDs for PFDA 5-40

FIGURES

Figure 1-1. Chemical structure of PFDA and related salts 1-1

Figure 2-1. Literature search for perfluorodecanoic acid and related salts 2-2

Figure 3-1. Ratio of extracellular water (% of body weight) in children vs. adults. Values (points)
are calculated from results in Friis-Hansen (1961) and plotted at the mid-point

for the corresponding age ranges evaluated 3-11

Figure 3-2. Comparison of PFDA PBPK model predictions to IV dosimetry data (circles) of Kim et

al. (2019) for a 1 mg/kg dose 3-21

Figure 3-3. Hepatic effects, human study evaluation heatmap 3-27

Figure 3-4. Evaluation results for animal studies assessing effects of PFDA exposure on liver

histopathology 3-32

Figure 3-5. Effects on liver histopathology following exposure to PFDA in short-term oral studies

in animals 3-35

Figure 3-6. Evaluation results for animal studies assessing effects of PFDA exposure on liver

serum biomarkers 3-36

Figure 3-7. Effects on serum liver biomarkers following exposure to PFDA in short-term oral

studies in animals 3-40

Figure 3-8. Evaluation results for animal studies assessing effects of PFDA exposure on liver

weight 3-41

Figure 3-9. Effects on relative liver weight following exposure to PFDA in short-term oral studies

in animals 3-43

Figure 3-10. Summary of evaluation of epidemiology studies of PFDA and antibody response to

vaccination 3-56

Figure 3-11. Summary of evaluation of epidemiology studies of PFDA and infectious disease 3-62

Figure 3-12. Summary of evaluation of epidemiology studies of PFDA and sensitization or

allergic response 3-66

Figure 3-13. Evaluation results for animal study assessing effects of PFDA exposure on host

resistance 3-73

Figure 3-14. Evaluation results for animal studies assessing effects of PFDA exposure on

immune function assays 3-74

Figure 3-15. Effects on immune function assays following exposure to PFDA in short-term oral

studies in animals 3-76

Figure 3-16. Evaluation results for animal studies assessing effects of PFDA exposure on

general/observational immune assays 3-77

Figure 3-17. Effects on general/observational immune assays following exposure to PFDA in

short-term oral studies in animals 3-79

This document is a draft for review purposes only and does not constitute Agency policy.

vii DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Figure 3-18. Evaluation results for animal studies assessing effects of PFDA exposure on blood

leukocyte counts 3-80

Figure 3-19. Effects on blood leukocyte counts following exposure to PFDA in short-term oral

studies in animals 3-82

Figure 3-20. Evaluation results for animal studies assessing effects of PFDA exposure on

immune histopathology and organ weights 3-83

Figure 3-21. Effects on immune organ histopathology following exposure to PFDA in short-term

oral studies in animals 3-85

Figure 3-22. Effects on immune organ weights following exposure to PFDA in short-term oral

studies in animals 3-87

Figure 3-23. Evaluation results for animal study assessing effects of PFDA exposure on immune

function assays for sensitization and allergic response 3-88

Figure 3-24. Study evaluation results for twenty-nine epidemiological studies of birth weight

and PFDA 3-101

Figure 3-25. Twenty-eight perinatal studies of birth weight measures and subsets considered for

different analyses 3-102

Figure 3-26.PFDA and birth weight z-scores (overall population) 3-104

Figure 3-27.PFDA and birth weight z-score (sex-stratified) 3-104

Figure 3-28. Overall study population mean birth weight results for 22 PFDA epidemiological

studies 3-108

Figure 3-29. Sex-specific mean birth weight results for 14 PFDA epidemiological studies: boys

are above reference line, girls are below 3-109

Figure 3-30. Low birth weight/small for gestational age heatmap 3-110

Figure 3-31. Dichotomous fetal growth restriction (small for gestational age and low birth

weight) forest plot 3-112

Figure 3-32. Study evaluation results for 17 epidemiological studies of birth length and PFDA 3-112

Figure 3-33. Overall study population mean birth length results for 15 PFDA epidemiological

studies 3-116

Figure 3-34. Sex-stratified birth length results for 10 PFDA epidemiological studies 3-117

Figure 3-35.Study evaluation results for 14 epidemiological studies of head circumference and

PFDA 3-118

Figure 3-36. Overall population head circumference results in 11 epidemiological studies 3-119

Figure 3-37. Sex-stratified head circumference results in 8 epidemiological studies 3-120

Figure 3-38. Study evaluation results for four epidemiological studies of postnatal growth and

PFDA 3-128

Figure 3-39. PFDA and postnatal growth-standardized weight measures (overall population) 3-130

Figure 3-40. PFDA and postnatal growth mean weight (in grams) 3-130

Figure 3-41. PFDA and postnatal standardized growth weight (sex-stratified; boys above dashed

line, girls below) 3-131

Figure 3-42. PFDA and postnatal growth - standardized height measures (overall population) 3-132

Figure 3-43. PFDA and postnatal growth mean height (in centimeters) 3-132

Figure 3-44. PFDA and postnatal growth - standardized height measures (sex-stratified; boys

above reference line, girls below) 3-133

Figure 3-45. PFDA and postnatal growth standardized head circumference (overall population) 3-134

Figure 3-46. PFDA and postnatal growth head circumference (sex-stratified; boys above

reference line, girls below) 3-134

Figure 3-47. PFDA and postnatal growth measures-body mass index, adiposity, ponderal index,

waist circumference (overall population) 3-135

This document is a draft for review purposes only and does not constitute Agency policy.

viii DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Figure 3-48. PFDA and postnatal growth rapid growth (overall population) and sex-specific (in

grams) 3-137

Figure 3-49. Study evaluation results for two epidemiological studies of anogenital distance and

PFDA 3-139

Figure 3-50. Study evaluation results for five epidemiological studies of gestational duration and

PFDA 3-140

Figure 3-51. Preterm birth forest plot-six studies based on the overall population 3-142

Figure 3-52. Overall population forest plot of 10 gestational age studies 3-144

Figure 3-53. Sex stratified forest plot of five gestational age studies 3-144

Figure 3-54. Study evaluation results for two epidemiological studies of spontaneous abortion

and PFDA 3-148

Figure 3-55. Developmental animal study evaluation heatmap 3-151

Figure 3-56. PFDA fetal body weight after gestational exposure 3-152

Figure 3-57. PFDA developmental effects 3-155

Figure 3-58. Semen parameters epidemiology study evaluation heatmap 3-167

Figure 3-59. Male reproductive hormones epidemiology study evaluation heatmap 3-169

Figure 3-60. Male pubertal development epidemiology study evaluation heatmap 3-171

Figure 3-61. Evaluation results for animal study assessing effects of PFDA exposure on male

reproductions 3-172

Figure 3-62. Effects on sperm evaluations following exposure to PFDA in short-term oral studies

in animals 3-174

Figure 3-63. Effects on male reproductive organ histopathology following exposure to PFDA in

short-term oral studies in animals 3-176

Figure 3-64. Effects on serum testosterone levels following exposure to PFDA in short-term oral

studies in animals 3-178

Figure 3-65. Effects on male reproductive organ weights following exposure to PFDA in short-

term oral studies in animals 3-180

Figure 3-66. Study evaluations for epidemiology studies of PFDA and female reproductive

effects 3-191

Figure 3-67. Female reproductive animal study evaluation heatmap 3-198

Figure 3-68. PFDA female reproductive effects 201

Figure 3-69. Study evaluation results for epidemiology studies of PFDA and serum lipids 3-211

Figure 3-70. Study evaluation results for epidemiology studies of PFDA and cardiovascular risk

factors other than serum lipids 3-216

Figure 3-71. Study evaluation results for epidemiology studies of PFDA and cardiovascular

disease 3-219

Figure 3-72. Study evaluation results for epidemiology studies of PFDA and diabetes and insulin

resistance 3-220

Figure 3-73. Study evaluation results for epidemiology studies of PFDA and adiposity 3-227

Figure 3-74. Evaluation results for animal study assessing effects of PFDA exposure on

cardiometabolic effects 3-231

Figure 3-75. Cardiometabolic effects following exposure to PFDA in short-term oral studies in

animals 3-234

Figure 3-76. Study evaluation results for epidemiology studies of PFDA and neurodevelopmental

effects 3-241

Figure 3-77. Study evaluation results for epidemiology studies of PFDA and thyroid effects 3-248

Figure 3-78. Thyroid hormone levels animal study evaluation heatmap 3-251

Figure 3-79. PFDA thyroid hormone levels after short-term oral exposure 3-254

This document is a draft for review purposes only and does not constitute Agency policy.

ix DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Figure 3-80. Endocrine histopathology animal study evaluation heatmap	3-255

Figure 3-81. PFDA endocrine histopathology	3-256

Figure 3-82. PFDA endocrine organ weights animal study evaluation heatmap	3-257

Figure 3-83. PFDA endocrine organ weight	3-259

Figure 3-84. Urinary effects human study evaluation heatmap	3-270

Figure 3-85. Evaluation results for animal studies assessing effects of PFDA exposure on urinary

effects	3-272

Figure 3-86. Kidney histopathology effects following exposure to PFDA in 28-day rat study	3-274

Figure 3-87. Urinary effects following exposure to PFDA in short-term oral studies in animals	3-278

Figure 3-88. PFDA general toxicity animal study evaluation heatmap	3-283

Figure 3-89. PFDA general toxicity effects	3-285

Figure 3-90. Study evaluation results for epidemiology studies of PFDA and cancer	3-289

This document is a draft for review purposes only and does not constitute Agency policy.

x	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

ABBREVIATIONS AND ACRONYMS

AIC

Akaike's information criterion

NCEA

National Center for Environmental

ALT

alanine aminotransferase



Assessment

AST

aspartate aminotransferase

NCI

National Cancer Institute

atm

atmosphere

NOAEL

no-observed-adverse-effect level

ATSDR

Agency for Toxic Substances and

NTP

National Toxicology Program



Disease Registry

NZW

New Zealand White (rabbit breed)

BMD

benchmark dose

ORD

Office of Research and Development

BMDL

benchmark dose lower confidence limit

PBPK

physiologically based pharmacokinetic

BMDS

Benchmark Dose Software

PFAAs

perfluoroalkyl acids

BMR

benchmark response

PFCA

perfluoroalkylcarboxylic acids

BUN

blood urea nitrogen

PND

postnatal day

BW

body weight

POD

point of departure

CA

chromosomal aberration

POD [AD J]

duration-adjusted POD

CASRN

Chemical Abstracts Service registry
number

QSAR

quantitative structure-activity
relationship

CHO

Chinese hamster ovary (cell line cells)

RD

relative deviation

CL

confidence limit

RfC

inhalation reference concentration

CNS

central nervous system

RfD

oral reference dose

CYP450

cytochrome P450

RGDR

regional gas dose ratio

DAF

dosimetric adjustment factor

RNA

ribonucleic acid

DMSO

dimethylsulfoxide

SAR

structure activity relationship

DNA

deoxyribonucleic acid

SCE

sister chromatid exchange

DTH



SD

standard deviation

EPA

Environmental Protection Agency

SDH

sorbitol dehydrogenase

ER

extra risk

SE

standard error

FDA

Food and Drug Administration

SGOT

glutamic oxaloacetic transaminase, also

FEVi

forced expiratory volume of 1 second



known as AST

GD

gestation day

SGPT

glutamic pyruvic transaminase, also

GDH

glutamate dehydrogenase



known as ALT

GGT

y-glutamyl transferase

TSCATS

Toxic Substances Control Act Test

GLP

good laboratory practices



Submission

GSH

glutathione

TWA

time-weighted average

GST

glutathione-S-transferase

UF

uncertainty factor

HBCD

hexabromocyclododecane

UFa

animal-to-human uncertainty factor

Hb/g-A

animal blood: gas partition coefficient

UFd

database deficiencies uncertainty factor

Hb/g-H

human blood: gas partition coefficient

UFh

human variation uncertainty factor

HEC

human equivalent concentration

UFl

LOAEL-to-NOAEL uncertainty factor

HED

human equivalent dose

UFs

subchronic-to-chronic uncertainty

HERO

Health and Environmental Research



factor



Online

WOS

Web of Science

i.p.

intraperitoneal





IRIS

Integrated Risk Information System





i.v.

intravenous





LC50

median lethal concentration





LD50

median lethal dose





LOAEL

lowest-observed-adverse-effect level





MN

micronuclei





MNPCE

micronucleated polychromatic
erythrocyte





MOA

mode of action





MTD

maximum tolerated dose





This document is a draft for review purposes only and does not constitute Agency policy.

xi	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Assessment Managers (Lead Authors)

J. Phillip Kaiser, Ph.D.

Lucina E. Lizarraga. Ph.D.

EPA/ORD/CPHEA

Authors

Xabier Arzuaga. Ph.D.

Thomas F. Bateson. Sc.D., M.P.H.
I. Allen Davis. M.S.P.H.

Andrew Kraft. Ph.D.

Elizabeth Radke. Ph.D.

Hongvu Ru. Ph.D.

Paul Schlosser. Ph.D.

I. Michael Wright. Sc.D.

EPA/ORD/CPHEA

Contributors

Michelle M. Angrish. Ph.D.

Krista Christensen. M.P.H., Ph.D.
Ingrid Druwe. Ph.D.

Barbara Glenn, Ph.D. (retired)
Andrew Hotchkiss, Ph.D.
Stephanie Kim, Ph.D.
Christopher Lau, Ph.D.

Geniece M. Lehman. Ph.D.

Susan Makris. M.S. (retired)
Anuradha Mudipalli, Ph.D.
Kathleen Newhouse. M.S.
Kristen Rappazzo, Ph.D.

Tammy Stoker, Ph.D.

Andre Weaver, Ph.D.

Erin Yost. Ph.D.
lay Zhao. Ph.D.

EPA/ORD/CPHEA

Chris Corton, Ph.D.
Jason C. Lambert, Ph.D.

EPA/ORD/CCTE

April Luke, M.S

Andrew A Rooney, Ph.D.
Kyla Taylor, Ph.D.

Dori Germolec, Ph.D.

EPA/OLEM
DTT/NIEHS

Nora Abdel-Gawad
Alexis Agbai
Angela Scafidi

Oak Ridge Associated Universities (ORAU) Contractor

This document is a draft for review purposes only and does not constitute Agency policy.

xii	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Robyn B. Blain, Ph.D.	ICF

Alexandra E. Goldstone, M.P.H.

Alexander J. Lindahl, M.P.H.

Christopher A. Sibrizzi, M.P.H.

Production Team

Maureen Johnson (CPHEA Webmaster)	EPA/ORD/CPHEA

Ryan Jones (HERO Director)

Dahnish Shams (Project Management Team)

Jessica Soto-Hernandez (Project Management Team)

Vicki Soto (Project Management Team)

Samuel Thacker (HERO Team)

Garland Waleko (Project Management Team)

Rebecca Schaefer	Oak Ridge Associated Universities (ORAU) Contractor

Executive Direction

Wayne Cascio, M.D. (CPHEA Director)	EPA/ORD/CPHEA

V. Kay Holt, M.S. (CPHEA Deputy Director)

Samantha Jones, Ph.D. (CPHEA Associate Director)

Kristina Thayer, Ph.D. (CPAD Director)

Andrew Kraft, Ph.D. (IRIS PFAS Team Lead and CPAD

Associate Director)

Paul White, Ph.D. (CPAD Senior Science Advisor)

Ravi Subramanian, Ph.D. (Acting CPAD Science Advisor)

Elizabeth Radke-Farabaugh, Ph.D. (Branch Chief)

Janice Lee, Ph.D. (Branch Chief)

Viktor Morozov, Ph.D. (Branch Chief)

Glenn Rice, Ph.D. (Branch Chief)

Garland Waleko, M.S. (Acting Branch Chief)

Reviewers

CPAD Executive Review Committee

Kristina Thayer	CPAD Division Director

Paul White	CPAD Senior Science Advisor

Glenn Rice	CPHEA/CPAD/TEAB-C Branch Chief

Karen Hogan	CPHEA/CPAD/Emeritus

Alan Stern	NJDEP (retired), Contractor

Agency Reviewers

This assessment was provided for review to scientists in EPA's program and regional offices.
Comments were submitted by:

Office of the Administrator/Office of Children's Health Protection

Office of Air and Radiation/Office of Air Quality Planning and Standards

This document is a draft for review purposes only and does not constitute Agency policy.

xiii	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Office of Chemical Safety and Pollution Prevention
Office of Land and Emergency Management
Office of Water
Region 2, New York City, NY
Region 8, Denver, CO
Interagency Reviewers

This assessment was provided for review to other federal agencies and the Executive Office of
the President (EOP). Comments were submitted by:

The White House

•	Office of Science and Technology Policy

•	Office of Management and Budget
Department of Defense

Department of Health and Human Services

•	Agency for Toxic Substances and Disease Registry

•	National Institute of Environmental Health Sciences

•	National Institute of Occupational Safety and Health

This document is a draft for review purposes only and does not constitute Agency policy.

xiv	DRAFT-DO NOT CITE OR QUOTE

-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

EXECUTIVE SUMMARY

Summary of Occurrence and Health Effects

Perfluorodecanoic acid (PFDA, CASRN 335-76-2)1, and its related salts are members of the
group per- and polyfluoroalkyl substances (PFAS). This assessment applies to PFDA as well as salts
(including non-metal or alkali metal salts) of PFDA that would be expected to fully dissociate in
aqueous solutions of pH ranging from 4-9 (e.g., in the human body). Thus, while this assessment
would not necessarily apply to non-alkali metal salts of PFDA due to the possibility of PFDA-
independent contributions of toxicity, it does apply to PFDA salts including ammonium
perfluorodecanoate (PFDA NH4, CASRN 3108-42-7) and sodium perfluorodecanoate (PFDA-Na,
CASRN 3830-45-3), and other non-metal or alkali metal salts of PFDA. The synthesis of evidence
and toxicity value derivation presented in this assessment focuses on the free acid of PFDA, given
the currently available toxicity data2.

Concerns about PFDA and other PFAS stem from the resistance of these compounds to
hydrolysis, photolysis, and biodegradation, which leads to their persistence in the environment.
PFAS are not naturally occurring in the environment; they are synthetic compounds that have been
used widely over the past several decades in industrial applications and consumer products
because of their resistance to heat, oil, stains, grease, and water. PFAS in the environment are
linked to industrial sites, military fire training areas, wastewater treatment plants, and commercial
products (see Section 1.1.3. Environmental Fate and Transport for information specific to PFDA).

The Integrated Risk Information System (IRIS) Program is developing a series of five PFAS
assessments (i.e., perfluorobutanoic acid [PFBA], perfluorohexanoic acid [PFHxA],
perfluorohexanesulfonic acid [PFHxS], perfluorononanoic acid [PFNA], perfluorodecanoic acid
[PFDA], and their associated salts) (see December 2018 IRIS Program Outlook) atthe requestof
EPA National Programs. Specifically, the development of human health toxicity assessments for
exposure to these individual PFAS represents only one component of the broader PFAS strategic
roadmap atthe EPA (https://www.epa.gov/pfas/pfas-strategic-roadmap-epas-commitments-

1 The CASRN given here is for linear PFDA; the source PFDA used in the animal toxicity study NTP (2018) was
reported to be >97% pure, giving this CASRN. For the human studies [e.g., Valvi etal. (2017)] the purity of the
PFDA source was not provided by the study authors. None of the available studies explicitly state that only
the linear form was used. Therefore, there is the possibility that some proportion of the PFDA used in the
studies were branched isomers and thus observed health effects may apply to the total linear and branched
isomers in a given exposure source.

2Candidate values for different salts of PFDA were also calculated by multiplying the candidate value for the
free acid of PFDA by the ratio of molecular weights. For example, for the ammonium salt the ratio would be:
mw ammonium salt _ 531 _ ^ Qgg same	0f conversion can be applied to other salts of PFDA, such

MW free acid	514

as the potassium or sodium salts, using the corresponding molecular weights.

This document is a draft for review purposes only and does not constitute Agency policy.

xv	DRAFT-DO NOT CITE OR QUOTE

-------
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

action-2021-2024). The systematic review protocol (AppendixA) for these five PFAS assessments
outlines the related scoping and problem formulation efforts, including a summary of other federal
and state assessments of PFDA. The protocol also lays out the systematic review and dose-
response methods used to conduct this review (see also Section 1.2). The systematic review
protocol was released for public comment in November 2019 and was updated based on those
public comments. Appendix A links to the updated version of the protocol which summarizes the
history of the revisions.

Human epidemiological studies have examined possible associations between PFDA
exposure and health outcomes, in particular liver serum biomarkers, antibody responses,
sensitization and allergic responses, fetal growth restrictions, semen parameters, reproductive
hormones, pubertal development, neurodevelopment, thyroid hormones, urinary effects, serum
lipids, adiposity, cardiovascular disease, atherosclerosis, and cancer. With the exception of immune
[i.e., decreased antibody responses] and developmental [i.e., decreased birth weight], the ability to
draw judgments regarding these associations based on the available human evidence is limited by
the overall quality of the epidemiological studies (studies were generally low confidence), the few
studies per health outcome, and, in some studies, the lack of a quantifiable measure of exposure.

Animal studies of PFDA exposure exclusively examined the oral exposure route, and
therefore no inhalation assessment was conducted nor was an RfC derived (Section 5.2.3). The
available animal studies of oral PFDA exposure examined a variety of noncancer endpoints,
including those relevant to liver, immune, developmental, male, and female reproductive,
endocrine, urinary, cardiometabolic and other health effects. Limited evidence was identified
evaluating PFDA-induced carcinogenicity in animals.

Overall, the available evidence indicates that PFDA exposure is likely to cause liver,
immune, developmental, and male and female reproductive effects in humans, given sufficient
exposure conditions3. Specifically, for liver effects, the primary support for this hazard conclusion
included evidence of increased relative liver weights, altered serum biomarkers of liver injury (e.g.,
serum enzymes) and histopathology (including necrosis) in rats. For immune effects, the primary
supporting evidence included decreased antibody responses in children. Developmental effects
were identified as a hazard based primarily on consistent findings of dose-dependent decreases in
fetal weight in mice supported by evidence of decreased birth weight from studies of exposed
humans in which PFDA was measured during pregnancy. The primary basis for the hazard
judgment on male reproductive effects involved coherent responses across sperm counts,
testosterone levels, and male reproductive histopathology and organ weights in adult male rats. For
female reproductive effects, the primary hazard judgement was based on decreased uterus weight
and estrous cycle effects in adult female rats. Selected quantitative data from these identified

3 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes only and does not constitute Agency policy.

xvi	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1	hazards were used to derive lifetime and subchronic organ-specific reference doses (osRfDs) (see

2	Table ES-1) and the overall lifetime and subchronic RfDs (see Table ES-2).

3	The available evidence suggests that PFDA exposure might have the potential to cause

4	cardiometabolic and neurodevelopmental effects in humans under sufficient exposure conditions4

5	based on findings from human studies; however, due to issues regarding inconsistency, imprecision

6	and/or sensitivity, these health hazards were not used in the derivation of toxicity values. Likewise,

7	some human and animal evidence was also identified for endocrine, urinary, and other health

8	effects (e.g., hematological), but the evidence is inadequate to assess whether PFDA may cause

9	these health effects in humans and was not advanced for the derivation of toxicity values.

Table ES-1. Organ-specific RfDs for health effects with evidence available to
synthesize and draw summary judgments for the derivation of toxicity values

Organ/ System

Integration
judgment

Toxicity
value

Value
(mg/kg-day)

Confidence

UFa

UFh

UFs

UFl

UFd

UFC

Basis

Immune

Evidence
indicates
(likely)

Lifetime
osRfD and
subchronic
osRfD

4 x lO"10

Medium

1

10

1

1

3

30

Decreased serum antibody
concentrations for both
tetanus and diphtheria in
children at age 7 years and
PFDA measured at age 5 years
Grandiean et al. (2012):
(Budtz-J0rgensen and
Grandiean, 2018a)

Developmental

Evidence
indicates
(likely)

Lifetime
osRfD and
subchronic
osRfD

3 x 10"10

Medium-
low

1

10

1

1

3

30

Decreased birth weight in
male and female children
(Wikstrom et al., 2020)

Liver

Evidence
indicates
(likely)

Lifetime
osRfD

NDa

Subchronic
osRfD

7 x 10"7

Medium

3

10

10

1

3

1,000

Increased relative liver weight
in SD female rats (NTP, 2018)

Male

Reproductive

Evidence
indicates
(likely)

Lifetime
osRfD

NDa

Subchronic
osRfD

5 x 10"6

Medium-
Low

3

10

10

1

3

1,000

Decreased absolute whole
epididymis weight in SD rats
(NTP, 2018)

Female
Reproductive

Evidence
indicates
(likely)

Lifetime
osRfD

NDa

4 Given the uncertainty in this judgment and the available evidence, this assessment does not attempt to
define what might be the "sufficient exposure conditions" for developing these outcomes (i.e., these health
effects are not advanced for dose-response analysis in Section 5).

This document is a draft for review purposes only and does not constitute Agency policy.

xvii	DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Organ/ System

Integration
judgment

Toxicity
value

Value
(mg/kg-day)

Confidence

UFa

UFh

UFs

UFl

UFd

UFC

Basis





Subchronic
osRfD

3 x 10"6

Medium-
Low

3

10

10

1

3

1,000

Increased number of days
spent in diestrus in SD rats
(NTP, 2018)

aFor hepatic, male reproductive, and female reproductive effects, derivation of candidate lifetime values was not
attempted given the high degree of uncertainty associated with using PODs from a 28-day rodent study to protect
against effects observed in a chronic setting.

ND = not determined; RfD = reference dose (in mg/kg-d) for lifetime exposure; subchronic RfD = reference dose (in
mg/kg-d) for less-than-lifetime exposure; osRfD = organ- or system-specific reference dose (in mg/kg-d);
UFA = animal to human uncertainty factor; UFC = composite uncertainty factor; UFD = evidence base deficiencies
uncertainty factor; UFH = human variation uncertainty factor; UFL = LOAEL to NOAEL uncertainty factor;
UFS = subchronic to chronic uncertainty factor.

Table ES-2. Overall Lifetime and subchronic RfDs

Organ/ System

Integration
judgment

Toxicity
value

Value
(mg/kg-day)

Confidence

UFa

UFh

UFS

UFL

UFD

UFC

Basis

Immune/devel
opmental

Evidence
indicates
(likely)

Lifetime
osRfD and
subchronic
osRfD

4 x 10"10

Medium

1

10

1

1

3

30

Decreased serum antibody
concentrations for tetanus and
diphtheria in children at age 7
years and PFDA measured at
age 5 vears Grandiean et al.
(2012): (Budtz-J0rgensen
and Grandiean, 2018a)

Decreased birth weight in
male and female children
(Wikstrom et al., 2020)

ND = not determined; RfD = reference dose (in mg/kg-d) for lifetime exposure; subchronic RfD = reference dose (in
mg/kg-d) for less-than-lifetime exposure; osRfD = organ- or system-specific reference dose (in mg/kg-d);
UFA = animal to human uncertainty factor; UFC = composite uncertainty factor; UFD = evidence base deficiencies
uncertainty factor; UFH = human variation uncertainty factor; UFL = LOAEL to NOAEL uncertainty factor;
UFS = subchronic to chronic uncertainty factor.

This document is a draft for review purposes only and does not constitute Agency policy.

xviii	DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Lifetime and Subchronic Oral Reference Dose (RfD) for Noncancer Effects

Both of the identified hazards with quantitative information to support the derivation of
candidate lifetime values (i.e., immune, and developmental), specifically decreased serum antibody
concentrations in children (male and female) fBudtz-Targensen and Grandiean. 2018al: fGrandiean
etal.. 20121 and decreased birth weight (male and female) fWikstrom et al.. 20201 were selected as
the basis for the RfD of 4 x 10"10 mg/kg-day.5 6 The PODs for these two osRfDs were nearly identical
(i.e., 1.07x 10"8 and 9.6 x 10"9, respectively), amounting to a rounding difference once the identical
UFs were applied. The marginally higher osRfD for immune effects was the value used for the RfD
as confidence in that values were higher than confidence in the value for decreased birth weight
BMDLi/2si)(Fii i)) values for decreased antibody concentrations for both tetanus and diphtheria at age
7 years and PFDA measured atage 5 years were nearly identical (1.07 x 10_8and 1.06 x 10"8mg/kg-
day, respectively) and were used as the point of departure (POD) for this endpoint. For decreased
birth weight in males and females fWikstrom etal.. 20201. a BMDLsrd(hed) of 9.6 x 10"9 mg/kg-day
was identified for this endpoint and was used as the POD. The osRfDs for both outcomes were
calculated by dividing the PODhed by an identical composite uncertainty factor of 30 to account for
inter individual differences in human susceptibility (UFh = 10), and deficiencies in the toxicity
evidence base (UFd = 3). It is important to emphasize that both critical effects supporting this RfD
are observed during the developmental period.

The same approach was selected as the basis for the subchronic RfD of 4 x 10"10 mg/kg-day.
The subchronic and lifetime RfDs are identical given that the duration extrapolation uncertainty
factor (UFs) is 1 for both values. A UFs of 1 was selected since the immune and developmental
osRfDs are based on effects observed during the developmental period after exposure during
gestation, which is recognized as a susceptible lifestage; therefore, exposure during this time
window can be considered more relevant to the induction of sensitive effects on these outcomes
than chronic and subchronic exposures (see section 5.2.1 and 5.2.2 for more details).

Confidence in the Oral Reference Dose (RfD) and subchronic RfD

The overall confidence in the RfD and subchronic RfD is medium and is driven by medium
confidence in the immune osRfD (the developmental osRfD was medium-low confidence), noting
that there was medium confidence in the quantification of the PODs for both immune fBudtz-
Targensen and Grandiean. 2018al: fGrandiean etal.. 20121 and developmental fWikstrom etal..

5 The candidate values for different salts of PFDA would be calculated by multiplying the candidate value for
the free acid of PFDA by the ratio of molecular weights. For example, for the ammonium salt the ratio would
be: MW ammonium salt = Hi = 1.033. This same method of conversion can be applied to other salts of PFDA,

MW free acid 514

such as the potassium or sodium salts, using the corresponding molecular weights.

6 Note that the RfD for the free acid presented in this document and an RfD for the anion of PFDA
(perfluorodecanoate, C10F19O2", CASRN 73829-36-4) would be practically identical given the molecular
weights between the two compounds differ by less than 0.5% (i.e., by the weight of a single hydrogen atom).

This document is a draft for review purposes only and does not constitute Agency policy.

xix DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1	2020) endpoints using BMD modeling. (Budtz-l0rgensen and Grandiean. 2018al:fGrandiean et al..

2	20121.

3	Noncancer Effects Following Inhalation Exposure

4	No studies that examine toxicity in humans or experimental animals following inhalation

5	exposure were available and no acceptable physiologically based pharmacokinetic (PBPK) models

6	are available to support route-to-route extrapolation; therefore, no RfC was derived.

7	Evidence for Carcinogenicity

8	Under EPA's Guidelines for Carcinogen Risk Assessment fU.S. EPA. 20051. EPA concluded

9	there is inadequate information to assess carcinogenic potential for PFDA by either oral or inhalation

10	routes of exposure. Therefore, the lack of adequate data on the carcinogenicity of PFDA precludes

11	the derivation of quantitative estimates for either oral (oral slope factor [OSF]) or inhalation

12	(inhalation unit risk [IUR]) exposure.

This document is a draft for review purposes only and does not constitute Agency policy.

xx	DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1.OVERVIEW OF BACKGROUND INFORMATION
AND ASSESSEMENT METHODS

1.1. BACKGROUND INFORMATION ON PERFLUORODECANOIC ACID
(PFDA)

Section 1.1 provides a brief overview of aspects of the physicochemical properties, human
exposure, and environmental fate characteristics of perfluorodecanoic acid (PFDA; CASRN
335-76-2), and its related salts that might provide useful context for this assessment. This overview
is not intended to provide a comprehensive description of the available information on these topics.
The reader is encouraged to refer to source materials cited below, more recent publications on
these topics, and the assessment systematic review protocol (see Appendix A).

1.1.1. Physical and Chemical Properties

PFDA and its related salts are members of the group per- and polyfluoroalkyl substances
(PFAS). Buck etal. (20111 define PFAS as fluorinated substances that "contain 1 or more C atoms
on which all the H substituents (present in the nonfluorinated analogues from which they are
notionally derived) have been replaced by F atoms, in such a manner that they contain the
perfluoroalkyl moiety CnF2n+i-)." More specifically, PFDA is classified as a perfluoroalkyl carboxylic
acid (PFCA) fOECD. 20181. PFCAs containing seven or more perfluorinated carbon groups are
considered long-chain PFAS (ATSDR. 2018b). Thus, PFDA is a long-chain PFAS. The chemical
structures of PFDA and some of its related salts are presented in Figure l-l7. The physical-chemical
properties of PFDA and these related salts are provided in Table 1-1.

7 While this figure shows the linear structures, the assessment may also apply to other non-linear isomers of
PFDA and related salts as described in the Executive Summary.

This document is a draft for review purposes only and does not constitute Agency policy.

1-1 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts



PFDA

335-76-2



PFDA
ammonia salt
3108-42-7



PFDA
sodium salt
3830-45-3

Figure 1-1. Chemical structure of PFDA and related salts.

Table 1-1. Physical-chemical properties of PFDA and related salts

Property (unit)

Value

PFDA
335-76-2

PFDA
NH4+ salt
3108-42-7

PFDA
Na salt
3830-45-3

Molecular weight (g/mol)

514a

531c

536d

Melting point (°C)

79.5a

82.6a*

84.4a*

Boiling point (°C)

218a

2i2a*

2i2a*

Density (g/cm3)

1.79a*

1.76a*

1.76a*

Vapor pressure (mm Hg)

1.73-3a

2.39e-02a*

2.39e-02a*

Henry's law constant (atm-m3/mole)

1.5e-10a*

1.5e-10a*

1.5e-10a*

Water solubility (mol/L)

5.25e-3a

1.86a*

1.86a*

PKa

-0.17b*

ND

ND

LogP

4.15a

7.39a*

7.39a*

Soil adsorption coefficient (L/kg)

397a*

397a*

397a*

Bioconcentration factor (BCF)

39.3a

29.8a*

29.8a*

aU.S. EPA (2019b) U.S. EPA CompTox Chemicals Dashboard:
https://comptox.epa.gov/dashboard/dsstoxdb/results?search=PFDA. Median experimental values used where available;
otherwise, median, or average predicted values used. All values from the U.S. EPA CompTox Chemicals Dashboard were
accessed on May 24, 2022.
bATSDR (2018a)

ND = no data.

*Predicted value.

This document is a draft for review purposes only and does not constitute Agency policy.

1-1	DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1.1.2. Sources, Production, and Use

PFAS are not naturally occurring in the environment fATSDR. 2018al. They are synthetic
compounds that have been used widely over the past several decades in consumer products and
industrial applications because of their resistance to heat, oil, stains, grease, and water. This class
of chemicals has been used in consumer products including stain-resistant fabrics for clothing,
carpets, and furniture; nonstick cookware; and personal care products (e.g., dental floss, cosmetics,
and sunscreen) fATSDR. 2018a. b).

PFDA has been used in stain and grease-proof coatings on food packaging, furniture,
upholstery, and carpet fHarbison et al.. 20151. Kotthoff et al. f 20151 analyzed a variety of consumer
products for PFAS. PFDA was detected in nano- and impregnation-sprays, outdoor textiles, carpet,
gloves, paper-based food contact materials, ski wax, and leather.

EPA has been working with companies in the fluorochemical industry since the early 2000s
to phase out the production and use of long-chain PFAS such as PFDA

(https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/risk-management-and-
polyfluoroalkyl-substances-PFASl. However, the production and use of PFAS has resulted in their
release to the environment through various waste streams. Also, because products containing
PFAS are still in use, they may continue to be a source of environmental contamination due to
disposal or breakdown in the environment fKim and Kannan. 20071.

No Chemical Reporting Data (CDR) on production volume are available in EPA's ChemView
(U.S. EPA. 2019a) for PFDA or its salts. As part of the National Defense Authorization Act for Fiscal
Year 2020 (Section 7321), 172 per- and polyfluoroalkyl substances including PFDA were added to
the EPA's Toxic Release Inventory (TRI) list (https://www.epa.gov/toxics-release-inventory-tri-
program /tri-listed-chemicalsl. The reporting requirements apply to a de minimus limit of 1% and a
manufacture, process, or otherwise use threshold of 100 lbs. Currently, there is no quantitative
information on PFDA releases to the environment from facilities manufacturing, processing, or
otherwise available in EPA's Toxic Release Inventory.

Wang etal. (2014b) estimated global emissions of PFDA from direct and indirect
(i.e., formation degradation of precursors) sources between 1951 and 2030 to be 8 metric tons
based on a lower estimate and 222 metric tons based on a higher estimate. The lower estimate
assumes that producers cease production and use of long-chain PFCAs and their precursors in line
with global transition trends. The higher estimate assumes that the emission scenario in 2015
remains constant until 2030.

1.1.3. Environmental Fate and Transport

Long-chain PFAS, including PFDA, are considered very stable and persistent in the
environment fATSDR. 2018a: Harbison etal.. 2015). and can be found world-wide in the
environment, wildlife, and humans f https://www.epa. gov/assessing-a nd-ma nagi ng-chem icals-
under-tsca/risk-management-and-polvfluoroalkvl-substances-PFASl. Long-chain PFAS, including

This document is a draft for review purposes only and does not constitute Agency policy.

1-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

PFDA, have been found at private and federal facilities, and have been associated with various
sources, including aqueous film forming foam (AFFF) for fire suppression, and PFAS manufacturers
and industries that use PFAS (e.g., textiles) fATSDR. 2018al.

Although specific data on PFDA are lacking, PFAS that are released to air have been found
exist in the vapor phase in the atmosphere and resist photolysis, but particle-bound concentrations
have also been measured fKim and Kannan. 20071. Wet and dry deposition are potential removal
processes for particle-bound PFAS in air fATSDR. 2018a],

In soil, the mobility of PFAS will vary depending on their soil adsorption coefficients (see
Table 1-1), with PFDA being moderately mobile. Uptake of soil PFAS to plants has been shown to
occur for similar, long-chain PFAS such as PFOA fATSDR. 2018al. Yoo etal. f20111 estimated a
grass-soil accumulation factor (grass concentration divided by soil concentration) of 0.10 for PFDA,
based on samples collected from a site with bio-solids-amended soil.

The potential for PFAS to bioaccumulate in aquatic organisms is dependent on their
bioconcentration factors (see Table 1-1), with the potential for PFDA to bioaccumulate being high
compared to most of the other PFAS for which these data are available.

1.1.4. Potential for Human Exposure, including Populations and Lifestages with Potentially

Greater Exposure

The general population may be exposed to PFAS via inhalation of indoor or outdoor air,
ingestion of drinking water and food, and dermal contact with PFAS-containing products fATSDR.
2018a: NLM. 2017. 2013). Exposure may also occur via hand-to-mouth transfer of materials
containing these compounds fATSDR. 2018a). However, the oral route of exposure has been
considered the most important route of exposure among the general population. This conclusion is
based on several studies that have investigated the various routes of PFAS exposure fKlaunig etal..
20151. Other authoritative sources on exposure assessment (e.g., ATSDR) continue to do human
biomonitoring studies on PFAS, including PFDA, and those sources should be consulted for the most
up-to-date information on PFDA exposure in humans.

Gebbinketal. (2015) modelled exposure to PFDA among the adult general population.
'Intermediate' exposure (i.e., based on median inputs for all exposure parameters) from direct and
indirect (i.e., precursor) sources was estimated to be 67 pg/kg-day. Of the pathways evaluated
(i.e., ingestion of dust, food, water; inhalation of air), direct intake of PFDA in the diet accounted for
the largest portion of exposure for the intermediate scenario.

The presence of PFAS in human blood provides evidence of exposure among the general
population. PFAS have been monitored in the human population as part of the National Health and
Nutrition Examination Survey (NHANES). PFDA was measured in serum samples collected in
2013-2014 from more than 2,000 survey participants (CDC. 2022). The results of these analyses
are presented in Table 1-2. Olsen etal. f20171 analyzed human plasma samples from 616 American
Red Cross (AMC) donors for PFAS in 2015. The results were compared to results of similar
analyses conducted in 2002-2001, 2006, and 2010. Geometric mean concentrations of PFDA

This document is a draft for review purposes only and does not constitute Agency policy.

1-3 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

declined 50% from 2000-2001 to 2015. PFDA has also been detected in cord blood and human
milk fATSDR. 2018a). For example, Lankova etal. (20131 detected PFDA in 10% of human milk
samples collected from 50 Czech women at concentrations ranging from <6 to 12 pg/mL indicating
that breastmilk is a potential route of exposure for infants. Exposure can also occur through hand-
to-mouth transfer of materials containing these compounds fATSDR. 2018bl or in infants through
ingestion of formula reconstituted with contaminated drinking water.

Populations that may experience exposures greater than those of the general population
may include individuals in occupations that require frequent contact with PFAS-containing
products, such as firefighters or individuals who install and treat carpets fATSDR. 2018a). Also,
because PFDA can be found in ski wax, individuals who engage in professional ski waxing may be
more highly exposed because PFAS in dust may become airborne and inhaled during this process
fHarbison et al.. 20151. Nilsson et al. f20101 observed a significant correlation between the number
of years individuals had worked as ski wax technicians and their blood levels of PFDA.

Populations living near fluorochemical facilities where environmental contamination has
occurred may also be more highly exposed fATSDR. 2018a). Yamada etal. (20141 estimated
exposure to PFDA and other PFAS among high seafood consumers and high freshwater fish
consumers in France. Depending on how non-detects were handled (set to zero or the limit of
detection), mean estimates for PFDA were 0.16 to 0.73 ng/kg-day for high seafood consumers, and
0.42 to 0.96 ng/kg-day for high freshwater fish consumers, as compared to the adult general
population (0.00 to 0.34 ng/kg-day). Thus, populations with a large portion of their diet from fish,
including some tribal groups, may experience disproportionally greater PFDA exposure.

Table 1-2. Serum PFDA concentrations based on NHANES 2013-2014
data (ng/L)

Population group3

Value

Total population (N = 2,168)

Geometric mean

0.185

50th percentile

0.200

95th percentile

0.700

3 to 5 years (N = 181)

Geometric mean

50th percentile

0.100

95th percentile

0.370

6 to 11 years (N = 458)

Geometric mean

50th percentile

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Population group3

Value

20 years and older (N = 1,766)
Geometric mean
50th percentile
95th percentile

0.193
0.200
0.800

aThis table provides only general context on serum PFDA levels from a single study and within a narrow time-
period (environmental PFDA levels are changing over time). Note that PFDA is expected to bioaccumulate over a
lifetime (see Sections 1.1.3 and 3.1). Up-to-date information from authoritative bodies should be used in any
decisional context.

bNot calculated because the proportion of results below the limit of detection was too high to provide a valid
result.

cLimit of detection was 0.1.

LOD = limit of detection.

Source: CDC (2022). Fourth National Report on Human Exposure to Environmental Chemicals.

Air and Dust

PFDA is not currently listed as a hazardous air pollutant under the Clean Air Act and has not
been evaluated under the National Air Toxics Assessment (https://www.epa.gov/national-air-
toxics-assessment) nor the Air Toxics Screening Assessment fhttps: //www.epa.gov/AirToxScreen].
However, PFDA was measured at concentrations ranging from 0.13 to 1.56 pg/m3 in the vapor
phase and 0.13 to 0.49 pg/m3 in the particle phase of air samples collected from an urban area of
Albany, New York in 2006 fKim and Kannan. 20071.

PFAS, including PFDA, have also been measured in indoor air and dust and may be
associated with the indoor use of consumer products such as PFAS-treated carpets or other textiles
(ATSDR. 2018a). For example, Strvnar and Lindstrom (2008) analyzed dust samples from 110
homes and 10 daycare centers in North Carolina and Ohio in 2000-2001 and detected PFDA in
30.4% of the samples. Similar analyses were conducted by Karaskova etal. f 20161 who collected
56 dust samples from 41 homes in the Czech Republic, Canada, and the U.S in 2013. PFDA was
detected in more than 80% of the samples with mean concentrations of 5.2, 8.5, and 6.9 ng/g for the
Czech Republic, Canada, and the U.S., respectively. Knobeloch etal. (2012) collected vacuum
cleaner dust from 39 homes in Wisconsin in 2008 and detected PFDA in 72% of the samples at a
median concentration of 5.7 ng/g. Fraser etal. (2013) analyzed dust samples collected from offices
(n = 31), homes (n = 30), and vehicles (n = 13) in Boston, MA in 2009. PFDA was detected in 97% of
the office samples at concentrations ranging from 5.3 to 492 ng/g, 43% of the home samples at
concentrations ranging from 7.0 to 26.8 ng/g, and 69% of the vehicle samples at concentrations
ranging from 5.4 to 70.1 ng/g. Indoor air samples (n = 4) from a town in Norway collected between
2005-2006 had a mean concentration of 3.4 pg/m3 for PFDA (Barber etal.. 2007).

Water

U.S. EPA conducted monitoring for several PFAS in drinking water as part of the third
Unregulated Contaminant Monitoring Rule (UCMR) fU.S. EPA. 2016cl. However, PFDA was not
among the 30 contaminants monitored. Kim and Kannan f20071 analyzed lake water, rainwater,

This document is a draft for review purposes only and does not constitute Agency policy.

1-5 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

snow, and surface water from Albany, New York, and reported concentrations of PFDA ranging
from non-detectto 8.39 ng/L. Konwick etal. (20081 observed elevated PFDA concentrations (30-
113 ng/L) in a river in Georgia near the site of a wastewater land application system associated
with carpet manufacturing. Washington et al. f20101 analyzed soil samples from agricultural fields
in Decatur, AL where wastewater treatment sludges had been applied. PFDA was the PFAS with the
highest concentration with a maximum of 990 ng/g.

AFFF Training Sites

PFDA was detected at an Australian training ground where AFFFs had been used (Baduel et
al.. 20151. and Braunigetal. f20171 suggested that PFAS were distributed via groundwater to biotic
and abiotic matrices in an Australian town impacted by PFAS from a nearby fire-fighting training
site. Mean concentrations of PFDA were 0.12 [ig/L in water, 0.4 [ig/kg dry weight in soil, <0.2 |J.g/kg
wet weight in grass, 0.24 ng/g in egg yolk, 0.21-9.7 |ig/L in cow, sheep, and horse serum, and
0.4 ng/L in human serum.

Military Sites

PFDA was detected at 10 U.S. military sites in 67.0% of the surface soil samples, and 48.5%
of the sediment samples fATSDR. 2018a: Anderson etal.. 20161. Table 1-3 provides the
concentrations of PFDA in soil, sediment, surface water, and groundwater at these military sites.

Table 1-3. PFDA levels at 10 military installations

Media

Value

Surface soil

Frequency of detection (%)

67.03

Median (ng/kg)

0.980

Maximum (ng/kg)

15.0

Subsurface soil

Frequency of detection (%)

12.50

Median (ng/kg)

1.40

Maximum (ng/kg)

9.40

Sediment

Frequency of detection (%)

48.48

Median (ng/kg)

1.90

Maximum (ng/kg)

59.0

Surface water

Frequency of detection (%)

52.00

Median (ng/kg)

0.067

Maximum (ng/kg)

3.20

Groundwater

Frequency of detection (%)

34.78

Median (ng/kg)

0.023

Maximum (ng/kg)

1.80

Source: Anderson et al. (2016): ATSDR (2018a).

This document is a draft for review purposes only and does not constitute Agency policy.

1-6 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Other Exposures

Schecter et al. T20121 collected 31 food samples from 5 grocery stores in Texas in 2009 and
analyzed them for persistent organic pollutants, including PFDA, which was not detected (LOD = 0.2
ng/mL) in any of the foods. Chen etal. f2018bl analyzed PFAS, including PFDA, in foods in Taiwan.
PFDA was detected in a wide range of foods at geometric mean concentrations ranging from
0.94 ng/mL in milk to 22.2 ng/g in eggs. Heo etal. (20141 analyzed a variety of foods and
beverages in Korea for PFAS. PFDA was detected in 1.3% of the fruit and vegetable samples at a
mean concentration of 0.0002 ng/g; 12.8% of the meat samples at a mean concentration of
0.132 ng/g; 13.5% of the dairy samples at a concentration of 0.041 ng/g; 19.0% of the beverage
samples at a mean concentration of 0.019 ng/L; and 45.5% of the fish and shellfish samples at a
mean concentration of 0.056 ng/g. Heo etal. f20141 also detected PFDA in tap water and bottled
water in Korea at mean concentrations of 1.19 and 0.014 ng/L, respectively. Perez etal. f20141
analyzed PFAS in 283 food items (38 from Brazil, 35 from Saudi Arabia, 36 from Serbia, and 174
from Spain). PFDA was detected in 4.5, 3.4, and 2.1% of the samples from Brazil, Serbia, and Spain,
respectively. The mean concentrations of PFDA in foods from these countries were 170, 267, and
772 pg/g, respectively. Stahl etal. f20141 characterized PFAS in freshwater fish from 164 U.S.
urban river sites and 157 near-shore Great Lakes sites. PFDA was detected in fish from 20% of the
urban river samples (median =
-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

populations; in vitro and in silico models; and ADME and pharmacokinetic studies) and EPA's
judgment on whether the studies would have a material impact on the assessment conclusions (i.e.,
identified hazards or toxicity values) presented in the public comment draft. The external peer
reviewers are asked to consider EPA's disposition of these newly identified studies and make
recommendations, as appropriate (see Charge Question 1).

The literature search queries the following databases (no date or language restrictions were
applied):

• PubMed fNational Library of Medicine 1

• Web of Science (Thomson Reuters)

• Toxline (National Library of Medicine)

• TSCATS fToxic Substances Control Act Test Submissions!

In addition, relevant literature not found through database searching was identified by:

• Review of studies cited in any PECO-relevant studies and published journal reviews;
finalized or draft U.S. state, U.S. federal, and international assessments (e.g., the draft
Agency for Toxic Substances and Disease Registry [ATSDR] assessment released publicly in
2018). In addition, studies included in ongoing IRIS PFAS assessments (PFHxA, PFHxS,

PFNA, PFDA) were also scanned for any studies that met PFBA PECO criteria.

• Searches of published PFAS SEMs (Carlson etal.. 2022: Pelch etal.. 2022) starting in 2021.

• Review of studies submitted to federal regulatory agencies and brought to the attention of
EPA. For example, studies submitted to EPA by the manufacturers in support of
requirements under the Toxic Substances Control Act (TSCA).

• Identification of studies during screening for other PFAS. For example, epidemiology
studies relevant to PFDA were sometimes identified by searches focused on one of the other
four PFAS currently being assessed by the Integrated Risk Information System (IRIS)
Program.

• Other gray literature (e.g., primary studies not indexed in typical databases, such as
technical reports from government agencies or scientific research groups; unpublished
laboratory studies conducted by industry; or working reports/white papers from research
groups or committees) brought to the attention of EPA.

All literature is tracked in the U.S. EPA Health and Environmental Research Online (HERO)
database (https://heronetepa.gov/heronet/index.cfm/proiect/page/proiect id/2614). The PECO
criteria (see Table 1-4) identify the evidence that addresses the specific aims of the assessment and
to guide the literature screening process.

This document is a draft for review purposes only and does not constitute Agency policy.

1-8 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 1-4. Populations, exposures, comparators, and outcomes (PECO)
criteria

PECO element

Evidence

Populations

Human: Any population and lifestage (occupational or general population, including children and other
sensitive populations). The following study designs will be included: controlled exposure, cohort, case
control, and cross-sectional. (Note: Case reports and case series will be tracked as potential supplemental
material.)

Animal: Nonhuman mammalian animal species (whole organism) of any lifestage (including preconception,
in utero, lactation, peripubertal, and adult stages).

Other: In vitro, in silico, or nonmammalian models of genotoxicity. (Note: Other in vitro, in silico, or
nonmammalian models will be tracked as potential supplemental material.)

Exposures

Human: Studies providing quantitative estimates of PFDA exposure based on administered dose or
concentration, biomonitoring data (e.g., urine, blood, or other specimens), environmental or occupational
setting measures (e.g., water levels or air concentrations, residential location and/or duration, job title, or
work title). (Note: Studies that provide qualitative, but not quantitative, estimates of exposure will be
tracked as supplemental material.)

Animal: Oral or inhalation studies including quantified exposure to PFDA based on administered dose,
dietary level, or concentration. (Note: Nonoral and noninhalation studies will be tracked as potential
supplemental material.) PFDA mixture studies are included if they employ an experimental arm that
involves exposure to a single PFDA. (Note: Other PFDA mixture studies are tracked as potential
supplemental material.)

Studies must address exposure to the following: PFDA (CASRN 335-76-2), or PFDA ammonium salt
(CASRN 3108-42-7) or PFDA sodium salt (CASRN 3830-45-3).

Comparators

Human: A comparison or reference population exposed to lower levels (or no exposure/exposure below
detection levels) or for shorter periods of time.

Animal: Includes comparisons to historical controls or a concurrent control group that is unexposed,
exposed to vehicle only or air only exposures. (Note: Experiments including exposure to PFDA across
different durations or exposure levels without including one of these control groups will be tracked as
potential supplemental material [e.g., for evaluating key science issues; Section 2.4 of the protocol].)

Outcomes

All cancer and noncancer health outcomes. (Note: Other than genotoxicity studies, studies including only
molecular endpoints [e.g., gene or protein changes; receptor binding or activation] or other nonphenotypic
endpoints addressing the potential biological or chemical progression of events contributing towards toxic
effects will be tracked as potential supplemental material [e.g., for evaluating key science issues; Section
2.4 of the protocol]).

1 In addition to those studies meeting the PECO criteria and studies excluded as not relevant

2 to the assessment, studies containing supplemental material potentially relevant to the specific

3 aims of the assessment were inventoried during the literature screening process. Although these

4 studies did not meet PECO criteria, they were not excluded. Rather, they were considered for use in

5 addressing the identified key science issues (see Appendix A.2.4) and other potential scientific

6 uncertainties identified during assessment development but unanticipated at the time of protocol

7 posting. Studies categorized as "potentially relevant supplemental material" included the following:

This document is a draft for review purposes only and does not constitute Agency policy.

1-9 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

• In vivo mechanistic or mode of action studies, including nonPECO routes of exposure
(e.g., intraperitoneal injection) and populations (e.g., nonmammalian models)

• In vitro and in silico models

• Absorption, distribution, metabolism, and excretion (ADME) and pharmacokinetic studies
(excluding models)8

• Exposure assessment or characterization (no health outcome) studies

• Human case reports or case series studies

The literature was screened by two independent reviewers with a process for conflict
resolution, first at the title and abstract level and subsequently the full-text level, using structured
forms in DistillerSR (Evidence Partners; https://distillercer.com/products/distillersr-systematic-
review-software/). Literature inventories for PECO-relevant studies and studies tagged as
"potentially relevant supplemental material" during screening were created to facilitate subsequent
review of individual studies or sets of studies by topic-specific experts.

1.2.2. Evaluation of Individual Studies

The detailed approaches used for the evaluation of epidemiologic and animal toxicological
studies used in the PFDA assessment are provided in the systematic review protocol (see
Appendix A.6). The general approach for evaluating PECO-relevant health effect studies is the same
for epidemiology and animal toxicological studies, although the specifics of applying the approach
differ; thus, they are described in detail in Appendices A.6.2 and A.6.3, respectively. Approaches for
study evaluation for mechanistic studies are described in detail in Appendix A.6.5.

The key concerns for the review of epidemiology and animal toxicological studies are
potential bias (systematic errors or deviations from the truth related to internal validity that affect
the magnitude or direction of an effect in either direction) and insensitivity (factors that limit the
ability of a study to detect a true effect and can lead to a false negative). For example, any types of
random measurement error that may lead to attenuation of study results (i.e., bias towards the
null). In evaluating individual studies, two or more reviewers independently arrived at judgments
regarding the reliability of the study results (reflected as study confidence determinations; see
below) regarding each outcome or outcome grouping of interest; thus, different judgments were
possible for different outcomes within the same study. The results of these reviews were tracked
within EPA's version of the Health Assessment Workplace Collaborative (HAWC). To develop these
judgments, each reviewer assigned a rating of good, adequate, deficient (or not reported, which
generally carried the same functional interpretation as deficient), or critically deficient (listed from

8Given the known importance of ADME data, this supplemental tagging was used as the starting point for a
separate screening and review of pharmacokinetics data (see Appendix A.9.2 for details).

This document is a draft for review purposes only and does not constitute Agency policy.

1-10 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

best to worst methodological conduct; see Appendix A.6 for definitions) related to each evaluation
domain representing the different characteristics of the study methods that were evaluated based
on the criteria outlined in HAWC.

Once all evaluation domains were evaluated, the identified strengths and limitations were
collectively considered by the reviewers to reach a final study confidence classification:

• High confidence: No notable deficiencies or concerns were identified; the potential for bias
is unlikely or minimal, and the study used sensitive methodology.

• Medium confidence: Possible deficiencies or concerns were noted, but the limitations are
unlikely to be of a notable degree or to have a notable impact on the results.

• Low confidence: Deficiencies or concerns were noted, and the potential for bias or
inadequate sensitivity could have a significant impact on the study results or their
interpretation. Low confidence results were given less weight than high or medium
confidence results during evidence synthesis and integration (see Sections 1.2.4 and 1.2.5).

• Uninformative: Serious flaw(s) were identified that make the study results unusable.
Uninformative studies were not considered further, except to highlight possible research
gaps.

Using the HAWC platform the two reviewers reached a consensus judgment regarding each
evaluation domain and overall (confidence) determination with conflict resolution by an additional
reviewer, as needed. The specific limitations identified during study evaluation were carried
forward to inform the synthesis (see Section 1.2.4) within each body of evidence for a given health
effect.

1.2.3. Additional Epidemiology Considerations

While the detailed methods for epidemiology study evaluation are described in the
systematic review protocol (see Appendix A.6.2.1), a few considerations have been developed
further; these are described here.

As noted above, study sensitivity is an important consideration given that it could lead to
false negative (i.e., null) result (Type II error) if a study is underpowered or not designed with
adequate sensitivity to detect an association that may exist A key element for study sensitivity,
along with others described in the systematic review protocol, is whether exposure
contrasts/gradients are sufficient across populations to detect differences in risk. For example, if
measurement error results in inaccurate exposure estimates this can lead to exposure
misclassification but also influence the ability to detect an association as well as an exposure-
response relationship which may be evidence of a biologic gradient

Confounding across PFAS is a potential source of uncertainty when interpreting the results
of epidemiology studies of individual PFAS (e.g., quantifying the effect of an individual PFAS can

This document is a draft for review purposes only and does not constitute Agency policy.

1-11 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

potentially be confounded by other PFAS). In order for confounding to occur, co-pollutants would
have to be associated with PFAS of interest, associated with the endpoint, and not act as an
intermediate in the causal pathway. One way to begin to assess whether co-exposure is occurring is
through examination of correlations. In a preliminary analysis of 22 studies in the inventory
reporting correlations, correlations differed across the PFAS (see Appendix A.6, Figure 6-2). While
some pairs have correlation coefficients consistently above 0.6 (e.g., PFNA and PFDA), the
correlations for most vary from 0.1 to 0.6 depending on the study. For this reason, it was not
considered appropriate to assume that co-exposure to other PFAS was necessarily an important
confounder in all studies. The potential for confounding across PFAS is incorporated in individual
study evaluations and assessed across studies in evidence synthesis. In most studies, it is difficult
to determine the likelihood of confounding without considering additional information not typically
included in individual study evaluation (e.g., associations of other PFAS with the outcome of
interest and correlation profiles of PFAS within and across studies). In addition, even when this
information is considered or the study authors perform analyses to adjust for other PFAS, it is often
not possible to fully disentangle the associations due to high correlations. This stems from the
potential for amplification bias in which bias can occur following adjustment of highly correlated
PFAS fWeisskopf et al„ 20181. Thus, in most studies, there may be some residual uncertainty about
the risk of confounding by other PFAS. A "Good" rating for the confounding domain is reserved for
situations where there is minimal concern for substantial confounding across PFAS as well as other
sources of confounding. Examples of this include results for a PFAS that predominates in a
population (such as a contamination event) or studies that demonstrate robust results following
multi-PFAS adjustment which would also indicate minimal concern for amplification bias. Because
of the challenge in evaluating individual studies for confounding across PFAS, this issue is also
assessed across studies during the evidence synthesis phase, as described in the systematic review
protocol (Appendix A, Section 9), primarily when there is support for an association with adverse
health effects in the epidemiology evidence (i.e., moderate, or robust evidence in humans, as
described below). Analyses used include comparing results across studies in populations with
different PFAS exposure mixture profiles, considering results of multi-pollutant models when
available, and examining strength of associations for other correlated PFAS. In situations where
there is considerable uncertainty regarding the impact of residual confounding across PFAS, this is
captured as a factor that decreases the overall strength of evidence (see Appendix A.10).

1.2.4. Data Extraction

The detailed data extraction approach is provided in Appendix A.8. Briefly, data extraction
and content management were carried out using HAWC for all health effects for animal studies and
some health effects for epidemiological studies. Data extraction elements that were collected from
epidemiological, animal toxicological, and in vitro studies is described in HAWC
(https://hawcprd.epa.gov/about/). Not all studies that meet the PECO criteria went through data
extraction: studies evaluated as being uninformative were not used to inform assessment

This document is a draft for review purposes only and does not constitute Agency policy.

1-12 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

judgments and therefore did not undergo full data extraction. All findings from informative studies
were considered for extraction, regardless of the statistical significance of their findings. The level
of extraction for specific outcomes within a study may differ (i.e., ranging from a narrative to full
extraction of dose-response effect size information). For quality control, data extraction was
performed by one member of the evaluation team and independently verified by at least one other
member. Discrepancies in data extraction were resolved by discussion or consultation within the
evaluation team.

1.2.5. Evidence Synthesis and Integration

For the purposes of this assessment, evidence synthesis and integration are considered
distinct but related processes (see Appendix A, Sections 9 and 10 for full details). For each assessed
health effect, the evidence syntheses provide a summary discussion of each body of evidence
considered in the review that directly informs the integration across evidence to draw an overall
judgment for each health effect The available human and animal evidence pertaining to the
potential health effects are synthesized separately, with each synthesis providing a summary
discussion of the available evidence that addresses considerations regarding causation that are
adapted from Hill fl9651. Mechanistic evidence is also synthesized as necessary to help inform key
decisions regarding the human and animal evidence; processes for synthesizing mechanistic
information are covered in detail in Appendix A, Section 9.2.

The syntheses of the human and animal health effects evidence focus on describing aspects
of the evidence that best inform causal interpretations, including the exposure context examined in
the sets of studies. Thus, data permitting, the evidence synthesis emphasizes studies of high and
medium confidence. Correspondingly, during data extraction when a relative abundance of
medium and high confidence studies was available for a given health outcome the low confidence
studies did not generally undergo full data extraction. Documentation of when this approach was
taken is noted in the specific health effect sections. When possible, results across studies are
compared using graphs and charts or other data visualization strategies. The synthesis of
mechanistic information informs the integration of health effects evidence for both hazard
identification (e.g., biological plausibility or coherence of the available human or animal evidence;
inferences regarding human relevance, or the identification of susceptible populations and
lifestages across the human and animal evidence) and dose-response evaluation (e.g., selection of
benchmark response levels, selection of uncertainty factors). Evaluations of mechanistic
information typically differ from evaluations of phenotypic evidence (e.g., from routine
toxicological studies). This is primarily because mechanistic data evaluations consider the support
for and involvement of specific events or sets of events within the context of a broader research
question (e.g., support for a hypothesized mode of action; consistency with known biological
processes), rather than evaluations of individual apical endpoints considered in relative isolation.

This document is a draft for review purposes only and does not constitute Agency policy.

1-13 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Following the synthesis of human and animal health effects data, and mechanistic data,
integrated judgments are drawn across all lines of evidence for each assessed health effect During
evidence integration, a structured and documented two-step process is used, as follows:

• Building from the separate syntheses of the human and animal evidence, the strength of the
evidence from the available human and animal health effect studies is summarized in
parallel, but separately, using a structured evaluation of an adapted set of considerations
first introduced by Sir Bradford Hill fHill. 19651. This process is conceptually similar to that
used by the Grading of Recommendations Assessment, Development, and Evaluation
(GRADE) (Morgan et al.. 2016: Guvatt etal.. 2011: Schiinemann et al.. 20111. which arrives
at an overall integration conclusion based on consideration of the body of evidence. These
summaries incorporate the relevant mechanistic evidence (or mode of action [MOA]
understanding) that informs the biological plausibility and coherence within the available
human or animal health effect studies. The terms associated with the different strength of
evidence judgments within evidence streams are robust, moderate, slight, indeterminate,
and compelling evidence of no effect.

• The animal, human, and mechanistic evidence judgments are then combined to draw an
overall judgment that incorporates inferences across evidence streams. Specifically, the
inferences considered during this integration include the human relevance of the animal
and mechanistic evidence, coherence across the separate bodies of evidence, and other
important information (e.g., judgments regarding susceptibility). Note that without
evidence to the contrary, the human relevance of animal findings is assumed. The final
output is a summary judgment of the evidence base for each potential human health effect
across evidence streams. The terms associated with these summary judgments are evidence
demonstrates, evidence indicates (likely), evidence suggests, evidence inadequate, and strong
evidence of no effect. The decision points within the structured evidence integration process
are summarized in an evidence profile table for each considered health effect

As discussed in the protocol (Appendix A), the methods for evaluating the potential
carcinogenicity of PFAS follow processes laid out in the EPA cancer guidelines fU.S. EPA. 20051:
however, for PFDA, data relevant to cancer were sparse which limited the extent of analysis that
was possible (see Section 3.3).

1.2.6. Dose-Response Analysis

The details for the dose-response employed in this assessment can be found in
Appendix All. Briefly, a dose-response assessment was performed for noncancer health hazards,
following exposure to PFDA via the oral route, as supported by existing data. For oral noncancer
hazards, oral reference doses (RfDs) are derived when possible. An RfD is an estimate, with
uncertainty spanning perhaps an order of magnitude, of an exposure to the human population
(including susceptible subgroups) that is likely to be without an appreciable risk of deleterious
health effects over a lifetime (U.S. EPA. 20021. The derivation of a reference value like the RfD

This document is a draft for review purposes only and does not constitute Agency policy.

1-14 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

depends on the nature of the health hazard conclusions drawn during evidence integration. For
noncancer outcomes, a dose-response assessment was conducted for evidence integration
conclusions of evidence demonstrates or evidence indicates (likely). In general, toxicity values are
not developed for noncancer hazards with evidence suggests conclusions (see Appendix A, Section
10.2 for exceptions). Consistent with EPA practice, the PFDA assessment applied a two-step
approach for dose-response assessment that distinguishes analysis of the dose-response data in the
range of observation from any inferences about responses at lower environmentally relevant
exposure levels (U.S. EPA. 2012a. 20051:

• Within the observed dose range, the preferred approach was to use dose-response
modeling to incorporate as much of the data set as possible into the analysis. This modeling
to derive a point of departure (POD) ideally includes an exposure level near the lower end
of the range of observation, without significant extrapolation to lower exposure levels.

• As derivation of cancer risk estimates and reference values nearly always involves
extrapolation to exposures lower than the POD; the approaches to be applied in these
assessments are described in more detail in Section A.11.2.

When sufficient and appropriate human and laboratory animal data are available for the
same outcome, human data are generally preferred for the dose-response assessment because use
of human data eliminates the need to perform interspecies extrapolations. For reference values,
this assessment will derive a candidate value from each suitable data set Evaluation of these
candidate values will yield a single organ/system-specific value for each organ/system under
consideration from which a single overall reference value will be selected to cover all health
outcomes across all organs/systems. While this overall reference value represents the focus of
these dose-response assessments, the organ/system-specific values can be useful for subsequent
cumulative risk assessments that consider the combined effect of multiple PFAS (or other agents)
acting at a common organ/system. For noncancer toxicity values, uncertainties in these estimates
are characterized and discussed.

For dose-response purposes, EPA has developed a standard set of models
f http: //www.epa.gov/bmds] that can be applied to typical data sets, including those that are
nonlinear. In situations where there are alternative models with significant biological support
(e.g., toxicodynamic models), those models are included as alternatives in the assessment(s) along
with a discussion of the models' strengths and uncertainties. EPA has developed guidelines on
modeling dose-response data, assessing model fit, selecting suitable models, and reporting
modeling results [see the EPA Benchmark Dose Technical Guidance (U.S. EPA. 2012a)]. For each
modeled response, a POD from the observed data was estimated to mark the beginning of
extrapolation to lower doses. The POD is an estimated dose (expressed in human-equivalent
terms) near the lower end of the observed range without significant extrapolation to lower doses.

This document is a draft for review purposes only and does not constitute Agency policy.

1-15 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1 The POD is used as the starting point for subsequent extrapolations and analyses. For noncancer

2 effects, the POD is used in calculating the RfD.

This document is a draft for review purposes only and does not constitute Agency policy.

1-16 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

2.LITERATURE SEARCH RESULTS

2.1. LITERATURE SEARCH AND SCREENING RESULTS

The database searches yielded 1057 unique records, with 536 records identified from
additional sources, such as posted National Toxicology Program (NTP) study tables, review of
reference lists from other authoritative sources fATSDR. 2018bl and searches of published PFAS
SEMs fPelch etal.. 20221 (see Figure 2-1). No unique studies were identified in submissions to EPA.
Of the 1057 studies identified, 595 were excluded during title and abstract screening, and 443 were
reviewed at the full-text level. Of the 443 studies screened at the full-text level, 262 were
considered to meet the populations, exposures, comparators, and outcomes (PECO) eligibility
criteria (see Table 1-4). The PECO criteria identify the evidence that addresses the specific aims of
the assessment and focuses the literature screening, including study inclusion/exclusion. In
addition to those studies meeting the PECO criteria, studies containing supplemental material
potentially relevant to the specific aims of the assessment were tagged during the literature
screening process. Although these studies did not meet PECO criteria, they were not excluded.
Rather, they were considered for use in addressing the identified key science issues and other
major scientific uncertainties identified during assessment development but unanticipated at the
time of protocol posting. Studies categorized as "potentially relevant supplemental material"
included the following:

• In vivo mechanistic or mode-of-action studies, including non-PECO routes of exposure (e.g.,
intraperitoneal injection) and non-PECO populations (e.g., nonmammalian models);

• In vitro and in silico models;

• Absorption, distribution, metabolism, and excretion (ADME) and pharmacokinetic (PK)
studies (excluding models);

• Exposure assessment or characterization (no health outcome) studies; and

• Human case reports or case-series studies

The studies meeting PECO at the full-text level included 234 epidemiologic studies, 14
animal studies, 1 PBPK model, and 8 in vitro/in vivo genotoxicity studies. Of the 1057 studies
screened, 374 were identified as supplemental material during title and abstract or full text
screening and tagged by topic area (e.g., mechanistic or MOA, non-PECO route of exposure, etc.).
High-throughput screening data on perfluorodecanoic acid (PFDA) are currently available from the
EPA's Chemicals Dashboard (U.S. EPA. 2021a). data were retrieved on November 2022) and

This document is a draft for review purposes only and does not constitute Agency policy.

2-1 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

relevant information is presented and analyzed in Appendix E, The last literature search update
used for the Toxicological Review was June 2022.

PFDA

Literature Searches (through June 2022)

Other
From draft ATSDR
assessment (n = 21)
Submitted to EPA (n = 0)
NTP report (n=l)
Published PFAS SEMs
^ (n = 514) J

PubMed WOS ToxLine TSCATS
(n = 868) (n = 850) (n = 123) (n = 1)

TITLE AND ABSTRACT SCREENING

Title & Abstract Screening
(1057 records after duplicate removal)

Excluded (n= 595)
• Not relevant to PECO (n = 595)

FULL TEXT SCREENING

Full-Text Screening

(n = 443)

Studies Meeting PECO (n = 262)

• Human health effects studies (n = 234)

• Animal health effect studies (n = 14)

• Genotoxicity studies (n = 8)

• PBPK models (n = 1)

• Accessory records, such as
supplementary material for included
studies (n = 5)

Excluded (n = 35)

not relevant to PECO (n = 11), review,
commentary, or letter (n = 15), abstract-only
(n = 8), unable to obtain full text (n = 1)

Tagged as Supplemental (n = 374)
mechanistic or MOA (n = 142), ADME (n =
31), exposure assessment or qualitative
exposure only (n = 125), mixture-only (n =
6), non-PECO route of exposure (n = 88),
case report or case study (n = 0), other (n
=112)

Figure 2-1. Literature search for perfluorodecanoic acid and related salts.

2.2. SUMMARY OF STUDIES MEETING PECO CRITERIA

Human and animal studies have evaluated potential effects to the liver, immune system,
developing fetus, male and female reproductive systems, endocrine, cardiometabolic,
neurodevelopmental, urinary, general toxicity and other organ systems (e.g., hematology) following
exposure to PFDA. The evidence base for these outcomes is synthesized in Sections 3.2.1-3.2.11. A
limited number of available studies in humans and animals informing of potential carcinogenic

This document is a draft for review purposes only and does not constitute Agency policy.

2-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

effects with PFDA exposure are summarized in Section 3.3. The single identified PBPK model is
discussed in Section 3.1.

Two hundred and thirty-four epidemiological studies were identified that report on the
potential association between PFDA and non-cancer and cancer human health effects (list of studies
filterable by health effect available at:

https://hawcprd.epa.gov/summary/visual/assessment/100500072/Epi-studies-of-PFDA-health-
effects/). The database of animal toxicity studies for PFDA consists of oral exposure studies (see
Table 2.1), including five dietary exposure studies in rats exposed for 7-14 days (Yamamoto and
Kawashima. 1997: Kawashima etal.. 1995: Permadi etal.. 1993: Takagi etal.. 1992.1991): two
drinking water studies in mice exposed for 12-49 days fLi etal.. 2022: Wang etal.. 20201. two
28-day gavage studies in rats and/or mice fFrawlev etal.. 2018: NTP. 20181. one 14-day oral study
(presumed to be gavage) in mice fLee and Kim. 20181 and one gestational exposure study in mice
via gavage with two exposure windows (GD 10-13 and 6-15) (Harris and Birnbaum. 1989). In
addition, three single exposure studies in animals via the oral route were identified with limited
utility for the evaluation of repeat-dose toxicity and the derivation of oral reference dose (RfD)
values (Kawabata etal.. 2017: Brewster and Birnbaum. 1989: Harris etal.. 1989).

Table 2-1. Animal toxicity studies examining health effects after PFDA
administration

Author (year)
Reference

Species, strain (sex)

Exposure route and duration

Dose range3

NTP (2018)

Rat, Harlan Sprague-
Dawley (male and
female)

Oral gavage; daily over 28 days

0, 0.156, 0.312, 0.625, 1.25 and
2.5 mg/kg-d

Frawlev et al.
(2018)

Rat, Harlan Sprague-
Dawley (female)

Oral gavage; daily over 28 days

0, 0.125, 0.25 and 0.5 mg/kg-d

Frawlev et al.
(2018)

Mouse, B6C3F1/N
(female)

Oral gavage; weekly over 28 days

0.04464, 0.0893, 0.179, 0.36
and 0.71 mg/kg-d (reported as
0, 0.3125, 0.625, 1.25, 2.5 and 5
mg/kg-wk)

Takagi et al.
(1991)

Rat, F344 (male)

Diet; daily over 14 days

0,10 mg/kg-d (reported as 0
and 0.01%)

Lee and Kim
(2018)

Mouse, ICR (male)

Uncharacterized (presumed to be
oral gavage); days 9,11, and 13
over 14 days

0 and 21.4 mg/kg-d (reported as
0 and 100 mg/kg)

Li et al. (2022)

Mouse, C57BL/6J
(female)

Drinking water; daily for 14 days

0 and 25 mg/kg-d

Wang et al.
(2020)

Mice, CD-I (male)

Drinking water; daily over 12-49
days

0,13 mg/kg-d (reported as 0
and 0.1 mM)

Permadi et al.
(1993)

Mouse, C57BL/6 (male)

Diet; daily over 10 days

0, 37.8 mg/kg-d (reported as 0
and 0.02%)

This document is a draft for review purposes only and does not constitute Agency policy.

2-3 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Author (year)
Reference

Species, strain (sex)

Exposure route and duration

Dose range3

Kawashima et al.
(1995)

Rat, Wistar (male)

Diet; daily over 7 days

0, 1.15,2.3, 4.6, and 9.22
mg/kg-d (reported as 0,00125,
0.0025, 0.005, and 0.01%)

Yamamoto and

Kawashima

(1997)

Rat, Wistar (male)

Diet; daily over 7 days

0 and 4.6 mg/kg-d (reported as
0 and 0.005%)

Takagi et al.
(1992)

Rat, Fisher F344 (male)

Diet; daily over 7 days

0 and 10 mg/kg-d (0 and 0.01%)

Harris and
Birnbaum (1989)

Mouse, C57BL/6N
(female)

Oral gavage; GD 10-13

0, 0.25, 0.5, 1, 2, 4, 8, 16, and 32
mg/kg-d

Harris and
Birnbaum (1989)

Mouse, C57BL/6N
(female)

Oral gavage; GD 6-15

0, 0.03, 0.1, 0.3, 1, 3, 6.4, and
12.8 mg/kg-d

Doses are presented as adjusted daily doses (ADD). Additional details on the ADD conversions can be found in the
HAWC project page for PFDA.

GD = gestational day.

1 Graphical representations of outcome-specific study evaluations are presented and

2 discussed within the hazard sections outlined above. Detailed rationales for each domain and

3 overall confidence rating are available in Health Assessment Workspace Collaborative (HAWC).

This document is a draft for review purposes only and does not constitute Agency policy.

2-4 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.PHARMACOKINETICS, EVIDENCE SYNTHESIS, AND
INTEGRATION

3.1. PHARMACOKINETICS

Perfluorodecanoic acid (PFDA) and its salts have characteristics of absorption, distribution,
metabolism, and excretion (ADME) comparable to other perfluoroalkyl acids (PFAAs) in that they
are readily absorbed by gastrointestinal tract following oral exposure irrespective of sexes and
species. Both animal and human data suggest that PFDA has a high affinity for protein binding and
efficient renal reuptake. Therefore, PFDA tends to accumulate in organs to the extent similar to or
greater than that of other PFAAs and has relatively slow clearance fDzierlenga etal.. 2019: Fuiii et
al.. 2015: Zhang etal.. 2013b). In general, PFDA accumulates primarily in liver, followed by kidney,
blood, and other tissues. PFDA is specifically a perfluorocarboxylic acid (PFCA), which are a subset
of PFAAs. Similar to other PFAAs, PFDA is also metabolically inert and therefore most of PFDA is
eliminated unchanged in urine and feces.

Of note, growing mechanistic evidence (both animal and human) suggests that renal
clearance becomes less efficient as the perfluorocarbon chain length increases fDzierlenga etal..
2019: Kudo. 2015: Lau. 20151. The findings support previous reports indicating that fecal
elimination may play an increasingly important role in elimination of long carbon chain-length of
PFAAs like PFDA (C10) as compared to other shorted chain of PFAAs (C £ 8) (Vanden Heuvel et al..
19911. Collectively, these PFDA pharmacokinetic data support the conclusions of Kudo f20151 and
Lau f20151 that PFDA has a much longer half-life than other shorter chained PFAAs (e.g., PFHxA).
While female rats administered PFDA tended to have a higher dose-normalized area under the
plasma concentration time curve (AUC) than males, fDzierlenga etal.. 20191 suggested that there
was no sex difference in PFDA half-life. However, calculating the average clearance across studies,
doses, and routes, the EPA obtains a value of 6.1 mL/kg-d in male rats and 4.3 mL/kg-day in female
rats, i.e., 30% lower in female rats. The elimination half-life of PFDA is generally much longer in
humans (4.5-12 years) than in rats (20-59 days) or mice (63-222 days). By comparison, Lau
f20151 provides estimated half-lives of 2.3-3.8 years for PFOA in humans (modestly lower than
PFDA), 5.4 years for PFOS (comparable to PFDA), but only 32 days for PFHxA. In rats PFOA has a
half-life of 2-6 days, PFOS a half-life of 38-71 days and PFHxA 0.4-1.6 hours. So, the qualitative
trend with chain-length and structure is similar, but there is an order of magnitude difference in
elimination of PFOA vs. PFOS in rats but in humans the difference between PFOA and PFOS is no
more than a factor of two.

This document is a draft for review purposes only and does not constitute Agency policy.

3-1 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.1.1. Absorption

Bioavailability (or fractional absorption) is typically estimated by comparing the AUC of
blood concentrations observed after an oral dose as compared to when the same dose is given IV, or
with the AUCs normalized to the dose. If kinetics is linear and oral uptake is less than 100% the
AUC after oral dosing will be less than that after IV dosing, and the fraction absorbed (Fabs) is
estimated as AUC (oral)/AUC (IV).

In the most recent animal study by Dzierlenga etal. (2019). Hsd: Sprague Dawley (SD) rats
were given PFDA or one of two other PFAA (perfluorohexanoic acid, PFHxA and perfluorooctanoic
acid, PFOA) by intravenous (IV) injection at 2 mg/kg or oral gavage (2,10, and 20 mg/kg). It was
found that the time to peak concentration (Tmax) increases with the chain length of PFAAs and
slightly with dose levels of oral administration for both sexes. For PFDA, Tmax (mean ± standard
error of mean, hour) increases from 8.27 ± 0.63 to 10.0 ± 0.06 hour, and from 9.01 ± 0.80 to
10.8 ± 1.2 hour, with increased gavage doses (2-20 mg/kg) of PFDA for males and females,
respectively. Peak concentration, Cmax (normalized with dose levels, mM/mmol/kg), also
appeared to be somewhat higher with increasing oral doses in female rats, but not male rats. Oral
bioavailability for PFDA was estimated to be 160-180% for both sexes. The nominal observation
of >100% absorption may be the result of enhanced reabsorption by intestinal transporters
fDzierlenga et al.. 20191. Other aspects of the results do not indicate nonlinearity; for example,
AUC/dose did not change significantly among oral doses of 2,10, and 20 mg/kg. But the peak
concentration after the IV dose (2 mg/kg), measured just 5 minutes after dosing, was lower than
the value estimated from the oral dose data, occurring 8-9 hours after the dose. Hence, there is no
clear explanation for this observation of AUC/dose being so much higher after oral vs. IV dosing.
Given the consistency of AUC/dose for the oral doses, it seems most likely that the error occurred
for the IV dose.

Kim etal. f20191 estimated Fabs = 0.87 ± 0.25 and 0.65 ± 0.08 in female and male SD rats,
respectively. On the other hand Dzierlenga et al. (2019) reported Fabs of 1.58-1.72 and 1.70-1.79
for male and female rats, respectively for 2-20 mg/kg doses, based on the serum AUC after oral vs.
IV dosing. It is not clear how to interpret these data nor resolve the discrepancy between the two
papers. In the Bayesian PK analysis of the rat data described in Appendix G 100% bioavailability
(Fabs = 1) was assumed and the variation in AUC/dose is then interpreted as variation and effective
uncertainty in the volume of distribution and clearance. Hence, the quantitative uncertainty is
accounted for.

Fuiii etal. (2015) measured PFDA PK in male and female FVB/NJcl mice dosed with
0.313 [imol/kg by IV administration and 3.13 [imol/kg by oral gavage. A shortcoming of the
experimental design is that serum concentration data were only collected up to 24 hours, making it
harder to estimate the PK parameters. However, based on the reported parameters the Tmax was
12 and 15 hours in male and female rats, respectively. Fuiii etal. f20151 reported the ratio of dose-
adjusted AUC after oral and IV exposures as 1.1 and 1.2 in male and female mice, respectively,

This document is a draft for review purposes only and does not constitute Agency policy.

3-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

indicating complete absorption. That these values are slightly greater than 1 may not only be due to
experimental variability, but also because clearance might have been slightly slower for the oral
dose, which was 10 times higher than the IV dose. Since the AUC/dose of values of Fuiii etal. f20151
are not significantly different for oral vs. IV dosing, the difference is presumed due to experimental
variability and the results are interpreted as showing 100% bioavailability (Fabs = 1) in mice.

Although there is no direct evidence of oral absorption of PFDA in humans, it can be
inferred from observations in epidemiological studies that identified positive associations between
PFDA concentrations in human tissues (e.g., blood or placenta) and environmental levels
(e.g., drinking water) (Stubleski et al.. 2016). Given these results for rats and mice and the lack of
controlled PK studies in humans, Fabs = 1 will be used for humans.

No data on absorption of PFDA through the respiratory tract or skin has been found. While
oral ingestion is considered the primary route of exposure, the contribution from these other routes
would need to be better evaluated in the scientific literature to determine their significance.

3.1.2. Distribution

General considerations

Upon absorption, PFDA moves rapidly through the body via the bloodstream to various
organs and tissues, mainly liver, lung, and kidney and, to a lesser extent, brain, and bone
fDzierlenga et al.. 2019: Vanden Heuvel et al.. 19911. In general, PFDA tends to accumulate in
organs to an extent greater than or similar to that of other PFAS. It has been suggested that the
extent of the covalent binding of PFDA with biological matrices (e.g., serum proteins) in blood and
tissues is critical to its distribution and bioaccumulation (Kudo. 2015: Vanden Heuvel et al.. 1992).
For instance, Kim etal. f20191 measured binding of PFDA to plasma proteins in vitro to incorporate
this factor into a PBPK model and reported that more than 99.7% was bound to protein in rat and
human plasma. These measured values were in line with animal experimental data reported by
Ylinen and Auriola (1990) that 99% of PFDA was bound with the serum proteins in Wistar rats with
a single intraperitoneal (IP) dose of 20 mg/kg PFDA. However, if distribution to tissues is assumed
to be limited by the product of free fraction and tissue blood flow, the PK distribution phase is
predicted to be much longer than observed. Hence the plasma binding must be labile, not strictly
limiting its distribution or clearance.

Of note, the degree of protein binding with PFDA affects not only its distribution but also the
elimination. Specifically, the PFAS-serum protein complex mediates glomerular filtration since
only the unbound fraction is filtered (Kudo. 2015). PFAS can then be extensively resorbed as fluid
carrying them passes down the renal tubules, with this resorption mediated by other PFAS-protein
complexes, specifically by organic anion transporters (Oat) proteins (Kudo. 2015). For instance,
Weaver etal. (2010) investigated the roles of rat renal Oat proteins in the deposition of
perfluorinated carboxylates with different chain lengths of carbons (C2-C18). The transport of
PFDA (C10) was measured from 10 to 300 mM with renal Oatproteins (Chinese hamster ovary cell

This document is a draft for review purposes only and does not constitute Agency policy.

3-3 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

line and kidney RNA from Sprague-Dawley rats). Of five Oat proteins (Oatl, 0at2, 0at3, Uratl, and
Oatplal), Oatplal appears to be the major Oat protein responsible for the reabsorption of C8
through CIO, with highest affinities for C9 and PFDA (CIO). These data collectively suggest that
chain length is a factor in the extent to which PFAAs are substrates of various basolateral and apical
transporters in renal proximal tubule cells, which in turn impacts the extent of elimination.
Moreover, since saturation of these transporters will lead to nonlinearity in elimination, one can
expect that PFAAs which are significant substrates will have greater nonlinearity in their
elimination (as a function of exposure level) compared to PFAAs which for which the transporters
have lesser affinity. While this is a general expectation, the PK data of Dzierlenga et al. (2019) did
not exhibit nonlinear elimination with single doses in the range of 2-20 mg/kg, and it is not known
if transporter saturation would occur with higher doses or multiple doses (leading to accumulation
of PFDA) in this dose range.

Animals (rats and miceJ

Distribution in rats and mice was examined in multiple toxicological studies of PFDA.
Vanden Heuvel et al. (1991) specifically evaluated [1-14C] PFDA pharmacokinetics in rats and
observed distribution into all tissues examined, including liver, kidney, heart, and gonads. Tissue
levels outside of the liver were less than 1% of the administered dose in male rats and less than 2%
of the dose in female rats. Vanden Heuvel etal. T19921 then examined the covalent binding of PFDA
to protein in male rat at 2 hour, 1 and 4 days after intraperitoneal dosing with 4.8 mg/kg [1-14C]
PFDA and found that ~0.1% of the administered dose was bound in plasma and liver and ~0.25%
was bound in testes (results independent of sample time). So, while a large fraction of PFDA in the
body is found in the liver, only a small fraction of that is covalently bound. While only a small
fraction of the total dose is covalently bound, the quantity could be enough to interfere with
estimation of long-term clearance or half-life.

Other investigators measured distribution into multiple tissues, most commonly kidney,
liver, and brain (Dzierlenga etal.. 2019: Kim etal.. 2019: Fuiii etal.. 2015). Although PFDA can be
found in the brain, the accumulation of PFDA was generally lower in the brain than in other organs
or tissues, while the highest levels were found in liver. For instance, Kawabata etal. (2017)
observed that the hepatic concentration of PFDA (mg/g tissue) was approximately 60 times higher
than that of the brain in Wistar rats given a single-oral dose of 50 mg PFDA/kg. This measurement
was made 9 days after the dose was administered, which should be a sufficient time for distribution
among the tissues to equilibrate but is short enough compared to overall clearance to represent a
significant portion of the administered dose.

Volume of Distribution in Rats and Mice

Ohmori etal. (2003) estimated the volume of distribution (Vd) for PFDA in male and female
Wistar rats (three each sex) after IV administration (48.64 mmol/kg BW) as 347.7 ± 15.2 and 441.1
± 55.1 mL/kg, respectively, for male and female Wistar rats (three each sex). The Vd of PFDA

This document is a draft for review purposes only and does not constitute Agency policy.

3-4 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

obtained by Ohmori etal. (20031 only varied slightly by sex though up to 2-fold larger than those of
other PFAAs tested in the same experiment (PFHA, PFOA, or PFNA). This sex difference is in
contrast with two most recent studies showing that Vd was larger in males than in females
fDzierlenga etal.. 2019: Kim etal.. 20191. For instance, Dzierlenga et al. f20191 investigated the
disposition of PFDA in Hsd: SD rats administered 2 mg/kg PFDA by IV and found Vd for the central
compartment (VI) was slightly larger in males (274 ±28 mL/kg) than females (238 ± 35 L/kg)
while the peripheral (V2) distribution was almost twice as large in males (355 ± 69 mL/kg) than in
females (186 ± 57 mL/kg). Summing VI and V2 for these results from Dzierlenga etal. (20191. the
total Vd in males is estimated to be 50% higher than females. Dzierlenga etal. (20191 also obtained
a larger Vd in males vs. females when PFDA was given by oral gavage, similar to their results from
IV dosing.

Kim etal. f20191 reported total volumes of distribution (i.e., not normalized to BW) for their
IV exposure: 0.1182 L and 0.0584 L for male and female rats, respectively. However, if one assumes
a BW of 0.25 kg then the Vd obtained is consistent with the reported Cmax values, i.e., Vd =
dose/Cmax. Given these absolute volumes of distribution and 0.25 kg BW, Vd values were
estimated to be 472.7 and 233.77 mL/kg for male and female rats in the Kim etal. (20191. which
are quite similar to the values reported by Dzierlenga et al. f20191 (see Table 3-1).

There are limited data on ADME properties of PFDA in mice. Fuiii etal. f20151 evaluated the
PK of PFDA in FVB/NJc mice aged 8-10 weeks using single IV dose (0.31 [imol/kg) and oral gavage
(3.13 [imol/kg). Unlike rats, while the Vd (mean ± standard deviation, mL/kg) was slightly larger in
males (250 ± 60) than females (200 ± 50) after IV administration, the difference in PFDA
distribution was not significant. Of note, once entering the body via IV administration, most of
PFDA were retained in the liver of mice (64-80% for males, 46-55% for females). The overall
distribution profiles of gavage route were similar to those of IV route fFuiii etal.. 20151.

The Vd values from the mouse and rat studies are summarized in Table 3-1 along with
results for rats from a hierarchical Bayesian analysis from partial pooling of the data, described in
Appendix G.

This document is a draft for review purposes only and does not constitute Agency policy.

3-5 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-1. Volume of distribution values reported for animal studies

Dose

Volume of distribution

Study

Strain

Route

(mg/kg)

(miykg)3

Male Rats

629 ± 97

Dzierlenga et al. (2019)

Hsd: SD

770.6 (485.8-1077.3)
586 ± 57

Dzierlenga et al. (2019)

Hsd: SD

Oral

579.3 (475.4-670.6)
411 ±46

Dzierlenga et al. (2019)

Hsd: SD

Oral

417.6 (357.8-475.7)
456 ± 35

Dzierlenga et al. (2019)

Hsd: SD

Oral

459.3 (396.3-523.8)
472.7 ± 37.2b

Kim et al. (2019)

317.9 (259.2-381.7)

Kim et al. (2019)

oral

448.5 (376.6-524.5)c

Ohmori et al. (2003)

Wistar

412.7(331.9-520.0)

Population mean (90% CI)

431.1 (108.9-696.4)

Female Rats

424 ± 92

Dzierlenga et al. (2019)

Hsd: SD

350.4 (294.7-405.2)
277 ± 35

Dzierlenga et al. (2019)

Hsd: SD

Oral

241.4 (203.1-278.9)
264 ± 36

Dzierlenga et al. (2019)

Hsd: SD

Oral

227.6 (190.9-262.9)
270 ± 40

Dzierlenga et al. (2019)

Hsd: SD

Oral

223.9 (187.6-260.3)
233.7 ± 17.8b

Kim et al. (2019)

252.9 (218.8-288.2)

Kim et al. (2019)

oral

450.9 (385.8-509.8)°
441.1 ±55.1

Ohmori et al. (2003)

Wistar

448.9 (413.9-482.7)

Population mean (90% CI)

313.4 (193.2-438.1)

Male Mice

Fuiiietal. (2015)

FVB/NJcl

0.16

250 ± 60

Female Mice

Fuiiietal. (2015)

FVB/NJcl

0.16

200 ± 50

Values in plain text are as reported for each study unless otherwise noted. Values in italics are the mean (90%
credible interval) from the Bayesian analysis described in Appendix G.

Kim et al. (2019) reported Vd as 118.18 ± 9.31 and 58.42 ± 4.46 mLfor male and female rats, respectively, after IV
exposures. These were normalized to an assumed 0.25 kg BW, which is consistent with Vd calculated as
dose/Cmax, given that Cmax is the initial concentration for IV dosing.

Kim et al. (2019) did not report Vd for oral doses.

This document is a draft for review purposes only and does not constitute Agency policy.

3-6 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Distribution of PFDA in mice and rats during pregnancy/gestation has not been evaluated.

Distribution in Humans

While PFDA is distributed throughout the body, tissue concentrations are expected to be
roughly 50% of the concentration in blood plasma based on the animal data presented above.

Based on this concentration ratio, most of the PFDA mass may be in various tissues since they
constitute over 90% of the body. Despite the expected mass apportionment, it is appropriate that
measurement of blood PFDA has been extensively applied to assess PFDA exposure for humans and
will be used in this review to estimate the risk from that exposure.

A recent study evaluated levels of several PFAS, including PFDA, in human serum as a
function of various measures of body composition as well as localized measurements of adipose
content throughout the body generated by dual-energy X-ray absorptiometry (DXA) and whole-
body magnetic resonance imaging (WB-MRI) fLind etal.. 20221. In women the study showed a
negative correlation between serum PFDA concentration and many measures of body fat, as well as
with the volume of areas of the body with high fat fractions, although much less so with the volume
of these regions. For example, there is a negative correlation between serum PFDA and the volume
of hips and inner thighs in women, but no correlation with the fat content of these regions fLind et
al.. 20221. In men the study showed no association between serum PFDA concentration and
measures of body composition. Given the minimal distribution of PFDA to adipose tissues seen in
rats (Kim etal.. 20191 one might expect essentially no effect of the volume of these tissues on serum
levels, as was seen in men. However, one would predict a negative correlation between Vd and
body fat, and in fact the results in women appear to be consistent with that prediction if glomerular
filtration increases with body mass or surface area, as will be discussed in the excretion section. It is
also possible that the correlation is due to variation in exposure related to body fat, where in the
male population exposure (per kg BW) was constant with body fat but for some reason exposure
decreased with body fat among the women. Matched estimates of exposure from dietary surveys or
samples, or matched measures of urinary clearance (PFAS concentrations in urine) are ultimately
needed to determine whether or not the correlations actually reflect PK variation.

Human Distribution During Pregnancy and Lactation PFDA can also be found in human
breast milk, placenta, embryo/fetal tissues, and cord blood (Mamsen etal.. 2019: Mamsen etal..
2017: Zhang etal.. 2013a: Liu etal.. 2011: Karrman et al.. 2009: Monrov etal.. 20081. Mamsen etal.
f20191 and Mamsen et al. f20171. examined fetal tissues after voluntary abortions (first trimester)
or intrauterine fetal death (second and third trimester; fMamsen et al.. 201911. More specifically,
Mamsen etal. (20171 reported time-matched maternal serum and fetal tissue levels from fetuses
between 36 and 65 days of age (i.e., between 5 and 10 weeks); the data appeared to show an
increasing trend in tissue concentration with fetal age, but the trend was not statistically significant.
When tissues were analyzed separately, PFDA concentration in placenta, liver and lung were
likewise found to increase with trimester, but were not detected in heart, CNS, or adipose tissue
fMamsen et al.. 20191. Also, the first trimester data were from women with a mean age of 26.5

This document is a draft for review purposes only and does not constitute Agency policy.

3-7 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

years, while the second and third trimester were from women with a mean age of 32.5 years.
Interestingly, first trimester maternal serum concentrations (mean 0.34 ng/mL) were somewhat
higher than the second and third trimester concentrations (mean 0.26 and 0.27 ng/mL,
respectively), though the difference was not statistically significant The ratio of placenta
concentration to first-trimester maternal serum indicates a strong time-trend in distribution to the
placenta, but this trend was also not statistically significant and the ratio of fetal liver and lung to
placenta did not show a consistent pattern with trimester (Mamsen et al.. 20191. In summary,
while some of the data are indicative of a time-dependence in the ratio of placental and fetal tissue
to maternal serum levels, none of those results are statistically significant and other aspects of the
data indicate that the ratio is constant. Therefore, it will be assumed that the relative volume of
distribution (L/kg BW) is constant during pregnancy in the PK modeling, as this also simplifies that
analysis.

To compare the distribution between tissues and maternal blood matrices among different
studies, adjustment should be made to correct for the distribution among blood components.
Specifically, Poothong et al. (20171 measured a mean ratio of 1.7 for serum: whole blood and 1.3 for
plasma: whole blood concentrations of PFDA. These factors will be used to adjust the subsequent
tissue: blood matrix ratios to tissue: plasma, when reported for whole blood or serum. If the ratio
of serum: whole blood concentration is 1.7 and hematocrit (hct) is 45%, then the mass fraction of
PFDA in plasma, given this ratio, would be Fp = 1.7 x (1-hct) = 93.5%. Using the reported plasma:
whole blood ratio and the same calculation, one obtains Fp = 1.3 x (1-hct) = 71.5%. Partitioning of
PFDA and other PFAAs between human plasma and blood cells were also investigated by Tin etal.
(20161. The estimated mass fraction in plasma (human samples) increased among perfluoroalkyl
carboxylates as the carbon chain length increased from C6 (mean 0.24) to Cll (0.87) with the mean
of 0.82 for PFDA (C10), which corresponds to a plasma: whole blood concentration ratio of 0.82/(1-
hct) = 1.5. Since this value is intermediate between the serum: whole blood and plasma: whole
blood values reported by Poothong etal. f20171. it will be used to convert tissue partitioning data
relative to whole-blood concentrations to serum-based concentrations below.

While the placenta shares circulation from the mother and fetus, it is the only tissue for
which PFDA concentrations in adult humans can be compared to plasma to evaluate overall
distribution. Mamsen et al. (20171 reported time-matched maternal serum and placenta tissue
levels from fetuses between 37 and 68 days of age (i.e., between 5 and 10 weeks), and obtained
mean placenta: maternal plasma ratio of 43%. The results for this ratio from Mamsen etal. f20191.
shown graphically, indicate mean values of about 40 and 55% for the second and third trimester,
respectively. The ratio of placenta to maternal serum (estimated from blood) at birth measured by
Zhang etal. (2013a) was 34% (both mean and median ratio; n = 32). These results for the placenta
are generally consistent with the volume of distribution (L/kg) measured in female rats, described
above. Given thatthe average Vd in female rats obtained above (0.38 L/kg) is based on a larger set
of studies, which show a fair amount of variability between them (indicating that results from a

This document is a draft for review purposes only and does not constitute Agency policy.

3-8 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

single study may not be reliable), the Vd in humans will be assumed to be the same as in rats; with
the results for male rats being used for men and results for female rats for women.

The observed ratio ranges from 25-55%, with most data between 30 and 40%. Mean
concentrations in lung and intestines were slightly greater than placenta (shown graphically in
Figure 2 ofMamsen etal. f2017)). while other tissues were below, and the reported mean ratio of
fetal tissue to maternal plasma was 27%. This indicates that distribution into the fetus as a whole is
50-80% of the range in placenta (34-55%).

Studies of the volume of distribution in newborns are not available, but one can reasonably
assume that it is similar to fetal tissues. Mamsen etal. (2017) specifically reported PFDA
concentrations in first trimester fetal liver, heart, intestine, lung, connective tissue, spinal cord, ribs,
and extremities. Results for individual tissues were only shown graphically, but most fetal tissues
had a mean concentration lower than the placenta, with mean maternal plasma concentration of
0.28 ng/g, placenta of 0.09 mg/g (43% of plasma), and fetal tissue of 0.05 ng/g (27% of plasma).
While Mamsen etal. (2019) had fewer data for PFDA in older fetuses, from their supplemental data
mean levels in first trimester fetal livers were less than those in the placenta (Mamsen et al.. 2017).
As will be discussed in the section on PK modeling below, the impact on distribution of the
pregnant mother with her fetus will not be large, but these results are informative of distribution
within the fetus, which will be imputed to newborns.

Several studies evaluated the cord serum: maternal serum ratio in humans at childbirth,
with the following median (mean) values reported or calculated from the reported median (mean)
concentrations in each matrix:

• Liu etal. C2011): 0.42 (0.39);

• Needham etal. (2011): 0.29 (mean not reported):

• Zhang etal. C2013a): 0.28 C0.251:

• Han etal. C2018): 0.38 (0.38 GM ratio);

• Yang etal. C2016a): 0.25 C0.351:

• Yang etal. f2016b1: 0.39 (0.43);

• Li etal. f2020a) (preterm): 0.23:

• Li etal. (2020a) (full-term): 0.35.

The average of the median values from these studies is 0.324, indicating that the placenta
creates a significant barrier for PFDA between maternal and fetal blood. But beyond this overall
average, fLi etal.. 2020a) observed a significant increase in the cord/maternal serum ratio between
preterm and full-term pregnancies, from a median ratio of 0.23 to 0.35. The authors evaluated the
correlation of the cord/maternal serum ratio with multiple placental transporters and identified a
significant, positive correlation with p-glycoprotein (MDR1) and multidrug resistance-associated
protein 2 (MRP2). These positive correlations, significant for full-term but not preterm pregnancies,
indicate that the placenta acts as a passive barrier to PFDA in early pregnancy and this function is
partly defeated by the expression of MDR1 and MRP2 transporters late in pregnancy.

This document is a draft for review purposes only and does not constitute Agency policy.

3-9 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

However, if the ratio of fetal serum to maternal serum is 0.324 (32.4%) and the ratio of fetal
tissue to maternal serum is 27%, then the ratio of fetal tissue to fetal serum would be
27%/32.4% = 83%, a much higher level of distribution than observed in adult rats and estimated
for the adult woman (38%) based on placental data described above.

Since the total body burden of PFDA in the human PK studies is unknown, it is not possible
to directly estimate Vd in humans. For male and female rats, the estimated (geometric mean)
Vd values are 448 and 287 mL/kg, respectively (see Table 3-1). As described above, the fetal tissue:
maternal plasma ratio varied between 0.25 and 0.55, with Mamsen et al. (2017) reporting a mean
fetal tissue: maternal plasma ratio of 0.27, which is 80% of the average Vd in female rats (assuming
1 L/kg body density). These data indicate that fetal tissue levels are close to maternal levels: if the
maternal Vd was that of female rats (0.287 L/kg) and fetal: maternal serum was 0.27, that implies
similar average concentration in the fetus as the mother, which is not indicated by the comparison
of fetal tissue and placenta concentration. Therefore, it will be assumed that Vd for women (both
prior to and during pregnancy) is equal to the geometric mean for female rats, 287 mL/kg. For
consistency, the Vd for men will be assumed equal to the average for male rats (448 mL/kg).

One can then ask if the somewhat different distribution into the fetus would impact the
overall distribution in the mother and fetus together (e.g., for PK modeling during pregnancy). If
one presumes that distribution into the fetus is fast compared to the rate of fetal development, such
that the concentration in maternal and fetal tissues remains at equilibrium and recognizes that the
fetus is less than 5% of the combined maternal and fetal mass, then the impact of slightly lower
distribution into the fetus on distribution in the mother and fetus will be minimal. Hence, human
maternal Vd is likely to be unchanged during pregnancy. The available data do not indicate a
difference greater than 10% or 20%.

Since PFDA binds strongly to serum proteins, one possible explanation for the apparently
higher distribution between fetal serum and tissues is that the fetus has a much lower level of these
proteins than an adult, allowing for a greater proportion of PFDA in fetal tissue vs. fetal serum.
However, data to support this hypothesis, i.e., measurements of PFDA binding in cord blood, are not
available. Pharmacokinetic modeling of PFOA dosimetry in humans by Goedenetal. (2019)
suggests another hypothesis: that the greater amount of extracellular water in the tissues of fetuses
and children (Friis-Hansen. 1961) leads to a greater distribution of PFAS into these tissues. The
amount of extracellular water in newborns was estimated to be 2.4 times higher than adults fFriis-
Hansen. 19611. Multiplying the volume of distribution from female rats (38%) by 2.4, one obtains
91%, which is much closer to the estimate of 81% obtained here. Hence, while the mechanism by
which distribution in a fetus, which we assume also applies to newborns, might not be the
difference in extracellular tissue water, the available quantitative data for extracellular water can
provide a reasonable prediction for the difference between newborns and adults, as well as the
transition between them (see Figure 3-1).

This document is a draft for review purposes only and does not constitute Agency policy.

3-10 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Ratio of ExtracellularTissue Water in Children vs. Adults

^ a

2.4 <

2.2
2

.2 1.8

** * r-

DC 1.6

1.4
1.2

1
0.8

) 1

3 4 5

Age (y)

7 8 £

) 1

Figure 3-1. Ratio of extracellular water (% of body weight) in children vs.
adults. Values (points) are calculated from results in Friis-Hansen (1961) and
plotted at the mid-point for the corresponding age ranges evaluated.

The interpolation function shown in Figure 3-1 can be multiplied by the adult Vd (L/kg) to
obtain the corresponding value for children under 10 years of age, as was done by Goeden et al.
f20191. However, an opposing factor is the approximately 20% larger blood volume as a fraction of
BW in young children compared to older children and adults fDarrowetal.. 19281. given that a high
fraction of PFHxS is bound to blood proteins. Hence, the extent and even the direction of any change
in Vd with age are uncertain and will require further PK studies to address.

Liu etal. (20111 also investigated correlations between PFDA concentrations in matched
maternal serum and breast milk samples collected from their subjects. The median value for the
concentration ratio between milk and maternal serum was 0.03:1, hence indicating a rather limited
level of lactational transfer to infants.

Effect of Liver Disease on Human Distribution

In a cross-sectional study by Yeungetal. (20131. the authors investigated the role of liver
disease in the deposition of PFDA by analyzing the distribution of PFDA in serum and liver using
samples from patients with hepatocellular carcinoma (HCC) and cirrhosis due to chronic hepatitis C
viral infection, (HCV); while the mean and median liver: serum ratios were higher in HCC (0.65 and
0.66) patients than HCV (0.41 and 0.33, respectively), the difference was not significantly different.
While the ratio of liver-to-serum PFDA concentration were not evaluated in control subjects of this
study, the comparison of absolute liver concentrations of other PFAS in healthy vs. diseased
samples indicated that pathological changes in diseased livers can alter the liver: serum PFAS
distribution.

This document is a draft for review purposes only and does not constitute Agency policy.

3-11 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.1.3. Metabolism

Vanden Heuvel et al. T19911 examined the metabolism of PFDA in male and female Wistar
rats administrated with a single intraperitoneal (IP) dose of [1-14C] PFDA (9.4 [imol/kg, 5 mg/kg).
The results showed that only parent compound of PFDA was found in urine or feces, suggesting that
there was no appreciable metabolism of PFDA. The findings are expected since PFDA is a
long-chain (CIO) PFAAs with chemical stability similar to that of other shorter length PFAA
chemicals (e.g., perfluorohexane sulfonic acid, PFHxS, C6). Although there have been no studies of
PFDA biotransformation following inhalation or dermal exposure, metabolism by these
administration routes is similarly not expected.

3.1.4. Excretion

In general excretion is one component of overall elimination of substances in the body, the
other being metabolism. Total elimination is often evaluated by observing the decline in
concentration of a compound in the blood or other tissues. Since PFDA does not undergo
appreciable metabolism, as discussed just above, the elimination data discussed below are
interpreted as measures of total excretion.

Excretion in Animals (rats and mice)

As observed for other PFAS, sex-specific elimination of PFDA was observed in rats. For
example, after IV administration (2 mg/kg PFDA), the dose-normalized serum
area-under-the-concentration curve (AUC/dose) was significantly higher in female rats
(3,065 mM h/mmol/kg) than male rats (1,875 mM h/mmol/kg) (Dzierlenga et al.. 2019). Similar
results were obtained for oral exposures of 2-20 mg/kg, with AUC/dose in female rats being 5,200-
5,500 mM h/mmol/kg vs. 2,960-3,320 mM h/mmol/kg in male rats fDzierlenga etal.. 20191. These
observations collectively suggest that elimination is slower in female rats than males, perhaps
because renal reuptake of PFDA is more efficient in female than male rats.

As noted earlier, the fecal excretion becomes increasingly important in elimination of long
carbon chain-length of PFAAs like PFDA (C10) as compared to shorter chain PFAS. For instance,
Kudo etal. (2001) attempted to evaluate the elimination of PFDA in Wistar rats (both sexes) with
intraperitoneally administration of PFDA using a single dose of 20 mg/kg. It was found that PFDA
was slowly excreted in urine, with only 0.2% of the dose being eliminated within 120 h. More of the
administered PFSA (~ 4%) was found in feces, indicating fecal excretion was a major route of the
elimination of PFDA for both sexes. Fecal excretion remained as the major route when rats were
intravenously injected with a dose of 25 mg/kg. Similarly, Vanden Heuvel etal. (1991) evaluated
the elimination of PFDA after 5 mg/kg intraperitoneal doses to male and female rats. Fecal
elimination accounted for 51% and 24% of the administered dose to the males and females,
respectively, over 28 days, while urinary excretion was less than 5% of the dose. These results are
partly contradicted by the data of Kim etal. f20191. who observed slightly over 3% total excretion

This document is a draft for review purposes only and does not constitute Agency policy.

3-12 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

in urine and feces after 120 hours, but that urine accounted for 25% of this excretion in male rats
(and 38% at 150 days) while urine was over or close to 50% of excretion in females.

Dzierlenga etal. f20191 also found that the total clearance (CLtot) of PFDA was extremely
low compared to other short-chain of PFAA compounds (e.g., PFHxA and PFOA) in both male and
female Hsd:Sprague-Dawley rats. These results were in line with previous findings ofVanden
Heuvel etal. f 19911 and Ohmori etal. f20031. and those of Kim etal. f20191. that PFAAs with
shorter carbon chain length tended to show higher CLtot.

Reported values of CLtot for rats are listed in Table 3-2. While the respective ranges of study
specific reported CLtot values for male and female rats indicate a degree of inter-laboratory
variability in the method of determination, the studies are all considered to be of adequate quality
and therefore there is no reason to preclude any one of them from an overall analysis. Therefore, a
hierarchical Bayesian analysis from partial pooling of all these data, described in Appendix G, was
performed in order to obtain overall population mean values and credible intervals for male and
female rats, listed in Table 3-2. These values (intervals) for CLtot are considered to be robust
estimates of average clearance in rats (and uncertainty therein).

Table 3-2. PFDA total clearance in rats and mice

Citation

Dose
(mg/kg)

Route

CLtot * (mL/d/kg)

Male Rats

Ohmori et al. (2003)

5.2 ± 1.3
5.32(3.66-6.77)

Kim et al. (2019)

3.04 ± 0.40a
1.61 (1.08-2.1)

Kim et al. (2019)

Oral

2.94(2.34-3.52)

Dzierlenga et al. (2019)

12.82 ±0.74
7.45(5.91-8.89)

Oral

7.44 ±0.31
5.04(4.36 - 5.75)

Oral

7.94 ±0.31
5.83(5.32-6.33)

Oral

8.11 ±0.22
5.71 (5.19-6.26)

Population mean (90% credible interval)

4.14(0.68 - 7.02)

Female Rats

Ohmori et al. (2003)

5.3 ±0.2
5.31 (4.44-6.29)

Kim et al. (2019)

3.24 ± 0.24a
2.07(1.84-2.31)

This document is a draft for review purposes only and does not constitute Agency policy.

3-13 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Kim et al. (2019)

Oral

2.61 (2.13-3.09)

Dzierlenga et al. (2019)

7.85 ±0.58
7.4(6.67-8.19)

Oral

4.61 ±0.22
3.78(3.41-4.16)

Oral

4.37 ±0.24
3.57(3.22-3.93)

Oral

4.61 ±0.24
3.77(3.39-4.14)

Population mean (90% credible interval)

4.06 (2.05 - 6.05)

Male Mice

Fuiii etal. (2015)

0.16

3.9°

Female Mice

Fuiii et al. (2015)

0.16

2.2°

* Values in plain text are as reported for each study unless otherwise noted. Values in italics are the mean (90%
credible interval) from the Bayesian analysis described in Appendix G

aReported absolute CL (mL/d) was divided by 0.25 kg; value is consistent with dose/AUCinf- reported.
bDzierlenga et al. (2019) indicates 3 rats/time-point used
Total of urinary and fecal clearance; see text below for details.

While Vanden Heuvel et al. (19911 also evaluated the elimination of PFDA in rats, they did
not report clearance values nor AUC values that could be used to calculate clearance. The half-lives
estimated from the decline in total body burden (based on 14C activity) were 23 and 43 days in
males and females, respectively, while the half-lives based on blood concentrations were 22 and 29
days, respectively (Vanden Heuvel et al.. 19911. These female half-lives are comparable to the beta-
phase half-lives reported for female rats by Dzierlenga et al. (20191 (18-44 days), though
somewhat lower than reported for female rats by Kim etal. (20191 (50-75 days). The half-life
estimates of Vanden Heuvel etal. (19911 for male rats are between the alpha-phase (1.7-2.1 days)
and beta-phase values (80-110 days) reported by Kim etal. (20191. but much less than those
reported by Dzierlenga etal. (20191 (215-300 days beta- or single-phase half-life). This range of
half-life values reflects the fact that half-life estimates are sensitive to noise in the experimental
data and study design, with Vanden Heuvel etal. (19911 having only measured elimination for 28
days, while Dzierlenga et al. (20191 measured plasma concentrations to 105 days and Kim et al.
(20191 to 150 days. Hence the results of Vanden Heuvel etal. (19911 appear to be generally
consistent with the other studies described here but will not be used in quantitative evaluation of
clearance.

Only Fuiii etal. (20151 evaluated the urinary and fecal clearance of PFDA in FVB/NJcl mice
using single IV dose (0.16 mg/kg) and oral gavage (1.6 [imol/kg). PFDA appeared to have the
smallest total (feces and urine) clearance as compared to short chained PFAAs (C < 8) (Fuiii etal..

This document is a draft for review purposes only and does not constitute Agency policy.

3-14 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

20151. Mouse urinary and fecal clearance were determined by dividing the total amounts excreted
in the urine and feces during a 24 hour period by the area under the curve (AUC) of the serum
concentration of each PFCA between 0 to 24 hours. Fecal elimination appeared to the primary
elimination route regardless of exposure routes (IV and oral gavage). For IV administration, there
were no marked differences in total clearance between sexes: 2.2 (1.4 and 0.8 mL/day/kg fecal and
urinary clearance, respectively) and 2.8 mL/day/kg (1.8 and 1.0 mL/day/kg fecal and urinary
clearance, respectively) for male and female mice, respectively. In comparison, the total clearances
for gavage-administered of PFDA were 3.9 mL/day/kg for male (3.6 and 0.3 mL/day/kg fecal and
urinary clearance, respectively) and 2.2 mL/day/kg for females (1.9 and 0.3 mL/day/kg fecal and
urinary clearance, respectively) fFuiii etal.. 20151. Since the toxicological studies being evaluated
used oral exposure, the oral PK results are considered most relevant and sex-specific PK
parameters are therefore suggested for calculating HEDs from corresponding points of departure in
male and female mice. The beta-phase half-lives obtained for male and female mice after oral
gavage are 1.4 day and 4.1 day, respectively (calculates as ln(2)/A2 from Table 1 of Fuiii etal.
(201511. However, since clearance was only observed for 24 h in the Fujii study, these half-life
estimates of half-life are considered uncertain and are not used for HED calculation. Instead, the
CLtot values in Table 3-2, which are determined from the amount of PFDA excreted in urine and
feces, will be used. While it would be preferable to have PK data from at least one other study in
mice, the results of Fuiii etal. f20151 are considered adequate for evaluating the relative clearance
in mice vs. humans.

Excretion in Humans

Fuiii etal. (20151 also estimated the elimination of PFDA in humans using 24-hour urine
samples collected from healthy volunteers, bile from patients who underwent biliary drainage, and
matched blood from both the healthy volunteers and patients. The clearance rate to urine and bile
from these data involves a straightforward calculation of the ratio of the daily amount excreted by
the route to the matched blood sample in a subject However, the fecal clearance rate is based on an
estimate of 98% resorption from the intestine (i.e., enterohepatic recirculation), that they obtained
by comparing their results for PFOA to direct observation of PFOA half-lives in humans by Olsen et
al. (20071: 98% intestinal resorption is required to match the total (urinary and biliary) excretion
otherwise estimated for PFOA with the previously measured PFOA half-life. It may be reasonable to
assume that these two compounds are resorbed in the intestines to a similar extent, but this
assumption is made in combination with use of biliary excretion data from five patients (three
female and two males), three suffering from choledocholithiasis, one from cholecystolithiasis, and
one from carcinoma of the head and pancreas. It is possible that anchoring the estimated fecal
clearance to the data from Olsen etal. (20071 in healthy subjects corrects for possible effects of
biliary disease, but these results should be considered with some caution.

The PFFDA urinary, biliary, fecal, and total clearances (sum of urinary and fecal clearance)
estimated by Fuiii etal. f20151 for humans were: 0.015 ± 0.01, 2.51 ± 2.1, 0.050 ± 0.04, and 0.066 ±

This document is a draft for review purposes only and does not constitute Agency policy.

3-15 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

0.05 (mean ± standard deviation, mL/kg/day). The difference between biliary clearance (not
included in the total) and fecal clearance is presumed to be the result of resorption after biliary
elimination. There may also be uncertainty due to the fact that Fuiii etal. f20151 only measured
PFDA levels for 24 hours and there was considerable variability in the results, as reflected by large
standard deviation (e.g., 2.51 ± 2.1 mL/day/kg for biliary clearance). On the other hand, collection
of 24-hour urine data provides a much better estimate of clearance by that route than extrapolating
from a single spot-sample.

An alternate approach to estimating human fecal clearance would be to assume that the
ratio of fecal/urinary clearance is similar in humans as in rats. Kim etal. (2019) observed a mean
fecal excretion 1.63 times higher than urinary excretion in male rats, but only 0.742 times urinary
excretion in female rats. Both of these ratios are considerably lower than the ratio of 3.3 estimated
by Fuiii etal. f2015I Given the uncertainty described above for the estimated fecal clearance of
Fuiii etal. (2015). these sex-specific ratios will be applied to the estimated human urinary clearance
from Fuiii etal. (2015) (0.015 mL/kg/day) to obtain total estimated urinary plus fecal clearance
rates of 0.039 mL/kg/day in men and 0.026 mL/kg/day in women.

In general, the total clearance profiles (urinary and fecal) of Fuiii etal. (2015) were
comparable between humans and mice: total clearance in humans decreased as a function of chain
length for C7-C9, then increased only slightly as the length increased further to C13, while mice
showed a clear decrease from C7-C10 followed by a clear increase with chain length from C10-C3.
In humans, the pattern in total clearance for C7 and higher was due to a shifting balance as fecal
clearance increased with chain length, but urinary clearance decreased.

