Health-Based Drinking Water Value Recommendations for PFAS in Michigan Report

2019

Michigan Science Advisory Workgroup

DR. JAMIE DEWITT

MR. KEVIN COX

DR. DAVID SAVITZ

HEALTH-BASED

DRINKING WATER VALUE

RECOMMENDATIONS

FOR PFAS IN MICHIGAN

Executive Director’s Foreword

This report accomplishes a key milestone in Michigan’s effort to

identify and reduce exposures to per- and polyfluoroalkyl

substances (PFAS) contamination. With it, we are now one step

closer to developing state drinking water standards for PFAS.

Michigan is a national leader at addressing PFAS

contamination. Through our unique, multi-agency approach,

Michigan’s PFAS Action Response Team (MPART) is

systematically identifying sources of PFAS contamination and

getting a better understanding of their occurrence throughout

our environment.

By using analytical techniques capable of finding PFAS as low

as 2 parts per trillion, we have found the presence of PFAS in the drinking water from thousands

of private residential wells near contaminated sites. We have also found PFAS in public water

supplies across the state. We tested over 1,700 supplies covering all community water supplies

plus schools and larger day cares with their own wells. We found PFAS in ten percent of the

supplies. While most of the PFAS levels were very low, three percent of the supplies have

required follow-up actions, and a few have required an alternate water source.

Unfortunately, we do not have federal drinking water standards, despite knowing they are in our

drinking water and that some PFAS have been associated with adverse health effects.

Recognizing that the USEPA is still likely several years away from providing any leadership on

PFAS drinking water standards, Michigan, like other states, was left to develop our own.

With Governor Gretchen Whitmer’s leadership, MPART formed a Science Advisory Workgroup

to navigate the science and standards from across the country to advise Michigan on drinking

water health-based values for PFAS. These health-based values will be used to inform the next

step of the drinking water rule-making process, which includes stakeholder involvement where

other factors will be considered.

I could not be more impressed with the thoughtful deliberation of our workgroup and the tireless

technical support from our staff. As the information in this report is given to EGLE for consideration

during the development of drinking water standards, we all owe them our sincere appreciation for

giving us a firm foundation on which to move forward with protecting Michiganders from

unacceptable levels of PFAS in their drinking water.

Steve Sliver,

Executive Director,

Michigan PFAS Action Response Team

Michigan Science Advisory Workgroup

Dr. Jamie DeWitt

Mr. Kevin Cox

Dr. David Savitz

Agency Support Staff to the Panel

Mr. Steve Sliver, Michigan Department of Environment, Great Lakes, and Energy

Mr. Kory Groetsch, Michigan Department of Health and Human Services

Dr. Jennifer Gray, Michigan Department of Health and Human Services

Dr. Eric Wildfang, Michigan Department of Environment, Great Lakes, and Energy

Ms. Chelsea Dickerson, Michigan Department of Environment, Great Lakes, and Energy

Report developed for the Michigan PFAS Action Response Team,

Lansing, Michigan

June 27, 2019

iii

The Michigan Science Advisory Workgroup

Dr. David Savitz

Dr. David Savitz, who chairs the advisory Workgroup, is a professor of

epidemiology in the School of Public Health at Brown University. He also serves

as associate dean for research, and holds joint appointments in obstetrics and

gynecology, and pediatrics in the Alpert Medical School. His epidemiological

research has addressed a wide range of public health issues including

environmental hazards in the workplace and community, reproductive health

outcomes, and environmental influences on cancer. He has done extensive work

on health effects of nonionizing radiation, pesticides, drinking water treatment by-products, and

perfluorinated compounds. He is the author of nearly 350 papers in professional journals and editor or

author of three books. He was president of the Society for Epidemiologic Research and the Society for

Pediatric and Perinatal Epidemiologic Research, and North American regional councilor for the

International Epidemiological Association. Dr. Savitz is a member of the National Academy of Sciences

Institute of Medicine. From 2013-2017 he served as vice president for research at Brown University. He

was a member of the C8 Science Panel that conducted some of the first epidemiologic research on

PFAS in the mid-Ohio Valley and has published a number of reports related to potential health effects

of PFAS. He recently chaired the Science Panel to advise MPART on the current research related to

toxicology, epidemiology, exposure pathways, and remediation of PFAS.