A recent evaluation of women with children 2-5 years of age by Kim etal. (2020b) found
that PFDA is decreased in women who have breast-fed by a factor of 1.3% (95% CI: 0.5, 2.1%) per
month of breast-feeding, indicating that this is a significant route of elimination for such mothers,
and correspondingly a source of exposure for their children. This rate of elimination is comparable
to that estimated for younger women, below, indicating that elimination roughly doubles during
breast-feeding. The specific bioassays being extrapolated from animals to humans only involved
exposure to young adult animals or during an initial pregnancy, when lactational excretion would
not be a factor. However, it could be significant for the estimation of dosimetry in human children,
useful for the interpretation of epidemiological data. The elimination that occurs during
breastfeeding would reduce the body burden in a mother who then becomes pregnant again, hence
the risk to her subsequent children. While reduction due to breast-feeding would not be predicted
for women who formula-feed their children, some reduction in maternal PFDA would also be
expected due to distribution to the fetus, along with the placenta, umbilical cord and amniotic fluid
that are lost at childbirth, independent of how the child is subsequently fed.

Zhang etal. (2013b) estimated the urinary clearance of PFDA from matched urine and blood
or serum samples from 86 healthy volunteers. The resulting median clearance rate in young
females (age <50 years, n = 20), 0.047 mL/kg/day, is three times higher than the urinary clearance

This document is a draft for review purposes only and does not constitute Agency policy.

3-16 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

estimated by Fuiii etal. (20151. The reason for the discrepancy between the results of Zhang etal.
(2013b) and Fuiii etal. (20151 is unclear, but a possible factor is that Zhang etal. (2013b) used
single urinary voids ("spot samples") to estimate clearance while Fuiii etal. f20151 collected 24-
hour urine samples, which avoids assumptions required to extrapolate from a spot sample to total
daily excretion. Given that Fuiii etal. f20151 used a more reliable method and provides a more
health-protective value, the total human clearance estimated from their results will be used for
human equivalent dose estimates.

Another factor to be considered is clearance through menstrual blood and serum loss. As
there is no known mechanism for resorption of PFDA from menstrual blood and serum (unlike
urinary and biliary/fecal pathways). Therefore, it is reasonable to assume that any fluid lost by this
process would carry with it the PFDA it contained.

Zhang etal. f2013bl calculated a rate for menstrual clearance assumed to apply for all PFAS
based on a study of PFOA and PFOS that estimated menstrual blood loss using measurements of the
blood quantity excreted (Harada et al.. 20051. This estimate was not specific to PFOA or PFOS and
might also be applied to PFDA. However, Harada etal. (20051 cite Hallberg etal. (19661 as the
source for a menstrual blood loss of 70 mL per cycle, but according to Hallberg, "the upper normal
limit of the menstrual blood loss is situated between 60-80 mL." Thus, 70 mL/cycle appears to be
closer to an upper bound for healthy women. On the other hand, Verner and Longnecker f20151
reviewed Hallberg et al. T19661 who evaluated both blood loss and total fluid loss from
menstruation and concluded that the fluid lost in addition to blood was likely to be serum, with the
corresponding serum binding proteins and associated PFAS. Including this serum loss and
assuming 12.5 menstrual cycles per year, Verner and Longnecker (20151 estimated an average
yearly total serum loss of 868 mL. Assuming a standard human body weight of 80 kg, the
corresponding average rate of clearance is 868 mL/ (365 days)/(80 kg) = 0.030 mL/kg-day.

Lorber etal. f20151 examined the effects of ongoing blood loss through menstruation or
through frequent blood withdrawal as a medical treatment. Male patients with frequent blood
withdrawal had serum concentrations 40-50% less than males from the general population for the
chemicals observed in the study (PFOA, PFNA, PFDA, PFHxS, and PFOS). Female patients also had a
lower serum concentration than females from the general public. Though the trend of lower PFAS
serum concentration in patients compared to the general public was consistent, there was not a
clear trend in relation to the number of recent blood draws or in the recency of the last blood draw.
This study's analysis of the impact of menstrual blood loss was purely a modeling exercise, which
was performed for PFOA and PFOS. The authors estimated a monthly blood loss of 35 mL (which is
similar to the median loss reported by Hallberg etal. (196611. 50% of which was serum, resulting in
a clearance of 17.5 mL/month, or 0.0073 mL/kg-day in an 80 kg woman. This value is also
chemical-independent and could be applied to PFHxS instead of the menstrual clearance estimated
by Verner and Longnecker f20151.

This document is a draft for review purposes only and does not constitute Agency policy.

3-17 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

In summary, the total estimated urinary plus fecal clearance based on Fuiii etal. (20151
(urinary) and Kim et al. (20191 (fecal/urinary) is 0.039 mL/kg-day in men and 0.026 mL/kg-day in
women. The estimated urinary clearance from Fuiii etal. f 20151 is considered particularly reliable
because 24-hour urine samples were used to determine the rate. Adding the menstrual clearance of
0.030 mL/kg-day based on the results of Verne r and Longnecker f 20151 yields total
CL = 0.056 mL/kg-day for women between 12.4 (menarche) and 50 years of age, except during
pregnancy and until menstruation resumes postpartum. These values are considered appropriate
for use in animal-human dose extrapolation and hence will be used in the calculation of data-
derived extrapolation factors (DDEFs) (see Section 3.1.7).

3.1.5. Summary of pharmacokinetic parameters

Summary rat, mouse, and human pharmacokinetic parameters (clearance, volume of
distribution, and fraction absorbed (Fabs)) from the preceding analyses are provided in Table 3-3,
along with overall half-lives calculated from the clearance and volume of distribution.

Table 3-3. Rat, mouse, and human pharmacokinetic parameters.

Sex and species

Clearance
(miykg-d)

Volume of
distribution
(mL/kg)

Tl/2a
(d)

References

Male rats

4.14

431.1

Kim etal. (2019)
Dzierlenga et al. (2019)
Ohmori et al. (2003)

Female rats

4.06

313.4

Rats (M + F)b

4.10

372.3

Male mice

3.9

250

Fuiii etal. (2015)

Female mice

2.2

200

Mice (M + F)b

3.1

225

Men

0.039

431.1 (men)0

7,662
(21 yr)

Fuiii etal. (2015)

Kim etal. (2019)

Verner and Longnecker (2015)

(This document.)

Women < 12.4 or > 50
years

0.026

313.4d

8,355
(23 yr)

Women 12.4-50 years
oldd

0.056

313.4d

3,879
(10.6 yr)

a Tl/2 = (volume of distribution [mL/kg

) x In (2) / (clearance [mL/kg-d]).

bAverage of separate male and female values.

cVd in women assumed equal the value for female rats, Vd in men assumed equal to male rats,
includes 0.03 mL/kg/d for menstrual clearance based on Verner and Longnecker (2015).

Some mechanistic insight can be gained by comparing the clearance values shown in Table
3-3 with species-specific glomerular filtration rate (GFR), with and without adjustment for serum
protein binding. Davies and Morris (19931 summarized GFR for multiple species. Using 0.25 kg as
the species average BW for the rat, the GFR/BW for rats is 7.55 L/kg-day, which is 1,800 and 1,900

This document is a draft for review purposes only and does not constitute Agency policy.

3-18 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

times higher than the population mean clearance in male and female rats, respectively. Considering
the time-period of (Davies and Morris. 19931. it seems appropriate to use their value for average
human BW, 70 kg, which results in an estimated GFR/BW of 2.57 L/kg/d in humans, which is
66,000 times greater than the estimated clearance for human males. Thus, GFR itself is not a
limiting factor for PFDA clearance in rats or humans.

Binding to serum proteins plays a likely role in these very large differences. As discussed
above in the context of distribution, PFDA binds to albumin with high affinity, which mediates
glomerular filtration since only the unbound fraction is filtered (Kudo. 20151. in addition to any role
played by renal transporters. Kim etal. (20191 measured reported PFDA free fractions (/free) of
0.00118 and 0.000112 in male and female rat plasma. Using these values, GFRx/free = 8.9 and 0.85
mL/kg-day in male and female rats. This alternative estimate of clearance for male rats is close to
the empirical population mean in Table 3-3 (6.8 mL/kg-day), which could be interpreted as
implying that there is moderate renal resorption. However, for female rats GFRx/free is 4.8-fold
lower than the empirical clearance of 4.06 mL/kg-day. Section 3.1.6 provides discussion of the fact
that the PBPK model of Kim etal. (20191. which assumes that tissue distribution is similarly limited
by the free fraction, under-predicts the short-term distribution of PFDA in rats. Hence, while we
expect that serum protein binding limits renal excretion (and tissue distribution) to some extent,
the reduction appears to be less than predicted by assuming that clearance is strictly limited to the
equilibrium free fraction. Alternately, there could simply be an error in the measured free fraction.

Kim etal. (20191 also measured reported average PFDA/free values of 0.00157 and 0.00123
in human males and females, respectively, which leads to GFRx/free = 4 and 3 mL/kg-day for men
and women, which are still 100 and 56 times greater than the respective estimated total clearance
values (which include fecal and menstrual elimination). Thus, it appears likely that there is
significant renal resorption of PFDA in humans, which acts above and beyond the limitation
predicted based on measured serum protein binding.

According to EPA's BW°75 guidelines fU.S. EPA. 20111 use of chemical-specific data for
dosimetric extrapolation such as described above is preferable to the default method of BW0 75
scaling. However, for the purpose of comparison, using the standard species BWs of 0.25 kg in rats
and 80 kg in humans, the clearance in humans is predicted to be 4.2 times lower than rats. Given
clearance rates of 6.8 and 4.7 mL/kg-day in male and female rats, one would then predict clearance
rates of 1.6 mL/kg-day in men and 1.1 mL/kg-day in women, which are respectively approximately
40 and 20 times higher than the respective clearance values listed in Table 3-3 from human PK data
for men and women of reproductive age. Thus, based on the PFDA-specific PK data, use of BW0 75
could lead to an over-prediction of human elimination, hence an over-prediction of human
equivalent doses (HEDs) of 20-40-fold.

3.1.6. Evaluation of PBPK and PK Modeling

The PFAS protocol (Supplemental Information document, Appendix A) recommends the use
of PBPK models as the preferred approach for dosimetry extrapolation from animals to humans,

This document is a draft for review purposes only and does not constitute Agency policy.

3-19 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

while allowing for the use of data-informed extrapolations (such as the ratio of serum clearance
values) for PFAS that lack a scientifically sound and/or sufficiently validated PBPK model. If
chemical-specific information is not available, the protocol then recommends that doses be scaled
alio metrically using body weight (BW)3/4 methods. Selection from amongst this hierarchy of
decisions considers both the inherent and chemical-specific uncertainty (e.g., data availability) for
each approach option. This hierarchy of recommended approaches for cross-species dosimetry
extrapolation is consistent with EPA's guidelines on using allometric scaling for the derivation of
oral reference doses (U.S. EPA. 2011). This hierarchy preferentially prioritizes adjustments that
result in reduced uncertainty in the dosimetric adjustments (i.e., preferring chemical-specific values
to underpin adjustments vs. use of default approaches).

A PBPK model is available for PFDA in rats and humans Kim etal. f20191. The
computational code for this model was obtained from the model authors and evaluated for
consistency with the written description in the published paper, the PK data for PFDA, known
physiology, and the accepted practices of PBPK modeling. Unfortunately, several flaws were found
in the model. One flaw, an error in the balance of blood flow through the liver, had only a moderate
impact on model predictions. A much larger issue is that the model had only been calibrated to fit
the oral PK data for rats and the set of model parameters selected by the model authors to match
those data included an oral bioavailability (BA) lower than is otherwise supported by the empirical
PK data. For example, the fraction absorbed by the male rat was effectively set to 25% in the model
when the empirical PK analysis showed 65 ± 8% bioavailability. Further, when the model was used
to simulate the intravenous PK data, which are data to which a PK model should be calibrated, the
parameters were found to be completely inconsistent with these data. Figure 3-2 compares results
obtained with a replication of the PBPK model, which exactly matches the published PBPK model
results for oral dosimetry, to the data and empirical PK fit for a 1 mg/kg IV dose to male rats.

The over-prediction (approximately three to four times higher than these key
pharmacokinetic data for male rats) of the IV data by the Kim etal. f 20191 model indicates that
distribution into the body is significantly under-predicted by the model, which was offset in the
simulations of oral dosimetry data by using an unrealistically low oral bioavailability. Initial efforts
to re-fit the model to the data did not produce acceptable fits to both the IV and oral dose PK data
and involved changing model assumptions in a way that would require separate experimental
validation before use. It was therefore determined that the published model structure and
underlying assumptions did not allow a sufficiently sound calibration of the model to the PK data,
given the currently available data.

This document is a draft for review purposes only and does not constitute Agency policy.

3-20 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Plasma

Time (day)

Figure 3-2. Comparison of PFDA PBPK model predictions to IV dosimetry data
(circles) of Kim et al. f20191 for a 1 mg/kg dose.

"Empirical PK fit" is the result of an empirical PK analysis shown by Kim et al. (2019) (digitized). EPA's replication of
the PBPK model exactly reproduces the PBPK model results of Kim et al. (2019) for oral dosimetry hence is
considered an accurate reproduction of the model. The discrepancy between the PBPK model prediction for a 1
mg/kg dose and the data demonstrates that the published model structure and parameters are very inconsistent
with the empirical data, hence that there is a significant flaw in the model.

The U.S. EPA also evaluated the use of a one-compartment PK compartment to explicitly
describe the time-dependent dosimetry of PFDA. Specifically, the population mean CLtot and Vd
from Table 3-3 for male rats were incorporated into a one-compartment PK model and evaluated
against independent PK data, specifically end-of-study serum levels from the NTP (20181 bioassay.
Details of the model and evaluation are provided in Appendix G. Based on those results, EPA
considers use of a one-compartment PK model to predict time- and dose-dependent changes in
PFDA serum concentration as being too unreliable, even with PK parameters that EPA otherwise
considers to be sound. While the species- and sex- dependent CLtot values in Table 3-3 are believed
to provide reasonably sound measures of average serum concentration vs. dose in rats, mice and
humans and hence a better basis for HED calculation than use of default BW3/4 scaling, the greater
level of precision that would be implied by integrating these parameters in a PK model which
nominally can predict the exact rate of accumulation and clearance over time is not supported.

Another factor in considering use of a PK model is the potential to extrapolate across life-
stages. However, as described in Distribution in Humans above, there are only limited data on PFDA
distribution during gestation, the interpretation of which is not entirely clear. There are no PK data
in young children or young animals to evaluate differences in that life-stage, nor data on clearance
during pregnancy vs. non-pregnant adults. Even if the model was judged to be adequate for low-

This document is a draft for review purposes only and does not constitute Agency policy.

3-21 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

dose extrapolation of dosimetry in adult animals, use of the PK model for some endpoints and not
others would create inconsistency in the extrapolation approach. Therefore, recognizing the range
of uncertainties and the poor performance of the model in predicting the NTP bioassay data, a
simple PK model such as that described in Appendix G will not be used for dosimetric extrapolation
despite its potential promise.

3.1.7. Approach for pharmacokinetic extrapolation of PFDA among rats, mice, and humans

Empirical PK data from all published studies, including Kim etal. (20191. were evaluated
and summarized above to obtain values for the volume of distribution (Vd, mL/kg) and total
clearance (CLtot, mL/kg-day) in male and female rats and mice, women of child-bearing age (<50
years of age) and men and older women (see Table 3-3). However, evaluation of a published PBPK
model fKim etal.. 20191 and a one-compartment PK model showed significant errors in the PBPK
model, and that the simpler PK approach also did not reliably predict PFDA serum concentrations
measured at the end of the NTP bioassay (also see Appendix G). An alternative to use of PK (or
PBPK) models for dosimetric extrapolation is use of data-derived extrapolation factors (DDEFs). As
stated in EPA's guidance for DDEFs (U.S. EPA. 20141. use of these factors "maximize the use of
available data and improve the scientific support for a risk assessment." As discussed above in the
section on Excretion, the estimated population average values of CLtot for male and female rats,
female mice and male and female humans are considered sufficiently sound for use in such
extrapolation and use of the alternative (default) approach, BW3/4 scaling, would lead to significant
errors in HED calculations. Therefore, application of DDEFs calculated from the clearance values
listed in Table 3-3 is considered the next preferred option in the absence of a reliable PK (or PBPK)
model.

Specifically, the ratio of sex-specific human clearance to clearance in the animal species and
sex used to identify a specific point-of-departure (POD) will be used to estimate HEDs for points of
departure (PODs) identified from bioassays performed with those animal species. For example, to
extrapolate from a POD from the NTP bioassay for an endpoint in male rats to human males,

HED = POD x (FabS,rat,m/Fabs,H) X CL||,m/CLrat,m,

where Fabs,H is the fraction absorbed in humans and CL,H,mis the clearance in human males, while
Fabs,rat,m is the fraction absorbed in male rats and CLrat,m is the clearance in male rats. The DDEF is
then (Fabs,rat,m/Fabs,H) x CLH/CLrat,m. As discussed in Section 3.1.1, Fabs is assumed to be 1 in rats due to
the range of results found, with the estimated clearance values based on this assumption. Fabs,H is
likewise assumed to be 1 but is shown here for generality.

For gestational effects the clearance in the female animal (dam) is assumed to determine
dosimetry to the fetus. While a higher clearance can be estimated for women of reproductive age by
including menstrual blood loss, menstruation does not occur during pregnancy and may not resume
until after weaning of the child. Female babies also clearly do not menstruate. Further, as described

This document is a draft for review purposes only and does not constitute Agency policy.

3-22 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

in Section 3.1.6, there is uncertainty in the clearance estimate for humans and there may be
differences in PK among human life-stages that cannot be quantified because of a lack of empirical
PK data during gestation, lactation, and childhood. While effects in adults do not involve
extrapolation across life-stages, the degree of accumulation of PFDA in rats during a 28-day
bioassay could be less than the accumulation during a comparable portion (4%) of the human
lifespan. Therefore, HEDs for developmental and immune effects have been calculated using the
health-protective CLh from Table 3-3 for women below age 12.4 and over age 50, which is based on
the average renal clearance for a mixed population of men and older women from Fuiii etal.
(20151: while assuming fecal clearance is 74.2% of renal clearance based on female rat data Kim et
al. f20191. 0.026 mL/kg-day = 2.6 x 105 L/kg-day, to assure an adequate level of protection. In
particular, the level of fecal clearance observed in male rats by Kim etal. f20191 was 163% of
urinary clearance and the fecal clearance estimated by Fuiii etal. f20151 for humans was 333% of
urinary clearance, but the estimate by Fuiii etal. (20151 was considered highly uncertain because of
the multiple assumptions involved and use of the female rat fecal/urinary ratio is health-protective
compared to both the male rat ratio and the uncertain human estimate.

Liver effects observed in adult female rats are assumed to be relevant to older women,
hence the same CLH (0.026 mL/kg-day) will be used to extrapolate those. Liver and reproductive
effects observed in adult male rats will be extrapolated using the CLH for men, 0.039 mL/kg-day.
Finally, reproductive effects observed in adult female rats will be extrapolated using the CLH for
women of reproductive age, 0.056 mL/kg-day.

The key assumption in calculating a DDEF for a given endpoint evaluated are then that for
effects observed in adult male and female rats or mice, the CL for the corresponding rat or mouse
sex and human sex (and relevant life-stage for women) from Table 3-3 are used to calculate the
DDEF. The following table then shows the resulting DDEFs.

Table 3-4. DDEF calculations

Sex and species of observation
(lifestage [endpoint])

CLA (miykg-d)

CLH (miykg-d)

DDEFa

Male rats (adult)

4.14

0.039

9.42 x 10"3

Female rats (adult [liver])

4.06

0.026

6.40 x 10"3

Female rats (adult [reproductive])

4.06

0.056

1.38 x 10"2

Mouse (developmental)

2.2

0.026

1.18 x 10"2

a DDEF = (CLH/CLA) x (Fabs,A/Fabs,H), with Fabs,A and Fabs,H assumed to be 1. Rat and mouse CL values from
Table 3-3. No data exist showing that CL in juveniles is different from adults.

When an internal dose POD, specifically a serum concentration, is obtained from human
epidemiological studies (Budtz-l0rgensen and Grandiean. 2018a: Grandiean etal.. 20121. the HED
will likewise be calculated as:

This document is a draft for review purposes only and does not constitute Agency policy.

3-23 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

HED = PODmt x CLh,

using the health-protective estimate for human clearance, CLH = 0.026 mL/kg-day = 2.6 x 10-5
L/kg-day, i.e., the value estimated for pregnant and breast-feeding women and their female
children.

Uncertainty in HED calculations for PFDA

The population mean male rat clearance value has a 90% credible interval from 2.6-fold
below to 1.6-fold above its value (see Table 3-2). However, application ofaDDEF assumes that a
steady-state concentration is reached, equal to dose/CL. When the end-of-study PFDA levels
observed by NTP T20181 are compared to the corresponding dose/CL using the mean CL estimated
for male rats the resulting dose/CL values are 2.7 to 1.4 times higher than the data. Hence the
uncertainty in use of the male rat CL in the current analysis is considered to be less than a factor of
3.

Likewise, the 90% credible interval for female rat CL only ranged from 2-fold below to 1.4-
fold above the population mean (see Table 3-2) but the estimated steady-state levels (dose/CL
values) are 2-3-fold higher than the end-of-study values measured by NTP (2018). so the
uncertainty from use of the mean CL for adult female rats is also considered to be less than 3-fold.

While these uncertainties in the male and female rat CL values could lead to over-prediction
of the HEDs, the clearance value used for humans was based on the results of Fuiii etal. f20151.
rather than the results of Zhang etal. (2013b) who obtained a rate of urinary clearance in young
women approximately three times higher than the (mixed population) estimate of Zhang et al.
(2013b). A more modest correction for fecal absorption (using the ratio of fecal/urinary elimination
observed in rats after IV dosing) was applied vs. the rate estimated by Fuiii etal. (2015). which was
roughly 3-fold higher than the rate used. An additional term for menstrual blood loss is only applied
for the extrapolation of reproductive effects observed in female rats. Hence, the population average
human clearance is unlikely to be lower than the values used for HED calculation and these choices
for human clearance is expected to offset the uncertainty in rat dosimetry, at least to some extent
The overall uncertainty in the animal-human extrapolation is therefore judged to be less than a
factor of 3, which is considered reasonable for a pharmacokinetic analysis.

Uncertainties in the extrapolation to developmental exposure and dosimetry in children
remain, given that developmental PK studies have not been conducted in rats and mice and there
are only limited developmental PK data for humans. (As described in Distribution in Humans, the
available data for distribution in human fetuses indicate that it is similar to distribution in adult
female rats, so there is no indication of a marked lifestage difference in the volume of distribution.)
There are likewise no data on clearance or excretion in early lifestages in comparison to adult
animals or humans, so there is uncertainty in the extent to which such differences may exist
However, since the available data do not indicate significant lifestage variation in clearance or
volume of distribution, the uncertainties from extrapolation across lifestages are judged unlikely to

This document is a draft for review purposes only and does not constitute Agency policy.

3-24 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

be greater than is accounted for by application of the standard human interindividual uncertainty
factor (UFH), of which a factor of three is typically attributed to pharmacokinetic uncertainty.

While the PK model parameter estimates seek to make the best use of the available
chemical- and species-specific data, there are also many uncertainties noted above, in particular for
humans. Therefore, we also evaluated the use of default BW3/4 scaling of total clearance (CL-BW),
i.e., if CLhuman = CLrat x (BWhuman/BWrat) - 0.25. The resulting clearance values for men and
women (scaled from male and female rats, respectively) are 40 and 20 times higher than the values
estimated from human data (Summary of pharmacokinetic parameters). Hence, estimates of human
equivalent doses using BW3/4 scaling of clearance would be significantly less health protective than
the proposed DDEFs (see Table 3-4). While it is plausible that human clearance is actually that high,
given the limited human PK data, lack of exposure control and quantification, and other
uncertainties discussed above, the clearance values obtained from chemical-specific data are
preferred and used in the derivations below since they are based on direct observation of human
excretion.

3.2. NONCANCER HEALTH EFFECTS

For each potential health effect discussed below, the synthesis describes the evidence base
of available studies meeting the PECO criteria, as well as the supplemental studies that most
directly inform questions relating to coherence, MOA, biological plausibility, or human relevance
during evidence integration.

For this section, evidence to inform organ/system-specific effects of PFDA in animals
following developmental exposure are discussed in the individual organ/system-specific sections
(e.g., liver effects in adult animals after gestational exposure are discussed in the liver effects
section). Given that spontaneous abortion and preterm birth are informative of both female
reproductive and developmental toxicity, these endpoints are also discussed in the sections for
Developmental and Female reproductive effects. General toxicity effects, including body weights
and survival, were summarized to aid in interpretation of other potential health effects considering
the association between PFDA exposure and induction of wasting syndrome (rapid and marked
reductions in body weight and food consumption) in animals (refer to Section 3.2.10 for more
details).

3.2.1. HEPATIC EFFECTS

Methodological considerations

Serum enzymes and other clinical markers of hepatocellular and biliary function were
evaluated across human and animal studies. For the animal studies, the results were interpreted
together with histopathology and liver weight measures to aid in the assessment of potential
adverse liver effects in response to PFDA exposure. Elevated serum levels of alanine
aminotransferase (ALT) and aspartate aminotransferase (AST) are useful indicators of

This document is a draft for review purposes only and does not constitute Agency policy.

3-25 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

hepatocellular damage that results in the release of these enzymes into the blood, with ALT
considered more specific and sensitive (Hall etal.. 2012: EMEA. 2008: Boone etal.. 20051. Alkaline
phosphatase (ALP) is localized to the bile canalicular membrane, and therefore, more indicative of
hepatobiliary damage. Increases in circulating ALP, y-glutamyltransferase (GGT; another
canalicular enzyme) and bile components (bilirubin and bile salts/acids) and are associated with
obstruction of hepatic bile flow and damage to biliary epithelial cells fHall etal.. 2012: EMEA. 2008:
Boone etal.. 2005). Blood proteins such as albumin, globulin, and total protein (amount of albumin
and globulin) are routinely evaluated in clinical chemistry. Albumin is synthesized in the liver and
then excreted into the bloodstream, where it is bound by fatty acids, cations, bilirubin, thyroxine
(T4), and other compounds. Globulins, a collection of blood proteins larger than albumin, is made
by both the liver and immune system. Decreased levels of these blood proteins can be good
indicators of protein loss due to kidney disease or impaired synthesis as a result of liver damage
CWhalan. 20151.

Human studies

Serum biomarkers

Eight epidemiology studies report on the relationship between PFDA exposure and liver
serum biomarkers. As discussed above, ALT and AST are considered reliable markers of
hepatocellular function/injury, while levels of ALP fBoone etal.. 20051. bilirubin, andy-GGT are
routinely used to evaluate hepatobiliary toxicity (Hall etal.. 2012: EMEA. 2008: Boone etal.. 2005).

Of the remaining seven studies, all were medium confidence, including five studies in adults
(four cross-sectional and one cohort) and one cohort (analyzed cross-sectionally) of children, and
one cross-sectional study examined all ages (3-79 years) (see Figure 3-3). In addition, one study, a
cross-sectional study of pregnant women, was considered uninformative due to lack of
consideration of potential confounding fliang etal.. 20141. All studies measured liver enzymes
using standard laboratory approaches.

This document is a draft for review purposes only and does not constitute Agency policy.

3-26 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Cakmak, 2022, 10273369 -

Jain and Ducatman, 2019, 5080621 -

Jiang, 2014, 2850910-

- D

Legend

Good (metric) or High confidence (overall)

Liu, 2022, 10273407-

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)

Mora, 2018,4239224-

Critically deficient (metric) or Uninformative (overall)

Nian M et al. 2019, 5080307-

Omoike, 2020, 6988477 -

Salihovic S et al. 2018, 5083555 -

¦f

Figure 3-3. Hepatic effects, human study evaluation heatmap. Refer to HAWC

for details on the study evaluation review: HAWC Human Hepatic Effects.

The results for the seven medium confidence studies are summarized in Table 3-5. While
most studies reported statistically significant associations with some clinical markers, the direction
of association varied across studies for individual markers and across markers within studies, and
the clinical relevance of the changes is unclear. The studies in adults were somewhat inconsistent
with respect to the direction and size of the effect estimates, though differences in the populations
and analyses complicate their comparison. Five of the studies examined general population adults
with cross-sectional measurement of exposure and outcome fCakmak etal.. 2022: Liu et al.. 2022:
Omoike et al.. 2020: Tain and Ducatman. 2019b: Nian et al.. 20191 (the latter also includes children).
Nian etal. f20191 observed positive associations with ALT, AST, GGT, and total bilirubin but an
inverse association with ALP. lain and Ducatman f2019b observed positive associations with ALT
and total bilirubin but inverse associations with AST and ALP. Liu etal. (20221 reported positive
associations with ALT, AST, GGT, and total bilirubin with exposure response gradients observed
across quartiles, but an inverse association with ALP. Cakmak et al. f20221 reported positive
associations with ALT, AST, ALP, GGT, and total bilirubin, but only GGT was statistically significant.
In contrast, Salihovic etal. f20181 examined changes in liver function with changes in PFDA
exposure over 10 years in elderly adults. They observed positive associations with ALT, ALP, and
GGT, but an inverse association with total bilirubin. The inconsistency across markers in the same
study and within markers across studies increases the uncertainty in the evidence, though it is
conceivable that the differences across markers could be explained by different mechanisms. Since
the individual markers reflect involvement of multiple types of hepatic cells, elevations in all of
them would not be expected, but there is no biological explanation for the reduced levels that were

This document is a draft for review purposes only and does not constitute Agency policy,

3-27 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

observed. Differences observed in Salihovic etal. (20181 could be explained by the different
population, or more likely, the examination of change in outcome over time vs. cross-sectional
analysis, but there is not enough data available to confirm this. Study sensitivity may also play a
role in the inconsistency. Nian etal. f20191 and Liu etal. T20221 had the largest exposure contrast
(IQR 0.9-1.0 ng/mL) and the strongest adverse associations. In the only available study in children,
the only marker examined (ALT) was found to be lower with higher exposure in girls.

Overall, there is reasonably consistent evidence of a positive association between exposure
to PFDA and ALT in adults, including positive associations in four of five available studies (three
statistically significant) and an exposure-response gradient observed in the one study that
examined quartiles of exposure. The larger associations in studies with adequate sensitivity (due to
larger exposure contrast) also increases certainty in the association. Evidence for other biomarkers
of hepatic function, particularly ALP and total bilirubin, are less consistent. The biological relevance
of the small observed changes in any of these liver enzymes is unclear. No studies of more clinical
or apical endpoints are available to provide coherence with these findings.

This document is a draft for review purposes only and does not constitute Agency policy.

3-28 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-5. Associations between PFDA and serum biomarkers of hepatic function in medium confidence
epidemiology studies

Reference

Population

Median
exposure
(IQR) or as
specified

Effect
estimate

Markers of hepatocellular injury

Markers of hepatobiliary injury

ALT

AST

ALP

GGT

Total bilirubin

Adults

Jain and

Ducatman

(2019b)

Cross-

sectional

(NHANES

2011-2014);

U.S.; 2883

adults

Serum
0.2

P (p-value)
per log-unit
change

Nonobese
0.003 (0.9)

Obese
0.01 (0.5)

Nonobese
-0.009 (0.6)

Obese
-0.01 (0.5)

Nonobese
-0.03 (0.01) *

Obese
-0.006 (0.7)

Nonobese
-0.003 (0.9)

Obese
0.003 (0.9)

Nonobese
0.05 (0.02) *

Obese
0.02 (0.3)

Omoike et
al. (2020)

Cross-
sectional
(NHANES
2005-12); U.S.
6652 adults

Serum
0.3

(20th-80th: 0.1-
0.5)

% change
(95% CI) per
1% increase

0.01 (-0.0, 0.03)

Salihovic et
al. (2018)

Cohort
(2001-14);
Sweden; 1002
elderly adults

Plasma at
baseline
(70 yrs)
0.3 (0.2-0.4)

P (95% CI) for

changes in
liver function
with change
in In-PFDA
over 10 years
(mixed
random
effects)

0.02 (0.01,0.04)*

0.2 (0.1, 0.2) *

0.06 (0.0,0.1)

-2.3 (-2.7, -1.9) *

Liu et al.
(2022)

Cross-
sectional
(2018-19);
China; 1303
adults

Serum
0.9 (0.5-1.4)

% difference
(95% CI) for
quartiles vs
Q1

02:
3.46 (1.84,
5.09)*
Q3:
8.03 (4.33,
11.86)*

02:
1.16 (0.13,
2.21)*

Q3:

4.63 (2.27, 7.05)*
Q4:

Q2:

-0.75 (-1.63, 0.13)
Q3:
-2.09 (-4.00, -
0.14)*

Q4:

Q2:

3.31 (1.41, 5.24)*
Q3:

8.97 (4.57,13.55)*
Q4:

19.37 (8.37, 31.48)*

Q2:

2.67 (1.50, 3.86)*
Q3:

6.23 (3.56, 8.98)*
Q4:

12.03 (5.53,18.94)*

This document is a draft for review purposes only and does not constitute Agency policy.

3-29 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Reference

Population

Median
exposure
(IQR) or as
specified

Effect
estimate

Markers of hepatocellular injury

Markers of hepatobiliary injury

ALT

AST

ALP

GGT

Total bilirubin

Q4:
15.51 (6.44,
25.35)*

11.75 (5.91,
17.91)*

-4.40 (-8.72, 0.13)

Nian et al.
(2019)

Cross-
sectional
(2015-16);
China; 1605
adults

Serum
0.9 (0.5-1.5)

% change
(95% CI) per
In-unit change

3.1 (0.1, 6.1) *

1.0 (-0.9, 3.0)

-3.8 (-5.4, -2.2)

2.2 (-0.9, 5.3)

4.3 (2.1, 6.6) *

Adults and Children

Cakmak et
al. (2022)

Cross-
sectional
(2007-2017);
Canada; 6109

Plasma
0.2

% change
(95% CI) per 1
GM increase

3.0 (-0.1, 6.2)

2.2 (-0.6, 5.0)

1.0 (-3.3, 5.6)

15.5 (2.2, 30.4)*

1.6 (-7.8, 11.9)

Children

Mora et al.
(2018)

Cross-
sectional
analysis of
cohort (1999—
2002), U.S.;
682 children
(7-8 yrs)

Plasma
0.3 (0.2-0.5)

P (95% CI)
with IQR
increase

-0.3 (-1.2, 0.5)
Boys: 0.1 (-1.3,
1.4)

Girls: -0.9 (-1.8,
-0.1) *

*p < 0.05.

IQR: interquartile range (i.e., 25th-75th percentile); NR = not reported.

This document is a draft for review purposes only and does not constitute Agency policy.

3-30 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Animal studies

The toxicity database for PFDA liver effects in experimental animals consists of three
28-day gavage studies fFrawlev etal.. 2018: NTP. 20181. five short-term studies (<14 days) via the
diet fYamamoto andKawashima. 1997: Kawashimaetal.. 1995: Permadi etal.. 1993: Takagi etal..
1992.19911. one short-term study via drinking water fWangetal.. 20201 and one developmental
study via gavage (Harris and Birnbaum. 19891. The studies included several strains of rats (S-D,
Wistar and Fischer 344) and mice (C57BL/6N, B6C3F1/N and CD-1[ICR]) and measured endpoints
considered informative for evaluation of liver toxicity, such as histopathology, serum biomarkers of
effects, and organ weights.

Histopathology

Liver histopathology was examined across five short-term oral exposure studies: two high
confidence studies in rats exposed via gavage (Frawlev etal.. 2018: NTP. 20181. a low confidence
study in mice exposed via the diet (Kawashima etal.. 19951 and two low confidence studies in mice
exposed via drinking water (Wang etal.. 20201. The primary issue contributing to the low
confidence rating for the Kawashima et al. f 19951. Li etal. f20221. Wang etal. f20201 studies was
the incomplete reporting of histopathology data (no information on incidence or severity) (see
Figure 3-4). Additional deficiencies were identified in the study evaluation domains for allocation
(non-reporting of randomization), nonreporting of blinding practices, and chemical administration
and characterization in Kawashima et al. (19951.

This document is a draft for review purposes only and does not constitute Agency policy.

3-31 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Reporting quality -

Allocation -

Observational bias/blinding -

Confounding/variable control -

Selective reporting and attrition -

Chemical administration and characterization -

Exposure timing, frequency and duration -

Endpoint sensitivity and specificity -

Results presentation -

Overall confidence -

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)

I Critically deficient (metric) or Uninformative (overall)
Not reported

Figure 3-4. Evaluation results for animal studies assessing effects of PFDA
exposure on liver histopathology. Refer to HAWC for details on the study
evaluation review.

Hepatocyte lesions were identified in male and female S-D rats at exposure doses of 0.5-
2.5 mg/kg-day across the two high confidence studies with 28-day exposure and these lesions were
not present in control animals fFrawlev etal.. 2018: NTP. 20181 (see Table 3-6 and Figure 3-5).
Cytoplasmic alterations that consisted of accumulation of eosinophilic granules within the
cytoplasm of centrilobular hepatocytes were observed in nearly all rats at doses of 0.625-
2.5 mg/kg-day in the study by NTP f20181. Cytoplasmic vacuolation that that was largely
centrilobular in distribution and characterized by accumulation of microvacuoles within the
cytoplasm was also reported in males and females at 1.25 and 2.5 mg/kg-day (10-100% incidence
rate) fNTP. 20181. Hepatocyte hypertrophy (i.e., increase in the size of primarily centrilobular
hepatocytes) was significantly elevated in these animals at similar doses (80-100% incidence)
(NTP. 20181. The severity of these lesions increased with dose, ranging from minimal to marked in
males and minimal to moderate in females. Minimal hepatocyte necrosis was increased in rats
across studies and sexes (Frawlev etal.. 2018: NTP. 20181 with incidence rates ranging from 10-
40%; a statistically significant trend was reported in females at doses >1.25 mg/kg-day in one study
fNTP. 20181. Frawlev etal. (20181 characterized the changes as centrilobular, single cell hepatocyte
necrosis occurring in female rats (males were not tested in the study). Hepatocyte necrosis in male
and female rats was described in the fNTP. 20181 report as "a few widely scattered, variably sized,

This document is a draft for review purposes only and does not constitute Agency policy,

3-32 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

randomly distributed foci of necrotic hepatocytes within the hepatic parenchyma mixed with
variable numbers of mononuclear inflammatory cells."

PFDA treatment had no appreciable effect on cellular infiltration in the liver in either male
or female rats up to a dose of 2.5 mg/kg-day after 28-day exposure fNTP. 20181 (see Figure 3-5).
The low confidence studies also observed hepatocyte changes in animals at higher PFDA doses
(4.6-25 mg/kg-day); however, data were only summarized qualitatively and, therefore, are not
displayed in Figure 3-5 (Li etal.. 2022: Wang etal.. 2020: Kawashima etal.. 1995). Kawashima et al.
(1995) described increases in lipid droplets, cell size (hypertrophy), peroxisome proliferation and
vacuolated nuclei in male Wistar rats in the two high-dose groups after 7-day exposure via the diet
(4.6 and 9.22 mg/kg-day). Similarly, increased hypertrophy and lipid accumulation was reported in
the liver of female C57BL/6J mice after exposure to 25 mg/kg-day for 14 days via drinking water.
Additionally, liver necrosis, steatosis, edema, and degeneration were found in male CD-I mice
exposed to a PFDA dose of 13 mg/kg-day via drinking water for 12 days (Wang etal.. 2020).
Although there is no information on incidence or severity, the findings from the Kawashima et al.
(1995). Li etal. (2022) and Wang etal. (2020) studies are coherent with observations from the high
confidence 28-day studies (e.g., vacuolation, hypertrophy and necrosis).

Altogether, PFDA induced a spectrum of morphological changes in rodent hepatocytes that
included cytoplasmic alterations and vacuolization, hypertrophy, and some evidence of structural
degenerative lesions (minimal necrosis accompanied in some cases by evidence of possible
inflammation) after short-term exposure. Furthermore, a general pattern of increased severity
(within and across lesions) was apparent with increasing dose.

This document is a draft for review purposes only and does not constitute Agency policy.

3-33 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-6. Incidence and severity of hepatocyte lesions in S-D rats exposed to PFDA in 28-day gavage studies

Animal group

Dose (mg/kg-d)

0.125

0.156

0.25

0.312

0.5

0.625

1.25

2.5

Cytoplasmic alterations

NTP (2018) - Female (n= 10 in all groups)

8 (minimal)

(minimal)

10
(mild)

NTP (2018) - Male (n= 10 in all groups)

(minimal)

10
(marked)

Cytoplasmic vacuolization

NTP (2018) - Female (n= 10 in all groups)

(minimal)

(moderate)

NTP (2018) - Male (n= 10 in all groups)

(mild)

(moderate)

Hypertrophy

NTP (2018) - Female (n= 10 in all groups)

(minimal)

(moderate)

NTP (2018) - Male (n= 10 in all groups)

(mild)

(moderate)

Necrosis

Frawlev et al. (2018) - Female (n= 8 in all
groups)

3 (minimal)

NTP (2018) - Female (n= 10 in all groups)

(minimal)

NTP (2018) - Male (n= 10 in all groups)

(minimal)

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study authors; shaded cells represent doses not
included in the individual studies. For example, the dose of 0.125 mg/kg-d was not used in the (NTP, 2018) study. Severity was normalized to a four-point
scale by the study authors as follows: minimal, mild, moderate, and marked.

This document is a draft for review purposes only and does not constitute Agency policy.

3-34 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Hepatocyte Cytoplasmic Alteration

Study Name

NTP. 2018, 4309127

Hepatocyte Cytoplasmic Vacuolization NTP. 2018, 4309127
Hepatocyte Hypertrophy NTP. 2018. 4309127

Hepatocyte Necrosis

Infiltration Cellular. Mixed Cell

Liver Histopathology, Abnormal

Outcome Confidence

High confidence

High confidence
High confidence

£ no significant change
A Statistically significant increase
/\ treatment-related increase

Frawley. 2018.4287119
NTP, 2018, 4309127

NTP 2018. 4309127

Kawashima. 1995.3858657
Wang, 2020, 6323927
Li. 2022.10273360

High confidence
High confidence

Low confidence
Low confidence
Low confidence

Study Type

28 Day Oral

28 Day Ora
28 Day Ora

7 Day Oral
12 Day Ora
14 Day Ora

Animal Description

Rat, Sprague-Dawley (Harian) (r
Rat. Sprague-Dawley (Harlan) (.
Rat, Sprague-Dawley (Harian) (
Rat, Sprague-Dawley (Harian) (
Rat, Sprague-Dawley (Harlan) ('
Rat, Sprague-Dawley (Harian) (,
Rat, Sprague-Dawley (Harlan) ( _
Rat, Sprague-Dawley (Harlan) ('
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (...
Rat, Wistar ( )

Mouse. CD-1 (o)

Mouse. C57BL/6J
-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Reporting quality -

Allocation -

Observational bias/blinding -

Confounding/variable control -

Selective reporting and attrition -

Chemical administration and characterization -

Exposure timing, frequency and duration -

Endpoint sensitivity and specificity -

Results presentation -

Overall confidence -

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NR Not reported

Figure 3-6. Evaluation results for animal studies assessing effects of PFDA
exposure on liver serum biomarkers. Refer to HAWC for details on the study
evaluation review.

1 Increases in ALT and AST, two markers of hepatocellular damage, were consistently

2 reported in S-D rats with 28-day gavage exposures and CD-I mice exposed for 12 days via the diet

3 (Wang etal., 2020: NTP, 20181 (see Table 3-7 and Figure 3-7). Increased ALT was reported in male

4 and female rats, although only effects in females showed a significant trend with 44% and 20%

5 changes from controls occurring at 1.25 and 2.5 mg/kg-day, respectively. AST levels increased in a

6 dose-dependent manner in both sexes, reaching statistical significance at all exposure doses in

7 males (13-42% compared to controls across 0.156-2.5 mg/kg-day) and at 1.25 and 2.5 mg/kg-day

8 in females (31% and 80% compared to controls, respectively). In mice exposed to a higher PFDA

9 dose (13 mg/kg-day), these enzymes were similarly elevated, increasing by 338% and 649%
10 relative to controls for ALT and AST, respectively (Wang etal.. 2020).

This document is a draft for review purposes only and does not constitute Agency policy,

3-36 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-7. Percent change relative to controls in hepatocellular serum
markers in short-term animal studies after PFDA exposure

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Alanine aminotransferase (ALT)

28 d gavage; female S-D rats
(NTP, 2018)

-3

28 d gavage; male S-D rats
(NTP, 2018)

12 d drinking water; male CD-I mice
(Wang et al., 2020)

338

Aspartate aminotransferase (AST)

28 d gavage; female S-D rats
(NTP, 2018)

-3

-8

28 d gavage; male S-D rats
(NTP, 2018)

12 d drinking water; male CD-I mice
(Wang et al., 2020)

649

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors. Shaded cells represent doses not included in the individual studies.

1 Markers of hepatobiliary function including ALP, bile salts/acids and bilirubin (total, direct

2 and indirect) were also altered in S-D rats after a 28-day exposure (NTP. 20181 (see Table 3-8 and

3 Figure 3-7). ALP levels increased significantly at doses >0.312 mg/kg-day in both males and

4 females (22-106% compared to controls). Levels of bile salts/acids and bilirubin (total, direct and

5 indirect) were elevated in male and female rats at doses >1.25 mg/kg-day; the effects showed a

6 significant trend and were large in magnitude (205-1,207% and 28-733% compared to controls for

7 bile salts/acids and bilirubin, respectively).

Table 3-8. Percent change relative to controls in hepatobiliary serum markers
in a 28-day rat study after PFDA exposure (NTP. 2018)

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Alkaline phosphatase (ALP)

Female S-D rats

106

Male S-D rats

Bile salts/acids

Female S-D rats

-6

205

658

This document is a draft for review purposes only and does not constitute Agency policy.

3-37 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Male S-D rats

-53

-39

440

1207

Total bilirubin

Female S-D rats

-6

-9

-10

356

Male S-D rats

350

Direct bilirubin

Female S-D rats

104

700

Male S-D rats

-22

-4

-7

733

Indirect bilirubin

Female S-D rats

-7

-11

-14

275

Male S-D rats

255

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors.

Albumin, globulin, and total protein were examined in S-D rats after 28-day exposure (NTP.
20181 (see Table 3-9 and Figure 3-7). Dose-related decreases in albumin were reported in males,
decreasing by 8% and 20% at 1.25 and 2.5 mg/kg-day, respectively. In females, statistically
significant increases in albumin levels were reported at 0.312 (11%) and 0.625 (13%) mg/kg-day
but there was no dose-response gradient A significant trend for globulin levels was found in both
males and females, with decreases of 9-42% at >0.156 mg/kg-day. The albumin and globulin
findings corresponded well with a decrease in total protein and increase in albumin/globulin (A/G)
ratio in animals. Statistically significant increases in the A/G ratio (13-47%) occurred in males and
females at all exposure doses (0.156-2.5 mg/kg-day) and total protein decreased significantly (4-
28%) in males at similar doses. In females, total protein decreased by 2% and 12% at the highest
doses (1.25 and 2.5 mg/kg-day, respectively), but a significant trend was not established.

This document is a draft for review purposes only and does not constitute Agency policy.

3-38 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-9. Percent change relative to controls in serum proteins in a 28-day
rat study after PFDA exposure fNTP. 20181

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Albumin

Female S-D rats

-10

Male S-D rats

-8

-21

Globulin

Female S-D rats

-9

-18

-21

-14

Male S-D rats

-13

-19

-27

-36

-42

Albumin/Globulin ratio

Female S-D rats

Male S-D rats

Total Protein

Female S-D rats

-2

-12

Male S-D rats

-4

-5

-10

-17

-28

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study

authors.

1 In summary, coherent effects across serum enzymes, biliary system components and blood

2 proteins that are consistent with altered liver function were reported in rats and mice after short-

3 term PFDA exposure. In mice, the serum enzyme changes were accompanied by a 40% reduction in

4 body weights at the high PFDA dose tested (13 mg/kg-day) (Wang etal.. 20201. Although the 28-

5 day rat study reported significant body weight reductions at >1.25 mg/kg-day, dose-related

6 changes in some serum biomarkers of hepatic injury occurred at doses lower (0.156-0.625 mg/kg-

7 day) than those associated with marked systemic toxicity (NTP. 20181.

This document is a draft for review purposes only and does not constitute Agency policy.

3-39 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Alanine Aminotransferase (ALT)

Aspartate Aminotransferase (AST)

Study Name

NTP. 2018. 4309127

Wang. 2020. 6323927
NTP. 2018. 4309127

Outcome Confidence

High confidence

Experiment Name

28 Day Oral

High confidence
High confidence

Wang. 2020. 6323927 High confidence

12 Day Oral
28 Day Oral

12 Day Oral

Animal Description

Rat. Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harian) (•_
Mouse, CD-1 ( J)
Rat. Sprague-Dawley (Harlan) (.
Rat, Sprague-Dawley (Harlan) (;.
Mouse. CD-1 (•_')

Trend Test Result

significant
not significant
not applicable
significant
significant
not applicable

PFDA Serum Liver Biomarkers

Alkaline Phosphatase (ALP)

Bile Salt/Acids

Direct Bilirubin

Indirect Bilirubin

Total Bilirubin (TBILI)

NTP. 2018. 4309127

NTP, 2018, 4309127

NTP. 2018. 4309127

High confidence
High confidence
High confidence
High confidence
High confidence

28 Day Oral
28 Day Oral
28 Day Oral
28 Day Oral
28 Day Oral

Rat. Sprague-
Rat. Sprague-
Rat, Sprague-
Rat. Sprague-
Rat. Sprague-
Rat, Sprague-
Rat, Sprague-
Ral, Sprague-
Rat, Sprague-
Rat. Sprague-

Dawley (Harlan) (.
Dawley (Harlan) (-
Dawley (Harlan) ("
Dawley (Harlan) ('
Dawley (Harlan) (^
Dawley (Harlan) (¦
Dawley (Harlan) (
Dawley (Harlan) (
Dawley (Harlan) (
Dawley (Harlan) (.

significant
significant
significant
significant
significant
significant
significant
significant
significant
significant

Albumin (A)

Albumin/Globulin (A/G) Ratio
Globulin (G)

Total Protein (TP)

NTP. 2018. 4309127

NTP. 2018, 4309127

NTP. 2018. 4309127

High confidence
High confidence
High confidence
High confidence

28 Day Oral
28 Day Oral
28 Day Oral
28 Day Oral

Rat, Sprague-Dawley (Harlan) (
Rat. Sprague-Dawley (Harlan) (-
Rat. Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (-.
Rat. Sprague-Dawley (Harlan) (^
Rat, Sprague-Dawley (Harlan) (.•
Rat, Sprague-Dawley (Harlan) (r
Rat, Sprague-Dawley (Harlan) (

not significant

significant

not significant

significant

# No significant change
A Statistically significant
V Statistically significant decrease

Dose (mg/kg-day)

100

Figure 3-7. Effects on serum liver biomarkers following exposure to PFDA in
short-term oral studies in animals (results can be viewed by clicking the HAWC
link).

Organ weight

The studies evaluating liver weight changes in animals consist of four high confidence
studies (see Figure 3-8): two 28-day gavage studies using female B6C3F1/N mice or male and
female S-D rats fFrawlevet al.. 2018: NTP. 20181. one 14-day study in female C57BL/6J mice
exposure via drinking water fLi etai. 20221 and one developmental study measuring effects in
female PO C57BL/6N mice exposed via gavage during gestational days (GD) 6-15 or 10-13 (Harris
andBirnbaum. 19891. The 28-day rat study by Frawlev et al. (20181 included three cohorts
exposed to similar experimental conditions. There are also four medium confidence studies that
administrated PFDA via the diet for 7-14 days in male Wistar rats fKawashima etal.. 19951. male
Fischer F344 rats fTakagi etal.. 1992.19911 or male C57BL/6N mice fPermadi etal.. 19931. Overall
confidence in these shorter duration studies was reduced to medium based primarily on
uncertainties surrounding the characterization of the test compound (no analytical verification
methods) and administered doses (lack of information on food consumption for estimating dietary
exposure doses) (see Figure 3-8). A low confidence study with incomplete reporting on liver
weight data (no information on sample size) is also available in Wistar rats exposed via the diet for
7 days fYamamoto and Kawashima. 19971 (see Figure 3-8).

This document is a draft for review purposes only and does not constitute Agency policy,

3-40 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

vLS"3 ^

Reporting quality

Allocation

Observational bias/blinding
Confounding/variable control
Selective reporting and attrition
Chemical administration and characterization
Exposure timing, frequency and duration
Endpoint sensitivity and specificity
Results presentation
Overall confidence

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NRj Not reported

Figure 3-8, Evaluation results for animal studies assessing effects of PFDA
exposure on liver weight. Refer to HAWC for details on the study evaluation
review.

Increased liver weight was consistently reported across all studies, species, strains, and
sexes (see Table 3-10 and Figure 3-9). Relative liver weight is often preferred over absolute liver
weight as it accounts for variations in body weight that may mask organ weight changes fBailev et
a!.. 20041. Statistically significant increases in relative liver weights were reported in rats and mice
at >0.089 mg/kg-day across the short-term studies, while reductions in terminal body weight
occurred in these animals at higher doses (>1.25 mg/kg-day) (see Section 3.2.10 on General toxicity
effects for more details). In general, the changes in relative liver weights demonstrated a dose and
time dependency. For example, dose-related increases in relative liver weights of 17-56%
compared to controls were reported in male Wistar/Fisher rats at doses 1.15-10 mg/kg-day after
7-14 days across three studies with dietary exposure (females were not examined) fKawashima et
al.. 1995: Takagi et al.. 1992.19911. In female P0 C57BL/6N mice exposed during gestion (GD 10-
13 and 6-15), relative liver weights increased by 12-127% compared to controls at doses of 1-16
mg/kg-day (Harris and Birnbaum. 19891. At a longer exposure duration (28 days), similar
magnitudes of relative liver weight increases were observed in female B6C3F1/N mice and
male/female S-D rats but at lower PFDA doses (16-81% at 0.089-0.71 mg/kg-day and 10-102% at
0.125-2.5 mg/kg-day, respectively). Further, in the studies that evaluated liver weight and other
relevant liver toxicity endpoints, the increases in liver weight corresponded with the reported
observations of hepatocellular histopathology f F rawlev etal.,2018: NTP. 20181 and alterations in
serum biomarkers of hepatocellular/biliary function fNTP, 20181.

This document is a draft for review purposes only and does not constitute Agency policy,

3-41 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-10. Percent change relative to controls in liver weight (relative to
body weight) due to PFDA exposure in short-term oral toxicity studies

Animal group

Dose (mg/kg-d)

0.03-0.045

6800

0.1-0.179

0.25-0.36

0.5-0.71

SZI-OI

2.0-3

4-4.6

6.4-8

9.22-12.8

16-25

32-37.8

7 d; male Wistar rats
(Kawashima et al., 1995)

7 d; male Fisher F344 rats
(Takagi et al., 1992)

14 d; male Fisher F344 rats
(Takagi et al., 1991)

28 d; male S-D rats
(NTP, 2018)

28 d; female S-D rats
(NTP, 2018)

102

28 d; female S-D rats -
Histopathology cohort
(Frawlev et al., 2018)

28 d; female S-D rats - MPS
cohort

(Frawlev et al., 2018)

28 d; female S-D rats - TDAR to
SRBC cohort
(Frawlev et al., 2018)

GD 10-13; pregnant P0 female

C57BL/6N mice

(Harris and Birnbaum, 1989)

-4

106

GD 6-15; pregnant P0 female

C57BL/6N mice

(Harris and Birnbaum, 1989)

106

127

10 d; male C57BL/6N mice
(Permadi et al., 1993)

100

14 d; Female C57BL/6J mice
(Li et al., 2022)

219

28 d; female B6C3F1/N mice
(Frawlev et al., 2018)

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors; shaded cells represent doses not included in the individual studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-42 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Liver Weight, Relative

I No significant change
I Statistically significant increase

Study Name Outcome Confidence

NTP, 2018, 4309127 High confidence

Takagi. 1992. 1320114 Medium confidence

Takagi. 1991, 2325496 Medium confidence
Kawashima, 1995. 3858657 Medium confidence

NTP. 2018.4309127 High confidence

Harris. 1989. 3858729 High confidence

Permadi. 1993. 1332452 Medium confidence

Liver Weight. Relative (Hematology Study) Frawley. 2018. 4287119 High confidence

Liver Weight. Relative (Hislopathology Cohort) Frawley, 2018. 4287119 High confidence

Liver Weight. Relative (MPS Cohort) Frawley. 2018.4287119 High confidence

Liver Weight. Relative (TDAR to SRBC Cohort) Frawley. 2018. 4287119 High confidence

Experiment Name

28 Day Oral
7 Day Oral
14 Day Oral
7 Day Oral
28 Day Oral

Gestational Oral (GD 10-13)
Gestational Oral (GD 6-15)
10 Day Oral PFDA
10 Day Oral
28 Day Oral
28 Day Oral
28 Day Oral
28 Day Oral

Animal Description

Rat. Sprague-Dawley (Harlan)
Rat. Fischer F344 (¦ ¦)
Rat. Fischer F344 ($)
Rat. Wistar (••")

Rat. Sprague-Dawley (Haiian)
P0 Mouse. C57BU8n <.)
P0 Mouse. C57BL/6n ( )
Mouse. C57BI/6 (J)

Mouse. C57BI/6 ( \)

Mouse. B6C3F1/N ( )
Rat, Sprague-Dawley (Harlan)
Rat. Sprague-Dawley (Harlan)
Rat. Sprague-Dawley (Harlan)

Trend Test Result

significant
not applicable
not applicable
not reported
significant
significant
significant
not applicable
not reported
significant
significant
significant
significant

PFDA Liver Weight

_A-A-A-A

0.1 1 10 100
mg.'kg-day

Figure 3-9. Effects on relative liver weight following exposure to PFDA in
short-term oral studies in animals (results can be viewed by clicking the HAVvC
link).

Mechanistic studies and supplemental information

The liver effects in response to oral exposure to PFDA in short-term animal studies consist
of increased serum biomarkers of liver function, increased liver weight and increased incidence of
hepatocellular lesions (e.g., cytoplastic alterations, vacuolation, and to a lesser extent necrosis).
Increased liver weight and hepatocellular hypertrophy can be associated with changes that are
adaptive in nature fHall et al.. 20121. and not necessarily indicative of adverse effects unless
observed in concordance with other clinical, pathological markers of overt liver toxicity (see PFAS
Protocol; Appendix A). As discussed in the protocol, Hall etal. (20121 was focused on framing liver
effects in the context of progression to liver tumors so additional information was considered when
evaluating noncancer liver effects for PFDA exposure. The additional information consists of
multiple in vitro and in vivo mechanistic studies in rodents (including peroxisome proliferator
activated receptor alpha (PPARa)-null mice) and limited studies in human-relevant models (mostly
in vitro systems but also studies in animal models with reduced PPARa sensitivity) as well as
evidence from other PFAS that help elucidate possible modes of action of PFDA liver toxicity.

Summary of mechanistic studies for PFDA

Mechanistic evidence relevant to potential PFDA-induced liver effects was collected from
the peer-reviewed literature and from in vitro high-throughput screening (HTS) assays accessed
through the EPA's CompTox Chemicals Dashboard f11ttps: //comptox. e p a. go v / d a s hboard 1 f fU.S.
EPA. 2021al: data were retrieved on November 2022). Given the relatively abundant evidence base
compared to most other PFAS, the available in vitro and in vivo mechanistic studies on PFDA were
considered in the context of what is known about the mode-of-action (M0A) for hepatic effects
elicited by related PFAS, including PF0A and PF0S, the most well-studied PFAS. M0A information
for PF0S and PF0A was based primarily on published reviews. As discussed in the systematic
review protocol (Appendix A), an Adverse Outcome Pathway (AOP)-type approach was employed
to organize and discuss the evidence according to the following levels of biological organization:
molecular interactions, cellular effects, organ effects, and organism effects. A summary of the

This document is a draft for review purposes only and does not constitute Agency policy.

3-43 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

mechanistic and supplemental evidence related to the potential mechanisms of hepatotoxicity for
PFDA is provided below. A detailed description of the methodology and results of the analysis
undertaken herein can be found in Appendices D and E.

Mechanistic evidence from in vivo and in vitro rodent cell models indicates that PFDA can
activate (potentially directly) several xenobiotic-sensing nuclear receptors and other cell signaling
pathways, namely PPARa, constitutive androstane receptor (CAR)/pregnant x receptor (PXR),
nuclear factor erythroid 2 related factor 2 (Nrf2), tumor necrosis factor alpha (TNFa), nuclear
factor kappa B pathway (NFkB) and c-Jun-N-terminal kinase (JNK)/activating transcription factor 2
(ATF-2) (see Appendix D.3.1 on Molecular initiating events for more details). PFDA exposure was
also associated with alterations in the hepatic expression and activity of XME enzymes, ROS
production and markers of oxidative damage (DNA oxidation and lipid peroxidation), disruption of
mitochondrial functions, induction of inflammatory responses, cellular damage/stress and
abnormal liver metabolic functions related to bile acid, glucose, and lipid metabolism in animals
(see Appendix D.3.2 on Cellular effects for more details). These molecular and cellular mechanisms
are associated with chemical-induced liver disorders such as steatohepatitis and fibrosis (Angrish
etal.. 2016: Cao etal.. 2016: Toshi-Barve etal.. 2015: Wahlang etal.. 20131 and provide support for
the biological plausibility of the observed liver effects in rats and mice after short-term PFDA
exposure (see synthesis of Animal studies in this Section for more details).

The available mechanistic information in human models is limited to a few in vitro studies
in the peer-reviewed literature and ToxCastandTox21 HTS assay results

fhttp s: //co mp tox. ep a. gov /dashboard!. The available evidence suggests some concordance with
responses evaluated in animal models. PFDA could modulate the activity of a number of human
nuclear receptor pathways potentially relevant to its mechanism(s) of hepatotoxicity. For example,
PFDA activated PPARa in primary and immortalized human liver cell lines (Rosenmai etal.. 2018:
Buhrke etal.. 2013: Rosen etal.. 2013) and Table E-2 for in vitro HTS assay results) and exhibited
direct binding towards the human PPARa in vitro (Ishibashi et al.. 2019). However, there was
reduced sensitivity in the binding and transcriptional activity towards the human PPARa compared
to the mouse, Baikal seal and polar bear isoforms in some studies (Ishibashi et al.. 2019: Routti et
al.. 2019: Wolf etal.. 2012: Wolf etal.. 2008). Reduced PPARa sensitivity in human versus rodent
models (i.e., rats and mice) has been previously demonstrated in studies with other perfluorinated
compounds fCorton etal.. 2018: Wolf etal.. 2012: Wolf etal.. 20081. Moreover, PFDA activated
nuclear receptors other than PPARa in human liver cell lines (i.e., PPARy, PXR and FXR) and
displayed high potency towards the human FXR in a receptor-ligand binding assays fBuhrke et al.
(2013). Rosen etal. (2013). Zhang etal. (2017) and T able E-2 for in vitro HTS assay results). At the
cellular level, PFDA elevated ROS production and induced markers of cellular stress and
cytotoxicity in human hepatoma HepG2 cells (see Appendix D.3.2 on Cellular effects for more
details).

This document is a draft for review purposes only and does not constitute Agency policy.

3-44 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

PPARa activation is described as one of the mechanisms through which perfluorinated
compounds induce liver toxicity in animals (ATSDR. 2018b: U.S. EPA. 2016a. b). PPARa appears to
be important for disruption of bile acid homeostasis and downstream effects related to bile acid
synthesis and transport mechanisms, as well as, signaling pathways associated with cellular stress
and anti-inflammatory responses in PFDA-exposed mice fLuo etal.. 20171. However, other
responses appear to occur, at least in part, independently of PPARa. Rosen etal. f20131 reported
transcriptional induction of PPARa-dependent and -independent genes in primary human
hepatocytes exposed to PFDA. Lim etal. (20211 showed that PFDA-mediated transcriptional
regulation of transporters involved in metabolism and xenobiotic biotransformation in HepaRG
cells was more consistent with activation of the ROS-sensitive transcription factor Nrf2 as opposed
to PPARa or CAR. Increased liver weight and activation of Nrf2 were reported after PFDA treatment
in both WT and PPARa-null mice fLuo etal.. 2017: Maher etal.. 20081. PFDA-mediated induction of
hepatic Mrp transporters involved in cholestasis was attenuated in mice devoid of Nrf2 or Kupffer
cell function (Maher et al.. 20081. A study that evaluated PFDA animal models known to be
generally resistant to PPARa activation (i.e., Guinea pigs and/or Syrian hamsters) displayed
histological responses indicative of hepatocellular stress, mitochondrial damage, hepatic lipid
accumulation and liver enlargement with PFDA exposure fVan Rafelghem etal.. 1987bl.
Noteworthy, hepatic lipid accumulation was characterized as more pronounced in Guinea pigs and
Syrian hamsters compared to rats and mice and the opposite was found for peroxisome
proliferation (Van Rafelghem etal.. 1987b). Finally, the NTP (20181 study that reported PFDA-
induced liver effects in rats exposed for 28 days, also evaluated the effects of the potent PPARa
inducer, Wyeth-14,643, on the liver. Similar to PFDA, Wyeth-14,643 caused increases in liver
weights, hepatocyte hypertrophy and changes in serum liver biomarkers (e.g., increased ALT, ALP
and AST) in rats; however, unlike PFDA, Wyeth-14,643 exposure was not associated with any
structural degenerative changes (i.e., hepatocyte necrosis).

Overall, the mechanistic evidence supports the biological plausibility of liver effects
observed in animal bioassays. Further, the available data indicate a likely role for both PPARa-
dependent and -independent mechanisms in the hepatotoxicity of PFDA in animals. Existing
evidence from in vitro studies and animal models considered more relevant to humans with respect
to PPARa sensitivity suggest that some responses may be conserved across species (including
activation of relevant nuclear receptor pathways [PPARa/y, PXR and FXR] and outcomes related to
hepatocellular stress, mitochondrial damage, lipid accumulation and liver enlargement). Taken
together, these data provide some support for the potential human relevance of the observed
hepatic effects in animals. Some uncertainties remain based on differences in experimental design
and/or confounding effects with cytotoxicity in in vitro test systems, as well as limited information
available from in vivo models to characterize the putative involvement of PPARa and other cell
signaling pathways in the mechanisms of hepatotoxicity of PFDA in animals and humans (see
Appendix D.3 and E.l for more details).

This document is a draft for review purposes only and does not constitute Agency policy.

3-45 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence from related PFAS

Based on the limitations in the mechanistic evidence for PFDA described above, studies
investigating the effects of structurally related long-chain PFAS (perfluoroalkyl sulfonic acids
containing > 6 carbons or PFCAs with > 7 carbons) are summarized herein, focusing on studies
conducted in null and humanized animal models identified from literature searches conducted for
other ongoing EPA PFAS IRIS assessments (i.e., PFHxS and PFNA) or in final EPA human health
assessments (i.e., PFOA and PFOS). Data in these models available for short-chain PFAS (e.g., PFBA)
are not summarized herein, as they were considered less relevant to PFDA exposure than those
data available for long chain PFAS, although extrapolations from other PFAS are all inherently
uncertain.

Gene expression profiling in response to exposure to several long-chain PFAS has been
evaluated in wild-type and PPARa-null mice and the results indicate a role for both PPARa-
dependent and independent pathways in the liver effects of these compounds. Gene expression
changes induced by PFOA, PFHxS and PFNA in wild-type mouse livers were largely attributable to
PPARa; however, a subset of transcriptional changes related to lipid metabolism, inflammation and
xenobiotic metabolism occurred in PPAR-a null mice that reflect potential activation of additional
nuclear receptors such as CAR and PPARy fRosen etal.. 2017: Rosen etal.. 2010: Rosen etal.. 20081.

Consistent with transcriptional regulation, the data support that these long-chain PFAS
induced tissue-level responses which are likely to be mediated by PPARa- dependent and
independent mechanisms. Increases in liver weight and hepatocyte hypertrophy and/or
proliferation were reported in PPARa wild-type and null mice exposed to PFOA (Das etal.. 2017:
Wolf etal.. 20081. Similarly, hepatomegaly (characterized by increased liver weight and cell size
and decreased DNA content) and hepatic lipid accumulation (indicating or leading to steatosis)
were observed with PFHxS or PFNA exposure in wild-type mice and mice devoid of PPARa function
fDas etal.. 20171. In contrast, these liver effects were only induced in wild-type animals treated
with the prototype PPARa agonist, Wyeth 14,643. Nakagawa etal. (20121 showed elevated levels
of hepatic triglycerides in wild-type, PPARa-null and humanized PPARa (hPPARa) mouse strains
exposed to ammonium perfluorooctanoate, but macrovesicular and/or microvesicular steatosis in
PPARa-null and hPPARa mice only. Additionally, PFOS and PFHxS decreased triglyceride and
cholesterol levels in plasma and increased triglycerides in the liver of APOE*3-Leiden CETP mice,
which exhibit attenuated clearance of apoB-containing lipoproteins and human-like lipoprotein
metabolism on a Western diet fBiiland etal.. 20111. Likewise, PFDA exposure was associated with
marked increases in hepatic lipid content (including triglyceride levels) and accumulation in rats
and mice (Kudo and Kawashima. 2003: Adinehzadeh and Reo. 1998: Kawashima etal.. 1995:
Sterchele etal.. 1994: Brewster and Birnbaum. 1989: Harrison etal.. 1988: Van Rafelghem etal..
1988b: Van Rafelghem etal.. 1987a). as well as, Guinea pigs and Syrian hamsters (Van Rafelghem et
al.. 1987bl. which like humans, appear to be less responsive to PPARa activation.

This document is a draft for review purposes only and does not constitute Agency policy.

3-46 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

The precise mechanism(s) of how these long chain PFAS induced hepatic lipid accumulation
and the potential association of this accumulation with progression to steatosis remain unclear.
Das etal. f20171 showed that PFOS, PFHxS, and PFNA, which are known to induce significant
hepatic lipid accumulation in animals, alter the expression of genes involved in fatty acid synthesis
and oxidation in mouse livers, and that these transcriptional changes are partly independent of
PPARa fDas etal.. 20171. The authors hypothesized that perfluorinated compounds disrupt the
balance of fatty acid synthesis and oxidation in favor of accumulation, which leads to steatosis. In
contrast, exposure to potent PPARa activators such as Wyeth 14,643, is not associated with
steatosis-like changes because, these compounds likely favor fatty acid oxidation over
synthesis/accumulation fDas etal.. 20171.

Collectively, studies in PPARa null and humanized animal models for structurally related
long chain PFAS are consistent with the plausible PPARa-dependent and independent mode of
action for PFDA liver toxicity and add further support to the potential human relevance of the
observed liver effects in animals. Further, the evidence suggests that these perfluorinated
compounds have the potential to induce steatosis, a well-known chemical-induced response that
can progress to steatohepatitis, fibrosis, and impaired liver function (Al-Ervani etal.. 20151.

Considerations for potentially adaptive versus adverse responses

Increases in liver weight and hepatocyte hypertrophy were observed in rodents with PFDA
administration in short-term oral studies (see Figure 3-5 and 3-9 above). Enlargement of the liver
and/or individual hepatocytes is a common chemical-induced response that can involve lipid
accumulation (e.g., micro- or macro-vesicular steatosis), organellar growth and proliferation
(e.g., peroxisomes, endoplasmic reticulum), increased intracellular protein levels (e.g., Phase I and
II enzymes), and altered regulation of gene expression (e.g., stress response, nuclear receptors)
(reviewed by Batt and Ferrari f 199511. Hepatocyte hypertrophy related to organelle growth and
proliferation in response to activation of xenobiotic-sensing receptors (primarily PPARa) is often
considered an adaptive response (Hall etal.. 20121. However, hepatocyte swelling is also
associated with cell death processes, oncosis or oncotic necrosis (Kleiner etal.. 20121. which
occurred in several liver diseases or conditions, such as ischemia-reperfusion injury, drug-induced
liver toxicity, and partial hepatectomy (Kass. 2006: Taeschke and Lemasters. 20031. Furthermore,
mechanistic evidence for PFDA and other long-chain PFAS suggests that in addition to PPARa
induction, these compounds activate non-PPARa-related mechanism of liver toxicity (see Appendix
D.3 and E.l for more details on the synthesis of PFDA-induced mechanisms of hepatotoxicity).

Hall etal. (20121 indicated that concordant histopathological evidence of degenerative or
necrotic changes (e.g., hepatocyte necrosis, fibrosis, inflammation, steatosis, biliary degeneration,
necrosis of resident cells within the liver) can be used to support the argument that liver
weight/hepatocyte enlargement are adverse (Hall etal.. 20121. In addition to increases in liver
weight and/or hepatocyte hypertrophy, PFDA caused cytoplasmic alterations and vacuolization as
well as, necrosis in rat hepatocytes across two high confidence 28-day gavage studies fFrawlev et

This document is a draft for review purposes only and does not constitute Agency policy.

3-47 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

al.. 2018: NTP. 20181. Cytoplasmic alterations of minimal to marked severity were observed in
nearly all male and female rats at >0.625 mg/kg-day fNTP. 2018], Cytoplasmic vacuolization of
minimal/mild to moderate severity occurred in males and females at >1.25 mg/kg-day fNTP. 20181.
Minimal necrosis was reported in females in the two 28-day studies with statistically significant
increases at the highest dose, 2.5 mg/kg-day fFrawlev etal.. 2018: NTP. 20181. Male rats were only
tested in one study, showing increased incidence ofhepatocyte necrosis, but the effect was not
dose-dependent fNTP. 20181. The lesions show a clear pattern of increased hepatocyte
damage/injury with dose, ranging from cytoplasmic changes to hypertrophy to necrosis fNTP.
20181. The necrotic lesions were accompanied in some cases by evidence of an initial inflammatory
response fNTP. 20181 and, although these changes were characterized as minimal, the findings
indicate some degree of structural degeneration considered adverse and that may progress to more
severe liver pathologies with increasing dose or exposure duration. Consistent with these
observations, steatosis, necrosis, edema, and degeneration were reported in mice at 13 mg/kg-day
and extensive lipid accumulation was reported in rats at 9.22 mg/kg-day in low confidence short-
term studies with PFDA administered orally (Wang etal.. 2020: Kawashima etal.. 19951. Acute i.p.
studies provide additional support for the accumulation of lipids in the liver with PFDA exposure
(see synthesis of Metabolic Effects in Appendix D.3.2), which is a key event leading to hepatic
steatosis fAngrish etal.. 20161. As discussed above, steatosis is a common liver response in animals
that is associated with exposure to perfluorinated compounds such as PFOA, PFHxS or PFNA.
Sustained steatosis can progress to steatohepatitis and other adverse liver diseases such as fibrosis
and cirrhosis (Angrish etal.. 20161.

Alterations in serum liver biomarkers were also present in rats that exhibited increases in
liver weight, hepatocyte hypertrophy and other histological lesions (i.e., necrosis) after 28-day
gavage exposure to PFDA fNTP. 20181. According to Hall etal. f20121. clinical markers of liver
damage and function can provide evidence in support of the adversity of concomitant increases in
liver weight/hepatocyte hypertrophy. These authors suggested that a weight-of-evidence approach
should be applied when evaluating clinical marker data, considering dose-dependent and
biologically significant changes in at least two of the following parameters: 2- to 3-fold increase in
ALT; change in biomarkers of hepatobiliary damage (e.g., AST, ALP and y-glutamyltranspeptidase
[yGT]); a change in biomarkers of liver dysfunction (e.g., albumin, bilirubin, bile acids/salts and
coagulation factors). PFDA increased ALT levels in female rats at >1.25 mg/kg-day fNTP. 20181:
similar changes were observed in male rats, but the effects did not show a significant trend.
Although the increases in circulating ALT levels in females were relatively small (20-44% or 1.2- to
1.4-fold), concordant changes in other clinical biomarkers occurred in these animals. Dose-
dependent increases in ALT and ALP were found in male and female rats at >0.156 mg/kg-day.
Similarly, levels of bile salts/acids and bilirubin were elevated in rats of both sexes at >1.25 mg/kg-
day, exhibiting marked changes (205-1,207% or 3.1- to 13.1-fold for bile acids/salts and 28-733%
or 1.3- to 8.3-fold for bilirubin).

This document is a draft for review purposes only and does not constitute Agency policy.

3-48 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Overall, application of the recommendations from Hall etal. (20121 clearly supports the
conclusion that PFDA exposure has multiple and coherent effects on liver histopathology, serum
biomarkers and liver weights in exposed animals (primarily rats) that meet the criteria for
adversity.

Evidence integration

There is slight evidence of an association between PFDA exposure and hepatic effects in
humans based on associations with liver biomarkers in the blood in eight studies (see Table 3-5).
Positive associations between exposure to PFDA and ALT were observed in four of five studies of
adults. However, there is inconsistency in the direction of association within other specific clinical
markers and lack of coherence across clinical markers that reduces the strength of the
evidence.

The evidence for PFDA-induced liver effects from short-term animal studies via the oral
route is considered moderate based on coherent effects across multiple endpoints relevant to the
assessment of liver toxicity (serum biomarkers, histopathology, and organ weight) (see Figures 3-5,
3-7 and 3-9). Increases in serum biomarkers of hepatocellular/hepatobiliary injury (ALT, AST, ALP,
bile salts/acids and bilirubin) fNTP. 2018] and liver weights were reported in male and female S-D
rats at>0.156 mg/kg-day after 28-day gavage exposure fFrawlev etal.. 2018: NTP. 20181. In
general, the responses were consistent in directionality across sexes and dose groups, exhibiting a
clear dose-response gradient Furthermore, the evidence for increased liver weights was consistent
across several species (rats and mice), strains (S-D, Wistar, Fischer F344, C57BL/6N, C57BL/6J and
B6C3F1/N) and exposure designs (gavage and dietary) (see synthesis of Organ weight in this
Section for more details). At higher doses (>0.5 mg/kg-day), a consistent pattern of hepatocellular
lesions was observed in S-D rats that included cytoplastic alterations and vacuolization,
hypertrophy, and necrosis fFrawlev etal.. 2018: NTP. 20181. The pattern of hepatocellular changes
showed a progression in severity within and across lesions with an increase in exposure dose,
which adds certainty to the interpretation of the evidence. In combination with the
histopathological findings, alterations in serum biomarkers and liver weights support the
development of adverse liver effects in rats after continuous PFDA exposure (see section on
Considerations for potentially adaptive versus adverse responses above). The evidence base is
limited in that there is an absence of studies via relevant exposure routes with durations longer
than 28 days (i.e., no subchronic and chronic exposure studies) examining potential hepatic effects
of PFDA exposure.

Analysis of mechanistic and supplementary data from in vivo and in vitro rodent models
provide experimental (e.g., liver weight changes after i.p. exposure) and biological support for the
phenotypic effects reported in the short-term oral studies summarized above. Exposure to PFDA
was associated with the activation of several molecular signaling pathways and altered cellular
functions hypothesized to be involved in the MOA for liver toxicity of related perfluorinated
compounds (see Summary on mechanistic and supplementary studies for PFDA and Appendices B

This document is a draft for review purposes only and does not constitute Agency policy.

3-49 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

and C for more details). Additionally, the evidence for PFDA-mediated liver effects implicates both
PPARa-dependent and -independent mechanisms.

The activation of PPARa in the MOA for non-cancer liver effects in rodents has implications
to human health assessment based on perceived differences in PPARa response between rats/mice
versus humans. PFDA can activate the human PPARa in vitro but it exhibits less sensitivity towards
the human isoform in comparison to other mammalian species. PFDA also interacts with other
nuclear receptors and cell signaling pathways relevant to its potential mechanism of hepatotoxicity
in both human and animal models. Furthermore, some hepatic responses in animals occurred, at
least in part, independent of PPARa or were found to be activated in human in vitro assays or
animal models that are more relevant to humans with respect to PPARa sensitivity (see Summary
on mechanistic studies for PFDA and Appendices D.3 and E.l for more details). These observations
are consistent with studies in PPARa null and humanized animals for other long-chain PFAS such as
PFOA, PFHxS and PFNA that suggest non-PPARa mechanisms of liver toxicity (see Evidence for
other PFAS for more details). Given that the precise role of PPARa in the non-cancer liver effects of
PFDA remains largely unknown and the possible involvement of PPARa-dependent and
independent pathways, the effects observed in animals are considered potentially relevant to
humans fSoldatow etal.. 20131.

Taken together, the available evidence indicates that PFDA exposure is likely to cause
hepatotoxicity in humans given sufficient exposure conditions9 (see Table 3-11). This conclusion is
based primarily on coherent liver effects in rats (and, to a lesser extent, mice) exposed to doses
>0.156 mg/kg-day for 28 days. The available mechanistic information overall 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.

9 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes only and does not constitute Agency policy.

3-50 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-11. Evidence profile table for PFDA exposure and liver effects

Evidence stream summary and interpretation

Evidence integration summary
judgment

Evidence from studies of exposed humans (see Section 3.2.1: Human studies

0®Q

Evidence indicates (likely)

Primary basis:

Two high confidence studies in rats
at >0.156 mg/kg-d after short-term
exposure

Human relevance:

Effects in rats are considered
relevant to humans (see Section
3.2.2: Mechanistic Evidence and
Supplemental Information)

Cross-stream coherence:

Alterations in serum liver
biomarkers were reported in
animals and in a few
epidemiological studies, although
the latter observations are
uncertain.

Susceptible populations and
lifestages: None identified, although
individuals with pre-existing liver
disease could potentially be at
greater risk

Other inferences: the MOA for
PFDA-induced liver effects is
unknown, although the available
evidence indicates the involvement
of PPARa-dependent and -
independent mechanisms

Studies and confidence

Summary and key
findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

Serum Biomarkers
7 medium confidence
studies (5 in adults, 1 in
adults and children, 1 in
children)

• 6/7 studies in adults
reported positive
associations
between some
clinical liver
function markers
and PFDA exposure

• 4/5 studies of ALT in
adults reported
positive
associations

• Exposure-response
gradient observed
in one study that
examined it

• Consistency for ALT

• Lack of expected
coherence across
related clinical
markers

• Unclear biological
significance of small
changes in ALT

®oo

Slight

Evidence of a positive
associations between PFDA
exposure and ALT in adults,
but there is a large degree
of uncertainty due to
inconsistency among other
clinical markers and lack of
clear adversity

Evidence from in vivo animal studies via the oral route (see Section 3.2.1: Animal studies)

Studies and confidence

Summary and key
findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

Histooatholosv

2 high and 3 low
confidence studies in
adult rats or mice

• 7-day dietary

• 12- and 14-day
drinking water

• 28-day gavage (2x)

• Hepatocellular
lesions (ranging
from cytoplasmic
alterations to
necrosis) at

>0.5 mg/kg-d across
high confidence
studies

• Other liver lesions
(i.e., lipid
accumulation and
edema) were found
in low confidence
studies at higher

• Consistency across
two high confidence
studies

• Coherent pattern of
hepatocellular
lesions across all
studies

• Increased severity
(within and across
lesions) with
increasing exposure

• No factors noted

0®Q

Moderate

Consistent and coherent
changes in serum
biomarkers, histopathology,
and liver weights, with the
strongest evidence in rats at
>0.156 mg/kg-d although
data are limited to short-
term studies. Taken
together, the coherent
changes across markers of
hepatic injury were judged

This document is a draft for review purposes only and does not constitute Agency policy.

3-51 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

doses (>4.6 mg/kg-
day)

as adverse (see
"Considerations for
potentially adaptive versus
adverse responses")

Serum biomarkers

2 high confidence
studies in adult rats or
mice

• Increased serum
markers of liver and
hepatobiliary
toxicity at
>0.156 mg/kg-d

• Consistency across
high confidence
studies

• Coherence across
serum markers

• No factors noted

• 12-day drinking
water

• 28-day gavage

• Dose-response
gradient for most
effects

Organ weight

4 high, 4 medium and 1
low confidence studies
in adult rats and mice.

• Increased relative
liver weights at
>0.089 mg/kg-d

• Consistency across
all studies, including
multiple species
and both sexes

• No factors noted

• 7-14-day dietary
(5x)

• 14 day drinking
water (lx)

• Dose-response
gradient

• Coherence with
serum markers and
histopathology

• 28-day gavage (2x)

• Gestational gavage
(lx)

Mechanistic evidence and supplemental information (see subsection above)

Biological events or
pathways

Primary evidence evaluated

Key findings, interpretation, and limitations

Evidence stream judgment

Molecular initiating
events— PPARa and
other cell signaling
pathways

Key findings and interpretation:

• Evidence of activation of PPARa, CAR/PXR, Nrf2, TNFa, NFkB and JUNK/ATF-
2 in rodent hepatic in vivo and/or in vitro models.

Evidence of PPARa-
dependent and -
independent pathways in
studies in rodents and

This document is a draft for review purposes only and does not constitute Agency policy.

3-52 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

• Some evidence of activation of PPARa/y, PXR and FXR in human liver cells
and/or cell-free binding assays

• The human FXR was a sensitive target for PFDA in vitro.

• Reduced sensitivity towards the human PPARa compared to Baikal seal,
polar bear and mouse PPARa isoforms in vitro.

Limitations: Lack of humanized in vivo models. Some inconsistencies in the vitro
results may be due to differences in experimental model and/or design or
confounding issues with cytotoxicity.

human in vitro models that
support the biological
plausibility of PFDA-induced
liver effects.

Cellular effects

Key findings and interpretation:

• Alterations in hepatic XMEs, oxidative stress, cell and mitochondrial
damage, inflammation, and liver metabolic functions in rodents.

• PPARa appears to be important for disrupting bile acid homeostasis in mice
and associated downstream effects.

• Activation of Nrf2 in wildtype and KO PPARa mice and observations of
hepatocellular stress, mitochondrial damage and lipid accumulation in
animal models known to be less responsive to PPARa activation (i.e., Guinea
pigs and/or Syrian hamsters) support involvement of PPARa-independent
mechanisms.

• PFDA increased ROS production and markers of cellular stress/cytotoxicity
in human hepatoma HepG2 cells.

Limitations: Few studies examining the role of PPARa and other cell signaling
pathways and no evidence in humanized vivo models. Inconsistencies in the in
vivo results are likely attributable to differences in experimental model and/or
design features.

Organ-level effects

Key findings and interpretation:

• Increased liver weights in rats, mice (both WT and PPARa-KO animals) and
in a rodent species known to be resistant to PPARa activation (i.e., Syrian
hamsters).

Limitations: Lack of evidence examining other organ-level effects, including
histological evidence.

This document is a draft for review purposes only and does not constitute Agency policy.

3-53 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.2.2. IMMUNE EFFECTS

Methodological considerations

Immune-related health effects evaluated from human and animal studies are grouped
according to immunotoxicity guidance from the World Health Organization/International
Programme on Chemical Safety (WHO/IPCS) and considered for evidence of major categories of
immunotoxicity: (1) immunosuppression, (2) immunostimulation, (3) sensitization and allergic
response, or (4) autoimmunity and autoimmune disease flPCS. 20121. Evidence for potential
immune effects is considered within these four categories because of common and related
mechanisms. Within each category, health effects data are organized and discussed from most to
least relevant for drawing hazard conclusions about immunotoxicity (IPCS. 20121. For human data,
clinical studies on disease or immune function assays are considered most relevant, then
general/observational immune assays (lymphocyte phenotyping or cytokines), and finally
endpoints such as hematology (i.e., blood leukocyte counts) are considered least relevant.

Similarly, animal data are presented from most to least relevant for immunotoxicity assessment as
described by WHO/IPCS as follows: host resistance, immune function assays, general/observational
immune assays, blood leukocyte counts and immune organ histopathology and weights flPCS.
20121. The available human and animal evidence provide relevant information for the assessment
of immunosuppression and sensitization or allergic response. However, the available evidence is
lacking or inappropriate to specifically address the potential for immunostimulation and
autoimmunity following PFDA exposure; therefore, these categories of potential immunotoxicity
are not discussed further.

Human studies

Epidemiology studies examining immune effects of PFDA exposure include studies on
antibody response, infectious diseases, and hypersensitivity-related outcomes, which includes
asthma, allergies, and atopic dermatitis. Outcomes related to immunosuppression were considered
within two subcategories: antibody response and infectious disease incidence. Several different
outcomes were included in the sensitization and allergic response category. The health effects
evidence from human studies is summarized below for each category.

Antibody response outcomes

The production of antigen-specific antibodies in response to an immune challenge
(e.g., vaccination in humans or injection with an antigen [e.g., sheep red blood cells] in rodents) is a
well-accepted measure of immune function included in risk assessment guidelines and animal
testing requirements for immunotoxicity fICH Expert Working Group. 2005: U.S. EPA. 1998: IPCS.
19961. Antibodies are proteins circulating in blood and other body fluids that bind to antigens and
thereby identify them for destruction or removal. The production, release, and increase in

This document is a draft for review purposes only and does not constitute Agency policy.

3-54 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

circulating levels of antigen-specific antibodies are important for protection against infectious
agents and preventing or reducing severity of influenza, respiratory infection, colds, and other
diseases as part of the humoral immune response. Reduced antibody production is an indication of
immunosuppression and may result in increased susceptibility to infectious diseases generally (i.e.,
not limited to those specifically studied).

There are five studies (six publications) that examined PFDA exposure and antibody
responses following vaccination for diphtheria or tetanus in children and adults; study evaluations
are summarized in Figure 3-10 and Table 3-12. These included three independent prospective
birth cohorts in the Faroe Islands, all medium confidence, one with enrollment in 1997-2000 and
subsequent follow-up to age 7 fGrandiean etal.. 20121 and age 13 fGrandiean etal.. 2017al. one
with enrollment in 2007-2009 and follow-up to age 5 fGrandiean etal.. 2017bl. and one with
enrollment in 1986-1987 and follow-up to age 28 (Shih etal.. 2021). Shih etal. (2021) also
examined antibody response to Hepatitis types A and B vaccination. These three cohorts are all
separate study populations born in the Faroe Islands and enrolled at different times and thus
considered independent of each other. The analyses, in Grandiean etal. f2017bl combined new
data from the cohort born in 2007-2009 with new follow-up data from the cohort born in 1997-
2000 fGrandiean etal.. 20121. which are labeled in the results table. There was also a cross-
sectional study of children in Greenland (Timmermann etal.. 2021). These studies were generally
well conducted, but exposure contrast was a concern in most of them, with median exposure levels
around 0.3 ng/mL and interquartile ranges around 0.2 ng/mL (exposure contrast was slightly
better in (Timmermann etal.. 2021)1. Potential for confounding across PFAS was considered in
individual studies evaluations as well as across studies in evidence synthesis (see below). In
addition to these developmental exposure studies, there was one study in healthy adult volunteers
in Denmark considered low confidence because of limited information provided on recruitment of
study subjects, lack of consideration of confounders, and a small study population (12 individuals)
leading to concerns with potential selection bias, confounding and low sensitivity (Kielsenetal..
20161.

This document is a draft for review purposes only and does not constitute Agency policy.

3-55 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Grandjean, 2016, 3858518-

Grandjean, 2017, 4239492

Kielsen, 2016. 4241223-

Shih, 2021. 9959487-

Timmermann, 2021. 9413315-

liH® _,BWC

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 3-10. Summary of evaluation of epidemiology studies of PFDAand
antibody response to vaccination. Refer to HAWC for details on the study
evaluation review: HAWC Human Vaccine Response Effects.

Domain and overall confidence ratings may vary by outcome; outcome-specific ratings and rationales are available
in HAWC and described in the relevant sections below. Multiple publications of the same study: Grandjean et al.
(2017a) also represents Grandjean et al. (2012).

The two prospective birth cohorts in the Faroe Islands with antibody levels measured
during childhood reported inverse associations between higher concentrations of serum PFDA and
lower anti-vaccine antibody levels for diphtheria and tetanus (see Table 3-12). Although results
were not always statistically significant, the general trend towards lower antibody levels was
apparent. Antibody levels were measured in individuals of several age groups (and therefore
different lengths of time since their initial vaccination or booster vaccination) and compared to
serum PFDA concentrations also measured at different ages. Although results were not always
statistically significant, inverse associations were observed in most (but not all) of these
comparisons. No biological rationale is understood as to whether one time period is more
predictive of an overall immune response and given the long half-life of PFDA (approximately 4.5-
12 years), there is reasonably high correlations across time periods fGrandiean etal.. 2017al.
Antibodies to diphtheria decreased with increasing PFDA concentrations in 11 of the 13 exposure
and outcome measurement timing combinations assessed. One of the two results that did not
support the trend was a statistically significant increase in diphtheria antibodies in children at
5 years of age (before receiving the 5-year booster) associated with increases in PFDA
concentrations at 18 months of age. This increase appears to be a response in this specific
exposure and outcome timing combination in the 2007-2009 cohort as there was an increase with
all PFAS measured at 18 months and outcome measured at 5 years of age in the 2007-2009 cohort
However, the 1997-2000 cohort from the same population and all other exposure and outcome
timing combinations, including in the 2007-2009 cohort when exposure was measured at birth,
resulted in a decrease of diphtheria antibodies (Grandjean etal.. 2017b). There is no clear
explanation for the discrepant findings for this specific exposure and outcome timing combination

This document is a draft for review purposes only and does not constitute Agency policy.

3-56 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

in the 2007-2009 cohort The only other result that did not show a decrease in diphtheria
antibodies was among 7-year-olds based on maternal PFDA concentration Grandiean etal. f20121
However, because a decrease in diphtheria antibodies was observed within 7-year-olds when PFDA
concentrations were measured at age 5, the lack of effect may be explained by differences in the
long-term influence of the maternal exposure measurement

Similar to the diphtheria results, tetanus antibodies had a decreasing trend with increasing
PFDA concentrations with few exceptions (10 of the 13 combinations indicated decreased antibody
levels). One of the exceptions is a statistically significant increase in tetanus antibodies in
7-year-olds with increasing maternal PFDA concentrations (similar to the discrepancy observed for
diphtheria for a similar exposure-outcome combination). Tetanus antibody levels at 13 years of age
were also increased with increasing PFDA concentrations measured in the children at ages 7 and
13 years of age (Grandiean etal.. 2017a). This may indicate that by 13 years of age, the effect of
maternal and childhood exposure is less relevant to tetanus antibody levels.

The other two studies of developmental exposure and antibody response to vaccination
reported less consistent findings. The cross-sectional results in Timmermann etal. f20211 differed
in direction of association based on the covariate set selected (with or without adjustment for area
of residence). 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. However, a subset
of the study population did have maternal samples available, and those results were also
inconsistent by vaccine. On the other hand, this study was the only one to examine the odds ratio
for not being protected against diphtheria (antibody concentrations <0.1 IU/mL), which has clear
clinical significance, and they reported an OR of 5.08 (95% 1.32,19.51) among children with known
vaccination records (adjusted for area of residence, consistent with continuous antibody results).
Shih etal. (2021). which examined antibody levels at age 28 with exposure measures at multiple
time points, reported inconsistent associations across exposure windows and vaccines. Results also
differed by sex, but without a consistent direction (i.e., stronger associations were sometimes
observed in women and sometimes men). Results were similarly inconsistent for antibodies to
Hepatitis A and B (not shown). Similar to the results in 13 year-olds in the other Faroe Islands
cohorts, this may indicate that by age 28, the effect of developmental exposure is less relevant.
Lastly, one low confidence study examined exposure to PFDA in adulthood and found inverse
associations with antibodies to both diphtheria and tetanus (statistically significant for diphtheria)
(Kielsen et al.. 2016).

It is plausible that the observed associations with PFDA exposure could be explained by
confounding across the PFAS. Exposure levels to other PFAS in the Faroe Islands populations were
considerably higher (PFOS 17 ng/mL, PFOA 4 ng/mL, PFNA1 ng/mL, PFDA 0.3 ng/mL at age 5
years in Grandiean etal. f20121. and there was a high correlation between PFDA and PFNA (r =
0.78) and moderate correlations with PFOS and PFOA (r = 0.39 and 0.35, respectively). The authors

This document is a draft for review purposes only and does not constitute Agency policy.

3-57 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

assessed the possibility of confounding in a follow-up paper fBudtz-Targensen and Grandiean.
2018a) that reanalyzed data from both Grandiean etal. f20121 and fGrandiean etal.. 2017bl for
benchmark analysis. In this re-analysis, estimates were adjusted for PFOS and PFOA. There were
variable attenuation of the observed effect estimates across the different analyses (though some of
the adjusted estimates were not estimable, likely due to collinearity), and PFNA was not adjusted
for in these models. However, associations with PFDA were stronger than for PFNA, and adjustment
by PFOS and PFOA did not eliminate the association, so confounding by co-occurring PFAS is
unlikely to fully explain the associations. Overall, while it is not possible to rule out confounding
across PFAS, the available evidence suggests that it is unlikely to explain the observed effects. Other
sources of potential confounding, including possible co-exposures such as PCBs, were controlled
appropriately.

Overall, in the two birth cohorts examining effect in children in the Faroe Islands, of the 26
paired antibody-to-PFDA exposure evaluations of diphtheria and tetanus antibody responses, 21 of
them support a decrease in antibodies with increasing PFDA concentration (see Table 3-12).
Although the results were not always statistically significant, the decreases were generally large,
with decreases in antibody concentration ranging from 2-25% per doubling of PFDA concentration.
The variability in some of the results could be related to differences in etiological relevance of
exposure measurement timing, differences in timing of the boosters since this was uncontrolled by
the study (children were vaccinated according to the official Danish/Faroese vaccination program),
as well as differences in timing of antibody measurements in relation to the last booster and PFDA
exposure measurement In addition, a cross-sectional study of children in Greenland reported a
large odds ratio for lack of protection against diphtheria following vaccination fTimmermann et al..
20211. and similar decreases in both diphtheria and tetanus antibodies were also observed in a
very small study in adults (n = 12) from Denmark based on a reduced change in antibodies after a
booster shot (Kielsen etal.. 2016). These associations were observed despite poor sensitivity
resulting from narrow exposure contrasts in all three studies, which increases confidence in the
association. There is some remaining uncertainty resulting from variability in the response by age
of exposure and outcome measures as well as vaccination (initial and boosters) in the Faroe Islands
childhood cohorts, and due to potential for confounding across PFAS. There is also uncertainty due
to inconsistent results in Timmermann etal. f20211 as well as a birth cohort with follow-up to
young adulthood in the Faroe Islands fShih etal.. 20211. However, the findings in children in the
Faroe Islands are based on both outcome measurement in childhood and prospective exposure
measurement, and the inconsistency may conceivably be attributed to these differences.

Table 3-12. Summary of PFDA exposure and selected data on antibody
response in humans

Reference,

PFDA exposure

Outcome measure timing

Diphtheria vaccine

Tetanus vaccine

timing and

(% Change in antibodies

confidence

with increase in PFDAb)

This document is a draft for review purposes only and does not constitute Agency policy.

3-58 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

concentration in
ng/mLa

(% Change in
antibodies with
increase in PFDAb)

Grandiean
etal. (2012),

Maternal; mean
(IQR):

Children (age 5),
prebooster

-21.7 (-35.7, -4.8)

-2.5 (-18.5,16.8)

Faroe
Islands,

0.3 (0.2-0.4)

Children (age 5),
postbooster

-18.8 (-30.5, -5.0)

-6.1 (-23.5, 15.3)

N = 380-

Children (age 7)

0.7 (-18.2, 24.0)

16.4 (-6.7, 45.2)

537,
medium

Children (age 5);
mean (IQR):

Children (age 5),
prebooster

-16.0 (-29.6, 0.3)

-13.6 (-26.3, 1.4)

Grandiean

0.3 (0.2-0.4)

Children (age 5),
postbooster

-8.7 (-20.6, 5.0)

-19.9 (-33.1, -3.9)

et al.

Children (age 7)

-14.4 (-28.4, 2.4)

-22.3 (-35.8, -5.8)

(2017a)

Faroe

Islands,

Children (age 13);
mean (IQR):
0.3 (0.2-0.4)

Children (age 13)

-3.7 (-22.0, 18.9)

18.7 (-11.8, 59.8)

1997-2000
cohort

Grandiean
et al.

At birth, not
reported

Children (age 5),
prebooster

-3.54 (-23.19,21.15)

-8.40 (-26.27, 13.79)

(2017b),
Faroe
Islands,
N = 349,

Infant (18 months);
median (25th—75th
percentile):
0.3 (0.2-0.4)

Children (age 5),
prebooster

2007-2009 cohort
25.52 (2.00, 54.48)

1997-2000 cohort
-22.87 (-60.92, 52.24)

2007-2009 cohort
-5.78 (-23.56, 16.13)

1997-2000 cohort
-14.47 (-56.88, 69.66)

medium

2007-2009
cohort
(unless
specified)

Children (age 5);
median (25th—75th
percentile):
0.3 (0.2-
0.5) ng/mL

Children (age 5),
prebooster

-8.99 (-23.63, 8.46)

-1.76 (-16.73, 15.91)

Shih et al.

(2021),

Faroe

Islands, N =

281,

medium

Cord blood;
median (IQR) 0.07
(0.06)

Adults (age 28)

Total: 7.29 (-11.2, 29.6)
Women:-1.39 (-24.8, 29.2)
Men: 16.16 (-10.6,51.0)

Total: -12.9 (-25.0, 1.2)
Women: -17.0 (-33.0,
3.0)

Men: -8.8 (-26.0,12.4)

Children (age 7);
0.22 (0.16)

Total: 37.89 (1.8, 86.8)

Women: 30.99 (-16.5,105.4)
Men: 43.8 (-4.4, 116.3)

Total: 3.2 (-18.5, 30.7)
Women: -2.6 (-31.3,
38.0)

Men: 8.3 (-21.1, 48.7)

Children (age 14);
0.28 (0.17)

Total: -7.2 (-35.1, 32.7)
Women: 39.4 (-28.7,172.8)
Men: -20.5 (-47.4, 20.3)

Total: -22.6 (-42.9, 4.9)
Women: -41.0 (-66.6,
4.1)

Men:-14.3 (-39.7,21.9)

Adults (age 22);
0.39 (0.26)

Total: 34.9 (4.9, 73.6)*
Women: 39.0 (2.2, 89.0)*

Men: 27.2 (-17.6, 96.3)

Total: -5.2 (-22.9, 16.4)
Women: -4.0 (-25.4,
23.5)

Men: -7.6 (-35.1, 31.6)

Adults (age 28);
0.34 (0.25)

Total: 19.6 (-1.2, 44.9)
Women: 24.7 (-2.9, 60.0)
Men: 12.8 (-16.2, 51.8)

Total: -5.3 (-18.7, 10.3)
Women: -1.9 (-19.6,
19.8)

Men: -9.9 (-28.8, 14.2)

Timmerman
n et al.
(2021),

Children (age 7-
12)

Children (age 7-12)

Adjusted for time since

vaccine booster,
breastfeeding duration
126(32, 289)

Adjusted for time since

vaccine booster,
breastfeeding duration
74 (12,169)

This document is a draft for review purposes only and does not constitute Agency policy.

3-59 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Greenland,
N = 314,
medium

Additionally adjusted for
area of residence
-39 (-70, 27)

Additionally adjusted for
area of residence
-29 (-61, 28)

Maternal (N = 57)

Children (age 7-12)

-39 (-84, 133)

95 (-45, 591)

Kielsen et al.
(2016),
Denmark
N = 12, low

Adult (10 days post
vaccination);
median (IQR):
0.3 (0.2-
0.3) ng/mL

Adult-change from
4 days to 10 days post
vaccination

-18.18 (-29.52, -5.00)

-8.31 (-18.10, 2.66)

aExposure timing is organized into groups based on maternal exposure, childhood exposure (including from birth
through age 13), and adult exposure.

bLinear regression (P or % change in antibody per 2-fold increase of PFDA). Numbers in parentheses are 95%
confidence intervals.

Bold font indicates p < 0.05.

Infectious disease

Direct measures of infectious disease incidence or severity such as respiratory tract
infections, pneumonia or otitis media are useful for evaluating potential immunotoxicity in humans.
Increases in incidence or severity of infectious disease can be a direct consequence of impaired
immune function whether the specific functional deficit has been identified or not. Given the clear
adversity of most infectious diseases, they are generally considered good measures for how
immunosuppression can affect individuals and communities. Physician diagnosis is the most
specific way to assess infectious diseases, but these are usually only available for severe diseases
and are less likely for diseases where treatment is not sought. Self-reported incidence or severity of
disease may be less reliable but may be the only way to assess diseases such as the common cold or
gastroenteritis which while less adverse, are more common and can thus provide information about
immunosuppression and susceptibility to more severe infections. In general, symptoms of infection
alone are not considered reliable measures of disease because of their lack of specificity. Antibody
levels in response to infection are also included in this section (differentiated from antibody levels
in response to vaccination, described above); the utility of these measures depends on the study
design and population due to various factors such as potential confounding and prevalence of
infection.

Six studies examined PFDA exposure and infectious disease outcomes in children and one
study examined disease severity in adults (see Figure 3-11). Three of these focused on the number
of episodes of infectious disease. One was a medium confidence prospective birth cohort study in
Japan which looked at the association of PFDA exposure with total infectious disease (including
otitis media, pneumonia, RS virus, and varicella) from birth to age 4 (Goudarzi etal.. 2017). with
outcomes ascertained using a questionnaire identifying physician diagnosed disease incidents. A
second medium confidence birth cohort in China identified cases of common cold or
bronchitis/pneumonia reported by parents with verification with medical records fWang etal..
20221. A low confidence cohort with PFDA exposure measured in childhood examined number of
episodes of parent-reported lower respiratory tract infections and common colds based on parent

This document is a draft for review purposes only and does not constitute Agency policy.

3-60 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

reports using an unvalidated questionnaire fKvalem etal.. 20201. Another prospective birth cohort
in examined days of infectious disease symptoms (fever, diarrhea, coughing, nasal discharge,
vomiting) with follow-up at 1-4 years fDalsager etal.. 20161. This study was considered low
confidence due to the non-specific nature of the symptoms reported, which may not represent
infectious disease. In the same birth cohort in Denmark, but with a larger sample size,
hospitalizations due to infectious disease were identified from a national registry fDalsager etal..
2021a). These two studies were evaluated separately due to their different samples and outcomes
measurement methods but should not be considered fully independent samples. Also in children,
one study examined antibody response to hand, foot, and mouth disease (HFMD) infection. This
birth cohort in China fZeng etal.. 2019bl measured antibody levels in infants at birth and 3 months
of age, a time-period expected to reflect passive immunity from maternal antibodies. This study is
low confidence because the outcome is broad and difficult to interpret and there are concerns for
confounding by timing of HFMD infection as well as other limitations. Lastly, one study examined
severity of COVID-19 illness in Denmark using biobank samples and national registry data
fGrandiean etal.. 20201. There was concern for selection bias in this study due to the expectation
that biobank samples were more likely to be available for individuals with chronic health concerns.
In addition, severity of COVID-19 is not a direct measure of immune suppression as other factors
may contribute to illness severity.

The results for this set of studies are summarized in Table 3-13. Results were overall
inconsistent. Positive associations (though mostly not statistically significant) between PFDA
exposure and specific infectious diseases were observed in some studies (diarrhea, common cold,
and lower respiratory infection in Wang etal. f20221. lower respiratory infections in Kvalem et al.
f20201. upper respiratory tract infections in Dalsager etal. f2021al. fever in Dalsager etal. f201611.
but inverse associations were observed in other studies. Where two studies were available for a
given infectious disease, the results were generally not in the same direction. The single study of
HFMD antibodies reported lower levels of protective antibody concentrations with higher PFDA
exposure and higher odds of having antibody levels below a clinically protective level (Zengetal..
2019b). Exposure contrast was limited across studies which makes it difficult to interpret the null
findings. Associations were slightly stronger in Wang etal. f20221. the only medium confidence
study with adequate sensitivity (due to slightly higher exposure levels and contrast), but this likely
does not fully explain the inconsistency in direction of association across studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-61 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Participant selection -
Exposure measurement -
Outcome ascertainment -
Confounding -
Analysis -
Sensitivity
Selective Reporting -
Overall confidence

++ ++ ++ ++

Legend

| Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)

Figure 3-11. Summary of evaluation of epidemiology studies of PFDA and
infectious disease. Refer to HAWC for details on the study evaluation review:

HAWC Human Infectious Disease Effects.

Table 3-13. Studies on PFDA and infectious disease in humans

Disease

Reference, confidence

Exposure
measurement

timing and
concentration

Disease
assessment
timing

PFDA Results

Total
infectious
disease3

Goudarzi et al. (2017)
medium

Maternal; median
(IQR): 0.3 (0.2-0.4)
ng/mL

From birth

to age 4

OR (95% CI):
Ql: Ref
Q2 1.00 (0.73, 1.35)
Q3 0.89 (0.66, 1.21)
Q4 0.80 (0.59, 1.08)

Dalsager et ai. (2021a),
medium

Maternal; median:
0.3

From birth
to age 4

HR (95% CI)
1.06 (0.93, 1.22)

This document is a draft for review purposes only and does not constitute Agency policy,

3-62 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Disease

Reference, confidence

Exposure
measurement

timing and
concentration

Disease
assessment
timing

PFDA Results

Lower
respiratory

tract
infection15

Wang et al. (2022),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

Age 1

OR (95% CI) for event
during first year of life per
loglO increase:
1.84 (0.36, 9.49)
IRR (95% CI) for count of
events per loglO increase:
0.85 (0.26, 2.79)

Dalsager et al. (2021a),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

From birth
to age 4

HR (95% CI)
1.06 (0.85, 1.32)

Kvalem et al. (2020)
medium

Child age 10;
median (IQR): 1.3
(0.9)

Age 10-16

RR (95% CI) per IQR
increase
1.09 (0.86,1.39)

Age 16 (last
12 months)

1.34 (0.84, 2.14)

Diarrhea

Dalsager et al. (2016) low

Maternal; median
(range): 0.3 (0.02-
1.0) ng/mL

Age 1-3

OR (95% CI) for proportion
of days with symptoms
Low exposure: Ref
Medium: 0.91 (0.53,1.56)
High: 0.91 (0.52,1.57)

Wang et al. (2022),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

Age 1

OR (95% CI) for event
during first year of life per
loglO increase:
3.36 (0.90, 12.63)
IRR (95% CI) for count of
events per loglO increase:
2.16(1.23, 3.79)*

This document is a draft for review purposes only and does not constitute Agency policy.

3-63 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Disease

Reference, confidence

Exposure
measurement

timing and
concentration

Disease
assessment
timing

PFDA Results

Dalsager et al. (2021a),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

From birth
to age 4

HR (95% CI) for Gl
0.81 (0.46, 1.43)

Common
cold
(No.
episodes/
frequency)

Wang et al. (2022),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

Age 1

OR (95% CI) for event
during first year of life per
loglO increase:
1.66 (0.48, 5.75)
IRR (95% CI) for count of
events per loglO increase:
1.05 (0.65,1.68)

Dalsager et al. (2021a),
medium

Maternal; median
(IQR): 0.6 (0.4-0.8)

From birth
to age 4

HR (95%) for upper
respiratory tract infection
1.16 (0.95,1.42)

Kvalem et al. (2020),
medium

Child age 10;
median (IQR): 1.3
(0.9)

Age 10-16

OR (95% CI) per IQR

increase:
Reference 1-2 colds
3-5 colds: 1.69 (0.46, 6.18)
>5: 1.36 (0.39, 4.80)

Age 16 (last
12 months)

Reference 0 colds
1-2 colds: 0.78 (0.55, 1.09)
>3: 0.56 (0.37, 0.84)*

Cough

Dalsager et al. (2016) low

Maternal; median
(range): 0.3 (0.02-
1.0) ng/mL

Age 1-3

OR (95% CI) for proportion
of days with symptoms
Low exposure: Ref
Medium: 0.63 (0.37,1.07)
High: 0.85 (0.50,1.46)

This document is a draft for review purposes only and does not constitute Agency policy.

3-64 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Disease

Reference, confidence

Exposure
measurement

timing and
concentration

Disease
assessment
timing

PFDA Results

Fever

Dalsager et al. (2016) low

Maternal; median
(range): 0.3 (0.02-
1.0) ng/mL

Age 1-3

OR (95% CI) for proportion
of days with symptoms
Low exposure: Ref
Medium: 1.07 (0.63,1.81)
High: 1.45 (0.85, 2.49)

Hand Foot
and Mouth
Disease Virus
Antibodies

Zeng et al. (2019b), low

Cord; median (IQR):
0.1 (0.01-0.2)

Birth and
Age 3 mo

OR (95% CI) for HFMD
antibody concentration
below clinically protective
level
Cord blood:
1.19 (0.82,1.71)
3 mo: 2.22 (1.42, 3.47)*

COVID-19
severity

Grandiean et al. (2020),
medium

Biobank prior to
illness; median

(IQR):
0.1(0.1-0.2)

Adulthood

OR (95% CI) for 1 unit
increase
Increased severity based on
hospitalization, admission
to intensive care and/or
death
0.53 (0.10, 2.84)

Bolded values are statistically significant. *p < 0.05.
includes Otitis media, pneumonia, RS virus, Varicella.

bLower respiratory tract infections include bronchitis, bronchiolitis, and pneumonia.

Sensitization and allergic response

Another major category of immune response is the evaluation of sensitization-related- or
allergic responses that are a result of aggravated immune reactions (e.g., allergies or allergic
asthma) to foreign agents (IPCS. 20121. A chemical may be either a direct sensitizer (i.e., promote a
specific IgE-mediated immune response to the chemical itself] or may promote or exacerbate a
hypersensitivity-related outcome without evoking a direct response. 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 measurements of health effects such as allergies or asthma, and skin prick test responses.
Observational tests such as measures of total IgE levels measure indicators of sensitivity and
allergic responses but are not a direct measurement of the response. The section is organized by
the different types of measurements, starting with functional measures as the most informative.

This document is a draft for review purposes only and does not constitute Agency policy.

3-65 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Seven cohorts (10 publications) examined hypersensitivity outcomes in children. Study
evaluations are summarized in Figure 3-12 and Table 3-14, Study sensitivity was a concern across
most of the studies, due to narrow exposure contrasts which makes interpretation of the null
findings difficult.

Participant selection -
Exposure measurement -
Outcome ascertainment -
Confounding
Analysis
Sensitivity
Selective Reporting
Overall confidence -

o?>

_l L_

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 3-12. Summary of evaluation of epidemiology studies of PFDA and
sensitization or allergic response. Refer to HAWC for details on the study
evaluation review: HAWC Human Hypersensitivity Effects.

Functional immune measures of sensitization or allergic response

Asthma

Six studies (eight publications) evaluated any asthma-related outcome in relation to PFDA
exposure. One case-control study in Taiwan examined asthma incidence (i.e., physician diagnosis
within the past year, identified from two hospitals), which is the most specific measure but may
result in under-ascertainment; this study was considered medium confidence (Zhou etal.. 2017b:
Zhu et ah, 2016: Dong etal.. 2013). Most available studies examined asthma prevalence (ever
diagnosed asthma) and were also considered medium confidence including four birth cohorts with
prenatal or cord PFDA blood measurements fBeck et al.. 2019: Zeng etal.. 2019a: Timmermann et
al.. 2017: Smitetal.. 20151 and one study with PFDA exposure measured in childhood fKvalem et
al.. 20201

Positive associations with asthma were observed in Dong etal. f20131 and '."immermann et
al. (2017) (see Table 3-15), including an exposure-response gradient observed in Dong etal.
f20131. However, in Timmermann etal. (2017). the association was observed only in a small

This document is a draft for review purposes only and does not constitute Agency policy.

3-66 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

number of subjects (4%, n = 22) that did not receive an MMR vaccine; the effects were statistically
significant when both the outcome and PFDA exposure were evaluated when the children were
5 years of age. There remained an increased risk for asthma diagnosis when these same children
were 13 years old. No association with childhood exposure was observed in the rest of the study
population (that received MMR vaccine), but a positive association was suggested (p >0.05) when
using maternal PFDA concentrations as an indication of prenatal exposure (Timmermann et al..
2017). The Taiwan case-control study used the child's current PFDA concentrations and observed
increased odds ratios in the highest quartile compared to the lowest quartile (concentrations not
reported for the quartiles) and in boys and girls with low or high testosterone or high estradiol as
well as in boys with low estradiol, indicating there was a modifying effect of sex hormones fZhou et
al.. 2017b: Zhu etal.. 2016: Dong etal.. 20131. Associations were stronger in boys than girls. Dong
etal. (2013) also observed a significant increase in asthma severity scores based on a 13-item
questionnaire assessing frequency, use of medicine, and hospitalizations in the highest quartile
with a significant increasing trend, but there was no difference in the asthma control test (five-item
questionnaire assessing control of asthma symptoms). The other four studies study fKvalem et al..
2020: Beck etal.. 2019: Zeng etal.. 2019a: Smit etal.. 20151 reported no increase in asthma with
PFDA exposure. The inconsistency may be accounted for at least in part by study sensitivity, as the
Taiwan study with a clear association (Dong etal.. 2013) had the highest PFDA exposure levels and
was based on asthma incidence within the past year, a more specific definition, less likely to suffer
from outcome misclassification, than whether the child ever had asthma ("ever asthma"). Still,
overall, there is considerable uncertainty due to the lack of association with asthma in most studies.

Dermal allergic measures - eczema

Four medium confidence birth cohorts from different geographic locations in five
publications (Chen etal.. 2018a: T immermann et al.. 2 017: Goudarzi etal.. 2016: Smit etal.. 2015:
Okada etal.. 2014) and one study with exposure measured in childhood (Kvalem etal.. 2020)
evaluated dermal allergic measures. While the studies used different terminology including
eczema, atopic eczema, and atopic dermatitis, all assessed presence of an itchy rash that was
coming and going for at least 6 months using the International Study of Asthma and Allergies in
Childhood questionnaire with the exception of Kvalem etal. f20201 which used a different
questionnaire. The dermal response conditions can represent hypersensitivity to antigen exposure
by way of any exposure route. None of the studies found a significant association between PFDA
exposure (either prior exposure, based on maternal or the child's earlier PFDA measurement, or
current exposure) and dermal allergic effects (see Table 3-15). However, a non-statistically
significant positive association for eczema was observed in Chen etal. (2018a) and for the children
without MMR vaccine in Timmermann etal. f20171. An inverse association (p > 0.05) was observed
in multiple studies fKvalem et al.. 2 0 2 0: T immermann et al.. 2 017: Smitetal.. 2015: Okada etal..

This document is a draft for review purposes only and does not constitute Agency policy.

3-67 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

20141 (in children with MMR vaccine for Mamsen etal. f201711. This inconsistency is not clearly
explained by study confidence or other factors.

Allergic sensitization/Skin prick test

Two medium confidence studies conducted skin prick tests. In Timmermann etal. (20171.
they examined five common allergens (birch/grass pollen, dog/cat dander, and house dust mites)
in 13-year old children from the Faroe Islands. A positive result was noted if the subjects
developed a wheal >3 mm in diameter. In Kvalem etal. f20201. a positive result was noted if there
was at least one positive test >3 mm at 10 and 16 years but the allergens tested were not described.
The relative risk of a positive test was slightly higher (p > 0.05) with PFDA exposure in Kvalem et
al. f20201 but there was no increase in the odds of having a positive test related to PFDA exposure
regardless of when the PFDA was evaluated (i.e., maternal, child at 5 years of age, or current
measurement at 13 years of age) in In Timmermann etal. (20171. Both studies had similar
exposure contrast

Observational immune measures of sensitization or allergic response

Two studies also analyzed observational measures including total IgE, eosinophil counts, or
eosinophil cationic protein fT immermann et al.. 2 017: Zhu etal.. 2016: Dong etal.. 20131: of these,
IgE measures are considered the most informative. Dong etal. (20131 observed a statistically
significant increase in total IgE, eosinophilic cationic protein concentration, and absolute
eosinophilic count with increasing current child PFDA concentrations in asthmatics, as well as
increased eosinophilic cationic protein concentrations in non-asthmatics in a population in Taiwan.
In the same medium confidence study, Zhu etal. f 20161 evaluated this further and found that the
positive association with IgE was observed in boys and girls with asthma, but only statistically
significant in boys. Zhu etal. f20161 expanded the evaluation to additional cytokines (IFN-y, IL-2,
IL-4, and IL-5) in subjects with and without asthma. While there were occasional statistically
significant decreases in the lower quartiles compared to quartile 1 (i.e., IL-2 in males and IFN-y in
females) there was no consistency or trend. In the second medium confidence study, Timmermann
etal. f20171 did not find any significant association between IgE levels in cord blood or blood
samples from children at age 7 and PFDA concentrations (either maternal concentrations or child's
concentration at age 5) in children from the Faroe Islands.

This document is a draft for review purposes only and does not constitute Agency policy.

3-68 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-14. Studies on PFDA and hypersensitivity-related outcomes in
humans

Reference

Study design
(location/study)

Exposure
measure timing

Disease
assessment
timing

Hypersensitivity
outcomes assessed

Study
confidence

Maternal exposure

Beck etal. (2019)

Prospective
(Denmark birth
cohort)

981

Maternal

Age 5

Asthma (ever)

Medium

Chen et al. (2018a)
Zeng et al. (2019a)

Prospective
(Shanghai Birth
Cohort)

687

Cord blood (log
transformed)

Age 2

Eczema

Medium

358

Age 5

Asthma (ever)

Goudarzi et al.
(2016)

Okada et al. (2014)

Prospective
(Japan/Hokkaido
Study of

Environment and
Children's Health
cohort 2003-
2013)

1,558

Maternal
(quartiles)

Age 4

Total allergic disease,
wheeze, eczema,
rhinoconjunctivitis
symptoms

Mediuma

2,062

Maternal
(quartiles)

From birth to
age 2

Wheeze, allergic
rhinoconjunctivitis
symptoms, eczema,
total allergic diseases

Smit et al. (2015)

Prospective
(Greenland,
Ukraine/
INUENDO birth
cohort)

1,024

Maternal (log
transformed)

Children
age 5-9

Asthma (ever), eczema,
wheeze

Medium

Timmermann et al.
(2017)

Prospective (Faroe
Island cohort;
1997-2000)

559

Maternal; child
age 5,13 (log
transformed)

Age 5, 7,13

Total IgE, asthma
(ever), allergies, allergic
rhinoconjunctivitis,
eczema, skin prick test

Medium
(low for
asthma)

Child exposure

Zhou et al. (2017b):
Zhu etal. (2016):
Dong et al. (2013)

Case-control
(Taiwan/ Genetic
and Biomarker
study for
Childhood
Asthma)

asthma
(231)
non
(225)

Child: current
(quartiles)

Children
age 10-15

Asthma incidence and
control, total IgE,
eosinophil count,
eosinophil cationic
protein

Medium

Kvalem et al. (2020)

Prospective
(Norway
Environment and
Child Asthma)

378

Child: 10 years

Age 10 and
16

Asthma (ever/current),
Eczema, skin prick test

Medium

aMedium vs. high confidence based primarily on sensitivity.

This document is a draft for review purposes only and does not constitute Agency policy.

3-69 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-15. Summary of PFDA and selected data on hypersensitivity in
humans

Reference

Exposure timing and
concentration3

Hypersensitivity
measurement
timing

PFDAb OR (95% CI) or as specified

Asthma

Smit et al. (2015)

Maternal, mean gest wk 24 or
25; geometric mean (5—95th
percentile): Ukraine 0.16 (0.07-
0.35) ng/mL, Greenland 0.42
(0.16-1.16) ng/mL

Child
(age 5-9)

Ever asthma

Ukraine: 0.80 (0.37,1.75) per 1 SD change
Greenland: 0.93 (0.73,1.19) per 1 SD change
Combined: 0.92 (0.73,1.16) per 1 SD change

Kvalem et al. (2020)

Child (age 10); median (IQR):
0.2 (0.1) ng/mL

Child (age 10)

Ever asthma

RR: 0.95 (0.78, 1.15)

Child (age 10-16)

Asthma between 10 and 16 years
RR: 0.89 (0.67-1.16)

Child (age 16)

Current asthma (last 12 months)
RR: 0.93 (0.71, 1.22)

Timmermann et al.
(2017)

Maternal, gest wk 34-36;
median (IQR): 0.3 (0.2-0.4
ng/mL)

Child (age 5)

Ever asthma
1.09 (0.72,1.65)

Child (age 13)

1.26 (0.83,1.92)

Child (age 5); median (IQR): 0.3
(0.2-0.4 ng/mL)

Child (age 5)

Ever asthma

No MMR: 4.04(1.05,15.50)c

Yes MMR: 0.71 (0.48,1.06), Interaction p = 0.02

Child (age 13)

No MMR: 2.87(0.84, 9.79)

Yes MMR: 0.71 (0.48,1.06), Interaction p = 0.03

Child (age 13) median (IQR): 0.3
(0.2-0.4 ng/mL)

Child (age 13)

Ever asthma
0.84 (0.55,1.29)

Beck et al. (2019)

Maternal, gest week 8-16;
median (IQR): 0.3 (0.2-0.4)
ng/mL

Child (age 5)

Ever doctor-diagnosed asthma
0.9 (0.60,1.44)

Ever self-reported asthma (>episodes of wheezing
lasting more than a day in past 12 months)
1.44 (0.87, 2.41)

Zeng et al. (2019a)

Cord blood median (IQR): 0.4
(0.2-0.5)

Child (age 5)

Ever asthma
0.63 (0.23, 1.72)

Girls: 0.21 (0.03, 1.47)
Boys: 1.09 (0.26, 4.50)

Dong et al. (2013)

Children, current; range: <0.1-
5.0 ng/mL

Child

(age 10-15)

Asthma incidence

Q2: 1.02 (0.58,1.80)

Q3: 1.30 (0.72, 2.33)

Q4: 3.22 (1.75, 5.94), p-trend < 0.001

This document is a draft for review purposes only and does not constitute Agency policy.

3-70 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Reference

Exposure timing and
concentration3

Hypersensitivity
measurement
timing

PFDAb OR (95% CI) or as specified

Zhou et al. (2017b)

Children, current; median
(IQR): 1.1 (0.9-1.5) ng/mL with
asthma, 1.0 (0.8-1.2) ng/nL
without asthma

Child

(age 10-15)

Asthma incidence
Low Testosterone:

M: 1.71 (0.75, 3.90); F: 1.24 (0.60, 2.56)

High Testosterone:

M: 3.16 (1.21, 8.25); F: 1.37 (0.63, 3.02)
Low Estradiol:

M: 1.21 (0.60, 2.46); F: 0.76 (0.27, 2.20)

High Estradiol:

M: 4.01 (1.46,11.06); F: 1.78 (0.94, 3.35)
No significant interaction with sex hormone category

Zhu et al. (2016)

Children, current

Child

(age 10-15)

Asthma incidence
Q4 vs, Q1

M: 3.45 (1.51, 7.88); p-trend = 0.003
F: 2.86 (1.16, 7.01); p-trend = 0.02

Allergic sensitization (positive skin prick test)

Kvalem et al. (2020)

Child (age 10); median (IQR):

Child (age 10)

RR: 1.15 (0.99, 1.35)

0.2 (0.1) ng/mL

Child (age 16)

RR: 1.12 (0.87, 1.45)

Timmermann et al.
(2017)

Maternal, gest wk 34-36;
median (IQR): 0.3 (0.2-0.4
ng/mL)

Child
(age 13)

1.02 (0.74,1.41)

Child (age 5); median (IQR): 0.3
(0.2-0.4 ng/mL)

Child
(age 13)

0.79 (0.59,1.05)

Child (age 13) median (IQR): 0.3
(0.2-0.4 ng/mL)

Child
(age 13)

0.81 (0.59,1.13)

Eczema

Kvalem et al. (2020)

Child (age 10); median (IQR):
0.2 (0.1) ng/mL

Child (age 10)

Ever doctor diagnosed:
RR: 0.93 (0.79, 1.09)

Child (age 10-16)

Ever between 10 and 16 years
RR: 0.86 (0.66, 1.12)

Child (age 16)

Current (last 12 months)
RR: 0.92 (0.68, 1.25)

Chen et al. (2018a)

Cord blood; median (range):
0.36 (
-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Reference

Exposure timing and
concentration3

Hypersensitivity
measurement
timing

PFDAb OR (95% CI) or as specified

Smit et al. (2015)

Maternal, mean gest wk 24 or
25; geometric mean (5—95th
percentile):

Ukraine 0.16 (0.07-0.35)
ng/mL,

Greenland 0.42 (0.16-1.16)
ng/mL

Child
(age 5-9)

Current:

0.95 (0.75,1.20) per 1 SD change
Ever:

0.88 (0.73,1.06) per 1 SD change

Timmermann et al.
(2017)

Maternal, gest wk 34-36;

median (IQR):

0.3 (0.2-0.4 ng/mL)

Child (age 13)

Ever: 0.92 (0.64,1.32)

Child (age 5); median (IQR): 0.3
(0.2-0.4 ng/mL)

Child (age 13)

Ever: 0.92 (0.64,1.31)

Child (age 13) median (IQR): 0.3
(0.2-0.4 ng/mL)

Child (age 13)

No MMR: 401.88 (0.09,1.84 x 106)c

Yes MMR: 0.88 (0.58,1.34), p-interaction = 0.2

aExposure timing is organized into groups based on maternal exposure (including cord blood), childhood exposure
(including from birth through age 13), and adult exposure.

bAII estimates are presented as OR (95% CI) for the odds of the outcome per 2-fold increase in PFDA concentration
unless otherwise stated.

cResults provided broken down by MMR vaccination status; yes (n = 537) or no (n = 22) when provided; some
results were not split by MMR vaccination status
Bold font indicates p < 0.05.

Animal studies

Animal toxicity studies examining effects on the immune system after PFDA exposure
include two 28-day gavage studies using S-D rats and/or B6C3F1/N mice (Trawlev etal.. 2018:
NTP. 20181 and a 14-day study in Balb/c mice (inferred as a gavage study based on information
provided, although the method of chemical administration was not specified) fLee and Kim. 20181.
Immune effects reported in these studies are discussed according to the immunotoxicity categories
and endpoint groupings outlined previously (see Section on Methodological considerations above
for more details). Most of the available evidence, including host resistance, immune function, and
observational assays, were conducted in female mice and rats, since female animals are preferred in
immunotoxicity testing due to increased sensitivity (Kadel and Kovats. 2018: Klein and Flanagan.
20161. Further, no chemical-specific information on potential sex-specific differences was
identified.

Immunosuppression
Host resistance

Host resistance assays measure the effects of toxicants on the overall immune function in
response to a challenge, usually from an infectious agent, and these assays are considered highly
relevant to the evaluation of immunotoxicity in the context of human health assessment (IPCS.
20121. Host resistance was evaluated in a 28-day gavage study in female B6C3F1/N mice

This document is a draft for review purposes only and does not constitute Agency policy.

3-72 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

considered medium confidence primarily due to the lack of reporting on the blinding of
investigators during assessment which raises some concerns for potential observational bias
fFrawlev etal.. 20181 (see Figure 3-13}. PFDA did not affect survival of mice challenged with three
dilution levels of Influenza virus (groups A-C) during the observational period after exposures
ended (Days 29-50 of the study); exposures ranged from 0.179-0.71 mg/kg-day (Refer to the
interactive HAWC link for additional details). The only effect noted was a slight decrease (7.8%) in
body weight at the highest exposure dose (0.71 mg/kg-day) on Day 29 in group C, the group
challenged with the highest level of influenza. In summary, host resistance appeared to be
unaffected by PFDA, although the evidence is limited to a single short-term study in mice.

Reporting quality -
Allocation -
Observational bias/blinding -
Confounding/variable control -I
Selective reporting and attrition -|
Chemical administration and characterization -
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity -
Results presentation -|
Overall confidence -

I 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)
Not reported

Figure 3-13. Evaluation results for animal study assessing effects of PFDA
exposure on host resistance. Refer to HAWC for details on the study
evaluation review.

Immune function assays

Markers of altered immune cell function or damage were evaluated in female B6C3F1/N
mice and female S-D rats exposed to doses of 0.045-0.71 and 0.125-0.5 mg/kg-day, respectively,
for 28 days via gavage fFrawlev etal.. 2018). Immune function assays included measures of:
(1) innate immunity such as mononuclear phagocyte system (MPS) activity in rats and natural killer
(NK) cell activity in rats and mice; (2) humoral-mediated immunity such as T-dependent antibody

This document is a draft for review purposes only and does not constitute Agency policy,

3-73 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

responses in rats and mice; and (3) cell-mediated immunity such as mixed leukocyte response in
mice and delayed-type hypersensitivity in rats and mice. These assays measure specific immune
system responses to a stimulus both at the cellular and organism level and can provide clear and
direct evidence of immunotoxicity flPCS. 20121. Overall, study confidence in experiments
conducted in both species was high for most endpoints, except delayed-type hypersensitivity
(DTH). The absence of information on the blinding or any other strategy used to mitigate potential
for observational bias resulted in a medium confidence rating for this endpoint (see Figure 3-14).

T%1

Reporting quality -
Allocation -
Observational bias/blinding -I
Confounding/variable control -
Selective reporting and attrition - +

Chemical administration and characterization - +

Exposure timing, frequency and duration -)
Endpoint sensitivity and specificity -
Results presentation -|
Overall confidence -

Legend

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)
NRi Not reported
~ Multiple judgments exist

Figure 3-14. Evaluation results for animal studies assessing effects of PFDA
exposure on immune function assays. Refer to HAWC for details on the study
evaluation review.

Dose-related decreases in specific activity of the MPS (cpm/mg of tissue) were reported in
rat liver (MPS was not examined in mice) at 0.125-0.5 mg/kg-day (15-45% compared to controls),
reaching statistical significance at the two highest doses (see Figure 3-15). Alterations in
phagocytic activity coincide with the liver histopathology (i.e., hepatocyte necrosis) and increased
liver weight (see Section 3.2.1 on Liver effects for more details) observed in the exposed animals.
Due to the increases in liver weight, it is possible that the effects on specific activity could represent
changes in hepatocyte numbers/size rather than alterations in the functional activity of tissue
macrophages fFrawlev etal.. 20181. However, a decreasing trend was also observed for total MPS

This document is a draft for review purposes only and does not constitute Agency policy,

3-74 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

activity (p = 0.051) and percent (%) uptake of sheep red blood cell (SRBC) by macrophages in the
liver (p = 0.029).

MPS activity was evaluated in other rat tissues such as the thymus, lung, kidney, and spleen.
In the thymus, MPS activity (total, specific and % SRBC uptake) was significantly increased at the
highest exposure dose (139-200% at 0.5 mg/kg-day) (Frawlev etal.. 2018) (see Figure 3-12).
However, the values for total activity and % uptake were two orders of magnitude lower than the
negative control tissue (kidney), which raises concerns about the biological significance of these
results. No treatment-related effects were found in MPS activity in the lung and spleen of rats (see
Figure 3-15).

Apart from the reduced MPS activity in rat liver after PFDA exposure, no treatment-related
effects were observed in other immune function assays evaluated in rats and mice (i.e., NK cell
activity and T-dependent antibody responses to SRBC in the spleen of rats and mice, mixed
leukocyte response in mouse spleen and DTH response to C. albicans in rats and mice). Despite a
general lack of findings from most immune function assays, the mild reductions in phagocytic
activity in rat liver suggest potential suppression of innate immunity after short-term PFDA
exposure, although uncertainties remain surrounding whether this finding might be attributable to
the observed liver toxicity.

This document is a draft for review purposes only and does not constitute Agency policy.

3-75 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study Design Animal Description Outcome Confidence Organ Endpoint Name Trend Test Result PFDA Immune Function Assays

Frawley, 2018, 4287119 28 Day Oral Rat, Sprague-Dawley (Harlan) ( ) High confidence Liver Mononuclear Phagocytic System, Specific Activity significant

•—• W- ~

Mononuclear Phagocytic System, Total Activity not significant

£ No significant change
^ Statistically significant increase
^7 Statistically significant decrease

Percent SRBC Uptake significant
Thymus Mononuclear Phagocytic System, Specific Activity significant

•—•—• •

•—•—• ~

Mononuclear Phagocytic System, Total Activity significant

•—•—• ~

Percent SRBC Uptake significant
Lung Mononuclear Phagocytic System, Specific Activity not significant

•—•—• ~

•—•—• •

Mononuclear Phagocytic System, Total Activity not significant
Percent SRBC Uptake not significant

•—•—• •

Kidney Mononuclear Phagocytic System, Specific Activity significant

•—•—• •

Mononuclear Phagocytic System, Total Activity significant
Percent SRBC Uptake significant
Spleen Mononuclear Phagocytic System, Specific Activity not significant

•—•—• •

Mononuclear Phagocytic System, Total Activity not significant
Percent SRBC Uptake not significant

•—•—• •

Natural Killer Cell Activity not significant
T-Dependent Antibody Response to SRBC not significant
Medium confidence N/A Delayed-Type Hypersensitivity to C. Albicans not significant

•—•—• •

Mouse, B6C3F1/N High confidence Spleen Mixed Leukocyte Response not significant

Natural Killer Ceil Activity not significant

—• • •

•

T-Dependent Antibody Response to SRBC not significant
Medium confidence N/A Delayed-Type Hypersensitivity to C. Albicans not significant

—• • •

—• —• •

1 1 1 1 1 1 1 1

-0.1 0 0.1 0.2 0.3 0.4 0.5 06 0.7 0.8
Dose (mg/kg-day)

Figure 3-15. Effects on immune function assays following exposure to PFDA in short-term oral studies in animals

(results can be viewed by clicking the HAWC link).

This document is a draft for review purposes only and does not constitute Agency policy,

3-76 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

General/observational immune assays

General or observational immune parameters were evaluated in two experiments (reported
in one study) using female B6C3F1/N mice and female S-D rats after 28-day gavage exposure
fFrawlev etal., 20181. The 28-day experiments were high confidence for most end points, except
bone marrow colony formation (see Figure 3-16). Key issues regarding observational bias/blinding
and results presentation (i.e., ambiguity surrounding sample size) reduced confidence to medium
for this endpoint. The assays included in the study are spleen cell immunophenotyping (rats and
mice), anti-CD3+-mediated T-cell proliferation (rats and mice), bone marrow DNA synthesis (rats
and mice) and bone marrow colony formation and differentials (rats only) fFrawlev etal.. 20181.
These assays can indicate changes in immune cell populations and mediators and are often used in
support of more predictive measures of immunotoxicity (i.e., host resistance and functional assays)
flPCS. 20121.

Reporting quality -
Allocation -I
Observational bias/blinding -j
Confounding/variable control -
Selective reporting and attrition -j
Chemical administration and characterization -
Exposure timing, frequency and duration -j
Endpoint sensitivity and specificity -
Results presentation -
Overall confidence J

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)
NR Not reported
* Multiple judgments exist

Figure 3-16. Evaluation results for animal studies assessing effects of PFDA
exposure on general/observational immune assays. Refer to HAWC for details
on the study evaluation review.

PFDA treatment caused dose-related reductions in absolute spleen cell numbers in mice
reaching up to 24% decrease compared to controls at the highest dose (0.71 mg/kg-day) (see
Table 3-16 and Figure 3-17). Likewise, absolute counts of splenic B-cells, T-cells, T-helper cells,

This document is a draft for review purposes only and does not constitute Agency policy,

3-77 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

cytotoxic T-lymphocytes, NK cells, and macrophages displayed a decreasing trend and achieved
statistical significance at doses >0.089 mg/kg-day; absolute counts of immature T-cells were not
affected by PFDA exposure (percent changes from controls are summarized in Table 3-16). The
relative percentages of spleen immune cell populations in mice were largely unchanged, except for
macrophages, which showed, dose-related reductions at similar doses (13-19% relative to controls
over 0.089-0.71 mg/kg-day). The mostly null findings in the relative percentage values of spleen
cell immunophenotypes likely reflect the observed spleen atrophy in animals (i.e., decreases in
spleen cell numbers and spleen weights [see synthesis on Histopathology and organ weights below
for more details]) fFrawlev etal.. 20181. Furthermore, a lack of treatment-related effects was
reported for other observational immune assays evaluated in mice (i.e., anti-CD3+-mediated T-cell
proliferation and bone marrow DNA synthesis) (see Figure 3-17). In rats, results were null with
PFDA exposure (0.125-0.5 mg/kg-day) in assays of spleen cell immunophe no typing (including
spleen cell numbers and immune cell populations), anti-CD3+-mediated T-cell proliferation and
bone marrow DNA synthesis, colony formation and progenitor cell populations (see Figure 3-17).

The reductions in absolute immune cell populations in mouse spleen provide evidence
consistent with potential immunosuppression following short-term PFDA exposure, although
uncertainties related to the overt organ toxicity (i.e., spleen atrophy) remain.

Table 3-16. Percent change relative to controls in absolute spleen cell
population counts in female B6C3F1/N mice exposed to PFDA exposure for
28-days fFrawlev et al.. 20181

Endpoint

Dose (mg/kg-d)

0.045

0.089

0.179

0.36

0.71

Spleen cell

-2

-8

-13

-8

-24

B-cell (lg+)

0.3

-10

-4

-27

Cytotoxic T-cell (CD4" CD8+)

-10

-19

-22

-10

-28

Helper T-cell (CD4+ CD8 )

-11

-13

-19

-12

-29

Immature T-cell (CD4+ CD8+)

-21

-53

-21

-16

-53

Macrophage (Mac3+)

-13

-21

-31

-25

-39

Natural Killer Cell (NK1.1+ CD3 )

-15

-18

-16

-18

T-cell (CD3+)

-9

-15

-22

-14

-28

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors; shaded cells represent doses not included in the individual studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-78 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study Design Animal Description

Frawley. 2018,4287119 28 Day Oral Mouse B6C3F1/N

Organ

Spleen

Outcome Confidence

High confidence

% No significant change
A Statistically significant increase
V Statistically significant decrease

Bone Marrow High confidence
Rat Sprague-Dawley (Harlan) Spleen High confidence

Bone Marrow High confidence

Medium confidence

Endpoint Name

Spleen Cell. Absolute Values

B-cell (lg+). Absolute Values

Cytotoxic T-cell (CD4- CD8+). Absolute Values

Helper T-cell (CD4+ CD8-), Absolute Values

Immature T-cell (CD4+ CD8+), Absolute Values

Macrophage (Mac3+). Absolute Values

Natural Killer Cell (NK1 1+ CD3-), Absolute Values

T-cell (CD3+), Absolute Values

B-cell (lg+). Percent Values

Cytotoxic T-cell (CD4- CD8+), Percent Values

Helper T-cell (CD4+ CD8-), Percent Values

Immature T-cell (CD4+ CD8+), Percent Values

Macrophage (Mac3+), Percent Values

Natural Killer Cell (NK1.1+ CD3-), Percent Values

T-cell (CD3+), Percent Values

Anti-CD3+ Mediated T-Cell Proliferation

DNA Synthesis

Spleen Cell, Absolute Values

B-cell (CD45+). Absolute Values

Cytotoxic T-cell (CD8+ CD5+), Absolute Values

Helper T-cell (CD4+ CD5+), Absolute Values

Macrophage (His36+), Absolute values

Natural Killer Cell (NK+ CD8+). Absolute Values

T-cell (CD5+), Absolute Values

B-cell (CD45+), Percent Values

Cytotoxic T-cell (CD8+ CD5+), Percent Values

Helper T-cell (CD4+ CD5+), Percent Values

Macrophage (His36+), Percent Values

Natural Killer Cell (NK+ CD8+), Percent Values

T-cell (CD5+), Percent Values

Anti-CD3+ Mediated T-Cell Proliferation

DNA Synthesis

Colony Formation and Cell Immunophenotyping

Animal Description

Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)
Harlan)

Trend Test Result

significant
significant
significant
significant
not significant
significant
significant
significant
not significant
not significant
not significant
not significant
significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant
not significant

PFDA General/Observational Immune Assays

v v

-~—~-

~ w

0.3 0.4 0.5
Dose (mg/kg-day)

Figure 3-17. Effects on general/observational immune assays following exposure to PFDA in short-term oral
studies in animals (results can be viewed by clicking the HAVVC link).

This document is a draft for review purposes only and does not constitute Agency policy.

3-79 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Blood leukocyte counts

Hematological evaluations of potential alterations in blood leukocyte (white blood cell]
counts with PFDA treatment comes from three high confidence experiments (reported in two
studies) with gavage exposure for 28 days: one in female B6C3F1/N mice (Frawlev etal.. 20181 and
two in male and female S-D rats (Frawlev etal., 2018: NTP. 20181 (see Figure 3-18). The
parameters evaluated included leukocyte counts and differentials (basophils, eosinophils,
lymphocytes, monocytes, and neutrophils). For lymphocytes, both absolute counts and total counts
(absolute plus large lymphocytes such as lymphoblasts or reactive lymphocytes) were provided.

Reporting quality -

!**•

Allocation -

Observational bias/blinding -

Confounding/variable control -

Selective reporting and attrition -

Chemical administration and characterization -

Exposure timing, frequency and duration -

Endpoint sensitivity and specificity -

Results presentation -

Overall confidence -

Legend

| Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NR Not reported

Figure 3-18. Evaluation results for animal studies assessing effects of PFDA
exposure on blood leukocyte counts. Refer to HAWC for details on the study
evaluation review.

The effects of PFDA exposure on blood leukocyte counts in animals are unclear (see Table 3-
17 and Figure 3-19). Frawlev etal. f20181 found no treatment-related effects on blood leukocyte
numbers and differentials in female mice and female rats with exposures up to 0.71 and
0.5 mg/kg-day, respectively (males were not examined). In a separate study by NTP (20181.
statistically significant changes were noted in circulating leukocytes in female rats (but not males)
at higher doses (>1.25 mg/kg-day). Specifically, the number of basophils increased by 157% and

This document is a draft for review purposes only and does not constitute Agency policy,

3-80 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

71% compared to controls at doses of 1.25 and 2.5 mg/kg-day, respectively, while the number of
monocytes increased by 41% at the high-dose group (2.5 mg/kg-day). Leukocyte and lymphocyte
(total and absolute) numbers were elevated at 1.25 mg/kg-day (37-41% compared to controls) but
not at 2.5 mg/kg-day (0% compared to controls). Conversely, eosinophil counts decreased up to
64% at a dose of 2.5 mg/kg-day. In general, the hematological data suggests increases in blood
leukocyte counts and populations in female rats. The biological significance of these findings is
uncertain given the inconsistencies in the directionality of changes across dose groups in the
fFrawlev etal.. 20181 study and, more importantly, the lack of coherent evidence in other endpoints
supportive of a potential immunostimulatory response following PFDA exposure. Additionally, the
observed hematological changes occurred mostly at high PFDA doses (>1.25 mg/kg-day) associated
with adverse systemic effects (see Section 3.2.9 on General toxicity effects for more details).

Table 3-17. Percent change relative to controls in blood leukocyte counts in
female S-D rats exposed to PFDA exposure for 28-days (NTP. 2018)

Endpoint

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Basophils

157

Eosinophils

-27

-18

-9

-64

Leukocytes

Lymphocyte (absolute)

Lymphocyte (total)

Monocytes

-12

Neutrophils

-9

-17

-6

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors; shaded cells represent doses not included in the individual studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-81 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Study Name

Outcome Confidence

Study Design

Animal Description

Trend Test Result

Basophils

Frawley, 2018,4287119

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan)

not significant

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan)

significant

28 Day Oral

Rat, Sprague-Dawley (Harlan)

not significant

Frawley. 2018, 4287119

High confidence

28 Day Oral

Mouse, B6C3F1/N ( )

not significant

Eosinophils

Frawley, 2018,4287119

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan)

not significant

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan)

significant

28 Day Oral

Rat, Sprague-Dawley (Harlan)

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Histopathology and organ weights

The data on immune histopathology and organ weights is described in one study using
female B6C3F1/N mice fFrawlev etai. 20181 and two studies in male and female rats (Frawlev et
al.. 2018: NTP. 20181 exposed to PFDA for 28 days via gavage. The NTP T20181 study was high
confidence for histopathology and organ weight measures. Frawlev etal. f20181 was considered
high confidence for the evaluation of organ weight but exhibited deficiencies in the presentation
and discussion of histopathological findings (lack of quantitative data), which resulted in a medium
confidence rating for this endpoint (see Figure 3-20).

Reporting quality-

Allocation -

Observational bias/blinding •
Confounding/variable control •
Selective reporting and attrition ¦
Chemical administration and characterization •
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity ¦
Results presentation ¦
Overall confidence ¦

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Not reported

* Multiple judgments exist

Figure 3-20. Evaluation results for animal studies assessing effects of PFDA
exposure on immune histopathology and organ weights. Refer to HAWC for

details on the study evaluation review.

Animal toxicity studies provide some evidence of immune organ histopathology (see
Figure 3-21). The bone marrow, lymph nodes, spleen and thymus were examined histologically in
male and female rats exposed to doses ranging from 0.125-2.5 mg/kg-day fFrawlev etal.. 2018:
NTP. 20181. No treatment-related effects were found in any of these organs at doses
<0.625 mg/kg-day across the two rat studies, but morphological changes were observed in bone
marrow and thymus in the study that tested higher doses (>1.25 mg/kg-day) fNTP. 20181.
Increased incidences of bone marrow hypocellularity (10/10 in males and females) and thymic
atrophy (9/10 in males and 8/10 in females) were observed in rats at the highest dose

This document is a draft for review purposes only and does not constitute Agency policy,

3-83 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

(2.5 mg/kg-day), while incidence of lymphocyte apoptosis in the thymus was increased in males
only at a dose of 1.25 mg/kg-day (8/10 rats). The aforementioned lesions ranged from mild to
moderate in severity and did not occur in the controls or in other exposure groups.

Changes in immune organ weights were reported in female mice and male/female rats
across three, 28-day gavage studies (Frawlev et al.. 2018: NTP. 20181 (see Table 3-18 and
Figure 3-17). The rat study by Frawlev etal. (2018) included three cohorts exposed to similar
experimental conditions. Statistically significant decreases in spleen weights (absolute and
relative) were observed across species and sexes at >0.179 mg/kg-day, reaching 55% in rats and
22% in mice relative to controls at the highest doses tested (2.5 and 0.71 mg/kg-day, respectively)
f Frawlev etal.. 2018: NTP. 20181. Although there were no notable histopathological findings in the
spleen, the organ weight reductions in mice are concordant with alterations in spleen cell numbers
and populations previously described (see synthesis on General/observational immune assays
above for additional details). Absolute and relative thymus weights decreased in a dose-dependent
manner (29-75% compared to controls) at >1.25 mg/kg-day in rats that exhibited thymic lesions
(atrophy and apoptosis) and marked body weight reductions in one study fNTP. 20181. In contrast,
another study reported increases in absolute and relative thymus weights in rats at lower PFDA
doses (0.125-0.5 mg/kg-day) but the results were not consistent across study cohorts and in most
cases did not show a dose-response dependency (Frawlev etal.. 2018) (see Table 3-21). As such,
the significance of the increases in thymus weights in rats is uncertain. Thymus weights in mice
were not impacted by PFDA treatment (up to 0.71 mg/kg-day) in one study (Frawlev etal.. 2018).

In summary, histopathological lesions were found in the bone marrow and thymus of rats
and decreased spleen and thymus weights were reported in mice and/or rats after short-term
PFDA exposure. The effects on spleen weights are coherent with reductions in spleen cell counts
and populations in mice at >0.089 mg/kg-day. The bone marrow and thymus lesions in rats were
only observed in the presence of marked reductions in body weight (12-38% relative to controls)
at PFDA doses >1.225 mg/kg-day, which provides a significant source of uncertainty. Indeed, bone
marrow hypocellularity and thymic atrophy have been linked to diet restriction in short-term rat
studies (Levin etal.. 1993) and PFDA-induced wasting syndrome characterized by decreased food
consumption and rapid weight loss has been well documented in animals (see Section 3.2.10 on
General toxicity for more details). As such, the toxicological significance of the histopathological
findings is uncertain.

This document is a draft for review purposes only and does not constitute Agency policy.

3-84 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Organ

Endpoint Name

Study Name

Outcome Confidence

Study Type

Animal Description

Trend Test Result

Bone Marrow

Histopathology

Frawley, 2018, 4287119

Medium confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ( )

not reported

Hypocellularity

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ( )
Rat, Sprague-Dawley (Harlan)

significant
significant

Lymph Node

Histopathology

Frawley, 2018, 4287119

Medium confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ( )

not reported

NTP. 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (c)
Rat, Sprague-Dawley (Harlan) (, )

not applicable
not applicable

Spleen

Histopathology

Frawley, 2018, 4287119

Medium confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (c)

not reported

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (i)
Rat, Sprague-Dawley (Harlan) (,-?)

not applicable
not applicable

Thymus

Histopathology

Frawley, 2018, 4287119

Medium confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (I)

not reported

Atrophy

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (2)
Rat, Sprague-Dawley (Harlan) (o)

significant
significant

Lymphocyte Apoptosis

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (c?)

significant

PFDA Immune Organ Histopathology

No significant changed Statistically significant increase Statistically significant decrease 1

• • •

• • • • •

-• •-

Dose (mg/kg-day)

Figure 3-21. Effects on immune organ histopathology following exposure to PFDA in short-term oral studies in
animals (results can be viewed by clicking the HAWC link}.

This document is a draft for review purposes only and does not constitute Agency policy,

3-85 ' DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-18. Percent change relative to controls in immune organ weights in
short-term animal studies after exposure to PFDA

Animal group

Dose (mg/kg-d)

0.045

0.089

0.125-
0.179

0.25-0.36

0.5-0.71

1.25

2.5

Spleen weight (absolute)
Male S-D rats NTP (2018)

-4

-26

-49

Spleen weight (absolute)
Female S-D rats NTP (2018)

-2

-9

-36

-55

Spleen weight (absolute)

Female C57BL/6N mice Frawlev et al.

(2018)

-3

2.8

-18

-6

-20

Spleen weight (relative)
Male S-D rats NTP (2018)

-1

-6

-19

Spleen weight (relative)
Female S-D rats NTP (2018)

-3

-5

-9

-27

-30

Spleen weight (relative)

Female C57BL/6N mice Frawlev et al.

(2018)

-3

-6

-16

-9

-22

Thymus weight (absolute)
Male S-D rats NTP (2018)

-1

-44

-75

Thymus weight (absolute)
Female S-D rats NTP (2018)

-20

-65

Thymus weight (absolute)

Female S-D rats: MPS cohort Frawlev et al.

(2018)

Thymus weight (absolute)

Female S-D rats; Histopathology cohort

Frawlev et al. (2018)

-5

-1

Thymus weight (absolute)

Female S-D rats; TDAR to SRBC cohort

Frawlev et al. (2018)

Thymus weight (relative)
Male S-D rats NTP (2018)

-2

-29

-61

Thymus weight (relative)
Female S-D rats NTP (2018)

-8

-46

Thymus weight (relative)

Female S-D rats: MPS cohort Frawlev et al.

(2018)

Thymus weight (relative)

Female S-D rats; Histopathology cohort

Frawlev et al. (2018)

-7

Thymus weight (relative)

Female S-D rats; TDAR to SRBC cohort

Frawlev et al. (2018)

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors; shaded cells represent doses not included in the individual studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-86 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Organ

Endpoint

Study Name

Outcome Confidence

Study Type

Animal Description

Trend Test Result

PFDA Immune Organ Weight

Spleen

Absolute

Frawley, 2018, 4287119
NTP, 2018, 4309127

High confidence
High confidence

28 Day Oral
28 Day Oral

Rat, Sprague-Dawley (Harlan) (2)
Rat, Sprague-Dawley (Harlan) ($)

not significant
not significant
not significant
significant

•—•—•

•—•—•—~—~

Rat. Sprague-Dawley (Harlan) (-)

significant

•—•—•—¥—~

Frawley. 2018. 4287119
Frawley, 2018, 4287119

High confidence
High confidence

28 Day Oral
28 Day Oral

Mouse. B6C3F1/N (+)

Rat, Sprague-Dawley (Harlan) (2)

significant
significant
not significant
not significant

•

Relative to Body

•—•—•
•—•—•
•—•—•

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat. Sprague-Dawley (Harlan) (V)

significant

•—•—*—~—~

Rat, Sprague-Dawley (Harlan) (s;)

significant

•—•—•—•—~

Frawley. 2018, 4287119

High confidence

28 Day Oral

Mouse, B6C3F1/N (+)

significant

•

Thymus

Absolute

Frawley, 2018, 4287119

High confidence

28 Day Oral

Rat. Sprague-Dawley (Harlan) ($)

not significant
not significant
significant

•—•—•
•—•—•

A—A— •

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (^)

significant

•—•—•—•—~

Frawley, 2018, 4287119

High confidence

28 Day Oral

Rat. Sprague-Dawley (Harlan) (. ')
Mouse. B6C3F1/N (?)

significant
not significant

Relative to Body

Frawley. 2018, 4287119

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ($)

not significant
not significant
not significant

•—•—•
•—•—•

A A •

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (Q)

significant

•—•—•—•—W

Rat, Sprague-Dawley (Harlan) (T)

significant

•—•—•—*—~

Frawley. 2018, 4287119

High confidence

28 Day Oral

Mouse, B6C3F1/N (J)

not significant

# No significant change A Statistically significant increase ^ Statistically significant decrease

0.01

0.1 1
Dose (mg/kg-day)

Figure 3-22. Effects on immune organ weights following exposure to PFDA in short-term oral studies in animals.

The rat study by Frawlev et al. (2018) included three cohorts exposed to similar experimental conditions. Results can be
viewed by clicking the HAWC link).

This document is a draft for review purposes only and does not constitute Agency policy,

3-87 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Sensitization and allergic response

Immune function assays

A 14-day study in male ICR mice exposed to a dose of 21.4 mg/kg-day examined the effect of
PFDA treatment on OVA-induced active systemic anaphylaxis fLee and Kim. 20181. a well-accepted
model for evaluating mast cell function and allergic reactions fie etal.. 2015: Evans et al.. 2014:
Ribeiro-Filho etal.. 20141. The study was rated as low confidence due to issues with reporting on
potential confounding effects (no information on general systemic toxicity measures; this excessive
dose would be expected to cause significant, overt toxicity given observations, including "wasting
syndrome", from other short-term studies with similar dosing paradigms; see Section 3.2.10 on
GENERAL TOXICITY EFFECTS for more details], experimental groups (no indication of
randomization) and the characterization of the test compound (no information on analytical
verification or specific method of administration) (see Figure 3-23),

Reporting quality -
Allocation -
Observational bias/blinding
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity
Results presentation
Overall confidence

V®e

i ift-

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

Figure 3-23. Evaluation results for animal study assessing effects of PFDA
exposure on immune function assays for sensitization and allergic response.

Refer to HAWC for details on the individual study evaluation review.

PFDA (21.4 mg/kg-day) exacerbated the response to OVA-induced active systemic
anaphylaxis in mice as indicated by a significant decrease in rectal temperature (i.e., hypothermia)
and significant elevation in serum levels of inflammatory mediators such histamine, TNFa and

This document is a draft for review purposes only and does not constitute Agency policy,

3-88 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

immunoglobulins (IgE and IgGl) compared to OVA treatment alone fLee and Kim. 20181.

Histamine is released in response to mast cell degranulation and plays a key role in immediate-type
hypersensitivity fAmin. 20121. The findings from the Lee and Kim f20181 study suggest possible
induction of immediate-type hypersensitivity, although the exposure dose was high compared to
doses associated with immunosuppressive responses in animals (0.089-2.5 mg/kg-day) and raises
concerns over potential confounding with general toxicity effects. Although the study provided no
information on general toxicity measures, PFDA exposure was associated with significant body
weight reductions at doses >1.25 mg/kg-day in oral exposure studies and the induction of wasting
syndrome in acute, i.p. injection studies at doses >20 mg/kg (see Section 3.2.10 on General toxicity
for more details).

Mechanistic studies and supplemental evidence

The available supplemental evidence most relevant to interpretation consists of an acute i.p.
injection study evaluating immunotoxicity endpoints in exposed rats and a few in vitro studies in
human and animal models examining possible mechanisms of immunotoxicity following PFDA
exposure.

An acute i.p. injection study investigating potential immune effects of PFDA exposure (20
and 50 mg/kg) in Fischer 344 rats showed reductions in the antibody (i.e., serum KLH-specific
IgG2a levels) and DTH responses to Keyhole limpet hemocyanin (KLH) in exposed animals (Nelson
etal.. 19921: the effects on the DTH response were not statistically significant but showed a
decreasing trend with increase in dose at each timepoint (40-46% and 38-47% compared to ad
libitum-fed controls after 8 and 30 days respectively). In addition, NK cell activity was increased in
rats after PFDA treatment fNelson etal.. 19921. Exposure to PFDA altered immune responses in
comparison to both ad libitum- and pair- fed controls with the exception of NK activity, which was
similarly elevated in PFDA-exposed rats and pair-fed (but not ad libitum) controls (Nelson etal..
19921. The acute toxicity of PFDA is characterized by a wasting syndrome, which induces rapid and
severe reductions in food consumption and body weight in rats at doses similar to those associated
with the immunomodulatory effects described above (20-100 mg/kg) (see Section 3.2.9 on General
toxicity effects for more details). The findings suggest that the antibody and DTH responses are
directly related to PFDA exposure, while the NK activity is likely a secondary effect of chemical-
induced wasting syndrome. Functional alterations in antibody and DTH responses after acute i.p.
exposure is supportive of the immunomodulatory effects observed after short-term PFDA
administration (see synthesis of Animal studies in this Section for more details).

Using an in vitro model to study mast cell functions and allergic inflammation, Lee and Kim
(20181 showed that PFDA exposure can elevate markers of mast cell degranulation (histamine,
^-hexosaminidase and intracellular calcium levels), increase gene expression and secretion of pro-
inflammatory cytokines involved in immune cell recruitment and activation (TNF-a, IL-1 (3, IL-6, and
IL-8) and induce NF-kB transactivation in IgE-stimulated rat basophilic leukemia (RBL-2H3) cells

This document is a draft for review purposes only and does not constitute Agency policy.

3-89 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

fLee and Kim. 20181. The data are consistent with the exacerbation of hypothermia and allergic
inflammatory mediators (histamine, TNFa, IgE and IgGl levels) in OVA-stimulated mice following
continuous high-dose oral exposure to PFDA (see synthesis of Animal studies in this Section for
more details) and suggest a plausible mechanism for PFDA-induced immediate-type
hypersensitivity.

Other potential mechanisms of PFDA-induced immune effects were evaluated in two studies
conducted in human and animal in vitro cell models. No effects on IgM secretion and surface
membrane expression were observed in human (F4 and Hurtwitz) or murine (HPCM2) B cell lines
at non-cytotoxic PFDA concentrations, but detergent-like activity (i.e., solubilization of cell
membranes) was reported in these lymphoblastoid cell lines at doses that caused significant
cytotoxicity fLevitt and Liss. 19861. Another study evaluated the effects of PFDA on cytokine
release in human primary and cultured leukocytes (Corsini etal.. 20121. Decreases in pro-
inflammatory (TNF-a and interleukin [IL]-6) and anti-inflammatory (IL-10 and interferon gamma
[IFN-y]) cytokine levels were reported in human peripheral blood leukocytes stimulated with
lipolysaccharide (LPS) and phytohemagglutinin, respectively, following PFDA exposure fCorsini et
al.. 20121. Leukocytes from female donors were generally more susceptible to alterations in
cytokine production (primarily TNFa) compared to male counterparts, although differential
responses across cytokine measures were apparent and may be explained in part by variability in
cell donors (Corsini etal.. 20121. Similarly, PFDA decreased TNFa levels and NF-kB activation
(measured as I-kB degradation, p65 phosphorylation and NF-kB gene reporter activity) in human
promyelocytic THP-1 cells stimulated with LPS but had no effects on PPARa-mediated
transactivation fCorsini etal.. 20121. Cell viability measured via the lactase dehydrogenase assay
was unaffected in this cell line by PFDA treatment fCorsini et al.. 20121. The data suggest that PFDA
suppresses cytokine release (i.e., TNFa) by interfering with the NF-kB pathway in stimulated
immune cells and that such effects may occur independently of PPARa activation.

Collectively, the mechanistic data indicate that PFDA can modulate NF-kB activation to
induce both pro- and anti- inflammatory responses in cultured immune cells, which may have
implications for the mechanisms of immunotoxicity of this compound.

Evidence Integration

Studies in humans and animals exposed to PFDA are available for the evaluation of potential
immunosuppression and sensitization or allergic responses.

The evidence of an association between PFDA exposure and immunosuppressive effects in
human studies is moderate. This is based on largely consistent decreases in antibody response
following vaccination (against two different infectious agents) in two medium confidence studies
describing results from two independent birth cohorts in the Faroe Islands with outcome
measurement in childhood. Reduced antibody response is an indication of immunosuppression and
may result in increased susceptibility to infectious disease flPCS. 20121. The antibody results

This document is a draft for review purposes only and does not constitute Agency policy.

3-90 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

present a consistent pattern of findings that higher prenatal, childhood, and adult serum
concentrations of PFDA were associated with suppression of at least one measure of the anti-
vaccine antibody response to common vaccines in two well-conducted birth cohorts in the Faroe
Islands and supported by a low confidence study in adults. An inverse association was observed in
21 of 26 evaluations, with a minimum of a 2% decrease in antibody concentration per doubling of
PFDA concentration at levels consistent with the general population in NHANES; six of these
evaluations were statistically significant and exhibited a large magnitude of effect (i.e., >18%
decrease in response). These associations were observed despite poor study sensitivity, which
increases confidence in the findings. There is some remaining uncertainty resulting from
variability in the response, including positive associations in a few exposure-outcome
combinations, differences in the responses by age of exposure and outcome measures as well as
timing of vaccination (initial and boosters), from potential confounding across PFAS, and from
inconsistency in two other medium confidence studies with outcome measurement in adults and
cross-sectional exposure measurement in children. Overall, the evidence supports an association
with immunosuppressive-type effects. These results are consistent with hazard identification
conclusions from the NTP f20161 monograph on immunotoxicity associated with exposure to PFOS
and PFOA, which concluded that PFOA and PFOS are presumed to be an immune hazard to humans
based largely on evidence of suppression of antibody responses in both human and animal studies
(NTP. 2016). Although no effects were observed in T-dependent antibody responses with PFDA in
one rat and one mouse study (both high confidence), other immunomodulatory responses were
observed in animals that indicate potential for immunosuppression (see summary of animal
evidence below for more details).

The database of animal studies examining PFDA-induced immunosuppressive responses
consists of two high or medium confidence studies in B6C3F1/N mice fFrawlev etal.. 20181 and/or
S-D rats (Frawlev et al.. 2018: NTP. 2018) exposed via gavage for 28 days. PFDA did not elicit a
strong immunotoxic response in animals as evidenced by the absence of treatment-related effects
in a host resistance assay and most immune function assays (NK cell activity and T-dependent
antibody responses to SRBC, mixed leukocyte response and DTH response to C. albicans).
Nevertheless, coherent responses that suggest potential immunosuppression by PFDA exposure
were observed, which is consistent with the human evidence. The immunomodulatory responses
included dose-related decreases in phagocytic activity of rat liver macrophages (MPS activity) at >
0.25 mg/kg-day and in immune cell population counts in mouse spleen at >0.089 mg/kg-day
(Frawlev etal.. 2018). but issues regarding overt organ toxicity (increased liver weight and
hepatocyte necrosis and spleen atrophy, respectively) introduce significant uncertainty (Frawlev et
al.. 2018). Additionally, morphological changes occurred in the bone marrow (hypocellularity) and
thymus (atrophy and lymphocyte apoptosis) of rats at PFDA doses associated with systemic toxicity
(i.e., decreased body weights at >1.25 mg/kg-day) fNTP. 20181: the changes are consistent with the
wasting syndrome that PFDA elicits and could represent secondary effects of the accompanying

This document is a draft for review purposes only and does not constitute Agency policy.

3-91 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

systemic toxicity. In light of the uncertainties in the available database, the evidence for potential
immunosuppression from short-term animal studies is considered slight.

Mechanistic evidence from a high-dose, i.p. injection study is supportive of potential PFDA-
induced immunosuppression (i.e., decreased antibody and DTH responses) in rats at >20 mg/kg
(Nelson etal.. 19921. Furthermore, an in vitro study using stimulated human primary and cultured
leukocytes suggests that PFDA is capable of inhibiting NF-kB transcription and suppressing
cytokine production (Corsini etal.. 20121. which may be relevant to its mechanisms of
immunotoxicity. Limitations in the mechanistic information include issues interpreting the
exposure context (i.e., acute, high-dose exposure) of the i.p. injection study and general lack of
studies in animal and human models that can provide support for the biological plausibility of
putative immunosuppression observed in human and animal studies.

There is slight evidence for sensitization and allergic responses from studies in humans, but
notable limitations and uncertainties in the evidence base remain. In human studies, the available
evidence for infectious disease and hypersensitivity was less consistent than the evidence on
immunosuppression and had more uncertainties resulting from a limited number of studies,
unexplained heterogeneity in outcome or results, variable exposure assessment approaches that
considered exposure at different times in relation to outcomes, and in some cases self-reported
outcomes. For asthma, two of the three available studies reported no association with PFDA
exposure. However, significant associations with asthma diagnosis and asthma-related outcomes,
including an exposure response gradient, were observed in one well-conducted (medium
confidence) study with adequate sensitivity (Dong etal.. 20131. This study also had the most
specific outcome definition, based on asthma incidence in the past year. These differences could
account for the inconsistency with other asthma studies, including the other medium confidence
study which examined "ever asthma". In addition, increases in biomarkers related to asthma were
reported in this study, providing biological plausibility to the apical association. Still, the number of
available studies is small, and poor sensitivity makes the null results difficult to interpret.

In animals, the single, short-term, low confidence study that examined endpoints relevant to
sensitization and allergic responses reporting findings coherent with immediate-type
hypersensitivity (i.e., exacerbation of hypothermia and markers of mast cell-mediated allergic
inflammation in OVA-induced mice) fLee and Kim. 20181: however, the high exposure dose used
(21.4 mg/kg-day) raises significant concerns about potential confounding effects by indirect
systemic toxicity and thus these coherent results were not interpreted to provide biological
plausibility for the findings in humans and the animal evidence was considered indeterminate (Lee
and Kim. 20181.

Altogether, considering the available evidence from human, animal and mechanistic studies,
the evidence indicates that PFDA exposure is likely to cause adverse immune effects, specifically
immunosuppression, in humans, given sufficient exposure conditions3 (see Table 3-19). The hazard
judgment is driven primarily by consistent evidence of reduced antibody response from human

This document is a draft for review purposes only and does not constitute Agency policy.

3-92 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

epidemiological studies (mostly from two birth cohort studies) at levels of 0.3 ng/mL (median
exposure in studies observing an adverse effect). The evidence in animals showed coherent
immunomodulatory responses at >0.089 mg/kg-day 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. A
small number of studies conducted via i.p. injection and in vitro exposure in human and rodent cell
culture models add some support for the biological plausibility of the observed phenotypic effects.
While there is some evidence that PFDA exposure might also have the potential to affect
sensitization and allergic responses, the human evidence underlying this possibility is uncertain
and without support from animal or mechanistic studies. Based on the antibody response data in
humans, children and young individuals exposed during critical developmental windows may
represent a potential susceptible population for the immunosuppressive effects of PFDA. 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 major source of uncertainty in the immunotoxicity database
of PFDA.

This document is a draft for review purposes only and does not constitute Agency policy.

3-93 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-19. Evidence profile table for PFDA exposure and immune effects

Evidence stream summary and interpretation

Inferences and summary
judgment

Evidence from studies of exposed humans (see Section 3.2.2: Human Studies)

Studies and confidence

Summary and key
findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

©©O

Evidence indicates (likely)

ImmunosuDDression
(Antibody response)
4 medium confidence
studies (3 in children)
and 1 low confidence
study

• Three studies in
children and one in
adults reported
decreased antibody
response following
vaccination with
higher PFDA
exposure

• Consistency overall
across vaccine type,
timing of vaccination,
and age at antibody
response
measurement
including in two
medium confidence
studies with
prospective exposure
measurement and
outcomes in children

• Associations observed
despite limited
sensitivity

• Potential for

confounding across
PFAS

ffiffiO

Moderate

Generally consistent
evidence for decreased
antibody responses. The

inconsistent and low
confidence evidence on
infectious disease did not
influence this judgment.

Primary basis:

Evidence of

immunosuppression from
human studies indicating
reduced antibody response in
children at levels of
approximately 0.3 ng/mL
(moderate evidence) and some
coherent findings in animals
(slight evidence) at
>0.089 mg/kg-d. Overall, other
forms of potential PFDA-
induced immunotoxicity,
including slight human
evidence for hypersensitivity-
related outcomes, were
interpreted with less certainty.

ImmunosuDDression
(Infectious diseases)
3 medium and 2 low
confidence cohort
studies

• Positive association
with infectious
diseases in one
medium and two low
confidence studies,
but inconsistency
across studies of the
same

infections/symptoms

• No factors noted

• Unexplained
inconsistency, though
limited sensitivity may
contribute

• Imprecision

Human relevance:

Coherent effects in human and
animal studies

Cross-stream coherence:
Evidence of

immunosuppression in both
animals and humans.

This document is a draft for review purposes only and does not constitute Agency policy.

3-94 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

Sensitization and
allergic resoonse
7 medium confidence
studies in children

• Significantly higher
odds of asthma
(OR = 3.2) in one
medium confidence
study. One
additional study
reported increased
odds of asthma with
higher PFDA
exposure, but only in
a small sub-group
that did not receive
MMR vaccine before
age 5

• Other studies
reported no
association with
hypersensitivity
outcomes

• Large effect size for
asthma incidence in
the only study with
adequate sensitivity
(based on exposure
contrast and outcome
definition)

• Exposure-response
gradient across
quartiles in same
study

• Potential for
confounding across
PFAS

• Unexplained
inconsistency across
studies

®oo

Slight

Sparse evidence for
hypersensitivity with some
concerns for unexplained
inconsistency and potential
confounding

Susceptible populations and
lifestages:

Based on the antibody
response data in humans,
children and fetuses may be at
higher risk of adverse effects.

Other inferences'.

MOA is unknown, but some
uncertain evidence from
human and animal in vitro
studies suggests a possible role
for NFkB in both pro- and anti-
inflammatory responses that
may be relevant to the
mechanism(s) of
immunotoxicity of PFDA.

Evidence from in vivo animal studies (see Section 3.2.2: Animal Studies)

Studies and confidence

Summary and Key
findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

ImmunosuDDression

2 high/medium
confidence studies in
mice and/or rats

• 28-day gavage (2x)

• Decreased hepatic
MPS activity in rats
at >0.5 mg/kg-d in
the presence of liver
toxicity (increased
liver weight and
hepatocyte necrosis)

• Decreased absolute
spleen cell
population counts in
mice at >0.89 mg/kg-
d in the presence of

• Coherence across
immune responses
(i.e., MPS activity in
rats and spleen cell
population counts and
spleen weights in
mice)

• Dose- response
gradient for MPS
activity, absolute
spleen cell counts and
spleen weights

• Lack of effects on host
resistance and most
immune function
assays

• Potential confounding
with overt organ or
systemic toxicity.

®oo

Slight
Coherent evidence of
potential
immunosuppression in rats

and mice at doses
>0.089 mg/kg-d across two
high/medium confidence
studies; however, there is
uncertainty due to potential

confounding effects with
overt organ/systemic toxicity.

This document is a draft for review purposes only and does not constitute Agency policy.

3-95 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

spleen atrophy
(decreased spleen
weights and total cell
counts)

• Bone marrow and
thymic lesions and
decreased thymus
weights in rats at
>1.25 mg/kg-d in the
presence of marked
body weight
reductions

• No effects in a host
resistance assay in
mice or other
immune function
assays conducted in
rats and mice at
doses up 0.71 mg/kg-
d

• High/medium

confidence studies

Sensitization and
allergic response

1 low confidence study
in mice

• 14-day

• Exacerbation of
hypothermia and
release of serum
inflammatory
markers
(i.e., histamine,

TNFcx, IgE and IgGl)
in OVA-stimulated
mice at 21.4 mg/kg-d

• Coherence across
markers of allergic
inflammation and
hypersensitivity

• Potential for
confounding by
systemic toxicity

QQQ

Indeterminate

Low confidence evidence
with considerable
uncertainty due to potential
confounding effects due to
high dose systemic toxicity.

Mechanistic evidence and supplemental information (see subsection above)

Biological events or
pathways

Primary evidence evaluated

Key findings, interpretation, and limitations

Evidence stream judgment

Mast cell function and
allergic response

Interpretation: PFDA may induce mast cell-mediated allergic inflammation via NFkB
activation.

A small number of
mechanistic studies in human

This document is a draft for review purposes only and does not constitute Agency policy.

3-96 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

Key findings'.

• Increases in markers of mast cell degranulation (histamine, P-hexosaminidase
and intracellular calcium levels), immune cell recruitment and activation (TNF-
a, IL-ip, IL-6, and IL-8 levels) and NFE1B transactivation in IgE-stimulated rat
RBL-2H3 cells.

Limitations: Single study available.

and rodent in vitro models
suggest a possible
involvement of NFHB in pro-
and anti-inflammatory
responses that may be
relevant to the mechanisms
of immunotoxicity of PFDA.
Supportive evidence of
immunosuppression in rats
was reported in an acute, i.p.
injection study. Although the
available evidence is limited
introducing significant
uncertainty, the findings
provide some support for the
biological plausibility of the
immune-related responses in
humans and animals.

Other mechanisms

Interpretation: PFDA may suppress cytokine production by inhibiting NFkB

activation.

Key findings:

• Attenuation of cytokine release (including TNFa) in stimulated human
peripheral blood leukocytes (leukocytes from female donors appeared to be
more susceptible to these effects); decreases in TNFa release and NF0B
activation but no effects on PPARa transactivation in stimulated human
promvelocvtic THP-1 cells (Corsini et al., 2012).

• No effects on IgM secretion and surface expression in human and murine B cell
lines exposed to noncvtotoxic PFDA concentrations (Levitt and Liss, 1986).

Limitations: Few studies available; cell donor variability introduces some uncertainty
in interpreting sex-specific differences in cytokine release from exposed human
primary leukocytes.

Other evidence

Interpretation: Results are consistent with immunosuppressive responses observed
in oral exposure studies.

Key findings:

• Decreases in antibody response and DTH in KLH-stimulated rats compared to
libitum and pair-fed controls; Increase in NK cell activity may be attributable to
PFDA-induced anorexia.

Limitations: Single study with high-dose, one-time i.p. exposure.

This document is a draft for review purposes only and does not constitute Agency policy.

3-97 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.2.3. DEVELOPMENTAL EFFECTS
Human studies

Studies of developmental endpoints related to PFDA are available for fetal and post-natal
growth restriction, spontaneous abortion, anogenital distance, birth defects, and gestational
duration outcomes (i.e., preterm birth and gestational age). Given that spontaneous abortion and
preterm birth could be driven by either female reproductive or developmental toxicity, these
endpoints are also discussed in the context of coherence in Section 3.2.5 on Female reproductive
effects.

Forty-eight epidemiological publications (across 46 different studies) examining examined
PFDA exposures in relation to developmental endpoints were identified in the literature search.

This included the following: eight studies on postnatal growth, 12 studies on gestational duration,
six on fetal loss, three on anogenital distance, two studies on birth defects, and 31 publications
(across 29 different studies) examined fetal growth restriction. Publications based on overlapping
populations in the same cohort were included in the synthesis only if they provided some unique
data for different endpoints. For example, the Bierregaard-Olesen etal. (2019) study from the
Aarhus birth cohort also provided birth length and head circumference measures in the overall
population and across sex that were not included in the main study by Bach etal. (2016). Therefore,
it is included in the fetal growth restriction count above and considered one study (population)
from two publications with separate analyses. This synthesis, and especially the evaluation of
consistency across studies, focuses on a primary study to avoid duplicative analyses or
overweighting of one study population. Although the results for the smaller sample size in this
study are not plotted, in this instance divergent primary birthweight (BWT) results are presented
for comparison in the text Another study by Gvllenhammar et al. (2018) was supplemented by a
second publication (Swedish Environmental Protection Agency. 2017) that provided mean BWT
data on a larger population from the same cohort. Supplemental data and communication from
study authors were used if they provided additional data or information, and US EPA calculated
confidence intervals and rescaled study results to provide comparisons based on a ln-unit change to
increase comparability.

Additional Methodological Considerations

As detailed in the PFAS Systematic Review Protocol (Appendix A) and Section 1.2.2, there
were multiple outcome-specific considerations for study evaluation that influenced the domain
ratings and the 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. Pregnancy hemodynamics represent a source of uncertainty as PFAS biomarkers
sampled late in pregnancy may be prone to bias potentially from either confounding or reverse
causality (see Appendix F for detailed discussion). Among the few fetal growth studies [e.g.,

This document is a draft for review purposes only and does not constitute Agency policy.

3-98 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

fGvllenhammar et al.. 20181 for PFDA] examining the potential for confounding by measures of
pregnancy hemodynamics (e.g., plasma albumin or GFR measures), there is very little direct
evidence that these measures were important confounders fGyllenhammar etal.. 2018: Meng et al..
2018: Sagiv etal.. 2018: Whitworth et al.. 20121 across different PFAS. Sample timing patterns
across studies were considered here to see if results among studies with early sampling (i.e.,
studies with any trimester 1 sampling) differed from those with later sampling (i.e., maternal
samples exclusively from trimester 2 through trimester 3, umbilical cord, placental or post-partum
maternal samples). More research is needed especially amongst studies with early samples and/or
with repeated measures during different stages of pregnancy to further clarify any potential impact
of this source of uncertainty in epidemiological studies using biomarkers. There is additional
uncertainty across all health endpoints due to potential confounding by co-occurring PFAS (see
Appendix A and F for methods and analyses, respectively). For fetal growth restriction and other
developmental endpoints, there may be more concern over potential PFAS co-exposure
confounding due to PFNA given higher correlations with PFDA and associations that are fairly
comparable in consistency and magnitude, as detailed in Appendix F. Although there is some
uncertainty as to whether other PFAS are plausible confounders here, studies were downgraded if
the authors did not rule out or account for these or other covariates that may be confounders.

For the exposure domain, all the available studies analyzed PFAS in serum or plasma using
standard methods. Given the long half-life of PFDA, samples collected during all three trimesters
(and shortly after birth) were considered representative of the most critical in utero exposure
windows for fetal growth and gestational duration measures. Various measures of postnatal
growth were included based on an assumed fetal programming mechanism (i.e., Barker hypothesis)
where in utero perturbations or exposures, such as poor nutrition, can lead to developmental
effects such as fetal growth restriction and ultimately adult-onset metabolic-related disorders (see
more on this topic in De Boo and Harding (20061 and Perng etal. (20161 syntheses for metabolic
disorders for other PFAS). There is some evidence that birth weight deficits from in utero
exposures can be followed by increased weight gain during rapid growth catch-up periods in early
childhood (Perng etal.. 20161. Therefore, the most critical exposure window for measures of
postnatal (and early childhood) weight and height change is assumed to be in utero. Thus, studies
were downgraded if exposure data were collected later during childhood concurrent with outcome
assessment (i.e., cross-sectional analyses).

Studies were also downgraded for study sensitivity, for example, if they had limited
exposure contrasts (i.e., limited exposure ranges or distributions) or small sample sizes, since this
can impact the ability of studies to detect statistically significant associations that may be present
(especially when sample size is reduced by estimating stratum-specific results such as by sex). In
the outcome domain, specific considerations included accuracy of fetal and early childhood
anthropometric measures and adequacy of case ascertainment for dichotomized (i.e., binary)
outcomes. Mismeasurement and incomplete case ascertainment can affect the accuracy of effect

This document is a draft for review purposes only and does not constitute Agency policy.

3-99 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

estimates by impacting both precision and validity. For example, the spontaneous abortion studies
were downgraded for incomplete case ascertainment in the outcome domain given that some
pregnancy losses go unrecognized early in pregnancy (e.g., before implantation). This incomplete
ascertainment, referred to as left truncation, can result in decreased study sensitivity and loss of
precision. Often, this type of error can result in bias towards the null if ascertainment of fetal loss is
not associated with PFDA exposures (i.e., non-differential). In some situations, differential loss is
possible and bias away from the null and can manifest as an apparent protective effect. Anogenital
distance (AGD) is an externally visible marker that has been shown in animal studies to be a
sensitive indicator of prenatal androgen exposure (lower androgen levels associated with
decreased AGD). It is associated with other reproductive tract abnormalities, including
hypospadias and cryptorchidism in human and animal males fLiu etal.. 2014: Sathvanaravana et al..
2010: Sal azar-Martinez etal.. 2004). The primary outcome-specific criteria for this outcome are the
use of clearly defined protocols for measurement, ideally multiple measures of each distance
(averaged), and minimal variability in the age of participants at measurement. In boys, measures
can be taken from the center of the anus to the posterior base of the scrotum (ASD) or from the
center of the anus to the cephalad insertion of the penile (APD).

Fetal and childhood growth restriction was examined through several endpoints including
low birth weight (LBW), small for gestational age (SGA), 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., considered good in the outcome domain) that were measured in multiple studies to allow for
an evaluation of consistency and any heterogeneity across studies that may be present). Some of
the adverse endpoints of interest examined here included fetal growth restriction endpoints based
on birth weight such as mean birth weight reductions (or variations of this endpoint such as
standardized birthweight z-scores), as well as categorical measures such as SGA births (e.g., lowest
decile of birthweight stratified by gestational age and other covariates) and LBW (i.e., typically
defined as <2,500 grams). Overall, birthweight measures are considered very accurate and, in these
studies, were derived predominately from medical records; therefore, the outcome domain
judgments reflect the high reliability of these endpoints. Sufficient details on the SGA percentile
definitions and stratification factors as well as sources of standardization for z-scores were
necessary for these endpoints to be considered good. LBW is a less preferred measure of fetal
growth restriction than SGA, especially if analyses include both term and preterm neonates. This is
because birth weight is dependent on both the rate of fetal growth and gestational duration, and
perturbation in each may arise from different etiologies.

Gestational duration measures were examined in epidemiological studies as either
continuous (i.e., per each gestational week) or dichotomized categorical endpoints such as preterm
birth (typically defined as gestational age <37 weeks). Although gestational age dating methods,
such as ultrasounds early in pregnancy are preferred, this and other approaches (e.g., last

This document is a draft for review purposes only and does not constitute Agency policy.

3-100 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

menstrual period recall], are expected to result in some decreased sensitivity as measurement
error could impact classification of SGA as well as PTB. Gestational duration measures were,
therefore, downgraded if based solely on last menstrual period estimates or if the method(s) were
not reported, and less uncertainty is anticipated in studies that compare and adjust for differences
between last menstrual period and ultrasound measurements. Any sources of error noted in the
classification of these endpoints are anticipated to be non-differential with respect to PFNA
exposure and, therefore, would not be considered a major concern for risk of bias, but could impact
precision and study sensitivity. Other measures of fetal growth may be subject to measurement
error (e.g., head circumference and body length measures) if the measures are less reproducible
(i.e., are subject to more interobserver differences). Thus, unless multiple measurements were
taken, these endpoints were given a rating of adequate fShinwell and Shlomo. 20031 Additional
details for domain-specific evaluation of epidemiological studies can be found in the PFAS
Systematic Review Protocol, Appendix A.

Growth Restriction - Neonatal Anthropometric Measures
Birth Weight

Legend

I Good (metric) or High confidence (overall)
~ Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
I Critically deficient (metric) or Umnformative (overall)

Figure 3-24. Study evaluation results for twenty-nine epidemiological studies of birth
weight and PFDA. Refer to HAWC for details on the study evaluation review: HAWC Human
Birth Weight.

"Confidence descriptors based on the mean birth weight or birth weight z-score endpoints.

This document is a draft for review purposes only and does not constitute Agency policy,

3-101 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

As shown in Figure 3-24, 29 different studies examined birth weight measures (either mean
BWT differences or standardized BWT scores) in relation to PFDA exposures. One uninformative
study fLee etal.. 20161 due to several critical study deficiencies in confounding, selection
participation and study sensitivity is not considered further below. Among the 28 included studies
based on maternal, umbilical cord or placental measures, eight reported standardized BWT
measures such as BWT z-scores (Gardener etal.. 2021: Wikstrom etal.. 2020: Workman etal.. 2019:
Xiao etal.. 2019: Gvllenhammar etal.. 2018: Meng etal.. 2018: Bach etal.. 2016: Wang etal.. 20161
with all but two f Gardener etal.. 2021: Xiao etal.. 20191 of these reporting both standardized and
mean BWT measures (see Figure 3-25). Twenty-six studies examined mean BWT either in the
overall population (i.e., both girls and boys) or both sexes including four fHall etal.. 2022: Lind et
al.. 2017a: Wang etal.. 2016: Robledo etal.. 20151 that reported sex-specific analyses only.

Fourteen studies in total reported sex-specific results in both sexes.

28 PFDA Perinatal Studies of Birth Weight (BWT) included in synthesis

20 Studies reporting
ONLY Mean BWT

6 Studies reporting BOTH
Mean BWT and standardized
BWT measures

2 Studies reporting
ONLY standardized
BWT measures

26 Different Mean
BWT Studies

15 Studies with Sex-Specific

BWT Results {5 High; 6
Medium; 3 Low Confidence)

22 Studies with Overall 4 Studies with Sex-

Population Results (6 Specific Results

High; 8 Medium; 8 Low Only
Confidence)

11 Studies with Sex-
Spec 1 lie and Overall
Population Results

Figure 3-25. Twenty-eight perinatal studies of birth weight measures and
subsets considered for different analyses.

Twenty-two of the 28 studies examining either standardized or mean BWT were
prospective birth cohort studies, while the remaining six (Xu etal.. 2019b: Gvllenhammar et al..
2018: Li etal.. 2017: Shi etal.. 2017: Callan etal.. 2016: Kwon etal.. 20161 were cross-sectional
studies (see Figure 3-25). For evaluation of patterns, studies that collected biomarker samples
concurrently or after birth were considered to be cross-sectional analyses [e.g., fHall etal.. 20221],
Five of the 28 PFDA studies relied on umbilical cord samples fXu etal.. 2019b: Cao etal.. 2018: Li et
al.. 2017: Shi etal.. 2017: Kwon etal.. 20161. and the recent medium confidence study by fHall et al..
20221 based their exposure characterization on PFDA placental measures sampled at birth. Twenty
studies had maternal blood measures that were sampled preconception (Robledo etal.. 20151 or
during trimester one (Buck Louis et al.. 2 018: Lind etal.. 2017a: Bach etal.. 20161. trimester three
(Gardener etal.. 2021: Luo etal.. 2021: Yao etal.. 2021: Kashino etal.. 2020: Gao etal.. 2019: Xiao et

This document is a draft for review purposes only and does not constitute Agency policy.

3-102 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

al.. 2019: Valvietal.. 2017: Callanetal.. 2016: Wangetal.. 20161 across multiple trimesters (Chenet
al.. 2021: Hiermitslev etal.. 2020: Wikstrom etal.. 2020: Workman etal.. 2019: Meng etal.. 2018:
Starling etal.. 2017: Woods etal.. 2017: Lenters etal.. 20161. after delivery f Gvllenhammar etal..
20181. The study by Mengetal. T20181 pooled samples from umbilical cord and multiple maternal
samples during trimesters 1 and 2.

Ten of the 2 8 included studies examining different B WT indices were rated high confidence
(Gardener etal.. 2021: Luo etal.. 2021: Yao etal.. 2021: Wikstrom etal.. 2020: Xiao etal.. 2019:

Buck Louis etal.. 2018: Lind etal.. 2017a: Valvi etal.. 2017: Bach etal.. 2016: Wang etal.. 20161.
while 10 were medium confidence fHall etal.. 2022: Chen etal.. 2021: Hiermitslev etal.. 2020:
Kashino etal.. 2020: Gvllenhammar etal.. 2018: Meng etal.. 2018: Woods etal.. 2017: K won etal..
2016: Lenters etal.. 2016: Robledo etal.. 20151 and eight were low confidence fGao etal.. 2019:
Workman et al.. 2019: Xu etal.. 2019b: Cao etal.. 2018: Li etal.. 2017: Shi etal.. 2017: Starling etal..
2017: Callan et al.. 20161. Among the 28 studies with mean BWT measures, 14 each had adequate
and deficient study sensitivity (see Figure 3-24). The evidence syntheses for mean BWT differences
detailed below primarily emphasizes the results from the twenty high or medium confidence
studies.

Standardized BWT Measures

Three of the eight studies reported smaller standardized BWT scores in relation to PFDA
exposures including one medium (Gvllenhammar et al.. 20181 and two high (Wikstrom etal.. 2020:
Xiao etal.. 20191 confidence studies (see Figure 3-26). One study not plotted by Gardener et al.
f20211 reported positive associations with increasing PFDA exposures, while four studies reported
null associations fWorkman etal.. 2019: Meng etal.. 2018: Bach etal.. 2016: Wang etal.. 20161. One
of the studies showing a null association in the quartile 4 (relative to quartile 1) and per each ln-
unit increase did show elevated but non-significant BWT scores of -0.10 and -0.13 for quartiles 2
and 3 (Bach etal.. 20161. Two of the studies (Wikstrom etal.. 2020: Gvllenhammar etal.. 20181 with
inverse associations in the overall population reported statistically significant BWT z-scores similar
in magnitude ((3 range: -0.14 to -0.15 per each ln-unit increase). The high confidence (Xiao etal..
20191 study reported associations about twice as large as these other studies ((3 = -0.39; 95%CI: -
0.94, 0.16) and were largely driven by associations in girls ((3 = -0.62; 95%CI: -1.28, 0.03) (see
Figure 3-27). One fWikstrom etal.. 20201 of two studies with categorical data showed evidence of
an inverse exposure-response relationship.

Study sensitivity did not seem to explain the four null studies as two were adequate (Bach
etal.. 2016: Wang etal.. 20161 and two were deficient (Workman etal.. 2019: Meng etal.. 20181. No
pattern in study results by exposure contrasts was evident either. There may be some evidence of
potential impact of pregnancy hemodynamics, as two of these three studies were based on later
biomarker samples.

This document is a draft for review purposes only and does not constitute Agency policy.

3-103 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

9 P [change in standardized BWT]

Bach etal . 2010.
3981534

Aarhus Birth Cohort (2008-2013).
Denmark. 1507 mother-infant pairs

IHvN

Adequate

Cohort
(Prospective)

Trimester 1-2

-0.1

Quartile 2

• , I

t—195% confidence interval

•0.13

Quartile 3

1 • \

0.02

Quartile 4

i —i

0.03

In-unt (ngfaO
Increase

H#H

W.kstrom et al..
2020.6311077

SELMA (2007-2010). Sweden. 1533
mother-infant pairs

|H«h|

Adequate

Cohort
(Prospective)

Trimester 1-2

-0.077

Quartile 2

-0085

Quartile 3

- • r1

•0.179

Quartlie 4

•—•—•.

•0 147

Imunt (ns-fnt)
Increase

Wang et al. 2016.
3858502

Taiwan Maternal and infant Cohort
Sludy (2000-2001), 223
mother-infant pairs

IHlohl

Adequate

Cohort
(Prospective)

Tnmester3

0.04

ln-unil (ng/mL)

X.»t> «l «t„ 2019,
S9ia«o9

l-auin Inlands (1994-1986).

WhN

IJnlkanril

Coin* l

(PlOSpRCliVRl

lnimralH 3

•0.39

In-tin 1 (nii'rnl)

i 9,1

Gyllenhammur el
al„ 2018,4236300

POPUP (1996-2011), Sweden, 381
mothw-infari! pairs

|Medlurn|

Oeficjenl

Cross-sediorwl

I
I

-0.1S

liwmil (ngrtm.)
irtciaasa

•—•—•!

Mangwlal., 201B,
4829851

ONBC (1996-2002!, D»rirn»rk, 353b
mother-infant pairs

|M«diuni|

DttfiCHtnl

(Prospective;

IliflMSlet 1-2

-0.014

In-unil (ngftri)

Workman el al..
2019.5387046

Canadian Healliiy Infant Longitudinal
Development (CHILD) Study
(2010-2012), Canada. 414
mother-infant pairs

|Low|

Defidant

(Prospective)

Trimester 2-3

-0.064

Iri-unil (ny-inL)

-0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 08 1

Figure 3-26.PFDA and birth weight z-scores (overall population)3. Refer to
Birth Weight-Z for details on the individual study evaluation review.

Abbreviations: BWT = Birth Weight

aStudies are sorted first by overall study confidence level then by Exposure Window examined.

Regression Exposure
Coefficient Comparison

, Taiwan Maternal and Infant Cohort |High|

Sluriy (2000-2001), 223
mother-infant pairs

-0.068
¦0-015
¦0.101
-0.132
0.04

Quorate 2
Quartile 3
Ouaiiik)4

Regression coefficient

0 |1 fdiamjB in slnmii»rdi/txi BWTl
© p [change in standardized BWT] p<0.0S
(—195% confidence Interval

2020. 6311677

. Taiwan Maternal and Infant Cohort |H>gh|

Study (2000-2001), 223
mother-infant pairs

-0167 Quartile 3
-0268 Quartile 4

-0.163 In-unit ing-iriL)

-0.14 In-unit (ngftnL)

¦0.62

ill (ntj-'ml)

Figure 3-27.PFDA and birth weight z-score (sex-stratified)a. Refer to Birth
Weight-Z Score Sex-Stratified for details on the individual study evaluation review

Abbreviations: BWT = Birth Weight

3Studies are sorted first by overall study confidence level then by Exposure Window examined.

1 Overall Population Results

2 Twenty-two studies (6 high and 8 each medium and low confidence) examined mean BWT

3 differences in the overall population (see Figure 3-28). Although some of these were not

4 statistically significant, 11 of the 22 studies reported some deficits including 4 high, 5 medium, and

5 2 low confidence studies. Eight studies in the overall population were null fChen etal.. 2021: Buck

6 Louis etal.. 2018: Mengetal.. 2018: Shi et al.. 2017: Starlinget al.. 2017: Woods etal.. 2017: Bach et

7 al.. 2016: Callan et al.. 20161 and three others reported increased mean BWT with increasing PFDA

8 exposures fGao etal.. 2019: Xu etal.. 2019b: Cao etal.. 20181. Five of the six studies with categorical

This document is a draft for review purposes only and does not constitute Agency policy,

3-104 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

data did not show definitive BWT deficits; however, the one study that reported deficits did
demonstrate an exposure-response relationship in the overall population fWikstrom etal.. 20201.

There was considerable variability in BWT deficits ((3 range: -29 to -101 g) per ln-unit
increases, with eight studies ranging from 29 to 72 grams. The high confidence study by Luo et al.
(20211 showed a statistically significant larger BWT deficit ((3 = -96.8 g; 95%CI: -178.0, -15.5 per
each ln-unit PFDA increase). For each ln-unit PFDA increase, statistically significant reductions
similar in magnitude were reported by the medium confidence studies by Swedish Environmental
Protection Agency f20171 ((3 = -94 g; 95%CI: -163, -25) and Kwon etal. T20161 f[3 = -101 g;
95%CI: -184.8, -17.7). The medium confidence study by Kashino et al. f20201 reported a large
deficit between PFDA exposure in the overall population ((3 = -31.4 g; 95%CI: -60.0, -2.7 per each
ln-unit increase). For each ln-unit PFDA increase, smaller non-statistically significant BWT deficits
in two high confidence studies by Yao etal. (20211 f[3 = -46.3 g; 95%CI: -131.1, 38.5) and Valvi et
al. (20171 ((3 = -59 g; 95%CI: -147, 26). The medium confidence study by Lenters etal. (20161
detected a BWT deficit ((3 = -31 g; 95%: -75,12 for each ln-unit PFDA increase) in single pollutant
multivariate models, although PFDA was not selected as an important independent predictor in
their multi-pollutant elastic net model adjusting for other PFAS exposures and phthalate
metabolites (see more details in Appendix F). The associations noted in many studies were evident
despite some limitations, such as low exposure levels and/or narrow contrasts which can decrease
study sensitivity and statistical power. In contrast to the medium and high confidence studies
which exhibited associations in the overall population, there was more heterogeneity in the low
confidence studies often noted by imprecision. Overall, 10 of the 22 studies of the overall

population with mean BWT data were deficient in study sensitivity given very low PFDA ranges and
median values (from 0.08 to 0.24 ng/mL) (see Table 3-20); this included 5 of the 8 null studies
fMeng etal.. 2018: Shi etal.. 2017: Starling etal.. 2017: Woods etal.. 2017: Call an etal.. 20161. Two
(Buck Louis et al.. 2018: Bach et al.. 20161 of the remaining three null studies also reported low
median and IQR values (0.20-0.30); thus, study sensitivity may partially explain some of these null
associations given the limited exposure contrasts.

Sex-Specific Results

Although they were not always consistent across sexes within each study, most studies
showed some mean BWT deficits in either or both sexes (see Figure 3-29). For example, nine
studies each in girls and boys showed some BWT reductions in relation to PFDA, including six out of
eleven medium and high confidence studies in boys and seven out of eleven medium and high
confidence studies in girls. Null associations were reported in two studies each for boys (Meng et
al.. 2018: Robledo etal.. 20151 and girls (Hiermitslev etal.. 2020: Swedish Environmental
Protection Agency. 20171. while increased BWT were reported in three studies in girls fCao etal..
2018: Lind etal.. 2017a: Shi etal.. 20171 and boys fCao etal.. 2018: Bach etal.. 2016: Wang etal..
20161.

This document is a draft for review purposes only and does not constitute Agency policy.

3-105 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Males

Among the five (2 high, 2 medium, and 1 low confidence) studies showing BWT deficits in
both sexes, three studies reported larger mean BWT deficits in boys fHall etal.. 2022: Kashino etal..
2020: Valvi etal.. 20171 while two did in girls fWikstrom etal.. 2020: Li etal.. 20171. The deficits
across sexes were quite variable per each unit change in PFDA exposures; with mean BWT deficits
ranging from - 20 g (Hiermitslev etal.. 20201 to -156 g (Swedish Environmental Protection Agency.
20171 in boys. Smaller per ln-unit PFDA changes of-24 g were noted in two studies f Kashino etal..
2020: Meng etal.. 20181 for girls compared to very large changes of -140 g fWang etal.. 20161 and -
254 g observed in Robledo etal. f20151. The medium confidence study by Hall etal. f20221
reported non-significant deficits only in tertile 3 for boys ((3 = -73.2 g; 95% CI: -307.2,160.8) and
girls ((3 = -50.3 g; 95%CI: -185.3, 84.7) relative to tertile 1.

Females

The high confidence study by Wang etal. T20161 reported a mean birth weight decrease
among girls only ((3 = -140 g; 95% CI: -260, -20) per each In increase. Among these girls, they also
reported large mean BWT deficits in PFDA quartiles 3 ((3 = -120 g; 95%CI: -330,100) and 4 ((3 =
-230 g; 95% CI: -440, -10) compared to the Q1 referent The high confidence study by fWikstrom
etal.. 20201 reported an exposure-response relationship among girls with BWT deficits ranging
from -42 to -116 g but only in quartile 4 ((3 = -27 g; 95%CI: -118, 64) for boys. Although deficits
were not seen in the high confidence (Bach etal.. 20161 study among 743 girls based on continuous
exposure expressions, large non-monotonic deficits were noted across all three upper PFDA
quartiles. In contrast, their sister publication (not shown on Figure 3-29) by Bierregaard-Olesen et
al. f20191 did report BWT deficits of 43 g (95%CI: -102,16) per each ln-unit increase in a subset of
334 girls

Overall, there was limited patterns in results across sexes or across study characteristics.
Among the studies showing mean BWT associations, six of nine studies in girls and five of nine
studies in boys were based on biomarker samples later in pregnancy or post-partum. This might be
indicative of potential bias related to pregnancy hemodynamics. Study sensitivity was limited in
half the studies but did not appear to explain the four null studies (two each were adequate and
deficient).

BWT Summary

Eighteen of 28 studies examining mean or standardized BWT measures in the overall
population or each sex including 17 of the 26 studies examining mean BWT measures. Eleven of 22
studies (and 9 of 14 medium and high confidence) examining mean BWT in the overall population.
Although there was not a clear sex-specific effect of PFDA, eight studies each in girls and boys
showed some mean BWT reductions; four studies showed deficits in both sexes. Few studies
examined non-linear relationships between PFDA and mean BWT. The lone study that reported

This document is a draft for review purposes only and does not constitute Agency policy.

3-106 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

deficits across categories, demonstrated an exposure-response relationship for mean BWT, while
one of two studies showed this for standardized BWT measures.

Eleven of the 22 studies of the overall population were deficient in study sensitivity with
very low PFDA contrasts which may partially explain some of these null associations. For example,
among the eight null studies examining mean BWT measures in the overall population, there was a
slight preponderance of deficient study sensitivity (five compared to three with adequate study
sensitivity). There was a definitive pattern by sampling timing as only two of the eleven studies
(including two of nine medium/high studies) reporting BWT deficits in the overall population had
early sampling biomarkers measures during pregnancy. The majority of sex-specific studies
reporting BWT deficits were also based on later biomarker sampling (defined here as trimester 2
exclusive onward).

Although the collective evidence is fairly consistent of an association between BWT and
PFDA there is considerable uncertainty given that pregnancy hemodynamic factors related to
sample timing may explain some of the reported BWT deficits.

This document is a draft for review purposes only and does not constitute Agency policy.

3-107 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Comparison

Regression coefficient

9 I? [change in mear BWT (g)]
© |1 [change in mean BWT =t)

* ' l

-31.36

In-unit (ng/mL)
increase

i • *

Ceo at Hi., 2013.
5080197

Zhoukou City Longitudinal Birth
Cohort (2013-2015), China, 282
mother-infant pars

|Law|

Deficient

(Prospective)

Atairth

122.5

Tartila 2

i —

Li et al.. 201 ?,
3981358

GBCS (2013). China, 321
mother-infant pairs

|Low|

Deficient

Al biltti

-4/.3

In-unit (riy/inL)

l • , i

Shi ctaL 2017.
3827535

Haidan Hospital (2012). China. 170
mother-infant pairs

|UM

Deficiont

Cross-sectional

Al birth

-1.3

In-unit (ng/mL)

Xu etal„ 2019,

Cross-sectional study (2016-2017).

|Low|

Deficient

Cross-sectional

Atairth

91.5

In-unit (ng/mL)

, i

5381338

China, 98 mother-nfan: pairs

Starling et a!.,
2017,3858473

Healthy Start cohort (2009-2014),
United States, S28 mother-infant

|Low|

Defioent

(Prospective)

Trimester 2-3

0.4

Quartite 2

11.5

ln-uni! (ng/mL)

Workman ot al.,
2019.5387046

Canadian Healthy Infant Longitudinal
Development (CHILD) Study
(2010-2012). Canada. 414

|Low|

Deficient

Cohort
(Prospective)

Trimester 2-3

ln-uni: (ng/mL)

i ,• i
1

Calian etal.,

AMETS (2008-2011), Australia, SS

|Low|

Deficiant

Cross-sectional

Trimester 3

In-unit (ng/ml)

2016,3858524

mother-Infant pairs

Gaoetal., 2019,
5387135

University '.2015-2016), China, 132
molher-inlanl pairs

|LaW|

Adeouate

(Prospective)

Tri master 3

29.9

Tertila 2

Terlile 3

300 -253 -200 -150 -100 -50 0 50

100 150 200 250 300

Figure 3-28. Overall study population mean birth weight results for 22 PFDA
epidemiological studies3*. (results can be viewed by clicking the HAWC link).

Abbreviation: BWT = Birth Weight

a Studies are sorted first by overall study confidence level then by Exposure Window examined.

b Meng et al. (2018) pooled samples from umbilical cord blood and maternal plasma during the first and second
trimesters. The remaining studies were all based on either one umbilical or maternal sample.

c- If a study presented regression coefficients for continuous exposure with multiple exposure units, only one unit
change is shown (e.g., (Bach et al,, 2016), with the exception of (Li et al., 2017), which displays both IQR and In-
unit (ng/mL) values.

d- The results displayed here for mean birth weight among 587 overall population participants in the POPUP Cohort
are from a larger population of participants (Swedish Environmental Protection Agency, 2017) compared to a
sample size of 381 in their 2018 publication Gvllenhammar et al. (2018).

e- Xu et al. (2019a) results are truncated for the 210.7 gram increase; the complete 95% CI ranges from -314.3 to
735.8 grams.

This document is a draft for review purposes only and does not constitute Agency policy,

3-108 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

KKTSSSSSS. m

ass?""

- 3a£3«"

SHS5T
ssssfc sswassitfr

— SSSr
aw-"* ea^aatea

iaf sstsxs? ™
as'"' sr.

JSSfSBSStBK

a1** svKsr^r*-
S?S3S SfflB3B»fc—"

SFMfc gSSgjST-

sfetteK-

"S£S°

iii*

Figure 3-29. Sex-specific mean birth weight results for 14 PFDA epidemiological
studies: boys are above reference line, girls are below a s. (results can be viewed by
clicking the HAWC link).

Abbreviation: BWT = Birth Weight

a Studies are sorted first by overall study confidence level then by Exposure Window examined.

b Meng et al. (2018) pooled samples from umbilical cord blood and maternal plasma during first and second
trimesters. The remaining studies were all based on either one umbilical or maternal sample.

c- If a study presented regression coefficients for continuous exposure with multiple exposure units, only one unit
change is shown.

d- The results displayed here for mean birth weight in the POPUP Cohort are from a larger population of
participants (Swedish Environmental Protection Agency. 2017) compared to a sample size of 381 in their 2018
publication Gvllenhammar et al. (2018).

This document is a draft for review purposes only and does not constitute Agency policy,

3-109 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

e- (Robledo et al„ 2015) regression coefficients for maternal serum PFDA are displayed. The complete 95% CI for
the male-8.4 gram difference ranges from -434.3 to 417.6 grams; the complete 95% CI for the female -254.4 gram
difference ranges from -766.7 to 258.1 grams.

f (Wang et al.. 2016) quartile results are truncated; the complete 95% CI for the -230 gram difference (Quartile 4)
ranges from -440 to -10 grams. Quartile results reported for females only.
s For evaluation of patterns of results, we considered studies that collected biomarker samples concurrently or
after birth to be cross-sectional analyses (e.g., Hall et al. (2022)).

Small for Gestational Age and Low Birth Weight

Participant selection

Exposure measurement

Outcome ascertainment

Confounding
Analysis -
Sensitivity
Selective Reporting -
Overall confidence -

Legend

I Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
I Critically deficient (metric) or Uninformative (overall)

Figure 3-30. Low birth weight/small for gestational age heatmap. Results can
be viewed by clicking the HAVVC link.

Five epidemiological studies included here examined associations between PFDA exposure
and different dichotomous fetal growth restriction endpoints, such as SGA (or related intrauterine
growth retardation endpoints) fWikstrom etal.. 2020: Xu et ah, 2019a: Wang et al.. 20161 or low
birth weight (LBW) fHiermitslev etal.. 2020: Mengetal.. 20181. Two studies were high confidence
fWikstrom et al.. 2020: Wangetal.. 20161. two were medium confidence fHiermitslev et al.. 2020:
Meng etal.. 20181 and one study was low confidence fXu etal.. 2019al. Three of these studies had
adequate study sensitivity fHiermitslev et al.. 2020: Wikstrom et al.. 2020: Wangetal.. 20161 while
two were deficient fXu etal.. 2019a: Meng etal.. 20181 (Figure 3-30).

T wo fWikstrom etal.. 2020: Wang etal.. 20161 of three SGA studies showed some adverse
associations, while one study was null fXu etal.. 2019a) (Figure 3-31). The high confidence study by
Wangetal. f20161 reported a statistically significant increased odds ratio (OR) (3.14;

95%CI: 1.07, 9.19) for SGA per each ln-unit PFDA increase among females. Increased risks were not

This document is a draft for review purposes only and does not constitute Agency policy.

3-110 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

detected among males (OR = 0.71; 95%CI: 0.33,1.52). The medium confidence study by (Wikstrom
etal.. 20201 showed that PFDA was associated with SGA based on a continuous measure (OR = 1.46;
95%CI: 1.06, 2.01 per each ln-unit increase), as well as categorical exposures (Q4: OR = 1.50;
95%CI: 0.94, 2.38 compared to Q1 referent). Results were stronger among females (OR = 1.62;
95%CI: 0.98, 2.67) than males (OR = 1.36; 95%CI: 0.90, 2.07) per each ln-unit increase.

Two studies reported relatively small ORs that were not statistically significant between
PFDA and risk of LBW, while another study showed an 80% increased risk of very LBW per each ln-
unit increase. The medium confidence study by fMeng etal.. 20181 reported a larger risk (OR = 1.8;
95%CI: 0.9, 4.0 per each ln-unit increase) for a very LBW (i.e., <2,260 grams) measure compared to
the typical LBW definition of <2,500 grams (OR = 1.3; 95%CI: 0.7, 2.15). There was also no
evidence of increased risk across PFDA quartiles or an exposure-response relationship, but the
study may have been impacted by sparse cell bias. A nonsignificant increased odds (OR = 1.15;
95%CI: 0.57, 2.33) was reported in the medium confidence study by Hiermitslev et al. (20201 per
each PFDA ln-unit increase.

SGA/LBW Summary

Although they were not always statistically significant, three fWikstrom etal.. 2020: Meng
etal.. 2018: Wang etal.. 20161 of the five studies examining either SGA, LBW, or very LBW showed
some increased risks with increasing PFDA exposures. There was no evidence of an exposure-
response relationships based on categorical data in one SGA and one LBW study. The relative risks
reported in the two LBW studies based on either categorical or continuous exposures (per each unit
increase) were consistent in magnitude (OR range: 1.2-1.3), while a larger risk was found (1.8) for
the very LBW endpoint SGA results were more variable based on sex-specific findings but both
studies showed larger risks among females. Two of the three studies with stronger results were
based on early biomarker sampling.

This document is a draft for review purposes only and does not constitute Agency policy.

3-111 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Population

998 ZOOZX 0«n mark, 3!

ACCEPT birth cohort (ant
2013 20*5). Gicur.lynJ.
rnoliiti-lnUirl piii's

Regression coefficient

Qu4ttfc>2

i-unit (ng-'mL)
ii "nil iiVJ'nil)

NOwbOt-» 111-3?;
Newborn* (n_37)
Neatboms (n=3?)
Newborns (n=37)

# SGA'LBiV Kelalvo
0 SUA-'LBW Rslalive
I—(05?s corfdence Hi

Quarto 2 Newborns <-»»1S33j

Ou«r.to3 Newborn* (n»1533)

QuaisHo4 Nuwbwiw (n-1533)

In unit fn^'mL) Newborns (n-l&ss)
ln-unil (no'mL) Naaborw (n—©9 J

Tuwan Maternal aid Intern Conor!
Study (2C0C-2OO') 223
motifer-Wsri pairs

Quanta 2 Newborn ooyt (i"801)

Quarto 3 Newborn ooyn (n«801)

Quarto 4 Newborn ooys (n»801)

In-unil tng-'mL: Newborn boy* (n-0i)
ireiww

In-unil (ng>'mL) Newborn toys (n=t17>

QuaiHo2
Qu.-tr-|l<' a
Quarto 4
In-unit (ng>mL)

(•-732)

Figure 3-31. Dichotomous fetal growth restriction (small for gestational age
and low birth weight) forest plotad. Results can be viewed at the HAWC link.

Abbreviations: SGA = Small for Gestational Age; LBW = Low Birth Weight

a- Studies are sorted first by overall study confidence level then by Exposure Window examined.

b- Low birth weight overall population data above blue reference line.

c- Overall population SGA data above Black reference line; sex-stratified SGA data below reference line.
d- Sex-Stratified SGA; Boys Above Dotted Line, Girls Below.

Birth Length Measures

_,l\V jr . „( _

Participant selection

Selective Reporting

Overall confidence

Adequate (metric) or Medium confidence (overall)
- | Doficiont (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)

Figure 3-32. Study evaluation results for 17 epidemiological studies of birth
length and PFDA. Refer to HAWC for details on the study evaluation review;

HAWC Human Birth Length.

Seventeen studies examined the relationship between PFDA exposures and mean or
standardized birth length measures including 15 studies that examined changes in the overall

This document is a draft for review purposes only and does not constitute Agency policy,

3-112 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

population and ten that examined sex-specific changes (see Figure 3-32). Two of these 10 reported
sex-specific analyses only fWang etal.. 2016: Robledo etal.. 20151. Most of the studies reported
mean birth length differences in relation to PFDA exposures, but two reported standardized birth
length measures across the sexes only fGyllenhammar etal.. 20181 or in both sexes as well as the
overall population (Xiao etal.. 20191.

Six of the 17 studies examining birth length measures in relation to PFDA were classified as
high confidence (Luo etal.. 2021: Xiao etal.. 2019: Buck Louis etal.. 2018: Valvi etal.. 2017: Bach et
al.. 2016: Wang etal.. 20161. four were medium confidence fChen etal.. 2021: Hiermitslev etal..
2020: Kashino etal.. 2020: Robledo etal.. 20151. and seven were low confidence fGao etal.. 2019:
Workman etal.. 2019: Xu etal.. 2019b: Cao etal.. 2018: Gvllenhammar etal.. 2018: Shi etal.. 2017:
Callan etal.. 20161 (see Figure 3-32). All but one of the ten medium and high confidence studies
were considered to have adequate study sensitivity, whereas the remaining six low confidence
studies were classified as deficient

Birth Length: Overall Population

Four (1 high and 3 low confidence studies fBierregaard-Olesen etal.. 2019: Xu etal.. 2019b:
Cao etal.. 2018: Callan etal.. 20161 of the 15 studies examining the overall population reported
increased birth length in relation to PFDA, while six studies were null fHiermitslev etal.. 2020:
Kashino etal.. 2020: Gao etal.. 2019: Buck Louis et al.. 2 018: Shi etal.. 2017: Valvi etal.. 20171 (see
Figure 3-33). Five (2 high, 1 medium and 2 low confidence) of the 15 studies reported reduced birth
length in the overall population. The high confidence study by Buck Louis etal. (20181 also did not
show an association for birth length and PFDA in the overall population. They did report that each
standard deviation increase in PFDA was associated with reductions in upper arm length ((3 = -0.09
cm; 95% CI: -0.14, -0.04); these were largely due to associations detected among White ((3 =
-0.21 cm; 95% CI: -0.31, -0.11) and Asian neonates ((3 = -0.15 cm; 95% CI: -0.25, -0.05). They also
reported reductions in upper thigh length ((3 = -0.14 cm; -0.21, -0.07) in the NICHD cohort with
the largest associations detected among White ((3 = -0.32 cm; 95% CI: -0.45, -0.19) and Asian
neonates ((3 = -0.18 cm; 95% CI: -0.30, -0.06).

The high confidence study by Xiao etal. T20191 reported similar birth length z-scores in the
overall population ((3 = -0.49; 95%CI: -1.00, 0.01), girls ((3 = -0.46; 95%CI: -1.07, 0.14), and boys ((3 =
-0.53; 95%CI: -1.17, 0.10). The high confidence study by Luo etal. f20211 showed a non-significant
birth length deficit ((3 = -0.23 cm; 95%CI: -0.64, 0.19) per each ln-unitPFDA increase. The medium
confidence study by Chen etal. (20211 detected a statistically significant birth length deficit (-0.27
cm; 95%CI: -0.53, -0.01 per each ln-unit increase), while the low confidence study by Workman et
al. (20191 reported a nonsignificant birth length deficit ((3 = -0.3 cm; 95%CI: -0.8, 0.2 per each ln-
unit increase). A small but precise deficit of 0.19 cm (95%CI: -0.36, 0.02 per each ln-unit) was
reported in the low confidence study by Gvllenhammar etal. T20181 for their standardized birth
length measures.

This document is a draft for review purposes only and does not constitute Agency policy.

3-113 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Among these five studies (two high, one medium and two low confidence) showing some
evidence of birth length deficits, there was limited evidence of exposure-response relationships
with only study fChen etal.. 20211 examining categorical data showing deficits in quartile 4 only (-
0.46 cm; 95%CI: -0.91, -0.01). There was a preponderance (four of five studies) of birth length
reductions in the overall population from studies based on later sampled biomarkers which may be
indicative of an impact of pregnancy hemodynamics. Study sensitivity did not seem to explain null
results, as five of these six studies had adequate ratings.

Birth Length: Sex-specific Results

Among the 10 studies with sex-specific results, seven different ones (4 high, 3 medium
confidence) showed some evidence of birth length deficits in relation to PFDA. This included four
studies each in girls and in boys (see Figure 3-34). Only the high confidence study by Xiao et al.
(2019) noted above found reduced standardized birth length measures in both girls and boys.

Three studies in girls were null (Hiermitslev etal.. 2020: Kashino etal.. 2020: Robledo etal.. 2015)
and two fCao etal.. 2018: Valvi etal.. 20171 showed increases in birth length with increasing PFDA
exposures. Four studies fChen etal.. 2021: Cao etal.. 2018: Shi etal.. 2017: Wang etal.. 20161 in
boys were null and two fHiermitslev etal.. 2020: Bierregaard-Olesen etal.. 20191 showed slight
non-significant increases in birth length.

In addition to the high confidence study by Xiao etal. (2019) noted above, three other
studies reported any suggestion of smaller birth length among boys. The medium confidence study
by Robledo etal. (2015) reported a non-statistically significant reduction among boys ((3 =
-1.15 cm; 95% CI: -3.65, 0.96 per each ln-unit based on maternal serum measures). Smaller birth
length deficits per each PFDA ln-unit increase were detected in the high confidence study by Valvi
etal. T20171 ((3 = -0.23 cm; 95%CI: -0.68, 0.22) and the medium confidence study by Kashino et al.
("20201 ((3 = -0.16 cm; 95%CI: -0.38, 0.07).

Including the Xiao etal. (2019) data above, four of the 10 studies in females reported some
birth length reductions. The high confidence study by Wang etal. (2016) reported non-statistically
significant deficits were detected among girls in quartile 4 ((3 = -0.75 cm; 95% CI: -2.09, 0.59) and
for each PFDA ln-unit increase ((3 = -0.47 cm; 95% CI: -1.23, 0.30). Birth length deficits similar in
magnitude ((3 = -0.44 cm; 95%CI: -0.79, -0.09 per each ln-unit PFDA increase) were detected
among girls in the medium confidence study by Chen etal. f20211. Smaller birth length changes
were detected among girls ((3 = -0.22 cm; 95%CI: -0.86, 0.43) in the Aarhus Birth Cohort
Bierregaard-Olesenetal. (2019) study.

Among the 10 studies in total, 7 different ones reported some evidence of sex-specific
associations between PFDA and reduced birth length, including 4 studies each in girls and in boys.
Few patterns were evident across study characteristics and study sensitivity did not appear to be
an explanatory factor for null studies. Sample timing also did not appear to be a strong determinant

This document is a draft for review purposes only and does not constitute Agency policy.

3-114 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

of the sex-specific study results as four of the seven different studies reporting reductions were
based on later biomarker sampling.

Birth Length Summary

Overall, ten different studies among the 16 in total showed some evidence of birth length
deficits in relation to PFDA exposures in either the overall population or in one/both sexes. Five (2
high, 1 medium, and 2 low confidence) of the 14 studies examining the overall population reported
birth length deficits that were consistent in magnitude (mean birth length deficit range: 0.19 to 0.30
per each unit increase). Birth length changes were a bit more variable in the seven studies (four
high, three medium confidence) that reported sex-specific deficits. Although three studies reported
mean birth length reductions around 0.20 cm, the remaining sex-specific studies ranged from -0.44
to -1.15 cm per each ln-unitPFDA increase. Four (2 high and 2 medium confidence) of the 10
studies each in boys and girls (3 high and 1 medium confidence) reported birth length deficits in
relation to PFDA.

Although some of these studies reported large differences in birth length, there was no
direct evidence exposure-response relationships in the few studies with categorical data. However,
the Wang etal. f20161 analysis in girls did show some large gradients in birth length among the
upper two quartiles. The fChen etal.. 20211 also showed larger deficits in quartile 4 relative to
quartile 1. Few patterns were evident across study characteristics, and study sensitivity did not
appear to be an explanatory factor for null studies. Seven of ten different studies reporting some
birth length reductions in the overall population (four of five) and across sexes (four of seven) were
based on later biomarker samples. This may be indicative of potential bias due to the impact of
pregnancy hemodynamics and adds some uncertainty.

This document is a draft for review purposes only and does not constitute Agency policy.

3-115 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure Window Regression
Coefficient

Exposure
Comparison

Regression coefficient

Bjerragaard-Olesen

Aaitius Birth Cohort (2038-20-3;.
Denmark, 702 mother-infa'it pairs

|H,gh

Adequate

(Prospective)

Trimester 1-2 0.2

In-unit (ng'mLl

5063648

Buck Louis et al.,
2018,

NICHD Fetal Growth Studies
(2000-2013), United States. 2106
rnolhur-Muril paiis

|H,gh

(Pr^SL)

Tnmester2 -0.082

In-un it (ng.'mL)

1 •-J-*

Luo et St., 2021.

Zfcujiang Hospital Cohort
(2017-2019), China, 221
motrtci-inrant pairs

I High

Adequate

(Prospective)

TnrreBter 3 -0.23

In-unit ¦ng.'mLJ

• |4 [change id mean 31
-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Exposure
Comparison

Regression coefficient

Bjoriegaaru-Oleson Aarhus Birth Cahort (2008-2013).
et al.. 2019. Denmark. 702 mother-infant pairs

5063648

Taiwan Maternal and Infant Cotvort
Sludy (2000-2001), 223
mother-infant pairs

ACCEPT birth cohort (2010-2011,
2013-2015), Greenland. 482
mcilhar-iritanl pairs

Hokkaido Study on Environment and
Children's Heafth (2003-2009).
Japan. 1985 mother-infant pairs

Zhoukou City Longitudinal Birtfi
Cohort (2013-2015). China. 282
mother-infant pairs

Shi et al., 2017.

3827535

BjerregaarS-Olesen Aaitius Birth Cohort (2008-2013),
' I.. 201D, Denmark. 702 inothcr-infant pairs

5083648

'.Vang ct al., 2016, Taiwan Maternal and Infant Cohort
3858502 Study (2000-2001). 223

mother-infant pairs

IHighl
IHighi

IMediumj

|Low|

|High|
IHighi

2020.6311632

ACCEPT birth cohort (2010-2011,
2013-2015), Greenland, 482
mother-infant pairs
Hokkaido Sludy on Environment ar
Children's Health (2003-2KS).

Zhoukou City Longitudinal Birth
Cohort (2013-2015), China, 282
mother-infant pairs

IHighl

IMediurnj
iMedlumi
IMediumi

I |Medlum|

|Low|

Adequate
Adequate

Adequate
Adequate
Deficient

Adequate
Adequate

Deficient

Adequate
Adequate
Adequate

Trimester 3
Trimester 3

(Prospective)

Cohort
(Prospective)

Cross-sectional

-0.2
-0.75
-0.47

-0 462

0.095

In-unit (ng/mL)
In-unit (ng/mL)

In-unit (ng/mL)
increase
In-unit (ng/mL)

'n-unMng^L'

In-unit (ng/mL)

Tortile 3
l-unlt (ng/mL)

In-unit (ng/mL)
increase
Quartilc 2

Q-oartile 3
O jartile 4
In-unit (ng/mL)

In-unit (ng/fnL)

In-unit (ng/mL)

In-unit (ng/mL)
increase

In-unit (ng/mL)
In-unit (ng/mL)

Cross-sectional

t—

# 6 [change in mea

BL(g>]

0 B [change in men

Bl. (g» p<0.05

H 95% confidence ii

Figure 3-34. Sex-stratified birth length results for 10 PFDA epidemiological
studies ab. Results can be viewed by clicking the HAWC link.

Abbreviation: BL= Birth Length

a- Studies are sorted first by overall study confidence level then by Exposure Window examined.
b- Xiao et al. (2019) reports birth length z-score data.

This document is a draft for review purposes only and does not constitute Agency policy,

3-117 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Head Circumference

Fourteen studies examined PFDA levels in relation to head circumference including five
high confidence studies fBierregaard-Olesen etal.. 2019: Xiao etal.. 2019: Buck Louis et al.. 2018:
Valvi etah. 2017: Wang etal.. 20161 and five medium confidence studies (Chen etal.. 2021:
Hiermitslev etal.. 2020: Kashino et al.. 2020: Lind et al.. 2017a: Robledo et al., 20151 (see Figure 3-
35). The four low confidence studies (Workman etal.. 2019: Xu et al.. 2019b: Gvllenhammar et al..
2018: Callan et al.. 20161 as well as fXiao etal.. 20191 were considered deficient in the study
sensitivity domain largely due to low exposure levels and/or narrow contrasts. The remaining nine
medium and high confidence studies had adequate ratings in the sensitivity domain.

Participant selection -I
Exposure measurement -I
Outcome ascertainment -

Overall confidence -I

a' a*' * *.»'''

•

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)

•

¦ gg *

•

«¦

•

Figure 3-3 5.Study evaluation results for 14 epidemiological studies of head
circumference and PFDA. Refer to HAWC for details on the study evaluation
review: HAWC Human Head Circumference.

One study provided standardized head circumference data fXiao etal.. 20191. while the
other 13 included in Figures 3-36 and 3-37 are based on mean head circumference differences.
Eight studies examined sex-specific results in both boys and girls fHiermitslev etal. 2020: Kashino
etal.. 2020: Bierregaard-Olesen etal.. 2019: Xiao etal.. 2019: Lind etal.. 2017a: Valvi etal.. 2017:
Wang etal.. 2016: Robledo et al. 20151 including three with sex-specific data only (Lind et al..
2017a: Wang et al.. 2016: Robledo et al, 20151. Eleven studies reported head circumference results
in the overall population (Chen etal.. 2021: Hiermitslev etal.. 2020: Kashino etal... 2020:
Bierregaard-Olesen et al.. 2019: Workman etal.. 2019: Xiao etal.. 2019: Xu etal.. 2019b: Buck Louis
etal.. 2018: Gvllenhammar et al.. 2018: Valvi etal.. 2017: Callan et al.. 20161.

This document is a draft for review purposes only and does not constitute Agency policy,

3-118 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Head Circumference-Overall Population Results

Only 2 of 11 studies in the overall population showed head circumference associations with
PFDA exposures. The medium confidence study by Hiermitslev etal. f202 01 reported a
nonsignificant decreased head circumference in the overall population (p = -0.15 cm; 95%CI: -0.37,
0.07 per each ln-unit increase). Slightly smaller but precise head circumference deficits were
reported in the medium confidence study by Kashino et al. (20201 f(3 = -0.10 cm; 95%CI: -0.24,
0.003 per each ln-unit PFDA increase). In contrast, nonsignificant increased head circumference in
the overall population was reported in relation to PFDA in three studies fChen etal.. 2021:
Workman et al.. 2019: Valvi et al.. 20171. No associations were reported between PFDA exposures
and mean or standardized head circumference measures in 6 of the 11 studies based on the overall
population, including three studies each with high fBierregaard-Olesen etal. 2019: Xiao et al..
2019: Buck Louis et al.. 2 018) and low confidence fXu etal.. 2019b: Gvllenhammar etal.. 2018:
Callan etal.. 20161 (see Figure 3-36).

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

# £ [change in mean HC (cm)]
© [5 (change in mean HC lcm)J pO.05
M 05% confidence Interval
I—| HC Z-Score

B|«r recjaard-OleHer

etal.. 2019.
bDB3S4B

Aaihua Bit III Cohurl (2008-2013),
Denmark. 702 mother-infant pairs

Hlgh|

Adequate

Cohort
(Prospective)

TilmasU* 1-2

liMjnil (njjfinL;

Buck Louis et al..
21518, 5016992

NICHD Fetal Growth Studies
(2009-2013), Uniter! States. 2106
mother-infant pairs

High|

Adequate

Cohort
(Prospective)

Trimester 2

-0.03

in-unit (ng.'mL)
increase

H^i-I

Cohort

in-unlt (ng.'mLj

3963872

Donmark. 604 mother-infant pairs

(Prospective)

Xiao etal., 2019.

Faroa Islands (199'l-1995), Faroe

Deficient

Cohort

Trimester 3

-0.101

In-unit (nq.'mL)

5913609

Islands, 172 mother-infant sairs

(Prospective)

Increase

Chen etal. 2021,

Shanghai Birth Cohort (2015-2017),

Medium|

Adequate

Cohort

Trimester 1-2

0.21

'n-unit (ng.'mL:

726398S

China. 214 mother-infant pairs

(Prospective)

incoaso

Hiermitslev et al,,
2020, 5880849

ACCEPT birth cohort {2010-2011.
2013-2015), Greenland, 482
mother-infant pairs

Meriium|

Adequate

Cohort
(Prospective)

Trimester 1-3

-0,15

in-unit (ng.'mL!

I <11

Kashino ct al.,
2020,6311032

Hokkaido Study on environment and
Children's Health (2003-2009).
Japan, 1985 mother-infant pairs

Modiuml

Adoquato

Cohort
(Prospective)

Trimester 3

-0.104

rn-unit (ng.'mL;

Gyilenhainmar et
al. 2018, 42383'JQ

POPUP (1996-2011), Sweeten. 381

|Low|

Deficient

Cross-sectonal

3 weeks post-birth

-0.04

ln-unit (ngWiL)

1 *T 1

Xu etal.. 2019.

Cross-sectional study (2016-2017),

|Low|

Deficient

Cross-sectional

At birth

-0.074

ln-unit (ng.'mL)

Workman et al..
201B, 5387046

Canadian Healthy Infant Longitudinal

|Low|

Deficient

Cohort

Trimester 2-3

0.17

m-unit i'ng.'mL}

(2010-2012). Canada, 414
mother-Maul pairs

Callan etal.,

AMETS (2008-2011). Australia. D8

|Low|

Deficient

Crass-sectional

Trimester 3

-0.07

ln-unit (ng.'mL)

2016, 3H5«5?4

motfier-infanl pairs

increase

-1 -0 8 -06 -0 4 -0.2 6 02

04 06 o!8 1

Figure 3-36. Overall population head circumference results in 11
epidemiological studies*®. Refer to the HAWC link.

Abbreviation: HC = Head Circumference

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b- Xiao et al. (2019) reports head circumference z-score data.

Head Circumference-Sex-specific Results

Among the eight studies fHiermitsIev et al.. 2020: Kashino et al.. 2020: Bierregaard-Olesen
etal. 2019: Xiao et al., 2019: Land etal. 201.7a: Valvi etal.. 2017: Wang etal.. 2016: Robledo etal,
20151 reporting sex-specific head circumference results in both male and female neonates, three
studies in girls and one in boys reported reductions with increasing PFDA exposures (see Figure 3-
37). The Lind etal. f20'17al study reported an exposure-response relationship based on PFDA
quartiles (range: -0.1. to -0.3 cm) in boys, but there was not much evidence of associations when
scaled to each ln-unit increase (p = -0.10 cm; 95%CI: -0.5, 0.3). The high confidence study by Wang

This document is a draft for review purposes only and does not constitute Agency policy,

3-119 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

etal. f20161 detected a non-significant decrease ((3 = -0.37 cm; 95% CI: -0.85, 0.10 per each ln-unit
increase) in mean head circumference only among girls. The medium confidence study bv Robledo
etal. f20151 reported larger but very imprecise head circumference reductions for girls (|3 =
-0.62 cm; 95% CI: -2.4,1.2 per each ln-unit PFDA increase}. The high confidence study by
Bierregaard-Olesen etal. f2019) showed a smaller non-significant result (j3 = -0.22 cm; 95%CI: -
0.65, 0.22 per each ln-unit increase]. In contrast, one high fValvi etal.. 20171: p = 0.51 cm;
95% CI: 0.13, 0.90} and one medium fLind etal.. 2017a): (3 = 0.3 cm; 95% CI: -0.1, 0.7) confidence
study each reported increased birth length for female neonates for each ln-unit PFDA increase, as
did the Bierregaard-Olesen etal. f20191 study (P = 0.19 cm; 95%CI: -0.19, 0.38 each ln-unit
increase) in males. Null associations were reported per each ln-unit increase in five studies in boys
fHiermitslevetal.. 2020: Kashino etal. 2020: Xiao etal.. 2019: Valvi etal.. 2017: Wang etal.. 20161
and three in girls fHiermitslev et al.. 2020: Kashino etal. 2020: Xiao et al.. 2019).

Four of eight available studies reported some head circumference reductions among boys
or girls including three that were based on early biomarker samples. In addition to the Unci et al.
f2017al study noted in boys above, four null studies examining different head circumference
measures in relation to continuous exposures reported non-significant and imprecise deficits
around -0.1 cm per each unit increase for either or both sexes fHiermitslev etal.. 2020: Kashino et
al.. 2020: Valvi etal. 2017: Wang etal.. 20161.

Overall Study Study Si
Confidence

SH«-| Adfc

eiti vity Design Expo&un

ot at, 2017, Farce Islam

531«»S>
2CI5.2SC1I97

anew Ma '.anal and Want Co-orl

'C'her-irrsnt pairs

Dei nark. 838 aether-intent pairs

IWeoWtr
(t'^aurr

Adequate

iPmHcive)

iProspeoive)

m-un*

Irvunrt ing.Y-_>

IV etal., ACCEPT bir.h cohi

ct 1-3 -0 03

Valvi ot at, 2017. Fo-'x stones :-->:<7.2000;
(J v. OI. 2016 Tiiwan Ma mini and Infant O

IPriftpK'Jw)

ln-umt ing-'r'->

H-1W5; Faroe
)V UniteC
Odcnse Chili) Cohere :20" 3-2012:.

Mantis, 172-null
Ll-E Study tl

IHghl

fctaSurr

ft'exSurr

I -0X5IS

to; G-oonlmd, ^82

Hekfcsoo 3:uc

Figure 3-37. Sex-stratified head circumference results in 8 epidemiological
studies a-b. Refer to the HAWC link.

Abbreviation: HC = Head Circumference

This document is a draft for review purposes only and does not constitute Agency policy,

3-120 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Xiao et al. (2019) reports head circumference z-score data.

Head Circumference-Summary

There was limited evidence of associations between PFDA and head circumference with 6
(2 high and 4 medium confidence) out of 14 studies reporting reductions in head circumference in
the overall population or either or both sexes. These reductions were reported in the over one-half
(6 of 10) of the high and medium confidence studies. Very limited evidence was found in the overall
population with only 2 of 11 studies. Four of eight sex-specific studies reported some head
circumference reductions with three of these occurring among female neonates. Four of these six
studies that reported some head circumference reductions in the overall population or either sex
were based on early biomarker sampling during or prior to pregnancy. In contrast to the null sex-
specific studies where only one of five studies had deficient study sensitivity, nearly all (five of six)
null studies in the overall population were rated as deficient Narrow exposure contrasts in many
studies of PFDA, likely limited statistical power and may have precluded the ability to detect
statistically significant associations that are small in magnitude.

Fetal Growth Restriction Summary

Eighteen of the 28 studies examining different BWT measures in relation to PFDA measures
in the overall population or either/both sexes, reported some evidence of associations. This
included 11 different studies (and nine of fourteen medium and high confidence) out of 22
examining mean BWT in the overall population. There was considerable variability in BWT deficits
(P range: -29 to -101 g per ln-unit increases) in the overall population, with seven studies ranging
from 31 to 59 g deficits per each ln-unit increase. These deficits were seen despite low exposure
levels and contrasts in many studies (Table 3-20). For example, among the nine medium and high
studies reporting it, the PFDA IQR in the overall study populations ranged from 0.07 to 0.37 ng/mL
and the median levels ranged from 0.11 to 0.55 ng/mL. Few studies examined nonlinear
relationships between PFDA and mean BWT. The lone study that reported deficits across
categories, demonstrated an exposure-response relationship for mean BWT, while one of two
studies showed this for standardized BWT measures. Twelve of the 13 studies reporting sex-
specific results showed some evidence of BWT deficits in either or both sexes. However, there was
not a clear sex-specific effect of PFDA. Eight studies each in girls and boys showed some reductions
and only four studies showed deficits in both sexes.

Although there was no evidence on an exposure-response relationships in the few studies
with categorical data, the majority of studies reporting results for either SGA, LBW, or very LBW
showed some increased risks with increasing PFDA exposures. Relative risks generally were fairly
modest in magnitude ranging from 1.2 to 1.8, with more variable and larger risks for SGA results
denoted among females.

This document is a draft for review purposes only and does not constitute Agency policy.

3-121 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Results were more mixed for birth length and head circumference. Ten of 16 studies in total
including reported birth length in relation to increasing PFDA exposures. This included 5 (2 high, 1
medium and 2 low confidence) of the 14 studies that reported birth length deficits fairly consistent
in magnitude in the overall population (range: 0.13 to 0.30 per each unit increase). In comparison
to the overall population results, birth length changes were more variable in the 10 studies that
examined stratified results by sex. Four out 10 studies each in boys (2 high and 2 medium
confidence) and girls (3 high and 1 medium confidence) birth length deficits in relation to PFDA.
The four studies that showed birth length deficits in girls were generally more consistent in
magnitude; one study reported mean birth length reductions of 0.20 cm and, the three others
ranged from -0.44 to -0.75 cm per each ln-unit increase. There was no direct evidence exposure-
response relationships in the few birth length studies with categorical data; but one analysis in girls
did show some large gradients in birth length among the upper two quartiles.

Five (2 high; 3 medium) out of 14 overall PFDA studies reported reductions in head
circumference in the overall population (2 of 11 studies) or either sex (3 of 7 studies); the 5 studies
showing some reductions accounted for one-half of the 10 high and medium confidence studies.
Head circumference results were less consistent but were comparable in magnitude to those seen
for birth length. The one analysis in boys that reported head circumference reductions did show an
exposure-response relationship. Although not monotonic across all quartiles, another study in girls
did show some large gradients in head circumference among the upper two quartiles.

Few explanatory factors were consistently identified by general study characteristics across
the FGR endpoints including exposure levels, study sensitivity, and sex differences. Given limited
exposure contrasts in many of the studies, this likely precluded sufficient statistical power to detect
associations small in magnitude and, especially when stratified by sex. There was a definitive
pattern by sampling timing as only two of the eleven studies (including two of nine medium/high
studies) reporting BWT deficits in the overall population had early sampling biomarkers measures
during pregnancy. Although there was no pattern in the sex-specific studies of birth length, most
(four of five) of the studies reporting some birth length reductions in the overall population were
based on later biomarker samples. The opposite was seen for studies of head circumference with
three of four studies in the overall population or either sex based on early samples. Although these
patterns were not consistent across endpoints, the dearth of birth weight and length results in the
overall study populations based on early or prepregnancy measures might be indicative of potential
bias due to the impact due to pregnancy hemodynamics on PFDA levels. Despite fairly consistent
evidence of an association between PFDA and different BWT-related measures, and more mixed for
other endpoints, there is considerable uncertainty given that some sample timing differences may
explain some of the reported fetal growth restriction deficits examined here.

This document is a draft for review purposes only and does not constitute Agency policy.