Mr. Kevin Cox

Kevin Cox is a Managing Toxicologist at NSF International. Prior to his current

role, Mr. Cox was a Supervising Toxicologist supporting NSF’s drinking water

additives and dietary supplement certification programs. As an expert in human

health risk assessment, Mr. Cox has authored numerous chemical risk

assessments evaluating exposure from unregulated drinking water contaminants,

dietary supplement ingredients, toy product materials, and pool and spa treatment

chemicals. Specific to PFAS, Mr. Cox has conducted a state-of-the-science analysis of published PFAS

risk assessments in support of NSF International drinking water programs. This analysis was recently

presented to Michigan water management professionals. Mr. Cox received his B.S. in biochemistry and

history from the University of Michigan and his MPH in Environmental Health Sciences - Toxicology

from the University of Michigan School of Public Health. He is currently an Associate Member of the

Society of Toxicology. Mr. Cox also holds a J.D. from the University of Michigan Law School and is a

member of the Michigan Bar Association.

Dr. Jamie DeWitt

Dr. Jamie DeWitt is an associate professor in the Department of Pharmacology and

Toxicology of the Brody School of Medicine at East Carolina University. Her

laboratory’s research program explores relationships between biological organisms

and their responses after exposure to environmental contaminants, with a specific

focus on the immune system and its interactions with the nervous system during

development and adulthood. The research program particularly focuses on

emerging aquatic contaminants, especially PFAS. With respect to PFAS, DeWitt has published 13

primary research articles, six review articles, two book chapters, and edited a book on PFAS toxicity.

She has served as an external reviewer for the United States Environmental Protection Agency

(USEPA) health effects assessment of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate

(PFOS), the United States National Toxicology Program’s immune effects assessment of PFOA and

PFOS, the United States Agency for Toxic Substances and Disease Registry toxicological profile for

PFASs, and was a member of the International Agency for Research on Cancer working group for the

assessment of the carcinogenicity of PFOA. Her laboratory currently assesses the immunotoxicity of

emerging PFAS that have been designed to replace those that have been phased out of production and

that are of concern in North Carolina. She double-majored in environmental science and biology for her

bachelor’s degree from Michigan State University and has doctoral degrees in environmental science

and neural science from Indiana University-Bloomington. She completed postdoctoral training in

ecotoxicology at Indiana University-Bloomington and in immunotoxicology at the USEPA in partnership

with the University of North Carolina at Chapel Hill.

Table of Contents

Executive Director’s Foreword ..................................................................................................... i

The Michigan Science Advisory Workgroup ............................................................................... iii

Executive Summary ................................................................................................................... 2

Approach ................................................................................................................................... 5

Workgroup Interpretation of the Charge .................................................................................. 5

Challenges and Limitations ..................................................................................................... 6

Process ...................................................................................................................................... 7

Selection of Toxicity Values ................................................................................................... 7

Uncertainty Factors ................................................................................................................ 7

Relative Source Contribution .................................................................................................. 8

Drinking Water Health-Based Value Derivation ...................................................................... 8

Confidence Statement ............................................................................................................ 9

PFAS Chemical Summary Sheets.............................................................................................10

Chemical Summary for PFNA ............................................................................................10

Chemical Summary for PFOA ............................................................................................12

Chemical Summary for PFHxA ..........................................................................................14

Chemical Summary for PFOS ............................................................................................16

Chemical Summary for PFHxS ..........................................................................................18

Chemical Summary for PFBS ............................................................................................20

Chemical Summary for GenX .............................................................................................22

Rationale for Individual HBVs ....................................................................................................25

Summary of Conclusions ..........................................................................................................26

Summary Table of Drinking Water HBVs ...............................................................................26

Figure 1. ............................................................................................................................27

Concluding Remarks .................................................................................................................27

References ...............................................................................................................................29

Appendix A: Acronym List ......................................................................................................33

Appendix B: MPART Motion for Creation of Science Advisory Workgroup, April 4, 2019 ......34

Appendix C: USEPA Method 537.1 Analyte List ....................................................................35