3-122 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-20. Summary of 33 studies (from 35 publications) of PFDA exposure in relation to fetal and postnatal
growth restriction measures sorted by overall confidence3

Author (Year)

Study

location/

Years

Exposure
median/IQR (range)
in ng/mL

SGA/LBW

Birth weight

Birth length

Postnatal
measures
(Wt, Ht)

High Confidence Studies

Wang et al. (2016)

Taiwan,
2000-2001

223

0.46/0.56-Boys
0.43/0.48-Girls
(0.16-1.57)

tsGA (Girls)*

0 (Boys)

-Girls*b

0 Boys

-Girls
0 Boys

- Wt Girls*

+ Wt Boys
-Ht Girls*

- Ht Boys

Bierregaard-Olesen
et al. (2019); Bach et
al. (2016)

Denmark,
2008-2013

1533

0.30/0.20
(
-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Author (Year)

Study

location/

Years

Exposure
median/IQR (range)
in ng/mL

SGA/LBW

Birth weight

Birth length

Postnatal
measures
(Wt, Ht)

Yaoetal. (2021)

China, 2010-
2013

369

0.55/0.37 (0.09-3.77)

-All

Gaoetal. (2022)

China, 2013-
2016

1350

1.82/1.44(0.21, 26.6)

0 RWG-Wt

(All)*
si RWG-Wt

(Girls)
t RWG-Wt

(Boys)
t RWG-Ht

(All)*
t RWG-Ht

(Girls)
t RWG-Ht
(Boys)

Starling et al. (2019)

USA, 2009-
2014

1410

0.1/0.1 (N/A)

0 -Wt
tRWG-Wt
+ Adiposity
(All/Boys/Girls)

Zhang et al. (2022)

China, 2013-
2016

2395

1.72/1.38(0.21, 27.8)

0 All
0 Girls
0 Boys

Medium Confidence Studies

Robledo et al. (2015)

MI/TX, USA,
2005-2009

234

0.45-Boysd
0.40-Girlsd
(N/A)

-Girls
0 Boys

0 Girls
- Boys

-Girls
0 Boys

Lenters et al. (2016)

Ukraine/
Poland/
Greenland,
2002-2004

1,321

0.16-0.40
(0.07-1.18)
range across 3
countries

-All

This document is a draft for review purposes only and does not constitute Agency policy.

3-124 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Author (Year)

Study

location/

Years

Exposure
median/IQR (range)
in ng/mL

SGA/LBW

Birth weight

Birth length

Postnatal
measures
(Wt, Ht)

Gvllenhammar et al.
(2018): Swedish
Environmental
Protection Agencv
(2017)e

Sweden,
1996-2001

381/587

0.24/0.14
(LOD-1.1)

-All*

0 Girls
- Boys*

-All

0 All

0 Wt, Ht

Woods et al. (2017)

OH, USA,
2003-2006

272

0.20/0.10
(0.2-0.3)f

0 All

Meng et al. (2018)

Denmark,
1996-2002

2,120

0.20/0.10
(N/A)

T LBW (All)
T VLBW (All)

- All
- Girls
0 Boys

Kwon et al. (2016)

S. Korea,
2006-2010

268

0.11/0.07
(0.04-0.41)

-All*

Chen et al. (2021)

China, 2013-
2015

214

1.73/1.47 (N/A)

0 All

- All*

- Girls*

0 Boys

Gao et al. (2019)

China, 2015-
2016

132

0.47 (LOD-3.15)

+ All

0 All

Hall et al. (2022)

0 Girls
- Boys

Hiermitslev et al.
(2020)

Greenland,
2010-

2011;2013-
2015

266

0.71/N/A (0.12-7.84)

t LBW (All)

-All
0 Girls
- Boys

0 All

-All
0 Girls
0 Boys

Kashino et al. (2020)

Japan, 2003-
2009

1591

0.6/0.5 (LOD-2.4)

-All
-Girls
- Boys

0 All

-All
0 Girls
0 Boys

Low Confidence Studies

Xu et al. (2019b)

China,
2016-2017

0.21/0.15
(0.1-0.58)f

0 SGA

+ All

This document is a draft for review purposes only and does not constitute Agency policy.

3-125 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Author (Year)

Study

location/

Years

Exposure
median/IQR (range)
in ng/mL

SGA/LBW

Birth weight

Birth length

Postnatal
measures
(Wt, Ht)

Li et al. (2017)

China, 2013

321

0.15/0.16
(LOD-2.12)

-All
-Girls
- Boys

Lee et al. (2018)

South Korea,
2012-2013

361

0.37/0.36
(0.04-1.25)

-Wt*,-Htb

Callan et al. (2016)

W. Australia
2003-2004

0.12/N/A
(0.03-0.39)

0 All

Cao et al. (2018)

China,
2013-2015

337

0.10/0.09
(0.04-0.22)g

+ All
+ Girls
0 Boys

0 All
+ Girls
0 Boys

0 All
-Girls
0 Boys

+ Wt Girls
- Wt Boys
0 Ht Girls
0 Ht Boys

Starling et al. (2017)

CO, USA,
2009-2014

598

0.10/0.10 c
(LOD-3.5)

0 All

Shi et al. (2017)

China, 2012

170

0.08/0.10
(LOD-0.60)

0 All
+ Girls
- Boys

0 All
0 Girls
- Boys

0 All
0 Girls
0 Boys

Jensen et al. (2020a)

Denmark,
2010-2012

589

0.26/N/A (N/A)

+ Adiposity
(All)

Workman et al.
(2019)

Canada, 2010-
2011

414

0.13/N/A (LOD-1.4)

-All

+ All

Abbreviations: LOD = limit of detection; N/A: not available; All = Overall population of boys and girls; IQR = interquartile range; HC = head circumference;
SGA = small for gestational age; LBW = low birth weight; VLBW = very low birth weight; Ht = height; Wt = weight; RWG = rapid weight gain.

Symbols: 0: null association; + : positive association; - : negative association; t: increased odds ratio; decreased odds ratio.

'Statistically significant findings based on p < 0.05.

aOverall confidence descriptor is for the birth weight endpoints when studies included prenatal and postnatal growth measures; four other studies had only
postnatal data Gao et al. (2022): Zhang et al. (2022): Starling et al. (2019): Lee et al. (2018).
bExposure-response relationships detected for categorical data.

CIQR calculated by subtracting the 25th percentile from the 75%; the 25th percentile estimated here as 0 given it was below the detection limit.
dRobledo et al. (2015) regression coefficients for maternal serum PFDA are displayed.

eSwedish Environmental Protection Agency (2017) results are displayed here for mean birth weight among 587 overall population participants in the POPUP
Cohort compared to a smaller sample size of 381 in the 2018 publication by Gvllenhammar et al. (2018).
f5th-95th percentiles.

This document is a draft for review purposes only and does not constitute Agency policy.

3-126 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

g10th-90th percentiles.

Note: "Developmental effects" indicated by increased odds ratio (t) for dichotomous outcomes, (+) for adiposity/body mass index and waist circumference,
and negative associations (-) for the other outcomes.

This document is a draft for review purposes only and does not constitute Agency policy.

3-127 DRAFT-DO NOT CITE OR QUOTE

-------
1

Growth Restriction -

Toxicological Review of Perfluorodecanoic Acid and Related Salts
Postnatal Growth (Infancy and earlv childhood up to 2 years of age )

Sensitivity

Soloctivo Reporting - ~
Overall confidence -

~Legend

Good (metric) or High confidence (overall)

~ Adequate (metnc) or Medium confidence (overall)
- Deficient (metric) or Low confidence (overall)

J Critically defioont (mothc) or Unmformative (overall)

~ Multiple judgments exist

Figure 3-38. Study evaluation results for four epidemiological studies of
postnatal growth and PFDAab. Refer to HAWC for details on the study evaluation
review: HAWC Human Postnatal Growth

a In Gvllenhammar et al. (2018), the outcomes height, weight, and body mass index are rated as Good, while the
outcome head circumference is rated as Adequate.

b In Starling et al. (2019), the outcome weight-for-age z-score (at 5 months) rated as Good, while the outcomes
length-for-age z-score and adiposity/fat mass at 5 months were rated as Adequate.

Eight studies were identified that assessed postnatal growth in relation to PFDA (see Figure
3-38) with each of these examining some measures of childhood weight and/or height in relation to
PFDA. Four studies were considered high fGao et al.. 2022: Zhang et al.. 2022: Starling etal.. 2019:
Wang etal.. 20161. one was medium fGvllenhammar etal.. 20181 and three were low confidence

flensen et al.. 2020a: Cao et al.. 2018: Lee et al.. 20181. Of the eight postnatal growth studies, four
each had adequate fGao etal.. 2022: Zhang etal.. 2022: Lee et al.. 2018: Wang et al.. 20161 and
deficient flensen et al.. 2020a: Starling et al.. 2019: Cao et al.. 2018: Gyllenhammar etal.. 20181
study sensitivity ratings largely owing to small exposure contrasts. Although there was some

This document is a draft for review purposes only and does not constitute Agency policy,

3-128 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

overlap across studies, limited serial measures during infancy as well as inconsistent age at
examinations and analyses may limit some comparisons here. For example, (Zhang etal.. 20221
examined growth up to 12 months and fStarling etal.. 20191 took measurements at 5 months only.
Both fWang etal.. 20161 and fLee etal.. 20181 examined postnatal growth at 2 years, while fCao et
al.. 20181 analyses were based on a mean of 19 months in participants. fGyllenhammar et al.. 20181
had serial measures of postnatal growth at 3, 6,12 and 18 months, flensen etal.. 2020al examined
different adiposity measures at 3 and 18 months, while (Gao etal.. 20221 examined growth
trajectory based on serial measurements at five time periods within the first 2 years (at birth, 42
days, 6 months, 12 months, and 24 months).

Postnatal Weight

Postnatal Weight: Overall Population

In the overall population, five postnatal studies (two high, one medium, and two low
confidence) examined PFDA in relation to either standardized (Zhang etal.. 2022: Starling etal..
2019: Gvllenhammar etal.. 20181 or mean weight measures in two low confidence studies (Cao et
al.. 2018: Lee etal.. 20181. All three standardized weight studies reported null associations
including fGyllenhammar et al.. 20181 for PFDA exposures and standard deviation scores (SDS) for
weight measured at 3 to 18 months (Figure 3-39). Similar to findings from f Zhang etal.. 20221
examining growth up to 12 months, fStarling etal.. 20191 also detected no difference in the overall
population at 5 months for either weight-for-age and weight-for-length z-scores across PFDA
tertiles and for each ln-unit increase.

Two low confidence studies examining mean weight differences in the overall population
during early childhood showed some deficits related to upper PFDA exposure (Figure 3-40) around
2 years of age. fCao etal.. 20181 reported a non-significant and imprecise postnatal weight change
(P= -130 g; 95%CI: -579, 319) in the overall population (mean age of examination of mean of 19
months) for tertile 3 (relative to tertile 1), but the opposite was seen for tertile 2. Despite their
limited exposure contrast, (Lee etal.. 20181 reported a no n-significant mean weight decrease at 2
years for each ln-unit PFDA increase ((3= -140 g; 95%CI: -310, 30) in the overall population. They
detected lower mean weight at 2 years across PFDA quartiles in a mono tonic fashion (e.g., (3 range:
-200 to -390 g). For example, they detected a statistically significant weight reduction ((3= -390 g;
95%CI: -770, -10) in quartile 4 (relative to quartile 1). Both studies were based on measurements
in children around 2 years of age.

This document is a draft for review purposes only and does not constitute Agency policy.

3-129 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

Zhang et at,.
2022- 9944433

Shanghai Birth Cohort (SBC)
(2013-2016), China. 239S
mother-infant pairs

|H0h|

Adequate

Cohort
(Prospective)

Trimester 2

0.007
0.009

fertile 2
Tertile 3

# p [change in weight Z-Scoro]
O P [cliariga in weight Z-scois] p«0.05
H 95% confidence interval

Starling ct al.,
2019, 5412449

Healthy Start Study (2009-2014),
United States, 1410 mother-infant
pairs

|Hgh|

Deficient

Cohort

Trimester 2-3

-0.01

In-unit (ng/mL)

' f 1

0,02

In-unit (ng/mL)

Gyllenhammar et

POPUP (1996-2011). Sweden, 381

|Medium|

Deficient

Cross-sectional

3 weeks post-birth

-0.087

In-unit (ng/mL)
lr-uiit (ng/mL)

al., 2018.4238300

mother-infant pairs

-0.082

increase

• 1 1

•0.066

In-uiit (ng/mL)

increase

¦0.055

In-unit (ng/mL)

_ 1

0 35 -0.3 -0.25 -0.2 -0.15

-0.1 -0.05 0 0 05 0 1 0.15 0.2 0.25

Figure 3-39. PFDA and postnatal growth-standardized weight measures
(overall population)3-0. Refer to the HAWC link.

a- Studies are sorted first by overall study confidence level then by Exposure Window examined.
b- Age at Outcome Measurement: Gyllenhammar et al. (2018) at 3 months, 6 months, 12 months, and 18 months
(ordered top to bottom); Starling et al. (2019) at 5 months; Zhang et al. (2022) between 42 days and 12 months.
c- Above the first blue line is Weight-for-Age Z-Score; between the two blue lines is Weight-for-Length Z-Score;
below the last blue line is standardized PNG weight.

Study

Population Overall Study Study Sensitivity
Confidence

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

Cao el al . 2018.
5080197

Zhoukou City Longitudinal Birth |Low| Deficient
Cohort (2013-2015). China, 282
mother-infant pairs

Cohort
(Prospective

At birth

140,8

-130.2

Tertile 2
Tertile 3

1 1 •-

• • 1

Lee etal.. 2018.
4238394

Environment and Development of |Low| Adequate
Children (EDC) Cohort. South Korea,

645 mother-child pairs

Cohort

0-2 years

130

Quartile 4

1 1 •"

-60

In-unit (ng/mL)
increase

1 9i 1

0 p [chang® in mean growth weight (g)J
Q p [change in mean growth weight (g)J p«0 05

•390

QuartBo4

H 95% confidence Interval

-140

ln-urut (ng/mL)

1 •—r<

Cao et al.. 2018.
5080197

Zhoukou City Longitudinal Birth |Low| Deficient
Cohort (2013-2015). China. 282
mother-infant pairs

Cohort
(Prospective)

At birth

23.8
-438.4

Tertile 2
Tertile 3

BOYS

l •

• 1—'

320,3

GIRLS

292

Tortile 3

-1.200 -1.000 -800

-600 -400 -200 0 200 400 600 800 1.000 1.200

Figure 3-40. PFDA and postnatal growth mean weight (in grams)a d. Refer to the
HAWC link.

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Age at Outcome Measurement: Cao et al. (2018) at 19 months; Lee et al. (2018 data measurements taken from
0-2 years is in black and at 2 years is in blue.

c- Overall population data above the black reference line; sex-stratified data below.
d- Sex-stratified: male infant data above the blue line; females below.

Postnatal Weight: Sex-specific

1 Four studies (three high and one low confidence) included PFDA sex-specific analyses with

2 one fCao etal.. 20161 reporting mean weight changes and three reporting standardized weight

3 measures fZhang et al.. 2022: Starling etal.. 2019: Wang et al.. 20161 (Figure 3-41). Two of the four

4 studies showed detected deficits in relation to PFDA albeit not consistent across sexes. The low

5 confidence fCao etal.. 20181 study detected imprecise contrasting changes in postnatal weight,

6 with non-significant decreases in the highest fertile for boys (p = -438 g; 95% CI: -980,103) but

7 increases among girls (p = 292 g; 95% CI: -501,1,085). Two of the sex-standardized weight studies

This document is a draft for review purposes only and does not constitute Agency policy,

3-130 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

reported null results for boys and girls based either on weight-for-age and weight-for-length
standardized measures (Figure 3-41). Starling et al. (20191 reported no difference in either sex at 5
months for weight-for-age and weight-for-length z-scores across PFDA tertiles or for each ln-unit
increase, as did Zhang etal. f20221 across PFDA tertiles for postnatal growth up to 12 months. In
contrast, Wang etal. f20161 detected statistically significant reductions among females only for
average childhood weight z-scores ((B = -0.32; 95% CI: -0.63, 0). No relationship was seen for age 2
for weight z-score in either sex, and the largest weight z-scores were among females detected at
birth and at age 11.

2022, <3&-l<1
-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

1 height in a monotonic fashion with the largest statistically significant weight difference detected in

2 quartile 4 ((3 = -1.11 cm; 95%CI: -1.86, -0.36).

Overall Study Study Sensitivity Design
Confidence

Exposure Regression
Window Coefficient

7twig et al , Shanghai Birth Cohort (SBC)

2022.9944433 (2013-2016}, Chm». 239S
mother-infant pars

Cohort Trimester 2

(Prospective)

Exposure
Comparison

-0.053
0.01

3 weeks post-birth -0.050
-0.052
-0.045
-0.038

Regression coofficioru

9 (3 [change in hoig-u Z-Scoro]
O P [change in hcignt Z Score] p<0.05
W 95% confidence interval

Figure 3-42. PFDA and postnatal growth - standardized height measures
(overall population)3 b. Refer to the HAWC link.

a- Studies are sorted first by overall study confidence level then by exposure window examined.
b- Age at Outcome Measurement: Gvllenhammar et al. (2018) at 3 months, 6 months, 12 months, and 18 months
(ordered top to bottom); Zhang et al. (2022 between 42 days and 12 months.

Study

Cao et al., 2018. Zhoukou City Longitudinal Birth
5080197 Cohort (2013-2015). China. 282

mother-inlani paire

Overall Study Study Sensitivity Design Exposure Regression Exposure

Confidence Window Coefficient Comparison

|Low| Deficient Cohort At birth 1.56 Tertile2
(Prospective)

Regression coefficient

Loo ct al., 2018, Environment and Dovclopmont of
4238394 Children (E DC) Cohort, South Korea,

645 mother-child pairs

Cohort 0-2 yoars

1.27
0.33

Tertile 3
OuartHo 4

-1.11
-0.44

9 (i [change in mean growth height (cm)]
0 & [change in mean growth height (cm)] p<0.0:
}—H 95% confidence interval

Cao et al.. 2018. Zhoukou City Longitudinal Birth
5080197 Cohort (2013-2015). China. 282

mother-infant pairs

Cohort Al birth
(Prospective)

2.45 Tortile 2 GIRLS

1.31 Tertile 3

-0 5 0 0.5

_JJ 2 L5 3 35 4

Figure 3-43. PFDA and postnatal growth mean height (in centimeters)3bcd

Refer to the HAWC link.

a- Age at Outcome Measurement: Cao et al. (2018) at 19 months; Lee et al. (2018) data measurements taken from
0-2 years is in black and at 2 years is in blue.

b- Sex-stratified data is located below the solid black line; boys are above the purple dotted line and girls are below.
c- Cao et al. (2018) female results have upper bounds that have been truncated; the upper bounds are 5.41 for
Tertile 2 and 4.5 for Tertile 3.

3 Postnatal Height: Sex-specific

4 Three studies (two high and one low confidence) examined height in relation to PFDA

5 across sexes including one fCao et al., 20181 examining mean and two studies examining

6 standardized measures (Zhang etal.. 2022: Wang etal., 20161. The low confidence study by (Cao et

7 al.. 20181 reported larger postnatal mean height increases among females (|3 range: 1.31 to 2.45

8 cm) than males ((3 range: 0.61 to 1.07 cm) across PFDA tertiles. The high confidence study by Zhang

9 etal. f20221 reported null associations for both sexes based on continuous and categorical PFDA
10 exposures (Figure 3-44). In contrast, the high confidence study by Wang etal. f20161 detected

This document is a draft for review purposes only and does not constitute Agency policy,

3-132 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

statistically significant reductions among females only for childhood height z-scores averaged from
birth, 2, 5, 8, and 11 years ((3 = -0.52; 95% CI: -0.80, -0.24). Smaller height z-scores were found for
all time periods for both male and females but was only statistically significant for females at ages 2
and 11. For example, they reported height z-score reductions ((3 = -0.61; 95%CI: -1.02, -0.23 per
each ln-unit PFDA increase) at age 2 among females and was much smaller among males ((3= -0.17;
95%CI: -0.63,0.30).

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

Zhang et al..
2022. 9944433

Shanghai Birth Cohort (SBC)
(2013-2016). China. 2395
mother-infant pairs

|High|

Adequate

Cohort
(Prospective)

Trimester 2

0.02
0

ln-unit(ng/mL)
increase

Tertile 2

0 (3 [chango in haight Z-Scorc]
O P (change in height Z-Score] p<0.05
(—¦| 95% confidence interval

|—(#—| BOYS
1 • 1

0.033

Tortilo 3

i—r# 1

Wang at al., 2016.
3858502

Taiwan Maternal and Infant Cohort
Study (2000-2001), 223
mother-infant pairs

|High|

Adequate

Cohort
(Prospective)

Trimester 3

-0.172

ln-unit (ng/mL)

Zhang etal..
2022. 9944433

Shanghai Birth Cohort (SBC)
(2013-2016). China. 2395
mother-infant pairs

|High|

Adequate

Cohort
(Prospective)

Trimester 2

0.03
0.065
0.011

In-unit(ng.'mL)
increase

Tertile 2
Tcrtic 3

f—+#—1 GIRLS

T-*—4

Wang at al.. 2016.
3858502

Taiwan Maternal and Infant Cohort
Study (2000-2001). 223
mother-infant pairs

|High|

Adequate

Cohort
(Prospective)

Trimester 3

-0.614

ln-unit (ng/mL)

1.2 -1 -0.8

06 -0.4 -0.2 0 0.2 04

Figure 3-44. PFDA and postnatal growth - standardized height measures (sex-
stratified; boys above reference line, girls below)a c. Refer to the HAWC link.

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Age at Outcome Measurement: Dong et al. (2019) between 2-11 years; Zhang et al. (2022) between 42 days and
12 months.
c Boys above reference line, girls below.

Postnatal Head Circumference

Three studies (one high, medium, and low confidence study each) examined post-natal
standardized head circumference including two studies fZhang etal.. 2022: Gvllenhammar et al..
20181 that reported standardized results only in the overall population (Figure 3-45) and one fCao
etal.. 2018) that examined mean head circumference data in the overall population as well as
across sexes (Figure 3-46). None of the three studies examining head circumference showed much
evidence of decreases in head circumference with increasing PFDA exposures. Null results were
detected in Zhang etal. (2022) for postnatal head circumference-for-age Z score up to 12 months of
age per each ln-unit increase and across PFDA tertiles and for Gvllenhammar et al. f20181 head
circumference SDS measures were based on four different time points (3, 6,12 and 18 months). In
the overall population, Cao etal. f20181 detected a sizeable mean head circumference increase in
PFDA fertile 2 (0.50 cm; 95%CI: -0.44,1.44) but was null in fertile 3. Results were null for boys,
while contrasting results were seen for fertile 3 ((3= -0.69 cm; 95% CI: -2.26, 0.88) and fertile 2 ((3=
0.67 cm; 95% CI: -0.79, 2.13) among girls.

This document is a draft for review purposes only and does not constitute Agency policy.

3-133 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

# (3 [charge in HC Z-Scorc]
O P [charge in HC Z Score] p<0.05

Zhang eta,,
2022. 9944433

Shanghai Birth Cohort (SBC)
(2013-201S). Cur,a. 2395
mother infant pairs

lHign|

-0.026

Tertile 2

H 95% confderce interval

Prospective)

1 •—<

-0.031

Tcitic 3

' • r 1

-0.08

ln-ui>it (itgi'inLj

I • l—i

Gvlanhamma' et

POPUP (1996-2011 i, SxvBder 381

I Low

Optician:

Cross-sectional 3 v.'eeks posl-biilh

-0 03

In-unil {nq-'mU

al , 2018. 4238300

TKjlhui-infant oairs

¦0.011

mciadsu
In-uril (rq.'rnL]

0.020

increase

0.058

increase

•

P.i3-

-0.25 -0.2 -0.15 -0.1 -0.05 0 0.05

0.1 0.15 0.2 0.25 0.3

Figure 3-45. PFDA and postnatal growth standardized head circumference
(overall population)31*. Refer to the HAWC link.

Abbreviation: HC= Head Circumference

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Age at Outcome Measurement: Zhang et al. (2022) between 42 days and 12 months; Gvllenhammar et al. (2018)
at 3 months, 6 months, 12 months, and 18 months (ordered top to bottom).

Study

Population

Overall Study

Study Sensitivity

Design

Exposure

Regression

Exposure

Confidence

Window

Coefficient

Comparison

Regression coefficient

# 3 [change in HC Z-Swxa]

Zhang et al,,
2022 3944433

Shanghai Rrth Cohort (SBC)
(2013-20161 CI' mil 23S5
r-o'-her-irfarit osira

HON

Adequate

Cohort
(Ptuspodivu)

Trimester 2

-a.02

ln-iinit(ng?ml)

BOYS

O 3 tcftungo in HCZ-Suxu] p<0.05
J-H 95% confidence interval

1— • 1

0.062

TeftBoa

1 1 '+ <

Cau el al. 2013.
5080-97

Ztoukou City Longitudinal Birth
Coh-art (2013-2015), China. 282
mother-infant oairs

(Lewi

D«5o«rH

Cohort
(Prospective)

At b'lth

0.02

TtH .il« 2

0.02

Terile 3

Zhang et al_
2022.8944433

Shangvwi Birth Cohort (SBC)
(2013 2016) China 239ti
rroihw-irlarl obhm

|H>qh|

Adequate

Cohort

(Prospective)

Tritwister2

0.08
-COM

0.062

ln-unrt(n<}'"mL)
in crease
Terile 2
Terile 3

GIRLS

h—• 1

h-*—'

Cao etal. 2018.

SOW 07

Ztaxikou City LoiigKiKfcial Biitli
Cohort (2013-2015). China 282
n-o:h«r-irfant pairs

|Lw;|

Dvflc-cnt

Cohort
;Piospeclivej

At tliltll

0.6?

Tcitik?

Terile 3

—

-1 -0.8 -0.8 -0.4 -3.2 Q 0.2

C.d 0 8 0.3 I

Figure 3-46. PFDA and postnatal growth head circumference (sex-stratified;
boys above reference line, girls below)3*1. Refer to the HAWC link.

Abbreviation: HC= Head Circumference

a. Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Age at Outcome Measurement: Cao et al. (2018) at 19 months (averaged); Zhang et al. (2022) between 42 days
and 12 months.

c- Zhang et al. (2022) reports standardized results based on head circumference z-scores, while Cao et al. (2018)
reports mean head circumference data (in cm).
d- Cao et al. (2018) upper and lower bounds have been truncated. For boys, for Tertile 2 the bounds are [-1.23,
1.27] and for Tertile 3 the bounds are [-1.19,1.24], For girls, for Tertile 2 the bounds are [-0.79, 2.13] and for
Tertile 3 the bounds are [-2.26,0.88],

Adiposity Measures fWaist Circumference/ Body Mass Index/Ponderal Index)

Three studies (two high and one low confidence) examined post-natal adiposity measures
including % fat mass increase as well as standardized waist circumference, BMI, and ponderal index
measures. Two studies detected increased adiposity relative to PFDA exposures, while one study
(Zhang etal.. 2022) reported null associations for BMI-for-age Z score per each In-unitPFDA

This document is a draft for review purposes only and does not constitute Agency policy,

3-134 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

increase in the overall population and across sexes (Figure 3-47). lensen etal. (2020a) showed null
associations for PFDA and waist circumference SDS in the overall population and across both sexes.
However, they did report increased adiposity measures in the overall population including body
mass index SDS ((3 = 0.42; 95%CI: 0.01, 0.84 per each ln-unit increase) with stronger associations
among females ((3 = 0.58; 95%CI; -0.03,1.19 per each ln-unit increase). Similarly, a statistically
significant association with larger ponderal index SDS ((3 = 0.60; 95%CI: 0.18,1.02 per each ln-unit
increase) was detected in the overall population and was driven by associations in females ((3 =
1.02; 95%CI: 0.40,1.64 per each ln-unit increase). Starling et al. (2019) reported a slight non-
significant increase in infant adiposity at 5 months of age for each ln-unit increase in PFDA ((3 =
0.59% fat mass increase; 95%CI: -0.27,1.44) with larger increases among males ((3 = 0.79% fat
mass increase; 95%CI: -0.46, 2.04) compared to females ((3 = 0.44% fat mass increase; 95%CI:
-0.82,1.69). The opposite was seen in their categorical analyses dichotomized at the median with
more adiposity in females ((3 = 0.70% fat mass increase; 95%CI: -0.78, 2.17) and males ((3 = 0.23%
fat mass increase; 95%CI: -1.39,1.85).

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression coefficient

% |1 (change in adiposity measures]

O P (change in adiposity measures) s<0.0!

Stalling et al..
2019. 5412449

Healtfiy Stall Study (2009-2014),
United States. 1410 mother-infant
pairs

:High|

Deficient

Colli* t

TriineshM 2-3

0.59

ln-unit (ng/mL)

|—195% confidence inteival

OCC <2010-2012). Denmark. 613

|Medium|

Cohort

0.42

ln-unit (ng/ml)

2020.6833719

mother-infant pans

(Prospective)

Zhang etal..
2022, 99<1
-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Starling etal. (20191 reported null associations for rapid weight gain measures based on
weight-for-age z-score (OR = 0.87; 95%CI: 0.50,1.52 per each ln-unit PFDA increase) and a weight-
for-age standard deviation growth rate between birth and 5-month follow-up (Figure 3-48). They
did, however, report a non-significant increase in rapid weight gain derived from weight-for-Iength
z-score (OR = 1.50; 95%CI: 0.84, 2.70) for categorical exposures above the median (0.2-3.5 ng/mL
relative to the referent up to 0.1 ng/mL). In the overall population, Gao etal. f20221 reported null
associations between PFDA and their weight-for-age and weight-for-Iength z-score endpoints across
all trajectory designations. Based on the weight-for-Iength z-score, the low-rising participants (e.g.,
growth trajectory starts with a low value and followed by an increased trend afterward) vs.
moderate-stable referent group (e.g., growth trajectory starts with a moderate value and followed
by stable growth afterward) had a non-significant OR for the overall population (0.78; 95%CI: 0.53,
1.16 ln-unit PFDA increase). Results were contrasting in females (OR = 0.48; 95% CI: 0.27, 0.8 per
each ln-unit PFDA increase) and males (OR = 1.30; 95%CI: 0.71, 2.37 per each ln-unit PFDA
increase).

Gao etal. (2022) reported a non-significant increased risk (OR = 1.54; 95%CI: 0.85, 2.82 per
each ln-unit PFDA increase) for length-for-age z-score among those participants that were
considered high-rising vs. the moderate-stable group with comparable risks detected amongst male
and females (OR range: 1.73-1.83). In a weighted quantile sum mixture model, they also detected
higher odds (OR = 1.59; 95% CI: 0.90, 2.82 per each ln-unit PFDA increase) among the high-rising
group (vs. moderate-stable) based on length-for-age z-scores, with PFDA having the highest weight
among the PFAS mixtures. Gao etal. (2022) reported non-significant inverse associations
comparable in magnitude based on head-circumference-for-age z-score for high-rising vs. moderate-
stable (OR= 0.66; 95%CI: 0.38,1.12 per each ln-unit PFDA increase) and low-stable vs. moderate-
stable participants (OR = 0.67; 95%CI: 0.49, 0.93 per each ln-unit PFDA increase). They reported a
statistically significant inverse association (OR = 0.51; 95%CI: 0.27, 0.99 per each ln-unit PFDA
increase) for low-rising vs. moderate-stable groups in the single PFAS model. They reported a
lower risk in the weighted quantile sum model (OR = 0.37; 95%CI: 0.18, 0.72), with PFDA having
the highest weight among the PFAS mixtures. In general, the low- and high-rising groups examined
by Gao etal. (2022) may be at most risk for metabolic syndrome, as evidenced by changes in
obesity and other health effects later in life. However, results were not consistent in the overall
population or across sexes for these different rapid growth measures. Therefore, there is no
compelling evidence of increased postnatal weight gain among those that may represent low birth
weight individuals with rapid weight gain trajectories (i.e., low-rising group).

Based on mixed results within and across these two studies, there is limited support that
accelerated growth in infancy is related to PFDA. Although there was some evidence of increased
risks occurring in the high-rising trajectory group which may be indicative of rapid weight gain for
those that experienced fetal growth restriction, however, the evidence is scant and inconsistent to
draw many conclusions in the overall population or across sexes.

This document is a draft for review purposes only and does not constitute Agency policy.

3-136 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Healthy Start Study (2009-2014),
United States. 1410 mother-intant
pairs

Cohort Trimester 2-3

|Hiflh|
IHiflhl

Cohort Trimester 2-3

1.8

Population Description

ln-unil (ng/mL> Rapid Growth vs. Non-Rapid Growth OVERALL

Low-Rlstng vs Moderate Stable Trajectory

High-Rising vs Moderate Stable Trajectory

High-Stable vs Moderate Stable Trajectory

Low-Stable vs Modorato Stable Trajectory

Low-Rising vs Moderate Stable Trajectory

High-Rising vs Moderate Stable Trajectory

High-Stable vs Moderate Stable Trajectory

Low-Stable vs Moderate Stable Trajectory
increase

In-unit(ngfmL) Low-Rising vs Moderate Stable Trajectoiy
increase

In-unit(ng'mL) High-Rising vs Moderate Stable Trajectory

In-unit(ng.-mL) High-Stable vs Moderate Stable Trajectory
increase

ln-gnrl(ng/mL> Low-Stable vs Moderate Stable Trajectory
Quantile 2 Rapid Growth vs. Non -Rapid Growth

In-unit (ng/mL)

ln-unil (ng/mL)

In-unit (ngM.)

increase
ln-unil (ng/mL)

ln-unil (ng/mL)

0.78 In-unit (ng/mL)
0 eS ln-unil (ng/mL)
1.19 In-unit (ng/mL)

1.13

Low-Rising vs Moderate Stable Trajectory
High-Rising vs Moderate Stable Trajectory
High-Stable vs Moderate Stable Trajectory
Low-Stable vs Moderate Stable Trajectory

ln-unil (ng/mL)

increase
In-unit(ng.YnL)

In-onit(nflfmL)

increase
In-unit(ng.'mL)

In-unittna'mL)

Low-Rising v.
High-Rising v
High-Stable v

Low-Rising v.
High-Rising v
High-Stable v
Low-Stablo v

Moderate Stable Trajectory
i Moderate Stable Trajectory
. Moderate Stable Trajectory
Modorato Stable Trajectory
Moderate Stable Trajectory
i Moderate Stable Traiectory
i Moderate Stable Trajectory
Modorato Stable Trajectory

Regression coefficient

# [Odds Ratio for Rapid Growth)
O [Odds Ratio for Rapid Growth) p<0 05
H 9S% confidence interval

Figure 3-48. PFDA and postnatal growth rapid growth (overall population) and sex-specific (in grams)3 '. Refer to
the HAWC link.

a Studies are sorted first by overall study confidence level then by Exposure Window examined.

b Age at Outcome Measurement: Starling et al. (2019) at 5 months, Gao et al. (2022) modeled data (collected at 42 days, 6 months, 12 months, and 24
months).

c- Weight-for-Age Z-Score data above the black reference line; weight-for-length below.
d- Overall population data above the blue line; Sex-stratified data below.
e- Sex-Stratified data: male infants above the blue dash-dotted line; females below.

f- Quantile 2 in Starling et al. (2019) represents dichotomized exposure at median (quantile 1 referent: LOD-0.1 ng/mL; quantile 2: 0.2-3.5 ng/mL).

This document is a draft for review purposes only and does not constitute Agency policy,
3-137 DRAFT—DO NOT CITE OR QUOTE "

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Postnatal Growth Summary

Overall, there were mixed results within and across the eight available postnatal PFDA
studies of early childhood. For example, two (one high and one low confidence) out of five different
studies measuring height and three (one high and two low confidence) out of six different studies
measuring weight reported some deficits in relation to PFDA. Interestingly, there were more
consistent results seen in three studies (Cao etal.. 2018: Lee etal.. 2018: Wang etal.. 2016) that
examined postnatal growth measures at age 2. For example, both studies showing some postnatal
height deficits in either the overall population or across sexes were based on participants examined
at 2 years of age. There was no evidence of associations between PFDA exposures and early
childhood head circumference, but two (one high and one low confidence) of three studies showed
some suggestion of increased postnatal adiposity. Only two studies examined rapid weight gain in
relation to PFDA and were fairly inconsistent within and across studies based on different weight
and length measurements.

Only three of the eight total studies reported categorical data which may inform presence of
non-linearity or exposure-response relationships. Only one of these three studies showed any
evidence of any monotonic deficits across PFDA categories. There were a fairly small number of
studies across each common endpoints; thus, a lack of patterns across study characteristics (except
age at examination) was not unexpected. For example, although there were no studies with good
ratings for study sensitivity, this did not appear to be an explanatory factor for the null studies.
However, limited exposure contrasts and statistical power may have hampered the ability to detect
associations small in magnitude especially among the sexes.

In summary, although the evidence was mixed across various postnatal measures and
different examination windows, with only minimal evidence of exposure-response relationships to
support the continuous exposure scaled results. One challenge in evaluating consistency across
heterogeneous studies includes disparate periods of follow-up and assessment (e.g., childhood age
at examination). Despite the mixed evidence shown here, there was some suggestion of more
consistency in studies that examined postnatal growth measures around 2 years of age. This may
reflect the challenges that exist to detect associations in children that experience fetal growth
restriction and subsequent rapid growth periods. Although, there was limited information and
evidence of rapid weight associations among the two studies that considered this. Overall, the
evidence for postnatal associations is slight largely due to the early childhood weight and adiposity
results along with inconsistency across the other measures.

This document is a draft for review purposes only and does not constitute Agency policy.

3-138 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Ano genital distance

Christensen. 2021, 9960218

Lind. 2017, 3858512

Tian. 2019, 5390052

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 3-49. Study evaluation results for two epidemiological studies of
anogenital distance and PFDA. Refer to the interactive HAWC link for additional
details: HAWC Human AGP

Three medium confidence birth cohorts (Figure 3-49) in Denmark fLind etal.. 2017al. China
fTian et al.. 20191. and the Faroe Islands fChristensen etal.. 20211 examined the association
between PFDA exposure and AGD at 3 months of age. All three studies examined boys while Lind et
al. (2017a) and Christensen et al. (20211 also included girls.

Among boys, Tian etal. (20191 reported smaller AGD at birth with higher PFDA exposure
(ASD p = -0.58, 95% CI: -1.11, -0.06; APD p = -0.63, 95% CI: -1.24, -0.01). Decrements were also
observed at 6 months (p >0.05), but not at 12 months, which may be due to greater heterogeneity in
size as children develop. A positive association was observed in Christensen et al. f20211 (Q2 (3 vs
Q1 = 1.4; 95% CI: 0.4, 2.5; Q3 p = 1.0; 95%CI: 0.0, 2.1; Q4 p = 1.3;95%CI: 0.3, 2.4). No association
was observed in Lind etal. (2017a). Exposure levels were considerably higher in Tian etal. (2019)
(median 2.1 vs. 0.2 and 0.3 ng/mL), but this does not explain the inconsistent direction of
association across studies.

For girls, there was an inverse association with PFDA for one of the two AGD measures
(AGDAC, measured from the center of anus to the top of clitoris) reported in Lind etal. f2017al.
They reported an association based on continuous exposure (P = -1.3, 95% CI: -2.8, 0.2), and
across upper two PFDA exposure quartiles in non-monotonic fashion (Q2 vs. Ql: p = 0.4;
95% CI: -1.3, 2.0; Q3: p = -0.7; 95% CI: -2.4, 0.9; Q4: -1.7; 95% CI: -3.6, 0.1, p trend= 0.04). An
association was also observed in the fourth quartile in the other AGD measure (AGDAF, measured
from center of anus to posterior fourchette), though it was not statistically significant (Q4 p = -1.0;
95% CI: -2.4; 0.4). No association was observed in Christensen et al. f20211.

AGD is a marker of androgen exposure, and thus an inverse association in AGD would be
expected to correspond with a decrease in testosterone. This was not observed in the single low
confidence study of testosterone in neonates (see Male and Female Reproductive Effects); however,
there is considerable uncertainty in the reproductive hormones evidence base. Thus, this lack of
coherence does not reduce confidence in the AGD findings. Reduced AGD is associated with
clinically relevant outcomes in males, including cryptorchidism, hypospadias, and lower semen

This document is a draft for review purposes only and does not constitute Agency policy.

3-139 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

quality and testosterone levels (Thankamony et al.. 2016). but adversity of reduced AGD is less
established in females. As noted above few studies of birth defects, US EPA did not identify any
epidemiological studies that examined PFDA in relation to congenital genitourinary defects, such as
cryptorchidism and hypospadias. Overall, the evidence for AGD is indeterminate given the mixed
results for various AGD measures across the sexes.

Gestational Duration Endpoints

\C£ • op*\ a\®v' ,

Participant selection
Exposure measurement
Outcome ascertainment -|
Confounding
Analysis -I
Sensitivity -
Selective Reporting -
Overall confidence A

Legend

| Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficieni (metric) or Low confidence (overall)
Critically deficient (melnc) or Uninformative (overall)

Figure 3-50. Study evaluation results for five epidemiological studies of
gestational duration and PFDA. Refer to HAWC for details on the study evaluation

review: HAWC Human Gestational Duration.

As shown in Figure 3-50,12 informative epidemiological studies examined PFDA in relation
to changes in gestational duration measures (i.e., gestational age or PTB}. All 12 examined
gestational age measures, while 6 included preterm birth. Four studies were high confidence

(Gardener etal.. 2021: Huo et al.. 2020: Lind et al.. 2017a: Bach etal.. 20161. five were medium (Hall
etal.. 2022: Yang etal.. 2022a: Hiermitslev et al.. 2020: Gvllenhammar etal.. 2018: Meng et al..
20181. and three studies were low owing largely to very limited exposure contrasts fGao et al..
2019: Workman etal.. 2019: Li etal.. 20171. One study had good sensitivity fHuo etal.. 20201. while

five were adequate fYang et al.. 2022a: Hiermitslev et al.. 2020: Gao etal.. 2019: Lind etal.. 2017a:
Bach et al. 20161 and six were deficient (Hall etal.. 2022: Gardener etal. 2021: Workman etal.
2019: Gvllenhammar etal.. 2018: Meng etal.. 2018: Li etal.. 20171. Ten of the 12 studies were

This document is a draft for review purposes only and does not constitute Agency policy,

3-140 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

prospective cohort or nested case-control studies (Hall etal.. 2022: Yang etal.. 2022a: Gardener et
al.. 2021: Hiermitslev etal.. 2020: Huo etal.. 2020: Gao etal.. 2019: Workman etal.. 2019: Meng et
al.. 2018: Lind etal.. 2017a: Bach etal.. 20161. and two were cross-sectional fGvllenhammar et al..
2018: Li etal.. 20171. For examination of consistency and between-study heterogeneity, we
examined the type of statistical analyses in addition to the type of study design. As part of this,
cross-sectional analyses are considered for any study that used maternal serum/plasma, umbilical
cord, or placental post-partum PFDA measures in relation to gestational duration even if the data
were derived from prospective cohort or nested case-control studies (Hall etal.. 2022: Yang etal..
2022a).

The epidemiological studies had maternal exposure measures that were sampled either
during trimester one fLind etal.. 2017al. two fHuo etal.. 20201. three f Gardener etal.. 2021: Gao et
al.. 20191 across multiple trimesters fHiermitslev etal.. 2020: Meng etal.. 2018: Bach etal.. 20161.
or had post-partum maternal or infant samples (Hall etal.. 2022: Yang etal.. 2022a: Gvllenhammar
etal.. 2018: Li etal.. 20171. All five of the cross-sectional studies/analyses had late sampling
(defined here as trimester 2 exclusive onward). Four (Hiermitslev etal.. 2020: Meng etal.. 2018:
Lind etal.. 2017a: Bach etal.. 20161 of the prospective cohort studies had early sampling (defined
here as having at least some trimester 1 maternal sampling), while the remaining two fGardener et
al.. 2021: Workman etal.. 20191 relied on late biomarker sampling.

Preterm Birth

Six studies examined PFDA and preterm birth including three studies each being high
(Gardener etal.. 2021: Huo etal.. 2020: Bach etal.. 20161 and medium confidence (Yang etal..
2022a: Hiermitslev etal.. 2020: Meng etal.. 20181 (Figure 3-51). Three studies showed some
evidence of increased risk of PTB with increasing PFDA exposures including two studies with early
biomarker sampling. Null associations for PTB were reported in the medium confidence study by
Yang etal. f2022al and high confidence study by Bach etal. f 20161. while a no n-significant inverse
association (OR = 0.65; 95%CI: 0.24,1.79 per each PFDA ln-unit increase) was reported in the
medium confidence study by Hiermitslev etal. (20201.

Although there was no evidence of an exposure-response relationship, the high confidence
study by Gardener etal. (20211 reported that participants in PFDA exposure quartile 4 had a
greater odd of PTB (OR = 1.82; 95%CI: 0.54, 6.19) relative to quartile 1. The medium confidence
study by Meng etal. f 20181 reported an increased OR of 1.6 for PTB (95% CI: 0.8, 3.0) in the PFDA
quartile 4, but no evidence of monotonicity or increased risk in the other quartiles. A larger
statistically significant result was detected for each ln-unit increase (OR = 2.2; 95%CI: 1.3, 3.8).
Associations between PFDA and different PTB measures (including overall and different sub-types)
were at or just below the null value based on continuous exposures in the high confidence study by
Huo etal. (20201. Similar patterns emerged across PFDA exposure tertiles, albeit non-significant
ORs with an exposure-response relationship was suggested for clinically indicated PTBs (T2 OR =
1.11; 95%CI: 0.50, 2.48; T3 OR= 1.30; 95%CI: 0.59, 2.89). This result seemed to be largely driven by

This document is a draft for review purposes only and does not constitute Agency policy.

3-141 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

results in female neonates (OR = 1.38; 95%CI: 0.61, 3.11 per each ln-unit PFDA increase) (sex-
specific data not shown on forest plots below}.

PTB Summary

Three (two high and one medium confidence) of six studies showed increased odds of PTB
with increasing PFDA exposures with risks ranging from 1.3 to 2.2. Although the number of studies
was small, two of these three studies showing increased risks were based on late biomarker
samples. No other patterns were evident by study confidence or other characteristics. For example,
study sensitivity did not seem to be an explanatory factor among the null studies. One of the four
studies with categorical data showed evidence of exposure-response relationships.

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Rogrossion coefficient

• PTB Refatve Risk (RR)
O PTB Relatve Risk (RR) p«0.05
H 95S"i5 confidence in larval

Bach etal., 2016.
3981534

Aarfius Birth Cofiori <2008-2013)
Denmark, 1507 mother-infant pairs

Adequate

Cohort
(Prospective)

Trimester 1-2

0.83

Quartile 2

~—#1

0.8?

Quackle 3

»—•->

1.06

Quarile 4

1 <•—

0.97

ln-unit (ng/rrl)
increase

Hue. etal.. 2020.
8835452

Shanghai Brh Cohort (2013-2016).
China, 7840 rrother-infan; pairs

|High|

Good

Cohort
(Prospective)

Trimester 2

1.04

0.85
0.86

Tertile 2

Tertile 3

ln-unit (no-'ml.)
increase

>-#r<

Gardener et al..
2021.7021199

Vanguard Pilct Study of the National
Children's Study (NCS) (2009).
Unitec Stales. 5420 mother-! nfart
pairs

IHigni

Deficient

Cohort
(Prospective)

Trimester 3

0.6

QuarJIe 2

• 1

1.13

QuarJIe 3

1.82

QuarJIe 4

Yang et al.. 2022.
10176604

Kashgar Birth Cohort (2018 2019).
China. 768 mother infant pairs

[Mediunl

Adequate

Nested
case control

At birth

1.06

ljvjil(ng.''mL)
increase

Mengetal.. 2018.
4829651

DNBC (1996-2002). DenmaiK 3535
mother-infant pairs

IMediurnl

Deficient

Cohort
(Prospective)

Trimester 1-2

QuarJIe 2

•—•—

t.1

Quaitile 3

i—•

1.6

QuarJIe 4

"• 1

2.15

In unit (rtg-mL)
increase

i h

~ 1

Hicrrmtslcv et al
2020.5660649

ACCEPT birth cohort <2010-2011.
2013-2015). Greenland 482
mother-infant pars

[Medium!

Adequate

Cohort
(Prospective)

Trimester 1 -3

0.649

ln-unit {n$'"ml)
increase

»——
i

-1 O 1

2 3 4

5 6 7

Figure 3-51. Preterm birth forest plot-six studies based on the overall
populationab. Refer to the HAWC link.

Abbreviation: PTB= Preterm Birth

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b For evaluation of patterns of results, we considered studies that collected biomarker samples concurrently or
after birth to be cross-sectional analyses [e.g., (Yang et al., 2022a)1.

Gestational Age

Twelve studies examined PFDA in relation to changes in gestational age. Two of these
studies reported only sex-specific data fHall etal.. 2022: Lind etal.. 2017al with three studies
reporting both sex-specific and overall population results fHiermitslev etal.. 2020: Meng et al..
2018:T,i et al.. 20171.

Gestational Age-Overall Population

Six of the ten studies based on the overall population were null including the high
confidence studies by Bach etal. f20161 and Huo etal. f20201. the medium confidence studies by

This document is a draft for review purposes only and does not constitute Agency policy,

3-142 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Hiermitslev etal. (20201. and the low confidence studies by Gao etal. (20191: Workman et al.
(20191: Li etal. (20171 (Figure 3-52). No patterns were seen by study sensitivity amongst these null
studies but four of the six had adequate or good domain ratings.

Four studies in the overall population (one high and three medium confidence) showed
some evidence of lower gestational age relative to PFDA in the overall population. The high
confidence study by Gardener et al. f20211 showed decreased gestational age in only PFDA quartile
4 with no exposure-response relationship evident (Q4 (3 = -0.26 weeks vs. Ql). Although it was null
for term births, there was an inverse association between gestational age and each PFDA ((3 = -0.72
weeks; 95%CI: -3.39,1.97 per each ln-unit increase) among preterm births in the medium
confidence study by Yang etal. f2022al. Two other medium confidence studies reported only slight
non-significant deficits (-0.12) per each ln-unit increase fGvllenhammar etal.. 2018: Meng et al..
20181. but the latter showed larger deficits in both exposure quartile 3 and 4 ((3 range: -0.20 to -
0.50 weeks, respectively). Three of these four studies reporting lower gestational age were based
on later biomarker sampling.

Gestational Age-Sex Specific

Two of the five studies in male neonates reported some gestational age deficits compared to
just one study in girls. The medium confidence study by Hiermitslev etal. f20201 and the low
confidence study by Li etal. f 20171 reported null findings for both boys and girls. The high
confidence study by Lind etal. (2017a) showed minimal evidence of associations in the upper
quartiles for either sex, although they reported an imprecise gestational age reduction of-0.21
weeks (95%CI: -0.66, 0.24) among girls that was incongruous with their categorical data. The
medium confidence study by Hall etal. (2022) reported no n-significant deficits in the upper tertile
for boys ((3 = -0.26 weeks; 95% CI: -0.77, 0.27). The medium confidence study by Meng et al. T20181
detected a statistically significant decrease for boys per each ln-unit increase ((3 = -0.25 weeks;
95% CI: -0.43, -0.04) that was similar in magnitude.

Gestational Age Summary

Overall, there was mixed evidence of associations between PFDA and gestational duration
endpoints. Only six of the twelve PFDA (two high and four medium confidence) studies showed
some evidence of associations with gestational age in either the overall population, term/pre-term
subsets, or either sex. Four of the six studies that showed some gestational age deficits were based
on later biomarker sampling which might be indicative of an impact of pregnancy hemodynamics.
Four of these studies had deficient study sensitivity ratings which may explain why some results
were not statistically significant especially among the sex-specific analyses. No patterns were seen
by study sensitivity among the six different null studies. There was limited evidence to draw any
conclusions from the three sex-specific findings given that only two of five studies among boys and
one study in girls detected any evidence of gestational age differences in relation to PFDA.

This document is a draft for review purposes only and does not constitute Agency policy.

3-143 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Study

Population

Overall Study
Confidence

Study Sensitivity

Design

Exposure
Window

Regression
Coefficient

Exposure
Comparison

Regression cooffleient

A & [association v. ti Gfij
G |> {ilWyOU.lIIWI 'I. I'l GA|

Ouch ol al. 2016.
3981534

Aahus Olrl'i CuliU!U2(X®-2013),
DanmarK, 1507 notw-rtfinl p*i»

IftgU

Adc«{uulc

(Prospective)

Dniester 1-2

0.1

Quur'Jte2

1 |«-H

I—I cort dense me-wi

0.1
0,1
0

Quarts 3
Quarita 4
Irmnl (iv'mL)

I »•—1

Hiatal. 2020.
CS3S4S2

ShmShal Birlh Collar. (2013-2016)
Ch:o», 2849 moUuK-Mart pair*

MgN

Good

Cohort
(Prospective)

TiknastwZ

0.04

Irvunil

»•»

Ua-Oorw «1
5021, 73211®

V«pgu«!fl Plo:Sljeyo'UwN»lioT*l
OhiM-aVs Slucy (NCS) (2S0S),
UniVwi S-sHm, 64211 nvi+M-hfav
pvbv

«flh|

Cohort
IPrDMXiOivol

Tiimesler3

0.04

-0.06
0.26

Uuarie 2
Quartfl 3
Ouoi'io 4

• ,

G/«9Trtamr-«r a:

« ,2018, 4238300

POPUP 11900-2011). Sweden. 381

mo*fyn-n

|V«Sum!

Odcient

Cross-sectional

.vwfcs posl-bit

-0.12

Irvunil

I •-L-«

Valgum., 2C22.

Ktshgti BMi Cohort (2018-20101

IVadrjm

Nested

Al 8.-th

•0.00

Irvunl [n^'mL)

trvun» (nglml)

10176804

China, 768 irottiar-i-tfmt pairs

casa-oantrol

-0.72

Incraass

Marg ei al 2013

, Canada, 414
motief in'an: pat-s

DefBtoW

Cohort
|PlD»p«tlrv«|

Trmes«or23

0.0i2

h Uli« (il^'llll)

1
1

Gaoetal. J019

Afftared Hosp4al of Capital Maninal

University (2U19 SKI'K). Clina 132
"io»!»!-¦ n'sn. pais

Howl

Adequate

fptoepodta)

Trt™ster3

0.31

TertiteJ

0C1

lertltoa

3.5

-2.5 -2

-1.5 -1 -0.5 0

05 1 1-S 2

Figure 3-52. Overall population forest plot of 10 gestational age studies-™1.
Refer to the HAWC link.

Abbreviation: GA = Gestational Age

a Studies are sorted first by overall study confidence level then by Exposure Window examined.
b Yang et al. (2022a), -0.7 per IQR Increase value is reported in the preterm birth population; the -0.08 per IQR
increase value is in the term birth population.
c- Gardener gestational age differences estimated from digitization of their Figure 4; 95%Cls were not estimable.
d- For evaluation of patterns of results, we considered studies that collected biomarker samples concurrently or
after birth to be cross-sectional analyses [e.g., (Yang et al., 2022a)l.

Study

Population

Overall Study Study Sensitivity
Confidence

Design

Exposure
Window

Coefficient

Exposure
Comparison

Regression coefficient

9 (SRVwiaSon -M h GA fsvkfl
o A (aMiMJison Kith GA f.\*|l 3
-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1 Gestational Duration Summary

2 Seven different studies out of 12 showed some associations between PFDA exposures and

3 different gestational duration measures with comparable levels of evidence in preterm birth and

4 gestational age. Five of these seven studies were based on later biomarker sampling which might be

5 indicative of an impact of pregnancy hemodynamics. Study sensitivity was limited in some studies

6 and could explain some of the null results and lack of statistical significance especially in the sex-

7 stratified analyses. Few other patterns were evident across sex or different study characteristics.

This document is a draft for review purposes only and does not constitute Agency policy.

3-145 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-21.Summary of 12 studies of PFDA exposure and gestational duration measures

Author

Study location/Years

Exposure median/IQR (range)
in ng/mL

Study sensitivity
domain judgment

PTB

High Confidence Studies

Bach et al. (2016)

Denmark, 2008-2013

1,507

0.30/0.20 (LOD-2.87)

Adequate

0 All

Gardener et al. (2021)

USA, 2009-2013

354

0.2/0.2 (LOD-2.6)

Deficient

-T All

-All

Huo et al. (2020)

China, 2013-2016

2,849

1.69/1.38 (N/A)

Good

T All
t Girls
0 Boys

0 All

Lind et al. (2017a)

Denmark, 2010-2012

636

0.30/0.10(0.1-1.8)

Adequate

0 Girls
0 Boys

Medium Confidence Studies

Gvllenhammar et al. (2018);

Sweden, 1996-2001

381

0.24/0.14 (LOD-1.1)

Deficient

-All

Swedish Environmental

Protection Agency (2017)a

Hall et al. (2022)

USA, 2010-2011

120

0.06/N/A (LOD-0.3)

Deficient

- Boys
Girls

Hiermitslev et al. (2020)

Greenland, 2010-2011;
2013-2015

266

0.71/N/A (0.12-7.84)

Adequate

nU All

0 All
0 Girls
0 Boys

Meng et al. (2018)

Denmark, 1996-2002

2,132

0.20/0.10 (N/A)

Deficient

T All*

-All
- Boys*

0 Girls

Yang et al. (2022a)

China, 2018-2019

768

0.035-cases; 0.027- controls
(range: 0.003-0.359)

Adequate

0 All

-All

Low Confidence Studies

Li et al. (2017)

China, 2013

321

0.15/0.16 (ND-2.12)

Deficient

0 All
0 Girls
0 Boys

Gao et al. (2019)

China, 2015-2016

132

0.47 (LOD-3.15)

Adequate

0 All

Workman et al. (2019)

Canada, 2010-2011

414

0.13/N/A (LOD-1.4)

Deficient

0 All

Abbreviations: PTB = preterm birth; GA = gestational age.

*p < 0.05; 0: no association; +: positive association; negative association; T: increased odds ratio; -l: decreased odds ratio.

Note: "Adverse effects" are indicated by both Increased ORs (T) for dichotomous outcomes and negative associations (-) for the other outcomes.

This document is a draft for review purposes only and does not constitute Agency policy.

3-146 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

aSwedish Environmental Protection Agency (2017) and Gvllenhammar et al. (2018) results are included here (both analyzed the POPUP cohort).

This document is a draft for review purposes only and does not constitute Agency policy.

3-147 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Birth Defects

Two studies examined birth defects in relation to PFDA exposures (Figure 3-53) with one
each having adequate and deficient study sensitivity. The medium confidence congenital heart
defect study by fOu etal.. 20211 showed increased risks for PFDA >0.53 ng/mL (vs. <0.53 ng/mL)
for all defect groups examined including septal defects (OR = 2.33; 95%CI: 1.00, 5.45), conotruncal
defects (OR = 2.58; 95%CI: 0.92, 7.25), and total heart defects (OR=1.83; 95%CI: 1.07, 3.12). The
low confidence Cao etal. (2018 study showed minimal evidence of associations between PFDA
exposures and all birth defects (OR = 1.37; 95%CI: 0.60, 3.08). There is considerable uncertainty in
interpreting results for broad all birth defect groupings which decreases study sensitivity given the
etiological heterogeneity across different birth defects. Overall, there was limited evidence of
associations between PFDA exposures and birth defects in the two available epidemiological
studies. However, there is insufficient data for any specific birth defects to draw further
conclusions.

Fetal Loss-Spontaneous Abortion

Buck Louis, 2016, 3858527 -
Jensen, 2015, 2850253-
Liew, 2020, 6387285
Mi, 2022, 10413561 H
Wang,2021,10176703
Wikstrom, 2021, 7413606 H

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 3-54. Study evaluation results for two epidemiological studies of
spontaneous abortion and PFDA. Refer to HAWC for details on the study
evaluation review: HAWC Human Spontaneous Abortion

Six (five medium and one low confidence) epidemiological studies (Mi etal.. 2022: Wang et
al.. 2021: Wikstrom etal.. 2021: Liew etal.. 2020: Louis etal.. 2016: lensen etal. 20151 reported on
the association between PFDA exposure and spontaneous abortion, which is defined as pregnancy
loss occurring before approximately 20-22 weeks gestation. This period can be further divided
into preclinical/early loss (occurring before implantation or before a pregnancy is clinically
recognized) and clinical loss (occurring at 5-28 weeks gestation). The study evaluations of the

This document is a draft for review purposes only and does not constitute Agency policy.

3-148 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

available studies are summarized in Figure 3-54. Two medium confidence studies were
prospective cohorts with high ascertainment of early losses, one of couples trying to conceive,
followed through delivery fLouis etal.. 20161 and one of women undergoing in vitro fertilization
fWang etal.. 20211. Three additional medium confidence studies assigned pregnant women from
existing cohorts as controls and enrolled cases with first trimester losses fWikstrom etal.. 20211.
throughout pregnancy fMi etal.. 20221. or identified cases via medical registry fLiew etal.. 20201.
One study considered low confidence. Tensen etal. (20151 is a cohort of pregnant women enrolled
at 8-16 weeks gestation and was deficient for participant selection due to the high risk of
incomplete case ascertainment (i.e., due to not including early losses and potential for loss to
follow-up). Missing early losses has the potential to bias the results towards the null or even in a
protective direction if there is a true effect but is unlikely to result in a spurious positive
association. This potential also existed in Liewetal. f20201. but this study was not downgraded to
low confidence as loss to follow-up was not a concern.

The results of the studies on spontaneous abortion are summarized in Table 3-22. Three of
six studies showed some evidence of increased risk of spontaneous abortion. This included two
studies (one medium and one low confidence) that reported strong positive associations between
PFDA exposure and spontaneous abortion, with large effect sizes and statistical significance fMi et
al.. 2022: Tensen etal.. 20151. In addition, another medium confidence study by fLiew etal.. 20201
reported a smaller (OR = 1.3; 95% CI: 0.7, 2.2) but not statistically significant positive association,
while another medium confidence study fWikstrom etal.. 20211 was largely null. Two medium
confidence studies, which were the only studies able to consider preclinical losses, reported inverse
(nonsignificant) associations (Wang etal.. 2021: Louis etal.. 20161. It is unlikely that the limitations
identified in the low confidence study would explain the observed positive associations, as bias in
Tensen etal. f20151 is expected to be towards or past the null. Thus, while there is some evidence of
an association with spontaneous abortion, there is considerable uncertainty due to inconsistency
across medium confidence studies. It is possible that this is related to the inclusion of preclinical
loss, but this is not clear based on available evidence.

This document is a draft for review purposes only and does not constitute Agency policy.

3-149 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-22. Associations between PFDA and spontaneous abortion in
epidemiology studies

Reference, study
confidence

Population

Median
exposure
(125th, 75th) in
ng/mLoras
specified

Spontaneous
abortion types
included

Effect estimate
description

Effect estimate (95%
CI)

Liew et al. (2020),

Case-control nested within pregnancy
cohort, Denmark; 438 women

0.2 (0.1-0.2)

Clinical, 12-22
weeks

OR (95% CI) for
quartiles vs. Q1

02:1.0(0.6,1.7)
03:1.1(0.7,1.9)
04: 1.3 (0.7, 2.2)

medium

Wikstrom et al.

Case-control nested within pregnancy
cohort, Sweden; 1,529 women

0.3 (0.2-0.3)

Clinical, first
trimester

OR (95% CI) for
doubling of
exposure

1.10(0.81, 1.53)

(2021), medium

Jensen et al.

Pregnancy cohort, Denmark; 392
women

0.3 (0.2-0.6)

Clinical, post
enrollment at
8-16 weeks

OR (95% CI) for
tertiles vs. Tl

T2: 1.9 (0.9, 3.8)
T3: 2.7 (1.3, 5.4)*

(2015). low

Louis et al. (2016),

Preconception cohort, U.S.; 344 women

0.4(0.2-0.6)

Total

HR (95% CI) for
tertiles vs. Tl

T2: 0.83 (0.49, 1.40)
T3: 0.68 (0.41, 1.14)

medium

Wang et al.

Preconception cohort of women
undergoing first IVF cycle, China, 305
women

0.5 (0.3-0.7)

Preclinical

RR (95% CI) for

log-unit

increase

0.67(0.16, 2.73)

(2021), medium

Mi et al. (2022).

Case-control nested within pregnancy
cohort, China; 88 women

0.8

Clinical (9-12
weeks)

OR (95% CI) for
above vs.
below median

5.00 (1.53,16.33)*

medium

Abbreviations: OR: odds ratio; HR: hazard ratio; RR: relative risk; Tl: Tertile 1; T2: Tertile 2; T3: Tertile 3: IVF: in
vitro fertilization.

* Denotes statistical significance atp < 0.05.

Animal studies

One toxicity study evaluated effects of PFDA on offspring (Harris and Birnbaum. 19891.
This gavage study in mice examined maternal health, fetal survival, growth, and morphological
development in two experiments covering different developmental windows. The two respective
experiments consisted of gavage administration of 0-32.0 mg/kg-day on GD 10-13 to examine the
developmental window related to cleft palate and hydronephrosis and gavage administration of 0-
12.8 mg/kg-day on GD 6-15 to examine the entire developmental window related to the major
period of organogenesis. The dams were necropsied on GD 18; the fetuses were removed from the
uterus and examined. The Harris and Birnbaum (19891 study was evaluated as high confidence for
most endpoints examined in both experiments (see Figure 3-55). Concerns were noted for fetal
body weight measures as the study failed to report fetal body weights by sex, which impacted the
results presentation domain and lowered the overall confidence of this endpoint to medium.

This document is a draft for review purposes only and does not constitute Agency policy.

3-150 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

S179

Reporting quality -
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -
Exposure timing, frequency and duration
Endpoint sensitivity and specificity
Results presentation
Overall confidence

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)

Deficient (metric) or Low confidence (overall)

Critically deficient (metric) or Uninformative (overall)

Not reported

Multiple judgments exist

Figure 3-55. Developmental animal study evaluation heatmap. Refer to HAWC
for details on the study evaluation.

1 Fetal growth

2 Fetal body weights were measured at GD 18 for each experiment (GD 10-13 or GD 6-15],

3 Both experiments reported a significant trend in fetal body weight with decreases >5% being

4 observed at >0.5 mg/kg-day (9.6-44%) for the GD 10-13 experiment and >3 mg/kg-day (6-50%)

5 for the GD 6-15 experiment (see Figure 3-56 and Table 3-23). The changes in fetal body weight

6 were of large magnitude and occurred at doses not associated with maternal toxicity. In the GD 10-

7 13 experiment, changes in fetal body weight were ~10% at doses ranging from 0.5-4 mg/kg-day

8 and were >40% atthe highest dose (32 mg/kg-day). In the GD 6-15 experiment, changes in fetal

9 body weight were 23% at 6.4 mg/kg-day and as large as 50% at the highest dose (12.8 mg/kg-day).

This document is a draft for review purposes only and does not constitute Agency policy,

3-151 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name Study Name

Outcome
Confidence

Study Design

Animal Description

Trend Test Response Dose
Result Units (mg/kg-day)

Fetal Body Weight Harris, 1989, 3858729 High confidence Gestational Oral (GD 10-13) F1 Mouse, C57BL/6n (c52) significant litter

Gestational Oral (GD 6-15) F1 Mouse, C57BL/6n (.;$) significant litter

0.25
0.5

2
4
8
16
32

0.03

0.1

0.3

6.4
12.8

PFDA Fetal Body Weight

O Statistically significant
0 Percent control response
M range

I 0 I

—I 1 1 1 1 1 1 1 1 1—

-55 -50 -45 -40 -35 -30 -25 -20 -15 -10 -5
Percent Control Response

Figure 3-56. PFDA fetal body weight after gestational exposure (results can be viewed by clicking the H.AWC link).

This document is a draft for review purposes only and does not constitute Agency policy,

3-152 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-23. Percent changes relative to controls in fetal body weight in a
developmental mouse study after PFDA exposure fHarris and Birnbaum.
19891

Endpoint

Dose (mg/kg-d)

0.25

0.5

Decreased fetal body weight for the GD 10-13
experiment

-4

-10

-11

-10

-17

-22

-44

Dose (mg/kg-d)

Endpoint

0.03

0.1

0.3

6.4

12.8

Decreased fetal body weight for the GD 6-15
experiment

-1

-3

-1

-4

-6

-23

-50

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors.

Maternal health

In the Harris and Birnbaum T19891 study, the health of the dams was assessed during both
experiments through examination of body weight, liver weight and survival. Both exposure
durations resulted in a significant trend in body weight change (defined as final body weight -
gravid uterus weight + empty uterus weight - initial body weight) for the dams with statistically
significant decreases in the two highest dose groups of both experiments. Body weight gain was
markedly decreased (-149% change from controls) in the 12.8 mg/kg-day group of the GD 6-15
experiment (see Figure 3-57). A significant trend was also reported for increased liver weight in
both the GD 10-13 and GD 6-15 experiments; refer to Section 3.2.1 for more detail on this effect
Maternal deaths were not observed in the GD 10-13 experiment, but 3 dams died in the high dose
group (12.8 mg/kg-day) of the GD 6-15 experiment This result is consistent with the overt toxicity
of PFDA at high doses (refer to Section 3.2.10 on General toxicity effects for more details).

Fetal viability

In the Harris and Birnbaum T19891 study, endpoints related to fetal viability were measured
at GD 18 for each experiment (i.e., groups dosed on GD 10-13 or GD 6-15). In both experiments,
there was no difference in total implantations per litter between the control and treated groups
indicating that the pregnancy rate was similar prior to exposure. However, following exposure, an
increase in percent resorptions per litter (defined as [total number of resorptions and dead
fetuses/number of total implantation sites] x 100) was observed in the high dose groups of both
experiments (170% and 344% for the GD 10-13 and GD 6-15 experiments, respectively) with
statistical significance reported for the GD 6-15 experiment (see Figure 3-33). A reduction in the
number of live fetuses per litter was also reported in high dose groups of both experiments (32%
and 36% for the GD 10-13 and GD 6-15 experiments, respectively) with statistical significance
reported for the GD 6-15 experiment Additionally, there was an increase in the number of dams

This document is a draft for review purposes only and does not constitute Agency policy.

3-153 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

that experienced total resorption in the high dose groups of both experiments (4/12 dams vs. 0/13
in controls for the GD 10-13 experiment; 3/10 dams vs. 0/12 in controls for the GD 6-15
experiment) though the number of litters with resorptions were not different between control and
treated groups (see Figure 333). Although these data might suggest an effect of maternal exposure
on fetal viability as increased resorptions and decreased number of live fetuses are indicative of
developmental toxicity per the U.S. EPA's Guidelines for Developmental Toxicity Risk Assessment
(U.S. EPA. 1991). effects on these endpoints were observed at doses that were also associated with
significant maternal toxicity.

Morphological development

In the Harris and Birnbaum T19891 study, morphological development was examined in
GD 18 fetuses for both the GD 10-13 and GD 6-15 experiments. This included external evaluation
of all fetuses, soft tissue evaluation of 50% of the litters in each dose group (using Bouin's fixation
and Wilson's free-hand sectioning technique), and skeletal evaluation of the remaining 50% of the
litters in each dose group (using alizarin red S staining of ossified bone). In the GD 6-15
experiment, PFDA exposure caused significant dose-related trends for multiple skeletal variations
(i.e., absence of fifth sternebrae, delay in braincase ossification, and delay in phalanges ossification)
(see Figure 3-57). The fetal incidence of delayed braincase ossification was significantly increased
at >0.03 mg/kg-day with the incidence rates ranging from 26 to 100%; it is unclear exactly which
cranial bones are included in "braincase ossification." The number of fetuses with absence of the
fifth sternebrae and delayed phalanges ossification was significantly increased at >6.4 mg/kg-day
ranging from 15 to 35%. The statistical analyses of the skeletal variations data were performed
independently by the U.S. EPA and not included in the original study. Litter incidence and
individual fetus per litter data were not reported for these effects. Data for skeletal variations were
reported as fetal incidence while data for individual fetus per litter is the preferred unit of analysis
for these effects. Absence of the fifth sternebrae and delayed phalanges ossific- and mortality at
12.8 mg/kg-day). Whereas skeletal variations were significantly increased, the GD 6-15
experiment reported no soft tissue or skeletal malformations. Per the U.S. EPA's Guidelines for
Developmental Toxicity Risk Assessment, a malformation is defined as "as a permanent structural
change that may adversely affect survival, development, or function" while a variation "is used to
indicate a divergence beyond the usual range of structural constitution that may not adversely
affect survival or health." Furthermore, skeletal variations are commonly associated with maternal
toxicity fCarnev and Kimmel. 20071 as was observed for the absence of the fifth sternebrae and
delayed phalanges ossification in mice exposed to PFDA. Based on the considerations above,
including a lack of malformations and/or that some skeletal variations were observed at the same
doses as maternal toxicity, the biological adversity for PFDA-induced skeletal variations is
considered unlikely. Thus, the greatest level of concern is interpreted for the delayed brain
ossification, although the significance of this variation (in terms of later biological consequences) is
unclear.

This document is a draft for review purposes only and does not constitute Agency policy.

3-154 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Effect

Outcome Confidence

Experiment Name

Endpoint Name

Animal Description

Trend Test Result

PFDA Developmental Effects

Maternal Body Weight

High confidence

Gestational Oral (GD 10-13)

Maternal Body Weight Gain, Corrected

P0 Mouse, C57BL/6n (2)

significant

Gestational Oral (GD 6-15)

Maternal Body Weight Gain, Corrected

P0 Mouse, C57BL/6n ( )

significant

•——•—

• • ~ ~

:etal survival

High confidence

Gestational Oral (GD 10-13)

Live Fetuses per Litter

F1 Mouse, C57BL/6n (
-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Mechanistic studies and supplemental information

In support for PFDA-induced developmental effects in humans and mice, fTruong etal..
20221 reported that PFDA caused morphological effects in embryonic zebrafish from a
developmental toxicity screening study. Of the 139 PFAS tested, PFDA was determined to be the
most potent for the induction of teratogenic effects. Similar results were reported in an additional
study using zebrafish (Ulhaq et al.. 20131. Ulhaq etal. (20131 reported that spinal curvature was a
common malformation observed in zebrafish embryos exposed to PFDA and of the seven PFAS
tested, PFDA was the second most potent for the induction of developmental toxicity.

Evidence Integration

Based on over 45 different epidemiological studies included here the evidence of an
association between PFDA exposure and developmental effects in humans is considered slight but
was supported by the moderate evidence in animals. The epidemiological evidence was strongest
and most consistent for fetal growth restriction and in particular for birth-weight related measures,
which were assessed by the most accurate growth restriction measures available. Out of 28 in total,
18 different studies showed some deficits for the overall population or for both/either sex across
various birth weight measures. For example, 11 out of 2 2 PFDA studies in the overall population
reported some birth weight deficits; this included 9 out of 14 medium and high confidence studies.
Although data were more mixed, there appeared to be some coherence across these and other pre-
natal growth measures with different postnatal growth parameters. For example, there was some
consistency across 2 (one high and one low confidence) of the 3 postnatal weight studies with a
common examination window (~2 years of age). The evidence for other endpoints was not as
strong or consistent, including 10 of 17 birth length studies that showed some associations.
Although the consistency varied somewhat across the developmental endpoints, the dearth of birth
weight and birth length results in the overall study populations based on early or prepregnancy
measures might be indicative of potential bias due to the impact due to pregnancy hemodynamics
on PFDA levels. Despite fairly consistent evidence of an association between PFDA and different
BWT-related measures, and more mixed for other endpoints, there is considerable uncertainty
given that some sample timing differences may explain some of the reported fetal growth
restriction deficits examined here.

Across the outcomes, this set of developmental studies was of good quality and generally
had a low risk of bias, as 34 out of the 45 studies across the six primary endpoints [fetal growth
restriction (including both birth weight and length measures), gestational duration, postnatal
growth, anogenital distance, birth defects, and spontaneous abortions] were either medium or high
overall confidence. Several studies demonstrated sufficient sensitivity to detect associations in the
overall population and across sub-groups. However, many studies lacked power to detect
statistical interactions or differences across populations especially those based on stratified
analyses. This often results from low exposure levels with limited contrasts in many of the study

This document is a draft for review purposes only and does not constitute Agency policy.

3-156 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

populations, which may have diminished the sensitivity of some studies to detect associations. As
such, any null findings for studies with endpoints which lacked sensitivity should not be interpreted
as supporting a lack of effect In addition to the outcomes discussed in this section, pubertal
development is discussed in the reproductive Sections (3.2.4 and 3.2.5) but could also be a
developmental effect. The evidence for both males and females was based on one medium
confidence study and was weak, but study sensitivity was again a concern.

As noted above, fetal growth restriction endpoints provided the strongest evidence for
adverse developmental effects among the available studies. In considering the dose-dependence of
the birth weight decreases, only one out of four medium or high confidence studies with categorical
PFDA exposure data showed an exposure-response relationship. In addition, 9 of 14 medium or
high confidence studies of the overall population as well as 9 of 14 sex-specific results showing
adverse results based on continuous exposure also offer support for a biological gradient.
Exposure-response relationships were less evident for other endpoints that were examined.

It can be challenging to identify patterns across heterogenous epidemiologic studies and
study populations in the current database given the low exposure levels and/or limited and
variable exposure contrasts. Examining birth weight differences in human populations is also
challenging, since it can be difficult to differentiate pathological deficits versus natural biological
variation. There was considerable variability in BWT deficits ((3 range: -29 to -101 g per ln-unit
increases) in the overall population, with seven studies ranging from 31 to 59 g deficits per each ln-
unit increase. The clinical significance of these changes may not be immediately evident, but effects
of this magnitude can increase the number of infants at higher risk for other co-morbidities and
mortality especially during the first year of life. These population-level changes may have a large
public health impact when these mean birth weight deficits shift the birth weight distribution to
include more infants in the low-birth-weight category. Additionally, decreased birth weight has
been associated with long-term adverse health outcomes fOsmond and Barker. 20001.

Supporting the human evidence, the large and dose-dependent effects on fetal body weight
observed across two independent experiments reported in the lone mouse study by Harris and
Birnbaum (1989) (medium confidence for this endpoint) are without evidence to the contrary and
thus provide moderate evidence coherent with the findings in humans. Following gestational PFDA
exposure, decreases in fetal body weight with a significant trend were consistently observed in
both experiments at >0.5 mg/kg-day, including doses (0.5-4 mg/kg-day) well below those that
produced maternal toxicity. The changes in fetal body weight were also large in magnitude with the
percent changes of up to 10% at the lower doses and ranging as high as 44-50% at the highest
doses tested in both experiments. The rodent data for decreased fetal body weight are coherent
with data from the human studies in which the strongest and most consistent evidence was for fetal
growth restriction. Although an increased fetal incidence of several skeletal variations (i.e., delayed
braincase and phalanges ossification and absence of fifth sternebrae) was observed, the delays in
brain ossification, which started at >0.03 mg/kg-day, well below doses eliciting maternal toxicity,

This document is a draft for review purposes only and does not constitute Agency policy.

3-157 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

were most notable. This change is potentially indicative of delayed development (which would be
coherent with the PFDA-induced changes on fetal body weight); however, the significance of this
variation (in terms of future adverse consequences), is unknown, and malformations, which are
known to be adverse, were not observed. On a related note, PFDA was reported to be teratogenic in
embryonic zebrafish fTruong etal.. 2022: Ulhaq etal.. 20131. There were also statistically
significant changes reported for fetal viability in mice (i.e., increased % of resorptions per litter and
reduced number of live fetuses per litter) at the highest dose tested in the GD 6-15 experiment
(Harris and Birnbaum. 19891: however, effects on fetal viability were observed at the same doses as
significant maternal toxicity, preventing the ability to draw conclusions at these doses.

A notable data gap exists, as animal studies evaluating the effect of PFDA on postnatal
development were not identified. Although data were limited and not entirely consistent, some
effects of PFDA on postnatal growth were observed in humans. Additionally, effects on postnatal
development (e.g., delayed eye opening; reduced postnatal growth) have been observed in rodents
exposed to other PFAS such as PFOA, PFBS, PFBA. Overall, the information for postnatal
developmental effects is limited, introducing uncertainty on whether more sensitive developmental
effects of PFDA might occur. An additional data gap is the lack of data to inform the potential
mechanisms for PFDA-induced fetal growth restriction effects.

Taken together, the available evidence indicates that PFDA exposure is likely to cause
developmental toxicity in humans given sufficient exposure conditions10 (see Table 3-24). This
conclusion is based primarily on findings of dose-dependent decreases in fetal weight in the only
available toxicology study, with mice gestationally exposed to PFDA doses >0.5 mg/kg-day and
supported by evidence of decreased birth weight from studies of exposed humans in which PFDA
was measured during pregnancy, primarily with median PFDA values ranging from 0.11 to
0.46 ng/mL. The conclusion is further supported by coherent epidemiological evidence for
biologically related effects (e.g., decreased postnatal growth and birth length).

10 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes only and does not constitute Agency policy.

3-158 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Table 3-24. Evidence profile table for PFDA exposure and developmental effects

Evidence stream summary and interpretation

Inferences and summary
judgment

Evidence from studies of exposed humans-fetal growth restriction (see Section 3.2.3: Human studies)

Studies, outcomes,
and confidence

Fetal growth restriction
(Mean birth weight/ z-
scores; Small for
gestational age/low
birth weight)

8 high, 10 medium, and
10 low confidence
studies

Key findings and
interpretation

18 of the 28 studies
reported some inverse
associations between
PFDA exposures and
standardized or mean
birth weight measures
including 17 of 26 studies
of mean birth weight

11 of 22 studies showed
evidence of mean birth
weight deficits in the
overall population,
including 9 out of 14
medium or high
confidence studies

9 of 14 studies in boys
and girls reported some
birth weight deficits
including 8 out of 11
medium and high
confidence studies in
girls and 7 out of 11 in
boys; 4 studies reported
deficits in both sexes.

3 of 5 studies of small for
gestational age or low
birth weight reported
increased risks in the

Factors that increase
strength or certainty

Consistent decreases
across different
populations and with
variable study
sensitivity

Most of the evidence
among high and
medium confidence
studies (e.g., 9 out of
14 medium or high
confidence studies
showed BWT deficits)

Dose-dependent
(evidence of linear
relationships) in
many studies
examining

continuous measures

Moderate or large
magnitude of effect
in many studies
(typically > -30 grams
per each In-unit)

Although some
variability is
anticipated for
observational studies
of heterogenous

Factors that decrease
strength or certainty

Substantial
uncertainty due to
the potential impact
of hemodynamic
changes among
studies showing birth
weight deficits,
especially based on
late biomarker
sampling defined at
trimester 2 or later,
e.g., 9 of 11 studies
in the overall
population and 6 of 9
studies in girls and 5
of 9 in boys

Uncertainty of
potential

confounding in some
studies due to some
highly correlated
PFAS like PFNA,
although an
evaluation of this
possibility concludes
that it would not fully
explain the observed
PFDA associations
(see Appendix F)

Evidence stream summary

0OO

Slight

Based on consistent evidence
for birth weight reductions,
the most sensitive endpoint,
with coherence across some
other developmental
endpoints (e.g., preterm birth,
post-natal growth, and other
fetal growth measures such as
birth length, small for
gestational age and low birth
weight); more mixed for other
endpoints like head
circumference and gestational
duration.

Evidence indicates (likely)

Primary basis:

Slight human evidence for fetal
and postnatal growth
restriction supported by
coherent moderate evidence in
animals and for some other
developmental endpoints in
humans.

Human relevance: Evidence in
animals is presumed relevant
to humans.

Cross-stream coherence:
Impaired fetal growth was
observed in both humans and
mice.

Susceptible populations and
lifestages:

Based on evidence of impaired
fetal growth from human and
animal studies, early lifestages
may be at higher risk.

Other inferences'.

No specific factors are noted.

This document is a draft for review purposes only and does not constitute Agency policy.

3-159 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

overall population; fairly
consistent in magnitude
(OR range: 1.2-1.8)

populations,
exposure

levels/sources, and
design/analysis
elements, coherence
with findings for
related outcomes,
most notably for
birth length and
postnatal growth
measures

• 1 out of 4 medium or
high confidence
studies with
categorical data
showed exposure-
response
relationships in
overall population as
well as in girls for
standardized and
mean BWT measures

• Imprecision of some
effect estimates

Fetal growth restriction
(Birth length)

6 high, 4 medium, and 7
low confidence studies

• 10 of 17 studies in total
including 5 (2 high, 1
medium and 2 low
confidence) of 15
examining the overall
population reported
some birth length
deficits (including 3 of
the 10 total medium or
high confidence studies)

• 7 (4 high and 3 medium
confidence) of 10 sex-
specific studies reported
some birth length
deficits; 4 studies each in
boys and girls

• Overall population
results were similar
in magnitude despite
between-study
sources of
heterogeneity
including different
exposure contrasts

• Sex-specific deficits
were often larger and
more variable than
the overall
population

• Substantial

uncertainty due to
the potential impact
of hemodynamic
changes among
studies showing birth
length deficits based
on later biomarker
sampling, e.g., 4 of 5
studies in overall
population and 4 of 7
sex-specific studies

Fetal growth restriction
(Head circumference)

5 high, 5 medium, and 4
low confidence studies
(9 adequate and 5

• 5 (2 high; 3 medium

confidence) of 14 studies
reported smaller head
circumference including
2 of 11 in overall

• Five of the 10 high
and medium
confidence studies
reported smaller
head circumference

• Limited evidence of
associations
especially in the
overall population
where five of the six

This document is a draft for review purposes only and does not constitute Agency policy.

3-160 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

deficient study
sensitivity)

population; and 3 of 7
sex-specific studies

in the overall
population or either
sex

• Two of the 6 studies
with adequate
sensitivity reported
some head
circumference
deficits across sexes

null studies had
deficient study
sensitivity

Evidence from studies of exposed humans-anogenital distance (see Section 3.2.3: Human studies)

Anogenital distance

3 medium confidence
studies

• Inverse association

between PFDA exposure
and anogenital distance
(AGD) in one of three
medium confidence
studies in boys and one
of two studies in girls

• Adverse association
in boys observed in 1
medium confidence
study

• Unclear adversity of
AGD decreases in
girls

• Although some
variability is
anticipated for
observational studies
of heterogenous
populations,
exposure

levels/sources, and

design/analysis

elements,

unexplained

inconsistency

QQQ

Indeterminate

Based on inconsistent results
across medium confidence
studies

Evidence from studies of exposed humans-gestational duration (see Section 3.2.3: Human studies)

Gestational Duration
(Preterm birth)

3 high and 3 medium
confidence studies

• 3 (2 high and 1 medium
confidence) of 6 preterm
birth studies reported
increased risk; 6 studies
had deficient study

• Risks fairly consistent
in magnitude (OR
range: 1.3 to 2.2).

• Some uncertainty
due to potential
impact of pregnancy
hemodynamics as 2
of 3 studies based on

©OO

Slight
Mixed evidence and
uncertainty due to the
potential impact of
hemodynamic changes among

This document is a draft for review purposes only and does not constitute Agency policy.

3-161 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

sensitivity; 5 adequate,
and 1 good

later biomarker
sampling

• Potential

confounding by PFAS
including highly
correlated PFNA;
limited evidence for
PFNA suggests would
not likely fully
explain PFDA
associations

studies with gestational
duration deficits

Gestational Duration
(Gestational age)

4 high, 5 medium, and
3 low confidence
studies

• 6 of 12 studies reported
lower gestational age; 4
of these 6 had deficient
study sensitivity

• No factors noted

• Unexplained
inconsistency,
although this may be
partially due to poor
sensitivity

• Substantial
uncertainty due to
the potential impact
of hemodynamic
changes among 4 of
6 studies showing
gestational age
deficits, especially
based on late
sampling (defined as
trimester 2 or later)

• Outcome may be
prone to some
measurement error

Evidence from studies of exposed humans-postnatal growth (see Section 3.2.3: Human studies)

Postnatal growth

• 3 (1 high and 2 low

confidence) of 6 studies

• Consistency across 2
of the 3 weight

• Potential

confounding across

®oo

Slight

This document is a draft for review purposes only and does not constitute Agency policy.

3-162 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

4 high, 1 medium and 3
low confidence studies

showed postnatal weight
deficits; with limited
sensitivity in some
studies (3 adequate; 3
deficient)

• 2 (1 high and 1 low
confidence) of 5 studies
showed postnatal height
deficits; with limited
sensitivity in some
studies (3 adequate; 2
deficient)

• 2 (1 high and 1 low
confidence) of 3 studies
showed increased
adiposity; with limited
sensitivity in some
studies (1 adequate; 2
deficient)

• Both high confidence
studies showed minimal
and mixed rapid weight
gain results; with limited
sensitivity in some
studies (1 adequate; 1
deficient)

studies with a
common

examination window
(2 years of age),
including one high
and one low
confidence study

PFAS for some
endpoints

• Unknown critical
window(s) for
childhood growth
endpoints;
assumption was in
utero period is most
relevant

Mixed results across different
measures, with limited study
sensitivity in some studies.
Results were more consistent
when a homogenous
population considered (~2
years of age).

Evidence from studies of exposed humans-spontaneous abortion (see Section 3.2.3: Human studies)

Soontaneous abortion

5 medium and 1 low
confidence studies

• Two medium and one
low confidence studies
reported increased odds
of spontaneous abortion
while 2 medium
confidence study

• Large effect size in
two studies (OR>2)

• Unexplained

inconsistency across
medium confidence
studies

®oo

Slight
Based on inconsistent
evidence across studies

This document is a draft for review purposes only and does not constitute Agency policy.

3-163 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

reported an inverse
association.

• Potential

confounding across
PFAS

Evidence from in vivo animal studies (see Section 3.2.3: Animal studies)

Studies, outcomes, and
confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream summary

Fetal growth

1 medium confidence
study (2 independent
experiments)

• Fetal body weight was
reduced at

>0.5 mg/kgday in the
GD 10-13 experiment
(maternal body weight
decreased at
>16.0 mg/kg-d).

• Fetal body weight was
reduced at >1.0 mg/kg-d
in the GD 6-15
experiment (maternal
body weight decreased
at >6.4 mg/kg-d, with
mortality at higher
doses).

• Consistency across
the medium
confidence GD 10-13
and GD 6-15
experiments.

• Dose-response
gradient observed
within experiments
and exposure
duration gradient
observed across
experiments

• Large magnitude of
effects (up to 50%).

• No factors noted

0®Q

Moderate

Based primarily on decreased
fetal growth at >0.5 mg/kg-d
in two independent
experiments from a single
study in mice. The reliability
and biological significance of
other, potentially related,
findings from this study are
unclear.

Fetal viability
1 high confidence study

• A treatment-related
increase in the
percentage of
resorptions per litter was
reported at 12.8 mg/kg-d
in dams treated from
GD 6-15.

• The number of live
fetuses per litter was
reduced at 12.8 mg/kg-d

• Coherence of effects
on percentage of
resorptions and
number of live
fetuses in a high
confidence study.

• Substantial concern
for potential
confounding as
decreased fetal
viability occurred at
the same dose as
maternal mortality.

This document is a draft for review purposes only and does not constitute Agency policy.

3-164 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Evidence stream summary and interpretation

Inferences and summary
judgment

in dams treated from
GD 6-15.

Moroholosical
develooment

1 medium confidence
study (two independent
experiments)

• Increased fetal
incidences of skeletal
variations (i.e., absence
of fifth sternebrae at
>6.4 mg/kg-d

• Delayed ossification of
the phalanges at
>6.4 mg/kg-d

• Delayed braincase
ossification at
>0.03 mg/kg-d).

• Dose-response
gradient for skeletal
and braincase
ossification
variations.

• Consistent increase
in variations across
two medium
confidence
experiments

• Unclear biological
relevance of
variations as no
malformations were
reported.

• Potential
confounding of
skeletal and
phalanges
ossification
variations at doses
causing overt
toxicity.

Mechanistic evidence and supplemental information (see subsection above)

Biological events or
pathways

Primary evidence evaluated

• Key findings, interpretation, and limitations

Evidence stream judgment

Other evidence

Interpretation 1: PFDA causes developmental toxicity in embryonic zebrafish.

Key findings:

• Of the 139 PFAS chemicals tested, PFDA was the most potent for the induction of
teratogenic effects in zebrafish (Truong et al., 2022)

• Of the 7 PFAS chemicals tested, PFDA was the second most potent for the
induction of developmental effects in zebrafish. Spinal curvature, a malformation,
was commonly reported in zebrafish embryos exposed to PFDA (Ulhaa et al.,
2013).

Limitations: A comprehensive list of tested/observed developmental endpoints was

not provided.

The findings in zebrafish
provide some support for the
biological plausibility of the
developmental effects in
humans and animals.

This document is a draft for review purposes only and does not constitute Agency policy.

3-165 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

3.2.4. MALE REPRODUCTIVE EFFECTS
Human studies

There are nine epidemiology studies that examined the association between PFDA exposure
and male reproductive effects. The outcomes included in these studies were semen parameters,
reproductive hormones, timing of pubertal development, and anogenital distance. The studies are
described below.

Semen evaluations

Semen concentration and sperm motility and morphology were considered the core
endpoints for the assessment of semen parameters. Key issues for the assessment of semen
parameters involve sample collection and sample analysis. Samples should be collected after an
abstinence period of 2-7 days, and analysis should take place within two hours of collection and
follow guidelines established by the World Health Organization (WHO. 20101. While exposure
would ideally be measured during the period of spermatogenesis rather than concurrent with the
outcome, a cross-sectional design is considered adequate because the period of spermatogenesis in
humans is fairly short (74 days plus 12 days of maturation) fSigman et al.. 19971. the half-life of
PFDA is long, and there is no concern for reverse causality with this outcome because it is not
expected the semen quality would influence PFDA concentrations in blood.

Four cross-sectional studies examined the relationship between PFDA and semen quality.
Based on the above considerations, three were evaluated as medium confidence overall
(Figure 3-34), though one of these was considered uninformative for the core endpoint sperm
motility due to the overnight delay between collection and analysis fBuck Louis et al.. 20151. One
study analyzed male partners from a preconception cohort in the U.S. fBuck Louis etal.. 20151. one
study enrolled young adult men whose mothers were enrolled in a national pregnancy cohort
f Petersen etal.. 20221. and one enrolled healthy young man being considered for military service
(Toensen et al.. 20131. The remaining study was low confidence due to multiple identified
deficiencies and was focused on men seeking infertility assessment (Huang et al.. 2019a). All four
studies analyzed PFDA in serum used appropriate methods, and, thus, exposure misclassification is
expected to be minimal.

This document is a draft for review purposes only and does not constitute Agency policy.

3-166 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Huang, 2019, 5406374-

Joensen, 2013. 29191 SO

Louis, 2015, 2851139-

Petersen KU etal. 2022-

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)

Figure 3-58. Semen parameters epidemiology study evaluation heatmap.

Refer to HAWC for details on the study evaluation review: HAWC Human Semen
Parameters.

The results for the association between PFDA exposure and semen quality are presented in
Table 3-25. The studies analyzed the outcomes differently so the effect estimates are not directly
comparable. None of the results were statistically significant, but there was a suggestion of a
decrease in motility with increased exposure in Toensenetal. (20131 and in concentration in Huang
etal. f2019al. but not in Petersen etal. f20221. Because the methods used to assess motility was
considered critically deficient in Buck Louis etal. f20151. it was not possible to evaluate its
consistency with the other medium confidence studies. For concentration and morphology, there
was no clear decrease in the medium confidence studies. However, PFDA levels in both studies
were lower than levels of other measured PFAS (<0.5 ng/mL) and the exposure contrasts were
narrow, which introduces concerns regarding sensitivity, i.e., lack of ability to detect and
association if present.

Table 3-25. Associations between serum PFDA and semen parameters in
epidemiology studies

Reference;

study
confidence

Population

Median
exposure

(IQR)
(ng/mL)

Effect estimate

Concentration
(x 106/mL)

Motility (%
motile)

Morphology
(% normal)

Huang et
al. (2019a);
low

Cross-sectional
study of men
seeking infertility
assessment
(2009-2010); 57
men

0.0 (range
0.0-1.2)

P (95% CI) for
1 In-unit
increase in
serum PFDA

-21.59 (-77.91,
34.73)

5.96 (-11.58,
23.50)

-0.02 (-0.10,
0.07)

This document is a draft for review purposes only and does not constitute Agency policy.

3-167 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Reference;

study
confidence

Population

Median
exposure

(IQR)
(ng/mL)

Effect estimate

Concentration
(x 106/mL)

Motility (%
motile)

Morphology
(% normal)

Petersen
et al.
(2022),
medium

Cross-sectional
analysis within
cohort of general
population men
(2017-2019),
Denmark; 1,041
men (18-20 yr)

0.2 (5th-
95th: 0.1-
0.3)

% difference
(95% CI) for
tertiles of PFDA
vsTl

T2: 3 (-9, 17)
T3: -3 (-15, 11)

T2: -1 (-6, 6)
T3: -3 (-9, 3)

T2: -1 (-11,10)
T3: 2 (-8,13)

Joensen et
al. (2013);
medium

Cross-sectional
study of men
evaluated for
military service
(2008-2009),
Denmark; 247
men (18-22 yr)

0.4 (0.3-
0.5)

P (95% CI) for
1-unit increase
in serum PFDA

Cubic root
transformed
0.22 (-0.76,
1.19)

Square
transformed
-1343 (-2759,
73.69)

Square root
transformed
-0.097 (-0.88,
0.69)

Buck Louis
et al.
(2015);
medium

Cross-sectional
analysis within
preconception
cohort (2005-
2009), U.S.;
462 men

0.5 (0.3-
0.6)

P (95% CI) for
1 In-unit
increase in
serum PFDA

-1.06 (-30.5,
28.3)

Uninformative

5.80 (-1.31,
12.9)

*p < 0.05.

Reproductive hormones

Testosterone and estradiol were considered the primary endpoints for male reproductive
hormones. Progesterone, LH, FSH, and SHBG were also reviewed where available. Key issues for
the evaluation of these studies were sample collection and processing (see Figure 3-59). For
testosterone, LH, and FSH, due to diurnal variation, blood sample collection should be in the
morning, and if not, time of collection should be accounted for in the analysis. If there is no
consideration of time of collection for these hormones, the study is classified as deficient for
outcome ascertainment and low confidence overall. A cross-sectional design was considered
appropriate for this outcome since levels of these hormones are capable of being rapidly
upregulated or downregulated and they are not expected to directly bind to or otherwise interact
with circulating PFAS.

Seven studies (eight publications) examined the relationship between PFDA and
reproductive hormones. Three studies were medium confidence cross-sectional studies in adults,
including Toensenetal. f20131 and Petersen etal. f20221. cross-sectional studies of young adult
men described above. An analysis of NHANES data in adult men fXie etal.. 20211 was also medium
confidence for estradiol but low confidence for testosterone due to potential outcome
misclassification as previously described. A cross-sectional study in adolescents (aged 13-15
years) (reported in Zhou etal. (2016) and Zhou etal. (2017b)) was low confidence due to concerns

This document is a draft for review purposes only and does not constitute Agency policy.

3-168 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

for confounding (e.g., pubertal indicators were not considered). Three studies, one a birth cohort in
Denmark fTensen et al.. 2020b) and two cross-sectional studies in China (Liu etal.. 2020b: Yao etal..
20191 examined associations in infants. Yao etal. f20191 and Yao etal. f20191 were low confidence
due to not accounting for time of day of sample collections (both studies) and potential concerns for
confounding fYao etal.. 20191. Liu etal. f2020bl was medium confidence due to less concern for
diurnal variation of the included hormone (progesterone).

Jensen, 2020, 6311643-

'
+

Joensen, 2013, 2919160-

Liu, 2020, 6569227-

Petersen KU et al. 2022-

Xie, 2021,8437891-

Yao Q et al. 2019, 5187556-

Zhou, 2016, 3856472-

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 3-59. Male reproductive hormones epidemiology study evaluation
heatmap. Refer to HAWC for details on the study evaluation review: HAWC Human
Male Reproductive Hormones.

*Outcome-specific ratings differed for this domain.

Given the differences in populations (adults, adolescents, newborns), evaluation of
consistency across studies is not straightforward. For testosterone, inverse associations between
PFDA exposure and testosterone levels were observed in two studies. Among the two medium
confidence studies for this outcome, Toensen etal. (20131 observed a decrease in log-transformed
testosterone with higher PFDA exposure in adult men, though this was not statistically significant
(P (95% CI) = -0.17 (-0.41, 0.0711. Petersen et al. f20221 reported no association with association.
Also in adults, but low confidence for testosterone, Xie etal. f20211 found positive associations
between PFDA exposure and free and total testosterone (statistically significant for free
testosterone, with exposure gradient observed across quartiles). In adolescent boys, the low
confidence study by Zhou etal. (20161 reported an inverse association ((3 (95% CI) = -0.26
(-0.41, -0.10)). In infants, one study fTensen et al.. 2020bl reported a positive association between
PFDA exposure and testosterone ((3=0.37, 95% CI: -0.11, 0.84, p=0.1) while no association was
observed in Yao etal. f20191.

This document is a draft for review purposes only and does not constitute Agency policy.

3-169 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

For estradiol, in adults in Toensen etal. (20131. there was also a decrease with higher PFDA
exposure ((3 (95% CI) = -0.22 (-0.48, 0.002)), but this was not observed in the other two studies in
adults fPetersen etal.. 2022: Xie etal.. 20211 or in adolescents in Zhou etal. f20161 or infants in Yao
etal. f20191. Toensen et al. f20131 also examined several other reproductive hormones and sex
hormone binding globulin (SHBG) in young men and found no evidence of association with PFDA
exposure for SHBG, luteinizing hormone, or inhibin-B, but did report a positive association with
follicle stimulating hormone (FSH) ((3 (95% CI) = 0.42 (-0.005, 0.85)). The increase in FSH would
be consistent with an increase in gonadotropin production as a compensatory response to a
decrease in testosterone. However, Petersen etal. (20221 found no association with FSH, LH, or
SHBG. In flensen etal.. 2020bl. inverse associations, though not statistically significant were
observed with DHEA, DHEAS, and Androstenedione. Liu etal. f2020bl found no association with
progesterone.

Pubertal development

Pubertal development is primarily assessed using established criteria, such as Tanner stage
ratings. For boys, Tanner staging involves evaluation of the development of genitalia (scrotum
appearance, testes, and penile size) and pubic hair. Stage 1 represents prepubertal development;
Stage 2, the onset of pubertal development, and Stage 5 represents full sexual maturity. Two
medium confidence birth cohorts in Denmark fErnstetal.. 20191 and the U.S. fCarwile etal.. 20211
examined timing of pubertal development with PFDA exposure. Ernst etal. (20191 used maternal
exposure measured in blood and prospectively identified pubertal onset with follow-up checks
every six months. In boys, they reported that there was no clear pattern of association between
PFDA exposure and Tanner stages of genital development or pubic hair, or other markers of
pubertal development such as axillary hair, acne, voice break, or first nocturnal ejaculation when
exposure was analyzed in tertiles. For each outcome, the mean age of onset was later in the middle
(0.16-0.21 ng/mL) vs. the lowest (0.08-0.15 ng/mL) tertile, but earlier in the highest tertile (0.22-
0.9 ng/mL). This pattern was also observed with a combined puberty indicator outcome, with boys
in the middle tertile reaching the indicator 4.59 months later (95% CI: -0.93,10.11) and the highest
tertile 2.83 months earlier (95% CI: -8.43, 2.77) than the lowest tertile. Carwile etal. (20211 used
exposure measured during mid-childhood (median 8 years) with follow-up to early adolescence
(median 13 years). Using a pubertal development score based on parental responses to scales of
multiple pubertal markers (voice deepening, body hair growth, facial hair growth, acne, and growth
spurt), they reported no association with PFDA exposure. This was consistent with their findings
for older age at peak heigh velocity (used as a proxy for pubertal development). Exposure contrast
was narrow in both studies (median 0.2 ng/mL, 10th-90th percentile 0.1-0.3 in Ernst etal. (20191.
0.3, 25th-75th percentile 0.2-0.5 in Carwile etal. (202111. which may have reduced study
sensitivity.

This document is a draft for review purposes only and does not constitute Agency policy.

3-170 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Carwile, 2021, 9959594

Ernst, 2019, 5080529-

Cy1

-p"

;^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)

Figure 3-60. Male pubertal development epidemiology study evaluation
heatmap. Refer to HAWC for details on the study evaluation review: HAWC Human
Male Pubertal Development.

Summary of human studies

Overall, there is inconsistent evidence for male reproductive effects of PFDA exposure. One
medium confidence study in adult men found reduced sperm motility and testosterone (Joensenet
al.. 20131 and one low confidence study also found an inverse association in adolescents (Zhou et
al.. 20161. This is coherent with an inverse association with anogenital distance in one medium
confidence study fTian etal.. 20191 (see Section 3.2.3 on Developmental Effects). However, the
other available studies did not report consistent findings for semen parameters and reproductive
hormones. No clear association was observed with estradiol or pubertal development.

Animal studies

Only one animal toxicity study evaluated male reproductive effects after PFDA exposure
fNTP. 20181. This study examined the following endpoints after a 28-day gavage exposure
(0, 0.156, 0.312, 0.625,1.25, and 2.5 mg/kg-day) in 7- to 8-week-old male Sprague-Dawley rats:
sperm evaluations, histopathology, hormone levels, and organ weights. The endpoints evaluated by
NTP f20181 are considered to be reliable measures for assessing male reproductive toxicity (Creasy
and Chapin. 2018: Creasy etal.. 2012: Sellers etal.. 2007: U.S. EPA. 1996b). The NTP (20181 study
was evaluated as high confidence for most endpoints examined with no notable concerns in any of
the study evaluation domains (see Figure 3-61). Concerns for potential insensitivity were identified
for sperm measures as the exposure duration (28 days) used for this experiment was insufficient to
fully detect potential effects on sperm development, resulting in a low confidence rating; this
potential bias is towards the null. In rats, spermatogenesis takes approximately 8 weeks for germ
cells to mature from spermatogonia to spermatozoa fCreasv and Chapin. 20181.

This document is a draft for review purposes only and does not constitute Agency policy.

3-171 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Reporting quality -
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -\
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity -
Results presentation -
Overall confidence

Legend

| Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)

S Critically deficient (metric) or Uninformative (overall)
Not reported
* Multiple judgments exist

Figure 3-61, Evaluation results for animal study assessing effects of PFDA
exposure on male reproductions. Refer to HAWC for details on the study
evaluation review: HAWC NTP f2018I

Sperm evaluations

Testicular and epididymal sperm counts and testicular sperm motility were only measured
for the three highest dose groups (0.625,1.25 and 2.5 mg/kg-day) (see Figure 3-62). Testicular
sperm counts are indicative of changes in sperm production in the testis, while epididymal counts
indicate both changes in testicLilar sperm production and storage of sperm in the epididymis;
therefore, both measures are considered informative for evaluating effects on sperm parameters
fCreasv and Chapin. 2018: Creasy et al.. 20121. Testicular sperm counts (absolute and relative to
organ weight) decreased dose- dependently at 0.625 and 1.25 mg/kg-day (-10% and -19-21%
change compared to controls, respectively) but not at the highest dose group (2.5 mg/kg-day). As
such, a clear trend for testicular sperm counts could not be established. A significant trend was
reported for cauda epididymal sperm counts with decreases of 11-30% compared to controls
across 0.625-2.5 mg/kg-day. NTP (2018) also reported sperm counts normalized to cauda
epididymis weight and observed no treatment-related effects (data not shown in Figure 3-62).
However, this is not considered a sensitive measure as sperm contributes to epididymal weight and
reporting findings as a ratio may mask reductions in sperm number fU.S. EPA. 1996b! A non-
statistically significant decrease in testicular sperm motility of 11% compared to controls was

This document is a draft for review purposes only and does not constitute Agency policy,

3-172 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1 reported at 2.5 mg/kg-day, but there was no clear dose-response effect In summary, the dose-

2 related decreases in sperm counts in the epididymis suggest that PFDA can affect sperm

3 parameters at doses > 0.625 mg/kg-day after 28-day exposure.

4 The findings on sperm measures from NTP f20181 are interpreted with caution as

5 sensitivity concerns for these outcomes are based on the exposure duration used in this study

6 which did not capture the entire process of spermatogenesis (approximately 8 weeks in rats)

7 (Creasy and Chapin. 20181.

This document is a draft for review purposes only and does not constitute Agency policy.

3-173 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study
Design

Outcome Confidence Target Organ

Endpoint Name

Animal Description

Trend Test Result

Response Units

Dose

(mg/kg-day)

NTP, 2018, 4309127 28 Day Oral

Low confidence Testes

Testicular Spermatid Count

Rat, Sprague-Dawley (Harlan) ( ')

not significant

10A6

0.625
1.25

O Statistically signifcant
0 Percent control response

H 95% CI

2.5

Testicular Spermatid Count per mg
Testis

Rat, Sprague-Dawley (Harlan) (o)

not significant

10A3/mg

0.625
1.25

2.5

Percent Motile Sperm

Rat, Sprague-Dawley (Harlan) ( ")

not significant

percent

0.625

1.25

2.5

Epididymis

Cauda Epididymis Sperm Count

Rat, Sprague-Dawley (Harlan) ( )

significant

millions

0.625

PFDA Sperm Evaluations

-50 -40 -30 -20 -10 0 10 20 30
Percent Control Response

Figure 3-62. Effects on sperm evaluations following exposure to PFDA in short-term oral studies in animals

(results can be viewed by clicking the HAWC link].

This document is a draft for review purposes only and does not constitute Agency policy,

3-174 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Histopathologv

Testicular and epididymal lesions were reported in the 28-day rat study by NTP f2018I
The testes were examined in all dose groups for histopathologic responses (see Figure 3-63).
Minimal to mild atrophy of the interstitial (Leydig) cells was observed in nearly all the rats exposed
to the two highest PFDA dose groups (8/10 and 10/10 for 1.25 and 2.5 mg/kg-day, respectively)
but not in the controls. Leydig cell atrophy is a response coherent with reduced sperm production
(Creasy and Chapin. 2018: Creasy etal.. 2012) and indicative of reduced androgen levels, which
were also observed in this study (see synthesis of Reproductive hormones in this Section). Mild
degeneration of the germinal epithelium and spermatid retention within the seminiferous tubules
was also increased in 4/10 rats from the high dose group; control group incidence was 1/10 and
0/10, respectively. The epididymis was examined in the three highest dose groups (0.625,12.5 and
2.5 mg/kg-day) (see Figure 3-63). Only the highest dose group (2.5 mg/kg-day) displayed mild
duct germ cell exfoliation in 4/10 rats examined compared to 1/10 rats in the control group and a
single marked case of hypospermia (1/10 rats) not observed in the controls. Sperm granuloma was
found in 1/10 rats in the controls but not in the exposed animals (data not shown in Figure 3-63).
NTP T20181 did not observe any histopathological effects on the preputial gland, seminal vesicle,
and prostate when examining animals in the control and high dose groups. In summary, there is
consistent evidence of histopathological observations indicative of mild degenerative changes in
the testes and epididymis at doses > 1.25 mg/kg-day after 28-day exposure. Note that these doses
are associated with significant body weight changes (see Evidence Integration section below for a
discussion on potential confounding due to co-occurring systemic toxicity at doses causing some
PFDA-induced male reproductive effects).

This document is a draft for review purposes only and does not constitute Agency policy.

3-175 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study Outcome
Design Confidence

Target
Organ

Endpoint Name

Animal Description

Trend Test
Result

Incidence

Dose

(mg/kg-day)

NTP, 2018, 4309127 28 Day Oral High confidence

Testes

Germinal Epithelium Degeneration

Rat, Sprague-Dawley (Harlan)

significant

1/10(10.0%)
0/10 (0.0%)

0.156

0.312

0.625

1.25

4/10 (40.0%)

2.5

Interstitial Cell Atrophy

Rat, Sprague-Dawley (Harlan) (,?)

significant

0/10 (0.0%)

0.156

0.312

0.625

8/10 (80.0%)

1.25

10/10(100.0%)

2.5

Seminiferous Tubule Spermatid Retention

Rat, Sprague-Dawley (Harlan) (e)

significant

0/10 (0.0%)

0.156

0.312

0.625

1.25

4/10 (40.0%)

2.5

Epididymis

Exfoliated Germ Cell, Epididymal Duct

Rat, Sprague-Dawley (Harlan) (
-------
Toxicological Review ofPerfluorodecanoicAcid and Related Salts

1 Reproductive hormones

2 NTP f20181 evaluated serum testosterone in all dose groups at study termination (see

3 Figure 3-64). A significant trend was reported with 25, 64, and 75% decreases in serum

4 testosterone when compared to controls for the 0.625,1.25, and 2.5 mg/kg-day dose groups,

5 respectively. Testosterone is essential for the development and maturation of the male

6 reproductive system, and it also plays a role in maintaining spermatogenesis and reproductive

7 functions in adults fToor and Sikka. 20171. The changes in serum testosterone levels at doses >

8 0.625 mg/kg-day are concordant with the reductions in sperm counts and Leydig cell damage in

9 adult male rats exposed to PFDA for 28 days (see synthesis on Sperm evaluations and
10 Histopathology in this section).

This document is a draft for review purposes only and does not constitute Agency policy.

3-177 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study Outcome Confidence Target Organ Endpoint Name Animal Description

Design

Trend Test Result Response Units Dose

(mg/kg-day)

NTP, 2018, 4309127 28 Day Oral High confidence Blood Testosterone (T) Rat, Sprague-Dawley (Harlan) (5) significant ng/mL

O Statistically significant
0 Percent control response
|—| 95% CI

significant ng/mL

PFDA Male Testosterone Levels

Percent Control Response

Figure 3-64. Effects on serum testosterone levels following exposure to PFDA in short-term oral studies in
animals (results can be viewed by clicking the HAWC link).

This document is a draft for review purposes only and does not constitute Agency policy.

3-178" DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review ofPerfluorodecanoicAcid and Related Salts

Organ weight

The right testis was measured at study termination in all dose groups, while epididymis
weights (both whole and the cauda segments) were evaluated in the three highest dose groups
(0.625,12.5, and 2.5 mg/kg-day) (NTP. 20181 (see Figure 3-65). Absolute weights are the
preferred measure for testis and epididymis as these organs appeared to be conserved even with
body weight changes (Creasy and Chapin. 2018: U.S. EPA. 1996b). A decreasing trend (p < 0.01) in
absolute testis weight was reported across the doses, reaching a -13% change compared to
controls at 2.5 mg/kg-day. Absolute epididymis weights for whole and cauda segments also
showed a decreasing trend (p < 0.01) and reported -10-11% and -23-25% change relative to
controls for the 1.25 and 2.5 mg/kg-day dose groups, respectively. Decreases in epididymis weight,
particularly in the cauda segment, may reflect reductions in sperm counts (Creasy and Chapin.
2018: Evans and Ganiam. 2011). which was observed to occur at similar doses (see synthesis on
Sperm evaluations in this Section). Overall, the data shows consistent dose-related decreases in
organ weights in the testis and epididymis at > 0.625 mg/kg-day after short-term exposure to
PFDA.

This document is a draft for review purposes only and does not constitute Agency policy.

3-179 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Study
Design

Outcome Confidence Target Organ

Endpoint Name

Animal Description

Trend Test Result

Response Units Dose

(mg/kg-day)

NTP, 2018, 4309127 28 Day Oral

High confidence Testes

Right Testis Weight, Absolute

Rat, Sprague-Dawley (Harlan) (-¦')

significant

g o

0.156

0.312

0.625

1.25

2.5

Epididymis

Cauda Epididymis Weight, Absolute

Rat, Sprague-Dawley (Harlan) ( )

significant

g o

0.625

1.25

2.5

Epididymis Weight, Absolute

Rat, Sprague-Dawley (Harlan) ( !)

significant

g o

0.625

PFDA Male Reproductive Organ Weights

0 Statistically significant
£ Percent control response
M 95% CI

-40 -35 -30 -25 -20 -15 -10 -5 0 5 10
Percent Control Response

Figure 3-65. Effects on male reproductive organ weights following exposure to PFDA in short-term oral studies in
animals (results can be viewed by clicking the HAWC link).

This document is a draft for review purposes only and does not constitute Agency policy,

3-180 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Mechanistic studies and supplemental information

Several studies have evaluated the potential mechanisms by which PFDA exposure may lead
to male reproductive effects. Experimental studies have investigated PFDA-induced effects on
Leydig cell steroidogenesis, androgen (AR) and estrogen (ER) receptor functions, aromatase
activity and androgen metabolism and excretion and the potential impact of indirect systemic
toxicity on the male reproductive effects of this chemical.

In vitro cell culture studies have evaluated PFDA-induced effects on Leydig cell functions
and steroidogenesis. Leydig cells are the primary site of testosterone synthesis (Creasy and Chapin.
20181. Cholesterol uptake by the mitochondria in Leydig cells is a critical step in human chorionic
gonadotropin (hCG)-induced testosterone production fScottetal.. 20091. In both immortalized
mouse (MA-10) Leydig cells (LCs) and primary rat LCs, exposure to PFDA significantly decreased
mitochondrial cholesterol uptake, and hCG-stimulated testosterone synthesis fBouirad et al.. 20001.
The PFDA exposure levels affecting hormone synthesis in MA-10 cells did not lead to increased
cytotoxicity measured as DNA damage, protein synthesis, and mitochondrial integrity (Bouirad et
al.. 20001. In contrast, PFDA showed a lack of activity in HTS assays from the EPA's ToxCast and
Tox21 database evaluating steroid hormone biosynthesis, including glucocorticoids, androgens,
estrogens, and progestogens in adrenal gland H295R cells fU.S. EPA f2019bl: refer to Appendix E.2
for more details on the HTS results).

The in vitro observations of PFDA-induced effects on Leydig cell functions are consistent
with both the 28-day gavage study in rats by NTP (20181 discussed above and high dose, i.p.
injection studies that exposed rodents (predominantly rats) to single PFDA doses ranging from 20
to 400 mg/kg and evaluated effects on histopathology, androgen levels, and androgen-responsive
reproductive organ weights after observational periods of 7 to 28 days fBookstaff et al.. 1990: Van
Rafelghem etal.. 1987b: Olson and Andersen. 19831. The i.p. injections studies report decreases in
serum testosterone and 5-a-dihydrotestosterone levels fBookstaff etal.. 19901. altered testicular
testosterone production fBookstaff et al.. 19901. and reduced androgen-responsive reproductive
organ weights in rats fBookstaff etal.. 1990: Olson and Andersen. 19831. Furthermore, these
studies report that PFDA exposure was associated with increased incidence of histopathological
effects considered indicative of androgen disruption and spermatogenic disturbance (Creasy and
Chapin. 2018: Creasy etal.. 20121. Effects observed in rats include increased seminal vesicle and
prostatic acini atrophy, and reduced seminal vesicle epithelial cell height, fBookstaff etal.. 19901.
and while mice appeared to be resistant to seminiferous tubule degeneration, rats, hamsters, and
guinea pigs were responsive to this PFDA-induced effect (Van Rafelghem et al.. 1987b).

Another mechanism by which PFDA could alter male reproductive function is via increased
hepatic metabolism and excretion of androgens or metabolic precursors such as cholesterol.
Bookstaff et al. (1990) performed an experiment in which castrated Sprague-Dawley rats were
supplemented with testosterone via sustained release capsules and then treated with vehicle or
PFDA. They observed that acute PFDA exposure (20-80 mg/kg, i.p.) had no effect on serum

This document is a draft for review purposes only and does not constitute Agency policy.

3-181 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

testosterone levels when the source of this hormone was the capsule rather than the testes. These
findings suggest that PFDA does not impact hepatic androgen metabolism and excretion, and that
decreases in serum testosterone levels observed after exposure are likely caused by a disruption in
steroidogenesis in the testis. This argument is supported by the reductions in testosterone
secretion in response to hCG stimulation in testicular tissue harvested from PFDA-exposed rats
evaluated in the same study fBookstaffetal.. 19901 and inhibition of hGC-mediated steroidogenesis
in cell culture rodent models using immortalized and primary Leydig cells described above
(Bouirad et al.. 20001.

Overall, the findings from available in vivo and cell culture studies provide support for an
effect of PFDA exposure on Leydig cell functions ultimately resulting in reduced steroidogenesis.

Separately, PFDA-induced effects on AR and ER functions and aromatase activity have been
evaluated in in vitro cell culture studies and HTS assays from the EPA's ToxCast and Tox21
platforms U.S. EPA (2019b): refer to Appendix E.2 for more details on the HTS results). AR and ER
are known to regulate male reproductive functions (Wan etal.. 2013: Wilson etal.. 20081 and
aromatase is a key enzyme in the conversion of androgens to estrogens, which is important for
sexual development and differentiation (Sweeney etal.. 2015: Hotchkiss et al.. 2008: Tones etal..
20061. Disruption of AR transactivation has been demonstrated in Chinese hamster ovary cells
(CHO-K1) fKieldsen and Bonefeld-largensen. 20131 and androgen sensitive TARM-Luc cells
fMcComb etal.. 20191 at PFDA concentrations that did not induce cytotoxicity. No significant effects
on ER transactivation were observed in human breast adenocarcinoma MCF-7 cells with PFDA
exposure alone (Li etal.. 2020b: Kieldsen and Bonefeld-l0rgensen. 20131 but in combination with
17(3-estradiol, PFDA displayed antiestrogenic activity measured by inhibition of ER transactivation
and downregulation of ER-responsive genes at non-cytotoxic concentrations (Li etal.. 2020b). In
HTS assays profiling AR and ER functions across multiple endpoints and in vitro test models, PFDA
displayed low activity for these receptors at concentrations closely associated with cytotoxicity
(Table E-3 in Appendix E.2). PFDA was active in 2 out of 17 AR assays (displaying binding activity
in rat prostrate tissue and induction of cell proliferation in human prostate carcinoma 22Rvl cells)
and in 2 out of 21 assays profiling the ERa (1 out of 2 independent assays measuring transcriptional
activity in HepG2 cells and an antagonist transactivation assays in human embryonic kidney
HEK293T cells). Consistent with the HTS results, the ToxCast model predictions suggest that PFDA
is inactive for both AR/ER agonist and antagonist activities (Table E-4 in Appendix E.2). Lastly,
PFDA exposure decreased aromatase activity in the human choriocarcinoma JEG-3 cell line under
conditions of cytotoxicity fKieldsen and Bonefeld-largensen. 20131 but no activity in a HTS assay
measuring aromatase inhibition in human breast cancer MCF-7 cells (Table E-5 in Appendix C).
Taken together, findings from in vitro cell culture studies and HTS assays do not provide consistent
and reliable evidence for potential effects of PFDA on AR or ER functions, or aromatase activity.
However, for the most part, these in vitro cell models are not derived from the male reproductive

This document is a draft for review purposes only and does not constitute Agency policy.

3-182 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

system and variability in the cellular/tissue environment may lead to differences in hormone
receptor/enzyme functions (Leehv etal.. 2016: Abdel-Hafiz and Horwitz. 20141.

In addition to the mechanisms described above, PFDA-induced wasting syndrome (see
General toxicity effects; Section 3.2.10) may indirectly affect the male reproductive system. This is
because severe decreases in body weight are known to alter reproductive functions fCreasv and
Chapin. 2018: U.S. EPA. 1996bl. Decreased body weight and food consumption were observed in
acute, i.p. injection studies at doses >40 mg/kg and lethality were reported in some studies at doses
>50 mg/kg (Bookstaff et al.. 1990: Van Rafelghem etal.. 1987b: Olson and Andersen. 19831.
Bookstaff et al. (19901 addressed the impact of PFDA-induced changes in body weight on male
reproductive endpoints by adding pair fed control rats that were weight-matched to each PFDA
treatment groups. The authors observed that single exposure to 20, 40, or 80 mg/kg of PFDA via
i.p. injection significantly decreased serum testosterone and DHT, testicular testosterone
production, seminal vesicle and prostate weights, and seminal vesicle epithelial cell height In pair
fed control animals, there were no significant responses in the male reproductive system except in
the group matched to the highest PFDA dose (80 mg/kg), which was associated with large
reductions in food intake (44%) and body weight (72%) and observed responses were attenuated
compared to PFDA exposure. These results indicate that PFDA-induced effects at the low and
medium doses were direct reproductive system effects and not secondary to chemical-induced
systemic effects. The body weight reductions in male rats observed in the 28-day gavage study at
1.25-2.5 mg/kg-day are consistent with moderate body weight changes (21-38%) that are not
associated with confounding effects from overt systemic toxicity in supplemental studies tailored to
examine that potential linkage.

Overall, the available evidence from in vivo and cell culture studies provides evidence of a
biologically plausible mechanism for PFDA-induced adverse responses in the male reproductive
system by disruption of steroidogenesis in Leydig cells, which in turn could impair reproductive
functions and spermatogenesis. Specifically, it appears that PFDA exposure can disrupt androgen
production in Leydig cells, which may lead to downstream histopathological effects, organ weight
changes, and decreased spermatogenesis. Disruptions in androgen levels/production is a known
pathway for chemical-induced alterations in spermatogenesis (Toor and Sikka. 2017: Sharpe.
20101. This support for biological plausibility is derived from studies in exposed animals and in
vitro animal models; studies informing the relatability of these data to exposed humans are
currently unavailable.

Evidence integration

The evidence of an association between PFDA exposure and male reproductive effects in
humans is limited to two medium (Tian etal.. 2019: Toensen etal.. 20131 and one low confidence
study (Zhou etal.. 20161. with findings suggesting potential decreases in testosterone, decreased
sperm motility, and anogenital distance (see Section 3.2.3 on Developmental Effects) with higher

This document is a draft for review purposes only and does not constitute Agency policy.

3-183 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

PFDA exposure. There are concerns over inconsistency and imprecision, thus, the evidence is
considered indeterminate.

The available evidence from a 28-day gavage study in rats and supportive data from i.p.
injection and cell culture studies in rodents provide moderate evidence of male reproductive
toxicity in animals with PFDA exposure. The 28-day rat study showed coherent effects across
several relevant endpoints, including sperm evaluations, histopathology, hormone levels and organ
weights (NTP. 20181. with most effects observed at doses below those shown to cause overt
toxicity. Adverse histopathological changes were observed at doses associated with body weight
decrements of potential concern. The study methods were considered high confidence for all
endpoints other than sperm evaluations, which were considered potentially insensitive due to an
inadequate exposure duration (i.e., biased towards the null; confidence is reduced specifically in the
interpreted reliability of null findings [i.e., sperm motility]). A consistent pattern of decreased
testicular and epididymal sperm counts occurred at >0.625 mg/kg-day, but only the effects in the
epididymis were dose related. Dose-related decreases in serum testosterone levels and testicular
and epididymal weights were also reported in rats at >0.625 mg/kg-day. The reductions in sperm
counts, serum testosterone levels and organ weights are coherent with the mild degenerative
changes found in testes and epididymis at similar doses, particularly Leydig cell atrophy, which is
associated with androgen deficiency and decreased spermatogenesis fCreasy etal.. 20121.
Consistent effects on serum androgen levels, male reproductive organ weights, and histopathology
were observed in rodents exposed to high doses of PFDA (>20 mg/kg) in, single, i.p. injection
studies. The adverse effects observed in the in vivo oral and i.p. exposure studies are biologically
consistent with a potential mechanism for PFDA-induced reproductive effects in which alterations
in Leydig cell functions result in decreased steroidogenesis and androgen levels (see synthesis on
Mechanistic studies and supplemental information above for more details).

Limitations of the animal evidence base include the availability of only a single, short-term
oral exposure study in a single species, and uncertainties regarding the potential impact of systemic
toxicity, particularly with regard to the observed histopathological effects. Significant reductions in
body weight were reported in the highest dose groups in the 28-day gavage study (21% at
1.25 mg/kg-day and 38% at 2.5 mg/kg-day; see Section 3.2.9 on General toxicity effects for more
details) (NTP. 20181. However, concern for nonspecific effects on the male reproductive system is
attenuated by the observed dose-related effects (i.e., sperm counts, testosterone levels and organ
weights) at a lower PFDA dose, not associated with body weight changes (0.625 mg/kg-day).
Likewise, an i.p. injection study that examined potential effects of PFDA-induced "wasting
syndrome" using pair-fed control rats observed androgenic deficiency and male reproductive
toxicity at 20 and 40 mg/kg that were independent from severe body weight depression at the
highest dose (72% at 80 mg/kg) (Bookstaff etal.. 19901. With respect to in vitro evidence, a
general lack of in vitro models derived from the male reproductive system, and models restricted to
rodents, limits the ability of the available evidence to inform potential pathways involved in PFDA-

This document is a draft for review purposes only and does not constitute Agency policy.

3-184 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

induced male reproductive toxicity and to elucidate conserved mechanisms across species,
including humans. Nonetheless, the mechanistic information from acute i.p. and in vitro animal
studies is both consistent and coherent with the oral exposure study evidence, and therefore,
provides support for the biological plausibility of the phenotypic responses. In the absence of
information to the contrary and given the conserved role of androgen-dependent pathways in male
reproductive functions across species (including humans), the available evidence is considered to
be relevant to humans.

A potentially susceptible population for PFDA-induced male reproductive effects are young
individuals exposed during critical developmental life stages (e.g. the masculinization
programming, which occurs prior to the differentiation of androgen-sensitive tissues and
determines penis size and anogenital distance fDentetal.. 20151. although no such studies were
available in the current animal evidence base and few epidemiological studies examining pubertal
development and anogenital distance were available. Androgens play a critical role in the normal
development of the male reproductive system and disruptions caused by exposures to reproductive
toxicants during gestation and early post-natal life stages can lead to agenesis of the male
reproductive system and/or infertility (Foster and Gray. 2013: Sharpe. 2010: Scott etal.. 20091.

Taken together, available evidence indicates that PFDA is likely to cause male reproductive
effects in humans under sufficient exposure conditions (see Table 3-26). This conclusion is based
primarily on a constellation of coherent evidence from a high confidence study in animals exposed
to 0.625-2.5 mg/kg-day for 28 days, with some support for biological plausibility provided by
mechanistic evidence from i.p. and cell culture models. Although no direct information on the
human relevance of the animal evidence is available, many aspects of the male reproductive system
are conserved across species, and the limited sensitivity in human studies may explain the lack of
associations observed. Uncertainties in the database of PFDA-induced male reproductive toxicity
includes the absence of subchronic, chronic, developmental, or multigenerational studies testing
these outcomes in animals (which, overall, are anticipated to be more sensitive than the available
short-term study design), and a general lack of adequate epidemiological or toxicological studies
evaluating the potential for effects of early life PFDA exposure on male reproductive system
development.

This document is a draft for review purposes only and does not constitute Agency policy.

3-185 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-26. Evidence profile table for PFDA exposure and male reproductive effects

Evidence stream summary and interpretation

Evidence integration
summary judgment

Evidence from studies of exposed humans (see Section 3.2.4: Human studies)

0®Q

Evidence indicates
(likely)

Primary basis:

Single, short-term study
(high confidence) in rats,
generally at > 0.625
mg/kg-d PFDA

Human relevance:

Effects in rats are
presumed relevant to
humans based on the
conserved role of
androgen-dependent
pathways in male
reproductive functions
across species.

Cross-stream coherence:
N/A, human evidence is
indeterminate.

Susceptible populations
and lifestages:

Studies and
confidence

Summary and key findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

Semen evaluations
3 medium and 1 low
confidence cross-
sectional studies (1 is
uninformative for
motility)

• Decreased motility
with increased
exposure in Joensen
etal. (2013).

• No clear decrease in
concentration or
morphology in three
medium confidence
studies, but sensitivity
is low.

• Large effect size for
motility in medium
confidence study

• Unexplained
inconsistency in
medium confidence
studies for motility

• Imprecision

©QQQ

Indeterminate

Coherent results in semen motility and
testosterone across a medium and a
low confidence study; inconsistency
and imprecision add uncertainty.

Reproductive
hormones
For estradiol: 2
medium and 1 low
confidence cross-
sectional studies

For testosterone: 1
medium and 3 low
confidence studies

• Decreased

testosterone in one of
three studies of adults
(one of two medium
confidence) and one
low confidence study
of adolescents. No
inverse association
observed in two
studies of infants.

• No factors noted

• Unexplained
inconsistency in
medium confidence
studies

• Imprecision

This document is a draft for review purposes only and does not constitute Agency policy.

3-186 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration
summary judgment

Pubertal development

• In one study, for
several indicators of
puberty, mean age of
onset was later in
middle vs. lowest
tertile of exposure,
but earlier in the
highest tertile. The
other study reported
no association with
timing of puberty.

• No factors noted

• Unexplained
inconsistency

Based on the potential
for exposure to cause
impaired androgen
function, males exposed
during critical windows
of androgen-dependent
development may be
susceptible.

Other inferences:
Mechanistic evidence
from rodent i.p. studies
and cell culture models
suggest that male
reproductive toxicity is a
primary target for PFDA
(likely through disruption
of Leydig cells and
steroidogenesis), even at
doses associated with
overt systemic toxicity
(i.e., moderate body
weight decreases).

2 medium confidence
cohort studies

Evidence from in vivo animal studies (see Section 3.2.4: Animal studies)

Studies and
confidence

Summary and key findings

Factors that increase
certainty

Factors that decrease
certainty

Evidence stream judgment

Sperm evaluations

1 low confidence study
(due to insensitivity) in
rats exposed for 28
days

• Decreases in testicular
and epididymal sperm
counts at

>0.625 mg/kg-d

• No effects on sperm
motility

• Low confidence (due
to the potential
insensitivity of a short
exposure duration) is
mitigated by
consistent effects

• Consistent effects for
decreased sperm
count across tissues

• Dose-response
gradient for
epididymal sperm
counts

• Lack of expected
dose-response for
testicular sperm
counts

0®Q

Moderate

Coherent effects across sperm counts,
serum testosterone levels and male
reproductive histopathology and organ
weights in a single, high confidence
study; some concerns about
insensitivity due to short-term
exposure.

Histopatholosv

• Mild degenerative
lesions in testes and
epididymis at >1.25
mg/kg-d

• Consistent pattern of
lesions across tissues

• Leydig cell atrophy is
coherent with
decreased sperm

• Potential

confounding by body
weight decreases,
although this concern
is mitigated by
findings from

1 high confidence
study in rats exposed
for 28 days

This document is a draft for review purposes only and does not constitute Agency policy.

3-187 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration
summary judgment

counts and
testosterone levels

supplemental
mechanistic studies.

• High confidence study

Reoroductive
hormones

1 high confidence
study in rats for 28
days

• Decreases in serum
testosterone levels at
>0.625 mg/kg-d

• Dose-response
gradient

• High confidence study

• No factors noted

Organ weight

1 high confidence
study in rats for 28
days

• Decreases in testis
and epididymis
weights at
>0.625 mg/kg-d

• Consistent effects
across tissues

• Coherence with sperm
counts histopathology
and testosterone
levels

• Dose-response
gradient

• High confidence study

• No factors noted

Mechanistic evidence and supplemental information (see subsection above)

Biological events or
pathways (or other
information)

Summary of key findings, interpretation, and limitations

Evidence stream judgment

Levdig cell androgen

Key findings and interpretation:

Evidence of altered Leydig cell function

function

• Impaired Leydig cell mitochondrial cholesterol uptake and testosterone
synthesis in two vitro rodent models.

• Altered testosterone secretion in rat testes and altered androgen levels,
reproductive organ weights and histopathology in rodent species after acute,
i.p. injection consistent with evidence of reduced steroidogenesis.

Limitations: few studies; in animal models only; acute, i.p. exposure at high doses
associated with systemic toxicity

and decreased androgen production
provide support for the biological
plausibility of the male reproductive
effects of PFDA.

This document is a draft for review purposes only and does not constitute Agency policy.

3-188 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration
summary judgment

Reoroductive hormone

Key findings and interpretation:

• Effects in a minority of in vitro studies/assays relating to the AR (receptor
binding, transactivation and cell proliferation) and ER pathways
(transactivation), and in one study on aromatase.

• ToxCast model predictions suggests that PFDA is inactive for AR/ER agonist and
antagonist activities.

Limitations: Mixed results across studies; some effects at cytotoxic levels; models

generally not in male reproductive tissues.

signaling

Other mechanisms

Key findings and interpretation:

• Generally, lack of support for potential role of hepatic androgen metabolism or
indirect systemic toxicity in PFDA-induced male reproductive effects in rodent
studies

Limitations: acute i.p. exposure; high dose; few studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-189 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.5. FEMALE REPRODUCTIVE EFFECTS

1 Human studies

2 Studies of possible female reproductive effects of PFDA are available for reproductive

3 hormones, fecundity (i.e., time to pregnancy), menstrual cycle characteristics, and endometriosis.

4 In addition, studies were available for spontaneous abortion and preterm birth which could be

5 driven by either female reproductive or developmental toxicity. These outcomes are reviewed in

6 Section 3.2.3 on Developmental effects in this assessment but are included in the consideration of

7 coherence across outcomes for female reproductive effects. The study evaluations for these

8 outcomes are summarized in Figure 3-66.

This document is a draft for review purposes only and does not constitute Agency policy.

3-190 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

w ^>ce

C^6 ^°°

o^6

Bach, 2015, 3981559-
Bach, 2018, 5080557
Barrett, 2015, 2850382
Carwile, 2021,9959594
Ernst. 2019, 5080529-
Jensen et a!., 2020, 6311643 -
Kim, 2020, 6833596 -
Liu, 2020, 6569227
Louis, 2012, 1597490
Lum, 2017, 3858516
Mccoy, 2017, 3858475
Singer, 2018, 5079732-
Timmermann. 2022, 10176553-
Vestergaard, 2012, 1332472
Wang, 2017, 3856459
Wang, 2021, 10176703
Wise, 2022, 9959470
Xie, 2021, 8437891 -
Yang et al„ 2022, 10176804

Zhou, 2016, 3856472

II 1 1 1 1 1 1

++
4-+
++

+¦#¦

++
++
++

¦f

++ ++

+ *

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 3-66, Study evaluations for epidemiology studies of PFDA and female
reproductive effects. Refer to HAWC for details on the study evaluation review:

HAWC Human Female Reproductive Effects.

This document is a draft for review purposes only and does not constitute Agency policy,

3-191 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reproductive hormones

Reproductive hormones examined in the evaluated studies include testosterone,
estradiol/estrogen, insulin like growth factor 1 (IGF-1), follicle stimulating hormone (FSH),
luteinizing hormone (LH), progesterone, prolactin, and inhibin B, as well as sex hormone binding
globulin (SHBG). Key issues for the evaluation of these studies were sample collection and
processing. For testosterone, LH, FSH, and prolactin, due to diurnal variation, blood sample
collection should be in the morning, and if not, time of collection must be accounted for in the
analysis. If there is no consideration of time of collection for these hormones, the study is classified
as deficient for outcome ascertainment and low confidence overall. The timing of PFDA exposure
relevant for influencing reproductive hormones is unclear and dependent on several factors, and
thus all exposure windows with available data were considered relevant for these endpoints of
interest, particularly given the long half-life of PFDA. This includes cross-sectional studies since
levels of these hormones are capable of being rapidly upregulated or downregulated and they are
not expected to directly bind to or otherwise interact with circulating PFAS.

Ten studies (Timmermann et al.. 2022: Yang etal.. 2022b: Xie etal.. 2021: Tensenetal..
2020b: Liu etal.. 2020b: Yao etal.. 2019: Zhang etal.. 2018a: McCoy etal.. 2017: Zhou etal.. 2016:
Barrett etal.. 20151 reported on associations between PFDA exposure and female reproductive
hormones. Four studies were medium confidence, including cross-sectional studies of healthy
adults in Norway (Barrett etal.. 20151 and the U.S. (Xie etal.. 20211 (latter is low confidence for
testosterone), a cross-sectional study of newborns in China (Liu etal.. 2 02 Obi, and a pregnancy
cohort in China (Yang etal.. 2022b). Most of the remaining six studies were low confidence. In
adults, this included an analysis of women with premature ovarian insufficiency in China (Zhang et
al.. 2018al and a cohort of pregnant women in Denmark fTimmermann etal.. 20221. In children and
adolescents, there was a cohort of adolescents in Taiwan fZhou etal.. 20161 and two studies in
infants, a cohort in Denmark flensen et al.. 2020bl and a cross-sectional study in China fYao etal..
20191. Lastly. McCoy etal. (20171 was considered uninformative due to multiple deficiencies in
study evaluation.

For estrogen, one study, a cohort in pregnant women with follow-up across pregnancy
(Yang etal.. 2022b) examined estrone (Ei), estradiol (E2), and estriol (E3) and reported an inverse
association between PFDA (median 0.8 ng/mL) and estrone (P [95% CI]: -0.12 (-0.24, -0.01)).
Associations with estradiol and estriol were in the same direction but not statistically significant.
The remaining studies examined only estradiol. In general population adults, an inverse, though
non-monotonic, association (P [95% CI] vs Q1 for Q2: -78.64 [-310.37, 153.09]; Q3: -183.04 [-
353.51,-12.56]; Q4: =117.92 [-285.64, 49.70]) was also reported in (Xie etal.. 20211 (median 0.1
ng/mL). Associations varied by age group, with inverse associations in adolescents and 12-49 year
olds, but a positive association in women 50 years of age and older. No association with PFDA was
reported with follicular estradiol in Barrett etal. f20151 (mean PFDA 0.3 ng/mL), or with blood

This document is a draft for review purposes only and does not constitute Agency policy.

3-192 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

estradiol in Zhang etal. (2018a) (median PFDA 0.4 ng/mL), Zhou etal. (20161 (median PFDA
1.0 ng/mL), or cord blood estradiol in Yao etal. (20191 (median PFDA 0.2 ng/mL).

For testosterone, since Barrett etal. f20151 did not examine associations with testosterone,
all of the available evidence is low confidence. None of the four available studies reported a
statistically significant association between PFDA and testosterone fXie etal.. 2021: Yao etal.. 2019:
Zhang etal.. 2018a: Zhou etal.. 20161. and the direction of association was not consistent across
studies (positive association in Yao etal. (20191 and Xie etal. (20211. inverse association in the
other two studies.

For other reproductive hormones, Barrett etal. (20151 also examined luteal phase
progesterone, finding a positive association with PFDA (0.472 (-0.043, 0.987)). Liu etal. f2020bl
examined progesterone in newborns and found no association with PFDA. Zhang etal. f2018al
examined FSH, LH, and prolactin and also found no association with PFDA. Tensen et al. f2020bl
reported inverse associations between PFDA and DHEA (p < 0.05), DHEAS, Androstenedione, and
17-OHP (p>0.05). Lastly, Timmermann etal. (20221 found a positive, though imprecise association
with prolactin during pregnancy (3.3% difference (95% CI -0.4, 7.2) per doubling of PFDA
concentrations).

Overall, the findings in reproductive hormones are primarily null, with a few inconsistent
associations observed. However, due to low exposure levels in most studies and the availability of a
small number of studies per population type (adult women, adolescents, newborns) and
reproductive hormones, the evidence is difficult to interpret.

Fecundity

There are six epidemiology studies that report on the association between PFDA exposure
and fecundity. Fecundity is the biological capacity to reproduce. Time to pregnancy, defined as the
number of calendar months or menstrual cycles from the time of cessation of contraception to
detection of pregnancy, is a primary outcome measure used to study fecundity. There are
challenges in studying this outcome as it is ideal to enroll women at the point when contraception is
discontinued, but this is generally limited to women trying to get pregnant who may not be
representative of the general population. An alternative approach is to enroll pregnant women and
ask for their recall of time to pregnancy, but this is subject to selection bias due to excluding women
who are unable to conceive, and thus are potentially most affected. Two studies were
preconception cohorts and considered medium confidence fLum etal.. 2017: Vestergaard etal..
20121. and two were pregnancy cohorts and considered low confidence fBach etal.. 2018: Bach et
al.. 20151 due to the potential for selection bias described above. Another fecundity-specific
consideration is the potential for confounding in parous women due to factors related to previous
pregnancies (Bach etal.. 20181. In addition to the studies of time to pregnancy, two studies
examined women undergoing infertility treatment; one medium confidence cohort examined
successful pregnancies using IVF fWang etal.. 20211 and one low confidence cross-sectional study
compared PFAS concentrations in women with different types of infertility (with male factor

This document is a draft for review purposes only and does not constitute Agency policy.

3-193 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

infertility as the control group) and associations with fertilization rate (Kim etal.. 2020c). A
summary of the study evaluations is presented in Figure 3-46, and additional details can be
obtained from HAWC.

The results for the association between PFDA exposure and time to pregnancy are
presented in Table 3-27. A fecundability ratio less than 1 indicates a decrease in fecundity/increase
in time to pregnancy. One study fBach etal.. 20181 reported longer time to pregnancy with higher
exposure in the fourth quartile, but only in parous women, which despite adjustment for
interpregnancy interval, may be more likely to be confounded. None of the other available studies
reported a decrease in fecundity/increase in time to pregnancy with higher exposure, though this
observed lack of association could be due to poor study sensitivity resulting from low exposure
levels. In addition to the time to pregnancy results, two studies fBach etal.. 2015: Vestergaardet
al.. 20121 also analyzed infertility as an outcome and found no increase with higher exposure.
Similarly, Wang etal. (2021) reported no increase in negative hcG test or clinical pregnancy failure
following IVF with higher PFDA exposure (associations indicated less pregnancy failure and test
negativity with higher exposure). Kim etal. (2020c) found no association between different
infertility factors (endometriosis, PCOS, genital tract infections, or idiopathic) compared to male
factor infertility. However, Kim etal. f2020cl did report an inverse, though imprecise, association
between PFDA exposure and fertilization rate ((3=-60.83, 95% CI: -129.25, 7.59).

Table 3-27. Associations between PFDA and time to pregnancy in
epidemiology studies

Reference,

study
confidence

Population

Median
exposure (IQR)
or as specified

Comparison
for effect
estimate

Fecundability ratio (FR)
(95% CI)

Vestergaard
etal. (2012),

Preconception cohort (1992-1995),
Denmark; 222 nulliparous women

0.1(0.1, 0.1)a

log-unit
increase

1.15 (0.89, 1.49)

medium

Above median
vs. below

1.40 (0.96, 2.03)

Bach et al.
(2018), low

Danish National Birth Cohort sub-
sample (1996-2002), Denmark, 638
nulliparous women and 613 parous
women

0.2 (0.1-0.2)

Quartiles vs.
Q1

Nulliparous
Q2: 1.13 (0.89, 1.43)
Q3: 1.02 (0.82, 1.28)
Q4: 1.11 (0.89, 1.39)

Parousb
Q2: 0.92 (0.68, 1.26)
Q3: 0.95 (0.71, 1.28)
Q4: 0.86 (0.65, 1.15)

Bach et al.
(2015), low

Aarhus pregnancy cohort (2008-13),
Denmark; 1,372 nulliparous women

0.3 (0.2-0.4)

0.1 ng/mL
increase

1.00 (0.97, 1.03)

Quartiles vs.
Q1

Q2: 1.08 (0.91, 1.28)
Q3: 0.98 (0.83, 1.16)
Q4: 1.08 (0.91, 1.28)

This document is a draft for review purposes only and does not constitute Agency policy.

3-194 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference,

Median

Comparison

study

exposure (IQR)

for effect

Fecundability ratio (FR)

confidence

Population

or as specified

estimate

(95% CI)

Lum et al.

LIFE preconception cohort (2005-09),
U.S.; 401 women

0.4 (0.2-0.6)

Tertiles vs. T1

T2: 0.7 (0.5, 1.1)
T3: 0.9 (0.6, 1.3)

(2017),

medium

*p < 0.05.

Participants with pregnancy.

bThese results were based on a model that corrected PFAS exposure based on an interpregnancy interval of
median length. An alternate model where interpregnancy interval was included as a covariate was statistically
significant in Q4. A model with no adjustment for interpregnancy interval was not significant but had a
monotonic decrease across quartiles (FRs of 0.92, 0.87, 0.78).

Pubertal development

Pubertal development is primarily assessed using established criteria, such as Tanner stage
ratings. In girls, Tanner staging involves evaluation of the development of breasts and pubic hair.
Stage 1 represents prepubertal development; Stage 2, the onset of pubertal development, and Stage
5 represents full sexual maturity. Age at menarche and age at peak height velocity (i.e., the age at
which a child experiences the largest increase in height) can also be used as measures of pubertal
development Three studies, including two medium confidence cohorts in Denmark (Ernst etal..
20191 and the United States (Carwile etal.. 20211 and one low confidence cross-sectional study
fWise etal.. 20221. examined timing of pubertal development with PFDA exposure.

Carwile etal. f20211 used exposure measured during mid-childhood (median 8 years) with
follow-up to early adolescence (median 13 years). Using a pubertal development score based on
parental responses to scales of multiple pubertal markers (breast development, body hair growth,
acne, growth spurt, and menarche), they reported less pubertal development in early adolescence
with higher exposure (P (95%) per doubling of exposure: -0.11 (-0.18, -0.03)). This was consistent
with their findings for older age at peak height velocity (0.23 (0.11, 0.35)) and older age at menarche
(HR (95% CI) per doubling of exposure: 0.91 (0.77,1.0611. Ernst etal. f20191 used maternal
exposure measured in blood and prospectively identified pubertal onset with follow-up checks
every six months. In girls, age at Tanner stages 2 and 3 for breast development were lower with
higher exposure, consistent with Carwile etal. (20211. though not statistically significant. No
association was observed for Tanner stages 4 and 5. No clear patterns for associations were
observed with pubic hair development, axillary hair, or age at menarche. Results for the second and
third tertiles were discordant for some outcomes (lower age at menarche and axillary hair
development in second tertile, higher in third). Looking at a combined puberty indicator outcome,
there was lower age at puberty (not significant) in the second tertile and no difference in the third
tertile compared to the first. Wise etal. (20221 did not report a clear association with age at
menarche (age was higher in both the first and third tertiles compared to the second), but this
study was low confidence due to concerns for lack of temporality between exposure and outcome
misclassification due to recall of age at menarche among adult women. Sensitivity was a concern for

This document is a draft for review purposes only and does not constitute Agency policy.

3-195 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

all three studies, as exposure contrast was narrow. Exposure levels and contrast were slightly
higher in Carwile etal. (20211 than the other studies (IQR 0.4 ng/mL vs 10th-90th percentile
difference of 0.2 ng/mL in Ernst etal. f20191. so it is possible that this is a basis for the clearer
associations in the former study.

Menstrual cycle characteristics

Four epidemiology studies report on the association between PFDA exposure and
menstrual cycle characteristics. Two were cohorts, one a preconception cohort already described
for fecundity (Lum etal.. 20171 and one a pregnancy cohort (Singer etal.. 20181. Two studies were
cross-sectional, one of participants in a preconception cohort fZhou etal.. 2017al and one of
general population black women of reproductive age fWise etal.. 20221. For any outcome related
to menstruation, there is potential for reverse causation because menstruation is one of the
mechanisms by which PFAS are removed from the body fWong etal.. 2014: Zhang etal.. 2013bl.
This potential bias could be away from the null with irregular and longer cycles. Thus, all four
studies were considered low confidence. There were no associations reported between menstrual
cycle length or irregularity and PFDA exposure, but due to limited sensitivity related to exposure
contrasts and low confidence in the studies, these findings are difficult to interpret.

Endometriosis

Two epidemiology studies report on the association between PFDA exposure and
endometriosis. Both studies were cross-sectional, which decreases confidence for this chronic
outcome due to the inability to establish temporality and the likely lack of measurement in the
relevant etiologic window. There is potential for reverse causality as described above since
endometriosis can influence the menstrual cycle and it is possible that this would act in a protective
direction since endometriosis can be associated with heavier and more frequent bleeding which
could increase elimination of PFDA from the body. Parity and related factors such as time since last
child have also been suggested as a source of reverse causality for this association as a longer inter-
pregnancy interval could allow more accumulation of PFAS levels (Wang etal.. 20171. but this was
not a major concern in this set of studies as one study adjusted for parity and the other performed a
sensitivity analysis with only women without a history of pregnancy. Still, because of the concern
related to menstrual cycle irregularity association with endometriosis, all the studies were
classified as low confidence, though one is considered higher quality within that classification; this
study included two groups of women, one group scheduled for surgery (laparoscopy or
laparotomy), and one group identified through a population database who underwent pelvic MRI to
identify endometriosis (Louis etal.. 20121. The remaining study was additionally deficient for
outcome ascertainment, specifically a case definition including only endometriosis-related
infertility among surgically confirmed cases (Wang etal.. 20171. which is likely to include less
severe or asymptomatic cases among the controls. The low confidence study with good outcome
ascertainment fLouis etal.. 20121 reported higher odds of endometriosis with higher exposure in

This document is a draft for review purposes only and does not constitute Agency policy.

3-196 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

the operative sample (OR = 2.95, 95% CI: 0.72,12.1), but lower odds in the population sample
(OR = 0.06, 95% CI: 0.00,12.3), though both estimates were imprecise. The low confidence study
by Wang etal. f20171 reported lower odds of endometriosis-related infertility with higher exposure
(OR vs. Tl: T2: 0.93 (95% CI: 0.51,1.70), T3: 0.74 (95% CI: 0.40,1.35). It is difficult to reconcile the
differing results considering the low number of studies, all of which were low confidence, and the
potential for reverse causality for this outcome.

Premature Ovarian Insufficiency

One low confidence study, a case-control study in China, examined the association between
PFDA exposure and premature ovarian insufficiency (POI) fZhang et al.. 2018bl In this study, POI
was defined as an elevated FSH level greater than 25 IU/L on two occasions more than four weeks
apart and oligo/amenorrhea for at least four months. Because this definition is closely tied to
menstruation, there are concerns for reverse causality as with the previous two outcomes, which
would be expected to be biased away from the null as there is reduced bleeding/elimination of
PFDA from the body. The study reported higher odds of POI (not statistically significant) with
higher PFDA exposure (OR (95% CI) for tertile 2 vs. 1: 1.03 (0.54,1.96), tertile 3 vs. 1: 1.36
(0.71,2.60), but given the lack of additional evidence and concerns for reverse causality, there is
considerable uncertainty in these results.

Animal studies

A single study in the database of toxicity studies for PFDA evaluated female reproductive
effects (NTP. 2018). The study examined the following endpoints after a 28-day gavage exposure
(0, 0.156, 0.312, 0.625,1.25, and 2.5 mg/kg-day) in adultfemale rats: organ weights,
histopathology, hormone levels, and estrous cycles. The NTP f20181 study was evaluated as high
confidence for all endpoints examined (see Figure 3-67). Although there is only a 28-day study
available, the duration of the study is sufficient for assessing female reproductive toxicity given that
significant effects on estrous cyclicity were observed as early as Day 21 of the 28-day study and the
mean estrous cyclicity length is reported to be 4.4 days amongst multiple sub-strains of Sprague
Dawley rats (Marty etal.. 2009).

This document is a draft for review purposes only and does not constitute Agency policy.

3-197 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

7.0^'

,0^

Reporting quality -
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity -
Results presentation -
Overall confidence -

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 3-67. Female reproductive animal study evaluation heatmap. Refer to

HAWC for details on the study evaluation review.

Estrous cycle

Female rats from the three highest dose groups (0,625,1.25, and 2.5 mg/kg-day) were
evaluated for changes in the estrous cycle due to PFDA exposure, as compared to controls. To
examine this endpoint, vaginal smears were performed for sixteen consecutive days before animals
were necropsied. Changes in the percent of time spent in each estrous stage (proestrus, estrus,
metestrus, diestrus) were affected by exposure (see Figure 3-68 and Table 3-28). Specifically, for
proestrus, the percentage of time spent was increased by 103 and 123% at 0.625 and
1.25 mg/kg-day, respectively but then decreased by 81% at 2.5 mg/kg-day. For metestrus, the
percentage of time spent was increased by 23% at 0.625 mg/kg-day but then decreased by 100% at
>1.25 mg/kg-day. A significant trend test was observed for the percentage oftime spentin estrus
with statistically significant decreases (42-84%} at >1.25 mg/kg-day (Figure 3-68 and Table 3-28).
Correspondingly, a significant trend test was observed for the percentage of time spent in diestrus
with statistically significant increases (27-63%) at >1.25 mg/kg-day (see Figure 3-68 and Table 3-
28), Estrous cyclicity was disrupted and all female rats remained in a continuous state of diestrus
at 2.5 mg/kg-day starting on Day 21 (Day 9 of the sixteen days in which vaginal cytology was
assessed). The sustained state of diestrus suggests that these animals may have been infertile (U.S.
EPA. 1996a). although this was not specifically evaluated. Although decreased body weight in

This document is a draft for review purposes only and does not constitute Agency policy,

3-198 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

female rats was observed at the same doses (body weight decreases were 12-36% at
>1.25 mg/kg-day; refer to Section 3.2.10 on General toxicity effects for more details) as effects on
estrous cyclicity, it is unclear if these effects are related and the effect on female reproductive
function is disproportionately more severe and concerning than the moderate changes in body
weight. Although body weight has been shown to fluctuate during the different estrous stages and
weight loss has been shown to correlate with disrupted estrous cyclicity in rats fTropp and Markus.
20011. it is not possible to determine if the decreases in body weight in female rats might be
responsible for the effects on estrous cyclicity observed in the NTP (20181 study. Furthermore,
even though no changes were observed on other stages of the estrous cycle (i.e., proestrus and
metestrus), the effects of PFDA on estrus and diestrus are still considered biologically relevant
given the potential influence that the lack of cyclicity may have on fertility, regardless of whether
the observed decrease in body weight may have partially contributed to these changes. Changes in
cycle length and the number of cycles during the study were not affected in the 0.625 and
1.25 mg/kg-day groups. Data for cycle length and number of cycles could not be determined for the
2.5 mg/kg-day group because estrous cyclicity was disrupted in all female rats at this dose and all
animals remained in a state of continuous diestrus starting at Day 21 until sacrifice.

Table 3-28. Percent changes relative to controls in time spent in each estrous

stage (proestrus, estrus, metestrus, diestrus) in female S-D rats exposed to

PFDA exposure for 28 days (NTP. 2018)

Endpoint

Dose (mg/kg-d)

0.625

1.25

2.5

% of Estrous cycle in diestrus

% of Estrous cycle in estrus

-22

-42

-84

% of Estrous cycle in metestrus

-100

% of Estrous cycle in proestrus

103

123

-81

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors.

Hormone levels

Testosterone was measured in all dose groups at study termination; it is unclear from the
study description if the study authors controlled for fasting or time of necropsy. A significant trend
test was observed with statistically significant increases reported at >0.312 mg/kg-day (see
Figure 3-68). Increases were monotonic and varied from 30% to 348% change from controls;
levels of circulating testosterone were increased more than two-fold at 1.25 mg/kg-day. Other sex
hormones (e.g., estradiol) were not measured in this study. The biological relevance of increased
testosterone to the development of PFDA-induced female reproductive toxicity is unclear.
Specifically, the association of increased testosterone and altered estrous cycling (e.g., prolonged
diestrus) requires further investigation.

This document is a draft for review purposes only and does not constitute Agency policy.

3-199 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Histopathology

Histological examination of the clitoral gland, ovaries, uterus, and mammary glands were
performed at study termination. Histopathology was examined for the ovaries at all doses; all other
reproductive tissues were examined only in the control and high-dose groups. Histological changes
due to PFDA treatment were not reported for any tissue examined including the uterus (see
Figure 3-68) even though PFDA-effects on estrous cyclicity and uterine weight were reported.

Organ weights

Uterine weights were measured in all dose groups at study termination. A significant trend
test was observed for both absolute and relative weights with the two highest dose groups reaching
statistically significant decreases for both measures (see Figure 3-68). Decreases reached -64%
and -44% change from controls for absolute and relative weights, respectively. Other organs
related to the female reproductive system were not measured. It should be noted that comparisons
of uterine weights were not made in rats that were in the same estrous stage. As noted below, many
studies in rats have shown that uterus weight decreases during diestrus. Therefore, it is unclear if
the reductions in uterus weight are a direct effect of PFDA or rather a secondary effect due to
prolonged diestrus owing to PFDA exposure.

This document is a draft for review purposes only and does not constitute Agency policy.

3-200 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Outcome Confidence Experiment Name Effect

Endpoint Name

Animal Description

Trend Test Result

NTP, 2018, 4309127 High confidence 28 Day Oral Estrous Cycle

% of Estrous Cycle in Diestrus

Rat, Sprague-Dawley (Harlan) (i!

significant

# No significant change

% of Estrous Cycle in Estrus

Rat. Sprague-Dawley (Harlan) (-

significant

A Significant increase

% of Estrous Cycle in Metestrus

Rat, Sprague-Dawley (Harlan) (

not significant

~ Significant decrease

% of Estrous Cycle in Proestrus

Rat, Sprague-Dawley (Harlan) (^

not significant

Hormone

Testosterone (T)

Rat, Sprague-Dawley (Harlan) (:

significant

Histopathology

Clitoral Gland Histopathology

Rat. Sprague-Dawley (Harlan) (-

not applicable

Mammary Gland Histopathology

Rat, Sprague-Dawley (Harlan) (

not applicable

Ovary Histopathology

Rat, Sprague-Dawley (Harlan) (_

not applicable

Uterus Histopathology

Rat. Sprague-Dawley (Harlan) (2

not applicable

Organ Weight

Uterus Weight. Absolute

Rat, Sprague-Dawley (Harlan) ("

significant

Uterus Weight, Relative

Rat, Sprague-Dawley (Harlan) (

significant

Estrous Cycle

Number of days in Diestrus

Rat, Sprague-Dawley (Harlan) (

significant

Number of days in Estrus

Rat, Sprague-Dawley (Harlan) (_

significant

PFDA Female Reproductive Effects

—» A A
-•-W—W

-AAA

W
W

•-V-T

mg/kg-day

Figure 3-68. PFDA female reproductive effects (results can be viewed by clicking the HAWC link:

https://hawcprd.epa.gOv/summary/data-pivot/assessment/100500072/pfda-female-reproductive-animal/I

This document is a draft for review purposes only and does not constitute Agency policy,

201 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Mechanistic studies and supplemental information

As discussed in the male reproductive section (see Section 3.2.4), PFDA-induced effects on
AR and ER functions and aromatase activity have been evaluated in in vitro cell culture studies and
high throughput screening (HTS) assays from ToxCastandTox21. Findings from in vitro cell
culture studies and HTS assays do not provide consistent evidence for potential effects of PFDA on
AR or ER functions, or aromatase activity. Additional in vivo and/or cell culture studies are
necessary to address inconsistencies in the available in vitro data and determine whether these
pathways might be disrupted by PFDA exposure. In an in vitro study, PFDA inhibited progesterone
production in mouse Leydig tumor cells, which the study authors postulated was due to oxidative
stress fZhao etal.. 20171. It is not possible to corroborate this effect with data from the lone
reproductive study in rats fNTP. 2018] given that progesterone was not measured in the fNTP.
20181 study. In the NTP T20181 study, Wyeth-14,643 (a PPARa agonist) was shown to cause effects
on estrous cyclicity similar to those reported for PFDA. However, mechanistic studies that
investigate the role of PPARa in PFDA-altered estrous cyclicity are not available.

Evidence Integration

There is indeterminate evidence of an association between PFDA exposure and female
reproductive effects in human studies, though the low confidence studies that were available had
concerns for study sensitivity which reduces the ability to interpret the observed null findings. A
significant inverse association between PFDA and anogenital distance in girls was observed in one
study (see Developmental Effects), which is relevant to female reproductive toxicity. The biological
relevance of this effect on anogenital distance is unclear given that an increase in this measure is
considered adverse in girls rather than a decrease per the U.S. EPA's Guidelines for Reproductive
Toxicity Risk Assessment. Furthermore, the available reproductive hormone evidence for PFDA
does not support an association. Previous studies have shown an association between increased
testosterone and increased anogenital distance in women (Mira-Escolano etal.. 20141. however the
human evidence is inadequate for examining PFDA-induced effects on testosterone in women.
Whereas increased testosterone was observed in female rats in the NTP (20181 study, the study
authors did not measure anogenital distance given that there was no developmental exposure in
the study. The increased testosterone observed in female rats is considered relevant to humans
and given the known association between increased testosterone and anogenital distance in
women, an increase in anogenital distance rather than a decrease would be expected in women
exposed to PFDA. Overall, there is little biological understanding of how hormonal perturbation or
other biological processes might result in a decrease in anogenital distance owing to PFDA
exposure.

In addition to the outcomes described in this Section, there is potential for two of the
outcomes described in the developmental section (refer to Section 3.2.3 for more details), preterm
birth and spontaneous abortion, to be related to female reproductive toxicity. The evidence for

This document is a draft for review purposes only and does not constitute Agency policy.

3-202 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

these outcomes was inconsistent. Given that most of the evidence for female reproductive effects
was null or inconsistent, there is little clear indication of an association. However, the exposure
levels in most of the study populations were low, which resulted in low sensitivity to detect an
effect, and thus these findings should not be interpreted as supporting a lack of effect.

The available data from a 28-day gavage study in rats provide moderate evidence that PFDA
exposure may cause female reproductive toxicity (see Table 3-29). The evidence is sparse. The
data are from a single animal study that did not evaluate fertility, pregnancy outcomes, multiple
hormone levels (only testosterone was measured), or markers of reproductive development. PFDA
was observed to cause effects on the following female reproductive parameters: organ weight
(i.e., decreased uterine weights at >1.25 mg/kg-day), hormone levels (i.e., increased testosterone
levels at >0.312 mg/kg-day), and estrous cycle (i.e., percentage of time spent in estrus and diestrus
at >1.25 mg/kg-day). One factor increasing the strength of the evidence is the severity of the effect
on estrous cyclicity; specifically, that PFDA induced a continuous state of diestrus in 100% of rats
treated at the highest dose tested (2.5 mg/kg-day), which could be indicative of reductions or
delays in fertility. However, some caution in the interpretation of the higher dose effects is
warranted given the significant decreases in body weight, particularly at 2.5 mg/kg-day (36%
decrease). Support for the adversity and concerning nature of prolonged diestrus and its
association with infertility is provided by the following text in the U.S. EPA's Guidelines for
Reproductive Toxicity Risk Assessment:

• "Persistent diestrus indicates temporary or permanent cessation of follicular development
and ovulation, and thus at least temporary infertility,"

• "Pseudopregnancy is another altered endocrine state reflected by persistent diestrus."

• "Significant evidence that the estrous cycle (or menstrual cycle in primates) has been
disrupted should be considered an adverse effect."

• "The greatest confidence for identification of a reproductive hazard should be placed on
significant adverse effects on sexual behavior, fertility or development, or other endpoints
that are directly related to reproductive function such as menstrual (estrous) cycle
normality, sperm evaluations, reproductive histopathology, reproductive organ weights,
and reproductive endocrinology."

Furthermore, prolonged diestrus is commonly reported in rodent models of impaired
fertility (Li etal.. 2017: Caldwell etal.. 2014: Miller and Takahashi. 2014: Mayer and Boehm. 2011)
and continuous diestrus is observed during reproductive senescence in aged female rats (Lefevre
and Mcclintock. 1988). There was also coherence between decreased uterus weight and increased
percentage of time spent in diestrus at >1.25 mg/kg-day. Previous studies have shown that
decreased uterus weight in rats is commonly observed during diestrus fWestwood. 2008: Vasilenko
etal.. 1981: Walaas. 1952: Boettiger. 19461. In addition to prolonged diestrus, PFDA decreased the

This document is a draft for review purposes only and does not constitute Agency policy.

3-203 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

percentage of time spent in estrus fNTP. 2018], which could indirectly cause infertility given that
rodents are sexually receptive only during estrus (Goldman et al.. 20071. The severe, PFDA-induced
decreased time spent in estrus is expected to result in decreased opportunities for mating in the
rats, and therefore reductions or delays in fertility. Unfortunately, no multi-gene rational studies of
PFDA were available to inform this hypothesis.

In this study, PFDA did not cause histopathological changes in female reproductive tissues.
Given the short-term duration of the lone animal study, it cannot be reasonably ruled out that
detectable histopathological effects could have become apparent with a longer observation
window. The short-term duration of the lone animal study does not reduce confidence in the
database for PFDA-induced female reproductive effects given that biologically relevant effects (e.g.,
prolonged diestrus) were still observed.

Taken together, the available evidence indicates that PFDA is likely to cause female
reproductive toxicity in humans under sufficient exposure conditions11 (see Table 3-29). This
conclusion is based primarily on evidence from a high confidence study in rats exposed to doses
ranging from 1.25-2.5 mg/kg-day PFDA for 28 days. The PFDA-induced disruption of estrous
cyclicity observed in female rats from the NTP study (NTP. 20181 and its implications for infertility
can be considered relevant to humans given that the mechanisms responsible for regulating female
reproductivity (e.g., estrous cyclicity in rats and menstrual cycling in humans) are similar between
rats and humans fGoldman etal.. 2007: Bretveld et al.. 20061. Given the sparse evidence base
(i.e., one short-term animal study and largely low confidence or null human studies) and the lack of
understanding for how PFDA exposure causes the observed reproductive effects or whether they
might progress with longer exposures, further studies that could inform this conclusion include
those that examine the effect of PFDA on female fertility and pregnancy outcomes in exposed
animals from subchronic, chronic, developmental, or multigenerational studies, as well as in vivo or
cell culture mechanistic studies.

11 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes only and does not constitute Agency policy.

3-204 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-29. Evidence profile table for PFDA exposure and female reproductive effects

Evidence stream summary and interpretation

Evidence
integration
summary judgment

Evidence from studies of exposed humans (see Section 3.2.5: Human studies)

0®Q

Evidence indicates
(likely)

Primary basis:
Evidence from a high
confidence study in
rats showing
biologically coherent
effects on uterus
weight and the
estrous cycle after
oral exposure to
PFDA at

>1.25 mg/kg-d for
28 days.

Human relevance:
Evidence in animals is
presumed relevant to
humans given that
mechanisms
regulating female
reproduction are
similar between rats
and humans.

Cross-stream
coherence:
N/A, human evidence
is indeterminate.

Studies,
outcomes, and
confidence

Summary and key findings

Factors that
increase
certainty

Factors that
decrease certainty

Evidence stream
summary

Reoroductive
hormones

4 medium and 5 low
confidence studies

• Inverse association between PFDA exposure and
estrogen observed in 2 studies. Most studies
reported no association with female
reproductive hormones, but sensitivity was
limited in most studies

• No factors
noted

• No factors noted

QQQ

Indeterminate

Within and across
outcomes, findings
were mixed, null,
and/or of low
confidence.
Interpretation of
the lack of an
association for
most outcomes in
these studies is
complicated by
poor sensitivity for
observing effects
due to low
exposure levels.

Fecundity

3 medium and 3 low
confidence studies

• One study reported longer time to pregnancy
with higher PFDA exposure, but only in parous
women. No association observed in other
studies, but sensitivity was limited.

• No factors
noted

• Unexplained
inconsistency,
although a lack
of association in
some studies
may be

attributable to

limited

sensitivity

Pubertal
develooment

2 medium and 1 low
confidence cohort
studies

• One study reported later age at pubertal onset
based on pubertal development score, age at
peak height velocity, and age at menarche. Two
other studies reported no clear association

• Coherence of
related
effects in one
study

• Unexplained
inconsistency

Menstrual cvcle

4 low confidence
studies

• No association observed between PFDA exposure
and menstrual cycle characteristics, but
sensitivity was limited.

• No factors
noted

• Potential for
reverse causality

This document is a draft for review purposes only and does not constitute Agency policy.

3-205 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence
integration
summary judgment

Endometriosis

2 low confidence
studies

• Higher odds of endometriosis with higher PFDA
exposure in women scheduled for laparoscopy or
laparotomy in one study, but lower odds of
endometriosis in a population-based sample in
the same study and a low confidence study.

• No factors
noted

• Unexplained
inconsistency
across low
confidence
studies

• Potential for
reverse causality

Susceptible
populations and
lifestages:

Based on altered
estrous cyclicity data
in animals, females of
reproductive age may
be at higher risk.

Other inferences:
No specific factors are
noted.

Evidence from in vivo animal studies (see Section 3.2.5: Animal studies)

Studies,
outcomes, and
confidence

Summary and key findings

Factors that
increase
certainty

Factors that
decrease certainty

Evidence stream
summary

Estrous cvcle

1 high confidence
study

• The percentage of time spent in estrus was
significantly decreased at >1.25 mg/kg-d.

• The percentage of time spent in diestrus was
significantly increased at >1.25 mg/kg-d.

• Estrous cyclicity was disrupted at 2.5 mg/kg-d
and all female rats in this dose group remained in
a continuous state of diestrus by Day 21.

• Large
magnitude of
effect and
concerning
severity

• In a high
confidence
study

• Dose-
response
gradient for
effects on the
percentage of
time spent in
estrus and
diestrus.

• Coherence
with reduced
uterus
weight.

• Lack of expected
coherence for
histopathology,
although
possibly
explained by
short exposure
duration

• Potential
confounding by
body weight
decreases.

0®Q

Moderate

Based on multiple,
coherent changes
in female
reproductive
endpoints, most
notably that PFDA
induced a
continuous phase
of diestrus, which
could be indicative
of infertility, in
100% of rats at
2.5 mg/kg-d.

This document is a draft for review purposes only and does not constitute Agency policy.

3-206 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence
integration
summary judgment

Organ weight

1 high confidence
study

• Decreased absolute and relative uterine weights
at >1.25 mg/kg-d.

• Dose-
response
gradient in a
high

confidence
study

• Potential

confounding by
body weight
decreases
(mitigated some
by comparable
effects on
absolute and
relative weights)

Hormone levels

1 high confidence
study

• Increased testosterone levels at >0.312 mg/kg-d.

• Dose-
response
gradient in a
high

confidence
study

• Unclear
biological
relevance of
increases

Histopathology

1 high confidence
study

• No PFDA-induced histopathological changes
were observed for the clitoral gland, ovaries,
uterus, and mammary glands.

• No factors
noted

• No factors noted

Mechanistic evidence and supplemental information (see subsection above)

Biological events
or pathways

Primary evidence evaluated
Key findings, interpretation, and limitations

Evidence stream
judgment

Hormone levels

Interpretation: PFDA inhibits progesterone

production.

Key findings:

• PFDA reduced progesterone production in mouse
Leydig tumor cells. The study authors suggested
that oxidative stress may be a possible
mechanism.

Limitations: Single study available, lack of evidence
examining effects on other sex hormones.

• Evidence of
decreased
progesterone
production
provides
limited
support for
the biological
plausibility of
the female
reproductive
effects of

This document is a draft for review purposes only and does not constitute Agency policy.

3-207 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence
integration
summary judgment

PFDA. It is not
possible to
corroborate
this effect
with data
from the lone
reproductive
study in rats
(NTP, 2018)
progesterone
was not
measured in
thefNTP,
2018) study.

This document is a draft for review purposes only and does not constitute Agency policy.

3-208 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.6. CARDIOMETABOLIC EFFECTS

Methodological considerations

Cardiometabolic risk refers to the likelihood of developing diabetes, heart disease, or
stroke. Contributors to this risk include a combination of metabolic dysfunctions mainly
characterized by insulin resistance, dyslipidemia, hypertension, and adiposity.

Human studies

There are 22 epidemiology studies that report on the relationship between PFDA exposure
and cardiometabolic effects, including serum lipids (12 studies), blood pressure (5 studies),
atherosclerosis (2 studies), cardiovascular disease (2 studies), ventricular geometry (1 study),
diabetes and insulin resistance (11 studies), adiposity and weight gain (6 studies), and metabolic
syndrome (2 studies).

Serum lipids

Cholesterol as found in, low-density lipoprotein (LDL) is one of the major controllable risk
factors for cardiovascular disease including coronary heart disease, myocardial infarction, and
stroke. Cholesterol levels are typically measured in the blood. Twenty-three studies
(28 publications) report on the association between PFDA exposure and serum lipids (e.g. total
cholesterol, lipoprotein complexes, and triglycerides). There were multiple outcome-specific
considerations for study evaluation that were influential on the ratings. First, for outcome
ascertainment, collection of blood during a fasting state is preferred for all blood lipid
measurements fNIH. 2020: Nigam. 20111 but lack of fasting was considered deficient for
triglycerides and LDL-cholesterol (which is typically calculated using levels of triglycerides, as well
as total cholesterol and HDL, using the Friedewald equation). This is because triglyceride levels
remain elevated for several hours after a meal fNigam. 20111. Self-reported high cholesterol was
also considered deficient due to the high likelihood of misclassifying cases as controls (Nataraian et
al.. 20021. Both of these issues are likely to result in nondifferential outcome misclassification and
to generally bias results towards the null. It was also considered important to account for factors
that meaningfully influence serum lipids, most notably use of cholesterol lowering medications and
pregnancy. Studies that did not consider these factors by exclusion, stratification, or adjustment
were considered deficient for the participant selection domain. All the available studies analyzed
PFDA in serum or plasma and serum lipids using standard, appropriate methods. As described in
Section 3.2.8 on Endocrine effects, reverse causation was considered but is unlikely to significantly
bias the results because PFAS, including PFDA, do not preferentially bind to serum lipids, so
exposure measurement was adequate for this outcome across all studies.

A summary of the study evaluations is presented in Figure 3-69, and additional details can
be obtained from HAWC. Three studies were excluded from further analysis due to critical

This document is a draft for review purposes only and does not constitute Agency policy.

3-209 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 deficiencies in at least one domain. Most studies (14) were classified as medium confidence, though

2 five of these were classified as low confidence for triglycerides and LDL cholesterol due to lack of

3 fasting as described above fBlomberg etal.. 2021: Tensen etal.. 2020a: Yang etal.. 2020: Zengetal..

4 2015: Starling et al.. 2014b! Six studies were classified as low confidence fVarshavsky etal.. 2021:

5 Khalil etal.. 2020: Lin etal.. 2020b: Koshv etal.. 2017: Christensen etal.. 2016: Fu etal.. 20141 for

6 all lipid endpoints. For the majority of studies, sensitivity to detect an effect was a concern due to

7 limited exposure contrast, and thus null associations are interpreted with caution. Potential for

8 confounding across PFAS was considered within individual study evaluations and synthesized

9 across studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-210 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

9-3^' O* GS6^' CN

Sternberg AJ et al. 2021
C-skrrsk, 2022, 10273389'
Chrrstensen, 2016, 3858533-
Dong. 2019. 5080195-
~under Let al. 2022-
Fu. 2014. 3749193-

Kang. 201S. 4937557

Seo, 2018, 4238334-

>&rshavsky JR et al. 2021
Yang J etal. 2020
Vang, 2018, 4238452-
Zeng, 2015, 2851005-
Zhang, 2019, 5033575-

—J

- -

*¦*

- -

-§•

- -

+»

-P

1*0

_
D

Legend

I Gccd (metric) or High confidence (overall)

~ | Adeq ja:e {metric) or Medium confidence (overai "i
Deficient (metric) or Low confidence (overal)
Critically defcient (metric) or Un informative (overa '

~ Mutt z e judgment exist

Figure 3-69. Study evaluation results for epidemiology studies of PFDA and
serum lipids. Refer to HAWC for details on the study evaluation review: HAWC
Human Serum Lipids.

Multiple publications of the same study: Dong et al. (2019) (on figure) includes Christensen et al. (2019) and Jain
and Ducatman (2019a). Liu et al. (2020a) (on figure) includes Liu et al. (2020a)

This document is a draft for review purposes only and does not constitute Agency policy,

3-211 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

The results for the association between PFDA exposure and blood lipids among the medium
confidence studies are presented in Table 3-30. Of the 14 medium confidence studies, 4 were in
general population adults, 3 were in pregnant women, and 7 were in adolescents and children. In
adults, the majority of studies reported higher total cholesterol with higher exposure, including
four in general population adults fCakmak etal.. 2022: Dunder etal.. 2022: Liu etal.. 2020a: Dong et
al.. 20191. two in pregnant women f Gardener etal.. 2021: Starling etal.. 2014al. This included
statistical significance in three studies fCakmak etal.. 2022: Dunder etal.. 2022: Gardener etal..
20211. and an exposure-response gradient in both studies that examined categorical exposure
(Gardener etal.. 2021: Liu etal.. 2020a). Results in children were less consistent. Four studies
reported statistically significant positive associations in at least one analysis f Averina et al.. 2 0 21:
Blomberg etal.. 2021: Tensen etal.. 2020a: Mora etal.. 20181. but other studies reported inverse
fTian et al.. 2020: Kang etal.. 2018: Zeng etal.. 20151 or null associations. In addition, to the
continuous serum lipids measurements, one study (Averina et al.. 20211 examined dyslipidemia as
a dichotomous outcome (defined as total cholesterol >5.17 mmol/L). They reported increased odds
oflipidemia with higher exposure (OR [95% CI] vs quartile 1: Q2: 2.34 [1.08, 5.05], Q3: 2.19 [1.01,
4.74]; Q4: 2.36 [1.08, 5.16]). Results for triglycerides were not available for all studies, but a
positive association was observed in two studies in adults fCakmak etal.. 2022: Dunder etal.. 20221
and one study in pregnant women fGardener etal.. 20211. while the other one study in adults and
two studies in pregnant women showed no association. An inverse association was observed in
Mora etal. (20181 in children; the direction of this association was not coherent with the reported
positive associations for total and LDL cholesterol in the same cohort, which increases uncertainty.
Other studies in children indicated no association with triglycerides.

Looking at the low confidence studies in adults (Varshavskv etal.. 2021: Khalil etal.. 2020:
Lin etal.. 2020b: Christensenetal.. 2016: Fu etal.. 20141 and adolescents fKoshv etal.. 20171. four
reported increases in total cholesterol fLin etal.. 2020b: Koshv etal.. 2017: Fu etal.. 20141 or
unspecified high cholesterol fChristensen et al.. 20161 with increased exposure, with one being
statistically significant (Koshv etal.. 20171. Two studies (Varshavskv et al.. 2 0 21: Khalil etal.. 20201
reported inverse results. The results of all the low confidence studies were interpreted with caution
due to serious limitations.

Overall, evidence for the association between PFDA exposure and serum lipids is
inconsistent, and this inconsistency cannot be easily explained by study confidence level or the
participant-demographics. This may be partly explained by narrow exposure contrasts which may
have reduced sensitivity and impaired the ability of some studies to observe an effect. However,
the strongest associations were observed in studies (Dong etal.. 2019: Mora etal.. 2018: Starling et
al.. 2014a) with low PFDA exposure levels (median <0.5 ng/mL). This could be an indication that
sensitivity in this body of evidence is adequate, or could be due to residual confounding, such as by
other PFAS or the demographics of the study population. There is some support for the PFAS
scenario, as PFDA was highly correlated with PFNA (0.7) and moderately correlated with PFOS and

This document is a draft for review purposes only and does not constitute Agency policy.

3-212 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 PFOA (0.4) in both Starling etal. (2014a) and Dongetal. (20191. and positive associations were

2 stronger for PFOA in Starling etal. (2014a) and for PFNA, PFOS, and PFOA in Dongetal. (20191.

3 Conversely, in Mora etal. f20181. PFDA was highly correlated with PFOA (0.7) and moderately

4 correlated with PFOS (0.6) and PFNA (0.5), but the observed positive associations were strongest in

5 PFDA, and thus are unlikely to be completely explained by confounding. Given available data, there

6 is not enough evidence to state conclusively whether confounding contributed to these results.

Table 3-30. Associations between PFDA and blood lipids in medium
confidence epidemiology studies

Reference

Population

Median exposure
in ng/mL(IQR)

Effect estimate

Total cholesterol

LDL

Triglycerides

General population, adults

Dong et al.
(2019)

Cross-sectional
study, U.S.
(NHANES 2003-
2014); 8,950
adults (20—
80 yrs)

0.2

P (95% CI) for
1-unit increase

6.6
(-8.5, 21.7)

10.7
(-8.5, 29.9)

Cakmak et al.
(2022)

Cross-sectional
study, Canada
(CH MS 2007-
2017); 6,045
participants

0.2 (GM)

% change for

increase
equivalent to
GM

2.8 (0.2, 5.3)*

10.7 (5.5,16.1)*

7.0 (1.0,13.2)*

Dunder et al.
(2022)

Cohort (2001-
2004), Sweden;
864 older adults
(70-80 yrs)

0.3 (0.2-0.4)

P (95% CI) for

change in
exposure and
outcome over
10 yrs

0.23 (0.14, 0.32)*

0.12 (0.03, 0.20)*

0.08 (0.04, 0.12)*

Liu et al.
(2020a)

Cross-sectional
analysis from
randomized
clinical trial of
weight loss; 326
overweight
adults

0.4 (0.2-0.5)

Means ± SE for
tertiles

Tl: 183.1 ±7.9
T2: 186.6 ±7.5
T3: 192.1 ± 7.6
p = 0.2

Tl:138.9± 11.3
T2: 119.7 ± 10.7
T3: 129.3 ± 10.8
p = 0.3

This document is a draft for review purposes only and does not constitute Agency policy.

3-213 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference

Population

Median exposure
in ng/mL(IQR)

Effect estimate

Total cholesterol

LDL

Triglycerides

Pregnant women

Starling et al.
(2014a)

Cross-sectional
analysis from
birth cohort
(2003-2004),
Norway; 891
women

0.09
(
-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference

Population

Median exposure
in ng/mL(IQR)

Effect estimate

Total cholesterol

LDL

Triglycerides

(followed to 18
mo)

Mora et al.

Birth cohort

(1999-2002),

U.S.;

682 children (7-
8 yrs)

0.3
(0.2-0.5)

P (95% CI) for
IQR increase

6.8
(3.6,10.1) *

similar for boys
and girls

3.2
(0.6, 5.8) *

similar for boys
and girls

-3.6
(-8.2, 1.0)
similar for boys
and girls

(2018)

Zeng et al.

Cross-sectional
analysis (2009-
2010), Taiwan;
225 adolescents
(12-15 yrs)

1.0

(range
-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Averma, 2021 7410155 -
Eao.2017 3800095 -
Baukov. 2021 7410153-
Chrislens=:r, 2016- 3S58533 -
ChristenEen, 2013 5080398-
Buar-g, 2019.5083534-
Koshy. 201- 4235478 -
Lira. 2017 38585C4 -
Liu, 2018 1S3724C -
Liu, 2321. 9944393 -
Mobacke, 2018. 4354133-
Stariin-j, 2014 2446659-
2021 7410195 -
Vara, 2018. 4238462

Legend

9 -Sofld (metric) or Rigt confidence (overall)
* I Adeqjase metric: or Medium coifiderce (overall)
- | Oefcient (metre) or Low confidence (overall)
9 Critically ceficiert imetric) or Un
-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

did not report a positive association with preeclampsia. The other two medium confidence studies
reported no increase in the odds of preeclampsia (Huang etal.. 2019b: Starling et al.. 2014a) or
gestational hypertension fHuang et al.. 2019bl. Associations were in the inverse direction in both
studies, but neither was statistically significant In addition, one low confidence study fVarshavskv
etal.. 20211 reported positive associations with continuous blood pressure (both systolic and
diastolic) during mid-gestation.

Table 3-31. Associations between PFDA and hypertensive disorders of
pregnancy in epidemiology studies

Reference,

study
confidence

Population

Median
exposure in
ng/mL (IQR)

Effect estimate

Gestational
hypertension

Preeclampsia

Starling et al.

(2014a).

medium

Nested case-control
study within cohort
in Norway; 1,046
women

0.1

HR (95 CI) for
above vs. below
median

0.81 (0.63, 1.05)

Huang et al.

(2019b).

medium

Cross-sectional study
in China; 674 women
at delivery

0.4(0.2-0.5)

OR (95% CI) for
tertiles vs T1

T2: 1.26(0.48, 3.31)
T3: 0.63 (0.20, 2.00)

T2: 1.16 (0.38, 3.53)
T3: 1.00 (0.31, 3.19)

Liu et al.
(2021a).
medium

Nested case-control
study within cohort
in China; 544 women

0.4(0.3-0.7)

OR (95% CI) for
tertiles vs T1

T2: 1.24(0.74, 2.06)
T3: 1.48(0.89, 2.45)

Birukov et al.
(2021). low

Cohort in Denmark;
1,436 women

0.6(0.5-0.9)

HR (95% CI) for
doubling of
exposure

1.35 (0.86,2.11)

0.93 (0.71, 1.22)

For atherosclerosis, there was a non-significant increase in the echogenicity of the intima-
media complex (a measure of the structural composition of the arterial wall that is an indicator of
early change in the carotid artery) and in the number of carotid arteries with atherosclerotic
plaques only in women in one medium confidence study (Lind etal.. 2017b). but no association with
atherosclerosis in the low confidence study, which did not stratify by sex (Koshv etal.. 2017). In the
single medium confidence study of ventricular geometry (Mobacke etal.. 2018). there was a small
but statistically significant decrease in relative wall thickness (RWT) ((3 = -0.02, 95% CI: -0.04,
-0.01) and increase in left ventricular end-diastolic volume ((3 = 0.95, 95% CI: 0.11,1.79). There is
some inconsistency in the literature about the adversity of decreased RWT, with some studies
indicating increased RWT is associated with hypertension fLi etal.. 20011 and concentric left
ventricular geometry (de Simone et al.. 2005). and others indicating decreased RWT is associated
with abnormal left ventricular geometry (Hashem etal.. 2015) and ventricular tachyarrhythmia
(Biton etal.. 2016). In either case, it is difficult to interpret these results without additional studies.

Overall, there is limited evidence of an association between PFDA exposure and
cardiovascular risk factors. One low confidence study reported a positive association with blood
pressure, and medium confidence studies reported associations with atherosclerosis and

This document is a draft for review purposes only and does not constitute Agency policy.

3-217 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

ventricular geometry, but no association was observed in medium confidence studies of blood
pressure.

Cardiovascular disease

Three studies examined cardiovascular disease and its association with PFDA exposure in
adults. All reported on coronary heart disease fHuang etal.. 2018: Christensen etal.. 2016:

Mattsson etal.. 20151. while one additionally examined total cardiovascular disease, congestive
heart failure, angina pectoris, myocardial infarction (heart attack), and stroke (Huang etal.. 20181.
Two studies were medium confidence (see Figure 3-71), including a case-control study nested
within a prospective cohort of farmers and other rural residents in Sweden fMattsson etal.. 20151.
while the other fHuang etal.. 20181 was based on NHANES, a nationally representative cross-
sectional survey in the U.S. The third study was low confidence and based on a survey of male
anglers in Wisconsin fChristensen et al.. 20161. The timing of exposure measurement in all three
studies was considered adequate, though the prospective measurement in Mattsson etal. (20151
may be more likely to capture the relevant etiologic period of these chronic outcomes. Exposure
levels in the medium confidence studies were similar (median = 0.2 ng/mL), and slightly higher in
the low confidence study (median = 0.5 ng/mL).

For coronary heart disease, Huang etal. f 20181 reported significantly higher odds with
higher exposure (see Table 3-33). Christensen et al. T20161 also reported higher odds, though not
statistically significant, while Mattsson etal. (20151 reported no increase. For other outcomes,
Huang etal. (20181 reported higher odds of total cardiovascular disease, angina pectoris, and
myocardial infarction, and stroke, though these were not statistically significant and only
myocardial infarction and angina pectoris had monotonic gradients across the quartiles (angina
pectoris Q2 vs. Ql: 1.16 (0.67,1.99), Q3: 1.21 (0.75,1.95), Q4: 1.23 (0.68,2.24); myocardial infarction
Q2: 0.99 (0.65,1.49), Q3: 1.32 (0.90,1.92), Q4: 1.38 (0.83,2.28)). There is not a clear explanation for
the differing results in the medium confidence studies; both had similar exposure levels (median
0.2 ng/mL). The populations in Mattsson et al. (20151 and Christensen etal. (20161 are fairly
homogeneous (farmers/rural residents in Sweden and male anglers in Wisconsin, respectively), in
contrast to the nationally representative sample in Huang etal. (20181. It is possible that the
prospective exposure measurement in Mattsson et al. (20151 played a role (vs. cross-sectional
measurement in Huang etal. f 20181 and Christensen etal. f201611. and the lack of additional
prospective studies makes this difficult to interpret Given that the timing of exposure
measurement in Mattsson etal. f20151 is more likely to be during the relevant etiologic window,
the lack of association in that study contributes to considerable uncertainty in this body of
evidence.

This document is a draft for review purposes only and does not constitute Agency policy.

3-218 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Participant selection -

Exposure measurement-

Outcome ascertainment-

Confounding-

Analysis -

Sensitivity -

Selective Reporting -

Overall confidence -

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)

Figure 3-71. Study evaluation results for epidemiology studies of PFDA and
cardiovascular disease. Refer to HAWC for details on the study evaluation review:
HAWC Human Cardiovascular Disease.

Table 3-32. Associations between PFDA and coronary heart disease in
epidemiology studies

Reference, study
confidence

Population

Median exposure
in ng/mL(IQR)

Coronary heart disease
OR (95% CI)

Mattsson et al.

Nested case-control study of farmers and rural
residents in Sweden, exposure measured 1990-
1991 and 2002-2003, cases identified through
2009, N = 462

0.2(0.1)

Q2: 0.87 (0.49, 1.60)
Q3: 1.13 (0.66, 1.94)
Q4: 0.92 (0.53, 1.60)

(2015), medium

Huang et al. (2018),

Cross-sectional study of general population in U.S.
(NHANES), N = 10,859

0.2 (0.2-0.4)

Q2: 1.50(0.97, 2.32)
Q3: 1.17(0.77, 1.79)
Q4: 1.84 (1.26, 2.69) *

medium

Christensen et al.

Cross-sectional study of male anglers in U.S.,
N = 154

0.5 (0.3-0.9)

1.12 (0.49, 2.18)

(2016), low

*p <0.05, NR: not reported.

1 Diabetes and insulin resistance

2 Twenty-one studies (23 publications) reported on the relationship between PFDA exposure

3 and diabetes, insulin resistance, fasting blood glucose, or gestational diabetes. A summary of the

4 study evaluations is presented in Figure 3-72, and additional details can be obtained from HAWC.

This document is a draft for review purposes only and does not constitute Agency policy.

3-219 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

^akmak. 2122.10273388 -

Duan, 2020, 551.3597
Fteiach, 2017. 3858513'

Goodrich. 202". S96C584 ¦

— j 2019 5381135

Sun, 2018. 4241053
Valvi, 2017, 3SS3872
V^lvi. 2021,8438216
Wing, 201S, SCTSfiee
Yu. 2021,7751046
Zhang,2015,2857764¦

¦f

•

1 +

++
++

Legend

Good (metric) or High confdeioe (overall}
Adequate (metric) at Med um confidence (overal j
Deficient (metric) or Low confidence (overall)
Uritically deficient (meSric) or b r irfcmnative (overa

Figure 3-72. Study evaluation results for epidemiology studies of PFDA and
diabetes and insulin resistance. Refer to HAWC for details on the study evaluation
review: HAWC Human Diabetes and Insulin Resistance.

Multiple publications of the same study: Christensen et al, (2019) includes Jain (2021 and Jain (2020a).

This document is a draft for review purposes only and does not constitute Agency policy,

3-220 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

For diabetes, due to concerns for reverse causality resulting from metabolic and behavioral
changes following a diabetes diagnosis, the optimal epidemiological studies would be longitudinal
cohort studies with repeated measurements before onset. Two medium confidence studies
evaluated PFDA exposure and incident diabetes fCharles etal.. 2020: Sun etal.. 20181. Sun et al.
f20181. a nested case-control study, found that at the highest tertile of PFDA exposure (range: 0.2-
1.95 ng/mL), there was a non-statistically significant inverse (i.e., "protective") association seen
with diabetes (OR = 0.7, 95% CI: 0.5,1.1). Charles etal. (2020). also a nested case-control study,
reported results that differed based on the selected control group; an inverse association was
observed with controls matched for birth year and year of blood collection, controlling for BMI (OR
= 0.89, 95% CI: 0.55,1.44), while a positive association was observed with controls additionally
matched for BMI (OR = 1.52, 95% CI = 0.76, 3.07), though neither was statistically significant.

For insulin resistance and blood glucose, there were several outcome-specific
considerations for study evaluation that were influential on the ratings. Homeostatic model
assessment (HOMA) is a method for assessing insulin resistance and (3-cell function from fasting
glucose and insulin measured in the plasma (Matthews etal.. 1985). The HOMA of insulin
resistance (HOMA-IR) is often used in studies evaluating future risk for diabetes and was
considered a primary outcome for this review along with fasting blood glucose. Measures of insulin
resistance and blood glucose, including HOMA-IR, are not interpretable in the presence of diabetes,
particularly if diabetes is treated with hypoglycemic medication since the treatment will affect
insulin production and secretion. Studies that did not consider diabetes status and use of diabetes
medications by exclusion, stratification, or adjustment were thus considered deficient for
participant selection. For the timing of the exposure measurement, unlike the criteria described for
diabetes, exposure and outcome can be assessed concurrently as insulin resistance and blood
glucose can represent short-term responses, and establishing temporality was not deemed a major
concern.

Sixteen studies examined associations between PFDA exposure and insulin resistance or
fasting blood glucose. Nine studies examined associations in adolescents and adults, five studies in
pregnant women, and two studies in children. Six studies did not consider diabetes status of
participants and were thus considered low confidence (Khalil etal.. 2020: Lin etal.. 2020b: Kang et
al.. 2018: Liu etal.. 2018: Fleisch etal.. 2017: Koshv etal.. 2017). The remaining ten studies were
medium confidence fCakmak et al.. 2022: Gardener etal.. 2021: Goodrich etal.. 2021: Valvi etal..
2021: Yu etal.. 2021: Duan etal.. 2020: Ren etal.. 2020: Christensen etal.. 2019: Tensen etal.. 2018:
Wang etal.. 20181six were low confidence.

Results of the insulin resistance and fasting blood glucose are presented in Table 3-34. In
all studies of insulin resistance, the results were generally null, and in the low confidence study by
Fleisch etal. (2017). an inverse association was observed. In the studies of fasting blood glucose,
there was again no clear positive association observed. It is possible that the null associations
could be due to poor sensitivity from narrow exposure contrasts in most of the studies, but a

This document is a draft for review purposes only and does not constitute Agency policy.

3-221 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 minority of studies had higher exposure levels with corresponding greater contrast and also found

2 no association. Additionally, null, and even inverse associations could be due to outcome

3 misclassification resulting from inclusion of participants with diabetes in some studies. However,

4 based on the current evidence, there is no indication that PFDA exposure is associated with greater

5 insulin resistance or higher fasting blood glucose levels.

This document is a draft for review purposes only and does not constitute Agency policy.

3-222 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-33. Associations between PFDA and insulin resistance in epidemiology studies

Reference

Confidence

Population

Median exposure
(IQR) in ng/mL or
as specified

Exposure change

Effect
estimate

Fasting
blood glucose

Insulin resistance
(HOMA-IR)

General population, adolescents, and adults

Goodrich et
al. (2021)

Medium

Cohort and cross-sectional study of
adolescents in U.S.; 310 in cohort and 137
in cross-sectional

NR (due to high
proportion below
the LOD)

"Not associated"

Koshv et al.
(2017)

Low

World Trade Center Health Registry
(WTCHR) who resided in NYC and were
born between Sept. 11,1993 and Sept. 10,
2001; U.S.; 402 adolescents

Control
0.1 (0.2)
WTCHR
0.1 (0.1)

In-unit change

Beta
coefficient
(95% CI)

-0.04 (-0.11, 0.03)#

Christensen et
al. (2019)

Medium

Cross-sectional study in U.S. (NHANES
2007-2014); 2975 individuals aged
20 years and older

0.2 (0.1-0.4)

Quartiles

Odds ratio
(95% CI)

Q2: 0.9 (0.7, 1.3)
Q3: 1.1 (0.7, 1.7)
Q4: 0.9 (0.6, 1.5)

Cakmak et al.
(2022)

Medium

Cross-sectional study in Canada (CHMS
2007-2017); 3,356-6,024 individuals 12
years and older

GM 0.2

Change equivalent
to GM

Percent
change

-0.3 (-1.4, 0.8)

5.3 (-3.5, 15.0)

Valvi et al.
(2017)

Medum

Birth cohort in Faroe Islands; 699 young
adults

0.2 (0.2-0.3)

Log2 change

P (95% CI)

Glucose AUC
Exposure at 7 yr
0.0 (-0.01,0.02)
Similar with exposure
at 14, 22, 28 yr and in
men and women

Exposure at 7 yr
0.03 (-0.03, 0.10)
Similar with exposure
at 14, 22, 28 yr and in
men and women

Khalil et al.
(2020)

Low

Cross-sectional study of firefighters in U.S.;
38 men

0.3 (0.2-0.3)

Log unit change

P (95% CI)

No association
(estimates reported
on figure)

Liu et al.
(2018)

Low

Cross-sectional analysis in weight loss
clinical trial
in U.S.; 621 adults
(30-70 yrs)

Male
0.4 (0.3-0.5)

Female
0.4 (0.3-0.6)

n/a

Spearman
correlation

0.08

0.05

This document is a draft for review purposes only and does not constitute Agency policy.
3-223 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference

Confidence

Population

Median exposure
(IQR) in ng/mL or
as specified

Exposure change

Effect
estimate

Fasting
blood glucose

Insulin resistance
(HOMA-IR)

Lin et al.

Low

Cross-sectional study of older adults living
near a high contamination area in Taiwan;
397 adults (55-75 yrs)

Median (range)
1.7(0.6-27)

Quartiles

P (95% CI)

Women
Q2: -4.83 (-13.34,

3.68)
Q3: -4.33 (-12.91,

4.26)
Q4: -5.72 (14.37,
2.94)

Men

Q2: -5.38 (-19.68,

8.92)
Q3: 2.67 (-11.7,
17.05)
Q4: 3.9 (-11.1, 18.9)

(2020b)

Duan et al.

Medium

Cross-sectional study in China; 294 adults

2.1 (1.0-4.1)

1% increase

Percent
change

0.009 (-0.002, 0.020)

(2020)

Pregnant women

Gardener et

Medium

Pregnancy cohort in U.S.; 433 pregnant
women

0.2 (0.1-0.3)

Quartiles

Means (95%
CI)

Insulin: No
association (estimates
reported on figure)

al. (2021)

Jensen et al.

Medium

Birth cohort in Denmark; 649 pregnant
women (15-49 yrs)

0.3 (0.2-0.5)

Two-fold change

% Change
(95% CI)

-1.3 (-3.6, 1.0)

-1.5 (-13.5,12.1)

(2018)

Wang et al.

Medium

1:2 matched case control of pregnant
women in China; 84 cases and 168
noncases

Controls
0.3 (0.2-0.4)

Cases
0.3 (0.2-0.4)

Dichotomous
exposure (tertiles
of outcome)

Odds ratio
(95% CI)

Medium vs. Lowest
FBG
1.3 (0.7-2.4)
Highest vs. Lowest
FBG
1.0 (0.5-1.8)

(2018)

Yu, 2021,

Medium

Pregnancy cohort in China; 2,747 pregnant
women

1.7(1.4)

Log-unit change

P (95% CI)

0.01 (-0.02, 0.04)
1 hr post glucose
tolerance test
0.12 (0.01, 0.22)

2 hr post
0.08 (-0.002, 0.17)

This document is a draft for review purposes only and does not constitute Agency policy.
3-224 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference

Confidence

Population

Median exposure
(IQR) in ng/mL or
as specified

Exposure change

Effect
estimate

Fasting
blood glucose

Insulin resistance
(HOMA-IR)

Ren et al.
(2020)

Medium

Pregnancy cohort in China; 856 pregnant
women

2.0 (1.3-3.2)

In-unit change

OR (95% CI)
for high
glucose

1.24 (0.87, 1.76)
1 hr post glucose
tolerance test
1.61 (1.10, 2.44)

Children

Fleisch et al.
(2017)

Low

Birth cohort in U.S.; 665 mother-children's
pairs

GM (IQR)
Mid-childhood
0.3(0.2, 0.5)

Quartiles

Beta
coefficient
(95% CI)

Mid-childhood
Q2: -7.1 (-22.1, 10.6)
Q3: -31.3 (-42.8,

-17.5) *
Q4: -21.5 (-34.0,
-6.7)*

Kane et al.
(2018)

Low

Cross-sectional study in South Korea; 150

children

(3-18 yrs)

0.06 (0.04-0.1)

In-unit change

Beta
coefficient
(95% CI)

-0.2 (-1.3, 0.9)

*p-value or p-trend < 0.05.

# HOMA-IRwas log-transformed.

Note: Not all results (e.g., sub-group analyses, different exposure classification) were extracted from each study if additional results did not change the
interpretation.

NR = not reported.

This document is a draft for review purposes only and does not constitute Agency policy.
3-225 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Six studies reported on the association between PFDA exposure and gestational diabetes
(Liu etal.. 2019b: Rahman etal.. 2019: Wang etal.. 2018: Valvi etal.. 2017: Zhang etal.. 20151. Four
studies were medium confidence, one was low confidence, and one fLiu etal.. 2019bl was
uninformative due to lack of control for confounding in single-pollutant models. The three medium
confidence studies were inconsistent, with one fValvi etal.. 20171 reporting higher odds of
gestational diabetes with higher exposure (OR for doubling of exposure: 1.2 (0.7,2.0)), but the
association was not statistically significant and non-monotonic (OR for tertile 2: 2.0 (0.9,4.1), tertile
3: 1.0 (0.5,2.3)). Two medium confidence studies reported close to null association with gestational
diabetes and PFDA exposure (OR: 1.02 (0.86,1.20) in the overall cohort in Rahman etal. (20191. OR
0.95 (0.78,1.16) in Yu etal. f202111. and the other medium confidence study Wang etal. T20181
reported a non-statistically significant inverse association (OR: 0.85 (0.30-2.92)). The low
confidence study fZhang et al.. 20151 reported no association (OR: 1.0 (0.7-1.5)).

Overall, for diabetes and insulin resistance, there were no clear associations with higher
PFDA exposures. Results were generally null or in the inverse direction. While it is possible that a
positive association with these outcomes exists but was obscured by poor sensitivity and/or bias,
there is no clear explanation for the inconsistency based on study confidence, design, or population.

Adiposity

Thirteen studies reported on the association between PFDA exposure and obesity or related
outcomes. Two studies were excluded due to critical deficiencies in participant selection (Yang et
al.. 20181 and confounding (Zhao etal.. 2022: Yang etal.. 20181. Of the 11 remaining studies, four
were cohorts that examined early life exposure to PFDA and adiposity at 18 months (Karlsen etal..
20171. at 4-8 years of age (Bloom etal.. 20221. at 5 years of age (Chen etal.. 2019: Karlsen etal..
20171. and at 13 years of age flanis etal.. 20211: one was a clinical trial of weight loss diets in adults
that examined weight change fLiu etal.. 20181: and one was a cohort of adults living near a uranium
processing site f Blake etal.. 20181. All of these were classified as medium confidence. Five studies
(three in adults and two in children) were cross-sectional (Lind etal.. 2022: Wise etal.. 2022:
Thomsen et al.. 2021: Domazetetal.. 2020: Christensen etal.. 20191 and were low confidence due
to the potential for reverse causation resulting from metabolic changes in obese individuals. The
evaluations are summarized in Figure 3-73.

This document is a draft for review purposes only and does not constitute Agency policy.

3-226 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

em*. 2018. SQ30657
Btaom, 2022 9959635
Chen. 2D 15. 50S0578-
Chrin*nt»n. 2019 S0803$e
Domszei. 2020, 6833700 -
Janii 2021 7410181
Karisen. 2017. 3858520-
Llnd, 2D22. 10176401 -
Liu. 2018. 1937240
TlHHnsen, 2021, 9959568
Wise 2022 9959470
Y»ng. 2016. 4238462
Zhao, 2022. 10273285-

Legerid

| Good (rnetnc) or High conlifjsnce (overall)
Adequate (melrie) or H#<(!inn confiijs-nce (ov«t«S)
D«f>ci«ft< or tow conf«J«o<» (owftl)

CritlcjKy (metric) or Uninfonmativ* jowtal)

Figure 3-73. Study evaluation results for epidemiology studies of PFDA and
adiposity. Refer to HAWC for details on the study evaluation review: HAWC Human
Adiposity.

Multiple publications of the same study: Christensen et al. (2019 includes Jain (2020al

The available studies look at several different outcomes and populations, so are generally
not directly comparable (see Table 3-35). In the five studies in adults, one medium confidence
study reported higher BMI with higher exposure fBlake etal.. 20181 and the other medium
confidence study reported greater weight gain following a weight loss trial (Liu etal. 20181. with
only the latter being statistically significant. Of the three low confidence cross-sectional studies,
two reported statistically significant inverse associations with BMI in women (Lind et al.. 2022:

Wise etal.. 20221. while the third also reported an inverse, though not statistically significant,
association with waist circumference. In children, one medium confidence birth cohort fKarlsen et
al.. 20171 reported a slightly higher proportion of overweight participants with higher exposure at
18 months when maternal exposure was modeled as a continuous variable (RR = 1.14,
95% CI 0.91,1.43), but this was not statistically significant and not monotonic when modeled in

This document is a draft for review purposes only and does not constitute Agency policy.

3-227 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

tertiles (RRT2 vs. T1 = 0.90 (95% CI: 0.71,1.15), T3 vs. T1 = 1.03 (95% CI: 0.82, 1.31))or in follow-
up of the children at 5 years. However, a cross-sectional analysis by Karlsen et al. (20171 in this
population at 5 years indicated lower BMI and incidence of children who were overweight with
higher exposure. A second medium confidence birth cohort study reported non-significant inverse
associations in girls and non-significant positive associations in boys at 5 years fChen etal.. 20191.
The other two medium confidence cohort studies, including a birth cohort with exposure
measurement in gestation and follow-up to 4-8 yrs (Bloom etal.. 20221 and a cohort with exposure
measurement in mid-childhood (age 8) and follow-up to age 13 (Tanis etal.. 20211 were null overall
with regard to BMI and fat mass. The two low confidence cross-sectional studies reported inverse
associations with fat mass fDomazet et al.. 20201 and measures of fat obtained with MRI and Dual
X-ray absorptiometry fThomsen et al.. 20211. Overall, there is some limited evidence of an
association between PFDA exposure and adiposity in adults in two medium confidence studies, but
there is considerable uncertainty, and this association was not observed in studies of children.

This document is a draft for review purposes only and does not constitute Agency policy.

3-228 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-34. Associations between PFDA and adiposity in epidemiology studies

Reference, study
confidence

Population

Median exposure (IQR) in
ng/mL

Effect estimate

BMI

Waist circumference

Other

Adults

Blake et al.
(2018). medium

Prospective cohort near a
uranium processing site in
U.S.; 210 adults

0.1(0.1, 0.2)

% change
(95% CI) for IQR
increase in exposure

0.7 (-1.3,2.7)

Christensen et
al. (2019). low

NHANES, cross-sectional in
U.S.; 2,975 adults (20+ yr)

0.2 (0.1, 0.4)

OR (95% CI) for
increased WC by
quartiles (ref Ql)

Q2: 0.9 (0.6, 1.2)
Q3: 0.9 (0.5, 1.5)
Q4: 0.8 (0.5, 1.3)

Liu et al. (2018).

medium

Clinical trial of weight loss
diet in U.S.; 621 adults

Male
0.4 (0.3-0.5)

Female
0.4 (0.3-0.6)

Mean difference

Weight gain following trial
Tl: 2.5 ±0.9
T2: 3.1 ±0.9
T3: 4.2 ±0.8,
p-trend: 0.03

Children

Karlsen et al.
(2017), medium

Prospective birth cohort in
Faroe Islands; 444 children
at 18 mo and 371 at 5 yr

P (95% CI) for BMI;
Relative risk for
overweight

18 mo 0.1 (-0.1, 0.3)
5 yr (-0.04 (-0.2,
0.1)

Overweight
18 mo 1.1 (0.9,1.4)
5 yr 1.0 (0.6,1.7)

Chen et al.
(2019). medium

Prospective birth cohort in
China; 404 children at 5 yr

0.4 (range 0.2-2.0)

P (95% CI) for log-unit
change

Girls: -0.2 (-0.4, 0.1)
Boys: 0.1 (-0.3, 0.5)

Girls: -0.7 (-1.5, 0.1)
Boys: 0.2 (-0.8,1.0)

Body fat percentage (%)
Girls:-1.1 (-2.3, 0.2)
Boys: 1.1 (-0.2, 2.3)

P (95% CI) for tertiles
(ref Tl)

Girls

T2: -0.1 (-0.7, 0.4)
T3: 0.0 (-0.6, 0.5)
Boys

T2: -0.2 (-0.8, 0.4)
T3: 0.2 (-0.5, 0.8)

Girls

T2: -0.6 (-2.2, 1.0)
T3: -0.5 (-2.2, 1.1)
Boys

T2:-0.9 (-2.5, 0.7)
T3: 0.5 (-1.1, 2.1)

Girls

T2: -0.6 (-3.2, 1.9)
T3: -1.5 (-4.1, 1.0)

Boys
T2: 0.5 (-1.5, 2.6)
T3: 2.0 (-0.1, 4.1)

NR = not reported.

This document is a draft for review purposes only and does not constitute Agency policy.

3-229 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Metabolic syndrome

The current criteria for clinical diagnosis of metabolic syndrome include the following:
larger waist circumference; elevated triglycerides >150 mg/dL (1.7 mmol/L); reduced HDL-C
<40 mg/dL (1.0 mmol/L) in males and <50 mg/dL (1.3 mmol/L) in females; elevated blood
pressure: systolic >130 and/or diastolic >85 mm Hg; and elevated fasting glucose >100 mg/dL
(Alberti et al.. 2009). Main considerations are that three abnormal findings out of five in the criteria
would qualify a person for the metabolic syndrome and that country- or population-specific cut
points for waist circumference should be used (Alberti et al.. 2009).

Three studies reported on the association between PFDA exposure and metabolic
syndrome. One study was uninformative due to critical deficiencies in participant selection fYang et
al.. 20181. The remaining two studies were cross-sectional, with one fChristensen et al.. 20191
being medium confidence and one being low confidence fLin etal.. 2020bl. Christensen et al. f20191
found an exposure-dependent, significant inverse association between PFDA exposure and
metabolic syndrome (OR: 0.72; 95%CI: 0.54, 0.97 within (PFDA); quartile 2: 0.93; 95%CI: 0.64,
1.35, quartile 3: 0.71; 95%CI: 0.43, 1.18, and quartile 4: 0.56; 95%CI: 0.31,1.01). Lin etal. C2020bl
also reported an inverse association (not statistically significant) in women (OR (95% CI) for
quartilesvs. Ql, Q2: 0.68 (0.33, 1.4); Q3: 0.78 (0.38,1.61); Q4: 0.51 (0.24, 1.08) but reported a
positive association (also not statistically significant) in men (Q2: 0.94 (0.31, 2.85); Q3: 1.43 (0.48,
4.22); Q4: 1.9 (0.63,5.77).

Animal studies

There is a single study available in experimental animals that evaluated endpoints related to
cardiometabolic effects following short-term exposure to PFDA fNTP. 20181. The study exposed
female and male SD rats to PFDA doses of 0, 0.156, 0.312, 0.625,1.25 and 2.5 mg/kg-day for 28 days
via gavage and included endpoints such as serum lipids, histopathology, and organ weights.
Confidence in the study was rated as high during study evaluation for these endpoints with no
outstanding issues regarding risk of bias or sensitivity (see Figure 3-74).

This document is a draft for review purposes only and does not constitute Agency policy.

3-230 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Reporting quality
Allocation
Observational bias/blinding
Confounding/variable control
Selective reporting and attrition
Chemical administration and characterization
Exposure timing, frequency and duration
Endpoint sensitivity and specificity
Results presentation
Overall confidence

ffi

,V2-1

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)
Not reported

Figure 3-74. Evaluation results for animal study assessing effects of PFDA
exposure on cardiometabolic effects. Refer to HAWC for details on the study
evaluation review.

Histopathology

The heart and blood vessel were examined histologically in rats in the control and high-dose
groups (2.5 mg/kg-day) at study termination (see Figure 3-75). An increase in the incidence of
granulomatous inflammation of the epicardium (2/10 rats; moderate severity) was reported in
high-dose females after PFDA exposure. Granulomas are focal, inflammatory tissue responses that
arise from a broad range of etiologies, including infectious and non-infectious processes fBoros and
Revankar. 20171. This lesion was not observed in exposed males or in the controls. Results for
blood vessel histopathology were null. The biological significance of the histopathological
observations in females is unknown given the sparse information available.

Serum lipids

Cholesterol is important for maintaining cell membrane integrity and transport and is also
used as a precursor for the synthesis of steroid hormones, bile acids and other substances in the
body. Triglycerides are an essential source of energy storage and production. Both cholesterol and
triglycerides are routinely evaluated in blood lipid panels as cardiovascular risk measures.
Cholesterol and triglyceride levels were measured in rat serum after 28-day exposure (see

This document is a draft for review purposes only and does not constitute Agency policy,

3-231 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-36 and Figure 3-75. Dose-related decreases in triglyceride levels were reported in male
and female rats exposed to PFDA, with the largest changes occurring in males at the highest doses
(35% and 52% compared to controls at 1.25 and 2.5 mg/kg-day, respectively). A downward trend
(p < 0.01) was reported for cholesterol levels in females, reaching 35% compared to controls at
2.5 mg/kg-day. In males, cholesterol decreased 14-38% compared to controls across 0.156-
2.5 mg/kg-day, but the effects did not display a significant trend. The findings should be
interpreted with caution given the known species differences in lipid metabolism and blood
cholesterol levels between rodents and humans that may impact the evaluation of the human
relevance of the observed responses (Getz and Reardon. 2012: Davidson. 2010).

Table 3-35. Percent change relative to controls in serum lipids in a 28-day rat
study after PFDA exposure fNTP. 20181

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Triglycerides

Male S-D rats

-2

-21

-35

-52

Female S-D rats

-7

-23

-27

Cholesterol

Male S-D rats

-27

-38

-27

-12

-14

Female S-D rats

-8

-9

-35

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors.

Organ weight

Terminal absolute and relative heart weights were measured in all exposed animals (see
Table 3-37 and Figure 3-75). It is unclear which metric (i.e., absolute, or relative) would be more
appropriate to evaluate effects on heart weight in the presence of significant body weight changes
fBailev et al.. 20041. As such, both absolute and relative measures were considered herein.
Absolute heart weight showed a decreasing trend (p < 0.01) in males and females, with 15-37%
decreases compared to controls at doses of 1.25 and 2.5 mg/kg-day. In contrast, changes in
relative heart weights did not show a significant trend. The reductions in absolute heart weight
coincide with reductions in body weight observed in these animals at the high-dose groups
(>1.25 mg/kg-day) (see Section 3.2.10 on General toxicity effects for additional details).

This document is a draft for review purposes only and does not constitute Agency policy.

3-232 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-36. Percent change relative to controls in heart weights in a 28-day
rat study after PFDA exposure fNTP. 20181

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Absolute heart weight

Male S-D rats

-2

-18

-37

Female S-D rats

-2

-15

-36

Relative heart weight

Male S-D rats

-1

Female S-D rats

-3

-2

-3

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study

authors.

This document is a draft for review purposes only and does not constitute Agency policy.

3-233 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Organ

Study Name

Outcome Confidence

Exposure Design

Species, Strain (Sex)

Trend Test Result

PFDA Cardiometabolic Effects

Triglyceride (TRIG)

Blood

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (o)

significant

• • •

—•

—¦V

Rat, Sprague-Dawley (Harlan) (', )

significant

—•

•

—•

Cholesterol (CHOL)

Blood

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (. ••)

not significant

•

—•

Rat, Sprague-Dawley (Harlan) ($)

significant

—•

•

—~

Histopathology

Heart

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (o)

not significant

•

—•

Epicardium, Granulomatous Inflammation

Heart

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (9)

not significant

•

—•

Histopathology

Blood Vessel

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (•)

not applicable

•

—•

Rat, Sprague-Dawley (Harlan) (V)

not applicable

•

—•

Heart Weight, Absolute

Heart

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (o)

significant

Rat, Sprague-Dawley (Harlan) (y)

significant

—•

Heart Weight, Relative

Heart

NTR 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (o)

not significant

—•

•

—•

Rat, Sprague-Dawley (Harlan) (9)

not significant

—•

—•—

—•

# No significant change^ Statistically significant increase ~ Statistically significant decrease I 0 5 ^ 1 15

J Dose (mg/kg-day)

Figure 3-75. Cardiometabolic effects following exposure to PFDA in short-term oral studies in animals (results can

be viewed by clicking the HAWC link: https://hawcprd.epa.gOv/summarv/data-pivot/assessment/100500072/pfda-
cardiometabolic-effects/1.

This document is a draft for review purposes only and does not constitute Agency policy,

3-234 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence integration

The evidence of an association between PFDA exposure and cardiometabolic effects in
humans is slight, with an indication of higher serum lipids, adiposity, cardiovascular disease, and
possible markers of atherosclerosis with higher PFDA exposure. While most results were imprecise
and not statistically significant, exposure contrasts for PFDA in the study populations were
relatively narrow, which is interpreted to result in low sensitivity to detect an effect. However,
there is inconsistency across studies for similar outcomes, so there is considerable uncertainty in
the evidence. There is no evidence of an association with diabetes, insulin resistance, and metabolic
syndrome, but the null results are difficult to interpret due to concerns for sensitivity.

Overall, the animal evidence is indeterminate given that the observed changes fail to
establish a coherent pattern of adverse cardiometabolic effects in animals following short-term
PFDA exposure. The evidence in animals is limited to a high confidence study in rats exposed via
gavage for 28 days that examined cardiovascular histopathology, serum lipids and heart weights
(NTP. 20181. Dose-related decreases in triglyceride levels occurred in males and females and
cholesterol also decreased dose-dependently in females. However, the biological significance of
these responses is unclear. Absolute heart weights decreased dose-dependently in rats at the
highest doses (>1.25 mg/kg-day) but confidence in the results is reduced by potential confounding
with decreased body weights and a lack of corroborative findings from histopathological
evaluations or other organ weight measures (relative heart weight was unchanged). A major
limitation in the animal toxicity database of this chemical is the lack of studies examining prolonged
or chronic oral exposures. In addition, for some cardiometabolic endpoints (i.e., serum lipids), it
would be preferred if studies were available in models that are more physiologically relevant to
humans given species differences in lipid metabolism between humans and rodents (Getzand
Reardon. 2012: Davidson. 20101. In the absence of such studies or mechanistic information on
these responses, the human relevance of effects on rodent lipid profiles cannot be determined.

Overall, evidence suggests that PFDA exposure has the potential to cause cardiometabolic
effects in humans under sufficient exposure conditions (see Table 3-38). This conclusion is based
on evidence of an association between PFDA exposure and certain cardiometabolic outcomes
(serum lipids, adiposity, cardiovascular disease, and atherosclerosis) in a small number of
epidemiological studies with median exposure levels from 0.1-0.4 ng/mL; however, issues with
inconsistency across studies raise considerable uncertainty. Moreover, evidence in animals is
sparse and largely uninterpretable regarding its relevance to humans.

This document is a draft for review purposes only and does not constitute Agency policy.

3-235 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-37. Evidence profile table for PFDA exposure and cardiometabolic effects

Evidence stream summary and interpretation

Evidence integration summary
judgment

Evidence from studies of exposed humans (see Section 3.2.7: Human studies)

®oo

Evidence suggests

Primary basis:

Some coherent effects in a small
number of medium confidence
epidemiological studies, but
data is largely inconsistent.
Evidence from a high confidence
rat study was indeterminate.

Human relevance:

The utility of the observed
serum lipid effects in rats for
informing human health hazard
is uncertain given the species
differences in lipid metabolism
between humans and rodents.

Cross-stream coherence,
susceptibility, and other
inferences:

No specific factors are noted.

Studies, outcomes, and
confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream
judgment

Serum lioids
14 medium and 6 low
confidence studies

• Five of six medium
confidence studies in
adults (including two
in pregnant women)
reported higher
serum total
cholesterol with
higher PFDA
exposure (p <0.05 in
three studies).

• In children, results
were inconsistent.

• Consistency of
direction of
association across
studies in adults for
total cholesterol.

• Exposure-response
gradient in the only
two studies that
examined
categorical
exposure.

• Imprecision in most
positive associations

• Lack of coherence
across measures (total
cholesterol and
triglycerides) in some
studies

®oo

Slight

Positive associations
between PFDA and serum
lipids, adiposity,
cardiovascular disease,
and atherosclerosis in
some studies, but with
the exception of total
cholesterol in adults,
findings were inconsistent
or incoherent across
studies. Exposure levels
were low, which may
explain the lack of
association in some
studies.

Other cardiovascular risk
factors

9 medium and 4 low
confidence studies

• Studies of blood
pressure in the
general population
were largely null.
Three of five studies
reported

hypertension or a
positive association
with blood pressure
among pregnant
women, but there
was inconsistency
among medium
confidence studies.

• There was a non-
significant increase

• No factors noted

• Unexplained
inconsistency across
studies for blood
pressure

• Imprecision in positive
associations observed
for blood pressure and
atherosclerosis

This document is a draft for review purposes only and does not constitute Agency policy.

3-236 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

in the number of
carotid arteries with
atherosclerotic
plaques in women in
one study.

• One study reported
statistically
significant changes
in ventricular
geometry.

Cardiovascular disease
2 medium and 1 low
confidence studies

• One medium and
one low confidence
studies reported
higher odds of
coronary heart
disease (the former
being statistically
significant), but
another medium
confidence study
was null.

• Higher odds of
angina pectoris,
myocardial
infarction, and
stroke were
reported in the
single study that
examined them.

• No factors noted

• Unexplained
inconsistency across
medium confidence
studies, possibly
related to timing of
exposure
measurement

• Imprecision in results
of specific
cardiovascular
conditions

Diabetes and insulin
resistance

15 medium and 6 low
confidence studies

• One study reported
higher odds of
gestational diabetes
with higher PFDA
exposure, but the

• No factors noted

• Unexplained

inconsistency across
studies

This document is a draft for review purposes only and does not constitute Agency policy.

3-237 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

association was non-
monotonic and not
statistically
significant. Other
studies reported
either null or inverse
associations with
gestational diabetes.

• Two studies of
incident diabetes
and 16 studies of
insulin resistance
indicated primarily
null associations
with PFDA exposure.

• Low sensitivity
across majority of
studies

Adiposity

6 medium and 5 low
confidence studies

• One study in adults
reported an increase
in weight gain
(significant trend)
and one reported
higher BMI with
higher PFDA
exposure, but other
studies reported null
or inverse
associations

• Low sensitivity
across studies

• No factors noted

• Unexplained

inconsistency across
studies

Metabolic svndrome
2 medium confidence
studies

• Inverse association
between metabolic
syndrome and PFDA

• No factors noted

This document is a draft for review purposes only and does not constitute Agency policy.

3-238 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

exposure in two
studies (one
reported a positive
association in men).

Evidence from in vivo animal studies (see Section 3.2.7: Animal studies)

Studies, outcomes, and
confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream
summary

Histopathologv

1 high confidence study in
rats for 28 d

• No significant effects
in heart and blood
vessel

histopathology in
rats up to 2.5 mg/kg-
d

• High confidence
study

• No factors noted

QQQ

Indeterminate

Lack of coherent, adverse
effects indicative of
cardiometabolic toxicity.

Serum lipids

1 high confidence study in
rats for 28 d

• Decreases in

triglyceride (males
and females) and
cholesterol levels
(females only) in rats
at >1.25 mg/kg-d for
28 d

• Dose-response
gradient for most
effects

• High confidence
study

• Unclear biological
significance of
decreases in lipids

Organ weight

1 high confidence study in
rats for 28 d

• Decreases in

absolute (but not
relative) heart
weight in rats at
doses >1.25 mg/kg-d

• Dose-response
gradient for
absolute heart
weights

• High confidence
study

• Unexplained
inconsistency across
heart weight measures

• Potential confounding
by body weight
decrease (particularly
since only absolute
weights affected)

C = cohort study; CS = cross-sectional study; CC = case-control study.

This document is a draft for review purposes only and does not constitute Agency policy.

3-239 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.7. NEURODEVELOPMENTAL EFFECTS
Human studies

N eurodevelopment

There are 13 studies (19 publications) of PFDA and neurodevelopmental outcomes in
humans. The study evaluations are summarized for Figure 3-76. In the case of multiple publications
for the same study population, they were evaluated under one record if the selection procedures for
the analysis population were similar but evaluated under different records if selection procedures
were significantly different (see figure footnote for details). All but one study fGump etal.. 20111
was medium confidence, however all but fNiu etal.. 20191 were deficient for study sensitivity due to
limited exposure contrast With the exception of Gump etal. f20111. all studies were birth cohorts
or case-controls studies nested in cohorts that evaluated maternal exposure to PFDA during
pregnancy and/or during childhood. Functionally there is considerable overlap between different
domains of neurodevelopment, but for the purposes of this review, the outcomes were categorized:
eight studies (9 publications) examined Attention Deficient Hyperactivity Disorder (ADHD),
attention, or related behaviors fDalsager etal.. 2021b: Harris etal.. 2021: Skogheim etal.. 2021:
Luo etal.. 2020: Vuong etal.. 2018: Haver etal.. 2017: Oulhote etal.. 2016: Liew etal.. 2015: Gump
etal.. 20111. eight studies (ten publications) examined cognition and summary measures of
neurodevelopment (Yao etal.. 2022: Harris etal.. 2021: Skogheim etal.. 2020: Niu etal.. 2019:

Harris etal.. 2018: Liew etal.. 2018: Lvall etal.. 2018: Vuong etal.. 2018: Vuong etal.. 2016: Wang
etal.. 20151. five studies examined autism spectrum disorder (ASD) or social behaviors (Skogheim
etal.. 2021: Shin etal.. 2020: Niu etal.. 2019: Lvall etal.. 2018: Liew etal.. 20151. three examined
motor effects fYao etal.. 2022: Niu etal.. 2019: Harris etal.. 20181. and one examined congenital
cerebral palsy fLiewetal.. 20141.

This document is a draft for review purposes only and does not constitute Agency policy.

3-240 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Dalsager, 2021, 9960591 -
Gump, 2011, 3858629-
Harris, 2018, 4442261 -
H0yer, 2017,4184660-
Liew, 2014, 2852208-
Liew, 2015, 2851010-
Liew, 2018, 5079744-
Luo, 2020, 7175034-
Lyall, 2018, 4239287-
Niu, 2019, 5381527-
Oulhote, 2016, 3789517-
Shin, 2020, 6507470 -
Skogheim, 2019, 5918847-

¦M-

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)

++ 44

Skogheim, 2021, 9959649-
Vuong, 2016, 3352166-

Wang, 2015, 3860120-
Yao, 2022, 10273386-

| ++ ++

Figure 3-76, Study evaluation results for epidemiology studies of PFDA and
neurodevelopmental effects. Refer to HAWC for details on the study evaluation
review: https://hawc.epa.gov/summary/visual/assessment/100500072/pfda-and-
neurodevelopmental-outcomes/.ac

^Multiple publications of the same study population:

Project Viva - Harris et al. (2018) also includes (Harris et al.. 2021)

HOME study - Vuong et al. (2016) also includes (Vuong et al., 2018)

bFour publications with data from the Danish National Birth Cohort were evaluated separately due to significantly
different selection procedures but should not be considered independent: (Liew et al., 2014); (Liew et al., 2015):
(Liew et al., 2018): (Luo et al., 2020).

Two publications with data from the Norwegian Mother Father and Child Cohort were evaluated separately due to
significantly different selection procedures but should not be considered independent: (Skogheim et al., 2020)
and (Skogheim et al., 2021).

This document is a draft for review purposes only and does not constitute Agency policy,

3-241 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Most of the eight studies (reported in nine publications) examining ADHD or related
behaviors reported associations with greater difficulties in attention or behavior problems, but
there is some inconsistency within and across studies and imprecision in the results. Results for the
medium confidence studies are displayed in Table 3-39. Notably, the two studies with the most
clinically relevant outcome measure fSkogheim etal.. 2021: Liew etal.. 20151 examined diagnosed
ADHD and found no increase in the odds of diagnosis (effect estimates were in the inverse
direction). The remaining medium confidence studies (including another publication using the same
population as Liew etal. (2015). resulting in six studies) examined scores on neurobehavioral
assessments including the Strengths and Difficulties Questionnaire (SDQ), the Child Behavior
Checklist (CBC), and the Behavior Rating Inventory of Executive Function (BRIEF). With the
exception of Luo etal. f20201. which reported inconsistent results across child ages, all of these
studies reported associations consistent with greater difficulties in attention or behavior problems
with higher PFDA exposure, though effect estimates were small in most studies. This included
statistically significant associations in Harris etal. (2021) and Oulhote etal. (2016) with SDQ scores
and an exposure-response gradient across categories in Harris etal. (2021) and H0ver etal. (2017).
However, in most studies, the confidence intervals were wide. It is possible that the limited study
sensitivity could explain the non-significant findings, but this would likely not explain the
inconsistency with studies of the more apical outcome of ADHD diagnosis, and thus there is
uncertainty in the findings overall. Finally, a low confidence cross-sectional study examined inter-
response time (IRT) at age 9-11 and found statistically significant decreases in IRT, which indicates
poor response inhibition (a primary deficit in children with ADHD) as the test is designed to reward
longer response times (Gump etal.. 2011).

For the other neurodevelopmental outcomes, results were less consistent. In the eight
studies of cognition and summary neurodevelopmental scores, Vuongetal. f20181. reported higher
odds of "at risk" scores for metacognition and global executive indices at ages 3 and 8 (statistically
significant for the global executive composite, OR 2.95, 95% CI: 1.20, 7.23). Nonstatistically
significant decreases in IQ or similar scores were reported in two studies (Harris etal.. 2018: Wang
etal.. 2015). but the remaining studies did not report associations with IQ (Liew etal.. 2018).
executive function (Harris etal.. 2021). communication and problem solving (Niu etal.. 2019).
working memory (Skogheim etal.. 2020). adaptive or language developmental quotient (Yao etal..
20221. or intellectual disability fLvall etal.. 20181. Among the five studies of ASD and social
behavior, four examined diagnosed ASD; three of these reported inverse associations (statistically
significant in one) fSkogheim etal.. 2021: Shin etal.. 2020: Liew etal.. 20151 and one reported a
null finding (Lvall etal.. 2018). One study examined personal-social skills and found a positive
association with problems which was statistically significant in girls (Niu etal.. 2019). Two of three
studies of motor effects reported non-statistically significant associations with reduced motor
performance fNiu etal.. 2019: Harris etal.. 20181. Lastly, one study of congenital cerebral palsy

This document is a draft for review purposes only and does not constitute Agency policy.

3-242 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 found no association with PFDA exposure fLiewetal.. 20141. Due to the poor sensitivity of the

2 available studies, it is difficult to interpret the primarily null results for these outcomes.

Table 3-38. Results for medium confidence epidemiology studies of PFDA
exposure and behavioral and attention effects

Study name,
reference(s)

Measured
Outcome

Exposure
measurement
timing

Estimate type
(adverse
direction®)

Sub-
population
/N

Group or
unit change

Exposure
Median (IQR)
or range
(quartiles)

Effect
Estimate

CI LCL

CI UCL

Norwegian
Mother,
Father, and
Child cohort

Diagnosed
ADHD

Maternal (2nd
trimester)

OR (1s)

1,801

0.02-0.13

Ref

0.13-0.17

0.86

0.65

1.13

Skogheim et

0.17-0.23

0.77

0.59

1.02

al. (2021)

0.23-1.5

0.61

0.46

0.81

Danish

National Birth
Cohort

Diagnosed
ADHD

Maternal (1st
trimester)

RR (1s)

760

Ln-unit
increase in
exposure

0.2 (0.1-0.2)

0.76*

0.64

0.91

Liew et al.
(2015)

Externalizing
problems at
7 yrs

OR (1s) (odds
of elevated
score)

2,421

Per doubling
of exposure

1.09

0.78

1.53

Luo et al.
(2020)

Internalizing
problems at
7 yrs

1.03

0.72

1.47

Total SDQ
score at 7 yrs

1.11

0.87

1.43

Externalizing
problems at
11 yrs

2,070

0.95

0.70

1.28

Internalizing
problems at
11 yrs

0.95

0.72

1.26

Total SDQ
score at 11
yrs

0.86

0.68

1.08

Odense child
cohort

ADHD
symptom
score on CBC

18 mo

IRR (1s)
(relative
difference in
score)

775

Per doubling
of exposure

0.2

0.98

0.88

1.09

Dalsager et

Maternal (1st
trimester)

1,113

0.3

1.02

0.95

1.09

al. (2021b)

18 mo

OR (1s) (odds
of elevated

775

Per doubling
of exposure

0.2

1.06

0.78

1.44

Maternal (1st
trimester)

score)

1,113

0.3

1.08

0.85

1.37

HOME study

Behavioral
regulation
index on
BRIEF

3 yrs

OR (1s) (odds
of elevated

208

Ln-unit
increase in

0.2

1.95

0.83

4.62

Vuong et al.
(2018)

8 yrs

score)

exposure

0.2

1.70

0.59

4.88

Project Viva

Externalizing
problems

7-11 yrs

Mean

Difference (1s)

628

<0.1-0.2

Ref

0.3-0.3

0.2

-0.5

0.9

This document is a draft for review purposes only and does not constitute Agency policy.

3-243 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study name,
reference(s)

Measured
Outcome

Exposure
measurement
timing

Estimate type
(adverse
direction®)

Sub-
population
/N

Group or
unit change

Exposure
Median (IQR)
or range
(quartiles)

Effect
Estimate

CI LCL

CI UCL

Harris et al.
(2021)

0.4-0.4

0.3

-0.4

1.0

0.5-1.9

0.5

-0.2

1.2

Internalizing
problems

<0.1-0.2

Ref

0.3-0.3

0.2

-0.4

0.7

0.4-0.4

0.4

-0.2

0.9

0.5-1.9

0.6

0.0

1.1

Total SDQ
score

<0.1-0.2

Ref

0.3-0.3

0.4

-0.6

1.3

0.4-0.4

0.7

-0.4

1.7

0.5-1.9

1.1*

0.1

2.1

Faroe Island
cohort

(Oulhote et
al.. 2016)

Externalizing
problems

5 yrs

Mean

Difference (1s)

508

Per doubling
of exposure

0.3 (0.2-0.4)

0.45*

0.02

0.87

Maternal (32
wks gestation)

539

0.3 (0.2-0.4)

0.26

-0.29

0.81

Internalizing
problems

5 yrs

Mean

Difference (1s)

508

Per doubling
of exposure

0.3 (0.2-0.4)

0.27

-0.11

0.65

Maternal (32
wks gestation)

539

0.3 (0.2-0.4)

0.26

-0.29

0.81

Total SDQ
score

5 yrs

Mean

Difference (1s)

508

Per doubling
of exposure

0.3 (0.2-0.4)

0.72*

0.07

1.38

Maternal (32
wks gestation)

539

0.3 (0.2-0.4)

-0.01

-0.98

0.96

INUENDO
(Biopersistent
organochlorine
s in diet and
human fertility)

(Hover et
al.. 2017)

SDQ

hyperactivity
score at 5-9
yrs

Maternal
(second
trimester
median)

Regression
Coefficient

1,023

In-unit
increase in
exposure

1.5 (10th-90th
0.7-3.4)

0.13

-0.10

0.36

Low

exposure

0.2-1.2

Ref

Medium
exposure

1.2-2.0

0.11

-0.22

0.44

High

exposure

2.0-18.8

0.13

-0.27

0.53

Total SDQ
score at 5-9
yrs

Maternal
(second
trimester
median)

Regression
Coefficient

1,023

In-unit
increase in
exposure

1.5 (10th-90th
0.7-3.4)

0.40

-0.15

0.95

Low

exposure

0.2-1.2

Ref

Medium
exposure

1.2-2.0

0.07

-0.71

0.85

High

exposure

2.0-18.8

0.65

-0.30

1.61

*p <0.05

This document is a draft for review purposes only and does not constitute Agency policy.

3-244 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

SDQ: Strengths and Difficulties Questionnaire. Externalizing problems calculated from conduct and hyperactivity
subscales; Internalizing problems calculate from emotional and peer subscales.

BRIEF: Behavior Rating Inventory of Executive Function

a The arrows indicate the direction the effect estimate will be if there is an association between PFHxS and reduced
behavior. For all the tests included here, higher scores indicate more difficulties/behavior problems/ADHD
diagnosis. For ratio measures such as odds ratios (OR), an effect estimates greater than 1 indicates more
difficulties/behavior problems, while for regression coefficients and mean differences, an effect estimates greater
than 0 indicates more difficulties/behavior problems

Animal studies

There are no available animal toxicity studies informing of potential neurodevelopmental
effects of PFDA via any relevant exposure route and duration.

Evidence Integration

The evidence for potential neurodevelopmental effects in humans is considered slight.
Associations between PFDA exposure and outcomes related to attention and behavior were
reported in multiple epidemiological studies, though there was inconsistency between these
findings and the more clinically relevant measure of ADHD diagnosis. Results for other
neurodevelopmental effects were largely inconsistent, though poor sensitivity due to limited
exposure contrast may explain the lack of association in some studies. No animal toxicity studies
are available. Altogether, based on the available human studies, the evidence suggests that PFDA
exposure might cause neurodevelopmental effects in humans under sufficient exposure conditions4
(see Table 3-40).

This document is a draft for review purposes only and does not constitute Agency policy.

3-245 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-39. Evidence profile table for PFDA neurodevelopmental effects

Evidence stream summary and interpretation

Evidence integration
summary judgment

Evidence from studies of exposed humans (see Section 3.2.7: Human studies)

®oo

Evidence suggests

Studies, outcomes, and
confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream judgment

ADHD and related
behaviors

7 medium and 1 low
confidence studies

• 5/6 studies examining
behavioral issues
and/or attention
problems reported
positive associations
but the two studies
examining ADHD
diagnosis (the most
clinically relevant
outcome) reported
inverse findings.

• Consistency in
direction of
association for studies
of behavior and
attention

• Unexplained
inconsistency with
studies of ADHD
diagnosis

• Imprecision in most
study results

®oo

Slight

There is some evidence of
greater problem behaviors
and decreased attention
with increasing PFDA
exposure but there is
remaining uncertainty due
to inconsistency and
imprecision.

Primary basis:

Slight evidence of attention
and behavior effects in
humans.

Human relevance, cross-
stream coherence,
susceptibility, and other
inferences:

Evidence comes from
studies in humans at a
susceptible lifestage (in
utero or childhood
exposure).