Appendix D: Timeline for the Science Advisory Workgroup’s Development of Drinking Water

HBVs .....................................................................................................................................36

Appendix E: Timeline of the Maximum Contaminant Level Development Process .................37

Executive Summary

Background: The Michigan PFAS Action Response Team (MPART), is a unique, multi-agency

proactive approach for coordinating state resources to address per- and polyfluoroalkyl

substances (PFAS) contamination. Agencies responsible for environmental protection, public

health, natural resources, agriculture, military installations, commercial airports, and fire

departments work together to ensure the most efficient and effective response. The work done

by MPART on drinking water supports the development of standards now that we have key

information, including:

• PFAS have been discovered in drinking water during investigations of contaminated sites

and a survey of all of Michigan’s public water supplies. Public health responses, such as

the provision of alternate water (e.g., point of use filters) have been necessary for

thousands of Michiganders based on the strength of the source, location, and the

concentrations found.

• The MPART Science Advisory Panel report issued in December 2018 indicated that

observational epidemiology literature supports the need for drinking water values below

the United States Environmental Protection Agency (USEPA) Lifetime Health Advisory

(LHA) level of 70 ppt PFOS and PFOA, individually or in combination, and included a

recommendation for establishing state drinking water standards for PFAS.

• The Michigan Department of Health and Human Services (MDHHS)-led MPART Human

Health Workgroup developed public health drinking water screening levels for five

individual PFAS in February 2019. Those screening levels will prompt further evaluation

and public health consultations at numerous public water supplies and residences across

the state including where detectable levels of PFOS and/or PFOA are below the USEPA

LHA.

On March 26, 2019, Governor Gretchen Whitmer announced that Michigan was establishing

enforceable state drinking water standards for PFAS. These standards, otherwise known as

Maximum Contaminant Levels (MCLs), under the federal Safe Drinking Water Act have

traditionally been established first by the USEPA and then adopted by the states. At this time,

however, the USEPA has not initiated its process for establishing PFAS MCLs, and its process

could take five or more years to complete. Michigan chose not to wait any longer for federal action.

Governor Whitmer called on MPART to form a Science Advisory Workgroup (Workgroup) to

review the existing and proposed PFAS standards from across the country and develop health-

based values (HBVs) to inform the initial phase of the rulemaking process for establishing state

drinking water standards. The workgroup was given until July 1, 2019 to develop the HBVs. On

April 4, 2019, MPART approved a motion to create the Workgroup. The Charge from MPART to

the Workgroup is included in Appendix B. The members of the Workgroup were announced on

April 11, 2019. The Workgroup was supported by MPART staff.

The Workgroup members are experts in the fields of epidemiology, toxicology, and risk

assessment. The composition of the Workgroup matches the typical fields of evaluation for HBV

developments. Dr. Jamie DeWitt provided the strong toxicological expertise and up-to-date

knowledge on PFAS toxicology as HBVs typically use laboratory animal toxicity studies.

Epidemiological information supports the laboratory animal data, and Dr. David Savitz provided

his epidemiological expertise in selection of health endpoints and relevance to humans. Tying

both toxicology and epidemiology together are risk assessment practices, and Mr. Kevin Cox

provided the expertise in that field. Taken together, this Workgroup was able to knowledgably

speak on the current state of PFAS health research and provide the scientific expertise needed

to efficiently develop HBVs on the requested timeline.

The evaluation and deliberations of the Workgroup occurred over a very limited timeframe

(Appendix D), which required frequent interaction. Much of that interaction occurred during 7 web

conferences between April 19 and May 29, 2019, culminating in an in-person meeting the weekend

of June 1-2, 2019. The Workgroup’s final conclusions were presented to MPART on June 27, 2019.