Other

neurodevelopmental
effects

14 medium confidence
studies

• Some studies reported
decreases in cognition
or motor scores, but
findings were
inconsistent across
studies. No association
was observed with
ASD/social behavior or
cerebral palsy.

• No factors noted

• Unexplained
inconsistency

This document is a draft for review purposes only and does not constitute Agency policy.

3-246 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.8. ENDOCRINE EFFECTS
Human studies

Thyroid effects

Twenty-three studies examined thyroid hormones and PFDA exposure. A summary of the
study evaluations is presented in Figure 3-77, and additional details can be obtained from HAWC.
Two studies were considered uninformative and excluded from further analysis due to critical
deficiencies in confounding and analysis (Seo etal.. 20181 or serious deficiencies in several domains
fKim etal.. 20111. Sixteen studies were classified as medium confidence and five studies were
classified as low confidence fLiu etal.. 2021b: Itoh etal.. 2019: Zhang etal.. 2018a: Ti etal.. 2012:
Bloom etal.. 20101. Of the medium confidence studies, five were cross-sectional, nine were
prospective cohorts, one was a retrospective cohort, and one was participants from a randomized
clinical trial of energy-reduced diets (functionally equivalent to a prospective cohort).

In addition to the general considerations described in Section 1.2.2, there were several
outcome-specific considerations for study evaluation that were influential on the ratings. First, for
outcome ascertainment, collection of blood during a fasting state and at the same time of day for all
participants (or adjustment for time of collection) is preferred for measurement of thyroid
hormones to avoid misclassification due to diurnal variation fvan Kerkhof etal.. 20151. Studies that
did not consider these factors (e.g., by study design or adjustment) were not excluded but were
considered deficient for the outcome ascertainment domain. For participant selection, it was
considered important to account for current thyroid disease and/or use of thyroid medications;
studies that did not consider these factors by exclusion or another method were considered
deficient for the participant selection domain. Concurrent measurement of exposure with the
outcome was considered appropriate for this outcome since circulating hormone levels can change
quickly in response to a change in exposure and the half-life of PFDA in humans is long. All the
available studies analyzed PFDA in serum or plasma using appropriate methods (as described in
the protocol). Thyroid hormones were analyzed using standard and well-accepted methods in all
studies. Overall, while most studies were considered medium confidence, nearly all of them had
limitations in outcome ascertainment and/or study sensitivity (primarily due to low PFDA
exposure levels in the study populations). These ascertainment issues and other (non-differential)
sources of measurement error are likely to bias the results towards the null, and thus null
associations are difficult to interpret

This document is a draft for review purposes only and does not constitute Agency policy.

3-247 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Atnrruzi R el al. 2020
Aimuzi, 2019. 5387078'
Berg V. 2016. 3350759'
Blate, 2018. 5030357
Bloorn, 2010. 757875
Cakmak. 2022, 10273369
Goo J et al. 2021
Irroue. 2019, 5918599
Itoh. 2019. 5915990
Ji, 2012. 2919189
Kang, 2018. 4937567
Kim HY et al. 2020'
Kim, 2011. 1424975'
Liang H et al. 2020 ¦
Liu M et al. 2021
Liu. 2018. 1937240
Reardon, 2019. 5412435
Seo. 2018. 4238334
Shah-Kulkarra, 2016. 3S59S21
Vfeng. 2013. 4241230
Vteng, 2014. 2850394
Yang, 2016. 3858535
Zhang. 2018. 5079365'

Legend

Good (metric) or H-.gh confidence (overall)
+ Adequate (metric) cr Medium confidence (overall}
Deficient (metr e) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overal )

Figure 3-77. Study evaluation results for epidemiology studies of PFDA and
thyroid effects. Refer to HAWC for details on the study evaluation review: HAWC
Human Thyroid Effects

Multiple publications of single study: Berg et al. (2017) includes Berg et al. (2015). Aimuzi et al. (2019) and Aimuzi
et al. (2020) examine the same birth cohort but are considered separately because the populations are different
(neonates/cord blood in Aimuzi et al. (2019) and pregnant women in Aimuzi et al. (2020). These studies should
not be considered fully independent.

This document is a draft for review purposes only and does not constitute Agency policy,

3-248 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Twelve studies examined associations with thyroid hormones in adults (11 for T4, 8 for T3,
and 11 for TSH), including those focused on pregnant women. Results were mixed across studies of
the same hormone, with no clear pattern to explain the inconsistency (e.g., study confidence or
population characteristics). For T4 (both free and total), two medium confidence studies flnoue et
al.. 2019: Blake etal.. 20181 and one low confidence study fLiu etal.. 2021bl in general population
adults and one medium confidence study in pregnant women during early gestation fAimuzi etal..
2020) reported small positive associations (higher levels with higher exposure), but these
differences were imprecise (wide confidence intervals andp >0.05) and in Inoue etal. (2019).
nonmonotonic across quartiles of exposure (Inoue etal. (2019): % difference (95% CI): Q2: 1.0
(-2.3, 4.3), Q3: 1.5 (-2.3, 5.4), Q4: -0.5 (-3.6, 2.7); Blake etal. C20181: 2.5% change,

95% CI:= -2.9, 8.3); Aimuzi etal. f20201: (3 (95% CI): 0.05 (-0.03, 0.13)). One low confidence study
fZhang et al.. 2 018al in women with premature ovarian insufficiency (POI) reported non-
statistically significant lower levels of free T4 with higher exposure

((3 (95% CI) = -1.19 (-2.66, 0.28)). The other seven available studies (six medium confidence) did
not report a positive or negative association (Cakmaketal.. 2022: Itoh etal.. 2019: Reardonetal..
2019: Liu etal.. 2018: Yang etal.. 2016a: Berg etal.. 2015: Wang etal.. 2014a). For T3, two medium
confidence studies in pregnant women reported positive associations fAimuzi et al.. 2 02 0: Wang et
al.. 2014al. with statistical significance in one fWang etal.. 2014al. (3 (95% CI) = 0.002 (0, 0.003);
Aimuzi etal. f20201: (3 (95% CI): 0.05 (-0.03, 0.13)). In contrast, three studies, one medium and two
low confidence reported inverse associations (lower T3 with higher PFDA exposure) (Berg et al.
(2015). mean differences vs. the Q1 (95% CI) referent: Q2: -0.04 (-0.08, 0.04), Q3: -0.05 (-0.08, 0),
Q4: -0.10 (-0.14, -0.06); Liu etal. (2021c): % change (95% CI) per ln-unitincrease in PFDA: -3.79
(-7.69, 0.27); and Zhang et al. f2018al in women with POI. (3 (95% CI) = -0.56 (-1.27, 0.16)). Effect
sizes in both directions were close to null. The remaining studies reported no association fltoh et
al.. 2019: Reardon etal.. 2019: Liu etal.. 2018: Yang etal.. 2016al. Of the 11 studies reporting on
TSH, two medium and two low confidence studies reported higher TSH with higher exposure flnoue
etal. f20191: % difference (95% CI): Q2: -2.7 (-21.4, 20.6), Q3: 0.4 (-21.5, 28.5), Q4: 3.6 (-16.7,
28.8):Blake etal. (2018): 11% change, 95% CI: -4.5, 28.8 and Zhang et al. (2018a):

(3 (95% CI) = 0.85 (-0.03,1.72); Liu etal. (2021b): % change (95% CI) per ln-unit increase in PFDA:
9.53 (-6.15, 27.92)), but these estimates were imprecise (wide confidence intervals) and, in Inoue et
al. f20191. were non-monotonic across quartiles of exposure. The results in Blake etal. f 20181.
which was the only study with repeated measures of TSH, were not robust as it changed to an
inverse association when only the first exposure measurements were included in the model, rather
than repeated measures. In addition, one medium confidence study reported a non-significant
inverse association (Cakmaketal. (2022): % change (95% CI) for a one mean increase in PFDA: -7.0
(-17.2, 4.4)). The remaining studies reported no association between TSH and PFDA exposure
fAimuzi etal.. 2020: Itoh etal.. 2019: Reardon etal.. 2019: Yang etal.. 2016a: Berg etal.. 2015:

Wang etal.. 2014a: Wang etal.. 20131. In addition, two medium confidence studies fKim etal..

This document is a draft for review purposes only and does not constitute Agency policy.

3-249 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

2020a: Kang etal.. 20181 examined associations in children and adolescents and reported no
association with free T4 or TSH (both studies) or T3 (Kim etal.. 2020a).

Seven medium confidence studies and one low confidence study examined associations with
thyroid hormones in neonates. For T4 (total or free), there were seven studies available, and only
one reported an association; Liang etal. f20201 reported an inverse association with T4 but not
total T4 (P (9% CI for ln-unit increase in PFDA: -5.07 (-9,78, -0.37)). The remaining studies reported
no (Guo etal.. 2021: Aimuzi etal.. 2019: Itoh etal.. 2019: Shah-Kulkarni etal.. 2016: Yang etal..
2016a: Wang et al.. 2014a). For total T3, two out of six medium confidence studies reported higher
T3 with higher PFDA Shah-Kulkarni etal. (2016): (3 (95% CI) for ln-unit increase in PFDA: 2.4 (-
0.27, 5.09), stronger association in girls ((3 = 3.93 vs 1.02); Liang etal. f20201: 0.06 (0.03, 0.09)). In
contrast, one study reported lower T3 (p < 0.05) in boys with maternal thyroid antibody negative
but higher T3 in boys with maternal thyroid antibody positive (p > 0.05) and girls fltoh etal.. 20191.
The remaining three studies reported no association (Guo etal.. 2021: Aimuzi etal.. 2019: Yang et
al.. 2016a: Wang etal.. 2014a). For TSH, eight studies were available. Three reported inverse
associations between TSH and PFDA exposure, but in Itoh etal. (2019). this was observed only in
boys with maternal thyroid antibody positive, while in Shah-Kulkarni etal. (2016). the association
was observed only in girls and not statistically significant. The association was observed in the
overall population in Wang etal. f2014al. but this was also not statistically significant In addition,
one study reported a positive association with TSH Liu etal. f2021bl. The remaining studies
reported no association (Guo etal.. 2021: Aimuzi etal.. 2019: Yang etal.. 2016a: Berg etal.. 2015).
It is possible that the lack of consistency was due to differences in the timing of exposure
measurement (maternal sampling at median 18 weeks in Berg etal. (2017). second trimester in
Itoh etal. (2019). third trimester in Wang etal. (2014a). and 1-2 days before delivery in Yang et al.
f2016al. and cord blood sampling in Shah-Kulkarni etal. f20161. Aimuzi etal. f20191. Guo et al.
f20211: Liu etal. f2021bl. but this is not possible to evaluate further due to the lack of multiple
studies per sampling period other than cord blood.

Overall, the evidence for the association between PFDA exposure and thyroid effects in
human studies is inconsistent. A few studies do suggest an association between thyroid hormones
and PFDA exposure, but other studies are null, and the direction of association is not consistent
across studies. Even in the studies that observed associations, there is not clear coherence across
outcomes, where one would expect a decrease in T4 and T3 to correspond with an increase in TSH,
or vice versa, though this could be explained by secondary hypothyroidism as discussed below. For
most studies, the exposure levels were low (median exposure was less than 0.5 ng/mL) and there
were narrow exposure contrasts, which along with potential for outcome misclassification in most
studies, reduced the study sensitivity and could have impaired the ability of these studies to
observe a true effect. However, this poor sensitivity would not explain the observed differences in
the direction of association, and thus considerable uncertainty remains.

This document is a draft for review purposes only and does not constitute Agency policy.

3-250 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Animal studies

Two studies in the database of toxicity studies for PFDA evaluated endocrine effects. One
study exposed female Sprague-Dawley rats for 28 days (0, 0.125, 0.25, and 0.5 mg/kg-day) and
examined the adrenal glands (weight and histopathology) fFrawlev et al.. 20181. The second study
examined the following endpoints in both male and female Sprague-Dawley rats after a 28-day
gavage exposure (0, 0.156, 0.312, 0.625,1.25, and 2.5 mg/kg-day): thyroid hormone levels,
histopathology, and organ weights ("INTP. 20181.

Thyroid hormones levels

Aft-

Reporting quality-
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity -
Results presentation -
Overall confidence -

Legend

| Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)

NR Not reported

Figure 3-78. Thyroid hormone levels animal study evaluation heatmap. Refer
to HAWC for details on the study evaluation review.

In the NTP ( 20181 study which was considered high confidence (see Figure 3-79), thyroid
hormones were measured in male and female rats exposed to 0-2.5 mg/kg-day for 28 days (see
Figure 3-79 and Table 3-41). For thyroid-stimulating hormone (TSH), a statistically significant
decreasing trend (18 to 55%) was observed in male rats, but a significant decrease compared to
controls was not reported at any dose. No statistically significant change for TSH was observed in
the female rats but increases ranged from 3 to 35% with the lowest effect occurring at
0.625 mg/kg-day. A statistically significant increasing trend was reported for T3 in male (22 to
88%) and female rats with significant increases (24-109%) reported at >1.25 mg/kg-day for
females only. A statistically significant decreasing trend in free thyroxine (fT4) was reported in

This document is a draft for review purposes only and does not constitute Agency policy,

3-251 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

male and female rats with significant decreases at >0.312 mg/kg-day in males (42-82%) and at
>1.25 mg/kg-day in females (39-74%). A statistically significant decrease in total thyroxine (tT4)
was observed in males only at 0.312 mg/kg-day and was unchanged in females at all doses. fT4 is
the preferred measurement over tT4 in adult animals given that the level of tT4 can be dependent
on the amount of serum binding proteins while fT4 is available to be utilized by the body. The
effects of PFDA on fT4 and TSH in male and female rats are consistent with secondary
hypothyroidism, which is characterized by decreased T4 and decreased or normal levels of TSH
(Lewinski and Stasiak. 2017). However, there is uncertainty in this conclusion given that changes
in fT4 and T3 are often expected to occur in the same direction, with T3 being the more active
hormone form and formation of T3 contingent upon the deiodination of fT4. The potential
mechanism and interpretation for an observation of decreasing fT4 with increasing T3 is unknown
and unexamined in the PFDA evidence base.

This document is a draft for review purposes only and does not constitute Agency policy.

3-252 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-40. Percent changes relative to controls in thyroid hormone levels in
a 28-day rat study after PFDA exposure fNTP. 20181

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Thyroid-stimulating hormone (TSH)

Female S-D rats

Male S-D rats

-18

-22

-41

-55

Triiodothyronine (T3)

Female S-D rats

-4

109

Male S-D rats

-24

-31

-22

Free thyroxine (fT4)

Female S-D rats

-39

-74

Male S-D rats

-6

-42

-44

-68

-82

Total thyroxine (tT4)

Female S-D rats

-9

Male S-D rats

-2

-26

-12

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study

authors.

This document is a draft for review purposes only and does not constitute Agency policy.

3-253 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name Outcome Confidence

Animal Description

PFDA Endocrine Hormones

Thyroid Stimulating Mormon® (TSH) NTP. 2018, 4309127 High confidence

Thyroxine (T4), Free

Thyroxine (T4), Total

Triiodothyronine (T3)

NTP. 2018. 4309127 High confidence

NTP, 2018,4309127 High confidence

NTP. 2018,4309127 High confidence

Thyroid Stimulating Hormone (TSH) NTP, 2018.4309127 High confidence

Thyroxine (T4). Free

Thyroxine (T4). Total

Triiodothyronine (T3)

NTP. 2018, 4309127 High confidence

NTP. 2018.4309127 High confidence

NTP, 2018,4309127 High confidence

28 Day Oral Rat, Sprague-Dawley (Harlan) (<¦') significant

28 Day Oral Rat Sprague-Dawley (Harlan) ( f) significant

28 Day Oral Rat Spraguo-Dawley (Harlan) (. ') not significant ug/dL

28 Day Oral Rat, Sprague-Dawley (Harlan) ( -) significant

28 Day Oral Rat, Sprague-Dawley (Harlan) () not significant

28 Day Oral Rat Spraguo-Dawley (Harlan) (i) significant

28 Day Oral Rat. Sprague-Dawley (Harlan) (i) not significant ug/dL

28 Day Oral Rat Sprague-Dawley (Harlan) (v) significant

0.156
0.312
0.625
1.25

0.156

0.312

0.625

1.25

2.5

0.156

0.312

0.625

1.25

2.5

0.156

0.312

0.625

1.25

2.5

0.156
0.312
0.625
1.25

0.156

0.312

0.625

1.25

2.5

0.156

0.312

0.625

1.25

2.5

0.156
0.312
0.625
1.25
2.5

% Stalistically significant
Percent control response
| petcenl control low

1 1 r~

-100-80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 200
Percent Control Response

Figure 3-79. PFDA thyroid hormone levels after short-term oral exposure (results can be viewed by clicking the
HAWC link: https://hawcprd.epa.gOv/summary/data-pivot/assessment/100000026/pfda-endocrine-hormones/l.

This document is a draft for review purposes only and does not constitute Agency policy,
3-254 DRAFT—DO NOT CITE OR QUOTE "

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Histopathology

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 3-80. Endocrine histopathology animal study evaluation heatmap.

Refer to HAWC for details on the study evaluation review.

Both the NTP (20181 and Frawlev etal. (20181 studies performed histopathological
examinations to examine PFDA-related effects. The NTP (20181 study was considered high
confidence while the Frawlev etal. f 2 018 study was evaluated as medium confidence due to
incomplete reporting of the null data (see Figure 3-80). NTP (20181 performed histopathological
examination of the thyroid gland, adrenal cortex and medulla, parathyroid gland, and pituitary
gland in both male and female rats fNTP. 20181. Histopathology was examined for the thyroid
gland at all doses; all other endocrine tissues were examined only in the control and high-dose
(2.5 mg/kg-day) groups. NTP (20181 reported that there were no tissue changes observed in any of
the examined organs in either sex (see Figure 3-81). Results from the histopathological
examination of the adrenal glands in female rats from the Frawlev etal. (20181 were qualitatively
reported as being unchanged by PFDA exposure (Frawlev etal. 20181.

This document is a draft for review purposes only and does not constitute Agency policy,

3-255 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Study Name

Outcome Confidence

Exposure Design

Species, Strain (Sex)

Trend Test Result

Adrenal Gland Histopathology

Frawley, 2018, 4287119

Medium confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (9)

not reported

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ( ')

not significant

Rat. Sprague-Dawley (Harlan) ( )

not significant

Parathyroid Gland Histopathology

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (S)

not significant

Rat, Sprague-Dawley (Harlan) (2)

not significant

Pituitary Gland Histopathology

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (5)

not significant

Rat, Sprague-Dawley (Harlan) (2)

not significant

Thyroid Gland Histopathology

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (:')

not significant

Rat, Sprague-Dawley (Harlan) (y)

not significant

PFDA Endocrine Effects

0 No significant change Significant increase ^7 Significant decrease

(mg/kg-d)

Figure 3-81. PFDA endocrine histopathology (results can be viewed by clicking the HAWC link: details:

https://hawcprd.epa.gOv/summary/data-pivot/assessnient/100500072/PFDA-Endocrine-Histopath-Animal/l.

This document is a draft for review purposes only and does not constitute Agency policy,

3-256 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Organ weight

Reporting quality -

Allocation

Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity -
Results presentation -
Overall confidence \

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NRj Not reported

Figure 3-82. PFDA endocrine organ weights animal study evaluation heatmap.

Refer to HAWC for details on the study evaluation review.

Both the NTP (2018; and Frawlev etal. (20181 studies evaluated PFDA effects on endocrine
organ weights and were considered high confidence for this outcome (see Figure 3-82). As
indicated above, both studies measured adrenal weights. Only the NTP (20181 study measured
thyroid weight; both sexes in rats demonstrated a statistically significant trend in relative thyroid
weight with statistically significant increases reported at >1.25 mg/kg-day in male rats (43% at
both 1.25 and 2.5 mg/kg-day) and at >0.312 mg/kg-day in female rats (27-45%), For absolute
thyroid weight in male rats, there was no significant trend, and no significant change was observed
at any dose tested. In female rats, there was no significant trend but significant increases (33-34%)
were observed at doses ranging from 0.312 to 1.25 mg/kg-day but not at the highest dose tested
(2.5 mg/kg-day). Relative (to body weight) thyroid weight is the preferred measure for this organ
particularly in the presence of body weight changes (Bailey et al.. 20041. Significant reductions in
body weight were observed in the NTP (20181 study at the two highest doses tested
(>1.25 mg/kg-day; refer to the Section 3.2.10 on General toxicity effects for additional details). For
adrenal weight in female rats, no significant changes were observed in rats at doses up to
1.25 mg/kg-day in both studies. A statistically significant decrease (36%) for absolute adrenal
gland weight was observed at the highest dose group (2.5 mg/kg-day) in female rats from the NTP

This document is a draft for review purposes only and does not constitute Agency policy,

3-257 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 (20181 study; no change was reported for relative adrenal weight in females in this study. A

2 statistically significant decrease (15-21%) was reported in absolute adrenal gland weight in male

3 rats at all dose groups. Conversely, relative adrenal weight in males was significantly increased

4 (50%) at the highest dose tested fNTP. 20181. The toxicological significance of the adrenal organ-

5 weight changes is unclear; the opposing direction of absolute and relative organ-weight changes

6 suggests a confounding effect of body-weight changes (refer to the General toxicity section for more

7 detail on body-weight effects) at the same doses. Furthermore, No PFDA-induced histopathological

8 changes on the adrenal gland were observed (see discussion above).

This document is a draft for review purposes only and does not constitute Agency policy.

3-258 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Adrenal Gland Weight, Absolute

Study Name Outcome Confidence Exposure Design

NTP, 2018,4309127 High confidence 28 Day Oral

Adrenal Gland Weight, Absolute (Histopathology Cohort) Frawley, 2018, 4287119 High confidence

Adrenal Gland Weight, Relative NTP, 2018, 4309127 High confidence

Adrenal Gland Weight, Relative (Histopathology Cohort) Frawley, 2018, 4287119 High confidence

Thyroid Weight, Absolute NTP, 2018,4309127 High confidence

Thyroid Weight, Relative NTP, 2018,4309127 High confidence

I # No significant change Significant increase ^ Significant decrease I

28 Day Oral
28 Day Oral

28 Day Oral

Species. Strain (Sex)

Rat, Sprague-Dawley (Harlan) (:
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (:
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (;
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (
Rat, Sprague-Dawley (Harlan) (

Trend Test Result

significant
significant
not significant
significant
not significant
not significant
not significant
not significant
significant
significant

0.01 0.1 1 10
mg/kg-d

PFDA Endocrine Organ Weight

~ ~ ~ ~ ~

•—•—•

•——•—•—A

•—•-—•
•—•—•—•—•

•—A—A—A—•

Figure 3-83. PFDA endocrine organ weight (results can be viewed by clicking the HAWC link:

https://hawcprd.epa.gOv/sumniary/data-pivot/assessment/100500072/PFDA-Endocrine-Organ-Weight/l.

This document is a draft for review purposes only and does not constitute Agency policy,

3-259 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Mechanistic studies and supplemental information

In support for PFDA-induced changes on thyroid hormones observed in rats from the
28-day NTP f20181. structurally related PFAS (e.g., PFNA; PFOA) have been shown to effect thyroid
hormone levels in rodents. Specifically, PFNA induced hypothyroxinemia in rodents.
Hypothyroxinemia has been defined in humans as a low percentile value of serum free T4 (ranging
from the 2.5th percentile to the 10th percentile of free T4), with a TSH level within the normal
reference range (Alexander etal.. 20171.

Additionally, multiple high-dose, intraperitoneal (i.p.) injection studies have demonstrated
that PFDA affects T3 and T4 serum levels. Specifically, decreases in serum T4 have been repeatedly
observed in rats exposed to PFDA via i.p. injection at doses ranging from 20 to 80 mg/kg (Gutshall
etal.. 1989.1988: Van Rafelghem et al.. 1987a: Langlev and Pilcher. 19851. Evaluations of PFDA
effects on T3 varied among i.p. studies in rats. Langlev and Pilcher f!9851 observed an initial
significant decrease in T3 levels starting at 12 hours post PFDA exposure (75 mg/kg i.p.) as
compared to pair-fed controls, which remained significantly decreased until day 4 of the study.
Following day four of the study, there were no significant differences in T3 serum levels between
pair-fed and PFDA-exposed animals, while serum T4 levels remained significantly diminished
through day eight of the study as compared to the pair-fed controls. Gutshall etal. T19891 also
reported significant decreases in T3 at 75 mg/kg i.p. in rats at 12 and 24 hours after PFDA
exposure. Conversely, no changes in T3 were observed in rats exposed via i.p. to PFDA at doses up
to 80 mg/kg-day (Gutshall etal.. 1988: Van Rafelghem et al.. 1987a). However, the inconsistencies
in PFDA effects on T3 levels could be due to differences in experimental design and the time at
which thyroid hormones were measured. In the studies that showed no effect on T3 levels in rats
f Gutshall etal.. 1988: Van Rafelghem etal.. 1987al. thyroid hormones were measured at 7 and
14 days after PFDA treatment compared to the positive studies that showed effects on hormone
levels at 12 to 48 hours after exposure.

Under normal physiologic conditions, neurons in the hypothalamus release thyroid
releasing hormone (TRH) to stimulate thyrotrophs of the anterior pituitary gland to release thyroid
stimulating hormone (TSH). TSH plays a number of important metabolic functions including
stimulation of the thyroid gland to release triiodothyronine (T3) and thyroxine (T4). When
increased T3 and T4 serum levels reach above a certain blood concentration threshold, secretion of
TRH from the hypothalamus is inhibited via a negative feedback loop.

To evaluate whether PFDA altered the ability of the pituitary and thyroid glands of the
PFDA exposed animals to respond to a physiological stimulation, Gutshall etal. (19891 challenged
male Wistar rats with 500 |J.g/kg TRH at 15 or 22 hours post a single, high-dose 75 mg/kg (i.p.)
PFDA exposure and found that although the percent response changes in T4 and T3 compared to
baseline (i.e., pre-TRH challenge) were similar between the control and PFDA exposed animals, the
absolute values for T4 and T3 in the sera from PFDA exposed animals was significantly less than
that of their control counterparts following TRH stimulation f Gutshall etal.. 19891. These data

This document is a draft for review purposes only and does not constitute Agency policy.

3-260 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

indicate that PFDA may alter the ability of the glands in the hypothalamic-pituitary-thyroid axis
(HPT) to respond to physiological stimulation fGutshall etal.. 19891. Additional studies would help
clarify whether this observation is relevant in other species and at lower, more physiologically
relevant levels of PFDA exposure. Impaired responsiveness of the hypothalamic-pituitary-thyroid
axis to hormonal stimulation could explain the results from the 28-day study in rats fNTP. 2018] in
which TSH and T4 were both decreased by PFDA exposure in male rats; this mechanistic
information does not however provide insight on why T3 was increased in the presence of
decreased TSH and T4.

Additionally, the high dose, i.p. study by Gutshall etal. (1989) showed that PFDA is able to
displace T4 from plasma proteins fGutshall etal.. 19891. The fate of the displaced (i.e., free) T4 is
unknown, but the authors postulated increased biliary excretion may be a potential route of T4 loss.
Using a fluorescence displacement assay, Ren etal. f20161 reported that PFDA binds to
transthyretin, a major transport protein for thyroid hormone, with the potential to displace T4 from
the transport protein in occupational exposure settings. It is unclear how these mechanistic data
which indicate that PFDA decreases protein binding of T4, support the PFDA-induced effects on
thyroid hormone homeostasis observed in rats from the NTP (20181 study. A decrease in protein
binding of T4 could result in increased fT4 (unbound form) and a decrease in tT4 (bound form).
Conversely, decreased fT4 was observed in rats while tT4 was decreased only at the mid-dose in
males and unchanged in female rats exposed to PFDA fNTP. 20181. Interestingly, evaluation of
unsaturated binding capacity of thyroid-binding proteins, measured by T3 uptake analysis showed
that T3 uptake was significantly reduced in the 80 mg/kg PFDA exposed animals as compared to
the pair-fed controls (Van Rafelghem et al.. 1987a). Under in vitro conditions, Ren etal. (2015)
reported binding of PFDA to the human thyroid receptor but that PFDA did not exhibit antagonistic
or agonistic effects on the thyroid receptor pathway fRen etal.. 20151.

Kelling etal. f!9871 sought to determine the effects of PFDA on the thyroid by evaluating
the hepatic activities of L-glycerol-3-phosphate dehydrogenase, malic enzyme, and
glucose-6-phosphate dehydrogenase, which are enzymes that are sensitive to thyroid status. The
activity of these enzymes is increased during hyperthyroidism and decreased during
hypothyroidism (Mariash et al.. 1980). Similar to the study performed by Langley and Pilcher, SD
male rats received a single, high dose i.p. injection of either 20, 40, or 80 mg/kg PFDA and then
hepatic subcellular fractions were prepared following euthanasia. These hepatic fractions were
then used to assay the activity of L-glycerol-3-phosphate dehydrogenase, lactate dehydrogenase,
malic enzyme, and glucose-6-phosphate dehydrogenase. PFDA significantly increased the activity
of L-glycerol-3-phosphate dehydrogenase, cytosolic lactate dehydrogenase and cytosolic malic
enzyme as compared to their pair-fed and ad libitum controls indicating that the increase of
enzyme activity is a direct result of PFDA exposure and not a secondary effect caused by decreased
food intake and subsequent loss in body weight fKelling etal.. 19871. Similar effects of PFDA on
L-glycerol-3-phosphate dehydrogenase and cytosolic malic enzyme in rats were also reported by

This document is a draft for review purposes only and does not constitute Agency policy.

3-261 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Gutshall etal. (19891. There was no significant difference in glucose-6-phosphate dehydrogenase
activity, hepatic DNA content or protein content. These data indicate that while evidence such as
decreased serum T4 in rats exposed to PFDA is suggestive of a lessened thyroid state, the activation
of thyroid sensitive enzymes is increased in rats exposed to PFDA.

Overall, there is uncertainty in the relevance of the mechanistic studies and supplemental
information to the thyroid effects observed in rats from the NTP f20181 study. Specifically, the
doses from these studies (20-80 mg/kg) are much higher than those used in the 28-day study
(0.156-2.5 mg/kg-day) (NTP. 20181 and have been shown to cause overt systemic toxicity
including a "wasting syndrome" (refer to the General toxicity section), which could confound the
interpretation of the mechanistic data. Additionally, the mechanistic studies and supplemental
information are shorter duration in which rats were exposed to PFDA via i.p. injection rather than
gavage as was done in the NTP f20181 study. Furthermore, a data gap exists because there are no
mechanistic studies available that determined the effect of PFDA on the activities of deiodinases,
which convert free T4 to T3. Data on how PFDA might affect deiodinase activity could inform the
mechanism by which PFDA was observed to decrease fT4 while increasing T3 in rats from the NTP
(20181 study.

Evidence Integration

There is indeterminate evidence of an association between PFDA exposure and endocrine
related effects in studies of exposed humans. The evidence is largely null, but there are concerns
for study sensitivity. The observed associations are inconsistent across studies and not coherent
across thyroid hormones.

There is indeterminate animal evidence of endocrine toxicity- specifically, thyroid effects,
with PFDA based on incoherent evidence from a single high confidence short term study in rats (a
second short term study examined adrenal effects). PFDA was shown to cause changes in thyroid
hormone levels, some of which may be interpreted as suggestive of secondary hypothyroidism, a
phenotype characterized by decreased T4 and decreased or normal levels of TSH (Lewinski and
Stasiak. 20171: however, the PFDA data are not entirely coherent with such a hypothesis.
Specifically, in the NTP (20181 study, significant trends were reported for decreased TSH and fT4
(but nottT4) in male rats at >0.312 mg/kg-day, while significant trends were also reported for
increased T3 (the latter findings are not coherent with hypothyroidism). Likewise, in females,
increased T3 and decreased fT4 was observed at >1.25 mg/kg-day. High dose PFDA exposure-
induced decreases in total T4 were consistently observed in multiple, high dose i.p. studies in rats.
The cause of secondary hypothyroidism is thought to be due to impaired responsiveness of the
hypothalamus-pituitary-thyroid axis (Lewinski and Stasiak. 20171. Consistent with this, PFDA was
shown to impair the response of the hypothalamic-pituitary-thyroid axis to TRH stimulation in rats
from a high dose i.p. study (Gutshall etal.. 19891. These data provide mechanistic insight and
biological plausibility for how PFDA could be decreasing serum levels of T4. Furthermore, there
was coherence with increased relative thyroid weight and decreased fT4 serum levels at

This document is a draft for review purposes only and does not constitute Agency policy.

3-262 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

>1.25 mg/kg-day in male and female rats. A previous study observed increased relative thyroid
weight in a rat model of methimazole-induced hypothyroidism (Soukup etal.. 20011. Also, an
enlarged thyroid is a symptom of hypothyroidism in humans flOEHC. 20141. In support for PFDA-
induced changes on thyroid hormone homeostasis, structurally related PFAS compounds (e.g.,
PFNA; PFOA) have been shown to effect thyroid hormone levels in rodents. However, several
aspects of the available animal data decrease the strength or certainty of the evidence informing
thyroid effects, which was only available from a single oral exposure study. Whereas the NTP
(20181 study reported changes in fT4 and TSH in rats that may indicate secondary hypothyroidism,
there was an increase in T3 that cannot be explained. Furthermore, there are no mechanistic
studies that determined the effect of PFDA on deiodinase activity that could offer insight on how
PFDA decreased fT4 and TSH while increasing T3. Additionally, while fT4 was decreased in male
and female rats from the NTP T20181 study, a consistent decrease in tT4 was not observed.

However as noted above, fT4 not tT4 is the preferred measure in adult animals. Whereas there was
potential coherence between decreased fT4 and increased thyroid weight in rats, it is unclear how
thyroid weight and T3 were increased in the absence of increased TSH or histopathological
changes.

Uncertainty is also associated with the mechanistic studies and supplemental information.
Specifically, inconsistent results were observed for effects on T3 in rats exposed to PFDA via i.p.
injection and results from the protein binding studies fGutshall etal.. 19891 suggest that PFDA
decreased protein binding of T4, which could result in increased fT4 and decreased tT4, which is
not consistent with the results from the NTP (20181 study. The mechanistic database is also limited
in that there are no studies that investigated the effects of PFDA on deiodinase activity.
Furthermore, the activities of thyroid-sensitive hepatic enzymes (e.g., L-glycerol-3-phosphate
dehydrogenase) were increased in rats exposed to PFDA via the i.p. route suggesting that thyroid
activity may not be decreased due to PFDA treatment. In general, the interpretation and relevance
of the mechanistic studies and supplemental information to thyroid effects observed in the NTP
(20181 study is unclear given that these studies used doses that were much higher (i.e., 20-80
mg/kg-day, as compared to <2.5 mg/kg-day) and associated with overt systemic toxicity.
Additionally, the mechanistic studies and supplemental information are of shorter duration and
rats were exposed to PFDA via i.p. injection rather than gavage as was done in the NTP (20181
study.

In addition to the uncertainty in the available evidence in adults, due to the sparse evidence
base available, concern remains for potential susceptible populations to PFDA-induced endocrine
effects in susceptible populations including young individuals exposed during gestation, early
childhood, and puberty. Importantly, T3 and T4 levels play critical roles in bone growth and brain
development (O'Shaughnessv etal.. 20191 at these various life stages. However, at the present time
few epidemiological studies and no animal toxicological studies have addressed the potential for
PFDA-induced effects in these populations. A primary delineating feature between adult animals

This document is a draft for review purposes only and does not constitute Agency policy.

3-263 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

and developing offspring is that adults have a considerable reserve thyroid hormone capacity
whereas developing offspring do not Thus, there is an elevated concern regarding the potential for
decreases in thyroid hormones during developmental life stages due to the critical endocrine
dependency of in utero and neonatal development.

Taken together, there is inadequate evidence across human, animal, and mechanistic data
to determine whether PFDA exposure would cause endocrine effects in humans. This conclusion is
based on inconsistent evidence from human studies and from a single high confidence rat study
investigating PFDA doses <2.5 mg/kg-day that reported largely incoherent effects on thyroid
hormone homeostasis and thyroid structure (i.e., increased T3, decreased TSH and T4; increased
thyroid weight; no histopathology) that cannot be interpreted based on the currently available
evidence base.

This document is a draft for review purposes only and does not constitute Agency policy.

3-264 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-41. Evidence profile table for PFDA exposure and endocrine effects

Evidence stream summary and interpretation

Evidence integration summary
judgment

Evidence from studies of exposed humans (see Section 3.2.6: Human studies)

QQQ

Inadequate Evidence

Primary basis: Single high
confidence study in rats showing
mixed effects on thyroid hormone
levels that cannot be reliably
interpreted.

Human relevance: Given the
general conservation of thyroid
function across rodents and
humans, evidence in animals is
presumed relevant to humans in
the absence of evidence to the
contrary.

Cross-stream coherence:

No factors noted.

Susceptible populations and
lifestages:

None identified, as a hazard is not
supported by the current evidence.

Other inferences:

None

Studies, outcomes,
and confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream judgement

Thvroid hormones
16 medium and 5 low
confidence studies

• Results from studies of
thyroid hormones were
inconsistent. Most
results were null, but
study sensitivity was
limited which hinders
interpretation. Positive
and inverse associations
were observed in a few
studies, but there was a
lack of consistency of
direction of association
across studies.

• No factors noted

• Unexplained
inconsistency

• Incoherence in
direction of
association across
hormones

• Lack of association
in studies with
limited sensitivity

QQQ

Indeterminate

While a subset of studies
suggests changes in thyroid
hormone levels with higher
levels of PFDA, there is
considerable uncertainty due
to inconsistency across
studies and endpoints.

Evidence from in vivo animal studies (see Section 3.2.6: Animal studies)

Studies, outcomes,
and confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream summary

Thvroid hormones

1 high confidence
study

• Significantly decreased
trend forTSH in males.

• Significant increased
trend forT3 in males.

• Increased T3 in females
at >1.25 mg/kg-d.

• Decreased fT4 was
reported at

>0.312 mg/kg-d in males

• Consistency for
decreased fT4 in
male and female
rats in a high
confidence study.

• Dose-response
gradient for
decreased TSH
(males only),
decreased fT4
(males and
females), and

• Lack of expected
coherence across
thyroid measures
(the pattern of
changes is
inconsistent with
any currently
available
understanding of
adverse thyroid-
related changes)

QQQ

Indeterminate

There is mixed evidence from
a single high confidence rat
study that reported largely
incoherent effects on thyroid
hormone homeostasis and
thyroid structure
(i.e., increased T3, decreased
TSH and fT4, but not tT4;

This document is a draft for review purposes only and does not constitute Agency policy.

3-265 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

and at >1.25 mg/kg-d in
females.

• No change in tT4

increased T3
(males and
females).

• Supportive
evidence for
decreased fT4
from supplemental
(mechanistic and
i.p.) studies.

•

• Unexplained
inconsistency
across T4 (free and
total)

measurements

increased thyroid weight; no
histopathology) that cannot
be reliably interpreted based
on the currently available
evidence base.

Histooatholosv

1 high confidence
study and 1 medium
confidence study

• No PFDA-induced
histopathological
changes were observed
for the thyroid gland,
adrenal cortex and
medulla, parathyroid
gland, and pituitary
gland.

• No factors noted.

• No factors noted

Organ weights

2 high confidence
studies

• Decreased absolute
adrenal gland weight in
males at >0.156 mg/kg-d
and females at

2.5 mg/kg-d (NTP,
2018).

• Increased relative
adrenal gland weight in
males at 2.5 mg/kg-d
(NTP, 2018).

• Increased absolute
thyroid in females at
0.312 to 1.25 mg/kg-d
but not at the highest

• Consistency for
increased relative
thyroid weight in
male and female
rats across two
high confidence
studies.

• Dose-response
gradient for
decreased
absolute adrenal
gland weight
(males and
females), increased
relative adrenal
gland weight

• No factors noted

This document is a draft for review purposes only and does not constitute Agency policy.

3-266 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

dose tested (2.5 mg/kg-
d) (NTP, 2018).

Increased relative
thyroid weight in males
at >1.25 mg/kg-d and
females at

>0.312 mg/kg-d (NTP,
2018).

(males only), and
increased relative
thyroid weight
(males and
females).

Coherence of
increased thyroid
weight and
decreased fT4.

Mechanistic evidence and supplemental information (see subsection above)

Biological events or pathways
(or other information)

Hypothalamic-pituitarv-thyroid
axis

Plasma protein binding

Activity of thyroid sensitive
hormones

Primary evidence evaluated
Key findings, interpretation, and limitations

Interpretation: The results suggest that PFDA may impair the ability
of the hypothalamic-pituitary-thyroid axis to respond to
physiological stimulation.

Key findings:

• Decreased T3 and T4 levels after TRH stimulation in vivo.
Limitations: one-time i.p. exposure; single study.

Interpretation: The results suggest that PFDA may impair the
binding of thyroid hormones to plasma transport proteins.
Key findings:

• PFDA decreased the plasma protein uptake of T3 and T4.
Limitations: one-time i.p. exposure; few studies.

Interpretation: The data indicate activation of thyroid sensitive
enzymes in a manner that suggests PFDA increases thyroid activity
in rats.

Key findings:

• PFDA increased the activities of L-glycerol-3-phosphate

dehydrogenase, cytosolic lactate dehydrogenase and cytosolic
malic enzyme, which are thyroid-sensitive hormones.

Limitations: one-time i.p. exposure; few studies.

Evidence stream summary

The mechanistic and
supplementary data provide
limited, inconsistent information
on how PFDA may be affecting
thyroid hormone homeostasis,
and the results may be
confounded by overt systemic
toxicity due to the high doses
used in the i.p. studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-267 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration summary
judgment

Binding to thvroid receotor

Interpretation: PFDA is capable of binding to the thyroid hormone

receptor.

Key findings:

Under in vitro conditions, PFDA was shown to bind to the human
thyroid hormone receptor. PFDA did not exhibit antagonistic or
agonistic effects on the thyroid receptor pathway.

Limitations: Single study available.

Other evidence

Interpretation: Effects after i.p. injection is consistent with results in

orally exposed rats.

Key findings:

Altered T3 and T4 levels.

Limitations: Effects on T3 levels were inconsistent among the i.p.
studies, which could be due to differences in experimental design
and the time at which thyroid hormones were measured

This document is a draft for review purposes only and does not constitute Agency policy.

3-268 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.9. URINARY EFFECTS
Human Studies

Nine epidemiology studies (14 publications) investigated the relationship between PFDA
exposure and urinary effects, including glomerular filtration rate (GFR) and uric acid (see
Figure 3-84). Two studies were considered uninformative due to lack of consideration of potential
confounding (Zhang et al.. 2019: Seo etal.. 2018). The remaining studies were classified as low
confidence primarily due to concerns for reverse causality (with potential for bias away from the
null). In essence, as described in Watkins etal. f20131. decreased renal function could plausibly
lead to higher levels of PFAS (including PFDA) in the blood due to reduced excretion. This
hypothesis is supported by data presented by Watkins etal. f20131. though there is some
uncertainty in their conclusions due to the use of modeled exposure data as a negative control and
the potential for the causal effect to occur in addition to reverse causality. The results least likely to
be affected by reverse causality were analyses in two studies stratified by glomerular filtration
stage, Tain (2019): (Zeng etal.. 2019c) and one study with a prospective design Blake etal. (2018).

Three studies fLin etal.. 2020b: Blake etal.. 2018: Oin etal.. 20161 reported associations
between PFDA exposure and impaired renal function (i.e., lower GFR, higher serum uric acid),
though only Blake etal. f 20181 was statistically significant and the associations in Oin etal. f20161
and Lin etal. f2020bl were limited to one sex (girls in Oin etal. T20161 and men in Lin et al.
(2020b)) (see Table 3-43). Conversely, Wang etal. (2019) reported higher GFR and lower odds of
chronic kidney disease with higher exposure. The remaining studies report null associations with
renal function, including the studies that stratified by glomerular function stage. Overall, there is
unexplained inconsistency in the direction of the association. More importantly, because of the
potential for reverse causation for this outcome, there is considerable uncertainty in interpreting
the available evidence.

This document is a draft for review purposes only and does not constitute Agency policy.

3-269 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Blake, 2018, 5080657-
Cakmak, 2022, 10273369-
Jain and Ducatman, 2019b, 7922952-
Lin, 2020, 6988476 -
Qin, 2016, 3981721-
Seo, 2018, 4238334-
Wang, 2019, 5080583-
Zeng, 2019, 5918630-
Zhang, 2019, 5083675-

+4-

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)

++
++

•

•f

¦f

Figure 3-84. Urinary effects human study evaluation heatmap. Refer to HAWC
for details on the study evaluation review: HAWC Human Urinary Effects.

Table 3-42. Associations between serum PFDA and urinary effects in low
confidence epidemiology studies

Reference

Population

Median
exposure

(IQR)
(ng/mL)

Result

Glomerular filtration rate

Blake et al.
(2018)

Prospective cohort of
residents near a
uranium processing site
(1990-2008); U.S.; 210
adults

0.1
(0.1-0.2)

Percent change (95% CI) in eGFR per IQR change in PFDA
-2.2 (-4.3, -0.1) *

Jain (2019)

Cross-sectional study
(NHANES) (2007-2014);
U.S.; 4,057 adults

0.2 in
GF-1 group

Adjusted geometric mean (95% CI) by glomerular function stage

GF stage
GF-1
GF-2
GF3-A
GF-3B/4

All participants
0.25 (0.24, 0.26)
0.27 (0.25, 0.29)
0.33 (0.26, 0.43)
0.23 (0.19,. 0.28)

Men
0.26 (0.25, 0.28)
0.28 (0.26, 0.31)
0.31 (0.25, 0.38)
0.21(0.21, 0.22)

Women
0.23 (0.22, 0.24)
0.26(0.24, 0.28)
0.37(0.35, 0.39)
0.24(0.19, 0.31)

Warig et ai.
(2019)

Cross-sectional study
(2015-2016); China;
1,612 adults

0.9 (0.5,1.5)

Mean change (95% CI) in eGFR per In-unit change in PFDA
1.04 (0.27,1.81) *

Uric acid

Mean (SD)

|3 (95% CI) in serum uric acid for quartiles vs Q1

This document is a draft for review purposes only and does not constitute Agency policy,

3-270' DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Reference

Population

Median
exposure

(IQR)
(ng/mL)

Result

Scinicariello
et al. (2020)

Cross-sectional study
(NHANES) (2009-2014);
U.S.; 4,917 adults

0.2 (0.01)

Without chronic kidney disease
Q2: 0.00 (-0.09, 0.10)
Q3: -0.05 (-0.17, 0.07)
Q4: 0.12 (0.00, 0.24)

With chronic kidney disease
Q2: 0.34 (-0.03, 0.72)
Q3: 0.19 (-0.13, 0.52)
Q4: 0.26 (-0.09, 0.61)

OR (95% CI) in hyperuricemia for quartiles vs Q1
Without chronic kidney disease
Q2: 0.94 (0.66, 1.34)
Q3: 0.86 (0.57, 1.25)
Q4: 1.30 (0.94, 1.80)

With chronic kidney disease
Q2: 1.32 (0.66, 2.65)
Q3: 0.98 (0.60, 1.61)
Q4: 1.26 (0.64, 2.46)

Zeng et al.
(2019c)

Cross-sectional study
(2015-2016); China;
384 adults

0.9(0.5-1.5)

Mean difference per log-unit change in PFDA
0.01 (-0.06, 0.08)

Qin et al.

Cross-sectional study
(2009-2010); Taiwan;
225 children and
adolescents (mean age:
13.6 yr)

0.9
(0.8-1.2)

Mean change (95% CI) in serum uric acid per In-unit change in PFDA

(2016)

All participants
0.08 (-0.11, 0.28)

Boys
0.05 (-0.23, 0.34)

Girls
0.18 (-0.09, 0.46)

OR (95% CI) for high uric acid per quartile change in PFDA

1.3 (0.8,1.9)

1.0(0.6, 1.7)

1.8 (0.9, 3.7)

Lin et al.

Cross-sectional study
(2016-2017); Taiwan;
397 older adults (55-75
yrs)

1.6(1.2-2.4)

(3 (95% CI) in serum uric acid for quartiles vs Q1

(2020b)

All participants
NR

Men

Q2: 0.31 (-0.38, 0.99)
Q3: 0.68 (-0.02, 1.37)
Q4: 0.68 (-0.04, 1.4)

Women
Q2: -0.09 (-0.45, 0.27)
Q3: -0.1 (-0.02, 1.37)
Q4: -0.18 (-0.54, 0.19)

Creatinine

Cakmak et
al. (2022)

Cross-sectional study
(2007-2017); Canada;
6,045 adults

Mean 0.2

% change per 1 mean increase in PFDA
-1.5 (-3.7, 0.7)

Chronic kidney disease

Wang et al.
(2019)

Cross-sectional study
(2015-2016); China;
1,612 adults

0.9 (0.5, 1.5)

OR (95% CI) for chronic kidney disease per In-unit change in PFDA
0.7 (0.6, 0.9) *

*p < 0.05.

1 Animal Studies

2 A 28-day study in female B6C3F1/N mice and two, 28-day studies in male and female S-D

3 rats are available to examine effects relevant to the evaluation of urinary system toxicity after PFDA

4 exposure fFrawlev etal.. 2018: NTP. 20181. The studies report on histopathology, serum

This document is a draft for review purposes only and does not constitute Agency policy.

3-271 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

biomarkers of effect and organ weights. Overall study confidence was high for most endpoints
evaluated in these studies with the exception of histopathology in Frawlev etal. (20181, which had
incomplete reporting of null data (results were only discussed qualitatively) resulting in a medium
confidence rating (see Figure 3-85).

Reporting quality-
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization -
Exposure timing, frequency and duration -
Endpoint sensitivity and specificity-
Results presentation -
Overall confidence A

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 3-85. Evaluation results for animal studies assessing effects of PFDA
exposure on urinary effects. Refer to HAWC for details on the study evaluation
review.

Histopathology

The kidney and urinary bladder were evaluated for histopathology across a high confidence
fNTP. 20181 and a medium confidence study fFrawlev et al.. 20181 in rats exposed for 28 days (see
Figure 3-86). NTP (20181 found no evidence of histopathological lesions in the urinary bladder of
males and females at the only dose examined (2.5 mg/kg-day). Chronic progressive
nephropathy (CPN) graded as minimal occurred in the kidneys of nearly all dose groups, including
controls, in this study fNTP. 20181 (see Figure 3-86). A reduction in the incidence of CPN was noted
in males and females at the highest dose tested (0% and 30% incidence at 2.5 mg/kg-day in females
and males respectively compared to 60% in controls) (NTP. 20181: but there was no clear dose-
response effect and incidences were in some instances increased at doses lower than
2.5 mg/kg-day (i.e., 0.156-1.25 mg/kg-day) in both sexes, as compared to controls. The other
28-day gavage study reported no effects in kidney histopathology in female rats up to doses of

This document is a draft for review purposes only and does not constitute Agency policy.

3-272 ' DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 0.5 mg/kg-day fFrawlev etal.. 20181. Taken together, the high dose decrease in CPN incidence in

2 rats in one study is not interpreted as biologically significant, and overall, the histopathology data

3 were considered null.

This document is a draft for review purposes only and does not constitute Agency policy.

3-273 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Study Name

Outcome
Confidence

Study
Design

Target Endpolnt Name
Organ

Animal Description

Trend Test
Result

Dose

(mg/kg-day)

| Statistically significant increase
3 No significant change

NTP, 2018. 4309127 High confidence 28 Day Oral Kidney Chronic Progressive Nephropathy Rat, Sprague-Dawley (Harlan) { ) significant 5/10(50.0%) 0

7/10(70.0%) 0.156

6/10(60.0%) 0.312
0.625

5/10(50.0%) 1.25

0/10 (0.0%) 2.5

Rat, Sprague-Dawley (Harlan) (.*) not significant 6/10(60.0%) 0

8/10(80.0%) 0.156

5/10(50.0%) 0.312

6/10(60.0%) 0.625

7/10(70.0%) 1.25

3/10(30.0%) 2.5

PFDA Male Reproductive Organ Histopathology

Figure 3-86. Kidney histopathology effects following exposure to PFDA in 28-day rat study (results can be viewed

by clicking the HAWC link: https://hawcprd.epa.gOv/summary/data-pivot/assessment/100500072/PFDA-Kidney-
Histopathology-effect-size-animal/l.

This document is a draft for review purposes only and does not constitute Agency policy,
3-274 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Serum biomarkers

Serum biomarkers of kidney injury and/or function, namely blood urea nitrogen (BUN) and
creatinine were measured in rats in one high confidence study fNTP. 20181 (see Table 3-44 and
Figure 3-87). Creatinine is a waste product of creatine metabolism produced in muscle tissue and
BUN is a waste product of protein metabolism in the liver. Both creatinine and BUN are removed
from the blood by the kidneys and often used as indicators of kidney function. Dose-related
increases in circulating BUN levels occurred in males and females, most notably at 1.25 and 2.5
mg/kg-day (25-50% compared to controls). In contrast, a significant downward trend was
reported for creatinine levels, reaching 4-11% decrease compared to controls at >1.25 mg/kg-day.
The decreases in creatinine levels were accompanied by significant decreases in glucose levels at
similar doses (31-51% compared to controls; data not shown in Table 3-44 or Figure 3-87) and
likely reflect the marked systemic toxicity associated with high-dose PFDA exposure (see Section
3.2.10 on General toxicity effects for more details) (NTP. 2018).

Table 3-43. Percent change relative to controls in serum biomarkers of
kidney function in a 28-day rat study after PFDA exposure fNTP. 20181

Animal group

Dose (mg/kg-d)

0.156

0.312

0.625

1.25

2.5

Blood urea nitrogen (BUN)

Male S-D rats

-9

-13

Female S-D rats

-2

Creatinine

Male S-D rats

-8

-11

Female S-D rats

-4

-5

-3

-4

-10

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study
authors.

Organ weight

Absolute and relative kidney weights were measured in the two 28-day gavage studies
using mice and/or rats fFrawlev etal.. 2018: NTP. 20181. There are some uncertainties
surrounding the most toxicologically relevant organ weight metric so both absolute and relative
kidney weights were evaluated herein (Craig etal.. 2015: Bailey etal.. 2004) (see Table 3-45 and
Figure 3-87). Absolute and relative kidney weights of female rats displayed an upward trend,
reporting increases of up to 11% and 13%, respectively, at a dose of 0.5 mg/kg-day in 1 out of 2
study cohorts exposed to similar experimental conditions fFrawlev etal.. 20181. Kidney weights
(absolute and relative) increased in response to PFDA exposure in the second study cohort, but the
changes were relatively small (0-5%) and a dose-related trend was not established. No appreciable
body weight changes were reported in this study up to the highest dose tested (0.5 mg/kg-day)
fFrawlev etal.. 2018). A separate study observed significant increases in relative kidney weight of
12-45% compared to controls in male and female rats at doses >0.625 mg/kg-day (NTP. 2018).

This document is a draft for review purposes only and does not constitute Agency policy.

3-275 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 Conversely, absolute kidney weight increased significantly in females by 9 and 15% at 0.312 and

2 0.625 mg/kg-day, respectively, but decreases were observed in both males and females at

3 2.5 mg/kg-day (10 and 15% from controls, respectively) fNTP. 20181. The apparent decreases in

4 absolute kidney weight at higher doses may be associated with concurrent reductions in body

5 weight occurring in the exposed animals (up 38% compared to controls at 2.5mg/kg-day) (see

6 Section 3.2.10 on General toxicity effects for more details) fNTP. 20181. In mice, kidney weights

7 were mostly unchanged by PFDA treatment (0.045-0.71 mg/kg-day) (Frawlev etal.. 20181. In

8 addition to the uncertainties due to confounding effects with decreased body weight at the highest

9 PFDA doses (>1.25 mg/kg-day), the observed kidney weight changes in rats are not supported by
10 significant histopathological findings in these animals fFrawlev et al.. 2018: NTP. 20181.

This document is a draft for review purposes only and does not constitute Agency policy.

3-276 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-44. Percent change relative to controls in kidney weights (absolute and relative to body weight) due to
PFDA exposure in short-term oral toxicity studies

Animal group

Dose (mg/kg-d)

0.045

0.089

0.125-0.179

0.25-0.36

0.5-0.71

1.25

2.5

Absolute kidney weight

28 d; female S-D rats -Histopathology
cohort

Frawlevetal. C20181

28 d: female S-D rats - MPS cohort Frawlev
etal. (2018)

28 d; female S-D rats
NTP(2018)

-15

28 d; male S-D rats
NTP(2018)

-1

-2

-10

28 d; female B6C3F1/N mice
Frawlev et al. (2018)

-1

-3

Relative kidney weight

28 d; female S-D rats -Histopathology
cohort

Frawlev et al. (2018)

28 d: female S-D rats - MPS cohort Frawlev
etal. (2018)

28 d; female S-D rats
NTP(2018)

28 d; male S-D rats
NTP(2018)

28 d; female B6C3F1/N mice
Frawlev et al. (2018)

-2

-5

-7

Bold values indicate instances where statistical significance (p < 0.05) compared to controls was reported by study authors; shaded cells represent doses not
included in the individual studies.

This document is a draft for review purposes only and does not constitute Agency policy.

3-277 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Effect Endpoint Name

Clinical Chemistry Blood Urea Nitrogen (BUN)

Creatinine (CREAT)

Organ Study Name Outcome Confidence Experiment Name

Blood NTP, 2018,4309127 High confidence 28 Day Oral

High confidence 28 Day Oral

Blood NTP, 2018,4309127 High confidence 28 Day Oral

High confidence 28 Day Oral

Species, Strain (sex)
Rat. Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)

Trend Test Result

significant
significant
significant
significant

PFDA Urinary Effects

-•—~—~

Histopathology Chronic Progressive Nephropathy

Kidney Histopathology
Urinary Bladder Histopathology

Kidney NTP, 2018,4309127 High confidence 28 Day Oral

High confidence 28 Day Oral
Kidney Frawley, 2018, 4287119 Medium confidence 28 Day Oral

Bladder NTP, 2018, 4309127 High confidence 28 Day Oral

High confidence 28 Day Oral

Rat. Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat. Sprague-Dawley (Harlan)

not significant
significant
not reported
not applicable
not applicable

Organ Weight Kidney Weight, Absolute (Histophatology Cohort) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral

Kidney Weight, Absolute (MPS Cohort) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral

Right Kidney Weight, Absolute Kidney NTP, 2018,4309127 High confidence 28 Day Oral

High confidence 28 Day Oral

Kidney Weight, Absolute (Hematology Study) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral

Kidney Weight, Relative (Histopathology Cohort) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral

Kidney Weight, Relative (MPS Cohort) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral

Right Kidney Weight, Relative Kidney NTP, 2018,4309127 High confidence 28 Day Oral

High confidence 28 Day Oral

Kidney Weight, Relative (Hematology Study) Kidney Frawley, 2018, 4287119 High confidence 28 Day Oral
f # No significant change Statistically significant increase Statistically significant decrease I

Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Mouse, B6C3F1/N ( - )
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Rat, Sprague-Dawley (Harlan)
Mouse, B6C3F1/N (9)

significant
not significant
significant
not significant
not significant
significant
not significant
significant
significant
not significant

Dose (mg/kg-day)

Figure 3-87. Urinary effects following exposure to PFDA in short-term oral studies in animals (results can be

viewed by clicking the HAWC link: https: //hawcprd.epa.gOv/summarv/data-pivot/assessment/100500072 /pfda-urinary-
effects/1.

This document is a draft for review purposes only and does not constitute Agency policy,

3-278 ' DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence Integration

The evidence for potential urinary system effects in humans is considered indeterminate.
Associations between PFDA exposure and impaired renal function were reported in two low
confidence epidemiological studies. However, there is considerable uncertainty in the
interpretation of these findings due to the potential for reverse causation and some unexplained
inconsistency in the direction of association across studies.

The evidence for potential urinary system effects in experimental animals is limited to three
high/medium confidence studies in rats fFrawlev etal.. 2018: NTP. 20181 and one high confidence
study in mice with exposure for 28 days fFrawlev etal.. 20181. Although alterations in BUN and
creatine levels were observed at >1.25 mg/kg-day in rats, there is no coherent pattern of effects
(BUN levels increased and creatinine levels decreased) or supportive information
(i.e., histopathology) to determine the toxicological relevance of the changes that occurred fNTP.
20181. Histopathological examinations of rat kidney and urinary bladder were mostly
unremarkable across two studies fFrawlev etal.. 2018: NTP. 20181. Finally, the interpretation of
the absolute and relative kidney weight changes in rats at doses >0.312 mg/kg-day is complicated
by the lack of coherent histopathological findings fFrawlev etal.. 2018: NTP. 20181. inconsistencies
in the direction of changes across experiments, and confounding effects from significant body
weight reductions at the highest doses tested (>1.25 mg/kg-day) fNTP. 20181. In summary, the
sparse and uncertain evidence from animal studies is considered indeterminate. The absence of any
long-term studies (subchronic/chronic) via the oral route or other relevant routes of exposure
increases uncertainty in the evaluation of potential urinary system toxicity in animals following
PFDA exposure.

Altogether, based on the available human and animal studies, there is inadequate evidence
to assess whether PFDA exposure can cause urinary system effects in humans (see Table 3-46).

This document is a draft for review purposes only and does not constitute Agency policy.

3-279 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-45. Evidence profile table for PFDA urinary effects

Evidence stream summary and interpretation

Evidence integration
summary judgment

Evidence from studies of exposed humans (see Section 3.2.8: Human studies)

QQQ

Inadequate Evidence

Studies, outcomes, and
confidence

Key findings and
interpretation

Factors that increase
strength or certainty

Factors that decrease
strength or certainty

Evidence stream judgment

Seven low confidence
studies

• Three studies reported
some associations
between PFDA
exposure and impaired
renal function (i.e.,
lower GFR or higher
serum uric acid).

• One study reported
associations in the
opposite direction and
three others were null

• No factors noted

• Low confidence studies
due to potential for
reverse causality

• Unexplained
inconsistency

QQQ

Indeterminate

There is some evidence of
urinary effects with PFDA
exposure across two low
confidence studies but
considerable concerns for
reverse causality and
inconsistency.

Primary basis:

Evidence from
epidemiological studies and
experimental animals is
indeterminate.

Human relevance, cross-
stream coherence,
susceptibility, and other
inferences:

No specific factors are
noted.

Evidence from in vivo animal studies (see Section 3.2.8: Animal studies)

Histopathology

1 high and 1 medium
confidence studies in rats
for 28 days

• Mostly null findings for
kidney and urinary
bladder

histopathology in rats
up to 2.5 mg/kg-d
across two studies.

• A high dose (2.5
mg/kg-d) decrease in
the incidence of CPN
in rats reported in one
study was not
interpreted as
biologically significant.

• No factors noted

QQQ

Indeterminate

Lack of coherent effects in
high and medium
confidence studies in rats
and mice exposed up to
2.5 mg/kg-d for 28 d.

This document is a draft for review purposes only and does not constitute Agency policy.

3-280 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence stream summary and interpretation

Evidence integration
summary judgment

Serum biomarkers

1 high confidence study in
rats for 28 d

• Increased BUN levels
and decreased
creatinine levels in rat
serum at >1.25 mg/kg-
d (alterations in
creatinine levels
coincide with body
weight reductions)

• High confidence study

• Lack of expected
coherence in the
directionality of BUN
and creatinine changes

• Potential confounding
by body weight
decreases

Organ weight

2 high confidence studies
(encompassing 4
experiments) in mice
and/or rats for 28 d

• Absolute and relative
kidney weight changes
in rats at doses
>0.312 mg/kg-d
(directionality of
effects varied across
experiments and
organ weight
measures); no effects
in mice up to
0.71 mg/kg-d

• High confidence
studies

• Unexplained

inconsistency across
experiments, species,
and organ weight
measures

This document is a draft for review purposes only and does not constitute Agency policy.

3-281 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

3.2.10. GENERAL TOXICITY

The potential for PFDA exposure-induced general toxicity is specifically discussed given
that PFDA has been shown to cause a "wasting syndrome" in rodents, which is characterized by
decreased food intake and reduced body weight fGoecke-Flora etal.. 19951. In animals, decreased
body weights can be indicative of non-specific overt toxicity and some effects that occur at doses
associated with this and other frank effects should be interpreted cautiously when drawing
conclusions about organ-/system-specific hazards. Thus, this section informs judgments drawn for
other potential health hazards, but a specific evidence integration judgment is not drawn.

Human Studies

No human studies were available to inform the potential for PFDA exposure to cause
general toxicity.

Animal Studies

Animal toxicity studies reporting general toxicity with repeated dose exposure to PFDA
include two 28-day gavage studies, four dietary exposure studies (7-14 days) in mice and/or rats,
and two drinking water studies (12-49 days) in mice. The endpoints measured in these studies
include body weight fLi etal.. 2022: Wang etal.. 2020: Frawlev etal.. 2018: NTP. 2018: Kawashima
etal.. 1995: Takagi etal.. 1992.19911. clinical observations fNTP. 20181 and survival fWang etal..
2020: NTP. 20181 (Figure 3-67). Three studies fLi etal.. 2022: Frawlev etal.. 2018: NTP. 20181
were evaluated as high confidence for all general toxicity endpoints tested (see Figure 3-88). Four
studies (Wang etal.. 2020: Kawashima etal.. 1995: Permadi etal.. 1993: Takagi etal.. 19921 were
evaluated as medium confidence for all general toxicity endpoints tested while the Takagi et al.
f!9911 study was evaluated as low confidence for the body weight endpoint (see Figure 3-67). Key
issues regarding study quality evaluation in the medium and low confidence studies were related to
exposure sensitivity (no analytical verification methods or quantitative data on food consumption),
allocation/randomization of animals into experimental groups, and deficiencies in data reporting
(see Figure 3-88).

This document is a draft for review purposes only and does not constitute Agency policy.

3-282 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Reporting quality -

Allocation -

Observational bias/blinding -

Confounding/variable control -

++ ++

Selective reporting and attrition -

Chemical administration and characterization -

Exposure timing, frequency and duration -

Endpoint sensitivity and specificity
Results presentation
Overall confidence

Legend

Good (metric) or High confidence (overall)

Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NR| Not reported
* Multiple judgments exist

I I •

Figure 3-88. PFDA general toxicity animal study evaluation heatmap. Refer to

HAWC for details on the study evaluation review.

Body weight

PFDA-induced body weight suppression was observed to be dose-dependent in short-term
animal studies in rats fFrawlev et al.. 2018: NTP. 2018: Kawashima etal.. 1995: Takagi etal.. 1992.
19911 and mice fLi et al.. 2022: Wang et al.. 2020: Frawlev et al.. 2018: Permadi et al.. 19931
(Figure 3-68). In rats treated with doses ranging from 1.0-10 mg/kg-day, reductions in mean body
weight and body weight gain ranged from 4-38% and 21-103% respectively, compared to controls.
In mice, changes in body weight were less than 5% at doses <0.71 mg/kg-day but decreases
reached 53% at 6.6 mg/kg-day. In the 28-day high confidence study that included multiple study
cohorts fFrawlev etal.. 20181. the study authors reported that 2 out of 88 rats in the 2.0 mg/kg-day
exposure group were euthanized due to marked reductions in body weight (>20%) occurring
within the first 5 days of the study initiation fFrawlev etal.. 20181. This evidence of PFDA-induced
acute toxicity was also observed in several single intraperitoneal (i.p.) injection studies as discussed
below. Furthermore, PFDA-induced decreased body weight in female rats was more severe with
longer treatment durations fFrawlev etal.. 20181. For example, body weight was decreased by 4%
at Day 15, by 13% at Day 22, and 22% at Day 29 at 2.0 mg/kg-day. Also, in this study, reduced body
weight was observed to be more sensitive to dose at Day 29 compared to earlier time points
(statistically significant at 1.0 mg/kg-day on Day 29 compared to 2.0 mg/kg-day for Days 15 and
22). The NTP f20181 study also showed similar results for multiple timepoint data for body weight.
For example, in male rats treated with the highest dose (2.5 mg/kg-day), body weight was
decreased by 13, 27, and 38% on Day 15, Day 22, and Day 29, respectively.

This document is a draft for review purposes only and does not constitute Agency policy,

3-283 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Clinical observations and survival

Clinical observations and survival data are available from a high confidence gavage study in
S-D rats exposed for 28 days fNTP. 20181. Additionally, a medium confidence study reported effects
on survival in male CD-I mice exposed to PFDA in the drinking water for 49 days fWang etal..
20201. PFDA exposure was associated (albeit not statistically significant) with thin appearance in
male and female S-D rats at the highest exposure dose tested (2.5 mg/kg-day) (see Figure 3-89).
The incidence rate was 30% in males and 10% in females compared to 0% for the corresponding
controls. Nasal/eye discharge was observed in 1 out 10 male rats in the control, 0.156, 0.0625,1.25
and 2.5 mg/kg-day exposure groups. No other clinical observations were reported. All exposed
animals survived and were euthanized at study termination. In summary, 28-day gavage exposure
to PFDA caused mild clinical symptoms in rats (thin appearance) but had no effect on survival in
this study. However as discussed above, Frawlevetal. f20181 reported that two (of 88) rats were
euthanized due to severe weight loss caused by 5 days of exposure to PFDA at 2.0 mg/kg-day. In
mice exposed to PFDA for up to 49 days, the mortality rate was reported to be significantly
increased at 6.6 mg/kg-day (Wang etal.. 20201.

This document is a draft for review purposes only and does not constitute Agency policy.

3-284 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint Name

Study Name

Outcome Confidence

Exposure Design

Species, Strain (Sex)

Observation Time

Trend Test Resul

Body Weight

Kawashima. 1995, 3858657

Medium confidence

7 Day Oral

Rat, Wistar (c?)

Day 7

not reported

Takagi 1992, 1320114

Medium confidence

7 Day Oral

Rat. Fischer F344 ( v)

Day 7

not reported

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (c )

Day 29

significant

Rat, Sprague-Dawley (Harlan) (7)

Day 29

significant

Body Weight (All Study Cohorts)

Frawley, 2018,4287119

High confidence

28 Day Oral

Rat. Sprague-Dawley (Harlan) (V)

Day 1

significant

Rat, Sprague-Dawley (Harlan) (7)

Day 8

significant

Rat. Sprague-Dawley (Harlan) ($)

Day 15

significant

Rat. Sprague-Dawley (Harlan) (y)

Day 22

significant

Rat. Sprague-Dawley (Harlan) (y)

Day 29

significant

Body Weight

Permadi. 1993, 1332452

Medium confidence

10 Day Oral PFDA

Mouse. C57BI/6 (tf)

Day 10

not reported

Body Weight (All Study Cohorts)

Frawley, 2018. 4287119

High confidence

28 Day Oral

Mouse. B6C3F1/N (¥)

Day 1

significant

Mouse, B6C3F1/N (7 )

Day 8

significant

Mouse. B6C3F1/N (¥)

Day 15

significant

Mouse. B6C3F1/N ($)

Day 22

significant

Mouse. B6C3F1/N (?)

Day 29

significant

Body Weight Gain (All Study Cohorts)

Frawley, 2018,4287119

High confidence

28 Day Oral

Rat. Sprague-Dawley (Harlan) ($)

Day 1- Day 8

significant

Rat, Sprague-Dawley (Harlan) (V)

Day 1- Day 15

significant

Rat. Sprague-Dawley (Harlan) (7)

Day 1- Day 22

significant

Rat, Sprague-Dawley (Harlan) (y)

Day 1 - Day 29

significant

Mouse. B6C3F1/N (?)

Day 1- Day 8

significant

Mouse, B6C3F1/N (¥)

Day 1- Day 15

significant

Mouse, B6C3F1/N ($)

Day 1- Day 22

significant

Mouse, B6C3F1/N ('+>)

Day 1- Day 29

significant

Nasal/Eye Discharge

NTP. 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) ( :')

Day 1 - 29

not reported

Thin Appearance

NTP. 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (y)

Day 1 - 29

not reported

Rat, Sprague-Dawley (Harlan) (o)

not reported

Survival

NTP, 2018, 4309127

High confidence

28 Day Oral

Rat, Sprague-Dawley (Harlan) (y)

Day 29

not reported

Rat. Sprague-Dawley (Harlan) (:f )

Day 29

not reported

PFDA General Toxicity Effects

•—•

# Dose

A Significant increase
~ ^ Significant decrease

mg/kg-day

Figure 3-89. PFDA general toxicity effects (results can be viewed by clicking the HAWC link:

https://hawcprd.epa.gOv/summarv/data-pivot/assessment/100000026/pfda-general-toxicitv-effects/I

This document is a draft for review purposes only and does not constitute Agency policy,

3-285 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Mechanistic studies and supplemental information

Several intraperitoneal (i.p.) studies using a single injection, have demonstrated that PFDA
induces a "wasting syndrome" in rodents, which is characterized by decreased food intake and
reduced body weight fGoecke-Flora et al.. 19951. In these studies, decreased body weight (5 to
72% compared to controls or pretreatment values) was observed in rats at doses ranging from 20
to 100 mg/kg PFDA (Unkila etal.. 1992: Bookstaff etal.. 1990: Chen etal.. 1990: Ylinen and Auriola.
1990: Gutshall et al.. 1988: Van Rafelghem and Andersen. 1988: Van Rafelghem et al.. 1988a: Kelling
etal.. 1987: Langlev and Pilcher. 1985: Olson and Andersen. 19831. Generally, across rodent
species, i.p. injection of PFDA at doses >20 mg/kg-day, even acutely, caused generalized acute
toxicity. Whereas significant decreases in food intake were also observed in rats at 40 to 80 mg/kg,
body weights were reduced compared to both ad-libitum and pair-fed controls suggesting that
PFDA-decreased body weight is not only related to reduced food intake but also a direct effect of
PFDA on body weight. In guinea pigs, body weight gain (32% decrease) and food intake (11%
decrease) were significantly reduced at 20 mg/kg PFDA via the i.p. route (Chime etal.. 19941. In a
study that tested multiple species, rats lost a maximum of 45% of their pretreatment body weight
at 50 mg/kg PFDA, hamsters lost 26% at 50 mg/kg and 41% at 100 mg/kg, and mice lost 25% at
150 mg/kg fVan Rafelghem etal.. 1987bl. Multiple other i.p. studies reported effects on body
weight and food intake, but the data were presented qualitatively or graphically, and percent
changes were not calculated. Doses for these studies ranged from 10 to 100 mg/kg (Kudo and
Kawashima. 2003: Wilson etal.. 1995: Chen etal.. 1994: Glauertetal.. 1992: Arand etal.. 1991:
Powers and Aust. 19861. Most of the studies described here utilized a single injection of PFDA,
highlighting the acute toxicity and rapid weight loss caused by PFDA treatment. It is important to
note that the doses used in the mechanistic/supplemental studies are much higher than the doses
in which body weight was decreased in some of the toxicity studies. For example, decreases in
body weight interpreted as biologically significant were observed in rats at >1.25 mg/kg-day from
the NTP C20181 study.

Summary of animal and mechanistic/supplemental information

The available studies for PFDA-induced general toxicity were mostly high and medium
confidence (see Figure 3-89) and evaluated endpoints related to general toxicity (body weight,
clinical observations, and survival) in multiple strains (S-D, Wistar, Fisher F344,

C57BL/6N and B6C3F1/N) of male and female rats and mice via gavage and dietary exposure for up
to 28 (Frawlev etal.. 2018: NTP. 2018: Kawashima etal.. 1995: Permadi etal.. 1993: Takagi etal..
1992.19911. Reduced body weight was consistently observed in all available animal studies, with
biologically significant effects occurring at doses as low as 1.25 mg/kg-day in rats from the NTP
(20181 study. The consistent effect of PFDA on body weight that appears to be time- and dose-
related coupled with clinical observations (i.e., thin appearance) in rats provide support for PFDA-
induced general toxicity. Furthermore, multiple acute i.p. studies across different species reported

This document is a draft for review purposes only and does not constitute Agency policy.

3-286 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

decreased body weight indicative of "wasting syndrome" at doses ranging from 20 to 100 mg/kg,
but primarily at >40 mg/kg-day.

3.2.11. OTHER HEALTH EFFECTS

Short-term oral exposure studies [high/medium confidence) in experimental animals
evaluated potential health effects related to the hematological, respiratory, digestive, dermal,
musculoskeletal, and adult nervous system (please see Section 3.2.7 for the synthesis of evidence
on neurodevelopmental effects). The available evidence from these animal studies is briefly
summarized below. Given the limitations of the evidence base and the lack of consistent or
coherent effects of PFDA exposure, there is inadequate evidence to determine whether any of the
evaluated outcomes below might represent potential human health hazards of PFDA exposure.
Additional studies on these health effects could modify these interpretations.

Animal studies

Other health effects

Hematological parameters were evaluated across two studies in male and/or female S-D
rats and one study in female B6C3F1/N mice, all with gavage exposure for 28-days fFrawlev etal..
2018: NTP. 20181. No significant effects were found in mice up to 0.71 mg/kg-day fFrawlev et al..
20181. In rats, mean corpuscular hemoglobin (amount of hemoglobin per red blood cell [RBC];
MCH) and mean corpuscular hemoglobin concentration (amount of hemoglobin per unit of RBC
volume; MCHC) decreased at the two highest doses (0.25 and 0.5 mg/kg-day) in one study fFrawlev
etal.. 20181: however, the changes did not show a dose-response gradient and were relatively small
(6-7% compared to controls). In the other rat study, a significant dose-related trend was reported
for several hematological parameters fNTP. 20181. Erythrocyte (RBCs) counts increased (9-23%)
in males and females and hematocrit (proportion of RBCs in blood; 6-16%) and hemoglobin (7-
19%) concentrations increased in females only at doses >1.25 mg/kg-day. These changes were
accompanied by decreases in reticulocyte counts (immature RBCs) of 54-91%, and slight decreases
in mean corpuscular volume (average volume of RBCs; decreases of 3-7%) and MCH (4%) and
slight increases in MCHC (2-4%) in males and females at similar doses. In addition, the platelet
count in females decreased by up to 30% in females at the highest dose group, 2.5 mg/kg-day. In
summary, although there is some potential evidence of hematological effects in rats with PFDA
exposure fNTP. 20181. the observed changes occurred mostly in the presence of significant
systemic toxicity (i.e., reduced body weights at >2.5 mg/kg-day), which limits the interpretation of
the findings.

Histopathology of the dermal, musculoskeletal, nervous, and special senses (eye and
harderian gland) systems was examined in the control and 2.5 mg/kg-day dose groups in adult S-D
rats in one 28-day study that reported null findings fNTP. 20181. The digestive and respiratory
systems were examined histologically in S-D rats across two, 28-days studies fFrawlev etal.. 2018:

This document is a draft for review purposes only and does not constitute Agency policy.

3-287 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

NTP. 20181. No lesions were identified in stomach or lungs of rats at doses of 0.125-0.5 mg/kg in
one study (Frawlev etal.. 20181. The second study found lesions in the esophagus, forestomach.
lungs and nose of exposed rats fNTP. 20181. Increased incidence of forestomach lesions
(epithelium hyperplasia, inflammation, and ulcer) was reported in males and inflammation was
reported in the lungs and esophagus of females. The incidence rates for these lesions were low
(10-20%) and restricted to the highest dose group (2.5 mg/kg-day). The nose lesions (epithelium
degeneration, hyperplasia, and chronic inflammation) were increased in both males and females
(10-50% incidence) across 0.158-2.5 mg/kg-day, but there was no clear dose-response
relationship, and these morphological changes were also observed in the control group (0-20%
incidence). Overall, the limited information available for these organ systems impedes further
evaluation of the biological significance of the histopathological results.

3.3. CARCINOGENICITY

3.3.1. CANCER
Human studies

Eight studies evaluated the risks of cancer associated with exposures to PFDA fVelarde et
al.. 2022: Liu etal.. 2021b: Omoike etal.. 2021: Lin etal.. 2020a: Tsai etal.. 2020: Wielsae etal..
2017: Christensen etal.. 2016: Hardell etal.. 20141. Five cancer studies by (Velarde etal.. 2022:
Omoike et al.. 2021: Lin etal.. 2020a: Wiels0e etal.. 2017: Christensen etal.. 20161 were evaluated
as 'Uninformative.'

The study of risks of prostate cancer (Hardell etal.. 20141 was low confidence due to
concern about the exposure measurement not representing the etiologically relevant time period,
potential for confounding, insufficiencies in the analysis, and concerns about sensitivity (see Figure
3-90). Hardell etal. f20141 reported a non-significantly increased risk of prostate cancer among
men with PFDA concentrations in blood that were above the median value. The study of risks of
thyroid cancer (Liu etal.. 2021b) was low confidence due to concern about the exposure
measurement not representing the etiologically relevant time period, deficiencies regarding the
outcome definition, and potential for confounding, (see Figure 3-90). Liu etal. f2021bl reported
significantly decreased risk of thyroid cancer associated with increasing quartiles of PFDA. The
study of risks of breast cancer fTsai etal.. 20201 was low confidence due to concern about the
exposure measurement not representing the etiologically relevant time period, potential for
confounding, and concerns about low sensitivity (see Figure 3-90). Tsai etal. (2020) reported non-
significantly increased risk of breast cancer per In-transformed unit increase in PFDA concentration
in blood among women <=50 years of age; and non-significantly decreased risk of breast cancer
per In-transformed unit increase in PFDA concentration in blood among women >50 years of age.
In summary, the available epidemiologic evidence on PFDA and the risks of cancer is limited and
generally uninformative.

This document is a draft for review purposes only and does not constitute Agency policy.

3-288 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicologiccil Review of Perfluorodecanoic Acid and Related Salts

Participant selection

Exposure measurement -

Outcome ascertainment

Confounding
Analysis
Sensitivity -
Selective Reporting -
Overall confidence

S H

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 3-90. Study evaluation results for epidemiology studies of PFDA and
cancer. Refer to HAWC for details on the study evaluation review: HAWC Human
Cancer.

Animal studies

There are no long-terra animal bioassay studies available for PFDA. One short-term study
reported null findings for neoplastic histopathology in male and female rats gavaged with doses of
0-2.5 mg/kg-day for 28 days fNTP. 20181. The study performed a complete necropsy of control
and PFDA-exposed groups, examining various tissues (i.e., esophagus, intestine, liver, pancreas,
salivary glands, stomach, bloodvessel, heart, adrenal cortex, adrenal medulla, parathyroid gland,
pituitary gland, thyroid gland, epididymis, preputial gland, prostate seminal vesicle, testes, clitoral
gland, ovary, uterus, bone marrow, lymph node, spleen, thymus, mammary gland, skin, bone, brain,
lung nose, eye, harderian gland, kidney and urinary bladder). However, the study was considered
low confidence for the assessment of carcinogenicity due to the inadequacy of the short-term
exposure duration for evaluating the long-term development of potential cancers. Although 28-day
studies may be able to provide some information on preneoplastic lesions, the study duration does
not cover the entire spectrum of tumor development and promotion for nearly all cancer types and
thus they are insensitive.

Mechanistic studies and supplemental information

The scope of the analysis for evaluating putative mechanisms of carcinogenicity for PFDA
focused on the synthesis of genotoxicity studies based on data availability. A more comprehensive

This document is a draft for review purposes only and does not constitute Agency policy,

3-289 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

and rigorous mode of action (MOA) investigation was not attempted due to the sparse and low
confidence human and animal studies available, as well insufficient information for the evaluation
of alternative carcinogenic mechanisms (e.g., mitogenesis, inhibition of cell death, cytotoxicity with
reparative cell proliferation and immune suppression) or considerations for human relevance of
tumor responses in animals, susceptible populations and lifestages and anticipated shape of dose-
response relationships. This is in agreement with the proposed framework for cancer MOA analysis
in the EPA Guidelines for Carcinogen Risk Assessment, which states that "the framework supports a
full analysis of mode of action information, but it can also be used as a screen to decide whether
sufficient information is available to evaluate or whether the data gaps are too substantial to justify
further analysis" fU.S. EPA. 20051.

Studies evaluating the genotoxic, mutagenic and clastogenic potential of PFDA from in vitro
assays with prokaryotic organisms and mammalian cells and in vivo assays in rats and mice are
summarized in Table 3-47. Mutagenicity test results in S. typhimurium (TA98, TA100, TA1535,
TA1537, and TA1538) and E. coli strains (WP2 uvrA pKMlOl) across several studies were
consistently negative for PFDA in the presence or absence of S9 rat liver metabolism system (NTP.
2005: Kim etal.. 1998: Godin etal.. 1992: Mvhr etal.. 19901. Similarly, PFDA had no effect on
mutation frequency in L5178Y mouse-lymphoma cells and in the HGPRT forward mutation assay in
Chinese hamster ovary (CHO) cells with or without S9 metabolic activation f Godin etal.. 1992:

Mvhr etal.. 1990: Rogers etal.. 19821.

PFDA was inactive for the in vitro transformation of BALB/C-3T3 mouse cells (Godin etal..
19921 and in the sister chromatic exchange (SCE) assays in CHO cells but induced chromosomal
aberrations indicative of clastogenic effects under conditions of S9 metabolic activity (Godin etal..
1992: Mvhr etal.. 19901. PFDA caused DNA double-strand breaks (DSB) in human gastric
adenocarcinoma AGS and SGC cell lines, although the details of the study exposure methodology
including information on the test article concentrations were not provided fLiu etal.. 2019al. The
mechanisms of PFDA-induced DSB were attributed to the downregulation of X-ray repair cross
complementing 4 (XRCC4) expression and nonhomologous end-joining (NHEJ) inactivation. These
events lead to impairment of DNA damage repair and inhibition of p5 3 expression and apoptosis,
contributing to the observed alterations in cell sensitivity to chemotherapy fLiu etal.. 2019a).
Elevated levels of DSB were also detected in mice with PFDA treatment (dosing regimen was not
specified) fLiu etal.. 2019al. Xu etal. f2019bl also showed increases in DNA strand breaks, 80HdG
formation and ROS levels, indicative of oxidative DNA damage in primary mouse hepatocytes
exposed to PFDA. In vivo experiments in rats showed increase in oxidative DNA damage (80HdG
levels) in liver tissue after dietary PFDA treatment at 10 mg/kg-day for 2 weeks (Takagi etal..
19911 but no effects were reported with a lower dose (1.4 mg/kg-day) administered via i.p. for up
to 8 weeks (Kim etal.. 19981. There were no effects on frequency of micronucleated polychromatic
or normochromatic erythrocytes in blood after repeated dose PFDA treatment (0.156-2.5 mg/kg-
day) via gavage fNTP. 20121. PFDA was not associated with induction of unscheduled DNA

This document is a draft for review purposes only and does not constitute Agency policy.

3-290 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 synthesis (UDS) in primary hepatocytes isolated from rats after single-dose exposure (>11 mg/kg);

2 however, increase in S-phase DNA synthesis was observed in the exposed rats (Godin etal.. 1992:

3 Mvhr etal.. 19901.

4 In summary, PFDA does not appear to elicit a strong genotoxic response as demonstrated by

5 the lack of activity in most assays described above, including mutagenicity tests in prokaryotic

6 organisms and mammalian cells; SCE and cell transformation assays in vitro; and UDS, oxidative

7 DNA damage and micronucleus assays in rats. Nevertheless, there is some evidence of potential

8 clastogenic effects in CHO cells, S-phase induction in rat hepatocytes, double strand DNA breaks in

9 human and mouse gastric cells and oxidative DNA damage in primary mouse hepatocytes.

This document is a draft for review purposes only and does not constitute Agency policy.

3-291 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 3-46. Test evaluating genotoxicity and mutagenicity

Test

Materials and methods

Results

Conclusions

References

Genotoxicity studies in prokaryotic organisms

Ames assay

5. typhimurium strains (TA98, TAIOO, TA1535, TA1537, and
TA1538) were tested with or without S9 rat liver
homogenate and with a pre-incubation period. PFDA
concentrations ranged from 33.3 to 10, 000 ng/plate.

No increase in the number of
reverent colonies was observed with
PFDA in any of the tester strains in
the presence or absence of S9
metabolic activation.

There is no evidence of PFDA
mutagenicity in 5. typhimurium
strains.

Godin et al.
(1992): Mvhret
al. (1990)

Ames assay

5. typhimurium strains (TA98 and TA1535) were incubated
with PFDA (1 to 100 g/plate) with or without S9.

Test results were negative in the
two strains tested irrespective of the
presence of S9.

There is no evidence of PFDA
mutagenicity in 5. typhimurium
strains.

Kim et al.
(1998)

Ames assay

5. typhimurium strains (TA98 and TA100) and £ coli strain
(WP2 uvrA pKMlOl) in the presence or absence of S9.
Concentrations of PFDA were 0-10,000 pg/plate.

Test results were negative in all
bacterial strains irrespective of the
presence of S9.

There is no evidence of PFDA
mutagenicity in 5. typhimurium and
£ coli strains.

NTP(2005)

Genotoxicity studies in mammalian cells - in vitro

Mutagenicity
assay

L5178Y mouse-lymphoma cells were treated with PFDA
(0.01-500 Mg/mL) for 24 h and plated in the presence of
selective agents to evaluate mutation frequency (ouabain,
excess thymidine, methotrexate, cytosine arabinoside and
thioguanine) and in non-selective medium to evaluate
survival.

Mutagenicity tests showed no
significant results in any of the
selective systems.

There is no evidence of PFDA
mutagenicity in L5178Y cells.

Rogers et al.
(1982)

CHO/HGPRT
forward
mutation assay

Chinese hamster ovary (CHO) cells were treated with PFDA
concentrations ranging from 0.005 to 0.5 mg/mL with or
without S9.

The results were negative for PFDA-
mediated induction of forward
mutations in the HGPRT locus in
CHO cells under conditions of S9
metabolic activation and
nonactivation.

There is no evidence of PFDA
mutagenicity in CHO cells in the
HGPRT forward mutation assay.

Godin et al.
(1992): Mvhret
al. (1990)

This document is a draft for review purposes only and does not constitute Agency policy.

3-292 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Test

Materials and methods

Results

Conclusions

References

Cytogenetic
assays in CHO
cells

CHO cells were treated with PFDA to evaluate induction of
sister chromatic exchange (SCE) and chromosomal
aberrations with or without S9. PFDA concentrations of
0.167 to 5,000 Mg/mL were tested in the SCE assays and
7.50 to 201 ng/mL were used in the chromosomal
aberration assay.

The results of the SCE assay were
negative in the presence or absence
of S9 metabolic activation. PFDA did
induce chromosomal aberrations at
151 and 201 Mg/mL but only under
conditions of metabolic S9
activation. Cytotoxicity was
observed at a concentration of
201 Mg/mL in the chromosomal
aberration assay.

Induction of chromosomal
aberrations provides evidence of
clastogenic activity of PFDA in
combination with S9. PFDA did not
cause DNA damage in the SCE assay.

Godin et al.
(1992): Mvhret
al. (1990)

In vitro

transformation
of BALB/C-3T3
cells

BALB/C-3T3 mouse cells were treated with PFDA at doses
of 40.0 to 650 ng/mL with or without S9.

PFDA failed to significantly increase
morphological transformation in
BALB/C-3T3 cells in the presence or
absence of S9 metabolism.

There is no evidence of malignant
transformation with PFDA in
cultured BALB/C-3T3 mouse cells.

Godin et al.
(1992)

DNA damage

(double-strand

breaks)

Human gastric adenocarcinoma AGS and SGC cell lines
treated with PFDA (concentration not specified).

PFDA induced double-strand DNA
breaks, reduced DNA repair activity,
altered expression of DNA repair
gene pathways (e.g., NHEJ),
inhibited apoptosis via p53
downregulation and affected
chemotherapy sensitivity of human
gastric cells.

PFDA can cause double strand DNA
damage in vitro by altering DNA
repair mechanisms.

Liu et al.
(2019a)

DNA damage
(strand breaks
and oxidative
damage
[80HdG])

Primary hepatocytes isolated from male C57BL/6 mice and
exposed to PFDA at doses of 0.1,1,10,100 mM.

PFDA increased DNA strand breaks
and levels of 80HdG and ROS in
primary mouse hepatocytes
(statically significant only at highest
dose for ROS but there was a dose-
response gradient).

There is evidence of oxidative DNA
damage with PFDA in vitro
exposure.

Xu et al.
(2019b)

Genotoxicity studies in mammalian species - in vivo

Unscheduled
DNA synthesis
(UDS) and S-
phase induction
assays

Adult male F344 rats were treated by oral gavage with a
dose of PFDA (5.5 to 44.0 mg/kg) and primary hepatocyte
cultures were prepared ~15-48 h after treatment to
examine nuclear labeling.

PFDA was found to be inactive in the
UDS assays but induced a significant
increase in the number of S-phase
cells at doses >11.0 mg/kg.

S-phase induction provides some in
vivo evidence of genotoxicity with
PFDA.

Godin et al.
(1992): Mvhret
al. (1990)

This document is a draft for review purposes only and does not constitute Agency policy.

3-293 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Test

Materials and methods

Results

Conclusions

References

Oxidative DNA
damage (80HdG)

Male Fischer F344 rats were treated with PFDA (0.01% or
10 mg/kg-d) via the diet for 14 d. DNA was isolated from
the liver and kidney of rats after treatment for analysis of
80HdG formation.

80HdG levels were significantly
increased by PFDA treatment in rat
liver but no effects were seen in the
kidney.

PFDA (10/mg/kg-d) caused oxidative
DNA damage in rat liver after
repeated dose exposure via the diet.

(Takagi et al.,
1991)

Oxidative DNA
damage (80HdG)

Female Sprague Dawley rats were treated with a dose of
10 mg/kg PFDA via i.p. once a week for a 2- or 8-week
period. DNA was isolated from rat liver after treatment for
analysis of 80HdG formation.

80HdG levels were not significantly
affected by PFDA treatment in the
two time points analyzed.

PFDA (1.4 mg/kg-d) did not cause
oxidative DNA damage in rat liver
after repeated dose exposure via
i.p. administration.

Kim et al.
(1998)

Micronucleus
assay

Male and female Sprague Dawley rats (5/group) were
exposed daily to PFDA by oral gavage at doses of 0, 0.156,
0.312, 0.625,1.25 and 2.5 (males only) mg/kg for 28 d.

Test results were negative for the
increase in frequency of
micronucleated polychromatic or
normochromatic erythrocytes in rat
blood.

There is no evidence of PFDA
(0.156-2.5 mg/kg-d) genotoxicity in
the erythrocyte micronucleus assay.

NTP (2012)

DNA damage

(double-strand

breaks)

Mice were exposed to PFDA via drinking water (dosing
regimen was not specified)

PFDA induced double-strand DNA
breaks in mouse gastric cells.

PFDA can cause double strand DNA
damage in vivo.

Liu et al.
(2019a)

CA = chromosomal aberration; CHO = Chinese hamster ovary; DNA = deoxyribonucleic acid; LD50 = median lethal dose; ROS = reactive oxygen species;
S-D = Sprague Dawley.

This document is a draft for review purposes only and does not constitute Agency policy.

3-294 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Evidence integration

The available evidence to evaluate the potential for PFDA exposure to lead to the
development of any cancer type consists of sparse and minimally informative studies in humans
and animals and limited mechanistic information from genotoxicity studies. Specifically, the single
low confidence study of prostate cancer (reporting an association that was not statistically
significant) in exposed humans, as well as the single, low confidence null study in rats with poor
sensitivity due to short-term duration are of limited utility for drawing a conclusion regarding
potential carcinogenicity with PFDA exposure. The results from genotoxicity studies were mostly
null, although a few studies provided some evidence of potential genotoxic effects in response to
PFDA (i.e., clastogenic effects in CHO cells, S-phase induction in rat hepatocytes, double strand DNA
breaks in human and mouse gastric cells and oxidative DNA damage in primary mouse
hepatocytes). Considering evidence for all potential cancer types across the available human,
animal and mechanistic studies and based on the EPA cancer guidelines (U.S. EPA. 2005). the
evidence base is judged to be inadequate to assess the carcinogenic potential of PFDA in humans.

This document is a draft for review purposes only and does not constitute Agency policy.

3-295 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

4.SUMMARY OF HAZARD IDENTIFICATION
CONCLUSIONS

4.1. SUMMARY OF CONCLUSIONS FOR NONCANCER HEALTH EFFECTS

The available evidence indicates hazards likely exist with respect to the potential for liver,
immune, developmental, and male and female reproductive effects in humans, given sufficient
PFDA exposure conditions12. Additionally, the available evidence suggests that PFDA exposure
might also have the potential to cause cardiometabolic and neurodevelopmental effects in humans
given sufficient exposure conditions. These judgments were derived primarily from epidemiological
studies and studies in experimental animals, the latter exposed to PFDA during short-term (7-28
days) and developmental (GD 6-15) oral exposures. On the other hand, there is inadequate
evidence for urinary, endocrine, and other health effects to determine the potential for health
hazards in humans with PFDA exposure. A summaiy of the justifications for
the evidence integration judgments for each of the main hazard sections is provided below.

The hazard identification judgment that the evidence indicates PFDA exposure is likely to
cause liver effects in humans, given sufficient exposure conditions12, is based on concordant effects
for increased liver weight, alterations in levels of serum biomarkers of liver injury (ALT, AST, ALP,
bile salts/acids, bilirubin and blood proteins), and some evidence of hepatocyte degenerative or
necrotic changes that provide support for the adversity of PFDA-induced liver toxicity reported in
high and medium confidence studies in rats and mice exposed to PFDA doses >0.156 mg/kg-day.
Although associations between serum ALT levels and PFDA exposure in epidemiological studies of
adults were observed, the epidemiology evidence overall is uncertain due to unexplained
inconsistency in the results for other clinical markers and a lack of clear evidence of adversity.
Mechanistic studies in rodents and limited evidence from in vitro studies and animal models
considered more relevant to humans provide support for the biological plausibility and human
relevance of the apical effects observed in animals and suggest a possible PPARa-dependent and
independent MOA for PFDA-induced liver toxicity.

The hazard identification judgment that the evidence indicates PFDA exposure is likely to
cause immunosuppression in humans, given sufficient exposure conditions12, is based on moderate
human evidence of immunosuppression primarily from two medium confidence studies in children
and one low confidence study in adults at levels of 0.3 ng/mL (median exposure in studies
observing an adverse effect). Although some evidence for coherent immunomodulatory responses

12 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes only and does not constitute Agency policy.

4-1 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

consistent with immunosuppression (decreases in phagocytic activity of liver microphages, spleen
cell counts and immune organ weights and immune organ histopathology) was identified in short-
term, high, and medium confidence studies in rats and mice at >0.089 mg/kg-day, the animal
evidence overall is uncertain. Issues with overt organ and general systemic toxicity pose
limitations with respect to the interpretation of the animal evidence. Although possible effects of
hypersensitivity-related responses were reported in one epidemiological study and one high-
exposure study in mice (21.4 mg/kg-day), outstanding uncertainties remain to draw specific
conclusions for this outcome.

The hazard identification judgment that the evidence indicates PFDA exposure is likely to
cause developmental toxicity, given sufficient exposure conditions13, is based primarily on
consistent findings of dose-dependent decreases in fetal weight in mice gestationally exposed to
PFDA doses >0.5 mg/kg-day, supported by evidence of decreased birth and childhood weight from
studies of exposed humans in which PFDA was measured during pregnancy. The conclusion is
further supported by coherent epidemiological evidence for biologically related effects (e.g.,
decreased birth length).

The hazard identification judgment that the evidence indicates PFDA exposure is likely to
cause potential adverse effects to the male reproductive system in humans, given sufficient
exposure conditions, is based on a coherent pattern of effects on sperm counts, testosterone levels,
and male reproductive histopathology and organ weights at doses >0.625 mg/kg-day in adult rats
exposed for 28 days (high confidence for most endpoints evaluated). Although the MOA for PFDA-
induced male reproductive effects is unknown, a few acute i.p. and in vitro rodent studies suggest a
possible mechanism via disruption of Leydig cell function and impaired steroidogenesis. Evidence
from a medium confidence epidemiological study reported non-statistically significant decreases in
testosterone levels and altered sperm parameters that are coherent with the effects observed in
animals. Although these findings were imprecise, the study had limited sensitivity to observe an
effect and no conflicting evidence was identified from studies of similar confidence.

The hazard identification judgment that the evidence indicates PFDA exposure is likely to
cause female reproductive toxicity in humans given sufficient exposure conditions is based
primarily on the results of a high confidence study in rats showing biologically coherent effects on
uterus weight and the estrous cycle after oral exposure to PFDA at >1.25 mg/kg-day for 28 days.
Although human studies are available for examining associations between PFDA and female
reproductive toxicity (e.g., fecundity), the results were mostly null, possibly due to their low
sensitivity for observing effects.

13 The "sufficient exposure conditions" are more fully evaluated and defined for the identified health effects
through dose-response analysis in Section 5.

This document is a draft for review purposes oniy and does not constitute Agency poiicy.

4-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

The hazard identification judgment that the evidence suggests PFDA exposure has the
potential to cause cardiometabolic effects in humans given sufficient exposure conditions 14is based
primarily on associations between PFDA and serum lipids, adiposity, cardiovascular disease, and
atherosclerosis in a few epidemiological studies. However, evidence is largely inconsistent across
studies, which adds considerable uncertainty. Evidence in experimental animals from a high
confidence rat study was indeterminate.

The hazard identification judgment that the evidence suggests PFDA exposure has the
potential to cause neurodevelopmental effects in humans given sufficient exposure conditions15 is
based on associations between PFDA exposure and outcomes related to attention and behavior,
although there is high degree of uncertainty due to inconsistencies and imprecision in the results.
No relevant animal studies were available.

Finally, there is inadequate evidence to evaluate the potential for PFDA exposure to cause
effects on the endocrine system, urinary system, and other health outcomes in adult humans
(i.e., respiratory, digestive, dermal, musculoskeletal, and hematological systems, and nonspecific
clinical chemistry). The available data from human and/or animal studies for these health
outcomes was largely limited or lacked consistency and coherence. Further, the absence of studies
examining the potential for effects of PFDA exposure on the thyroid in developing organisms, or on
mammary glands, represent data gaps in light of associations observed for other PFAS, such as
PFBS, PFOA and PFOS CATSDR. 2018b: U.S. FPA. 20181. see Table 4-1 below.

Table 4-1. Hazard conclusions across published EPA PFAS human health
assessments

Health Outcome

EPA PFAS Assessments3'11

PFDA

PFBA

PFBS

GenX
Chemicals

PFOAc

PFOSc

Thyroid

Human: +
Animal: +/-

Human: +/-
Animal: +/-

Liver

Human: +
Animal: +

Human: -
Animal: +

Developmental

+/"

Human: +
Animal: +

Reproductive

+/"

Human: -
Animal: +/-

Immunotoxicity

+/"

Human: +
Animal: +

Human: +/-
Animal: +

14 Given the uncertainty in this judgment and the available evidence, this assessment does not attempt to
define what might be the "sufficient exposure conditions" for developing these outcomes (i.e., these health
effects are not advance for dose-response analysis in Section 5).

15 Given the uncertainty in this judgment and the available evidence, this assessment does not attempt to
define what might be the "sufficient exposure conditions" for developing these outcomes (i.e., these health
effects are not advance for dose-response analysis in Section 5).

This document is a draft for review purposes only and does not constitute Agency policy.

4-3 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Health Outcome

EPA PFAS Assessments3'11

PFDA

PFBA

PFBS

GenX
Chemicals

PFOAc

PFOSc

Renal

+/-

Human: +/-
Animal: +/-

Hematological

+/-

Ocular

Serum Lipids

+/-

Human: +
Animal: +

Human: +

Hyperglycemia

Human: -
Animal: -

Animal: +/-

Nervous System

+/-e

Human: -
Animal: -

Animal: +/-

Cardiovascular

+/-

Cancer

+/-

+/"

+/-

a Assessments used multiple approaches to summarizing their non-cancer hazard conclusions; for comparison
purposes, the conclusions are presented as follows: V =evidence demonstrates or evidence indicates (e.g.,

PFDA), or evidence supports (e.g., PFBS);=suggestive evidence;= inadequate evidence (e.g., PFDA) or
equivocal evidence (e.g., PFBS); and = sufficient evidence to conclude no hazard (no assessment drew this
conclusion).

bThe assessments all followed the EPA carcinogenicity guidelines (2005); a similar presentation to that used to
summarize the noncancer judgments is applied for the cancer hazard conclusions, as follows: V = carcinogenic to
humans or likely to be carcinogenic to humans;= suggestive evidence of carcinogenic potential;=
inadequate information to assess carcinogenic potential; and = not likely to be carcinogenic to humans (no
assessment drew this conclusion).

c The U.S. EPA (2016b) and U.S. EPA (2016a) PFOA and PFOS assessments did not use structured language to
summarize the noncancer hazard conclusions. The presentation in this table was inferred from the hazard
summaries found in the respective assessments; however, this is for comparison purposes only and should not be
taken as representative of the conclusions from these assessments. Those interested in the specific noncancer
hazard conclusions for PFOA and PFOS must consult the source assessments.
d No data available for this outcome for this PFAS, so 'entered by default.
eData available for PFDA includes neurodevelopmental outcomes in humans.

This document is a draft for review purposes only and does not constitute Agency policy.

4-4 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

4.2. SUMMARY OF CONCLUSIONS FOR CARCINOGENICITY

Given the limited scope and utility of the available evidence across human, animals and
genotoxicity studies, the evidence is judged to be insufficient to determine whether PFDA exposure
(via any exposure route) might affect the development of any specific cancer types. In accordance
with EPA cancer guidelines (U.S. EPA. 20051 a weight of evidence descriptor of inadequate to
assess the carcinogenic potential is assigned for PFDA.

4.3. CONCLUSIONS REGARDING SUSCEPTIBLE POPULATIONS AND LIFE
STAGES

Understanding of potential areas of susceptibility to the identified human health hazards of
PFDA can help to inform expectations of variability in responses across individuals, as well as
uncertainties and confidence in candidate toxicity values (see Section 5.2). The available human
and animal studies indicate that early life represents a susceptible lifestage for the effects of PFDA
exposure. Two medium confidence studies reported immune effects (i.e., decreased antibody
response) in children exposed to PFDA during gestation and childhood (Grandiean etal.. 2017b)
and (Grandiean etal.. 2017a: Grandiean etal.. 2012). Additionally, developmental effects (i.e., fetal
growth restriction, gestational duration, postnatal growth and spontaneous abortion) were
reported in multiple high quality studies (Buck Louis etal.. 2018: Gvllenhammar etal.. 2018: Meng
etal.. 2018: Lind etal.. 2017a: Swedish Environmental Protection Agency. 2017: Valvi etal.. 2017:
Woods etal.. 2017: Bach etal.. 2016: Kwon etal.. 2016: Lenters etal.. 2016: Wang etal.. 2016:
Robledo etal.. 20151. The strongest and most consistent evidence was observed for fetal growth
restriction. Potentially coherent with these epidemiological observations, effects in developing
rodents (decreased fetal body weight, skeletal variations, decreased live fetuses per litter) after
maternal exposure also support the potential for early life susceptibility. Young individuals may
also be susceptible to PFDA-induced male reproductive effects. Although no animal studies and
only a few human studies are available examining reproductive effects in early lifestages
(i.e., pubertal development and anogenital distance), effects on sperm motility and testosterone
were consistently reported in exposed human and rodent adults fNTP. 2018: Zhou etal.. 2016:
Toensen et al.. 2013). Given the potential for PFDA to impair androgen function, boys exposed
during critical developmental lifestages may be susceptible as exposure during gestation and early
postnatal life stages could result in agenesis of the male reproductive system and/or infertility.

Although inconclusive, some effects on thyroid hormone homeostasis were observed in
adult rats fNTP. 20181. Although no studies are available that assessed the effect of PFDA on
thyroid hormones in developing organisms, young individuals exposed during gestation, early
childhood and puberty may be a susceptible population given thatT3 and T4 levels play critical
roles in bone growth and brain development (O'Shaughnessv etal.. 2019) at these lifestages
(i.e., both pregnancy and early life). PFDA was also observed to disrupt estrous cyclicity in female

This document is a draft for review purposes only and does not constitute Agency policy.

4-5 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 rats with potential implications for impaired fertility fNTP. 20181. Therefore, although the current

2 evidence does not explicitly address the potential for a linkage between these observations and

3 impaired fertility in women, women of reproductive age may also be susceptible to the effects of

4 PFDA exposure.

This document is a draft for review purposes only and does not constitute Agency policy.

4-6 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

5.DERIVATION OF TOXICITY VALUES

5.1. NONCANCER AND CANCER HEALTH EFFECT CATEGORIES
CONSIDERED

The available evidence indicates that oral exposure to PFDA is likely to cause adverse
hepatic, immune, developmental, and male and female reproductive effects in humans given
sufficient exposure conditions3, based on epidemiological and animal toxicity studies. This section
aims to characterize the dose levels associated with these identified hazards and derive toxicity
values as presented below. Additionally, the available evidence suggests PFDA exposure might
have the potential to cause cardiometabolic and neurodevelopmental effects in humans given
sufficient PFDA exposure conditions4 based on a limited number of epidemiological studies;
however, the results are considered too uncertain to support the derivation of toxicity values. For
all other health effects (i.e., endocrine, urinary, hematology, special senses [eye and harderian
gland], dermal and musculoskeletal systems), the evidence is inadequate to assess the hazard
potential; therefore, these endpoints were not considered for the derivation of toxicity values.

There are no available studies to inform the potential for PFDA to cause adverse health
effects via inhalation exposure, therefore, the derivation of reference concentrations (RfC) is
precluded (see Section 5.2.4). Likewise, evidence pertaining to the evaluation of carcinogenicity
was considered inadequate to assess carcinogenic potential of PFDA in humans, precluding the
derivation of cancer toxicity values via any exposure route (see Section 5.3).

5.2. NONCANCER TOXICITY VALUES

The noncancer toxicity values (i.e., RfDs) derived in this section are estimates of an
exposure for a given duration to the human population (including susceptible subgroups and/or life
stages) that are likely to be without an appreciable risk of adverse health effects (Section 1.2.1).
The RfD derived in Section 5.2.1 corresponds to chronic, lifetime exposure and is the primary focus
of this document In addition, a less-than-lifetime toxicity value (referred to as a "subchronic RfD")
is derived in Section 5.2.2. This subchronic RfD can be useful for certain decision purposes
(e.g., site-specific risk assessments with less-than-lifetime exposures). Both the lifetime and
subchronic RfD include organ/system-specific RfDs (osRfDs) associated with each health effect
considered for point of departure (POD) derivation, as supported by the available data. These
toxicity values might be useful in some contexts (e.g., when assessing the potential cumulative
effects of multiple chemical exposures occurring simultaneously). Section 5.2.3 summarizes that no
information exists to inform the potential toxicity of inhaled PFDA or to derive an inhalation
reference concentration (RfC).

This document is a draft for review purposes only and does not constitute Agency policy.

5-1 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

5.2.1. Oral Reference Dose (RfD) Derivation

Study/Endpoint Selection

As outlined in the sections below, data sufficient to support dose-response analyses for oral
PFDA exposure were available for all identified human health hazards (see Section 4.1): hepatic,
immune, developmental, and male and female reproductive effects. Rationales for study selection
and the specifics of RfD calculations, as well as the determination of confidence in quantitative
estimates are detailed in this section.

The following general considerations were used to prioritize studies for estimating points of
departure (PODs) for potential use in toxicity value derivation. Dependent on the evidence for each
identified hazard, high or medium confidence human studies that were deemed influential to the
hazard conclusions and suitable for dose-response analysis were prioritized for POD derivation and
compared to PODs derived from animal data when possible. Human studies were available for
developmental and immunotoxicity effects. For other health effects (i.e., hepatic, and male and
female reproductive effects), only evidence from animal studies was considered influential for
hazard identification and, therefore, these data were prioritized for dose-response assessment.
Given the lack of comprehensive subchronic or chronic animal studies, medium and high confidence
short-term studies in animals of longer exposure duration (e.g., 28 days versus 7 or 14 days) and
with exposure levels near the lower dose range of doses tested across the evidence base were
preferred along with medium or high confidence animal studies evaluating exposure during
development These types of medium and high confidence human and animal studies increase the
confidence in the resultant RfD because they represent data with lower risk of bias and reduce the
need for low-dose and exposure duration extrapolation (see Appendix C.l.l,).

A summary of endpoints and rationales considered for toxicity value derivation is presented

below.

Hepatic effects

The hazard conclusions for PFDA-induced liver effects are based primarily on moderate
evidence from short-term animal studies (see Section 3.2.1). In humans, an association between
PFDA exposure and ALT levels in the blood was identified, but there was considerable uncertainty
due to inconsistent results for other clinical markers. As such, only animal studies were considered
for dose-response analysis. The database of animal studies examining liver effects includes several
short-term studies in rats and mice fWang etal.. 2020: Frawlev etal.. 2018: NTP. 2018: Yamamoto
and Kawashima. 1997: Kawashimaetal.. 1995: Permadi etal.. 1993: Takagi etal.. 1992.1991:

Harris and Birnbaum. 1989). In particular, two high confidence studies in S-D rats gavaged with
PFDA for 28-days were prioritized for the derivation of candidate values because they included
several hepatic endpoints that together provided coherent evidence of liver toxicity with PFDA
exposure across histopathology, organ weights and/or clinical chemistry fFrawlev etal.. 2018: NTP.
20181 (see Table 5-1). Additionally, these studies had the longest exposure duration (28 days) and

This document is a draft for review purposes only and does not constitute Agency policy.

5-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

examined the lower range of PFDA doses (dose range of observed effects is 0.156-2.5 mg/kg-day)
across the available studies examining hepatic effects.

PFDA induced changes in serum liver biomarkers, hepatocyte lesions and increased liver
weights in rats across the two 28-day studies fFrawlev etal.. 2018: NTP. 20181. Although some of
the individual changes have the potential to represent adaptive responses (e.g., increased liver
weights and hypertrophy), the constellation of coherent liver effects, most notably consistent
effects across multiple serum biomarkers of hepatocyte and biliary injury and histological findings
of structural hepatocyte degeneration (necrosis), provide clear evidence of adversity (see
"Consideration for potentially adaptive versus adverse responses" under Section 3.2.1 for more
details). Alterations in the levels of serum enzymes such as ALT, AST and ALP and other functional
biomarkers (bile salt/acids, bilirubin, and blood proteins [albumin, globulin, and total protein])
were reported in the 28-day study that evaluated clinical chemistry fNTP. 20181. Increases in AST
and ALP levels were consistent across sexes and dose groups and generally occurred at lower doses
that did not induce significant body weight changes or other general systemic effects (0.156-
0.625 mg/kg-day PFDA). Similarly, dose-related increases in relative liver weights were reported
in male and female rats at >0.125 mg/kg-day across the two 28-day studies (Frawlev etal.. 2018:
NTP. 20181. As discussed in Section 3.2.1, relative liver weight is generally preferred over absolute
liver weight; as information on the former were available, changes in absolute liver weight were not
considered for dose-response analyses. Since there is no clear indication of sex-specific differences
in sensitivity with respect to PFDA-induced liver effects in the available animal toxicity studies, data
for both male and female S-D rats for these endpoints were advanced for dose-response modeling.

Corroborative hepatocyte lesions such as cytoplastic alterations and vacuolization,
hypertrophy and necrosis were reported in rats at higher doses (>0.625 mg/kg-day) across the two
28-day studies prioritized for dose-response analysis fFrawlev etal.. 2018: NTP. 20181. The
histopathological observations showed a clear progression in severity across lesions and dose
groups. These findings provide additional support for the adversity of the progressive effects on the
liver with PFDA exposure but were not prioritized for dose-response analysis due to the presence
of more sensitive liver endpoints (i.e., serum AST and ALP levels, and relative liver weight; see
Table 5-1).

Table 5-1. Endpoints considered for dose-response modeling and derivation
of points of departure for liver effects in animals

Endpoint

Study reference
and confidence

Exposure
route and
duration

Test strain,
species, and
sex

POD derived?

Notes

Increased serum ALT

NTP (2018):
high confidence

Gavage, 28 d

S-D rat, male
and female

Dose-dependent effects were
only observed in females and
occurred at higher doses
compared to other liver
findings

This document is a draft for review purposes only and does not constitute Agency policy.

5-3 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

Study reference
and confidence

Exposure
route and
duration

Test strain,
species, and
sex

POD derived?

Notes

Increased serum AST

NTP (2018):
high confidence

Gavage, 28 d

S-D rat, male
and female

Yes

Dose-dependent effects were
consistent across sexes and
concordant with liver weight
and liver histopathology
findings

Increased serum ALP

NTP (2018):
high confidence

Gavage, 28 d

S-D rat, male
and female

Yes

Effects were consistent across
sexes and dose groups and
concordant with liver weight
and liver histopathology
findings.

Other serum biomarkers
(increased bile salts/acids
and bilirubin, and
decreased albumin and
globulin)

NTP (2018):
high confidence

Gavage,
28 days

S-D rat, male
and female

Effects were mostly consistent
across sexes but occurred at
higher doses compared to
other liver findings

Hepatocyte lesions

NTP (2018):
high confidence
(cytoplasmic
alterations and
vacuolization,
hypertrophy,
and necrosis)

Gavage,
28 days

S-D rat, male
and female

Effects were consistent across
sexes and studies but
occurred at higher doses
compared to other liver
findings

Frawlev et al.
(2018): hiah
confidence
(necrosis)

Gavage, 28 d

S-D rat, male

Increased relative liver
weight

NTP (2018):
high confidence

Gavage, 28 d

S-D rat, male
and female

Yes

Dose-dependent effects were
consistent across studies,
cohorts, sexes and were
concordant with serum
biomarker and liver
histopathology findings. There
was no reason to prioritize
one dataset over the other.

Frawlev et al.
(2018): hiah
confidence

Gavage, 28 d

S-D rat, female
(included 3
experimental
cohorts)

Yes

1 Immune Effects

2 As described in Section 3.2.2, the strongest evidence for immune effects was from

3 epidemiological studies that provided moderate evidence of immunosuppression (Shih etal.. 2021:

4 Timmermann etal.. 2021: Grandieanetal.. 2017b: Grandiean etal.. 2017a: Kielsenetal.. 2016:

5 Grandiean etal.. 20121: thus, this outcome was prioritized for dose-response analysis and studies of

6 hypersensitivity (which collectively provided slight human evidence) were not considered. Given

7 the uncertainties with the animal data described in Section 3.2.2 that would be expected to strongly

8 impact quantitative estimates (e.g., influence of systemic toxicity), only the human data were

9 considered for the derivation of PODs.

This document is a draft for review purposes only and does not constitute Agency policy.

5-4 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

The two medium confidence epidemiology studies of antibody response following
vaccination providing the primary support for the hazard judgment were conducted in different
birth cohorts of the Faroe Islands population (see Table 5-2). These studies include measures of
PFDA exposure taken perinatally (pregnancy week 32 to 2 weeks postpartum), at 18 months, and at
5, 7, and 13 years, and measures of antibody levels at 5, 7, and 13 years for both diphtheria and
tetanus. The relevant etiologic window of exposure for this outcome is not known. Although there
were some heterogeneous results (see Section 3.2.2), the direction of association across these
combinations of different timings of exposure and outcome measurement were generally
consistent, indicating immunosuppression (i.e., decreased antibody response with higher
exposure). However, selecting the most informative exposure-outcome combination(s) for POD
derivation is complicated by the lack of a clear etiologic window. In a follow-up publication without
new data, the study authors performed benchmark dose modeling for a subset of the data
presented in Grandiean etal. (2012). specifically antibody levels at age 7 and PFDA concentrations
at age 5, and antibody levels at age 5 (prebooster) and perinatal PFDA concentrations (Budtz-
T0rgensen and Grandiean. 2018b). These were selected by the authors due to the strong inverse
associations observed and the results were considered reasonably representative of the study
results overall. After review of the BMD methods and additional modeling details fBudtz-Iargensen
and Grandiean. 2018bl for completeness and appropriateness (see Appendix C.l "Benchmark Dose
Response Modeling Results from Human Studeis," EPA utilized their analytic regression results for
this assessment

Budtz-l0rgensen and Grandiean (2018a) fit multivariate models of PFDA measured at age 5
years, against log2-transformed anti-tetanus antibody concentrations measured at the 7 year-old
examination controlling for sex, exact age at the 7 year-old examination, and booster type at age 5
years. Three model shapes of PFDA were evaluated by Budtz-largensen and Grandiean f2018al: a
linear model, a piecewise-linear model with a knot at the median, and a logarithmic function.
Ultimately, the linear model was found to have the best fit In the absence of a clear definition of an
adverse effect for a continuous endpointlike antibody concentrations, a default BMR of one SD
change from the control mean may be selected, as suggested in EPA's Benchmark Dose Technical
Guidance Document (U.S. EPA. 2012a). A lower BMR can also be used if it can be justified on a
biological and/or statistical basis. Regression coefficients for PFDA as the only PFAS in the model
were used to estimate the BMD and BMDL for a BMR of one standard deviation (SD) change in log2-
transformed anti-tetanus antibody concentration and for a BMR of Vi standard deviation (SD)
change in log2-transformed anti-tetanus antibody concentration (see Appendix C.l.l for details).
Budtz-l0rgensen and Grandiean (2018a) also fit multivariate models of PFDA controlling for both
PFOS and PFOA and BMD and BMDL estimates from those results were also derived in Appendix
C.l.l.

Statistically, the Technical Guidance additionally suggests that studies of developmental
effects can support lower BMRs, and BMRs of Vi SD (or BMRs of 5% rather than 10%) are routinely

This document is a draft for review purposes only and does not constitute Agency policy.

5-5 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

applied to rodent developmental toxicity study endpoints due to the sensitive lifestage.

Biologically, a BMR of Vi SD is considered a reasonable choice as anti-tetanus antibody
concentrations prevent against tetanus, which is a rare, but severe and sometimes fatal infection,
with a case-fatality rate in the U.S. of 13% during 2001-2008 fLiang etal.. 20181. The case-fatality
rate can be more than 80% for early lifestage cases fPatel and Mehta. 19991. Selgrade f20071
suggests that specific immuno-toxic 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 than just a single immuno-
toxic effect. Thus, decrements in the ability to maintain effective levels of tetanus antitoxins
following immunization may be indicative of wider immunosuppression in these children exposed
to PFDA. Taken together, the severity of this indicator of developmental immunosuppression and
the sensitive lifestage is interpreted to support the use of a BMR of Vi SD

A blood concentration for tetanus antibodies of 0.1 IU/mL is sometimes cited in the tetanus
literature as a 'protective level' and fGrandiean etal.. 2017bl noted that the Danish vaccine
producer Statens Serum Institut recommended the 0.1 IU/mL "cutoff" level "to determine whether
antibody concentrations could be considered protective;" and Galazka and Kardymowicz
fl9891mentions the same concentration, but Galazka et al. f!9931argues:

"The amount of circulating antitoxin needed to ensure complete immunity against
tetanus is not known for certain. Establishment of a fixed level of tetanus antitoxin
does not take into consideration variable conditions of production and adsorption of
tetanus toxin in the anaerobic area of a wound or a necrotic umbilical stump. A given
serum level could be overwhelmed by a sufficiently large dose of toxin. Therefore, there
is no absolute protective level of antitoxin and protection results when there is
sufficient toxin-neutralizing antibody in relation to the toxin load fPassen and
Andersen. 19861."

As a check, EPA evaluated how much extra risk would have been associated with a BMR set
at a cutoff value of 0.1 IU/mL. Using the observed distribution of tetanus antibodies at age 7 years
in log2(IU/mL), EPA calculated that 2.8% of those values would be below the cutoff value of 0.1
IU/mL. A BMR of Vi SD resulted in 7.9% of the values being below that cutoff which is 5.1% extra
risk and shows that the generic guidance that a BMR of Vi SD can provide a reasonably good
estimate of 5% extra risk.

This document is a draft for review purposes only and does not constitute Agency policy.

5-6 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-2. Endpoints considered for dose-response modeling and derivation
of points of departure for immune effects in humans

Endpoint

Study reference and confidence

POD
derived?

Notes

Antibody

concentrations for
diphtheria and
tetanus

Grandiean et al. (2012) [Birth cohort
1997-2000 with follow-up to age 7]
and (Grandiean et al., 2017a) [Birth
cohort 1997-2000 with follow-up to
age 131: Grandiean et al. (2017b)
[Birth cohorts from 1997-2000 &
2007-2009 with follow-up to age 5];
medium confidence

Effect was generally coherent with epidemiological
evidence for other antibody effects. However, while
these results contribute to understanding the hazard
for PFDA, the analytic models in these specific
publications used log-transformed exposure and log-
transformed outcome variables and such log-log
models cannot be used for BMD calculations and
thus PODs were not derived.

Antibody

concentrations for
diphtheria and
tetanus

Budtz-J0rgensen and Grandiean
(2018a): Birth cohorts 1997-2000 &
2007-2009 using different analyses of
combined data from Grandiean et al.
(2012) and (2017a) medium
confidence

Yes

Effect was large in magnitude and generally coherent
with epidemiological evidence for other antibody
effects. Results were based on analytic models using
log-transformed outcome and untransformed
exposure which were suitable for BMD calculations
and POD derivations (see Appendix C.l.l for more
details on BMD modeling results).

Developmental effects

Uncertainties in the human evidence of developmental effects resulted in a judgment of
slight (see Section 3.2.3); however, the database includes several well-conducted medium and high
confidence epidemiological studies reporting birth weight deficits of varying magnitude in male or
female neonates or both. Birth weight deficits (and several other developmental endpoints) were
generally larger and more consistent among studies that sampled maternal serum later in
pregnancy including postpartum measures. This suggests that those samples may be most prone to
potential bias from changing pregnancy hemodynamics, but the complex patterns of influence due
to pregnancy hemodynamics are not completely understood. Nevertheless, the apparent influence
of pregnancy hemodynamics introduces considerable uncertainty in the interpretation of these
associations of PFDA-induced developmental effects and was a major contributing factor in the
overall evidence integration judgement for this health effect (see Section 3.2.3). Despite these
concerns regarding sample timing, decreased birth weight was the focus of dose-response analysis,
given the accuracy in measurement of the endpoint, and the abundance of high-quality studies.
There is considerably less uncertainty related to pregnancy hemodynamics in studies based on
maternal serum samples collected during the first trimester.

Twenty-eight epidemiology studies (8 high and 10 medium confidence) evaluated
associations between PFDA and fetal growth restriction, including 26 studies examining mean birth
weight. Given the abundance of high confidence studies, low and medium confidence studies were
not considered for POD derivation; thus, four high confidence studies were considered as they
provided consistent evidence of associations within the overall population and across both sexes.
Among the eight high confidence studies detailed in Table 5-3, two studies Buck Louis etal. (2018):

This document is a draft for review purposes only and does not constitute Agency policy.

5-7 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Bach etal. (20161 were not considered further, as they did not find evidence of an inverse
association between PFDA exposures and mean birth weight in the overall population. Two studies
were not advanced because they reported vastly different findings across the sexes Lind et al.
f2017al: Wang etal. f20161 with no clear biological explanation for this inconsistency (see
discussion in Section 3.2.3).

Three of the four remaining studies examined PFDA during trimester three Luo et al.
(20211: Yao etal. (20211: Valvi etal. (20171 and one examined PFDA across trimesters one and two
(Wikstrom etal.. 20201.Two high confidence studies Valvi etal. (20171 and Wikstrom etal. (20201
were selected for dose-response quantification. In the (Wikstrom etal.. 20201 study, 96% of
samples were collected during the first trimester and the remaining during the early weeks of the
second trimester; sensitivity analyses showed no differences when trimester two samples excluded.
The Valvi etal. f20171 has a unique design that may increase study sensitivity by sampling all
participants during the same gestational week (i.e., 34). These two studies had a low overall risk of
bias and reliable exposure measurements with sufficient exposure contrasts (PFDA
median/interquartile ranges: 0.26/0.15 and 0.28/0.16 ng/mL, respectively for Wikstrom et al.
(20201: Valvi etal. (201711 and other characteristics that allowed for adequate study sensitivity to
detect associations (see Table 5-4). As noted above, the Valvi etal. f20171 and Wikstrom etal.
f20201 studies selected for dose-response quantification reported results consistent in magnitude
that allowed the consideration of sex-specific and overall population results. A limitation of the
Valvi etal. (20171 study advancing to dose-response is that it did not have early trimester samples
(trimester 3 only) and may be prone to some potential bias due to pregnancy hemodynamics (see
more details in Appendix F). Despite these important concerns regarding sample timing, as noted
above, derivation of a POD(s) for developmental outcomes using the Valvi, 2017 study was
considered potentially informative to toxicity value derivation for birth weight effects reported by
fWikstrom etal.. 20201.

The one available high confidence animal study that examined developmental toxicity in
mice treated with PFDA (Harris and Birnbaum. 19891 provided moderate evidence of
developmental toxicity (see Section 3.2.3). Several endpoints from this study were considered to be
suitable for POD derivation (see Table 5-5) and for comparison to PODs derived from the human
studies. Harris and Birnbaum (19891 reported developmental effects in C57BL/6N mice treated
either on GD 10-13 (0-32 mg/kg-day) or GD 6-15 (0-12.8 mg/kg-day). Harris and Birnbaum
f!9891 reported statistically significant changes for increased % resorptions per litter and
decreased number of live fetuses GD 6-15 component of the study. However, these effects were not
considered for dose-response analysis because their interpretation is confounded by overt
maternal toxicity (i.e., mortality) observed at the same dose. Statistically significant and dose-
dependent decreases in fetal body weight were also observed in both the GD 10-13 and the GD 6-
15 experiments. Data for decreased fetal body weight from the GD 6-15 experiment were
prioritized for dose-response analysis over data from the GD 10-13 experiment, since the former

This document is a draft for review purposes only and does not constitute Agency policy.

5-8 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 experiment encompasses a larger developmental window. Statistically significant and dose-

2 dependent increases in variations (i.e., delayed braincase and phalanges ossification and absence of

3 fifth sternebrae) were also reported, but there were methodological concerns and uncertainty

4 regarding the adversity of these endpoints (see Section 3.2.3) that precluded their consideration for

5 dose-response analysis.

Table 5-3. Mean Birth Weight deficit studies considered for dose-response
modeling and derivation of points of departure for developmental effects in
humans

Study reference and
confidence

Population-Overall
Population, Sex-
specific and All
Births vs. Term
Births only

PFDA
Biomarker
Sample
Timing

POD
derived?

Notes

Valvi etal. (2017):
high confidence

Overall Population;
Sex-specific; All
Births

Trimester 3

Yes

Effect was large in magnitude and coherent with findings in
mice and epidemiological evidence for other biologically
related effects (e.g., decreased postnatal growth and birth
length).

Wikstrom et al.
(2020). hiah
confidence

Overall Population;
Sex-specific; All
Births

Trimesters 1-
2

(94% in Tl)

Yes

Effect was statistically significant, large in magnitude, and
coherent with findings in mice and epidemiological evidence
for other biologically related effects (e.g., decreased postnatal
growth and birth length).

Luo et al. (2021), hiah
confidence

Overall Population;
Term Births

Trimester 3

Effect size was statistically significant and moderate in
magnitude.

Results are coherent with findings in mice and
epidemiological evidence for other biologically related effects
(e.g., preterm birth, postnatal growth, and other fetal growth
measures such as birth length).

Yao et al. (2021), hiah
confidence

Overall Population;
Sex-specific; All
Births

Trimester 3

Effect size was moderate in magnitude.

Wang etal. (2016):
high confidence

Sex-specific; Term
Births

Trimester 3

Study reported sex-specific findings that were not consistent
across male and female neonates.

Bach et al. (2016):
high confidence

Sex-specific; Term
Births

Trimester 1

Study reported sex-specific findings that were not consistent
across male and female neonates.

Buck Louis et al.
(2018). hiah
confidence

Overall Population;
Term Births

Trimester 2

Study did not detect inverse associations between mean birth
weight and PFDA.

Lind et al. (2017a),
high confidence

Sex-specific; All
Births

Trimester 1

Study reported sex-specific findings that were not consistent
across male and female neonates.

This document is a draft for review purposes only and does not constitute Agency policy.

5-9 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-4. Endpoints considered for dose-response modeling and derivation
of points of departure for developmental effects in animals

Endpoint

Study reference and
confidence

Exposure route
and duration

Test strain,
species, and sex

POD
derived?

Notes

Increased % resorptions
per litter

Harris and Birnbaum
(1989): hiah
confidence

Gavage,
GD 6-15

C57BL/6N mouse,
male and female

Effect was observed at the same
dose as significant maternal
mortality.

Decreased live fetuses
per litter

Harris and Birnbaum
(1989): hiah
confidence

Gavage,
GD 6-15

C57BL/6N mouse,
male and female

Effect was observed at the same
dose as significant maternal
mortality.

Decreased fetal body
weight

Harris and Birnbaum
(1989): medium
confidence

Gavage,
GD 10-13

C57BL/6N mouse,
male and female

Fetal body weight data from GD
10-13 was not advanced in lieu of
the more sensitive data available
from GD 6-15.

Decreased fetal body
weight

Harris and Birnbaum
(1989): medium
confidence

Gavage,
GD 6-15

C57BL/6N mouse,
male and female

Yes

Effect displayed a dose-response
trend and was coherent with other
developmental changes in mice
and humans.

Skeletal variations (i.e.,
delayed braincase
ossification; absence of
fifth sternebrae;
delayed phalanges
ossification)

Harris and Birnbaum
(1989): hiah
confidence

Gavage,
GD 6-15

C57BL/6N mouse,
male and female

The adversity and interpretation of
these effects is unclear (see
Section 3.2.3)

Male reproductive effects

The hazard conclusions for PFDA-induced male reproductive effects are driven by moderate
evidence from a single, high confidence study in rats gavaged for 28 days fNTP. 20181. The
available evidence from human studies was indeterminate (see Section 3.2.4); thus, there was no
further consideration of these human studies for POD derivation.

The single, 28-days study in adult male rats examining reproductive effects was considered
low confidence for sperm evaluations based on potential reduced sensitivity due to inadequate
exposure duration. Otherwise, the study would have been considered high confidence for sperm
measures and was considered high confidence for other, related male reproductive endpoints.

Thus, the coherent results across multiple measures, including sperm evaluations, in this well-
conducted study provide support for advancing the study for dose-response modeling. Effects in
male rats included significant decreases in testicular and epididymal sperm counts at doses
>1.25 mg/kg-day fNTP. 20181. Although there are concerns over exposure sensitivity for sperm
evaluations, the alterations in sperm counts are supported by concordant effects for histopathology
and organ weight measures in the testis and epididymis evaluated. The decreases in absolute

This document is a draft for review purposes only and does not constitute Agency policy.

5-10 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

epididymal sperm counts (but not testicular sperm counts) displayed a dose-response gradient and
thus were prioritized for POD derivation (see Table 5-5).

A consistent pattern of mild degenerative changes was detected in the testes and
epididymis of exposed rats at the two highest doses fNTP. 20181. These doses were associated with
moderate body weight decreases (21-38%) but concerns over potential confounding with overt
systemic toxicity were mitigated by mechanistic evidence suggesting that male reproductive effects
are only affected by severe changes in body weight (72%; see Mechanistic studies and
supplemental information in Section 3.2.4). Increased incidence of Leydig cell atrophy was
observed at doses >1.25 mg/kg-day, which is consistent with reductions in spermatogenesis and
serum testosterone levels reported in this same 28-day rat study and with mechanistic evidence
that suggests PFDA targets Leydig cells and disrupts steroidogenesis (see Mechanistic studies and
supplemental information in Section 3.2.4). As such, this endpointwas selected for dose-response
modeling (see Table 5-5). Other corroborative histopathological lesions (germinal epithelium
degeneration, seminiferous tubule spermatid retention, epididymal duct germ cell exfoliation and
hypospermia in the epididymis) were not advanced, as these lesions occurred mostly in the high-
dose group (2.5 mg/kg-day) and had low to medium incidence rates (10-40% compared to 0-10%
for controls). Finally, decreases in absolute testicular and epididymal weights and serum
testosterone levels identified in rats were also advanced for POD derivation. Absolute weights are
the preferred measure for testis and epididymis as these organs appeared to be conserved even
with body weight changes (Creasy and Chapin. 2018: U.S. EPA. 1996b). The changes in organ
weights and testosterone levels demonstrated a dose-response effect and were concordant with
other male reproductive findings occurring at similar doses (>1.25 mg/kg-day) (NTP. 2018).

Table 5-5. Endpoints considered for dose-response modeling and derivation
of points of departure for male reproductive effects in animals

Endpoint

Study
reference and
confidence

Exposure route
and duration

Test strain,
species, and
sex

POD derived?

Notes

Decreased testicular
sperm counts

NTP (2018): low
confidence

Gavage,
28 d

S-D rat, male

Effects provide
corroborative evidence of
male reproductive toxicity
but were not dose
dependent.

Decreased absolute
epididymis sperm
counts (cauda)

NTP (2018): low
confidence due
to concern for
potential
insensitivity

Gavage,
28 d

S-D rat, male

Yes

Effects displayed a dose-
response pattern and were
coherent with other male
reproductive findings

Leydig cell atrophy

NTP (2018): hiah
confidence

Gavage,
28 d

S-D rat, male

Yes

Effects were coherent with
other male reproductive
findings and mechanistic
evidence supporting
biological plausibility

This document is a draft for review purposes only and does not constitute Agency policy.

5-11 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

Study
reference and
confidence

Exposure route
and duration

Test strain,
species, and
sex

POD derived?

Notes

Other

histopathological
lesions in the testes
and epididymis

NTP (2018): hiah
confidence

Gavage,
28 d

S-D rat, male

Effects provide
corroborative evidence of
male reproductive toxicity
but were less sensitive
compared to other findings

Decreased serum
testosterone levels

NTP (2018): hiah
confidence

Gavage,
28 d

S-D rat, male

Yes

Effects displayed a dose-
response pattern and were
coherent with other male
reproductive system
findings

Decreased absolute
testis weight

NTP (2018): hiah
confidence

Gavage,
28 d

S-D rat, male

Yes

Decreased absolute
epididymis weight
(cauda and whole)

NTP (2018): hiah
confidence

Gavage,
28 d

S-D rat, male

Yes

Female reproductive effects

The available human evidence was judged to be indeterminate and thus these data were not
considered for dose-response analysis (see Section 3.2.5). Only one animal study (NTP. 20181
evaluated female reproductive effects due to PFDA exposure; the study was evaluated as high
confidence for all endpoints examined and provided moderate evidence for female reproductive
toxicity. The NTP f20181 study reported reproductive effects in female rats exposed to PFDA
(doses of 0, 0.156, 0.312, 0.625,1.25, and 2.5 mg/kg-day) via gavage for 28 days (see Table 5-6).
Statistically significant dose-dependent changes were observed for the number of days spent in
estrus and diestrus and for absolute and relative uterus weights; these endpoints were advanced
for POD derivation. Although Bailey etal. (2004) provided guidance on the preferred measure
(relative or absolute) for many organs (e.g., liver), both relative and absolute uterus weight were
carried forward for POD derivation because it is unclear which is the preferred measure for this
organ. Endpoints related to estrous cyclicity were also advanced for POD derivation. Under normal
conditions, the estrus stage is highlighted by sexual receptivity f Goldman etal.. 20071. PFDA was
shown to decrease the number of days spent in estrus in female rats, which could result in
decreased opportunities for mating and ultimately in reductions or delays in fertility. PFDA was
also reported to cause a continuous state of diestrus (NTP. 2018). Per the U.S. EPA's Guidelines for
Reproductive Toxicity Risk Assessment, "Persistent diestrus indicates temporary or permanent
cessation of follicular development and ovulation, and thus at least temporary infertility"; please
refer to Section 3.2.5 for a more detailed discussion. Whereas the study authors also reported
increased testosterone in female rats, this effect was not considered further because its biological
relevance to the development of PFDA-induced female reproductive toxicity is unclear.

This document is a draft for review purposes only and does not constitute Agency policy.

5-12 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-6. Endpoints considered for dose-response modeling and derivation
of points of departure for female reproductive effects in animals

Endpoint

Study reference
and confidence

Exposure route
and duration

Test strain,
species, and
sex

POD
derived?

Notes

Decreased estrus time

NTP (2018): hiah
confidence

Gavage, 28 d

S-D rat,
female

Yes

Effect displayed a dose-
response trend and was
coherent with other female
reproductive changes.

Increased diestrus
time

NTP (2018): hiah
confidence

Gavage, 28 d

S-D rat,
female

Yes

Effect displayed a dose-
response trend and was
coherent with other female
reproductive changes.

Decreased absolute
and relative uterus
weight

NTP (2018): hiah
confidence

Gavage, 28 d

S-D rat,
female

Yes

Effect displayed a dose-
response trend and was
coherent with other female
reproductive changes.

Increased
testosterone

NTP (2018): hiah
confidence

Gavage, 28 d

S-D rat,
female

The toxicological significance
of this effect in females for the
purposes of this assessment is
unclear.

Estimation or Selection of Points of Departure (PODs) for RfD Derivation

Consistent with EPA's Benchmark Dose Technical Guidance fU.S. EPA. 2012al. the BMD and
95% lower confidence limit on the BMD (BMDL) were estimated using a BMR selected to represent
a minimal, biologically significant level of change. The BMD technical guidance fU.S. EPA. 2012a)
sets up a hierarchy by which BMRs are selected, with the first and preferred approach using a
biological or toxicological basis to define what minimal level of response or change is biologically
significant If that biological or toxicological information is lacking, the BMD technical guidance
recommends alternative BMRs, specifically a BMR of 1 standard deviation (SD) from the control
mean for continuous data or a BMR of 10% extra risk (ER) for dichotomous data (see Appendix D
for more details). In cases when a biological or toxicological basis to define what minimal level of
response or change is biologically significant is lacking, a BMR of less than 1 SD is also considered
when there are concerns about the severity of the effect, or effects occur in a sensitive lifestage. The
BMRs selected for dose-response modeling of PFDA-induced health effects are listed in Table 5-7
along with the rationale for their selection.

Table 5-7. Benchmark response levels selected for BMD modeling of PFDA
health outcomes

Endpoint

BMR

Rationale

Liver effects

Increased serum enzymes in adult
rats (ALT and ALP)

1 standard deviation

No information is readily available that allows for
determining a minimally biologically significant response.

This document is a draft for review purposes only and does not constitute Agency policy.

5-13 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

BMR

Rationale

The BMD Technical Guidance (U.S. EPA, 2012a)
recommends a BMR based on 1 standard deviation (SD) for
continuous endpoints when biological information is not
sufficient to identify an appropriate BMR.

Increased relative liver weight in
adult rats

10% relative deviation

A 10% increase in liver weight is considered a minimally
biologically significant response level in adult animals and
has been used as the BMR for benchmark dose modeling in
prior IRIS assessments.

Immune effects

Decreased antibody
concentrations for diphtheria and
tetanus in children

Zi standard deviation

Diphtheria and tetanus are serious and sometimes fatal
infections. Immunomodulatory effects observed in children
may be broadly indicative of developmental
immunosuppression impacting these children's ability to
protect against a range of immune hazards. In addition,
childhood represents a sensitive lifestage. Given the
potential severity of this outcome, a BMR of both 1 SD and
Zi SD were considered (see additional discussion in
Appendix C.l.l). Ultimately, it was concluded that a BMR of
Zi SD is best supported based on the severity of the
outcome and the sensitive lifestage.

Developmental effects

Decreased birth weight in humans

5% extra risk of exceeding
adversity cutoff (hybrid
approach)

A 5% extra risk is commonly used for dichotomous
developmental endpoints as recommended by Benchmark
Dose Technical Guidance (U.S. EPA, 2012a). For birth
weight, a public health definition of low birth weight exists,
and the hybrid approach was used to estimate the dose at
which the extra risk of falling below that cut-off equaled
5%.

Decreased fetal weight in mice

5% relative deviation

A 5% change was used because the developmental effects
were observed during a sensitive lifestage. A 5% change in
markers of growth/development in gestational studies
(e.g., fetal weight) is considered a minimally biologically
significant response level and has been used as the BMR for
benchmark dose modeling in prior IRIS assessments (U.S.
EPA, 2012b. 2004. 2003).

Male reproductive effects

Increased Leydig cell atrophy in
adult rats

10% extra risk

No information is readily available that allows for
determining a minimally biological significant response. A
10% ER is recommended as the standard BMR for
dichotomous endpoints in the absence of information for a
biologically based BMR (U.S. EPA, 2012a).

Decreased epididymal sperm
counts in adult rats

1 standard deviation

No information is readily available that allows for
determining a minimally biological significant response.

Decreased serum testosterone in
adult rats

The BMD Technical Guidance (U.S. EPA, 2012a)
recommends a BMR based on 1 SD for continuous
endpoints when biological information is not sufficient to
identify an appropriate BMR.

Decreased testicular weight in
adult rats

This document is a draft for review purposes only and does not constitute Agency policy.

5-14 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

BMR

Rationale

Decreased epididymal weight in
adult rats

Female reproductive effects

Decreased estrus time in adult rats

5% relative deviation

Given that the PFDA-induced alterations in estrous cyclicity
are possible indicators of infertility, which is an outcome of
serious concern to the human population, a BMR of 5% RD
is selected for these effects. Further support for the BMR of
5% RD is provided by the large magnitude of these effects.
Specifically, PFDA induced a continuous state of diestrus in
100% of rats at the highest dose tested.

Increased diestrus time in adult
rats

Decreased absolute and relative
uterus weight in adult rats

1 standard deviation

No information is readily available that allows for
determining a minimally biologically significant response.
The BMD Technical Guidance (U.S. EPA, 2012a)
recommends a BMR based on 1 SD for continuous
endpoints when biological information is not sufficient to
identify an appropriate BMR.

Where modeling was feasible, the estimated BMDLs were used as points of departure
(PODS, see Table 5-7). Further details, including the modeling output and graphical results for the
model selected for each endpoint, can be found in Appendix C. Where dose-response modeling was
not feasible, or adequate modeling results were not obtained, NOAEL or LOAEL values were
identified based on biological rationales when possible and used as the POD. NOAELs and LOAELs
were determined based on the dose at which biologically significant changes were identified, which
takes precedence over statistical significance. For example, for relative liver weight, a 10% change
is generally viewed as a biologically significant level of change, taking into consideration the study-
specific variability. If no biological rationale for selecting the NOAEL/LOAEL is available, statistical
significance was used as the basis for selection. The PODs (based on BMD modeling or
NOAEL/LOAEL selection) for the endpoints advanced for dose-response analysis are presented in
Table 5-7.

Application of data-derived extrapolation factors for animal-human extrapolation ofPFDA
toxicological endpoints and dosimetric interpretation of epidemiological endpoints

Table 5-8 displays the POD and estimated HED PODs for liver, immune, developmental, and
male and female reproductive endpoints from animal and/or human studies selected for the
derivation of candidate values. Given that the available studies tested the free acid form of PFDA,
normalization from a salt to the free acid using a molecular weight conversion was not performed,
but formulas for providing such conversions are included in later tables.

This document is a draft for review purposes only and does not constitute Agency policy.

5-15 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-8. PODs considered for the derivation of PFDA candidate values

Endpoint

Study/
Confidence

Strain/
Species/Sex

POD
type/model

POD (mg/kg-
day)

POD internal
concentration3
(mg/L)

PODhed"
(mg/kg-day)

Liver effects

Increased AST

28-d study (NTP,
2018); hiah

SD rat, male

BMDLisd,
Hill CV

0.123

1.16 x 10"3

confidence

SD rat, female

NOAELc
(1% increase)

0.625

4.00 x 10"3

Increased ALP

SD rat, male

NOAELd
(9% increase)

0.156

1.47 x 10"3

SD rat, female

NOAELc
(14% increase)

0.156

1.00 x 10"3

Increased relative
liver weight

SD rat, male

BMDLiord,
Hill CV

0.170

1.60 x 10"3

SD rat, female

BMDLiord,
Hill CV

0.112

7.17 x 10"4

28-day study
(Frawlev et al.,
2018); high
confidence

SD rat, female
(histopathology
study cohort)

BMDLiord,
Exp2 CV

0.222

1.42 x 10"3

SD rat, female
(MPS study
cohort)

BMDLiord,
Linear CV

0.187

1.20 x 10"3

SD rat, female
(TDAR study
cohort)

NOAELc
(2% increase)

0.125

8.00 x 10"4

Immune effects (developmental)

Decreased serum
anti-tetanus
antibody
concentrations in
children at age 7
yrs and PFDA
measured at age 5
yrs

Budtz-

J0rgensen and

Grandiean

(2018a):

Grandiean et

al. (2012):

medium

confidence

Human, male and
female

BMDLi/2sd
Linear

4.11 x 10"4

1.07 x 10"s

Decreased serum
anti-diphtheria
antibody
concentrations at
age 7 yrs and
PFDA

concentrations at
age 5 yrs

Grandiean et
al. (2012):
Budtz-

J0rgensen and

Grandiean

(2018a):

medium
confidence

Human, male and
female

BMDLi/2sd
Linear

4.07 x 10"4

1.06 x 10"s

Decreased serum

anti-tetanus

antibody

Grandiean et
al. (2012):

Human, male and
female

BMDLi/2sd
Linear

7.02 x 10"4

1.83 x 10"s

This document is a draft for review purposes only and does not constitute Agency policy.

5-16 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

Study/
Confidence

Strain/
Species/Sex

POD
type/model

POD (mg/kg-
day)

POD internal
concentration3
(mg/L)

PODhed"
(mg/kg-day)

concentrations at
age 5 yrs and
perinatal
(pregnancy week
32-2 wks
postpartum) PFDA
concentrations

Budtz-

J0rgensen and

Grandiean

(2018a):

medium
confidence

Decreased serum
anti-diphtheria
antibody
concentrations at
age 5 yrs and
perinatal
(pregnancy week
32-2 wks
postpartum) PFDA
concentrations

Grandiean et
al. (2012):
Budtz-

J0rgensen and

Grandiean

(2018a):

medium
confidence

Human, male and
female

BMDLi/2sd
Linear

2.57 x 10"4

6.68 x 10"9

Developmental effects

Valvi et al.
(2017): hiah
confidence'

Human, male and
female

BMDL5RD,
Hybrid

2.8 x 10"4

7.3 x 10"9

Valvi et al.
(2017): hiah
confidence'

Human, male

BMDL5RD,
Hybrid

2.2 x 10"4

5.7 x 10"9

Decreased birth

Valvi et al.
(2017): hiah
confidence'

Human, female

BMDL5RD,
Hybrid

2.4 x 10"4

6.2 x 10"9

weight

(Wikstrom et
al., 2020): hiah
confidence5

Human, male and
femaleh

BMDL5RD,
Hybrid

3.7 x 10"4

9.6 x 10"9

(Wikstrom et
al., 2020): hiah
confidence®

Human, male

BMDL5RD,
Hybrid

3.3 x 10"4

8.6 x 10"9

(Wikstrom et
al., 2020): hiah
confidence®

Human, female

BMDL5RD,
Hybrid

3.1 x 10"4

8.1 x 10"9

Decreased fetal
body weight

Developmental
study (GD 6-15)
(Harris and
Birnbaum,
1989): medium
confidence

C57BL/6N mouse,
male and female

NOAEL
(4% decrease)

1.18 x 10"2

Male reproductive effects

Decreased cauda
epididymis sperm
count

28-d study (NTP,
2018); low
confidence

SD rat, male

BMDLisd, Exp3
CV

0.963

9.07 x 10"3

This document is a draft for review purposes only and does not constitute Agency policy.

5-17 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Endpoint

Study/
Confidence

Strain/
Species/Sex

POD
type/model

POD (mg/kg-
day)

POD internal
concentration3
(mg/L)

PODhed"
(mg/kg-day)

Increased Leydig
cell atrophy

28-day study
(NTP, 2018);
high confidence

NOAELd
(0% change)

0.625

5.89 x 10"3

Decreased serum
testosterone

NOAELd
(25% decrease)

0.625

5.89 x 10"3

Decreased
absolute testis
weight

BMDLisd,
Linear CV

1.074

1.01 x 10"2

Decreased
absolute cauda
epididymis weight

BMDLisd,
Linear CV

0.582

5.48 x 10"3

Decreased
absolute whole
epididymis weight

BMDLisd,
Linear NCV

0.546

5.14 x 10"3

Female reproductive effects

Decreased
number of days
spent in estrus

28-d study (NTP,
2018); high
confidence

BMDLsrd,
Linear CV

0.128

1.77 x 10"3

Increased number
of days spent in
diestrus

SD rat, female

BMDL5Rd, Exp2
CV

0.200

2.76 x 10"3

Decreased relative
uterus weight

NOAELc
(12% increase)

0.625

8.63 x 10"3

Decreased
absolute uterus
weight

NOAELc
(12% increase)

0.625

8.63 x 10"3

a Blood concentration PODs determined from human epidemiological analyses.

b For PODs based on animal toxicity studies, PODHED = POD x DDEF, where the DDEF is taken from Table 3-4 based
on the species, sex and endpoint being extrapolated. For POD internal concentrations (PODint; i.e., PODs from
human epidemiological studies), PODHED = POD x CLH, with CLH = 2.6 x 10-5 L/kg-d. For details, see Approach for
pharmacokinetic modeling of PFDA in rats and humans.

cNo models provided adequate fit; therefore, a NOAEL approach was selected.

dAfter visual inspection, data were not considered amenable for BMD modeling due to obvious non-monotonicity
in the dose-response; therefore, a NOAEL approach was used instead.

eHighest dose group was dropped to allow for adequate model fit.

'Trimester 3 maternal biomarker samples.

g96% of samples during the first trimester and the remaining during the early weeks of the second trimester;
sensitivity analyses showed no differences when trimester 2 samples excluded.

hSex-specific results were available for both males and females separately; these were consistent in magnitude
with the overall result.

This document is a draft for review purposes only and does not constitute Agency policy.

5-18 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Derivation of Candidate Lifetime Toxicity Values for the RfD

Under EPA's A Review of the Reference Dose and Reference Concentration Processes fU.S. EPA.
20021 and Methods for Derivation of Inhalation Reference Concentrations and Application of
Inhalation Dosimetry fU.S. EPA. 19941. five possible areas of uncertainty and variability were
considered in deriving the candidate values for PFDA. The identified potential areas of
susceptibility to PFDA exposure-induced health effects, including in children and possibly in
women of reproductive age (see Section 4.3), can help inform UF value selection and, subsequently,
confidence in toxicity values. An explanation of these five possible areas of uncertainty and
variability and the values assigned to each as a designated UF to be applied to the candidate PODhed
values are listed In Table 5-9 below. For liver and male and female reproductive effects,
quantitative information is limited to studies in which animals were exposed for <28 days. For each
of these identified hazards, very little information is available to assess the extent to which the
specific changes caused by PFDA exposure for 28 days might be expected to worsen with PFDA
exposure for a lifetime. Separately, human equivalent PODs for these endpoints were much less
sensitive (several orders of magnitude) than the PODs for developmental and immune effects from
the epidemiology studies (see Table 5-9). As such, for liver, male reproductive, and female
reproductive effects, derivation of candidate lifetime values was not attempted given the high
degree of uncertainty associated with using PODs from a 28-day rodent study to protect against
effects observed in a chronic setting. However, these endpoints were considered for the derivation
of the subchronic RfD (see Section 5.2.2).

Developmental effects observed in mice from the Harris and Birnbaum (19891 study, albeit
observed after exposure during a sensitive lifestage, were not considered for derivation of a
candidate lifetime value. Specifically, given the availability of PODs for developmental effects from
high confidence human studies that were observed to be more sensitive than the POD from the
rodent study (by 6-7 orders of magnitude; see Table 5-10), the available human data were given
preference. It is important to note that the (Valvi etal.. 20171 study was not considered for the
derivation of candidate toxicity values for developmental effects given the limitations described
above. However, the PODs determined from the (Valvi et al.. 20171studv are informative for the
PODs and resulting RfDs for developmental effects based on birth weight data from the (Wikstrom
etal.. 20201 study.

Table 5-9. Uncertainty factors for the development of the candidate lifetime
toxicity values for PFDA

Value

Justification

UFa

A UFaof 1 is applied to developmental and immunological effects observed in humans.

UFh

A UFh of 10 is applied for interindividual variability in humans in the absence of
quantitative information on potential differences in pharmacokinetics and
pharmacodynamics relating to PFDA exposure in humans.

This document is a draft for review purposes only and does not constitute Agency policy.

5-19 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Value

Justification

UFS

A UFs of 1 is applied to developmental delays (i.e., decreased birth body weight)
fWikstrom et al. (2020): and reduced antibody responses in children Grandjean et al.
(2012): Budtz-J0rgensen and Grandiean (2018a). The developmental period is
recognized as a susceptible lifestage when exposure during a time window of development
is more relevant than lifetime exposure in adulthood (U.S. EPA, 1991). Additional
considerations for the UFS for immune effects are discussed below.

UFl

A UFl of 1 is applied for LOAEL-to-NOAEL extrapolation when the POD is a BMDL or a
NOAEL. BMDLs were available for both the developmental and immune effects in the
epidemiology studies advanced for candidate value derivation.

UFd

A UFd of 3 is applied to account for deficiencies and uncertainties in the database.
Although limited, the evidence base in laboratory animals consists of high/medium
confidence short-term studies in rodents and a high confidence developmental study in
mice. The database for PFDA also includes several high/medium confidence
epidemiological studies most informative for immune and developmental effects, which
are sensitive effects of PFDA exposure. However, uncertainties remain regarding the lack
of studies examining effects with long-term exposure in adults—including in women of
reproductive age (which may have increased susceptibility), studies of potential multi-
generational effects, and studies of postnatal development, neurotoxicity, and thyroid
toxicity after PFDA exposure during development. In all, the data are too sparse to
conclude with certainty that the quantified developmental effects are likely to be the most
sensitive; thus, a UFD of 1 was not selected. However, a UFD of 10 was also not selected
give the availability of data from well-conducted studies on a range of health outcomes in
multiple species, including sensitive evaluations of developmental and immune endpoints
in humans. See discussion below for additional details.

UFC

See Table 5-10

Composite Uncertainty Factor = UFA x UFH x UFS x UFL x UFD

As described in EPA's A Review of the Reference Dose and Reference Concentration Processes
(U.S. EPA. 20021 the interspecies uncertainty factor (UFa) is applied to account for extrapolation of
animal data to humans, and accounts for uncertainty regarding the pharmacokinetic and
pharmacodynamic differences across species. The datasets considered for derivation of candidate
lifetime values were from human studies, so a UFa = 1 was applied to all PODs after the application
of dosimetric approaches for estimation of HEDs as described above.

For immune effects, both a duration extrapolation uncertainty factor (UFs) = 3 and a value
of UFs = 1 were considered to account for extrapolation from less than chronic data, ultimately
selecting a UFs = 1. A UFs=10 was not considered as the developmental period is recognized as a
susceptible lifestage for these types of effects and therefore exposure during this time window can
be considered more relevant than exposure in adulthood (U.S. EPA. 19911. The reduced antibody
responses were measured in children 5-7 years of age. The HED calculations used for these
immune effects assume chronic exposure, so an RfD based on them will assure that serum PFDA
levels remain below the POD irrespective of exposure duration. Also, development is recognized as
a sensitive period for effects on immune system responses. According to the WHO/IPCS
Immunotoxicity Guidance for Risk Assessment, developmental immunotoxicity encompasses the
prenatal, neonatal, juvenile and adolescent lifestages and should be viewed differently from the
immune system of adults from a risk assessment perspective (IPCS. 20121. Special considerations

This document is a draft for review purposes only and does not constitute Agency policy.

5-20 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

for developmental immunotoxicity include increased dose sensitivity, potential for effects to
become permanent even after cessation of exposure, broader spectrum of adverse effects and
"rewiring of the immune system" flPCS. 20121. which indicates a greater health risk for early-life
exposures to immunotoxicants compared to adults. Given PFDA's long half-life and the expectation
that the children and their mothers have been exposed to elevated levels of PFDA for many years,
the observed effects on immune response are considered to be the result of a cumulative, prolonged
exposure to the subjects from conception until the age when the response was evaluated. Further,
the consequences of perturbed immune system function (in this case, suppressed antibody
responses leading potentially to increased disease) during development are expected to be
generally more severe and longer lasting than those that manifest in healthy adults. Taken
together, the observed immune effects in children considered to be the result of prolonged
exposure to PFDA and the enhanced susceptibility of the developmental immune system to
chemical pollutants, attenuate concerns of potentially increased sensitivity with longer-term
exposures. As such, a UFs =1 rather than a UFs = 3 was applied for immune effects in children.
Uncertainties regarding possible more sensitive latent effects of these impacts on the immune
system during early-life exposures leading to unpredictable outcomes later in life, for example in
other susceptible lifestages of reduced immunocompetence such as pregnancy and most notably
old age, are addressed as part of the justification for selecting a database uncertainty factor (UFd) >
1, as discussed below.

For PFDA, both a UFd = 10 and a UFd = 3 were considered due to the limited database
(e.g., the lack of a two-generation developmental/reproductive toxicity study) and a UFd = 3
ultimately was applied. Typically, the specific study types lacking in a chemical's database that
influence the value of the UFd to the greatest degree are developmental toxicity and
multigenerational reproductive toxicity studies. The PFDA database does include a medium
confidence fHarris and Birnbaum. 19891 developmental toxicity study in mice. Despite its quality,
however, that study fails to cover potential transgenerational impacts of longer-term exposures
evaluated in a two-generation study. The 1994 Reference Concentration Guidance (U.S. EPA. 19941
and 2002 Reference Dose Report (U.S. EPA. 20021: (U.S. EPA. 20021 support applying a UFd in
situations when such a study is missing. The 2002 Reference Dose Report (U.S. EPA. 20021: (U.S.
EPA. 20021 states that "[i]f the RfD/RfC is based on animal data, a factor of 3 is often applied if
either a prenatal toxicity study or a two-generation reproductive study is missing." Consideration
of the PFDA, PFBA (a short-chain perfluoroalkyl carboxylic acid),1617 PFBS (a short-chain

16The systematic review protocol for PFDA (see Appendix A) defines perfluoroalkyl carboxylic acids with
seven or more perfluorinated carbon groups and perfluoralkane sulfonic acids with six or more
perfluorinated carbon groups as 'long-chain" PFAS. Thus, PFHxA and PFBA are considered short-chain PFAS,
whereas PFHxS is considered a long-chain PFAS.

17IRIS Toxicological Review of Perfluorobutanoic Acid (PFBA, CASRN 375-22-4) and Related Salts (U.S. EPA.
20221.

This document is a draft for review purposes only and does not constitute Agency policy.

5-21 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

perfluoroalkane sulfonic acid with a 4-carbon backbone),18 PFHxA (a short-chain perfluoroalkyl
carboxylic acid), and PFHxS (a long-chain perfluoroalkane sulfonic acid)19 databases together,
however, diminish the concern that the availability of a multigenerational reproductive study
would result in reference values far below those currently derived for PFDA. Although limited in
their ability to assess reproductive health or function, measures of possible reproductive toxicity
occurred at doses equal to or higher than those that resulted in effects in other organ systems
(e.g., thyroid, liver) when measured after exposure to PFDA for 28 days (NTP. 2019). Similar
results were observed for the animal databases for PFOA and PFOS indicating reproductive effects
were not uniquely sensitive markers of toxicity for these long-chain PFAS fATSDR. 2018b). Further,
no notable male or female reproductive effects were observed in epidemiological or toxicological
studies investigating exposure to PFHxS fMDH. 20191. Therefore, considering the limited chemical-
specific information alongside information gleaned from structurally related compounds, the lack
of a multigenerational reproductive study is not considered a major concern relative to UFd
selection for PFDA.

The lone animal developmental study (Harris and Birnbaum. 1989) for PFDA also did not
evaluate postnatal developmental effects. Effects on postnatal development (e.g., delayed eye
opening; reduced postnatal growth) have been observed in rodents exposed to other long-chain
PFAS such as PFOA fATSDR. 2018bl. Overall, the available information on potential PFDA-induced
postnatal developmental effects is sparse, introducing uncertainty as to whether more sensitive
developmental effects of PFDA might occur and may be of concern relative to UFd selection.

Another gap in the PFDA database is the lack of measures of thyroid toxicity in gestationally
exposed offspring or after longer-than-28-day PFDA exposures, and the lack of a developmental
neurotoxicity study. Thyroid hormones are critical in myriad physiological processes and must be
maintained at sufficient levels during times of brain development in utero and after birth. Although
no PFDA-specific data on thyroid hormone levels following gestational exposure are available,
effects on thyroid hormone homeostasis were observed in a study in adult rats exposed to PFDA for
28 days (NTP. 2018). and disrupted thyroid signaling has been shown to be a consequence of
exposure to other PFAS (U.S. EPA. 2021b). Therefore, anticipating that potentially sensitive effects
due to PFDA exposure also could have been observed had thyroid hormone levels been measured in
the Harris and Birnbaum (1989) developmental study, or in longer-term studies, is reasonable.
Thus, the lack of data for PFDA-induced effects on thyroid levels in developing animals or with
prolonged exposure or data on potential thyroid dependent neurodevelopmental effects is a source
of uncertainty.

18 Human health toxicity values for perfluorobutane sulfonic acid (CASRN 375-73-5) and related compound
potassium perfluorobutane sulfonate (CASRN 29420-49-3"lfU.S. EPA. 2021b"!

19 Health Based Guidance for Water: Toxicological Summary for: Perfluorohexane sulfonate (PFHxS), MDH
(2019)

This document is a draft for review purposes only and does not constitute Agency policy.

5-22 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Lastly, the potential for sensitive effects following long-term exposure durations represents
an area of uncertainty for the PFDA database. While the potential for more sensitive effects is
mitigated mostly by the availability of very sensitive PODs (compared to other PODs) for
developmental effects from human studies, there are no comprehensive subchronic and chronic
animal studies available for PFDA. The longest exposure study treated mice for 30-49 days via
drinking water but tested only one high-PFDA dose (6.6 mg/kg-day) and evaluated limited
endpoints (body weight and survival) (Wangetal.. 20201. No chemical-specific information is
available to judge the degree to which the existing endpoints in the PFDA Toxicological Review
would be more sensitive with extended durations. Given that the PODs used to derive candidate
values were from studies of developmental exposure, this uncertainty cannot be fully addressed
through the application of a UFs. Specifically, for immune effects, there is a lack of epidemiological
studies or studies in animals examining the effects of PFDA exposures that encompass later
developmental periods (e.g., late childhood and adolescence) or other potentially susceptible
lifestages such as pregnancy and old age. In addition, the available studies include limited or no
evaluation of immunotoxicity categories other than immunosuppression, namely sensitization and
allergic response, and autoimmunity and autoimmune disease.

Given the residual concerns for potentially more sensitive effects outlined above, a database
uncertainty factor is considered necessary. Specifically, a value of 3 was selected for the UFd to
account for the uncertainty surrounding the lack of an evaluation of postnatal or multigenerational
effects in animals, specific investigations of potential effects on thyroid function after
developmental exposure or neurodevelopmental effects, and comprehensive long-term studies in
multiple species.

The uncertainty factors described in Table 5-9 and the text above were applied and the
resulting candidate values are shown in Table 5-10. The candidate values are derived by dividing
the PODhed by the composite uncertainty factor as shown below.

Candidate values for PFDA= PODhed^-UFc

This document is a draft for review purposes only and does not constitute Agency policy.

5-23 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-10. Candidate values for PFDA

Endpoint

Study/
Confidence

Strain/
Species/
Sex

PODhed
(mg/kg-d)

UFa

UFh

UFs

UFl

UFd

UFC

Candidate

value
(mg/kg-d)a

Immune effects (developmental)

Decreased serum
anti-tetanus antibody
concentration in
children at age 7 yrs
and PFDA measured
at age 5 yrs

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.07 x 10"s

4 x 10"10

Decreased serum
anti-diphtheria
antibody levels at age
7 yrs and PFDA
concentrations at age
5 yrs

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.06 x 10"s

4 x 10"10

Decreased serum
anti-tetanus antibody
levels at age 5 years
and perinatal
(pregnancy week 32-
2 wks postpartum)
PFDA concentrations

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.83 x 10"s

6 x 10"10

Decreased serum
anti-diphtheria
antibody levels at age
5 yrs and perinatal
(pregnancy week 32-
2 wks postpartum)
PFDA concentrations

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

6.68 x 10"9

2 x 10"10

Developmental effects

Decreased birth
weight

(Wikstrom et al.,
2020) high
confidence

Human,
male and
female

9.6 x 10"9

3 x 10"10

(Wikstrom et al.,
2020) high
confidence

Human,
male

8.6 x 10"9

3 x 10"10

(Wikstrom et al.,
2020) high
confidence

Human,
female

8.1 x 10"9

3 x 10"10

aThe candidate values for different salts of PFDA would be calculated by multiplying the candidate value for the
free acid of PFDA by the ratio of molecular weights. For example, for the ammonium salt the ratio would be:
mw ammonium salt _ 531 _ ^ Q33 same method of conversion can be applied to other salts of PFDA, such as

Jl/fT \7 fvnn rt ^ IS-} C1 A 11 '

the potassium or sodium salts, using the corresponding molecular weights.

This document is a draft for review purposes only and does not constitute Agency policy.

5-24 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

5.2.2. Selection of Lifetime Toxicity Value(s)

Selection of organ/system-specific oral reference doses (osRfDs)

From among the candidate values presented in Table 5-10, organ/system-specific RfDs
(osRfDs) are selected for the individual organ systems identified as hazards in Section 3. The osRfD
values selected were associated with decreased serum antibody concentrations in children for
immune effects and decreased birth weight for developmental effects. The confidence decisions
about the studies, evidence base, quantification of the POD, and overall osRfD are fully described in
Table 5-11, along with the rationales for selecting those confidence levels. In deciding overall
confidence, confidence in the evidence base is prioritized over the other confidence decisions. The
overall confidence in the osRfD for immune effects is medium, and the confidence in the osRfD for
developmental effects is medium-low. Selection of the overall RfD is described in the following
section.

Table 5-11. Confidence in the organ/system-specific (osRfDs) for PFDA

Confidence
categories

Designation

Discussion

Immune (developmental) osRfD = 4 x io~10 mg/kg-d

Confidence in
study3 used to
derive osRfD

High

Confidence in Grandiean et al. (2012): Budtz-J0rgensen and Grandiean (2018a) was
rated as medium primarily due to relatively limited PFDA exposure contrasts, which can
decrease study sensitivity in general. (HAWC link). Given that the results in this study were
statistically significant, EPA concluded that while there were potential study sensitivity
concerns at the evaluation stage, the results clearly showed that those concerns were not
borne out, and confidence in this study to derive an osRfD was judged to be high.

Confidence in
evidence base
supporting this
hazard

Medium

Confidence in the evidence base for immune effects is medium based on consistent findings of
reduced antibody responses from two medium confidence birth cohort studies (Grandiean et
al., 2012): (Grandiean et al., 2017a): (Grandiean et al., 2017b) and a low confidence study
in adults (Kielsen et al., 2016). Short-term studies in animals of high/medium confidence
provide supportive evidence of immunosuppression after PFDA exposure (Frawlev et al.,
2018): (NTP, 2018). Some residual uncertainties regarding unexplained inconsistency and
potential confounding by other co-occurring PFAS from epidemiological studies and issues with
concomitant overt target organ and systemic toxicity in animal studies lower confidence in the
available evidence for this hazard. Other limitations include the lack of epidemiological studies
or long-term/chronic studies in animals examining effects on the immune system across
different developmental life stages and immunotoxicity categories, including sensitization and
allergic response and autoimmunity and autoimmune disease.

This document is a draft for review purposes only and does not constitute Agency policy.

5-25 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Confidence
categories

Designation

Discussion

Confidence in
quantification of
the PODhed

Medium

Confidence in the quantification of the POD and osRfD is medium. The POD is based on BMD
modeling at the lower end of the range of the observed data and a BMDLi/2sd estimate that is
associated with a small degree of uncertainty due to potential confounding by PFOA (see
Appendix D.l.l for more details). The POD for decreased tetanus antibodies at age 7 yrs was
judged to be medium confidence based on a good model fit and was supported by the nearly
identical POD for decreased diphtheria antibodies at age 7 yrs. Both PODs support the osRfD.
An estimate for human clearance was applied to estimate the PODhed using PFDA-specific
pharmacokinetic information, the latter of which involves some residual uncertainty (see
discussion on Uncertainty in the pharmacokinetic modeling of PFDA above). There is also
uncertainty as to the most sensitive window of vulnerability with respect to the
exposure/outcome measurement timing (BMDs/BMDLs were estimated from PFDA levels
measured at age 5 or perinatally and anti-tetanus antibody concentrations measured at age 7
or 5): (Grandiean et al., 2017b) reported that estimated PFDA "concentrations at 3 mo and 6
mo showed the strongest inverse associations with antibody concentrations at age 5 yrs,
particularly for tetanus." Thus, it is possible that adverse effects during infancy could be more
sensitive than between ages 5 and 7 yrs.

Overall
confidence in
osRfD

Medium

The overall confidence in the osRfD is medium and is driven by medium confidence in the
evidence base for immune effects, the quantification of the POD, and the study used for BMD
modeling.

Developmental osRfD = 3 x 10~10 mg/kg-d

Confidence in
study3 used to
derive osRfD

Medium

Confidence in the Wikstrom et al. (2020) study for hazard identification was rated as hiah
(HAWC link) for developmental effects. The study was selected for dose-response analysis due
to low overall risk of bias and reliable exposure measurements which had sufficient exposure
contrasts and other characteristics that allowed for adequate study sensitivity to detect
associations. The Wikstrom et al. (2020) study demonstrated associations consistent in
magnitude for boys, girls, and the overall population. Overall, mean birth weight was
considered the most precise and accurate endpoint and not anticipated to be subject to much
error. This study was advanced for dose-response analysis, given no presumed impact of
pregnancy hemodynamics given the early sampling (96% from trimester 1). Wikstrom et al.
(2020) also adjusted for sample timing in their multivariate models and show no differences in
models also restricted to trimester 1 samples only. Some uncertainty remains on the potential
for confounding by other PFAS (concern primarily for PFNA) which were not examined in this
study. Given the potential quantitative impact of this uncertainty, confidence in the use of this
study for dose-response analysis was judged as medium rather than high.

Confidence in
evidence base
supporting this
hazard

Medium-low

Confidence in the evidence base for developmental effects is medium. There was consistent
evidence for reduced birth weight among multiple human studies, including high quality
studies. However, unlike the Wikstrom et al. (2020) study used here and noted above, some
uncertainty remains in many studies given the predominance of associations that were
detected for studies with later pregnancy sampling. The human database also showed some
coherence across different measures of fetal growth restriction. In animals, the lone
developmental study reported effects on fetal growth that are coherent with effects observed
in humans. Some residual uncertainty regarding potential confounding by other co-occurring
PFAS from epidemiological studies lowers confidence in the available evidence for this hazard.

Confidence in
quantification of
the PODHED

Medium

Confidence in the quantification of the POD and osRfD is medium given the POD was based on
a BMD hybrid approach within the range of the observed data and dosimetric adjustment was
based on PFDA-specific pharmacokinetic information, the latter of which involves some
residual uncertainty (see discussion on Uncertainty in the pharmacokinetic modeling of PFDA
above).

Overall
confidence in
osRfD

Medium-low

The overall confidence in the osRfD is medium-low and is driven by medium-low confidence in
the evidence base for developmental effects (i.e., fetal growth restriction).

This document is a draft for review purposes only and does not constitute Agency policy.

5-26 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

aAII study evaluation details can be found on HAWC.

Selection of overall oral reference dose (RfD) and confidence statement

Organ/system-specific and overall RfD values for PFDA selected in the previous section are
summarized in Table 5-12.

Table 5-12. Organ/System-specific and overall lifetime RfDs for PFDA

System

Toxicity Value

Basis

PODhed
(mg/kg-d)

UFC

osRfD or RfD
(mg/kg-d)

Confidence

Immune
(developmental)

osRfD

Decreased
antibody
concentrations
for both tetanus
and diphtheria in
children at age 7
yrs and PFDA
measured at age
5 yrs

1.07 x 10-S based
on BMDLy2sDfrom
Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a)

4 x 10"10

Medium

Developmental

osRfD

Decreased birth
weight in males
and females

9.6 x 10"9 based
on BMDI_5%rd
from (Wikstrom
etal., 2020)

3 x 10"10

Medium-low

Immune
/developmental

Overall lifetime
RfD

Decreased
antibody
concentrations
for both tetanus
and diphtheria in
children at age 7
yrs and PFDA
measured at age
5 yrs

Decreased birth
weight in males
and females

1.07 x 10_s based
on BMDLy2sDfrom
Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a)

9.6 x 10"9 based
on BMDI_5%rd
from (Wikstrom
etal., 2020)

4 x 10"10

Medium

From the identified human health effects of PFDA and derived osRfDs for immune and
developmental effects (see Table 5-12), an overall RfD of 4 x 10~w mg/kg-day based on decreased
serum antibody concentrations and decreased birth weight in humans was selected. As
described in Table 5-12, confidence in the RfD is medium, based on medium confidence in the
immune osRfD (the developmental osRfD was medium-low confidence), noting that there was
medium confidence in the quantification of the PODs for both immune (Budtz-l0rgensen and
Grandiean. 2018al: f Grandiean etal.. 20121 and developmental fWikstrom etal.. 20201 endpoints
using BMD modeling. This RfD is considered to be representative of both immune and
developmental effects given the close proximity (~1.5-fold) of the developmental and immune

This document is a draft for review purposes only and does not constitute Agency policy.

5-27 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

PODs and resulting osRfDs and that both critical effects are observed during the developmental
period. There is a slight difference in the immune and developmental osRfDs due to numerical
rounding in the RfD calculation (immune osRfD = 1.07 x 10_8/30 = 3.6 x 10"10= 4 x 10"10;
developmental osRfD = 9.6 x 10_9/30 = 3.2 x 10"10 = 3 x 10-10). Although the value associated with
the immune osRfD is slightly higher than the developmental osRfD, this value is chosen as the
representative overall RfD given that it is higher confidence (medium vs medium-low for the
developmental osRfD) and is considered appropriate given the definition of the RfD being a value
with uncertainty of up to an order of magnitude. Selection of this overall RfD is presumed to be
protective of all other potential health effects in humans, based on the currently available evidence.
Finally, the immune osRfD and developmental osRfD are based on effects observed in males and
females indicating that the overall RfD would be protective for both sexes.

Overall, the immune and developmental endpoints from epidemiological studies of PFDA
were preferentially advanced for the derivation of candidate lifetime values. For immune effects,
osRfDs were derived for decreased serum antibody levels (for both diphtheria and tetanus) in
children (male and female) at different timing of exposure and outcome measurement
combinations, specifically antibody levels at age 7 and PFDA concentrations at age 5, and antibody
levels at age 5 and perinatal PFDA concentrations fBudtz-largensen and Grandiean. 2018al (see
Table 5-8). The toxicity value (osRfD) for immune effects of 4 x 10"10 mg/kg-day was based on
deleterious effects observed in children showing decreased antibody concentrations for both
tetanus and diphtheria at age 7 years related to serum PFDA concentrations measured at age 5
years. The PODs for decreased tetanus and diphtheria antibody concentrations were nearly
identical (BMDLi/2sd[hed] of 1.07 x 10 8 mg/kg-day for tetanus and 1.06 mg/kg-day for diphtheria)
and were close to the PODs for other outcome-exposure combinations (see Table 5-10), which
further supports the selected osRfD. Although both tetanus and diphtheria are rare in the U.S., the
findings that PFDA exposure reduced antibody responses may be broadly indicative of
developmental immunosuppression impacting overall immune function in these children. The
lowest serum PFDA concentration measured at age 5 years was 0.05 ng/mL and the 10th% was 0.2
ng/mL (Grandiean and Bateson. 2021) so the estimated BMDy2sD (0.411 ng/mL) for this endpoint in
the single PFAS model is well within the observed range. No information was available to judge the
fit of the model in the range of the BMDLs (see Appendix C.l.l for more details).

For developmental effects, given that the candidate toxicity values are identical (see Table
5-10), the osRfD of 3 x 10"10 mg/kg-day (BMDL5RD[HED] of 9.6 x 10 9 mg/kg-day) based on
reduced birth weight in males and females from the Wikstrom etal. f20201 study was selected.
Although this osRfD is not based on the lowest POD for reduced birth weight from the (Wikstrom et
al.. 2020) study, it is more representative of the general human population (males and females
combined) than the comparisons in males or females only. There is some uncertainty with PODs
considered from the Valvi etal. f20171 study because it is not based on early sampling and may be
prone to bias from pregnancy hemodynamics to some unknown degree. As discussed in Appendix F,

This document is a draft for review purposes only and does not constitute Agency policy.

5-28 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

there is only one developmental study (Gvllenhammar etal.. 20181 for PFDA that collected and was
able to analyze maternal hemodynamics data such as GFR and/or albumin. This study did not
report any evidence of confounding following statistical adjustment of different GFR measures for
any of the PFAS examined, which is consistent with no demonstrated confounding by either GRR
fManzano-Salgado etal.. 20171: fWhitworth etal.. 20121 or albumin fSagiv etal.. 20181 for other
PFAS examined in other studies. However, existing meta-analyses for both PFOA fSteenland etal..
20181 and PFOS (Dzierlenga etal.. 20201 only detected birth weight deficits for later trimester
sampling (e.g., beyond trimester 1). A similar detailed analysis was precluded for PFDA given that
there are only two studies that examined any first trimester measures. Overall, there was limited
evidence of any patterns of larger birth weight associations with sample timing for PFDA, but
possible associations could not be evaluated further given limited available data as well as
disparate exposure measures, distributions, and contrasts being examined. In contrast, the
Wikstrom etal. (20201 study was prioritized for RfD derivation as it was a high confidence study
that sampled maternal plasma in the first and second trimester thereby reducing uncertainty
relating to pregnancy hemodynamics. Further confidence in the osRfD derived from the (Wikstrom
etal.. 20201 study is provided by the fact that the PODs from the (Wikstrom etal.. 20201 and (Valvi
etal.. 20171 studies are relatively close (see Table 5-8 above). While not presented in this
Toxicological Review, additional birth weight studies were BMD modeled to provide a sensitivity
analysis for the comparison of birth weight effects; please see Table C-8 of the Supplemental
Appendices. These studies are either medium confidence and/or have later trimester sampling and
thus not considered in the dose-response analysis. The PODs from these birth weight studies are
relatively close (varying by ~3-fold), providing further confidence in using the POD from the
(Wikstrom etal.. 20201 study for RfD derivation. In addition to the quantitative implications, the
close proximity of the BMDLs from a multitude of birth weight studies increases the confidence in
deriving osRfDs despite slight evidence of developmental effects in humans.

5.2.3. Subchronic Toxicity Values for Oral Exposure (Subchronic Oral Reference Dose [RfD])

Derivation

In addition to providing an RfD for lifetime exposure in health systems, this document also
provides an RfD for less-than-lifetime ("subchronic") exposures. Datasets considered for the
subchronic RfD were based on endpoints advanced for RfD derivation in Table 5-8. Given thatthe
developmental and immune effects were observed in humans exposed to PFDA during susceptible
lifestages (postnatal growth/development and immune system effects in children at ages 5-7),
these endpoints were also considered for the derivation of candidate subchronic values, applying
identical uncertainty factors to those used for the lifetime candidates values (see Table 5-14 below).

Similar to the derivation of the lifetime RfD, the developmental effects observed in mice
from the Harris and Birnbaum f!9891 study were not advanced for the derivation of candidate
subchronic values. The developmental PODs from human studies are 6-7 orders of magnitude more
sensitive than the POD from the rodent study (see Table 5-9), and were, therefore, prioritized. In

This document is a draft for review purposes only and does not constitute Agency policy.

5-29 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 addition, endpoints for hepatic, male reproductive toxicity, and female reproductive toxicity

2 observed in the 28-day rodent study fNTP. 20181 were considered for the derivation of subchronic

3 toxicity values. As compared to the large uncertainty in extrapolating the available 28-day studies

4 to lifetime PFDA exposure in the context of the RfD, it was considered reasonable to try to

5 extrapolate the 28-day study data for the purposes of deriving subchronic candidate values.

6 The use of animal data for hepatic, male reproductive, and female reproductive endpoints

7 required the application of different uncertainty factors than those used for developmental and

8 immune effects in humans and can be found in Table 5-13.

Table 5-13. Uncertainty factors for the development of the candidate
subchronic values for PFDA

Value

Justification

UFa

A UFa of 1 is applied to developmental and immunological effects observed in
epidemiological studies.

A UFa of 3 is applied to account for uncertainty in characterizing the pharmacokinetic
and pharmacodynamic differences between mice or rats and humans following oral
PFDA exposure. Aspects of the cross-species extrapolation of pharmacokinetic
processes have been accounted for by using a DDEF to convert internal doses in rodents
to administered doses in humans using evidence on clearance; however, some residual
pharmacokinetic uncertainty remains as does the potential for pharmacodynamic
differences. Availability of chemical-specific data justify the selection of a UF of 3 for
PFDA. See discussion below for more details.

UFh

UFS

A UFS of 1 is applied to developmental delays (i.e., decreased birth body weight)
Wikstrom et al. (2020): and reduced antibody responses in children (Budtz-
J0rgensen and Grandiean, 2018a): (Grandiean et al., 2012).The developmental
period is recognized as a susceptible life stage when exposure during a time window of
development is more relevant than subchronic exposure (U.S. EPA, 1991).

A UFS of 10 is applied to liver, male reproductive, and female reproductive effects in
adult animals (increased AST levels, decreased epididymis weight and decreased
number of days in estrus, respectively) because of the short exposure duration (28 d)
and the presumption that effects would worsen with longer exposures. See discussion
below for more details.

UFl

A UFl of 1 is applied for LOAEL-to-NOAEL extrapolation when the POD is a BMDL or a
NOAEL. All PODs considered for candidate subchronic values were BMDLs.

This document is a draft for review purposes only and does not constitute Agency policy.

5-30 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Value

Justification

UFd

A UFd of 3 is applied to account for deficiencies and uncertainties in the database.
Although limited, the evidence base in laboratory animals consists of high/medium
confidence short-term studies in rodents and a high confidence developmental study in
mice. The database for PFDA also includes several high/medium confidence
epidemiological studies most informative for immune and developmental effects.
However, uncertainties remain regarding the lack of studies of potential multi-
generational effects, and studies of postnatal development, neurotoxicity, and thyroid
toxicity during developmental lifestages. In all, the data are too sparse to conclude with
certainty that the quantified developmental effects are likely to be the most sensitive;
thus, a UFd of 1 was not selected. However, a UFD of 10 was also not selected give the
availability of data from well-conducted studies in multiple species, including
developmental and short-term rodent studies examining a range of potentially sensitive
health outcomes and sensitive evaluations of developmental and immune endpoints in
humans.

UFC

See Table 5-11 and
Table 5-15

Composite Uncertainty Factor = UFA x UFH x UFS x UFL x UFD

As described above under Derivation of Candidate Lifetime Toxicity Values for the RfD, and in
fU.S. EPA. 20021. five possible areas of uncertainty and variability were considered in deriving the
candidate subchronic values for PFDA. In general, the explanations for these five possible areas of
uncertainty and variability and the values assigned to each as a designated UF to be applied to the
candidate PODhed values are listed above and in Table 5-13, including the UFd which remained at 3
due to data gaps discussed previously in the derivation of the lifetime RfD. One UF that differs
between subchronic and chronic RfDs is that for effects (i.e., decreased fetal body weight, increase
AST levels, decreased whole epididymis weight and decreased estrus time) observed in rodents a
UFAof 3 was applied to account for pharmacokinetic and pharmacodynamic differences between
rodents and humans following oral PFDA exposure. As is usual in the application of this
uncertainty factor, the pharmacokinetic uncertainty is mostly addressed through the application of
an adjustment factor, in this case, chemical-specific dosimetric data for estimating human
equivalent doses (see Approach for pharmacokinetic extrapolation of PFDA among rats, mice, and
humans). This leaves some residual uncertainty around the pharmacokinetics and the uncertainty
surrounding differences in pharmacodynamic differences between animals and humans. Typically,
a UFa of 3 is applied for this uncertainty when either BW3/4 scaling or chemical-specific information
is used for dose extrapolation. This is the case for developmental, male reproductive and female
reproductive endpoints. For the liver endpoint, available mechanistic and supplemental
information is considered further in determining the most appropriate value for the UFa to account
for the uncertainty.

Evidence from in vitro studies suggest that PFDA interacts with several human receptor
pathways relevant to its mechanism of hepatotoxicity, including PPARa. PFDA can bind and
activate PPARa in vitro, but reduced sensitivity towards the human PPARa versus other
mammalian isoforms (i.e., mouse and Baikal seal) is apparent flshibashi et al.. 20191:fWolf et al..
20121:fWolf etal.. 20081 and similar findings have been demonstrated for some other

This document is a draft for review purposes only and does not constitute Agency policy.

5-31 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

perfluorinated compounds. If PPARa were the only operant MOA for noncancer effects in the liver,
this observation might support reducing the remaining portion of the UFa to 1, as it could be argued
that humans are not as sensitive as wild-type rats to the hepatic effects of PFDA exposure (note:
without evidence to the contrary, as mentioned in the previous paragraph, the toxicodynamic
portion of this UF is typically assigned a value of 3 assuming responses manifest in humans could
be more sensitive than those observed in animals). Although PPARa appears to be an important
mechanism of PFDA-induced liver toxicity in animals and reduced sensitivity in PPAR activation in
humans compared to rodents has been suggested, available evidence for PFDA in PPARa null mice,
human in vitro assays and in vivo animal models more relevant to humans with respect to PPARa
sensitivity (i.e., guinea pigs and Syrian hamsters) suggest that liver effects occur, at least in part,
independent of PPARa (see Summary of mechanistic studies for PFDA in Section 3.2.1). A plausible
PPARa-dependent and independent MOA for liver effects is also supported by studies in null and
humanized animal models of structurally related long-chain PFAS [C7-C9] (see Evidence for other
PFAS in Section 3.2.1), which are mostly lacking for PFDA (a few studies in null mice but no
humanized models). Considering the remaining uncertainty in additional MOAs that appear active
in PFDA-induced liver effects, and the relative contribution of these MOAs to toxicity in humans as
compared to rodents, uncertainties surrounding a potential multifaceted MOA for PFDA-induced
liver effects, the value of 3 was selected for the UFa for the purposes of deriving candidate
subchronic toxicity values for hepatic effects.

EPA states that for "short-term and longer-term reference values, the application of a UF
analogous to the subchronic-to-chronic duration UF also needs to be explored, as there may be
situations in which data are available and applicable, but they are from studies in which the dosing
period is considerably shorter than that for the reference value being derived" (U.S. EPA. 2002).

This is the case for hepatic, male reproductive and female reproductive endpoints derived from the
28-dav NTP f20181 study. Although there is no chemical-specific information to evaluate the
potential for increased sensitivity with exposures longer than 28-days (e.g., a 90-day subchronic
study), the following considerations are outlined to inform the application of the UFs for duration
extrapolation. (U.S. EPA. 2002)

With regards to female reproductive toxicity, PFDA-induced effects on estrous cyclicity
were observed to be of large magnitude in the 28-day study. Specifically, PFDA induced a
continuous state of diestrus in 100% of rats treated at the highest dose tested (2.5 mg/kg-day) by
Day 21 (by Day 9 of the sixteen days in which vaginal cytology was assessed) fNTP. 20181. Based
on these data, it is possible that PFDA-induced effects on estrous cyclicity could become more
sensitive or lead to more severe downstream effects like infertility with longer exposure durations.
For male reproductive effects, the study duration (28 days) was insufficient to cover the entire
period of spermatogenesis in rats (~8 weeks), raising concerns about reduced sensitivity for some
of the endpoints evaluated and selected for POD derivation (i.e., sperm evaluations). For liver
effects, increases in relative liver weights demonstrated a time dependency across short-term

This document is a draft for review purposes only and does not constitute Agency policy.

5-32 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

exposures. Relative liver weight increased by 17-56% at 1.15-10 mg/kg-day in rats exposed for 7-
14 days and by 12-127% at 1-16 mg/kg-day in mice exposed during gestion (GD 10-13 and 6-15).
Similar magnitudes of liver weight increases were achieved in rodents after 28-day exposure but at
lower PFDA doses (10-102% at 0.125-2.5 mg/kg-day in rats and 16-81% at 0.089-0.71 mg/kg-day
in mice). The limited data for liver weight suggest potential increase in sensitivity with increasing
duration, although there is no information on how liver weight or other sensitive liver endpoints
(increased AST and ALP levels) are impacted by longer-term exposures (>28 days). Considering the
potential for some health effects (prolonged diestrus, sperm measures and increased liver weight)
to worsen with increasing duration and the large uncertainty associated with the lack of any
chemical-specific data on whether the effects observed in the short-term study worsen after
subchronic exposure, a UFs of 10 is selected for the purposes of deriving candidate subchronic
toxicity values from the 28-day toxicity data.

The uncertainty factors described in Table 5-13 and the text above were applied and the
resulting candidate subchronic values are shown in Table 5-14. The candidate values are derived
by dividing the PODHED by the composite uncertainty factor as shown below.

Candidate values for PFDA = PODhed ^ UFc

This document is a draft for review purposes only and does not constitute Agency policy.

5-33 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-14. Candidate values for deriving the subchronic RfD for PFDA

Endpoint

Study/
Confidence

Strain/
Species/
Sex

PODhed
(mg/kg-d)

UFa

UFh

UFs

UFl

UFd

UFC

Candidate

value
(mg/kg-d)a

Immune effects (developmental)

Decreased serum
anti-tetanus antibody
concentrations in
children at age 7 yrs
and PFDA measured
at 5 yrs

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.07 x 10"s

4 x 10"10

Decreased serum
anti-diphtheria
antibody
concentrations at
age 7 yrs and PFDA
concentrations at
age 5 yrs

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.06 x 10"s

4 x 10"10

Decreased serum
anti-tetanus antibody
concentrations at
age 5 yrs and
perinatal (pregnancy
week 32-2 wks
postpartum) PFDA
concentrations

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

1.83 x 10"s

6 x 10"10

Decreased serum
anti-diphtheria
antibody
concentrations at
age 5 yrs and
perinatal (pregnancy
week 32-2 wks
postpartum) PFDA
concentrations

Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean
(2018a): medium
confidence

Human,
male and
female

6.68 x 10"9

2 x 10"10

Developmental effects

Decreased birth
weight

Wikstrom et al.
(2020): hiah
confidence

Human,
male and
female

9.6 x 10"9

3 x 10"10

Wikstrom et al.
(2020): hiah
confidence

Human,
male

8.6 x 10"9

3 x 10"10

Wikstrom et al.
(2020): hiah
confidence

Human,
female

8.1 x 10"9

3 x 10"10

Liver effects

Increased AST

SD rat,
male

1.16 x 10"3

1,000

1 x 10"6

This document is a draft for review purposes only and does not constitute Agency policy.

5-34 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

28-d study NTP
(2018); hiah

SD rat,
female

4.00 x 10"3

1,000

4 X 10"6

Increased ALP

confidence

SD rat,
male

1.47 x 10"3

1,000

1 X 10"6

SD rat,
female

1.00 x 10"3

1,000

1 X 10"6

SD rat,
male

1.60 x 10"3

1,000

2 x 10"6

SD rat,
female

7.17 x 10"4

1,000

7 x 10"7

Increased relative

28-d study
Frawlev et al.
(2018): hiah
confidence

SD rat,

female

(histopath

ology

study

cohort)

1.42 x 10"3

1,000

1 x 10"6

liver weight

SD rat,

female

(MPS

study

cohort)

1.20 x 10"3

1,000

1 x 10"6

SD rat,

female

(TDAR

study

cohort)

8.00 x 10"4

1,000

8 x 10"7

Male reproductive effects

Decreased cauda
epididymis sperm
count

28-d study NTP
(2018); low
confidence

SD rat,
male

9.07 x 10"3

1,000

9 x 10"6

Increased Leydig cell
atrophy

28-d study NTP
(2018); hiah

5.89 x 10"3

1,000

6 x 10"6

Decreased serum
testosterone

confidence

5.89 x 10"3

1,000

6 x 10"6

Decreased absolute
testis weight

1.01 x 10"2

1,000

1 x 10"5

Decreased absolute
cauda epididymis
weight

5.48 x 10"3

1,000

5 x 10"6

Decreased absolute
whole epididymis
weight

5.14 x 10"3

1,000

5 x 10"6

Female reproductive effects

Decreased number of
days spent in estrus

28-day study NTP
(2018); hiah

SD rat,
female

1.77 x 10"3

1,000

2 x 10"6

Increased number of
days spent in diestrus

confidence

2.76 x 10"3

1,000

3 x 10"6

This document is a draft for review purposes only and does not constitute Agency policy.

5-35 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Decreased relative
uterus weight

8.63 x 10"3

1,000

9 x 10"6

Decreased absolute
uterus weight

8.63 x 10"3

1,000

9 x 10"6

Jl/fT \7 fvnn rt ^ IS-} C1 A 11 '

the potassium or sodium salts, using the corresponding molecular weights.

Selection ofSubchronic Toxicity Value(s)

As described above, candidate subchronic values for several health effects associated with
PFDA exposure were derived. The subchronic osRfD values selected were associated with
decreased serum antibody concentrations for developmental immune effects, decreased birth
weight for developmental effects, increased relative liver weight for liver effects, decreased whole
epididymis weight for male reproductive effects and increased number of days spent in diestrus for
female reproductive effects. As discussed earlier, these subchronic osRfDs may be useful for certain
decision purposes (i.e., site-specific risk assessments with less-than-lifetime exposures).

Confidence in each subchronic osRfD is described in Table 5-15 and this considers confidence in the
study used to derive the quantitative estimate, the overall health effect, specific evidence base, and
quantitative estimate for each subchronic osRfD.

Table 5-15. Confidence in the subchronic organ/system specific RfDs
(subchronic osRfDs) for PFDA

Confidence categories

Designation3

Discussion

Immune (developmental) subchronic osRfD = 4 x 10~10 mg/kg-d

Confidence in study used
to derive the subchronic
osRfD

High

Confidence in Grandiean et al. (2012): Budtz-J0rgensen and Grandiean
(2018a) was rated as medium primarily due to relatively limited PFDA
exposure contrasts, which can decrease study sensitivity in general. (HAWC
link). Given that the results in this study were statistically significant, EPA
concluded that while there were potential study sensitivity concerns at the
evaluation stage, the results clearly showed that those concerns were not
borne out, and confidence in this study to derive an osRfD was judged to be
high.

Confidence in evidence
base supporting this hazard

Medium

Confidence in the evidence base for immune effects is medium based on
consistent findings of reduced antibody responses from 2 medium
confidence birth cohort studies (Grandiean et al., 2012): (Grandiean et
al., 2017a): (Grandiean et al., 2017b) and a low confidence study in
adults (Kielsen et al., 2016). Short-term studies in animals of hiah/medium
confidence provide supportive evidence of immunosuppression after PFDA
exposure (Frawlev et al., 2018): (NTP, 2018). Some residual uncertainties
regarding unexplained inconsistency and potential confounding by other co-
occurring PFAS from epidemiological studies and issues with concomitant
overt target organ and systemic toxicity in animal studies lower confidence
in the available evidence for this hazard. Other limitations include the lack
of epidemiological studies or long-term/chronic studies in animals examining
effects on the immune system across different developmental life stages and

This document is a draft for review purposes only and does not constitute Agency policy.

5-36 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Confidence categories

Designation3

Discussion

immunotoxicity categories, including sensitization and allergic response and
autoimmunity and autoimmune disease.

Confidence in the
quantification of the
PODhed

Medium

Confidence in the quantification of the POD and osRfD is medium. The POD is
based on BMD modeling at the lower end of the range of the observed data
and a BMDLi/2sd estimate that is associated with a small degree of
uncertainty due to potential confounding by PFOA (see Appendix D.l.l for
more details). The POD for decreased tetanus antibodies at age 7 yrs was
judged to be medium confidence based on a good model fit and was
supported by the nearly identical POD for decreased diphtheria antibodies at
age 7 yrs. Both PODs support the osRfD. A health-protective estimate for
human clearance was applied to estimate the PODhed using PFDA-specific
pharmacokinetic information, the latter of which involves some residual
uncertainty (see discussion on Uncertainty in the pharmacokinetic modeling
of PFDA above). There is also uncertainty as to the most sensitive window of
vulnerability with respect to the exposure/outcome measurement timing
(BMDs/BMDLs were estimated from PFDA levels measured at age 5 or
perinatally and anti-tetanus antibody concentrations measured at age 7 or
5): (Grandiean et al., 2017b) reported that estimated PFDA
"concentrations at 3 mo and 6 mo showed the strongest inverse associations
with antibody concentrations at age 5 yrs, particularly for tetanus." Thus, it
is possible that adverse effects during infancy could be more sensitive than
between ages 5 and 7 yrs.

Overall confidence in
subchronic osRfD

Medium

The overall confidence in the osRfD is medium and is driven by medium
confidence in the evidence base for immune effects, the quantification of
the POD, and the study used for BMD modeling.

Developmental subchronic osRfD = 3 x io~10 mg/kg-d

Confidence in study3 used
to derive osRfD

Medium

Confidence in the Wikstrom et al. (2020) study for hazard identification
was rated as hiah (HAWC link) for developmental effects. The study was
selected for dose-response analysis due to low overall risk of bias and
reliable exposure measurements which had sufficient exposure contrasts
and other characteristics that allowed for adequate study sensitivity to
detect associations. The Wikstrom et al. (2020) study demonstrated
associations consistent in magnitude for boys, girls, and the overall
population. Overall, mean birth weight was considered the most precise and
accurate endpoint and not anticipated to be subject to much error. This
study was advanced for dose-response analysis, given no presumed impact
of pregnancy hemodynamics given the early sampling (96% from trimester
1). Wikstrom et al. (2020) also adjusted for sample timing in their
multivariate models and show no differences in models also restricted to
trimester 1 samples only. Some uncertainty remains on the potential for
confounding by other PFAS (concern primarily for PFNA) which were not
examined in this study. Given the potential quantitative impact of this
uncertainty, confidence in the use of this study for dose-response analysis
was judged as medium rather than high.

Confidence in evidence
base supporting this hazard

Medium-low

Confidence in the evidence base for developmental effects is medium.

There was consistent evidence for reduced birth weight among multiple
human studies, including high quality studies. However, unlike the
Wikstrom et al. (2020) study used here and noted above, some
uncertainty remains in many studies given the predominance of associations
that were detected for studies with later pregnancy sampling. The human
database also showed some coherence across different measures of fetal
growth restriction. In animals, the lone developmental study reported
effects on fetal growth that are coherent with effects observed in humans.

This document is a draft for review purposes only and does not constitute Agency policy.

5-37 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Confidence categories

Designation3

Discussion

Some residual uncertainty regarding potential confounding by other co-
occurring PFAS from epidemiological studies lowers confidence in the
available evidence for this hazard.

Confidence in
quantification of the
PODhed

Medium

Confidence in the quantification of the POD and osRfD is medium given the
POD was based on a BMD hybrid approach within the range of the observed
data and dosimetric adjustment was based on PFDA-specific
pharmacokinetic information, the latter of which involves some residual
uncertainty (see discussion on Uncertainty in the pharmacokinetic modeling
of PFDA above).

Overall confidence in osRfD

Medium-low

The overall confidence in the osRfD is medium and is driven by medium-low
confidence in the evidence base for developmental effects (i.e., fetal growth
restriction).

Liver subchronic osRfD = 7 x 10~7 mg/kg-d

Confidence in study3 used
to derive osRfD

High

Confidence in the NTP (2018) study was rated hiah based on good or
adequate ratings for most study quality domains (HAWC link) and
characteristics that make it suitable for deriving toxicity values, including the
relevance of the exposure paradigm (route, duration, and exposure levels),
use of a relevant species, and the study size and design.

Confidence in evidence
base supporting this hazard

Medium

Confidence in the evidence base for liver effects is medium. Coherent liver
effects for histopathology, serum biomarkers and organ weights were
observed across short-term rodent studies (primarily two high confidence
28-d studies) that are supported by mechanistic studies of biological
plausibility and possible human relevance. Uncertainties remain due to the
absence of longer-term toxicity studies (28 d) and limited information from
available epidemiological studies and in vivo models to characterize the role
of PPARa and other signaling pathways in the mechanisms of hepatotoxicity
of PFDA in both humans and animals.

Confidence in
quantification of the
PODhed

Medium

Confidence in the quantification of the POD and osRfD is medium given the
POD was based on BMD modeling within (at the lower end) the range of the
observed data and dosimetric adjustment was based on PFDA-specific
pharmacokinetic information, the latter of which involves some residual
uncertainty (see discussion on Uncertainty in the pharmacokinetic modeling
of PFDA above).

Overall confidence in the
subchronic osRfD

Medium

The overall confidence in the osRfD is medium and is primarily driven by
medium confidence in both the evidence base supporting this hazard and
the quantification of the POD using BMD modeling of data from a high
confidence study.

Male reproductive subchronic osRfD = 5 x 10~6 mg/kg-d

Confidence in study3 used
to derive osRfD

High-medium

Confidence in the NTP (2018) study was rated hiah-medium (HAWC link)
since most of male reproductive measures were rated as high, including the
basis for the subchronic osRfD (decreased whole epididymis weight), with
the exception of sperm measures which suffered from insensitivity due to
short-term exposure. This is supported by the study evaluation results (good
or adequate ratings for most study quality domains) and characteristics that
make it suitable for deriving toxicity values, including the relevance of the
exposure paradigm (route, duration, and exposure levels), use of a relevant
species, and the study size and design.

Confidence in evidence
base supporting this hazard

Medium-low

Confidence in the evidence base for male reproductive effects is medium to
low. Coherent effects across several relevant measures, including, sperm
parameters, histopathology, serum testosterone levels and organ weights

This document is a draft for review purposes only and does not constitute Agency policy.

5-38 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Confidence categories

Designation3

Discussion

were observed in a high confidence 28-d rat study. The findings are
supported by coherent evidence from a limited number of epidemiological
and mechanistic studies. In spite of the available evidence, some
outstanding uncertainties in the database remain, including the absence of
longer-term exposure studies (>28 d), developmental or multigenerational
studies that evaluate effects in both adults and developing humans and
animals. Given these evidence base uncertainties, it is likely that this osRfD is
under-protective of all male reproductive effects.

Confidence in
quantification of the
PODhed

Medium

Confidence in the quantification of the POD and osRfD is medium given the
POD was based on BMD modeling within the range of the observed data and
dosimetric adjustment was based on PFDA-specific pharmacokinetic
information, the latter of which involves some residual uncertainty (see
discussion on Uncertainty in the pharmacokinetic modeling of PFDA above).

Overall confidence in the
subchronic osRfD

Medium-low

The overall confidence in the osRfD is medium-low and is primarily driven by
the medium-low confidence in the evidence base. The high confidence in
the study and medium confidence in the quantification of the POD does not
fully mitigate the uncertainties associated with medium-low confidence in
the evidence base.

Female reproductive subchronic osRfD = 3 x 10~6 mg/kg-d

Confidence in study3 used
to derive osRfD

High

Confidence in the NTP (2018) studv is hiah (HAWC link) given the studv
evaluation results (i.e., rating of good in all evaluation categories) and
characteristics that make it suitable for deriving toxicity values, including the
relevance of the exposure paradigm (route, duration, and exposure levels),
use of a relevant species, and the study size and design.

Confidence in evidence
base supporting this hazard

Medium-low

Confidence in the evidence base for female reproductive effects is medium-
low. There were consistent and coherent effects on uterus weight and the
estrous cycle in a single high confidence study. Despite the available
evidence, limitations of the evidence base for female reproductive effects
include the lack of informative human studies and the lack of a subchronic
study in animals as well as lack of studies that examined the effect of PFDA
on female fertility and pregnancy outcomes in exposed animals. There are
also no developmental or multigenerational studies that evaluated effects in
both adults and developing humans and animals. Given these evidence base
uncertainties, it is likely that this osRfD is under-protective of all female
reproductive effects.

Confidence in
quantification of the
PODhed

Medium

Overall confidence in the
subchronic osRfD

Medium-low

aAII study evaluation details can be found on HAWC.

1 Selection ofSubchronic RfD and Confidence Statement

2 Organ/system-specific and overall subchronic RfD values for PFDA selected in the previous

3 section are summarized in Table 5-16.

This document is a draft for review purposes only and does not constitute Agency policy.

5-39 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

Table 5-16. Organ/system-specific and overall subchronic RfDs for PFDA

System

Toxicity Value

Basis

PODhed (mg/kg-d)

UFC

osRfD
(mg/kg-d)

Confidence

Immune (developmental)

Subchronic
osRfD

Decreased serum
antibody

concentrations for
both tetanus and
diphtheria in children
at age 7 yrs and PFDA
measured at age 5 yrs

1.07 x 10-S based on
BMDL/2sd from
Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean (2018a)

4 x 10"10

Medium

Developmental

Subchronic
osRfD

Decreased birth
weight in males and
females

9.6 x 10"9 based on
BMDL5%rD from
Wikstrom et al.
(2020)

3 x 10"10

Medium-low

Liver

Subchonic
osRfD

Increased liver weight
in SD female rats

7.17 x 10"4 based on
BMDLio%rd from NTP
(2018)

1,000

7 x 10"7

Medium

Male reproductive

Subchronic
osRfD

Decreased absolute
whole epididymis
weight in SD rats

5.14 x 10"3 based on
BMDLisd from NTP
(2018)

1,000

5 x 10"6

Medium-low

Female reproductive

Subchronic
osRfD

Increased number of
days spent in diestrus
in SD rats

2.76 x 10"3 based on
BMDL5%rD from NTP
(2018)

1,000

3 x 10"6

Medium-low

Immune/developmental

Overall

subchronic RfD

Decreased antibody
concentrations for
both tetanus and
diphtheria in children
at age 7 yrs and PFDA
measured at age 5 yrs

Decreased birth
weight in males and
females

1.07 x 10_S based on
BMDLy2sDfrom
Grandiean et al.
(2012): Budtz-
J0rgensen and
Grandiean (2018a)

9.6 x 10"9 based on
BMDL5%rD from
(Wikstrom et al.,
2020)

4 x 10"10

Medium

1 From the identified subchronic osRfDs (see Table 5-16), an overall subchronic RfD of 4 x

2 10~10 mg/kg-day based on decreased serum antibody concentrations and decreased birth

3 weight in humans was selected. As described in Table 5-15, confidence in the RfD is medium,

4 based on medium confidence in the immune osRfD (the developmental osRfD was medium-low

5 confidence), noting that there was medium confidence in the quantification of the PODs for both

6 immune fBudtz-Targensen and Grandiean. 2018al: f Grandiean etal.. 20121 and developmental

7 fWikstrom etal.. 20201 endpoints using BMD modeling. This RfD is considered to be representative

8 of both immune and developmental effects given the close proximity (~1.5-fold) of the

9 developmental and immune PODs and resulting osRfDs and that both critical effects are observed
10 during the developmental period (see Section 5.2 for more details).

This document is a draft for review purposes only and does not constitute Agency policy.

5-40 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

As described above, the toxicity value of 4 xlO10 mg/kg-day for decreased serum antibody
concentrations for both diphtheria and tetanus at age 7 and PFDA measured at age 5 was selected
for immune effects Budtz-largensen and Grandiean f2018al: Grandiean etal. f20121: and the
toxicity value of 3 x 1010 mg/kg-day based on reduced birth weight from the Wikstrom et al.
f20201 study was selected for developmental effects.

The PODs calculated in Table 5-9 from 28-day studies in rodents were selected for each
health effect for the derivation of the candidate subchronic toxicity values based on several
considerations, including whether there is an endpoint with less uncertainty and/or greater
sensitivity, and whether the endpoint is protective of both sexes and all life stages.

For liver effects, the toxicity value of 7 x 10"7 mg/kg-day (BMDL10RD[HED] of 7.17 x 10"4
mg/kg-day) for increased liver weight in female rats in the NTP f20181 study was selected as the
liver osRfD because it is a reliable marker of hepatotoxicity and represents a more sensitive
reference value than other liver endpoints considered for dose-response modeling (see Table 5-8).
For male reproductive effects, endpoints with a high confidence rating (i.e., increased Leydig cell
atrophy, decreased serum testosterone, decreased testis weight, and decreased epididymis weight
[whole and cauda]) were prioritized over endpoints which suffered from potential sensitivity issues
due to short-term study exposure (i.e., decreased epididymal sperm counts). Since the PODs for the
prioritized endpoints were similar (HEDs ranging from 5.14 x 10"3-1.01 x 10-2) and consistent with
mechanistic evidence that suggest PFDA targets Leydig cells and causes decreased steroidogenesis
and androgen deficiency (see section 3.2.4), the most sensitive POD based on a BMDLISD(HED) of
5.14 x 10"3 mg/kg-day for decreases in whole epididymis weights was selected for derivation,
resulting in a subchronic toxicity value of 5 x 10-6 mg/kg-day for male reproductive effects. Lastly,
the osRfD of 3 x 10"6 mg/kg-day (BMDL5RD[HED] of 2.76 x 10"3 mg/kg-day) based on increased
number of days spent in diestrus was selected for female reproductive effects given its association
with infertility as provided by the U.S. EPA's Guidelines for Reproductive Toxicity Risk Assessment.
This endpoint is also supported by concomitant decreases in estrus time (BMDL5RD[HED] of
1.77 x 10"3 mg/kg-day), for which the association with infertility is less clear.

The subchronic osRfDs for liver, male reproductive and female reproductive effects derived
from short-term animal data were several orders of magnitude higher than the subchronic osRfDs
for immune and developmental effects in humans; therefore, they were not considered to be
sufficiently protective for consideration in the selection of the overall subchronic RfD. Also, in the
case of male and female reproductive effects, confidence in the respective osRfDs was lower
compared to the immune osRfD (medium-low versus medium) due to deficiencies in the evidence
base for these health effects.

5.2.4. Inhalation Reference Concentration (RfC) Derivation

No studies examining inhalation effects of short-term, subchronic, chronic or gestational
exposure for PFDA in humans or animals have been identified, precluding the derivation of an RfC.

This document is a draft for review purposes only and does not constitute Agency policy.

5-41 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

5.3. CANCER TOXICITY VALUES

Considering the limitations in the evidence base across human, animal, and mechanistic
studies of PFDA (see Section 3.3) and in accordance with the Guidelines for Carcinogen Risk
Assessment fU.S. EPA. 20051. EPA concluded that the evidence is inadequate to assess
carcinogenic potential of PFDA in humans. The lack of adequate carcinogenicity data for PFDA
precludes the derivation of quantitative estimates of either oral (oral slope factor, OSF) or
inhalation (inhalation unit risk; IUR) exposure.

This document is a draft for review purposes only and does not constitute Agency policy.

5-42 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

REFERENCES

Abdel-Hafiz. HA: Horwitz. KB. (2014). Post-translational modifications of the progesterone
receptors [Review], J Steroid Biochem Mol Biol 140: 80-89.
http://dx.doi.Org/10.1016/i.isbmb.2013.12.008.

Adinehzadeh. M: Reo. NY. (1998). Effects of peroxisome proliferators on rat liver phospholipids:
sphingomyelin degradation may be involved in hepatotoxic mechanism of
perfluorodecanoic acid. Chem Res Toxicol 11: 428-440.
http: / /dx. doi. o r g /10.10 21 /tx9 7015 51

Aimuzi. R: Luo. K: Chen. 0: Wang. H: Feng. L: Ouvang. F: Zhang. 1. (2019). Perfluoroalkyl and

polyfluoroalkyl substances and fetal thyroid hormone levels in umbilical cord blood among
newborns by prelabor caesarean delivery. Environ Int 130: 104929.
http://dx.doi.Org/10.1016/i.envint.2019.104929.

Aimuzi. R: Luo. K: Huang. R: Huo. XN: Nian. M: Ouvang. FX: Du. YT: Feng. LP: Wang. WY: Zhang. 1.
(2020). Perfluoroalkyl and polyfluroalkyl substances and maternal thyroid hormones in
early pregnancy. Environ Pollut 264: 114557.
http://dx.doi.Org/10.1016/i.envpol.2020.114557.

Al-Ervani. L: Wahlang. B: Falkner. KC: Guardiola. 11: Clair. HB: Prough. RA: Cave. M. (2015).

Identification of Environmental Chemicals Associated with the Development of Toxicant-
associated Fatty Liver Disease in Rodents. Toxicol Pathol 43: 482-497.
http://dx.doi.org/10.1177/0192623314549960.

Alberti. KG: Eckel. RH: Grundy. SM: Zimmet. PZ: Cleeman. II: Donato. KA: Fruchart. TC: Tames. WP:
Loria. CM: Smith. SC. (2009). Harmonizing the metabolic syndrome: A joint interim
statement of the International Diabetes Federation Task Force on Epidemiology and
Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World
Heart Federation; International Atherosclerosis Society; and International Association for
the Study of Obesity. Circulation 120: 1640-1645.
http://dx.doi.org/10.1161/CIRCULATIONAHA.109.192644.

Alexander. EK: Pearce. EN: Brent. GA: Brown. RS: Chen. H: Dosiou. C: Grobman. W: Laurberg. P:
Lazarus. TH: Mandel. ST: Peeters. R: Sullivan. S. f20171. 2017 Guidelines of the American
Thyroid Association for the diagnosis and management of thyroid disease during pregnancy
and the postpartum. Thyroid 27: 315-389. http://dx.doi.org/10.1089/thy.2016.Q457.

Amin. K. (2012). The role of mast cells in allergic inflammation. Respir Med 106: 9-14.
http://dx.doi.Org/10.1016/i.rmed.2011.09.007.

Anderson. RH: Long. GC: Porter. RC: Anderson. IK. (2016). Occurrence of select perfluoroalkyl

substances at U.S. Air Force aqueous film-forming foam release sites other than fire-training
areas: Field-validation of critical fate and transport properties. Chemosphere 150: 678-685.
http://dx.doi.Org/10.1016/i.chemosphere.2016.01.014.

Angrish. MM: Kaiser. TP: Mcqueen. CA: Chorlev. BN. (2016). Tipping the balance: Hepatotoxicity and
the 4 apical key events of hepatic steatosis [Review], Toxicol Sci 150: 261-268.
http: / /dx. doi. o r g /10.10 9 3 /toxsci /kfwO 18.

Arand. M: Coughtrie. MW: Burchell. B: Oesch. F: Robertson. LW. (1991). Selective induction of

bilirubin UDP-glucuronosyl-transferase by perfluorodecanoic acid. Chem Biol Interact 77:
97-105. http://dx.doi.Org/10.1016/0009-2797f91190008-U.

ATSDR (Agency for Toxic Substances and Disease Registry). (2018a). An overview of perfluoroalkyl
and polyfluoroalkyl substances and interim guidance for clinicians responding to patient

This document is a draft for review purposes only and does not constitute Agency policy.

R-l DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

exposure concerns. Interim guidance revised 5/7/2018.
https://stacks.cdc.gov/view/cdc/77114.

ATSDR (Agency for Toxic Substances and Disease Registry). (2018b). Toxicological profile for

perfluoroalkyls. Draft for public comment [ATSDR Tox Profile], Atlanta, GA: U.S. Department
of Health and Human Services, Centers for Disease Control and Prevention.
https://www.atsdr.cdc.gov/toxprofiles/tp200.pdf.

Averina. M: Brox. 1: 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. http://dx.doi.Org/10.1016/i.envres.2021.110740.

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. http://dx.doi.org/10.1289/EHP1493.

Bach. CC: Bech. BH: Nohr. EA: Olsen. I: Matthiesen. NB: Bonefeld-largensen. EC: Bossi. R: Henriksen.
TB. (2016). Perfluoroalkyl acids in maternal serum and indices of fetal growth: The Aarhus
Birth Cohort Environ Health Perspect 124: 848-854.
http://dx.doi.org/10.1289/ehp.1510046.

Bach. CC: Bech. BH: Nohr. EA: Olsen. 1: Matthiesen. NB: Bossi. R: Uldbierg. N: Bonefeld-l0rgensen.

EC: Henriksen. TB. (2015). Serum perfluoroalkyl acids and time to pregnancy in nulliparous
women. Environ Res 142: 535-541. http://dx.doi.Org/10.1016/i.envres.2015.08.007.

Baduel. C: Paxman. CI: Mueller. IF. (2015). Perfluoroalkyl substances in a firefighting training
ground (FTG), distribution and potential future release. J Hazard Mater 296: 46-53.
http://dx.doi.Org/10.1016/i.ihazmat.2015.03.007.

Bailey. SA: Zidell. RH: Perry. RW. (2004). Relationships between organ weight and body/brain
weight in the rat: What is the best analytical endpoint? Toxicol Pathol 32: 448-466.
http://dx.doi.Org/10.1080/01926230490465874.

Bao. WW: Oian. ZM: Geiger. SD: Liu. E: Liu. Y: Wang. SO: 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 Environ 607-608: 1304-1312. http://dx.doi.Org/10.1016/i.scitotenv.2017.07.124.

Barber. TL: Berger. U: Chaemfa. C: Huber. S: Tahnke. A: Temme. C: Tones. KC. (2007). Analysis of per-
and polyfluorinated alkyl substances in air samples from Northwest Europe. J Environ
Monit9: 530-541. http://dx.doi.org/10.1039/b701417a.

Barrett. ES: Chen. C: Thurston. SW: Haug. LS: Sabaredzovic. A: Fieldheim. FN: Frvdenberg. H: Lipson.
SF: Ellison. PT: Thune. I. (2015). Perfluoroalkyl substances and ovarian hormone
concentrations in naturally cycling women. Fertil Steril 103: 1261-1270.el263.
http://dx.doi.Org/10.1016/i.fertnstert2015.02.001.

Batt. AM: Ferrari. L. (1995). Manifestations of chemically-induced liver-damage. Clin Chem 41:
1882-1887.

Beck. IH: Timmermann. CAG: Nielsen. F: Schoeters. G: Tahnk. C: Kvhl. HB: Hast. A: Tensen. 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.
http://dx.doi.org/10.1186/sl2940-019-0541-z.

Berg. V: Nast. TH: Hansen. S: Elverland. A: Vevhe. AS: Torde. R: Odland. 10: 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. http://dx.doi.Org/10.1016/i.envint2015.01.007.

Berg. V: N0st. TH: Pettersen. RD: Hansen. S: Vevhe. AS: Torde. R: Odland. 10: Sandanger. TM. (2017).
Persistent organic pollutants and the association with maternal and infant thyroid
homeostasis: a multipollutant assessment Environ Health Perspect 125: 127-133.
http: / /dx. doi. o r g /10.12 89 /E H P15 2.

This document is a draft for review purposes only and does not constitute Agency policy.

R-2 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Biiland. S: Rensen. PCN: Pieterman. ET: Maas. ACE: van Per Hoorn. TW: van Erk. MT: Havekes. LM:
Willems van Diik. K: Chang. SC: Ehresman. DT: Butenhoff. TL: Princen. HMG. (2011).
Perfluoroalkyl sulfonates cause alkyl chain length-dependent hepatic steatosis and
hypolipidemia mainly by impairing lipoprotein production in AP0E*3-Leiden CETP mice.
Toxicol Sci 123: 290-303. http: / / dx. do i. o r g /10.109 3 /toxs ci /kfr 14 2.

Birukov. A: Andersen. LB: Andersen. MS: Nielsen. TH: Nielsen. F: Kvhl. HB: l0rgensen. IS: Grandiean.
P: Dechend. R: Tensen. TK. (2021). Exposure to perfluoroalkyl substances and blood
pressure in pregnancy among 1436 women from the Odense Child Cohort. Environ Int 151:
106442. http ://dx. doi.org/10.1016/i.envint.2021.106442.

Biton. Y: Goldenberg. I: Kutvifa. V: Baman. 1. R.: Solomon. S: Moss. AT: Szepietowska. B: Mcnitt. S:
Polonskv. B: Zareba. W: Barsheshet. A. (2016). Relative Wall Thickness and the Risk for
Ventricular Tachyarrhythmias in Patients With Left Ventricular Dysfunction. J Am Coll
Cardiol 67: 303-312. http://dx.doi.org/1 0.1016/i.iacc.2015.10.076.

Bierregaard-Olesen. C: Bach. CC: Long. M: Wiels0e. M: Bech. BH: Henriksen. TB: Olsen. 1: Bonefeld-
l0rgensen. EC. (2019). Associations of Fetal Growth Outcomes with Measures of the
Combined Xenoestrogenic Activity of Maternal Serum Perfluorinated Alkyl Acids in Danish
Pregnant Women. Environ Health Perspect 127: 17006.
http://dx.doi.org/10.1289/EHP1884.

Blake. BE: Pinnev. 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. http://dx.doi.Org/10.1016/i.envpol.2018.07.042.

Blomberg. AT: Shih. YH: Messerlian. C: Targensen. LH: Weihe. P: Grandiean. P. (2021). Early-life

associations between per- and polyfluoroalkyl substances and serum lipids in a longitudinal
birth cohort. Environ Res 200: 111400. http://dx.doi.Org/10.1016/i.envres.2021.111400.

Bloom. MS: Commodore. S: Ferguson. PL: Neelon. B: Pearce. TL: Baumer. A: Newman. RB: Grobman.
W: Tita. A: Roberts. T: Skupski. D: Palomares. K: Nageotte. M: Kannan. K: Zhang. C: Wapner.
R: Vena. TE: Hunt. KT. (2022). Association between gestational PFAS exposure and Children's
adiposity in a diverse population. Environ Res 203: 111820.
http://dx.doi.Org/10.1016/i.envres.2021.111820.

Bloom. MS: Kannan. K: Spliethoff. HM: Tao. L: Aldous. KM: Vena. TE. (2010). Exploratory assessment
of perfluorinated compounds and human thyroid function. Physiol Behav 99: 240-245.
http://dx.doi.Org/10.1016/i.physbeh.2009.02.005.

Boettiger. EG. (1946). Changes in the glycogen and water content of the rat uterus. Journal of

Cellular and Comparative Physiology 27: 9-14. http://dx.doi.org/10.1002/icp.1030270103.

Bookstaff. RC: Moore. RW: Ingall. GB: Peterson. RE. (1990). Androgenic deficiency in male rats
treated with perfluorodecanoic acid. Toxicol Appl Pharmacol 104: 322-333.
http: //dx.doi.org/10.1016/0041-008XC90190306-F.

Boone. L: Meyer. D: Cusick. P: Ennulat. D: Bolliger. AP: Everds. N: Meador. V: Elliott. G: Honor. D:

Bounous. D: Tordan. H. (2005). Selection and interpretation of clinical pathology indicators
of hepatic injury in preclinical studies [Review], Vet Clin Pathol 34: 182-188.
http: / /dx.doi.org/10.1111 /i.l 939-165X.2005.tb00041 ,x.

Boros. PL: Revankar. SG. (2017). Granulomatous diseases. 2017.

Bouirad. N: Vidic. B: Gazouli. M: Cultv. M: Papadopoulos. V. (2000). The peroxisome proliferator
perfluorodecanoic acid inhibits the peripheral-type benzodiazepine receptor (PBR)
expression and hormone-stimulated mitochondrial cholesterol transport and steroid
formation in Leydig cells. Endocrinology 141: 3137-3148.
http://dx.doi.Org/10.1210/endo.141.9.7678.

This document is a draft for review purposes only and does not constitute Agency policy.

R-3 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Braunig. T: Baduel. C: Heffernan. A: Rotander. A: Donaldson. E: Mueller. TF. (2017). Fate and
redistribution of perfluoroalkyl acids through AFFF-impacted groundwater. Sci Total
Environ 596-597: 360-368. http://dx.doi.Org/10.1016/i.scitotenv.2017.04.095.

Bretveld. RW: Thomas. CMG: Scheepers. PTT: Zielhuis. GA: Roeleveld. N. (2006). Pesticide exposure:
the hormonal function of the female reproductive system disrupted? [Review], Reprod Biol
Endocrinol 4: 30. http://dx.doi.org/10.1186/1477-7827-4-3Q.

Brewster. DW: Birnbaum. LS. (1989). The biochemical toxicity of perfluorodecanoic acid in the

mouse is different from that of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol
99: 544-554. http: //dx.doi.org/10.1016/0041 -008Xf89190161-0.

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. http://dx.doi.org/10.1289/ehp.1307621.

Buck Louis. GM: Zhai. S: Smarr. MM: Grewal. 1: Zhang. C: Grantz. KL: Hinkle. SN: Sundaram. R: Lee. S:
Honda. M: Oh. 1: Kannan. K. (2018). Endocrine disruptors and neonatal anthropometry,
NICHD Fetal Growth Studies - Singletons. Environ Int 119: 515-526.
http://dx.doi.Org/10.1016/i.envint.2018.07.024.

Buck. RC: Franklin. 1: Berger. U: Conder. TM: Cousins. IT: de Voogt. P: Tensen. AA: Kannan. K: Maburv.
SA: van Leeuwen. SP. (2011). Perfluoroalkyl and polyfluoroalkyl substances in the
environment: terminology, classification, and origins [Review], Integr Environ Assess
Manag 7: 513-541. http ://dx.doi.org/10.1002/ieam.258.

Budtz-l0rgensen. E: Grandiean. P. (2018a). Application of benchmark analysis for mixed

contaminant exposures: Mutual adjustment of perfluoroalkylate substances associated with
immunotoxicity. PLoS ONE 13: e0205388.
http://dx.doi.org/10.1371/iournal.pone.0205388.

Budtz-l0rgensen. E: Grandiean. P. (2018b). Computational details for the paper "Application of
benchmark analysis for mixed contaminant exposures: Mutual adjustment of
perfluoroalkylate substances associated with immunotoxicity".

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.
http://dx.doi.Org/10.1016/i.toxlet2013.01.025.

Cakmak. S: Lukina. A: Karthikevan. S: Atlas. E: Dales. R. (2022). The association between blood PFAS
concentrations and clinical biochemical measures of organ function and metabolism in
participants of the Canadian Health Measures Survey (CHMS). Sci Total Environ 827:
153900. http://dx.doi.Org/10.1016/i.scitotenv.2022.153900.

Caldwell. ASL: Middleton. LI: Timenez. M: Desai. R: Mcmahon. AC: Allan. CM: Handelsman. DT:

Walters. KA. (2014). Characterization of reproductive, metabolic, and endocrine features of
polycystic ovary syndrome in female hyperandrogenic mouse models. Endocrinology 155:
3146-3159. http:/ /dx.doi.org/10.1210/en.2014-l 196.

Callan. AC: Rotander. A: Thompson. K: Hevworth. I: Mueller. TF: Odland. 10: Hinwood. AL. (2016).

Maternal exposure to perfluoroalkyl acids measured in whole blood and birth outcomes in
offspring. Sci Total Environ 569-570: 1107-1113.
http://dx.doi.Org/10.1016/i.scitotenv.2016.06.177.

Cao. L: Ouan. XB: Zeng. WT: Yang. XO: Wang. MI. (2016). Mechanism of hepatocyte apoptosis
[Review], Journal of Cell Death 9: 19-29. http: //dx.doi.org/10.4137/TCD.S39824.

Cao. W: Liu. X: Liu. X: Zhou. Y: Zhang. X: Tian. H: Wang. 1: Feng. S: Wu. Y: Bhatti. P: Wen. S: Sun. X.
(2018). Perfluoroalkyl substances in umbilical cord serum and gestational and postnatal
growth in a Chinese birth cohort. Environ Int 116: 197-205.
http://dx.doi.Org/10.1016/i.envint.2018.04.015.

Carlson. LA: Angrish. M: Shirke. AY: Radke. EG: Schulz. B: Kraft. A: Tudson. R: Patlewicz. G: Blain. R:
Lin. C: Vetter. N: Lemeris. C: Hartman. P: Hubbard. H: Arzuaga. X: Davis. A: Dishaw. LV:

This document is a draft for review purposes only and does not constitute Agency policy.

R-4 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Druwe. IL: Hollinger. H: Tones. R: Kaiser. TP: Lizarraga. L: Noves. PD: Taylor. M: Shapiro. AT:
Williams. AT: Thayer. KA. (2022). Systematic evidence map for over one hundred and fifty
per- and polyfluoroalkyl substances (PFAS) [Review], Environ Health Perspect 130: 56001.
http://dx.doi.org/10.1289 /EHP10343.

Carney. EW: Kimmel. CA. (2007). Interpretation of skeletal variations for human risk assessment:
delayed ossification and wavy ribs. Birth Defects Res B Dev Reprod Toxicol 80: 473-496.

Carwile. TL: Seshasavee. SM: Aris. IM: Rifas-Shiman. SL: Claus Henn. B: Calafat. AM: Sagiv. SK: Oken.
E: Fleisch. AF. (2021). Prospective associations of mid-childhood plasma per- and
polyfluoroalkyl substances and pubertal timing. Environ Int 156: 106729.
http://dx.doi.Org/10.1016/i.envint.2021.106729.

CDC (Centers for Disease Control and Prevention). (2022). National Report on Human Exposure to
Environmental Chemicals. Atlanta, GA: U.S. Department of Health and Human Services.
https://www.cdc.gov/exposurereport/pdf/FourthReport UpdatedTables Volumel Mar20
18.pdf.

Charles. D: Berg. V: N0st. TH: Huber. S: Sandanger. TM: Rvlander. C. (2020). Pre- and post-

diagnostic blood profiles of perfluoroalkyl acids in type 2 diabetes mellitus cases and
controls. Environ Int 145:106095. http://dx.doi.Org/10.1016/i.envint.2020.106095.

Chen. H: Huang. CY: Wilson. MW: Lay. LT: Robertson. LW: Chow. CK: Glauert. HP. (1994). Effect of
the peroxisome proliferators ciprofibrate and perfluorodecanoic acid on hepatic cell
proliferation and toxicity in Sprague-Dawley rats. Carcinogenesis 15: 2847-2850.
http://dx.doi.org/10.1093/carcin/15.12.2847.

Chen. L: Tong. C: Huo. X: Zhang. 1: Tian. Y. (2021). Prenatal exposure to perfluoroalkyl and
polyfluoroalkyl substances and birth outcomes: A longitudinal cohort with repeated
measurements. Chemosphere 267: 128899.
http://dx.doi.Org/10.1016/i.chemosphere.2020.128899.

Chen. LC: Borges. T: Glauert. HP: Knight. SAB: Sunde. RA: Schramm. H: Oesch. F: Chow. CK:

Robertson. LW. (1990). Modulation of selenium-dependent glutathione peroxidase by
perfluorodecanoic acid in rats: Effect of dietary selenium. J Nutr 120: 298-304.
http://dx.doi.Org/10.1093/in/120.3.298.

Chen. 0: Huang. R: Hua. L: Guo. Y: Huang. L: Zhao. Y: Wang. X: Zhang. 1. (2018a). Prenatal exposure
to perfluoroalkyl and polyfluoroalkyl substances and childhood atopic dermatitis: A
prospective birth cohort study. Environ Health 17:1-12.
http://dx.doi.org/10.1186/sl2940-018-0352-7.

Chen. 0: Zhang. X: Zhao. Y: Lu. W: Wu. 1: Zhao. S: Zhang. I: Huang. L. (2019). Prenatal exposure to
perfluorobutanesulfonic acid and childhood adiposity: A prospective birth cohort study in
Shanghai, China. Chemosphere 226: 17-23.
http://dx.doi.Org/10.1016/i.chemosphere.2019.03.095.

Chen. WL: Bai. FY: Chang. YC: Chen. PC: Chen. CY. (2018b). Concentrations of perfluoroalkyl
substances in foods and the dietary exposure among Taiwan general population and
pregnant women. J Food Drug Anal 26: 994-1004.
http://dx.doi.Org/10.1016/i.ifda.2017.12.011.

Chinie. E: Kentish. P: Tarnot. B: George. M: Gibson. G. (1994). Induction of the CYP4A subfamily by
perfluorodecanoic acid: The rat and the guinea pig as susceptible and non-susceptible
species. Toxicol Lett 71: 69-75. http://dx.doi.org/10.1016/0378-4274(94)90200-3.

Christensen. TVR: Bangash. KK: Weihe. P: Grandiean. P: Nielsen. F: Tensen. TK: Petersen. MS. (2021).
Maternal exposure to perfluoroalkyl chemicals and anogenital distance in the offspring: A
Faroese cohort study. Reprod Toxicol 104: 52-57.
http://dx.doi.0rg/lO.lOl6/i.reprotox.202i.O6.Ol6.

This document is a draft for review purposes only and does not constitute Agency policy.

R-5 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Christensen. KY: Raymond. M: Meiman. T. (2019). Perfluoroalkyl substances and metabolic
syndrome. Int J Hyg Environ Health 222: 147-153.
http://dx.doi.Org/10.1016/i.iiheh.2018.08.014.

Christensen. KY: Raymond. M: Thompson. BA: Anderson. HA. (2016). Perfluoroalkyl substances in
older male anglers in Wisconsin. Environ Int91: 312-318.
http://dx.doi.Org/10.1016/i.envint.2016.03.012.

Corsini. E: Sangiovanni. E: Avogadro. A: Galbiati. V: Viviani. B: Marinovich. M: Galli. CL: Dell'Agli. M:
Germolec. DR. (2012). In vitro characterization of the immunotoxic potential of several
perfluorinated compounds (PFCs). Toxicol Appl Pharmacol 258: 248-255.
http://dx.doi.Org/10.1016/i.taap.2011.ll.004.

Corton. TC: Peters. TM: Klaunig. IE. (2018). The PPARa-dependent rodent liver tumor response is not
relevantto humans: addressing misconceptions [Review], Arch Toxicol 92: 83-119.
http://dx.doi.org/10.1007/s00204-017-2094-7.

Craig. EA: Yan. Z: Zhao. 01. (2015). The relationship between chemical-induced kidney weight
increases and kidney histopathology in rats. J Appl Toxicol 35: 729-736.
http: / /dx. doi. o r g /10.10 0 2 /i at. 3 0 3 6.

Creasy. D: Bube. A: de Riik. E: Kandori. H: Kuwahara. M: Masson. R: Nolte. T: Reams. R: Regan. K:
Rehm. S: Rogerson. P: Whitney. K. (2012). Proliferative and nonproliferative lesions of the
rat and mouse male reproductive system [Review], Toxicol Pathol 40: 40S-121S.
http://dx.doi.org/10.1177/0192623312454337.

Creasy. DM: Chapin. RE. (2018). Chapter 17 - Male reproductive system. In Fundamentals of
Toxicologic Pathology (Third Edition). London, UK: Academic Press of Elsevier.
http: //dx.doi.org/10.1016 /B978-0-12-809841 -7.00017-4.

Dalsager. L: Christensen. N: Halekoh. U: Timmermann. CAG: Nielsen. F: Kvhl. HB: Husbv. S:

Grandiean. P: Tensen. TK: Andersen. HR. (2021a). 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.
http://dx.doi.Org/10.1016/i.envint.2021.106395.

Dalsager. L: Christensen. N: Husbv. S: Kvhl. H: Nielsen. F: H0st. A: Grandiean. P: Tensen. 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. http://dx.doi.Org/10.1016/i.envint2016.08.026.

Dalsager. L: Tensen. TK: Nielsen. F: Grandiean. P: Bilenberg. N: Andersen. HR. (2021b). No

association between maternal and child PFAS concentrations and repeated measures of
ADHD symptoms at age 2Vi and 5 years in children from the Odense Child Cohort
Neurotoxicol Teratol 88: 107031. http://dx.doi.Org/10.1016/i.ntt.2021.107031.

Darrow. DC: Soule. HC: Buckman. TE. (1928). Blood volume in normal infants and children. J Clin
Invest 5: 243-258. http://dx.doi.org/10.1172/TCT100156.

Das. KP: Wood. CR: Lin. MT: Starkov. AA: Lau. C: Wallace. KB: Corton. TC: Abbott. BP. (2017).

Perfluoroalkyl acids-induced liver steatosis: Effects on genes controlling lipid homeostasis.
Toxicology 378: 37-52. http://dx.doi.Org/10.1016/i.tox.2016.12.007.

Davidson. MH. (2010). Update on CETP inhibition. 4: 394-398.
http://dx.doi.Org/10.1016/i.iacl.2010.08.003.

Davies. B: Morris. T. (1993). Physiological parameters in laboratory animals and humans [Review],
Pharm Res 10: 1093-1095. http://dx.doi.Org/10.1023/A:1018943613122.

De Boo. HA: Harding. TE. (2006). The developmental origins of adult disease (Barker) hypothesis
[Review], Aust N Z J Obstet Gynaecol 46: 4-14. http://dx.doi.org/10.ll 11 /i.1479-
828X.2006.00506.x.

de Simone. G: Daniels. S. R: Kimball. TR: Roman. MT: Romano. C: Chinali. M: Galderisi. M: Devereux.
RB. (2005). Evaluation of concentric left ventricular geometry in humans - Evidence for age-

This document is a draft for review purposes only and does not constitute Agency policy.

R-6 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

related systematic underestimation. Hypertension 45: 64-68.
http://dx.doi.Org/10.1161/01.HYP.0000150108.37527.57.

Dent. MP: Carmichael. PL: Tones. KC: Martin. FL. (2015). Towards a non-animal risk assessment for
anti-androgenic effects in humans [Review], Environ Int 83: 94-106.
http://dx.doi.Org/10.1016/i.envint.2015.06.009.

Domazet. SL: Tensen. TK: Wedderkopp. N: Nielsen. F: Andersen. LB: Gr0ntved. 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.
http://dx.doi.0rg/lO.lOl6/i.envres.202O.llOllO.

Dong. GH: Tung. KY: Tsai. CH: Liu. MM: Wang. D: Liu. W: Tin. 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. http://dx.doi.org/10.1289/ehp.1205351.

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.
http://dx.doi.0rg/lO.lOl6/i.ecoenv.2Oi9.O2.O6l.

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.
http://dx.doi.Org/10.1016/i.envint.2019.105295.

Dunder. L: Lind. PM: Salihovic. S: Stubleski. 1: Karrman. A: Lind. L. (2022). Changes in plasma levels
of per- and polyfluoroalkyl substances (PFAS) are associated with changes in plasma lipids -
A longitudinal study over 10 years. Environ Res 211: 112903.
http://dx.doi.Org/10.1016/i.envres.2022.112903.

Dzierlenga. AL: Robinson. VG: Waidvanatha. S: Devito. Ml: Eifrid. MA: Gibbs. ST: Granville. CA:

Blystone. CR. (2019). Toxicokinetics of perfluorohexanoic acid (PFHxA), perfluorooctanoic
acid (PFOA) and perfluorodecanoic acid (PFDA) in male and female Hsd:Sprague dawley SD
rats following intravenous or gavage administration. Xenobiotica 50: 1-11.
http://dx.doi.Org/10.1080/00498254.2019.1683776.

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. http://dx.doi.Org/10.1097/EE9.0000000000000095.