Conclusions: The Workgroup undertook a methodical approach to evaluate existing and

proposed standards from across the country for the 18 PFAS analytes considered under USEPA

Method 537.1 (Appendix C). They focused on those PFAS that they determined had enough peer

reviewed studies on which to base their conclusions. What they considered, and the logic behind

their approach, has been carefully documented in individual chemical summaries for each

compound that has a derived HBV in the following table:

Summary Table of Drinking Water Health-Based Values

Specific

PFAS

Drinking Water Health-

based Value

Chemical Abstract

Services Registry

Number (CASRN)

PFNA

6 ng/L (ppt)

375-95-1

PFOA

8 ng/L (ppt)

335-67-1

PFHxA

400,000 ng/L (ppt)

307-24-4

PFOS

16 ng/L (ppt)

1763-23-1

PFHxS

51 ng/L (ppt)

355-46-4

PFBS

420 ng/L (ppt)

375-73-5

GenX

370 ng/L (ppt)

13252-13-6

The Workgroup also recommended MPART and water supply operators screen analytical results

for other long-chain PFAS (eight carbons and above for carboxylates and six carbons and above

for sulfonates) included in USEPA Method 537.1 at the lowest concentration proposed for any of

the compounds, which is 6 ppt. Based on the similarity in toxicity for the long-chain PFAS, the

Workgroup recommends use of the HBV for PFNA (6 ng/L [ppt]) as a screening level for all other

long-chain PFAS included on the USEPA Method 537.1 analyte list for which the Workgroup did

not develop an individual HBV. Those other long-chain PFAS included in USEPA Method 537.1

are: NEtFOSAA (CASRN: 2991-50-6); NMeFOSAA (CASRN: 2355-31-9); PFDA (CASRN: 335-

76-2); PFDoA (CASRN: 307-55-1); PFTA (CASRN: 376-06-7); PFTrDA (CASRN: 72629-94-8);

and PFUnA (CASRN: 2058-94-8). While there is not enough information available at this time to

support HBVs and drinking water standards for them, these compounds are expected to produce

similar health effects. Additional monitoring, research for potential sources, notification of the

public, and efforts to reduce exposure are warranted.

The Workgroup recognizes that their conclusions in some cases deviate modestly from those of

other organizations. Evolving science and professional judgement can account for the variation.

The variation is not substantial, however, and the values are trending lower nationally over time.

Approach

Workgroup Interpretation of the Charge

The Workgroup was conscience of the importance and responsibility placed upon its efforts to

identify public health toxicity values for certain PFAS as described within the Charge. Prior to

initiating its efforts, the Workgroup sought and received clarification on the scope of the Charge.

Given the relatively short timeframe for which to accomplish the tasks set forth within Charge, the

Workgroup confirmed that the focus of the effort was to utilize the existing and proposed national-

and state-derived PFAS assessments to inform its decision-making process as opposed to

conducting a full systematic review of the available scientific literature on PFAS.

Additionally, as one of the outputs of the Charge is to inform State of Michigan on drinking water

health-based values for PFAS, it was important to understand if the State of Michigan had any

paradigms in place that the Workgroup must follow when deriving drinking water health-based

values. The response received from the State of Michigan indicated that the Workgroup was only

limited to applying a scientifically defensible approach as described within the Charge. With these

issues clarified, the Workgroup approached the tasks set forth in the charge in the following

manner:

1) Initially, PFAS analytes were identified within USEPA Method 537.1 for which published

or externally peer reviewed PFAS drinking water criteria or reference doses (RfDs) existed

and the derivation of such values was done in a scientifically defensible manner. This

approach resulted in the selection of PFOA, PFOS, PFHxS, PFHxA, PFBS, PFNA and

GenX as PFAS analytes for which the Workgroup would then develop individual public

health toxicity values. The remaining PFAS values within USEPA Method 537.1 were later

considered as to whether a class-based or group-based public health toxicity value could

be applied.

2) For each of the selected PFAS analytes, the Workgroup evaluated the identified points of

departure (defined as the point on a toxicological dose-response curve corresponding to

an estimated low effect level or no effect level) and rationale from published risk

assessments and assessed the underlying key studies that served as the basis for the

published values. From this review, the merits of each available point of departure was

discussed among the Workgroup and critical studies and points of departures for each of

the seven identified PFAS analytes were identified to form the basis of public health toxicity

values described further herein.