EMEA (European Medicines Agency). (2008). Non-clinical guideline on drug induced hepatotoxicity.
(EMEA/CHMP/SWP/150115/2006). London, UK.

http://www.ema.europa.eu/docs/en GB/document library/Scientific guideline/2009/09/
WC500003355.pdf.

Ernst. A: Brix. N: Lauridsen. LLB: Olsen. 1: Parner. ET: Liew. Z: Olsen. LH: Ramiau-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.
http://dx.doi.org/10.1289/EHP3567.

Evans. H: Killoran. KE: Mitre. E. (2014). Measuring Local Anaphylaxis in Mice. J Vis Exp 14: e52005.
http://dx.doi.org/10.3791/520Q5.

Evans. TT: Ganiam. VK. (2011). Chapter 2 - Reproductive anatomy and physiology. In RC Gupta (Ed.),
Reproductive and Developmental Toxicology (lsted., pp. 7-32). San Diego, CA: Academic
Press, http://dx.d0i.0rg/l0.1016/B978-0-12-382032-7.10002-5.

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.
http://dx.d0i.0rg/l 0.1289/EHP303.

This document is a draft for review purposes only and does not constitute Agency policy.

R-7 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Foster. P: Gray. LE. Tr. (2013). Toxic responses of the reproductive system. In CD Klaassen (Ed.),

Casarett& Doull's toxicology: The basic science of poisons (8th ed., pp. 861-906). New York,
NY: McGraw-Hill Education, https://www.mheducation.com/highered/product/casarett-
doull-s-toxicology-basic-science-poisons-9th-edition-klaassen/9781259863745.html.

Fraser. AT: Webster. TF: Watkins. DT: Strvnar. Ml: Kato. K: Calafat. AM: Vieira. VM: Mcclean. MP.

(2013). Polyfluorinated compounds in dust from homes, offices, and vehicles as predictors
of concentrations in office workers' serum. Environ Int 60: 128-136.
http://dx.doi.Org/10.1016/i.envint.2013.08.012.

Frawlev. RP: Smith. M: Cesta. MF: Haves-Bouknight. S: Blvstone. C: Kissling. GE: Harris. S: Germolec.
D. (2018). Immunotoxic and hepatotoxic effects of perfluoro-n-decanoic acid (PFDA) on
female Harlan Sprague-Dawley rats and B6C3F1 /N mice when administered by oral gavage
for 28 days. J Immunotoxicol 15: 41-52.
http://dx.doi.org/10.1080/1547691X.2018.1445145.

Friis-Hansen. B. (1961). Body water compartments in children: Changes during growth and related
changes in body composition. Pediatrics 28: 169-181.

Fu. Y: Wang. T: Fu. 0: Wang. P: Lu. Y. (2014). Associations between serum concentrations of

perfluoroalkyl acids and serum lipid levels in a Chinese population. Ecotoxicol Environ Saf
106: 246-252. http://dx.doi.Org/10.1016/i.ecoenv.2014.04.039.

Fuiii. 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. http://dx.doi.org/10.1539/ioh.14-0136-QA.

Galazka. A: Kardvmowicz. B. (1989). Immunity against diphtheria in adults in Poland. Epidemiol
Infect 103: 587-593. http://dx.doi.org/10.1017/s0950268800030983.

Galazka. AM: Milstien. IB: 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; Cutts, FT. http://apps.who.int/iris/bitstream/handle/10665/58891/WHO-
EPI-GEN-93.12-mod2-eng.pdf?sequence=38&isAllowed=y.

Gao. K. e: Zhuang. T: Liu. X: Fu. I: Zhang. 1: Fu. I. ie: Wang. L: Zhang. A: Liang. Y: Song. M: Tiang. 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.
http://dx.doi.org/10.1021/acs.est.9b00715.

Gao. Y: Luo. I: Zhang. Y: Pan. C: Ren. Y: Zhang. 1: Tian. Y: Cohort. SB. (2022). Prenatal Exposure to
Per- and Polyfluoroalkyl Substances and Child Growth Trajectories in the First Two Years.
Environ Health Perspect 130: 37006. http://dx.doi.org/10.1289/EHP9875.

Gardener. H: Sun. 0: Grandiean. P. (2021). PFAS concentration during pregnancy in relation to
cardiometabolic health and birth outcomes. Environ Res 192: 110287.
http://dx.doi.Org/10.1016/i.envres.2020.110287.

Gebbink. WA: Berger. U: Cousins. IT. (2015). Estimating human exposure to PF0S isomers and PFCA
homologues: the relative importance of direct and indirect (precursor) exposure. Environ
Int 74: 160-169. http://dx.doi.Org/10.1016/i.envint2014.10.013.

Getz. GS: Reardon. CA. (2012). Animal models of atherosclerosis [Review], Arterioscler Thromb
Vase Biol 32: 1104-1115. http://dx.doi.org/10.1161 /ATVBAHA.l 11.237693.

Glauert. HP: Srinivasan. S: Tatum. VL: Chen. LC: Saxon. DM: Lay. LT: Borges. T: Baker. M: Chen. LH:
Robertson. LW: Chow. CK. (1992). Effects of the peroxisome proliferators ciprofibrate and
perfluorodecanoic acid on hepatic cellular antioxidants and lipid peroxidation in rats.
Biochem Pharmacol 43: 1353-1359. http://dx.doi.org/10.1016/0006-2952r92190513-I.

Godin. CS: Mvhr. BC: Lawlor. TE: Young. RR: Murli. H: Cifone. MA. (1992). Assessment of the

potential genotoxicity of perfluorodecanoic acid and chlorotrifluoroethylene trimer and

This document is a draft for review purposes only and does not constitute Agency policy.

R-8 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

tetramer acids. Fundam Appl Toxicol 18: 557-569. http://dx.doi.org/10.1016/Q272-
0590C92190115-X.

Goecke-Flora. CM: Wvman. TF: Tarnot. BM: Reo. NY. (1995). Effect of the peroxisome proliferator

perfluoro-n-decanoic acid on glucose transport in the isolated perfused rat liver. Chem Res
Toxicol 8: 77-81. http://dx.doi.org/10.1021/tx00043a01Q.

Goeden. HM: Greene. CW: Tacobus. TA. (2019). A transgenerational toxicokinetic model and its use in
derivation of Minnesota PFOA water guidance. J Expo Sci Environ Epidemiol 29: 183-195.
http://dx.doi.org/10.1038/s41370-018-0110-5.

Goldman. TM: Murr. AS: Cooper. RL. (2007). The rodent estrous cycle: Characterization of vaginal
cytology and its utility in toxicological studies [Review], Birth Defects Res B Dev Reprod
Toxicol 80: 84-97. http://dx.doi.org/10.1002/bdrb.20106.

Goodrich. TA: Alderete. TL: Baumert. BO: Berhane. K: Chen. Z: Gilliland. FD: Goran. MI: Hu. X: Tones.
DP: Margetaki. K: Rock. S: Stratakis. N: Valvi. D: Walker. PI: Conti. DV: Chatzi. L. (2021).
Exposure to Perfluoroalkyl Substances and Glucose Homeostasis in Youth. Environ Health
Perspect 129: 97002. http://dx.doi.org/10.1289/EHP9200.

Goudarzi. H: Mivashita. C: Okada. E: Kashino. I: Chen. CI: Ito. S: Araki. A: Kobavashi. S: Matsuura. H:
Kishi. R. (2017). Prenatal exposure to perfluoroalkyl acids and prevalence of infectious
diseases up to 4years of age. Environ Int 104: 132-138.
http://dx.doi.Org/10.1016/i.envint.2017.01.024.

Goudarzi. H: Mivashita. C: Okada. E: Kashino. I: Kobavashi. S: Chen. CI: 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.
http://dx.doi.Org/10.1016/i.envint.2016.05.020.

Grandiean. P: Andersen. EW: Budtz-largensen. E: Nielsen. F: Malbak. K: Weihe. P: Heilmann. C.
(2012). Serum vaccine antibody concentrations in children exposed to perfluorinated
compounds. JAMA 307: 391-397. http://dx.doi.org/10.1001/iama.2011.2Q34.

Grandiean. P: Bateson. T. (2021). RE: Benchmark analysis for PFAS immunotoxicity. Available
online at (accessed

Grandiean. P: Heilmann. C: Weihe. P: Nielsen. F: Mogensen. UB: Budtz-l0rgensen. E. (2017a). Serum
vaccine antibody concentrations in adolescents exposed to perfluorinated compounds.
Environ Health Perspect 125: 077018. http://dx.doi.org/10.1289/EHP275.

Grandiean. P: Heilmann. C: Weihe. P: Nielsen. F: Mogensen. UB: Timmermann. A: Budtz-I0rgensen.

E. (2017b). Estimated exposures to perfluorinated compounds in infancy predict attenuated
vaccine antibody concentrations at age 5-years. J Immunotoxicol 14: 188-195.
http://dx.doi.org/10.1080/1547691X.2017.1360968.

Grandiean. P: Timmermann. CAG: Kruse. M: Nielsen. F: Vinholt. PI: Boding. L: Heilmann. C: M0lbak.
K. (2020). Severity of COVID-19 at elevated exposure to perfluorinated alkylates. PLoS ONE
15: e0244815. http://dx.doi.org/10.1371/iournal.pone.0244815.

Gump. BB: Wu. 0: Dumas. AK: Kannan. K. (2011). Perfluorochemical (PFC) exposure in children:
Associations with impaired response inhibition. Environ Sci Technol 45: 8151-8159.
http://dx.doi.org/10.1021/esl03712g.

Guo. 1: Zhang. I: Wang. Z: Zhang. L: Oi. X: Zhang. Y: Chang. X: Wu. C: Zhou. Z. (2021). Umbilical cord
serum perfluoroalkyl substance mixtures in relation to thyroid function of newborns:
Findings from Sheyang Mini Birth Cohort Study. Chemosphere 273: 129664.
http://dx.doi.Org/10.1016/i.chemosphere.2021.129664.

Gutshall. DM: Pilcher. GD: Langlev. AE. (1988). Effect of thyroxine supplementation on the response
to perfluoro-n-decanoic acid (PFDA) in rats. J Toxicol Environ Health 24: 491-498.
http://dx.doi.org/10.1080/15287398809531179.

This document is a draft for review purposes only and does not constitute Agency policy.

R-9 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Gutshall. DM: Pilcher. GD: Langlev. AE. (1989). Mechanism of the serum thyroid hormone lowering
effect of perfluoro-n-decanoic acid (PFDA) in rats. J Toxicol Environ Health 28: 53-65.
http://dx.doi.org/10.1080/15287398909531328.

Guvatt. G: Oxman. AD: Akl. EA: Kunz. R: Vist. G: Brozek. 1: Norris. S: Falck-Ytter. Y: Glasziou. P:
DeBeer. H: laeschke. R: Rind. D: Meerpohl. 1: Dahm. P: Schiinemann. HI. (2011). GRADE
guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin
Epidemiol 64: 383-394. http://dx.doi.Org/10.1016/i.iclinepi.2010.04.026.

Gvllenhammar. I: Diderholm. B: Gustafsson. 1: Berger. U: Ridefelt. P: Benskin. TP: 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.
http://dx.doi.Org/10.1016/i.envint.2017.12.002.

Hall. AP: Elcombe. CR: Foster. TR: Harada. T: Kaufmann. W: Knippel. A: Kiittler. K: Malarkev. DE:
Maronpot. RR: Nishikawa. A: Nolte. T: Schulte. A: Strauss. V: York. Ml. (2012). Liver
hypertrophy: a review of adaptive (adverse and non-adverse) changes-conclusions from
the 3rd International ESTP Expert Workshop [Review], Toxicol Pathol 40: 971-994.
http://dx.doi.org/10.n77/0192623312448935.

Hall. SM: Zhang. S: Hoffman. K: Miranda. ML: Stapleton. HM. (2022). Concentrations of per- and
polyfluoroalkyl substances (PFAS) in human placental tissues and associations with birth
outcomes. Chemosphere 295: 133873.
http://dx.doi.Org/10.1016/i.chemosphere.2022.133873.

Hallberg. L: Hogdahl. AM: Nilsson. L: Rvbo. G. (1966). Menstrual blood loss-a population study.

Variation at different ages and attempts to define normality. Acta Obstet Gynecol Scand 45:
320-351. http://dx.doi.org/10.3109/00016346609158455.

Han. W: Gao. Y: Yao. 0: Yuan. T: Wang. Y: Zhao. S: Shi. R: Bonefeld-Torgensen. EC: Shen. X: Tian. Y.

(2018). Perfluoroalkyl and polyfluoroalkyl substances in matched parental and cord serum
in Shandong, China. Environ Int 116: 206-213.
http://dx.doi.Org/10.1016/i.envint.2018.04.025.

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. http://dx.doi.Org/10.1016/i.envres.2004.12.003.

Harbison. RD: Bourgeois. MM: Tohnson. GT. (2015). Hamilton & Hardy's industrial toxicology (6th
ed.). Hoboken, NJ: John Wiley & Sons, Inc.

http://www.wiley.eom/WileyCDA/WileyTitle/productCd-0470929731.subiectCd-
CH50.html.

Hardell. E: Karrman. A: van Bavel. B: Bao. 1: Carlberg. M: Hardell. L. (2014). Case-control study on
perfluorinated alkyl acids (PFAAs) and the risk of prostate cancer. Environ Int 63: 35-39.
http://dx.doi.Org/10.1016/i.envint.2013.10.005.

Harris. MH: Oken. E: Rifas-Shiman. SL: Calafat. AM: Bellinger. DC: Webster. TF: White. RF: Sagiv. SK.
(2021). Prenatal and childhood exposure to per- and polyfluoroalkyl substances (PFAS) and
child executive function and behavioral problems. Environ Res 202: 111621.
http://dx.doi.Org/10.1016/i.envres.2021.111621.

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.
http://dx.doi.Org/10.1016/i.envint.2018.03.025.

Harris. MW: Birnbaum. LS. (1989). Developmental toxicity of perfluorodecanoic acid in C57BL/6N
mice. Fundam Appl Toxicol 12: 442-448. http://dx.d0i.0rg/l0.1016/0272-0590(89)90018-
3.

This document is a draft for review purposes only and does not constitute Agency policy.

R-10 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Harris. MW: Uraih. LC: Birnbaum. LS. (1989). Acute toxicity of perfluorodecanoic acid in C57BL/6
mice differs from 2,3,7,8-tetrachlorodibenzo-p-dioxin. Fundam Appl Toxicol 13: 723-736.
http://dx.doi.org/10.1016/0272-0590r89190330-8.

Harrison. EH: Lane. TS: Luking. S: Van Rafelghem. MT: Andersen. ME. (1988). Perfluoro-n-decanoic
acid: Induction of peroxisomal beta-oxidation by a fatty acid with dioxin-like toxicity. Lipids
23: 115-119. http://dx.doi.Org/10.1007/BF02535290.

Hashem. MS: Kalashvan. H: Chov. 1: Chiew. SK: Shawki. AH: Dawood. AH: Becher. H. (2015). Left

ventricular relative wall thickness versus left ventricular mass index in non-cardioembolic
stroke patients. Medicine (Baltimore) 94: e872.
http://dx.doi.Org/10.1097/MD.0000000000000872.

Heo. 11: Lee. TW: Kim. SK: Oh. IE. (2014). Foodstuff analyses show that seafood and water are major
perfluoroalkyl acids (PFAAs) sources to humans in Korea. J Hazard Mater 279: 402-409.
http://dx.doi.Org/10.1016/i.ihazmat.2014.07.004.

Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-
300.

Hiermitslev. MH: Long. M: Wielsae. M: Bonefeld-Iargensen. EC. (2020). Persistent organic pollutants
in Greenlandic pregnant women and indices of foetal growth: The ACCEPT study. Sci Total
Environ 698: 134118. http://dx.doi.Org/10.1016/i.scitotenv.2019.134118.

Hotchkiss. AK: Rider. CV: Blvstone. CR: Wilson. VS: Hartig. PC: Anklev. GT: Foster. PM: Gray. CL:

Gray. LE. (2008). Fifteen years after "Wingspread"-environmental endocrine
disrupters and human and wildlife health: where we are today and where we need to go
[Review], Toxicol Sci 105: 235-259. http://dx.d0i.0rg/l0.1093/toxsci/kfn030.

Haver. BB: Bonde. TP: Tattenborg. SS: Ramlau-Hansen. CH: Lindh. C: Pedersen. HS: Toft. G. (2017).

Exposure to perfluoroalkyl substances during pregnancy and child behaviour at 5 to 9 years
of age. Horm Behav 101: 105-112. http: //dx.doi.Org/10.1016/i.yhbeh.2017.ll.007.

Huang. M: Tiao. 1: Zhuang. P: Chen. X: Wang. 1: Zhang. Y. (2018). Serum polyfluoroalkyl chemicals are
associated with risk of cardiovascular diseases in national US population. Environ Int 119:
37-46. http://dx.doi.Org/10.1016/i.envint.2018.05.051.

Huang. 0: Liu. L: Wu. Y: Wang. X: Luo. L: Nan. B: Zhang. 1: Tian. M: Shen. H. (2019a). Seminal plasma
metabolites mediate the associations of multiple environmental pollutants with semen
quality in Chinese men. Environ Int 132: 105066.
http://dx.doi.Org/10.1016/i.envint.2019.105066.

Huang. R: Chen. 0: Zhang. L: Luo. K: Chen. L: Zhao. S: Feng. L: Zhang. I. (2019b). Prenatal exposure to
perfluoroalkyl and polyfluoroalkyl substances and the risk of hypertensive disorders of
pregnancy. Environ Health 18: 5. http://dx.doi.org/10.1186/sl2940-018-0445-3.

Huo. X: Zhang. L: Huang. R: Feng. L: Wang. W: Zhang. I: Cohort. SB. (2020). Perfluoroalkyl

substances exposure in early pregnancy and preterm birth in singleton pregnancies: a
prospective cohort study. Environ Health 19: 60. http: //dx.doi.org/10.1186/sl2940-020-
00616-8.

ICH Expert Working Group. (2005). Immunotoxicity studies for human pharmaceuticals S8 11. In
ICH Harmonised Tripartite Guideline.

Inoue. K: Ritz. B: Andersen. SL: Ramlau-Hansen. CH: Haver. BB: Bech. BH: Henriksen. TB: Bonefeld-
Targensen. EC: Olsen. 1: Liew. Z. (2019). Perfluoroalkyl substances and maternal thyroid
hormones in early pregnancy; findings in the Danish National Birth Cohort Environ Health
Perspect 127: 117002. http://dx.doi.org/10.1289/EHP5482.

IPCS (International Programme on Chemical Safety). (1996). Principles and methods for assessing
direct immunotoxicity associated with exposure to chemicals. Geneva, Switzerland: World
Health Organization, http://www.inchem.org/documents/ehc/ehc/ehcl80.htm.

This document is a draft for review purposes only and does not constitute Agency policy.

R-ll DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

IPCS (International Programme on Chemical Safety). (2012). Guidance for immunotoxicity risk

assessment for chemicals. (Harmonization Project Document No. 10). Geneva, Switzerland:
World Health Organization, https: //apps.who.int/iris/handle/10665 /330098.

IOEHC (Institute for Quality and Efficiency in Health Care). (2014). Deciding whether or not to treat
subclinical hypothyroidism, https: //www.informedhealth.org/underactive-thvroid-
hypothyroidism.2713.en.pdf?all backgrounds=Q&all details=0&all lexicons=0&all reports=
0&overview=l&print=l&theme=Q.

Ishibashi. H: Hirano. M: Kim. EY: Iwata. H. (2019). In vitro and in silico evaluations of binding

affinities of perfluoroalkyl substances to baikal seal and human peroxisome proliferator-
activated receptor a. Environ Sci Technol 53: 2181-2188.
http://dx.doi.org/10.1021/acs.est.8b07273.

Itoh. S: Araki. A: Mivashita. C: Yamazaki. K: Goudarzi. H: Minatova. M: AitBamai. Y: Kobavashi. 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.
http://dx.doi.Org/10.1016/i.envint.2019.105139.

Taeschke. H: Lemasters. II. (2003). Apoptosis versus oncotic necrosis in hepatic
ischemia/reperfusion injury. Gastroenterology 125: 1246-1257.
http://dx.doi.org/10.1016/S0016-5085r03101209-5.

Tain. RB. (2019). Synergistic impact of co-exposures to toxic metals cadmium, lead, and mercury

along with perfluoroalkyl substances on the healthy kidney function. Environ Res 169: 342-
347. http://dx.doi.Org/10.1016/i.envres.2018.ll.037.

Tain. RB. (2020a). 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. http://dx.doi.Org/10.1016/i.envpol.2020.115207.

Tain. RB. (2020b). Variabilities in concentrations of selected perfluoroalkyl acids among

normotensives and hypertensives across various stages of glomerular function. Arch
Environ Occup Health 76: 1-11. http://dx.doi.org/10.1080/19338244.2020.1732856.

Tain. RB. (2021). Perfluoroalkyl acids and their isomers, diabetes, anemia, and albuminuria:
Variabilities with deteriorating kidney function. Ecotoxicol Environ Saf 208: 111625.
http://dx.doi.Org/10.1016/i.ecoenv.2020.111625.

Tain. RB: Ducatman. A. (2019a). Roles of gender and obesity in defining correlations between
perfluoroalkyl substances and lipid/lipoproteins. Sci Total Environ 653: 74-81.
http://dx.doi.Org/10.1016/i.scitotenv.2018.10.362.

Tain. RB: Ducatman. A. (2019b). 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.
http: //dx.doi.org/10.1097/TOM.00000000000Q1532.

Tanis. TA: Rifas-Shiman. SL: Seshasavee. SM: Sagiv. S: Calafat. AM: Gold. PR: Coull. BA: Rosen. CT:

Oken. E: Fleisch. AF. (2021). Plasma concentrations of per- and polyfluoroalkyl substances
and body composition from mid-childhood to early adolescence. J Clin Endocrinol Metab
106: e3760-e3770. http://dx.d0i.0rg/l0.1210/clinem/dgabl87.

Te. I. -G: Kim. D. uk-Sil: Kim. S. -W: Lee. S: Lee. HS: Park. E. -K: Khang. D: Kim. SH. (2015). Tyrosol

suppresses allergic inflammation by inhibiting the activation of phosphoinositide 3-kinase
in mast cells. PLoS ONE 10: e0129829. http://dx.doi.org/10.1371/iournal.pone.0129829.

Tensen. RC: Andersen. MS: Larsen. PV: Glintborg. D: Dalgard. C: Timmermann. CAG: Nielsen. F:

Sandberg. MB: Andersen. HR: Christesen. HT: Grandiean. P: Tensen. TK. (2020a). 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.
http://dx.d0i.0rg/l 0.1289/EHP5184.

This document is a draft for review purposes only and does not constitute Agency policy.

R-12 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Tensen. RC: Glintborg. D: Gade Timmermann. CA: Nielsen. F: Kvhl. HB: Frederiksen. H: Andersson.
AM: Tuul. A: Sidelmann. TT: Andersen. HR: Grandiean. P: Andersen. MS: Tensen. TK. (2020b).
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.
http://dx.doi.Org/10.1016/i.envres.2019.109101.

Tensen. RC: Glintborg. D: Timmermann. CAG: Nielsen. F: Kvhl. HB: Andersen. HR: Grandiean. P:

Tensen. TK: Andersen. M. (2018). Perfluoroalkyl substances and glycemic status in pregnant
Danish women: The Odense Child Cohort. Environ Intll6: 101-107.
http://dx.doi.Org/10.1016/i.envint.2018.04.010.

Tensen. TK: Andersen. LB: Kvhl. HB: Nielsen. F: Christesen. HT: Grandiean. P. (2015). Association
between perfluorinated compound exposure and miscarriage in Danish pregnant women.
PLoS ONE 10: e0123496. http://dx.doi.org/10.1371/iournal.pone.0123496.

Ti. K: Kim. S: Kho. Y: Paek. D: Sakong. 1: Ha. 1: Kim. S: Choi. K. (2012). Serum concentrations of major
perfluorinated compounds among the general population in Korea: dietary sources and
potential impact on thyroid hormones. Environ Int 45: 78-85.
http://dx.doi.Org/10.1016/i.envint.2012.03.007.

Tiang. W: Zhang. Y: Zhu. L: Deng. 1. (2014). Serum levels of perfluoroalkyl acids (PFAAs) with isomer
analysis and their associations with medical parameters in Chinese pregnant women.
Environ Int 64: 40-47. http://dx.doi.Org/10.1016/i.envint2013.12.001.

Tin. H: Zhang. Y: Tiang. W: Zhu. L: Martin. TW. (2016). Isomer-Specific Distribution of Perfluoroalkyl
Substances in Blood. Environ Sci Technol 50: 7808-7815.
http://dx.doi.org/10.1021/acs.est.6b01698.

Toensen. UN: Vevrand. B: Antignac. TP: Blomberg Tensen. M: Petersen. TH: Marchand. P: Skakkebaek.
NE: Andersson. AM: Le Bizec. B: T0rgensen. N. (2013). PFOS (perfluorooctanesulfonate) in
serum is negatively associated with testosterone levels, but not with semen quality, in
healthy men. Hum Reprod 28: 599-608. http: //dx.doi.org/10.1093 /humrep/des425.

Tones. MEE: Boon. WC: Proietto. 1: Simpson. ER. (2006). Of mice and men: The evolving phenotype
of aromatase deficiency [Review], Trends Endocrinol Metab 17: 55-64.
http://dx.doi.Org/10.1016/i.tem.2006.01.004.

Toshi-Barve. S: Kirpich. I: Cave. MC: Marsano. LS: McClain. CI. (2015). Alcoholic, nonalcoholic, and
toxicant-associated steatohepatitis: Mechanistic similarities and differences [Review],

CMGH 1: 356-367. http://dx.doi.org/10.1016/i.icmgh.2015.05.006.

Kadel. S: Kovats. S. (2018). Sex Hormones Regulate Innate Immune Cells and Promote Sex
Differences in Respiratory Virus Infection. Front Immunol 9.
http://dx.doi.org/10.3389/fimmu.2018.Q1653.

Kang. H: Lee. HK: Moon. HB: Kim. S: Lee. 1: 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.
http://dx.doi.Org/10.1016/i.scitotenv.2018.07.177.

Karaskova. P: Venier. M: Melvmuk. L: Becanova. I: Voita. S: Prokes. R: Diamond. ML: Klanova. I.

(2016). Perfluorinated alkyl substances (PFASs) in household dust in Central Europe and
North America. Environ Int 94: 315-324. http: //dx.doi.Org/10.1016/i.envint.2016.05.031.

Karlsen. M: Grandiean. 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. http://dx.doi.Org/10.1016/i.reprotox.2016.08.002.

Karrman. A: Domingo. TL: Llebaria. X: Nadal. M: Bigas. E: Van bavel. B: Lindstrom. G. (2009).

Biomonitoring perfluorinated compounds in Catalonia, Spain: concentrations and trends in
human liver and milk samples. Environ Sci Pollut Res Int Environ Sci Pollut Res Int 17: 750-
758.

This document is a draft for review purposes only and does not constitute Agency policy.

R-13 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Kashino. I: Sasaki. S: Okada. E: Matsuura. H: Goudarzi. H: Mivashita. C: Okada. E: Ito. YM: Araki. A:
Kishi. R. (2020). Prenatal exposure to 11 perfluoroalkyl substances and fetal growth: A
large-scale, prospective birth cohort study. Environ Int 136: 105355.
http://dx.doi.Org/10.1016/i.envint.2019.105355.

Kass. GEN. (2006). Mitochondrial involvement in drug-induced hepatic injury. Chem Biol Interact
163: 145-159. http://dx.doi.Org/10.1016/i.cbi.2006.06.007.

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.
http: / /dx. doi. o r g /10.10 9 3 /toxsci /kfxO 5 8.

Kawashima. Y: Kobavashi. H: Miura. H: Kozuka. H. (1995). Characterization of hepatic responses of
rat to administration of perfluorooctanoic and perfluorodecanoic acids at low levels.
Toxicology 99: 169-178. http: //dx.doi.org/10.1016/0300-483Xr95103027-D.

Kelling. CK: Van Rafelghem. Ml: Menahan. LA: Peterson. RE. (1987). Effects of perfluorodecanoic
acid on hepatic indices of thyroid status in the rat Biochem Pharmacol 36: 1337-1344.
http://dx.doi.Org/10.1016/0006-2952f87190091-8.

Khalil. N: Ducatman. AM: Sinari. S: Billheimer. D: Hu. C: Littau. S: Burgess. TL. (2020). Per- and

polyfluoroalkyl substance and cardio metabolic markers in firefighters. J Occup Environ
Med 62: 1076-1081. http://dx.doi.Org/10.1097/TOM.0000000000002062.

Kielsen. K: Shamim. Z: Ryder. LP: Nielsen. F: Grandiean. P: Budtz-largensen. E: Heilmann. C. (2016).
Antibody response to booster vaccination with tetanus and diphtheria in adults exposed to
perfluorinated alkylates. J Immunotoxicol 13: 270-273.
http://dx.doi.org/10.3109/1547691X.2015.1067259.

Kim. HY: Kim. KN: Shin. CH: Lim. YH: Kim. II: Kim. BN: Hong. YC: Lee. YA. (2020a). The relationship
between perfluoroalkyl substances concentrations and thyroid function in early childhood:
A prospective cohort study. Thyroid 30: 1556-1565.
http://dx.doi.org/10.1089/thy.2019.Q436.

Kim. K: Bennett. DH: Calafat. AM: Hertz-Picciotto. I: Shin. HM. (2020b). Temporal trends and
determinants of serum concentrations of per- and polyfluoroalkyl substances among
Northern California mothers with a young child, 2009-2016. Environ Res 186: 109491.
http://dx.doi.Org/10.1016/i.envres.2020.109491.

Kim. S: Choi. K: Ti. K: Seo. 1: Kho. Y: Park. 1: Kim. S: Park. S: Hwang. I: Teon. 1: Yang. H: Giesv. TP.

(2011). Trans-placental transfer of thirteen perfluorinated compounds and relations with
fetal thyroid hormones. Environ Sci Technol 45: 7465-7472.
http://dx.doi.org/10.1021/es202408a.

Kim. SC: Hong. IT: Tang. ST: Kang. WS: Yoo. HS: Yun. YP. (1998). Formation of 8-oxodeoxyguanosine
in liver DNA and hepatic injury by peroxisome proliferator clofibrate and perfluorodecanoic
acid in rats. J Toxicol Sci 23: 113-119. http://dx.doi.Org/10.2131/its.23.2 113.

Kim. ST: Choi. El: Choi. GW: Lee. YB: Cho. HY. (2019). Exploring sex differences in human health risk
assessment for PFNA and PFDA using a PBPK model. Arch Toxicol 93: 311-330.
http://dx.doi.Org/10.1007/s00204-018-2365-v.

Kim. SK: Kannan. K. (2007). Perfluorinated acids in air, rain, snow, surface runoff, and lakes:

relative importance of pathways to contamination of urban lakes. Environ Sci Technol 41:
8328-8334. http://dx.doi.org/10.1021/es072107t.

Kim. YR: White. N: Braunig. I: Viiayasarathv. S: Mueller. IF: Knox. CL: Harden. FA: Pacella. R: Toms.
LL. (2020c). Per- and poly-fluoroalkyl substances (PFASs) in follicular fluid from women
experiencing infertility in Australia. Environ Res 190: 109963.
http://dx.doi.Org/10.1016/i.envres.2020.109963.

Kieldsen. LS: Bonefeld-l0rgensen. EC. (2013). Perfluorinated compounds affect the function of sex
hormone receptors. Environ Sci PollutRes Int 20: 8031-8044.
http: //dx.doi.org/10.1007/sl 1356-013-1753-3.

This document is a draft for review purposes only and does not constitute Agency policy.

R-14 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Klaunig. TE: Shinohara. M: Iwai. H: Chengelis. CP: Kirkpatrick. TB: Wang. Z: Bruner. RH. (2015).

Evaluation of the chronic toxicity and carcinogenicity of perfluorohexanoic acid (PFHxA) in
Sprague-Dawley rats. Toxicol Pathol 43: 209-220.
http://dx.doi.org/10.n77/0192623314530532.

Klein. SL: Flanagan. KL. (2016). Sex differences in immune responses [Review], Nat Rev Immunol
16: 626-638. http://dx.doi.org/10.1038/nri.2016.90.

Kleiner. DE: Weerasinghe. P: Buia. LM. (2012). Oncosis: An important non-apoptotic mode of cell
death. Exp Mol Pathol 93: 302-308. http://dx.doi.Org/10.1016/i.yexmp.2012.09.018.

Knobeloch. L: Imm. P: Anderson. H. (2012). Perfluoroalkyl chemicals in vacuum cleaner dust from
39 Wisconsin homes. Chemosphere 88: 779-783.
http://dx.doi.Org/10.1016/i.chemosphere.2012.03.082.

Konwick. BT: Tomv. GT: Ismail. N: Peterson. IT: Fauver. RT: Higginbotham. D: Fisk. AT. (2008).

Concentrations and patterns of perfluoroalkyl acids in Georgia, USA surface waters near and
distantto a major use source. Environ Toxicol Chem 27: 2011-2018.
http://dx.doi.Org/10.1897/07-659.l.

Koshv. TT: Attina. TM: Ghassabian. A: Gilbert. I: 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.
http://dx.doi.Org/10.1016/i.envint.2017.08.003.

Kotthoff. M: Miiller. I: liirling. H: Schlummer. M: Fiedler. D. (2015). Perfluoroalkyl and

polyfluoroalkyl substances in consumer products. Environ Sci Pollut Res Int 22: 14546-
14559. http://dx.doi.org/10.1007/sl 1356-015-4202-7.

Kudo. N. (2015). Metabolism and pharmacokinetics. In JC DeWitt (Ed.), Toxicological Effects of
Perfluoroalkyl and Polyfluoroalkyl Substances (pp. 151-175). New York, NY: Springer
International Publishing, http://dx.doi.org/10.1007/978-3-319-15518-0 6.

Kudo. N: Kawashima. Y. (2003). Induction of triglyceride accumulation in the liver of rats by

perfluorinated fatty acids with different carbon chain lengths: Comparison with induction of
peroxisomal beta-oxidation. Biol Pharm Bull 26: 47-51.
http://dx.doi.org/10.1248/bpb.26.47.

Kudo. N: Suzuki. E: Katakura. M: Ohmori. K: Noshiro. R: Kawashima. Y. (2001). Comparison of the
elimination between perfluorinated fatty acids with different carbon chain length in rats.
Chem Biol Interact 134: 203-216. http://dx.doi.org/10.1016/S0009-2797f01100155-7.

Kvalem. HE: Nvgaard. UC: Ladrup 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.
http://dx.doi.Org/10.1016/i.envint.2019.105259.

Kwon. EI: Shin. IS: Kim. BM: Shah-Kulkarni. S: Park. H: Kho. YL: Park. EA: Kim. YT: Ha. EH. (2016).
Prenatal exposure to perfluorinated compounds affects birth weight through GSTM1
polymorphism. J Occup Environ Med 58: el98-e205.
http://dx.doi.Org/10.1097/TOM.0000000000000739.

Langlev. AE: Pilcher. GD. (1985). Thyroid, bradycardic and hypothermic effects of perfluoro-n-
decanoic acid in rats. J Toxicol Environ Health 15: 485-491.
http://dx.doi.org/10.1080/15287398509530675.

Lankova. D: Lacina. 0: Pulkrabova. 1: Haislova. 1. (2013). The determination of perfluoroalkyl

substances, brominated flame retardants and their metabolites in human breast milk and
infant formula. Talanta 117: 318-325. http: //dx.doi.Org/10.1016/i.talanta.2013.08.040.

Lau. C. (2015). Perfluorinated compounds: An overview. In JC DeWitt (Ed.), Toxicological effects of
perfluoroalkyl and polyfluoroalkyl substances (pp. 1-21). New York: Springer.
http://dx.doi.org/10.1007/978-3-319-15518-0 1.

This document is a draft for review purposes only and does not constitute Agency policy.

R-15 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Lee. ES: Han. S: Oh. TE. (2016). Association between perfluorinated compound concentrations in
cord serum and birth weightusing multiple regression models. Reprod Toxicol 59: 53-59.
http://dx.doi.Org/10.1016/i.reprotox.2015.10.020.

Lee. IK: Kim. SH. (2018). Correlation between mast cell-mediated allergic inflammation and length
of perfluorinated compounds. J Toxicol Environ Health A 81: 1-12.
http://dx.doi.org/10.1080/15287394.2018.1440188.

Lee. YA: Kim. TH: lung. HW: Lim. YH: Bae. S: Kho. Y: Hong. YC: Shin. CH: Yang. SW. (2018). The serum
concentrations of perfluoroalkyl compounds were inversely associated with growth
parameters in 2-year old children. Sci Total Environ 628-629: 226-232.
http://dx.doi.Org/10.1016/i.scitotenv.2018.02.050.

Leehv. KA: Regan Anderson. TM: Daniel. AR: Lange. CA: Ostrander. TH. (2016). Modifications to
glucocorticoid and progesterone receptors alter cell fate in breast cancer [Review], J Mol
Endocrinol 56: R99-R114. http://dx.doi.org/l 0.1530/TME-l 5-0322.

Lefevre. 1: Mcclintock. MK. (1988). Reproductive senescence in female rats - A longitudinal-study of
individual-differences in estrous cycles and behavior. Biol Reprod 38: 780-789.
http://dx.doi.Org/10.1095/biolreprod38.4.780.

Lenters. V: Portengen. L: Rignell-Hvdbom. A: Tonsson. BA: Lindh. CH: Piersma. AH: Toft. G: Bonde.
TP: Heederik. D: Rvlander. 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.
http://dx.doi.org/10.1289/ehp.1408933.

Levin. S: Selmer. D: Z. R. (1993). Effects of two weeks of feed restriction on some common
toxicologic parameters in Sprague-Dawley rats. Toxicol Pathol 21: 1-14.
http://dx.doi.org/10.n77/019262339302100101.

Levitt. D: Liss. A. (1986). Toxicity of perfluorinated fatty acids for human and murine B cell lines.
Toxicol Appl Pharmacol 86: 1-11. http://dx.doi.org/10.1016/0041-0Q8xf86)90394-7.

Lewiriski. A: Stasiak. M. (2017). Physiopathology, diagnosis, and treatment of secondary
hypothyroidism. In Hypothalamic-Pituitary Diseases. Basel, Switzerland: Springer
International Publishing, http://dx.doi.org/10.1007/978-3-319-38681-2 7-1.

Li. 1: Cai. D: Chu. C: Li. 00: Zhou. Y: Hu. LW: Yang. BY: Dong. GH: Zeng. XW: Chen. D. (2020a).
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.
http://dx.doi.org/10.1021/acs.est.0c00829.

Li. I: Cao. H: Feng. H: Xue. 0: Zhang. A: Fu. 1. (2020b). Evaluation of the estrogenic/antiestrogenic
activities of perfluoroalkyl substances and their interactions with the human estrogen
receptor by combining in vitro assays and in silico modeling. Environ Sci Technol 54:
14514-14524. http://dx.doi.org/10.1021/acs.est0c03468.

Li. KM: Zhao. 0: Fan. ZY: Tia. SY: Liu. 0: Liu. FY: Liu. SL. (2022). The toxicity of perfluorodecanoic
acid is mainly manifested as a deflected immune function. Mol Biol Rep 49: 4365-4376.
http://dx.doi.Org/10.1007/sll033-022-07272-w.

Li. LP: Shigematsu. Y: Hamada. M: Hiwada. K. (2001). Relative wall thickness is an independent
predictor of left ventricular systolic and diastolic dysfunctions in essential hypertension.
Hypertens Res 24: 493-499. http://dx.doi.org/10.1291/hypres.24.493.

Li. M: Zeng. XW: Oian. ZM: Vaughn. MG: Sauve. S: Paul. G: Lin. S: Lu. L: Hu. LW: Yang. BY: Zhou. Y:
Oin. XD: Xu. SL: Bao. WW: Zhang. YZ: Yuan. P: Wang. T: Zhang. C: Tian. YP: Nian. M: Xiao. X:
Fu. C: Dong. GH. (2017). Isomers of perfluorooctanesulfonate (PFOS) in cord serum and
birth outcomes in China: Guangzhou Birth Cohort Study. Environ Int 102: 1-8.
http://dx.doi.0rg/lO.lOl6/i.envint.2Oi7.O3.OO6.

This document is a draft for review purposes only and does not constitute Agency policy.

R-16 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Liang. H: Wang. Z: Miao. M: Tian. Y: Zhou. Y: Wen. S: Chen. Y: Sun. X: Yuan. W. (2020). Prenatal

exposure to perfluoroalkyl substances and thyroid hormone concentrations in cord plasma
in a Chinese birth cohort Environ Health 19: 127. http://dx.doi.org/10.1186/sl2940-02Q-
00679-7.

Liang. 1L: 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. http://dx.doi.org/10.15585/mmwr.rr6702al.

Liew. Z: Luo. 1: Nohr. EA: Bech. BH: Bossi. R: Arah. OA: Olsen. 1. (2020). Maternal plasma

perfluoroalkyl substances and miscarriage: a nested case-control study in the Danish
National Birth Cohort Environ Health Perspect 128: 47007.
http://dx.doi.org/10.1289 /EHP6202.

Liew. Z: Ritz. B: Bach. CC: Asarnow. RF: Bech. BH: Nohr. EA: Bossi. R: Henriksen. TB: Bonefeld-

l0rgensen. EC: Olsen. 1. (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.
http://dx.doi.org/10.1289/EHP2754.

Liew. Z: Ritz. B: Bonefeld-l0rgensen. EC: Henriksen. TB: Nohr. EA: Bech. BH: Fei. C: Bossi. R: von
Ehrenstein. OS: Streia. E: Uldall. P: Olsen. 1. (2014). Prenatal exposure to perfluoroalkyl
substances and the risk of congenital cerebral palsy in children. Am J Epidemiol 180: 574-
581. http://dx.doi.org/10.1093/aie/kwu 179.

Liew. Z: Ritz. B: von Ehrenstein. OS: Bech. BH: Nohr. EA: Fei. C: Bossi. R: Henriksen. TB: Bonefeld-
l0rgensen. EC: Olsen. 1. (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.
http://dx.doi.org/10.1289/ehp.1408412.

Lim. 11: Suh. Y: Faustman. EM: Cui. TY. (2021). Regulation of transporters by perfluorinated
carboxylic acids in HepaRG cells. Drug Metab Dispos 50: 1396-1413.
http://dx.doi.org/10.1124/dmd.121.000477.

Lin. HW: Feng. HX: Chen. L: Yuan. XT: Tan. Z. (2020a). Maternal exposure to environmental

endocrine disruptors during pregnancy is associated with pediatric germ cell tumors.
Nagoya J Med Sci 82: 323-333. http://dx.doi.Org/10.18999/nagims.82.2.323.

Lin. TW: Chen. MK: Lin. CC: Chen. MH: Tsai. MS: Chan. DC: Hung. KY: Chen. PC. (2020b). 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. http://dx.doi.org/10.1016/i.iiheh.2020.113607.

Lind. DV: Priskorn. L: Lassen. TH: Nielsen. F: Kvhl. HB: Kristensen. DM: Christesen. HT: l0rgensen.
IS: Grandiean. P: Tensen. TK. (2017a). Prenatal exposure to perfluoroalkyl substances and
anogenital distance at 3 months of age in a Danish mother-child cohort Reprod Toxicol 68:
200-206. http://dx.doi.Org/10.1016/i.reprotox.2016.08.019.

Lind. PM: Lind. L: Salihovic. S: Ahlstrom. H: Michaelsson. K: Kullberg. 1: Strand. R. (2022). Serum
levels of perfluoroalkyl substances (PFAS) and body composition - A cross-sectional study
in a middle-aged population. Environ Res 209: 112677.
http://dx.doi.Org/10.1016/i.envres.2022.112677.

Lind. PM: Salihovic. S: van Bavel. B: Lind. L. (2017b). Circulating levels of perfluoroalkyl substances
(PFASs) and carotid artery atherosclerosis. Environ Res 152: 157-164.
http://dx.doi.0rg/lO.lOl6/i.envres.2Ol6.lO.OO2.

Liu. B: Wei. B: Mo. M: Song. Y: Tang. C: Tang. P: Guo. X: Tan. C: Liu. S: Huang. D: Oiu. X. (2021a).
Exposure to perfluoroalkyl substances in early pregnancy and the risk of hypertensive
disorders of pregnancy: A nested case-control study in Guangxi, China. Chemosphere 288:
132468. http://dx.doi.Org/10.1016/i.chemosphere.2021.132468.

This document is a draft for review purposes only and does not constitute Agency policy.

R-17 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Liu. C: Xu. X: Huo. X. ia. (2014). Anogenital distance and its application in environmental health
research [Review], Environ Sci PollutRes Int 21: 5457-5464.
http://dx.doi.Org/10.1007/sll356-014-2570-z.

Liu. F: Fan. Z: Song. N: Han. M: Yan. M: Guo. LH: Tihui. 1: Liu. S. (2019a). XRCC4, which is inhibited by
PFDA, regulates DNA damage repair and cell chemosensitivity. J Cell Biochem 120: 12665-
12676. http://dx.doi.org/10.1002/icb.28534.

Liu. G: Dhana. K: Furtado. ID: Rood. 1: Zong. G: Liang. L: Oi. L: Bray. GA: Deionge. L: Coull. B:

Grandiean. P: Sun. 0. (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. http://dx.doi.org/10.1371/iournal.pmed.1002502.

Liu. G: Zhang. B: Hu. Y: Rood. 1: Liang. L: Oi. L: Bray. GA: Deionge. L: Coull. B: Grandiean. P: Furtado.
ID: Sun. 0. (2020a). Associations of Perfluoroalkyl substances with blood lipids and
Apolipoproteins in lipoprotein subspecies: the POUNDS-lost study. Environ Health 19: 5.
http://dx.doi.Org/10.1186/sl2940-020-0561-8.

Liu. H: Pan. Y: Tin. S: Li. Y: Zhao. L: Sun. X: Cui. 0: Zhang. B: Zheng. T: Xia. W: Zhou. A: Campana. AM:
Dai. 1: Xu. S. (2020b). Associations of per-/polyfluoroalkyl substances with glucocorticoids
and progestogens in newborns. Environ Int 140: 105636.
http://dx.doi.Org/10.1016/i.envint.2020.105636.

Liu. 1: Li. 1: 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.
http://dx.doi.Org/10.1016/i.envint.2011.05.001.

Liu. IT: Cui. XX: Tan. YW: Dong. PX: Ou. YO: Li. 00: Chu. C: Wu. LY: Liang. LX: Oin. ST: Zeeshan. M:

Zhou. Y: Hu. LW: Liu. RO: Zeng. XW: Dong. GH: Zhao. XM. (2022). Per- and perfluoroalkyl
substances alternatives, mixtures and liver function in adults: A community-based
population study in China. Environ Int 163: 107179.
http://dx.doi.Org/10.1016/i.envint.2022.107179.

Liu. M: Zhang. G: Meng. L: Han. X: Li. Y: Shi. Y: Li. A: Turvk. ME: Zhang. 0: Tiang. G. (2021b).

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 56: 6144-6151. http://dx.doi.org/10.1021/acs.estlc02850.

Liu. X: Zhang. L: Chen. L: Li. I: Wang. 1: Zhao. Y: Liu. L: Wu. Y. (2021c). Identification and

prioritization of the potent components for combined exposure of multiple persistent
organic pollutants associated with gestational diabetes mellitus. J Hazard Mater 409:
124905. http://dx.doi.Org/10.1016/i.ihazmat.2020.124905.

Liu. X: Zhang. L: Chen. L: Li. I: Wang. Y: Wang. 1: Meng. G: Chi. M: Zhao. Y: Chen. H: Wu. Y. (2019b).

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. http://dx.doi.Org/10.1016/i.envint.2019.03.035.

Lorber. M: Eaglesham. GE: Hobson. P: Toms. LM: Mueller. IF: Thompson. IS. (2015). The effect of
ongoing blood loss on human serum concentrations of perfluorinated acids. Chemosphere
118: 170-177. http://dx.doi.Org/10.1016/i.chemosphere.2014.07.093.

Louis. GM: Sapra. KT: Barr. DB: Lu. Z: Sundaram. R. (2016). Preconception perfluoroalkyl and

polyfluoroalkyl substances and incident pregnancy loss, LIFE Study. Reprod Toxicol 65: 11-
17. http://dx.doi.Org/10.1016/i.reprotox.2016.06.011.

Louis. GMB: Peterson. CM: Chen. Z: Hediger. ML: Croughan. MS: Sundaram. R: Stanford. IB: Fuiimoto.
VY: Varner. MW: Giudice. LC: Kennedy. A: Sun. L: Wu. 0: Kannan. K. (2012).
Perfluorochemicals and endometriosis: The ENDO study. Epidemiology 23: 799-805.
http://dx.doi.Org/10.1097/EDE.0b013e31826cc0cf.

This document is a draft for review purposes only and does not constitute Agency policy.

R-18 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Lum. KT: 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. http://dx.doi.Org/10.1097/EDE.0000000000000552.

Luo. D: Wu. WX: Pan. YA: Du. BB: Shen. Ml: Zeng. LX. (2021). Associations of prenatal exposure to
per- and polyfluoroalkyl substances with the neonatal birth size and hormones in the
growth hormone/insulin-like growth factor axis. Environ SciTechnol 55: 11859-11873.
http://dx.doi.org/10.1021/acs.est.lc02670.

Luo. 1: Xiao. 1: Gao. Y: Ramlau-Hansen. CH: Toft. G: Li. 1: Obel. C: Andersen. SL: Deziel. NC: Tseng. WL:
Inoue. K: Bonefeld-largensen. EC: Olsen. 1: Liew. Z. (2020). Prenatal exposure to
perfluoroalkyl substances and behavioral difficulties in childhood at 7 and 11 years.

Environ Res 191: 110111. http://dx.doi.Org/10.1016/i.envres.2020.110111.

Luo. M: Tan. Z: Dai. M: Song. D: Lin. 1: Xie. M: Yang. 1: Sun. L: Wei. D: Zhao. 1: Gonzalez. FT: Liu. A.
(2017). Dual action of peroxisome proliferator-activated receptor alpha in
perfluorodecanoic acid-induced hepatotoxicity. Arch Toxicol 91: 897-907.
http://dx.doi.Org/10.1007/s00204-016-1779-7.

Lvall. 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. http://dx.doi.org/10.1289/EHP1830.

Maher. TM: Aleksunes. LM: Dieter. MZ: Tanaka. Y: Peters. TM: Manautou. IE: 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.
http://dx.doi.org/10.1093/toxsci/kfnl77.

Mamsen. LS: Biorvang. 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. http://dx.doi.org/10.1016/i.envint2019.01.010.

Mamsen. LS: Tonsson. BAG: Lindh. CH: Olesen. RH: Larsen. A: Ernst. E: Kelsev. 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.
http://dx.doi.Org/10.1016/i.scitotenv.2017.04.058.

Manzano-Salgado. CB: Casas. M: Lopez-Espinosa. MT: Ballester. F: Iniguez. C: Martinez. D: Costa. 0:
Santa-Marina. L: Pereda-Pereda. E: Schettgen. T: Sunver. 1: Vriiheid. M. (2017). Prenatal
exposure to perfluoroalkyl substances and birth outcomes in a Spanish birth cohort.

Environ Int 108: 278-284. http://dx.doi.Org/10.1016/i.envint.2017.09.006.

Mariash. CN: Kaiser. FE: Schwartz. HL: Towle. HC: Oppenheimer. TH. (1980). Synergism of thyroid-
hormone and high-carbohydrate diet in the induction of lipogenic enzymes in the rat -
Mechanisms and implications. J Clin Invest 65: 1126-1134.
http: //dx.doi.org/10.1172 /TCT109766.

Marty. MS: Allen. B: Chapin. RE: Cooper. R: Daston. GP: Flaws. TA: Foster. PMD: Makris. SL:

Mvlchreest. E: Sandler. D: Tvl. RW. (2009). Inter-laboratory control data for reproductive
endpoints required in the OPPTS 870.3800/OECD 416 reproduction and fertility test Birth
Defects Res B Dev Reprod Toxicol 86: 470-489. http://dx.doi.org/10.1002/bdrb.20208.

Matthews. PR: Hosker. TP: Rudenski. AS: Navlor. BA: Treacher. DF: Turner. RC. (1985). Homeostasis
model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and
insulin concentrations in man. Diabetologia 28: 412-419.
http://dx.doi.Org/10.1007/BF00280883.

Mattsson. K: Rignell-Hvdbom. A: Holmberg. S: Thelin. A: Tonsson. BA: Lindh. CH: Sehlstedt. A:

Rvlander. L. (2015). Levels of perfluoroalkyl substances and risk of coronary heart disease:

This document is a draft for review purposes only and does not constitute Agency policy.

R-19 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Findings from a population-based longitudinal study. Environ Res 142: 148-154.
http://dx.doi.Org/10.1016/i.envres.2015.06.033.

Mayer. C: Boehm. U. (2011). Female reproductive maturation in the absence of kisspeptin/GPR54
signaling. Nat Neurosci704-U297. http ://dx.doi.org/10.1038 /nn.2818.

McComb. 1: 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. http:/ /dx.doi.org/10.1007/sl 2403-019-00318-8.

McCoy. TA: Bangma. IT: Reiner. TL: Bowden. TA: Schnorr. 1: Slowev. M: O'Learv. T: Guillette. LI:

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. http://dx.doi.Org/10.1016/i.scitotenv.2017.06.137.

MDH (Minnesota Department of Health). (2019). Health Based Guidance for Water: Toxicological
Summary for: Perfluorohexane sulfonate (PFHxS).

https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfhx
s.pdf.

Meng. 0: Inoue. K: Ritz. B: Olsen. 1: Liew. Z. (2018). Prenatal exposure to perfluoroalkyl substances
and birth outcomes; an updated analysis from the danish national birth cohort Int J Environ
Res Public Health 15: 1832. http://dx.doi.org/10.3390/iierphl5091832.

Mi. X: Lin. SO: Zhang. XF: Li. 11: Pel. 1,1: Tin. F: Liao. 0: Xie. LM: Wei. LC: Hao. CI: Zhang. YW: Li. W.

(2022). Maternal perfluorinated compound exposure and risk of early pregnancy loss: A
nested case-control study [Letter], Biomed Environ Sci 35: 174-179.
http://dx.doi.org/10.3967/bes2022.026.

Miller. BH: Takahashi. IS. (2014). Central circadian control of female reproductive function
[Review], Front Endocrinol (Lausanne) 4: 195.
http://dx.doi.org/10.3389/fendo.2013.00195.

Mira-Escolano. MP: Mendiola. 1: Minguez-Alarcon. L: Melgareio. M: Cutillas-Tolin. A: Roca. M: Lopez-
Espin. 11: Noguera-Velasco. TA: Torres-Cantero. AM. (2014). Longer anogenital distance is
associated with higher testosterone levels in women: a cross-sectional study. BJOG 121:
1359-1364. http://dx.doi.org/10.llll/1471-0528.12627.

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. http://dx.doi.org/10.1016/i.envint.2018.03.033.

Monrov. 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. http://dx.doi.Org/10.1016/i.envres.2008.06.001.

Mora. AM: Fleisch. AF: Rifas-Shiman. SL: Woo Baidal. TA: 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.
http://dx.doi.Org/10.1016/i.envint.2017.ll.008.

Morgan. RL: Thayer. KA: Bero. L: Bruce. N: Falck-Ytter. Y: Ghersi. D: Guvatt. G: Hooiimans. C:

Langendam. M: Mandrioli. D: Mustafa. RA: Rehfuess. EA: Roonev. AA: Shea. B: Silbergeld. EK:
Sutton. P: Wolfe. MS: Woodruff. TT: Verbeek. TH: Hollowav. AC: Santesso. N: Schiinemann. HI.
(2016). GRADE: Assessing the quality of evidence in environmental and occupational health.
Environ Int92-93: 611-616. http://dx.doi.Org/10.1016/i.envint.2016.01.004.

Mvhr. BC: Lawlor. TE: Young. RR: Murli. H: Cifone. MA. (1990). Genotoxicity assessment of

perfluorodecanoic acid using a battery of in vitro and in vivo/in vitro assays (pp. 80). Myhr,
BC; Lawlor, TE; Young, RR; Murli, H; Cifone, MA.

Nakagawa. T: Ramdhan. DH: Tanaka. N: Naito. H: Tamada. H: Ito. Y: Li. Y: Havashi. Y: Yamagishi. N:
Yanagiba. Y: Aovama. T: Gonzalez. FT: Nakaiima. T. (2012). Modulation of ammonium

This document is a draft for review purposes only and does not constitute Agency policy.

R-20 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

perfluorooctanoate-induced hepatic damage by genetically different PPARa in mice. Arch
Toxicol 86: 63-74. http://dx.doi.Org/10.1007/s00204-011-0704-3.

Nataraian. S: Lipsitz. S. R.: Nietert. PI. (2002). Self-report of high cholesterol - Determinants of
validity in US adults. Am J Prev Med 23: 13-21. http://dx.doi.org/10.1016/s0749-
3797C02100446-4.

Needham. LL: Grandiean. P: Heinzow. B: l0rgensen. PI: Nielsen. F: Patterson. DG: Siodin. A: Turner.
WE: Weihe. P. (2011). Partition of environmental chemicals between maternal and fetal
blood and tissues. Environ SciTechnol 45: 1121-1126.
http://dx.doi.org/10.1021/esl019614.

Nelson. PL: Frazier. DE. Tr: Ericson. IE: Tarr. Ml: Mathes. LE. (1992). The effects of

perfluorodecanoic acid (PFDA) on humoral, cellular, and innate immunity in Fischer 344
rats. Immunopharmacol Immunotoxicol 14: 925-938.
http://dx.doi.org/10.3109/08923979209009242.

Nian. M: Li. 00: Bloom. M: Oian. ZM: Svberg. KM: Vaughn. MG: Wang. SO: Wei. 0: Zeeshan. M:

Gurram. N: Chu. C: Wang. 1: Tian. YP: Hu. LW: Liu. KK: Yang. BY: Liu. R0: 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. http://dx.doi.Org/10.1016/i.envres.2019.02.013.

Nigam. PK. (2011). Serum lipid profile: fasting or non-fasting? Indian J Clin Biochem 26: 96-97.
http://dx.doi.org/10.1007/sl 2291-010-0095-x.

NIH (National Institutes of Health). (2020). The PhenX Toolkit: protocol - lipid profile. Available
online at https://www.phenxtoolkit.org/protocols/view/40201 (accessed

Nilsson. H: Karrman. A: Westberg. H: Rotander. A: van Bavel. B: Lindstrom. G. (2010). A time trend
study of significantly elevated perfluorocarboxylate levels in humans after using fluorinated
ski wax. Environ Sci Technol 44: 2150-2155. http://dx.doi.org/10.1021/es9034733.

Niu. I: 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.
http://dx.doi.org/10.1186/sl2940-019-0493-3.

NLM (National Library of Medicine). (2013). HSDB: Perfluoro-n-nonanoic acid. Available online at
https://pubchem.ncbi.nlm.nih.gOv/source/hsdb/8040 (accessed

NLM (National Library of Medicine). (2017). HSDB: Perfluorohexanesulfonic acid. Available online
at https://pubchem.ncbi.nlm.nih.gov/source/hsdb/8274 (accessed

NTP (National Toxicology Program). (2005). Study information: CEBS accession number 002-
02652-0001-0000-8: Perfluorodecanoic acid. Available online at
https://tools.niehs.nih.gov/cebs3/ntpViews/?studyNumber=002-02 65 2-0001-0000-8
(accessed August 25, 2017).

NTP (National Toxicology Program). (2012). Study information: CEBS accession number 002-
02652-0003-0000-0: Perfluorodecanoic acid. Available online at
https://tools.niehs.nih.gov/cebs3/ntpViews/?studyNumber=002-02 65 2-0003-0000-0
(accessed August 25, 2017).

NTP (National Toxicology Program). (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://ntp.niehs.nih.gov/ntp/ohat/pfoa pfos/pfoa pfosmonograph 508.pdf.

NTP (National Toxicology Program). (2018). 28-day evaluation of the toxicity (C20615) of

perfluorodecanoic acid (PFDA) (335-76-2) on Harlan Sprague-Dawley rats exposed via
gavage [NTP], U.S. Department of Health and Human Services.
http://dx.doi.Org/10.22427/NTP-DATA-002-02652-0004-0000-l.

This document is a draft for review purposes only and does not constitute Agency policy.

R-21 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

NTP (National Toxicology Program). (2019). Toxicity studies of perfluoroalkyl carboxylates

administered by gavage to Sprague Dawley (Hsd:Sprague Dawley SD) Rats. Toxic Rep Ser.
http://dx.doi.org/10.22427/NTP-TOX-97.

O'Shaughnessv. KL: Thomas. SE: Spring. S. R: Ford. TL: Ford. RL: Gilbert. ME. (2019). A transient
window of hypothyroidism alters neural progenitor cells and results in abnormal brain
development Sci Rep 9: 4662. http://dx.doi.org/10.1038/s41598-019-4Q249-7.

OECD (Organisation for Economic Co-operation and Development). (2018). Toward a new

comprehensive global database of per- and polyfluoroalkyl substances (PFASs): Summary
report on updating the OECD 2007 list of per- and polyfluoroalkyl substances (PFASs).
(ENV/JM/MONO(2018)7). Paris, France, http://www.oecd.org/chemicalsafety/portal-
perfluorinated-chemicals/.

Ohmori. K: Kudo. N: Katavama. K: Kawashima. Y. (2003). Comparison of the toxicokinetics between
perfluorocarboxylic acids with different carbon chain length. Toxicology 184: 135-140.
http://dx.doi.Org/10.1016/S0300-483Xf02100573-5.

Okada. E: Sasaki. S: Kashino. I: Matsuura. H: Mivashita. C: Kobavashi. S: Itoh. K: Ikeno. T: Tamakoshi.
A: Kishi. R. (2014). Prenatal exposure to perfluoroalkyl acids and allergic diseases in early
childhood. Environ Int 65: 127-134. http://dx.doi.Org/10.1016/i.envint.2014.01.007.

Olsen. GW: Burris. TM: Ehresman. DT: Froehlich. TW: Seacat. AM: Butenhoff. TL: Zobel. LR. (2007).

Half-life of serum elimination of perfluorooctanesulfonate,perfluorohexanesulfonate, and
perfluorooctanoate in retired fluorochemical production workers. Environ Health Perspect
115: 1298-1305. http://dx.doi.org/10.1289/ehp.10009.

Olsen. GW: Mair. DC: Lange. CC: Harrington. LM: Church. TR: Goldberg. CL: Herron. RM: Hanna. H:
Nobiletti. IB: Rios. TA: Reagen. WK: Ley. CA. (2017). Per- and polyfluoroalkyl substances
(PFAS) in American Red Cross adult blood donors, 2000-2015. Environ Res 157: 87-95.
http://dx.doi.Org/10.1016/i.envres.2017.05.013.

Olson. CT: Andersen. ME. (1983). The acute toxicity of perfluorooctanoic and perfluorodecanoic
acids in male rats and effects on tissue fatty acids. Toxicol Appl Pharmacol 70: 362-372.
http://dx.doi.org/10.1016/0041 -008xC83190154-0.

Omoike. OE: Pack. RP: Mamudu. HM: Liu. Y: Strasser. S: Zheng. S: Okoro. 1: Wang. L. (2020).

Association between per and polyfluoroalkyl substances and markers of inflammation and
oxidative stress. Environ Res 196: 110361.
http://dx.doi.Org/10.1016/i.envres.2020.110361.

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.
http://dx.doi.Org/10.1016/i.envres.2020.110329.

Osmond. C: Barker. DT. (2000). Fetal, infant, and childhood growth are predictors of coronary heart
disease, diabetes, and hypertension in adult men and women [Review], Environ Health
Perspect 108: 545-553. http://dx.doi.org/10.2307/3454545.

Ou. Y: Zeng. X: Lin. S: Bloom. MS: Han. F: Xiao. X: Wang. H: Matala. R: Li. X: Ou. Y: Nie. Z: Dong. G: Liu.
X. (2021). Gestational exposure to perfluoroalkyl substances and congenital heart defects: A
nested case-control pilot study. Environ Int 154: 106567.
http://dx.doi.Org/10.1016/i.envint.2021.106567.

Oulhote. Y: Steuerwald. U: Debes. F: Weihe. P: Grandiean. P. (2016). Behavioral difficulties in 7-year
old children in relation to developmental exposure to perfluorinated alkyl substances
[Review], Environ Int 97: 237-245. http://dx.doi.Org/10.1016/i.envint.2016.09.015.

Passen. EL: Andersen. BR. (1986). Clinical tetanus despite a protective level of toxin-neutralizing
antibody [Case Report], JAMA 255: 1171-1173.
http://dx.doi.Org/10.1001/iama.1986.03370090093029.

Patel. TC: Mehta. BC. (1999). Tetanus: Study of 8,697 cases. Indian J Med Sci 53: 393-401.

This document is a draft for review purposes only and does not constitute Agency policy.

R-22 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Pelch. KE: Reade. A: Kwiatkowski. CF: Merced-Nieves. FM: Cavalier. H: Schultz. K: Wolffe. T:

Varshavskv. T. (2022). The PFAS-Tox Database: A systematic evidence map of health studies
on 29 per- and polyfluoroalkyl substances. Environ Int 167: 107408.
http://dx.doi.Org/10.1016/i.envint.2022.107408.

Perez. F: Llorca. M: Kock-Schulmever. M: Skrbic. B: Oliveira. LS: da Boit Martinello. K: Al-Dhabi. NA:
Antic. I: Farre. M: Barcelo. D. (2014). Assessment of perfluoroalkyl substances in food items
at global scale. Environ Res 135: 181-189. http://dx.doi.Org/10.1016/i.envres.2014.08.004.

Permadi. H: Lundgren. B: Andersson. K: Sundberg. C: Depierre. TW. (1993). Effects of perfluoro fatty
acids on peroxisome proliferation and mitochondrial size in mouse liver: dose and time
factors and effect of chain length. Xenobiotica 23: 761-770.
http://dx.doi.org/10.3109/00498259309166782.

Perng. W. ei: Rifas-Shiman. SL: Kramer. MS: Haugaard. LK: Oken. E: Gillman. MW: Belfort. MB.

(2016). Early Weight Gain, Linear Growth, and Mid-Childhood Blood Pressure A Prospective
Study in Project Viva. Hypertension 67: 301-308.
http://dx.doi.org/10.1161/HYPERTENSIONAHA.115.06635.

Petersen. KU: Haervig. KK: Flachs. EM: Bonde. TP: Lindh. C: Hougaard. KS: Toft. G: Ramlau-Hansen.
CH: T0ttenborg. SS. (2022). Per- and polyfluoroalkyl substances (PFAS) and male
reproductive function in young adulthood; a cross-sectional study. Environ Res 212:

113157. http://dx.doi.Org/10.1016/i.envres.2022.113157.

Poothong. S: Thomsen. C: Padilla-Sanchez. TA: Papadopoulou. E: Haug. LS. (2017). Distribution of
novel and well-known poly- and perfluoroalkyl substances (PFASs) in human serum,
plasma, and whole blood. Environ SciTechnol 51: 13388-13396.
http://dx.doi.org/10.1021/acs.est.7b03299.

Powers. RH: Aust. SD. (1986). The effects of nonadecafluoro-n-decanoic acid on serum retinol and
hepatic retinyl palmitate hydrolase activity in male Sprague-Dawley rats. J Biochem Toxicol
1: 27-42. http://dx.doi.Org/10.1002/ibt2570010204.

Oin. XD: Oian. Z: Vaughn. MG: Huang. 1: Ward. P: Zeng. XW: Zhou. Y: Zhu. Y: Yuan. P: Li. M: Bai. Z:

Paul. G: Hao. YT: Chen. W: Chen. PC: Dong. GH: Lee. YL. (2016). Positive associations of
serum perfluoroalkyl substances with uric acid and hyperuricemia in children from Taiwan.
Environ Pollut 212: 519-524. http://dx.doi.Org/10.1016/i.envpol.2016.02.050.

Rahman. ML: Zhang. C: Smarr. MM: Lee. S: Honda. M: Kannan. K: Tekola-Avele. 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.
http://dx.doi.Org/10.1016/i.envint.2019.01.027.

Reardon. ATF: Khodavari Moez. E: Dinu. I: Goruk. S: Field. CI: Kinniburgh. DW: Macdonald. AM:
Martin. TW: 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. http://dx.doi.Org/10.1016/i.envint.2019.04.023.

Ren. XM: Oin. WP: Cao. LY: Zhang. 1: Yang. Y: Wan. B: Guo. LH. (2016). Binding interactions of
perfluoroalkyl substances with thyroid hormone transport proteins and potential
toxicological implications. Toxicology 366-367: 32-42.
http://dx.doi.0rg/lO.lOl6/i.tox.2Ol6.O8.Oll.

Ren. XM: Zhang. YF: Guo. LH: Oin. ZF: Lv. OY: Zhang. LY. (2015). Structure-activity relations in

binding of perfluoroalkyl compounds to human thyroid hormone T3 receptor. Arch Toxicol
89: 233-242. http://dx.doi.Org/10.1007/s00204-014-1258-v.

Ren. Y: Tin. L: Yang. F: Liang. H: Zhang. Z: Du. 1: Song. X: Miao. M: Yuan. W. (2020). Concentrations of
perfluoroalkyl and polyfluoroalkyl substances and blood glucose in pregnant women.
Environ Health 19: 88. http://dx.doi.org/10.1186/sl2940-020-0064Q-8.

This document is a draft for review purposes only and does not constitute Agency policy.

R-23 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Ribeiro-Filho. T: Piuvezam. MR: Leite. FC: Costa. HF: Calheiros. AS: Torres. RC: de Azevedo. CT:

Martins. MA: Dias. C. daS: Bozza. PT. (2014). Curine inhibits mast cell-dependent responses
in mice. J Ethnopharmacollll8-1124. http://dx.doi.Org/10.1016/i.iep.2014.06.041.

Robledo. CA: Yeung. E: Mendola. P: Sundaram. R: Maisog. 1: Sweeney. AM: Barr. DB: Louis. GM.

(2015). Preconception maternal and paternal exposure to persistent organic pollutants and
birth size: the LIFE study. Environ Health Perspect 123: 88-94.
http://dx.doi.org/10.1289/ehp.1308Q16.

Rogers. AM: Andersen. ME: Back. KC. (1982). Mutagenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin
and perfluoro-n-decanoic acid in L5178Y mouse-lymphoma cells. MutatRes 105: 445-449.
http://dx.doi.org/10.1016/0165-7992r82190192-0.

Rosen. MB: Das. KP: Roonev. 1: Abbott. B: Lau. C: Corton. TC. (2017). PPARa-independent

transcriptional targets of perfluoroalkyl acids revealed by transcript profiling. Toxicology
387: 95-107. http://dx.doi.org/10.1016/i.tox.2017.05.013.

Rosen. MB: Das. KP: Wood. CR: Wolf. CT: Abbott. BP: Lau. C. (2013). Evaluation of perfluoroalkyl acid
activity using primary mouse and human hepatocytes. Toxicology 308: 129-137.
http://dx.doi.Org/10.1016/i.tox.2013.03.011.

Rosen. MB: Lee. IS: Ren. H: Vallanat. B: Liu. 1: Waalkes. MP: Abbott. BP: Lau. C: Corton. TC. (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. http: //dx.doi.org/10.1093 /toxsci/kfn025.

Rosen. MB: Schmid. TR: Corton. TC: Zehr. RP: Pas. KP: Abbott. BP: Lau. C. (2010). Gene expression
profiling in wild-type and PPAR alpha-null mice exposed to perfluorooctane sulfonate
reveals PPAR alpha-independent effects. PPAR Research 2010: 794739.
http: / /dx.doi.org/10.1155/2010/794739.

Rosenmai. AK: Ahrens. L: le Godec. T: Lundqvist. 1: Oskarsson. A. (2018). Relationship between

peroxisome proliferator-activated receptor alpha activity and cellular concentration of 14
perfluoroalkyl substances in HepG2 cells. J Appl Toxicol 38: 219-226.
http: / /dx. doi. o r g /10.10 0 2 /i at. 3 515.

Routti. H: Berg. MK: Lille-Lang0v. R: 0vgarden. L: Hariu. M: Pietz. R: Sonne. C: Goks0vr. A. (2019).
Environmental contaminants modulate the transcriptional activity of polar bear (Ursus
maritimus) and human peroxisome proliferator-activated receptor alpha (PPARA). Sci Rep
9: 6918. http://dx.doi.org/10.1038/s41598-019-43337-w.

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. http: / /dx. doi. o r g /10.10 9 3/aie/kwx332.

Salazar-Martinez. E: Romano-Riquer. P: Yanez-Marquez. E: Longnecker. MP: Hernandez-Avila. M.
(2004). Anogenital distance in human male and female newborns: a descriptive, cross-
sectional study. Environ Health 3:8-13. http://dx.doi.org/10.1186/1476-069x-3-8.

Salihovic. S: Stubleski. 1: 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. http://dx.doi.Org/10.1016/i.envint.2018.04.052.

Sathyanaravana. S: Beard. L: Zhou. C: Grady. R. (2010). Measurement and correlates of ano-genital
distance in healthy, newborn infants. Int J Androl 33: 317-323.
http://dx.doi.Org/10.llll/i.1365-2605.2009.01044.x.

Schecter. A: Malik-Bass. N: Calafat. AM: Kato. K: Colacino. TA: Gent. TL: Hvnan. LS: Harris. TR: Malla.
S: Birnbaum. L. (2012). Polyfluoroalkyl compounds in Texas children from birth through 12
years of age. Environ Health Perspect 120: 590-594.
http://dx.doi.org/10.1289/ehp.1104325.

This document is a draft for review purposes only and does not constitute Agency policy.

R-24 PRAFT-PO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Schiinemann. H: Hill. S: Guvatt. G: Akl. EA: Ahmed. F. (2011). The GRADE approach and Bradford
Hill's criteria for causation. J Epidemiol Community Health 65: 392-395.
http://dx.doi.org/10.1136/iech.2010.119933.

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. http://dx.doi.Org/10.1016/i.chemosphere.2020.127446.

Scott. HM: Mason. II: Sharpe. RM. (2009). Steroidogenesis in the fetal testis and its susceptibility to
disruption by exogenous compounds [Review], Endocr Rev 30: 883-925.
http://dx.doi.Org/10.1210/er.2009-0016.

Selgrade. MK. (2007). Immunotoxicity: The risk is real [Review], Toxicol Sci 100: 328-332.
http://dx.doi.org/10.1093/toxsci/kfm244.

Sellers. RS: Mortan. D: Michael. B: Roome. N: Tohnson. IK: Yano. BL: Perry. R: Schafer. K. (2007).
Society of toxicologic pathology position paper: Organ weight recommendations for
toxicology studies [Review], Toxicol Pathol 35: 751-755.
http://dx.doi.Org/10.1080/01926230701595300.

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. http://dx.doi.Org/10.1016/i.envint.2018.01.025.

Shah-Kulkarni. S: Kim. BM: Hong. YC: Kim. HS: Kwon. El: Park. H: Kim. YT: Ha. EH. (2016). Prenatal
exposure to perfluorinated compounds affects thyroid hormone levels in newborn girls.
Environ Int 94: 607-613. http://dx.doi.Org/10.1016/i.envint2016.06.024.

Sharpe. RM. (2010). Environmental/lifestyle effects on spermatogenesis [Review], Philos Trans R
Soc Lond B Biol Sci 365: 1697-1712. http://dx.doi.org/10.1098/rstb.2009.0206.

Shi. Y: Yang. L: Li. I: Lai. 1: Wang. Y: Zhao. Y: Wu. Y. (2017). Occurrence of perfluoroalkyl substances
in cord serum and association with growth indicators in newborns from Beijing.
Chemosphere 169: 396-402. http://dx.doi.Org/10.1016/i.chemosphere.2016.ll.050.

Shih. YH: Blomberg. AT: Bind. MA: Holm. D: Nielsen. F: Heilmann. C: Weihe. P: Grandiean. 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.
http://dx.doi.org/10.1080/1547691X.2021.1922957.

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.
http://dx.doi.Org/10.1016/i.envres.2020.109514.

Shinwell. ES: Shlomo. M. (2003). Measured length of normal term infants changes over the first two
days of life. J Pediatr Endocrinol Metab 16: 537-540.
http://dx.doi.Org/10.1515/ipem.2003.16.4.537.

Sigman. M: Lipshultz. LI: Howards. SS. (1997). Evaluation of the subfertile male. In LI Lipschultz; SS
Howards (Eds.), Infertility in the male (3rd ed., pp. 173). St. Louis, MO: Mosby.

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. http://dx.doi.Org/10.1016/i.envres.2018.05.019.

Skogheim. TS: Villanger. GD: Wevde. KVF: Engel. SM: Suren. P: 0ie. MG: Skogan. AH: Biele. G: Zeiner.
P: Overgaard. KR: Haug. LS: Sabaredzovic. A: Aase. H. (2020). 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. http://dx.doi.Org/10.1016/i.iiheh.2019.10.003.

Skogheim. TS: Wevde. K. iell Vegard F: Aase. H: Engel. SM: Suren. P: Pie. MG: Biele. G: Reichborn-

Kiennerud. T: Brantsaeter. AL: Haug. LS: Sabaredzovic. A: Auveung. B: Villanger. GD. (2021).

This document is a draft for review purposes only and does not constitute Agency policy.

R-25 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Prenatal exposure to per- and polyfluoroalkyl substances (PFAS) and associations with
attention-deficit/hyperactivity disorder and autism spectrum disorder in children. Environ
Res 202: 111692. http://dx.doi.Org/10.1016/i.envres.2021.111692.

Smit. LA: Lenters. V: Haver. BB: Lindh. CH: Pedersen. HS: Liermontova. I: Tonsson. BA: Piersma. AH:
Bonde. TP: 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.
http: / /dx. doi. o r g /10.1111 /all. 12 60 5.

Soldatow. VY: Lecluvse. EL: Griffith. LG: Rusvn. I. (2013). In vitro models for liver toxicity testing.
Toxicology Research 2: 23-39. http: //dx.doi.org/10.1039 /c2tx20051a.

Soukup. T: Zacharova. G: Smerdu. V: Tirmanova. I. (2001). Body, heart, thyroid gland and skeletal
muscle weight changes in rats with altered thyroid status. Physiol Res 50: 619-626.

Stahl. LL: Snyder. BP: Olsen. AR: Kincaid. TM: Wathen. IB: Mccarty. HB. (2014). Perfluorinated

compounds in fish from U.S. urban rivers and the Great Lakes. Sci Total Environ 499: 185-
195. http://dx.doi.Org/10.1016/i.scitotenv.2014.07.126.

Starling. AP: Adgate. TL: Hamman. RF: Kechris. K: Calafat. AM: Dabelea. D. (2019). Prenatal exposure
to per- and polyfluoroalkyl substances and infant growth and adiposity: The healthy start
study. Environ Int 131: 104983. http://dx.doi.Org/10.1016/i.envint2019.104983.

Starling. AP: Adgate. TL: 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. http://dx.doi.org/10.1289/EHP641.

Starling. AP: Engel. SM: Richardson. DB: Baird. DP: Haug. LS: Stuebe. AM: Klungsovr. K: Harmon. 0:
Becher. G: Thomsen. C: Sabaredzovic. A: Eggesbo. M: Hoppin. TA: Travlos. GS: Wilson. RE:
Trogstad. LI: Magnus. P. er: Longnecker. MP. (2014a). Perfluoroalkyl Substances During
Pregnancy and Validated Preeclampsia Among Nulliparous Women in the Norwegian
Mother and Child Cohort Study. Am J Epidemiol 179: 824-833.
http://dx.d0i.0rg/l 0.1093 /aie /kwt43 2.

Starling. AP: Engel. SM: Whitworth. KW: Richardson. DB: Stuebe. AM: Daniels. TL: Haug. LS: Eggesba.
M: Becher. G: Sabaredzovic. A: Thomsen. C: Wilson. RE: Travlos. GS: Hoppin. TA: Baird. DP:
Longnecker. MP. (2014b). Perfluoroalkyl substances and lipid concentrations in plasma
during pregnancy among women in the Norwegian Mother and Child Cohort Study. Environ
Int62: 104-112. http://dx.doi.Org/10.1016/i.envint2013.10.004.

Steenland. K: Barry. V: Savitz. D. (2018). Serum perfluorooctanoic acid and birthweight: an updated
meta-analysis with bias analysis. Epidemiology 29: 765-776.
http://dx.d0i.0rg/l 0.1097/EPF.0000000000000903.

Sterchele. PF: Vanden Heuvel. TP: Davis. TW: Shrago. E: Knudsen. 1: Peterson. RE. (1994). Induction
of hepatic acyl-CoA-binding protein and liver fatty acid-binding protein by
perfluorodecanoic acid in rats. Lack of correlation with hepatic long-chain acyl-CoA levels.
Biochem Pharmacol 48: 955-966. http://dx.doi.org/10.1016/0006-2952f9419Q366-2.

Strvnar. MI: Lindstrom. AB. (2008). Perfluorinated compounds in house dust from Ohio and North
Carolina, USA. Environ Sci Technol 42: 3751-3756. http://dx.doi.org/10.1021/es7032Q58.

Stubleski. I: Salihovic. S: Lind. L: Lind. PM: van Bavel. B: Karrman. A. (2016). Changes in serum

levels of perfluoroalkyl substances during a 10-year follow-up period in a large population-
based cohort Environ Int 95: 86-92. http://dx.doi.Org/10.1016/i.envint.2016.08.002.

Sun. 0: Zong. G: Valvi. P: Nielsen. F: Coull. B: Grandiean. 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. http://dx.doi.org/10.1289/EHP2619.

Swedish Environmental Protection Agency. (2017). [Rapport till Naturvardsverkets halsorelaterade
miljoovervakning: Utvardering av samband mellan mammors POP-belastning under
graviditets- och amningsperioden och deras barns halsa], (2215-17-008).

This document is a draft for review purposes only and does not constitute Agency policy.

R-26 PRAFT-PO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Sweeney. MF: Hasan. N: Soto. AM: Sonnenschein. C. (2015). Environmental endocrine disruptors:
Effects on the human male reproductive system [Review], Rev Endocr Metab Disord 16:
341-357. http://dx.doi.org/10.1007/slll54-016-9337-4.

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 DNAof rats. Cancer Lett 57: 55-60.
http://dx.doi.Org/10.1016/0304-3835f91190063-N.

Takagi. A: Sai. K: Umemura. T: Hasegawa. R: Kurokawa. Y. (1992). Hepatomegaly is an early

biomarker for hepatocarcinogenesis induced by peroxisome proliferators. J Environ Pathol
Toxicol Oncol 11: 145-149.

Thankamonv. A: Pasterski. V: Ong. KK: Acerini. CL: Hughes. IA. (2016). Anogenital distance as a
marker of androgen exposure in humans. Andrology 0: 1-10.
http://dx.d0i.0rg/l 0.1111 /andr. 12156.

Thomsen. ML: Henriksen. LS: Tinggaard. 1: Nielsen. F: Tensen. TK: Main. KM. (2021). Associations
between exposure to perfluoroalkyl substances and body fat evaluated by DXA and MRI in
109 adolescent boys. Environ Health 20: 73. http://dx.doi.org/10.1186/sl2940-021-
00758-3.

Tian. Y: Liang. H: Miao. M: Yang. F: Ti. 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.
http://dx.doi.org/10.1093/humrep/dez058.

Tian. Y: Miao. M: Ti. 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. http://dx.doi.Org/10.1016/i.envpol.2020.115426.

Timmermann. CA: Budtz-l0rgensen. E: Tensen. TK: Osuna. CE: Petersen. MS: Steuerwald. U: Nielsen.
F: Poulsen. LK: Weihe. P: Grandiean. P. (2017). Association between perfluoroalkyl
substance exposure and asthma and allergic disease in children as modified by MMR
vaccination. J Immunotoxicol 14: 39-49.
http://dx.doi.org/10.1080/1547691X.2016.1254306.

Timmermann. CAG: Andersen. MS: Budtz-l0rgensen. E: Bove. H: Nielsen. F: Tensen. RC: Bruun. S:
Husbv. S: Grandiean. P: Tensen. TK. (2022). Pregnancy exposure to perfluoroalkyl
substances and associations with prolactin concentrations and breastfeeding in the Odense
Child Cohort. J Clin Endocrinol Metab 107: e631-e642.
http://dx.doi.org/10.1210/clinem/dgab638.

Timmermann. CAG: Pedersen. HS: Weihe. P: Bierregaard. P: Nielsen. F: Heilmann. C: Grandiean. 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. http://dx.d0i.0rg/l0.1016/i.envres.2021.111712.

Toor. TM: Sikka. SC. (2017). Developmental and reproductive disorders: Role of endocrine
disruptors in testicular toxicity. In RC Gupta (Ed.), Reproductive and developmental
toxicology (2nded., pp. 1111-1121). San Diego, CA: Academic Press.
http://dx.d0i.0rg/l 0.1016/B978-0-12-804239-7.00059-7.

Tropp. 1: Markus. El. (2001). Effects of mild food deprivation on the estrous cycle of rats. Physiol
Behav 73: 553-559. http://dx.doi.org/10.1016/s0031-9384f01100487-5.

Truong. L: Rericha. Y: Thunga. P: Marvel. S: Wallis. D: Simonich. MT: Field. TA: Cao. D: Reif. DM:

Tanguav. RL. (2022). Systematic developmental toxicity assessment of a structurally diverse
library of PFAS in zebrafish. J Hazard Mater 431: 128615.
http://dx.doi.Org/10.1016/i.ihazmat.2022.128615.

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

This document is a draft for review purposes only and does not constitute Agency policy.

R-27 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

breast cancer in Taiwanese women. Environ Int 142: 105850.
http://dx.doi.Org/10.1016/i.envint.2020.105850.

U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental toxicity risk
assessment Fed Reg 56: 63798-63826.

U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
reference concentrations and application of inhalation dosimetry [EPA Report],
(EPA600890066F). Research Triangle Park, NC.

https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=71993&CFID=51174829&CFTOKE
N=25006317.

U.S. EPA (U.S. Environmental Protection Agency). (1996a). Guidelines for reproductive toxicity risk
assessment [EPA Report], (EPA/630/R-96/009). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=3 00 Q4Y0B.txt.

U.S. EPA (U.S. Environmental Protection Agency). (1996b). Guidelines for reproductive toxicity risk
assessment Fed Reg 61: 56274-56322.

U.S. EPA (U.S. Environmental Protection Agency). (1998). Health effects test guidelines OPPTS

870.7800 immunotoxicity. (EPA 712-C-98-351). Washington, DC: Prevention, Pesticides and
Toxic Substances, U.S. Environmental Protection Agency.

http://www.epa.gov/ocspp/pubs/frs/publications/Test Guidelines/series870.htm.

U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and
reference concentration processes. (EPA630P02002F). Washington, DC.
https://www.epa.gov/sites/production/files/2014-12/documents/rfd-final.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2003). Toxicological review of methyl ethyl
ketone (CAS No. 78-93-3): In support of summary information on the Integrated Risk
Information System (IRIS). (EPA 635/R-03/009). Washington, DC.
http: / /cfpub.epa.gov/ncea/iris /iris documents/documents/toxreviews /0071tr.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2004). Toxicological review of boron and

compounds. In support of summary information on the Integrated Risk Information System
(IRIS) [EPA Report], (EPA/635/04/052). Washington, DC: U.S. Environmental Protection
Agency, IRIS. http://nepis.epa.gov/exe/ZyPURL.cgi?Dockey=Pl 006CK9.txt

U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment
[EPA Report], (EPA630P03001F). Washington, DC.
https://www.epa.gov/sites/production/files/2013-
09/documents/cancer guidelines final 3-25-05.pdf.

U.S. EPA fU.S. Environmental Protection Agency). (2011). Recommended use of body weight 3/4 as
the default method in derivation of the oral reference dose. (EPA100R110001). Washington,
DC. https://www.epa.gov/sites/production/files/2013-09/documents/recommended-use-
of-bw34.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2012a). Benchmark dose technical guidance
[EPA Report], (EPA100R12001). Washington, DC: U.S. Environmental Protection Agency,
Risk Assessment Forum, https: //www.epa.gov/risk/benchmark-dose-technical-guidance.

U.S. EPA (U.S. Environmental Protection Agency). (2012b). Toxicological review of tetrahydrofuran.
In support of summary information on the integrated risk information system (IRIS) (pp. 1-
207). (EPA/635/R-11/006F). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2014). Guidance for applying quantitative data to
develop data-derived extrapolation factors for interspecies and intraspecies extrapolation
[EPAReport], (EPA/100/R-14/002F). Washington, DC: Risk Assessment Forum, Office of
the Science Advisor, https://www.epa.gov/sites/production/files/2015-
01 /documents /ddef-final.pdf.

This document is a draft for review purposes only and does not constitute Agency policy.

R-28 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

U.S. EPA (U.S. Environmental Protection Agency). (2016a). Health effects support document for

perfluorooctane sulfonate (PFOS) [EPA Report], (EPA 822-R-16-002). Washington, DC: U.S.
Environmental Protection Agency, Office of Water, Health and Ecological Criteria Division.
https://www.epa.gov/sites/production/files/2016-
05/documents/pfos hesd final 508.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2016b). Health effects support document for
perfluorooctanoic acid (PF0A) [EPA Report], (EPA 822-R-16-003). Washington, DC: U.S.
Environmental Protection Agency, Office of Water, Health and Ecological Criteria Division.
https://www.epa.gOv/sites/production/files/2016-05/documents/pfoa hesd final-
plain.pdf.

U.S. EPA (U.S. Environmental Protection Agency). (2016c). The Third Unregulated Contaminant
Monitoring Rule. Available online at https: //www.epa.gov/dwucmr/third-unregulated-
contaminant-monitoring-rule (accessed
U.S. EPA (U.S. Environmental Protection Agency). (2018). Human health toxicity values for
perfluorobutane sulfonic acid (CASRN 375-73-5) and related compound potassium
perfluorobutane sulfonate (CASRN 29420-49-3) (public comment draft). (EPA-823-R-18-
307). Washington, DC.

U.S. EPA (U.S. Environmental Protection Agency). (2019a). ChemView [Database], Retrieved from

https://chemview.epa.gov/chemview
U.S. EPA (U.S. Environmental Protection Agency). (2019b). CompTox Chemicals Dashboard
[Database], Research Triangle Park, NC. Retrieved from
https://comptox.epa.gov/dashboard
U.S. EPA (U.S. Environmental Protection Agency). (2021a). CompTox chemicals dashboard.

Washington, DC. Retrieved from https: //comptox.epa.gov/dashboard
U.S. EPA (U.S. Environmental Protection Agency). (2021b). Human health toxicity values for
perfluorobutane sulfonic acid (CASRN 375-73-5) and related compound potassium
perfluorobutane sulfonate (CASRN 29420-49-3) [EPA Report], (EPA/600/R-20/345F).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development. https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=350888.

U.S. EPA (U.S. Environmental Protection Agency). (2022). IRIS toxicological review of

perfluorobutanoic acid (PFBA) and related salts. (EPA/635/R-22/277F).

Ulhaq. M: Carlsson. G: Orn. S: Norrgren. L. (2013). Comparison of developmental toxicity of seven
perfluoroalkyl acids to zebrafish embryos. Environ Toxicol Pharmacol 36: 423-426.
http://dx.doi.Org/10.1016/i.etap.2013.05.004.

Unkila. M: Pohianvirta. R: Tuomisto. 1. (1992). Acute toxicity of perfluorodecanoic acid and cobalt
protoporphyrin in aTCDD-sensitive and a TCDD-resistant rat strain. Chemosphere 25:
1233-1238. http://dx.doi.org/10.1016/0045-6535f92190135-E.

Valvi. D: Hailund. K: Coull. BA: Nielsen. F: Weihe. P: Grandiean. P. (2021). Life-course Exposure to
Perfluoroalkyl Substances in Relation to Markers of Glucose Homeostasis in Early
Adulthood. J Clin Endocrinol Metab 106: 2495-2504.
http://dx.doi.org/10.1210/clinem/dgab267.

Valvi. D: Oulhote. Y: Weihe. P: Dalgard. C: Bierve. KS: Steuerwald. U: Grandiean. P. (2017).

Gestational diabetes and offspring birth size at elevated environmental pollutant exposures.
Environ Int 107: 205-215. http://dx.doi.Org/10.1016/i.envint.2017.07.016.
van Kerkhof. LWM: Van Dvcke. KCG: Tansen. EHT. M: Beekhof. PK: van Oostrom. CTM: Ruskovska. T:
Velickova. N: Kamcev. N: Pennings. TLA: van Steeg. H: Rodenburg. W. (2015). Diurnal
variation of hormonal and lipid biomarkers in a molecular epidemiology-like setting. PLoS
ONE 10: e0135652. http://dx.doi.org/10.1371/iournal.pone.0135652.

This document is a draft for review purposes only and does not constitute Agency policy.

R-29 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Van Rafelghem. MT: Andersen. ME. (1988). The effects of perfluorodecanoic acid on hepatic

stearoyl-coenzyme A desaturase and mixed function oxidase activities in rats. Fundam Appl
Toxicol 11: 503-510. http://dx.doi.org/10.1016/0272-0590f88190114-5.

Van Rafelghem. MT: Inhorn. SL: Peterson. RE. (1987a). Effects of perfluorodecanoic acid on thyroid
status in rats. Toxicol Appl Pharmacol 87: 430-439. http://dx.doi.org/10.1016/0Q41-
008Xf87190248-l.

Van Rafelghem. MT: Mattie. PR: Bruner. RH: Andersen. ME. (1987b). Pathological and hepatic
ultrastructural effects of a single dose of perfluoro-n-decanoic acid in the rat, hamster,
mouse, and guinea pig. Fundam Appl Toxicol 9: 522-540. http://dx.doi.org/10.1016/Q272-
0590f87190034-0.

Van Rafelghem. MT: Noren. CW: Menahan. LA: Peterson. RE. (1988a). Interrelationships between
energy and fat metabolism and hypophagia in rats treated with perfluorodecanoic acid.
Toxicol Lett 40: 57-69. http://dx.doi.org/10.1016/0378-4274C88190183-X.

Van Rafelghem. MT: Vanden Heuvel. TP: Menahan. LA: Peterson. RE. (1988b). Perfluorodecanoic acid
and lipid metabolism in the rat. Lipids 23: 671-678.
http://dx.doi.org/10.1007/BF02535666.

Vanden Heuvel. TP: Kuslikis. BI: Peterson. RE. (1992). Covalent binding of perfluorinated fatty acids
to proteins in the plasma, liver and testes of rats. Chem Biol Interact 82: 317-328.
http: //dx.doi.org/10.1016/0009-2797r92190003-4.

Vanden Heuvel. TP: Kuslikis. BI: Van Rafelghem. MT: Peterson. RE. (1991). Disposition of

perfluorodecanoic acid in male and female rats. Toxicol Appl Pharmacol 107: 450-459.
http://dx.doi.Org/10.1016/0041-008Xf91190308-2.

Varshavskv. TR: Robinson. IF: Zhou. Y: Puckett. KA: Kwan. E: Buarpung. S: Aburaiab. R: Gaw. SL: Sen.
S: Gao. SM: Smith. SC: Park. IS: Zakharevich. I: Gerona. RR: Fisher. ST: Woodruff. TT. (2021).
Organophosphate flame retardants, highly fluorinated chemicals, and biomarkers of
placental development and disease during mid-gestation. Toxicol Sci 181: 215-228.
http://dx.doi.org/10.1093/toxsci/kfab028.

Vasilenko. P: Adams. WC: Frieden. EH. (1981). Uterine size and glycogen-content in cycling and
pregnant rats - Influence of relaxin. Biol Reprod 25:162-169.
http://dx.doi.Org/10.1095/biolreprod25.l.162.

Velarde. MC: Chan. AFO: Saio. MET. V: Zakharevich. I: Melamed. 1: Uv. GLB: Teves. TMY: Corachea.
ATM: 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 Ptl: 131545.
http://dx.doi.Org/10.1016/i.chemosphere.2021.131545.

Verner. MA: Longnecker. MP. (2015). Comment on "enhanced elimination of

perfluorooctanesulfonic acid by menstruating women: evidence from population-based
pharmacokinetic modeling" [Letter], Environ Sci Technol 49: 5836-5837.
http://dx.doi.org/10.1021/acs.est.5b00187.

Vestergaard. S: Nielsen. F: Andersson. AM: Hi0llund. NH: Grandiean. P: Andersen. HR: Tensen. TK.
(2012). Association between perfluorinated compounds and time to pregnancy in a
prospective cohort of Danish couples attempting to conceive. Hum Reprod 27: 873-880.
http://dx.doi.org/10.1093/humrep/der450.

Vuong. A: Yolton. K: Webster. GM: Siodin. A: Calafat. AM: Braun. TM: 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.
http://dx.doi.0rg/lO.lOl6/i.envres.2Ol6.Ol.OO8.

Vuong. AM: Yolton. K: Wang. Z: Xie. C: Webster. GM: Ye. X: Calafat. AM: Braun. TM: Dietrich. KN:

Lanphear. BP: Chen. A. (2018). Childhood perfluoroalkyl substance exposure and executive

This document is a draft for review purposes only and does not constitute Agency policy.

R-30 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

function in children at 8 years. Environ Int 119: 212-219.
http://dx.doi.Org/10.1016/i.envint.2018.06.028.

Wahlang. B: Beier. TI: Clair. HB: Bellis-Tones. HT: Falkner. K: Mcclain. CT: Cave. MC. (2013). Toxicant-
associated steatohepatitis [Review], Toxicol Pathol 41: 343-360.
http://dx.doi.org/10.1177/0192623312468517.

Walaas. 0. (1952). Effect of oestrogens on the glycogen content of the rat uterus. European Journal
of Endocrinology 10: 175-192. http: //dx.doi.Org/10.1530 /acta.0.0100175.

Wan. HT: Mruk. DP: Wong. CK: Cheng. CY. (2013). Targeting testis-specific proteins to inhibit

spermatogenesis: lesson from endocrine disrupting chemicals [Review], Expert Opin Ther
Targets 17: 839-855. http://dx.doi.org/10.1517/14728222.2013.791679.

Wang. B: Fu. 1: Gao. K: Liu. 0: Zhuang. L: Zhang. G: Long. M: Na. 1: Ren. M: Wang. A: Liang. R: Shen. G:
Li. Z: Lu. 0. (2021). Early pregnancy loss: Do Per- and polyfluoroalkyl substances matter?
Environ Int 157: 106837. http://dx.doi.Org/10.1016/i.envint.2021.106837.

Wang. B: Zhang. R: Tin. F: Lou. H: Mao. Y: Zhu. W: Zhou. W: Zhang. P: Zhang. 1. (2017). Perfluoroalkyl
substances and endometriosis-related infertility in Chinese women. Environ Int 102: 207-
212. http://dx.doi.org/10.1016/i.envint.2017.03.003.

Wang. D: Gao. 0: Wang. T: Kan. Z: Li. X: Hu. L: Peng. CY: Oian. F: Wang. Y: Granato. D. (2020). Green
tea polyphenols and epigallocatechin-3-gallate protect against perfluorodecanoic acid
induced liver damage and inflammation in mice by inhibiting NLRP3 inflammasome
activation. Food Res Int 127: 108628. http://dx.doi.Org/10.1016/i.foodres.2019.108628.

Wang. I: Zeng. XW: Bloom. MS: Oian. Z: Hinvard. LI: Belue. R: Lin. S: Wang. SO: Tian. YP: Yang. M:
Chu. C: Gurram. N: Hu. LW: Liu. KK: Yang. BY: Feng. D: Liu. RO: Dong. GH. (2019). Renal
function and isomers of perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS):
Isomers of C8 Health Project in China. Chemosphere 218: 1042-1049.
http://dx.doi.Org/10.1016/i.chemosphere.2018.ll.191.

Wang. Y: Adgent. M: Su. PH: Chen. HY: Chen. PC: Hsiung. CA: Wang. SL. (2016). Prenatal exposure to
perfluorocarboxylic acids (PFCAs) and fetal and postnatal growth in the Taiwan maternal
and infant cohort study. Environ Health Perspect 124: 1794-1800.
http://dx.doi.org/10.1289/ehp.1509998.

Wang. Y: Rogan. WT: 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. http: //dx.doi.Org/10.1016/i.iiheh.2015.07.002.

Wang. Y: Rogan. WT: Chen. PC: Lien. GW: Chen. HY: Tseng. YC: Longnecker. MP: Wang. SL. (2014a).
Association between maternal serum perfluoroalkyl substances during pregnancy and
maternal and cord thyroid hormones: Taiwan maternal and infant cohort study. Environ
Health Perspect 122: 529-534. http://dx.doi.org/10.1289/ehp.1306925.

Wang. Y: Starling. AP: Haug. LS: Eggesbo. M: Becher. G: Thomsen. C: Travlos. G: King. D: Hoppin. TA:
Rogan. WT: Longnecker. MP. (2013). Association between perfluoroalkyl substances and
thyroid stimulating hormone among pregnant women: a cross-sectional study. Environ
Health 12: 76. http://dx.doi.org/10.1186/1476-069X-12-76.

Wang. Y: Zhang. L: Teng. Y: Zhang. 1: Yang. L: Li. I: Lai. 1: 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. http://dx.doi.Org/10.1016/i.ies.2018.03.016.

Wang. Z: Cousins. IT: Scheringer. M: Buck. RC: Hungerbiihler. K. (2014b). Global emission

inventories for C4-C14 perfluoroalkyl carboxylic acid (PFCA) homologues from 1951 to
2030, Part I: production and emissions from quantifiable sources [Review], Environ Int 70:
62-75. http://dx.doi.Org/10.1016/i.envint2014.04.013.

Wang. Z: Shi. R: Ding. G: Yao. 0: Pan. C: Gao. Y: Tian. Y. (2022). Association between maternal serum
concentration of perfluoroalkyl substances (PFASs) at delivery and acute infectious diseases

This document is a draft for review purposes only and does not constitute Agency policy.

R-31 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

in infancy. Chemosphere 289: 133235.
http://dx.doi.Org/10.1016/i.chemosphere.2021.133235.

Washington. TW: Yoo. H: Ellington. TT: Tenkins. TM: Libelo. EL. (2010). Concentrations, distribution,
and persistence of perfluoroalkylates in sludge-applied soils near Decatur, Alabama, USA.
Environ Sci Technol 44: 8390-8396. http://dx.doi.org/10.1021/eslQ03846.

Watkins. DT: Tosson. 1: Elston. B: Bartell. SM: Shin. HM: Vieira. VM: Savitz. DA: Fletcher. T: Wellenius.
GA. (2013). Exposure to perfluoroalkyl acids and markers of kidney function among
children and adolescents living near a chemical plant. Environ Health Perspect 121: 625-
630. http://dx.doi.org/10.1289/ehp.1205838.

Weaver. YM: Ehresman. DT: Butenhoff. TL: Hagenbuch. B. (2010). Roles of rat renal organic anion
transporters in transporting perfluorinated carboxylates with different chain lengths.
Toxicol Sci 113: 305-314. http: / / dx. do i. o r g /10.10 9 3/toxsci/kfp275.

Weisskopf. MG: Seals. RM: Webster. TF. (2018). Bias amplification in epidemiologic analysis of

exposure to mixtures. Environ Health Perspect 126. http://dx.doi.org/10.1289/EHP2450.

Westwood. FR. (2008). The female rat reproductive cycle: A practical histological guide to staging
[Review], Toxicol Pathol 36: 375-384. http: / /dx.doi.org/10.1177/0192623308315665.

Whalan. IE. (2015). A toxicologist's guide to clinical pathology in animals: Hematology, clinical
chemistry, urinalysis. Switzerland: Springer International Publishing.
http: / /dx.doi.org/10.1007/978-3-319-15853-2.

Whitworth. KW: Haug. LS: Baird. DP: Becher. G: Hoppin. TA: Skiaerven. R: Thomsen. C: Eggesbo. M:
Travlos. G: Wilson. R: Longnecker. MP. (2012). Perfluorinated compounds and subfecundity
in pregnant women. Epidemiology 23: 257-263.
http: / /dx.doi.org/10.1097/EDE.0b013e31823h5031.

WHO (World Health Organization). (2010). WHO laboratory manual for the examination and

processing of human semen. In WHO laboratory manual for the examination and processing
of human semen (5 th ed.). Geneva, Switzerland.

Wielsae. M: Kern. P: Bonefeld-largensen. EC. (2017). Serum levels of environmental pollutants is a
risk factor for breast cancer in Inuit: a case control study. Environ Health 16:56.
http://dx.doi.org/10.1186/sl2940-017-0269-6.

Wikstrom. S: Hussein. G: Lingroth Karlsson. A: Lindh. CH: Bornehag. CG. (2021). Exposure to
perfluoroalkyl substances in early pregnancy and risk of sporadic first trimester
miscarriage. Sci Rep 11: 3568. http://dx.doi.org/10.1038/s41598-Q21-82748-6.

Wikstrom. S: Lin. PI: Lindh. CH: Shu. H: Bornehag. CG. (2020). Maternal serum levels of

perfluoroalkyl substances in early pregnancy and offspring birth weight Pediatr Res 87:
1093-1099. http://dx.doi.Org/10.1038/s41390-019-0720-l.

Wilson. MW: Lay. LT: Chow. CK: Tai. HH: Robertson. LW: Glauert. HP. (1995). Altered hepatic

eicosanoid concentrations in rats treated with the peroxisome proliferators ciprofibrate and
perfluorodecanoic acid. Arch Toxicol 69: 491-497.
http: / /dx.doi.org/10.1007/s002040050203.

Wilson. VS: Blvstone. CR: Hotchkiss. AK: Rider. CV: Gray. LE. Tr. (2008). Diverse mechanisms of anti-
androgen action: Impact on male rat reproductive tract development [Review], IntJ Androl
31: 178-187. http:/ /dx.doi.org/10.1111/i.l 365-2605.2007.00861 .x.

Wise. LA: Wesselink. AK: Schildroth. S: Calafat. AM: Bethea. TN: Geller. RT: Coleman. CM: Fruh. V:
Claus Henn. B: Botelho. TC: Harmon. OE: Thirkill. M: Wegienka. GR: Baird. DP. (2022).
Correlates of plasma concentrations of per- and poly-fluoroalkyl substances among
reproductive-aged Black women. Environ Res 203: 111860.
http://dx.doi.0rg/lO.lOl6/i.envres.202i.lll86O.

Wolf. CI: Schmid. IE: Lau. C: Abbott. BP. (2012). Activation of mouse and human peroxisome

proliferator-activated receptor-alpha (PPARa) by perfluoroalkyl acids (PFAAs): further

This document is a draft for review purposes only and does not constitute Agency policy.

R-32 PRAFT-PO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

investigation of C4-C12 compounds. Reprod Toxicol 33: 546-551.
http://dx.doi.Org/10.1016/i.reprotox.2011.09.009.

Wolf. CT: Takacs. ML: Schmid. TE: Lau. C: Abbott. BP. (2008). Activation of mouse and human
peroxisome proliferator-activated receptor alpha by perfluoroalkyl acids of different
functional groups and chain lengths. Toxicol Sci 106: 162-171.
http://dx.doi.org/10.1093/toxsci/kfnl66.

Wong. F: Macleod. M: Mueller. IF: Cousins. IT. (2014). Enhanced elimination of perfluorooctane
sulfonic acid by menstruating women: evidence from population-based pharmacokinetic
modeling. Environ Sci Technol 48: 8807-8814. http://dx.doi.org/10.1021/es500796y.

Woods. MM: Lanphear. BP: Braun. TM: McCandless. LC. (2017). Gestational exposure to endocrine
disrupting chemicals in relation to infant birth weight: A Bayesian analysis of the HOME
Study. Environ Health 16: 115. http://dx.doi.org/10.1186/sl2940-017-0332-3.

Workman. CE: Becker. AB: Azad. MB: Moraes. TT: Mandhane. PI: Turvev. SE: Subbarao. P: Brook. TR:
Sears. MR: Wong. CS. (2019). Associations between concentrations of perfluoroalkyl
substances in human plasma and maternal, infant, and home characteristics in Winnipeg,
Canada. Environ Pollut249: 758-766. http://dx.doi.Org/10.1016/i.envpol.2019.03.054.

Xiao. C: Grandiean. P: Valvi. D: Nielsen. F: Tensen. 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. http://dx.doi.org/10.1210/clinem/dgzl47.

Xie. X: Feng. LP: Weng. XO: Liu. S: Chen. TM: Guo. XR: Gao. XY: Fei. OY: Hao. G: ling. CX. (2021).

Perfluoroalkyl and polyfluoroalkyl substance exposure and association with sex hormone
concentrations: Results from the NHANES 2015-2016. Environ Sci Eur 33: 69.
http://dx.doi.org/10.n86/sl2302-021-00508-9.

Xu. C: Yin. S: Liu. Y: Chen. F: Zhong. Z: Li. F: Liu. K: Liu. W. (2019a). Prenatal exposure to chlorinated
polyfluoroalkyl ether sulfonic acids and perfluoroalkyl acids: Potential role of maternal
determinants and associations with birth outcomes. J Hazard Mater 380: 120867.
http://dx.doi.Org/10.1016/i.ihazmat.2019.120867.

Xu. M: Zhang. T: Lv. C: Niu. 0: Zong. W: Tang. 1: Liu. R. (2019b). Perfluorodecanoic acid-induced
oxidative stress and DNA damage investigated at the cellular and molecular levels.
Ecotoxicol Environ Saf 185: 109699. http://dx.doi.Org/10.1016/i.ecoenv.2019.109699.

Yamada. A: Bemrah. N: Vevrand. B: Pollono. C: Merlo. M: Desvignes. V: Sirot. V: Marchand. P:

Berrebi. A: Cariou. R: Antignac. TP: Le Bizec. B: Leblanc. TC. (2014). Dietary exposure to
perfluoroalkyl acids of specific French adult sub-populations: High seafood consumers, high
freshwater fish consumers and pregnant women. Sci Total Environ 491-492: 170-175.
http://dx.doi.Org/10.1016/i.scitotenv.2014.01.089.

Yamamoto. A: Kawashima. Y. (1997). Perfluorodecanoic acid enhances the formation of oleic acid in
rat liver. Biochem J 325 (Pt 2): 429-434. http: //dx.doi.org/10.1042 /bi3250429.

Yang. BY: Wu. 1: Niu. XL: He. CT: Bloom. MS: Abudoukade. M: Abulizi. M: Xu. AM: Li. BB: Li. L: Zhong.
XM: Wu. OZ: Chu. C: Luo. YN: Liu. XX: Zeng. XW: Yu. YT: Dong. GH: Zou. XG: Liu. T. (2022a).
Low-level environmental per- and polyfluoroalkyl substances and preterm birth: A nested
case-control study among a Uyghur population in northwestern China. Exposure and Health
14: 793-805. http: //dx.doi.org/10.1007/sl 2403-021-00454-0.

Yang. T: Wang. H: Du. H: Fang. H: Han. M: Wang. Y: Xu. L: Liu. S: Yi. T: Chen. Y: Tiang. 0: He. G. (2022b).
Exposure to perfluoroalkyl substances was associated with estrogen homeostasis in
pregnant women. Sci Total Environ 805: 150360.
http://dx.doi.Org/10.1016/i.scitotenv.2021.150360.

Yang. 1: Wang. H: Du. H: Fang. H: Han. M: Xu. L: Liu. S: Yi. 1: Chen. Y: Tiang. 0: He. G. (2020). Serum
perfluoroalkyl substances in relation to lipid metabolism in Chinese pregnant women.
Chemosphere 273: 128566. http://dx.doi.Org/10.1016/i.chemosphere.2020.128566.

This document is a draft for review purposes only and does not constitute Agency policy.

R-33 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

Yang. L: Li. T: Lai. T: Luan. H: Cai. Z: Wang. Y: Zhao. Y: Wu. Y. (2016a). Placental transfer of

perfluoroalkyl substances and associations with thyroid hormones: Beijing prenatal
exposure study. Sci Rep 6: 21699. http://dx.doi.org/10.1038/srep21699.

Yang. L: Wang. Z: Shi. Y: Li. 1: Wang. Y: Zhao. Y: Wu. Y: Cai. Z. (2016b). Human placental transfer of
perfluoroalkyl acid precursors: Levels and profiles in paired maternal and cord serum.
Chemosphere 144: 1631-1638. http://dx.doi.Org/10.1016/i.chemosphere.2015.10.063.

Yang. 0: 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: Across-sectional study. Sci Total Environ 621: 1542-1549.
http://dx.doi.Org/10.1016/i.scitotenv.2017.10.074.

Yao. 0: Gao. Y: Zhang. Y: Oin. 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.
http://dx.doi.Org/10.1016/i.chemosphere.2021.131521.

Yao. 0: 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.
http://dx.doi.Org/10.1016/i.envint.2019.03.047.

Yao. 0: Vinturache. A: Lei. X: Wang. Z: Pan. C: Shi. R: Yuan. T: Gao. Y: Tian. Y. (2022). Prenatal
exposure to per- and polyfluoroalkyl substances, fetal thyroid hormones, and infant
neurodevelopment Environ Res 206: 112561.
http://dx.doi.Org/10.1016/i.envres.2021.112561.

Yeung. LW: Guruge. KS: Tanivasu. S: Yamashita. N: Angus. PW: Herath. CB. (2013). Profiles of

perfluoroalkyl substances in the liver and serum of patients with liver cancer and cirrhosis
in Australia. Ecotoxicol Environ Saf 96: 139-146.
http://dx.doi.Org/10.1016/i.ecoenv.2013.06.006.

Ylinen. M: Auriola. S. (1990). Tissue distribution and elimination of perfluorodecanoic acid in the
rat after single intraperitoneal administration. Pharmacol Toxicol 66: 45-48.
http://dx.doi.Org/10.llll/i.1600-0773.1990.tb00700.x.

Yoo. H: Washington. TW: Tenkins. TM: Ellington. 11. (2011). Quantitative determination of

perfluorochemicals and fluorotelomer alcohols in plants from biosolid-amended fields using
LC/MS/MS and GC/MS. Environ Sci Technol 45: 7985-7990.
http://dx.doi.org/10.1021/eslQ2972m.

Yu. G: Tin. M: Huang. Y: Aimuzi. R: Zheng. T: Nian. M: Tian. Y: Wang. W: Luo. Z: Shen. L: Wang. X: Du.
0: Xu. W: Zhang. 1. (2021). Environmental exposure to perfluoroalkyl substances in early
pregnancy, maternal glucose homeostasis and the risk of gestational diabetes: A prospective
cohort study. Environ Int 156: 106621. http: //dx.doi.Org/10.1016/i.envint.2021.106621.

Zeng. X: Chen. 0: Zhang. X: Li. H: Liu. 0: Li. C: Ma. M: Zhang. 1: Zhang. W: Zhang. 1: Huang. L. (2019a).
Association between prenatal exposure to perfluoroalkyl substances and asthma-related
diseases in preschool children. Environ Sci Pollut Res Int 26: 29639-29648.
http://dx.doi.Org/10.1007/sll356-019-05864-x.

Zeng. XW: Bloom. MS: Dharmage. SC: Lodge. CI: Chen. D: Li. S: Guo. Y: Roponen. M: Talava. P:

Hirvonen. MR: Ma. H: Hao. YT: Chen. W: Yang. M: Chu. C: Li. 00: Hu. LW: Liu. KK: Yang. BY:
Liu. S: Fu. C: Dong. GH. (2019b). 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.
http://dx.doi.Org/10.1016/i.scitotenv.2019.01.325.

Zeng. XW: Lodge. CI: Dharmage. SC: Bloom. MS: Yu. Y: Yang. M: Chu. C: Li. 00: Hu. LW: Liu. KK: Yang.
BY: Dong. GH. (2019c). Isomers of per- and polyfluoroalkyl substances and uric acid in

This document is a draft for review purposes only and does not constitute Agency policy.

R-34 DRAFT-DO NOT CITE OR QUOTE

-------
1

Toxicological Review of Perfluorodecanoic Acid and Related Salts

adults: Isomers of C8 Health Project in China. Environ Int 133: 105160.
http://dx.doi.Org/10.1016/i.envint.2019.105160.

Zeng. XW: Oian. Z: Emo. B: Vaughn. M: Bao. 1: Oin. XD: Zhu. Y: Li. 1: Lee. YL: Dong. GH. (2015).

Association of polyfluoroalkyl chemical exposure with serum lipids in children. Sci Total
Environ 512-513: 364-370. http://dx.doi.Org/10.1016/i.scitotenv.2015.01.042.

Zhang. C: Sundaram. R: Maisog. 1: 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. http://dx.doi.Org/10.1016/i.fertnstert.2014.10.001.

Zhang. R: Ye. 1: Wei. 0: Li. M: Xu. K: Li. Z: Lin. W: Liu. P: Chen. R: Ma. A: Zhou. Z. (2018a). Plasma
concentration of 14 perfluoroalkyl acids (PFAAs) among children from seven cities in
Guangdong, China. Sci Total Environ 616-617: 1469-1476.
http://dx.doi.Org/10.1016/i.scitotenv.2017.10.167.

Zhang. S: Tan. R: Pan. R: Xiong. 1: Tian. Y: Wu. 1: Chen. L. (2018b). Association of perfluoroalkyl and
polyfluoroalkyl substances with premature ovarian insufficiency in Chinese women. J Clin
Endocrinol Metab 103: 2543-2551. http://dx.doi.org/10.1210/ic.2017-02783.

Zhang. T: Sun. H: Lin. Y: Oin. X: Zhang. Y: Geng. X: Kannan. K. f2013al. 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.
http://dx.doi.org/10.1021/es400937y.

Zhang. T: Zhang. B: Bai. X: Yao. Y: Wang. L: Shu. Y: Kannan. K: Huang. X: Sun. H. (2019). Health status
of elderly people living near e-waste recycling sites: association of e-waste dismantling
activities with legacy perfluoroalkyl substances (PFASs). Environ Sci Technol Lett 6: 133-
140. http://dx.doi.Org/10.1021/acs.estlett9b00085.

Zhang. Y: Beesoon. S: Zhu. L: Martin. TW. (2013b). Biomonitoring of perfluoroalkyl acids in human
urine and estimates of biological half-life. Environ Sci Technol 47: 10619-10627.
http://dx.doi.org/10.1021/es401905e.

Zhang. Y: Pan. C: Ren. Y: Wang. Z: Luo. 1: Ding. G: Vinturache. A: Wang. X: Shi. R: Ouvang. F: Zhang. 1:
Li. 1: 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. http://dx.doi.Org/10.1016/i.scitotenv.2021.151303.

Zhang. YM: Dong. XY: Fan. LI: Zhang. ZL: Wang. 0: Tiang. N: Yang. XS. (2017). Poly- and

perfluorinated compounds activate human pregnane X receptor. Toxicology 380: 23-29.
http://dx.doi.Org/10.1016/i.tox.2017.01.012.

Zhao. W: Cui. R: Wang. 1: Dai. 1. (2017). Inhibition effects of perfluoroalkyl acids on progesterone
production in mLTC-1. J Environ Sci 56: 272-280.
http://dx.doi.Org/10.1016/i.ies.2016.08.026.

Zhao. Y: Tin. H: Ou. I: Zhang. S: Hu. S: Xue. 1: Zhao. M. (2022). The influences of perfluoroalkyl
substances on the rheumatoid arthritis clinic. BMC Immunol 23: 10.
http://dx.doi.org/10.1186/sl 2865-022-00483-7.

Zhou. W: Zhang. L: Tong. C: Fang. F: Zhao. S: Tian. Y: Tao. Y: Zhang. I. (2017a). Plasma perfluoroalkyl
and polyfluoroalkyl substances concentration and menstrual cycle characteristics in
preconception women. Environ Health Perspect 125: 067012.
http://dx.doi.org/10.1289 /EHP1203.

Zhou. Y: Hu. LW: Oian. ZM: Chang. IT: 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.
http://dx.doi.Org/10.1016/i.envint.2016.05.018.

Zhou. Y: Hu. LW: Oian. ZM: Geiger. SD: Parrish. KL: Dharmage. SC: Campbell. B: Roponen. M: Talava.
P: Hirvonen. MR: Heinrich. 1: Zeng. XW: Yang. BY: Oin. XD: Lee. YL: Dong. GH. (2017b).

This document is a draft for review purposes only and does not constitute Agency policy.

R-35 DRAFT-DO NOT CITE OR QUOTE

-------
Toxicological Review of Perfluorodecanoic Acid and Related Salts

1 Interaction effects of polyfluoroalkyl substances and sex steroid hormones on asthma

2 among children. Sci Rep 7: 899. http: //dx.doi.org/10.1038/s41598-017-01140-5.

3 Zhu. Y: Oin. XD: Zeng. XW: Paul. G: Morawska. L: Su. MW: Tsai. CH: Wang. SO: Lee. YL: Dong. GH.

4 (2016). Associations of serum perfluoroalkyl acid levels with T-helper cell-specific

5 cytokines in children: By gender and asthma status. Sci Total Environ 559: 166-173.

6 http://dx.doi.Org/10.1016/i.scitotenv.2016.03.187.

This document is a draft for review purposes only and does not constitute Agency policy.

R-36 DRAFT-DO NOT CITE OR QUOTE

-------