3) With critical studies and points of departure identified for each individual PFAS, the

Workgroup then identified appropriate uncertainty factors to derive public health toxicity

values. From these public health toxicity values, the Workgroup recommended specific

drinking water exposure paradigms, accounting for sensitive sub-populations, and applied

selected relative source contribution factors to derive the drinking water health-based

values described further herein.

4) Lastly, consideration was given to the remaining PFAS analytes from USEPA Method

537.1 that were not selected for the development of individual criteria as to whether a

class-based or grouping-based evaluation approach would be appropriate. As described

below, the Workgroup concluded that a screening level approach was valid to assess

longer-chain PFAS based on the lowest derived drinking water health-based values.

Based on guidance from the Director of EGLE’s Drinking Water and Environmental Health

Division, PFAS chemical summary sheets were used to capture the necessary information for the

MCL rulemaking process. The Workgroup and MPART staff used this format to provide maximum

transparency on the decisions and rationale for drinking water health-based value development

for each PFAS.

The chemical summary sheets describe:

• The critical study or studies, point of departure from each study, and conversion to a

human equivalent dose;

• Uncertainty factors and a calculated toxicity value;

• Exposure parameters, and methodology for calculation of a drinking water health-based

value.

Challenges and Limitations

The premises for the Workgroup’s efforts to provide evidence-based conclusions for informing the

regulation of PFAS in drinking water are compelling. Policy needs to provide clarity on what levels

of specific chemicals are believed to be protective of public health and develop a mechanism to

monitor and mitigate pollutants such as PFAS where needed. The Workgroup identified and

made optimal use of the scientific evidence that is available to provide guidance, drawing on its

knowledge of research methods and quantitative risk assessment. Furthermore, the Workgroup

approached the issue free of bias, and as a panel, has a wide range of expertise and familiarity

with the research on PFAS. However, the nature of this process is inherently subject to

uncertainty and other equally qualified experts presented with the same scientific data the

Workgroup drew upon might well make somewhat different conclusions. A number of other

organizations have been through a similar exercise in providing guidance on acceptable drinking

water contaminant levels, and while there are not extreme differences, there is not complete

convergence either. As described in some detail below, a series of inputs were needed to derive

the Workgroup’s estimates and make that sequence of decisions as transparent as possible for

those who wish to compare these conclusions to those made by other agencies. Like all the

others, they are based exclusively on toxicology studies given the ability to quantify exposure-

response relationships with great precision, but there is a loss of certainty in applying these

estimates to free-living human populations. In most cases, there is epidemiologic evidence

pertaining to the same health endpoints used in toxicology, and where there is such convergent

evidence (e.g., immune function, development), confidence in the applicability of the experimental

studies to human populations is enhanced. Finally, it should be noted that the scientific evidence

on PFAS is expanding rapidly and that with new studies, the guidelines may well need to be

revised. While it would be inefficient to do so frequently, on some periodic basis of several years,

it would be useful to repeat the process that generated this report to determine where changes

may be needed.

Process

Selection of Toxicity Values

Adverse health effects reported following exposure to PFAS in laboratory animal models and

epidemiological studies have been summarized in myriad peer-reviewed and publicly available

documents, including those generated by other state agencies. Most recently, the Agency for

Toxic Substances and Disease Registry (ATSDR), compiled a toxicological profile for 14 PFAS

that comprehensively summarizes evidence from publicly available published studies (ATSDR,

2018). This, and other summary documents, as well as the published studies themselves, were

relied on to determine points of departure, as well as the toxicity values that protect the most

sensitive populations and reflect a level that is unlikely to lead to adverse health effects if those

sensitive populations are exposed over a lifetime or during a sensitive period (i.e., during

development). The toxicity values are therefore designed to be protective of all exposed

populations. For all of the PFAS examined, points of departure were selected from studies with

laboratory animal models. This approach does not negate findings associated with

epidemiological studies, but reflects that humans experience uncontrolled and imperfectly

documented rather than controlled, precisely measured exposures. Additionally, these points of

departure reflect adverse health effects that occur at low doses and that are supported by the

weight-of-evidence across endpoints and between findings in humans and laboratory animal

models. Therefore, the process to select points of departure used the available scientific evidence

to identify an adverse health effect that occurred at a low dose, was supported by findings in other

studies, was relevant to humans, and would be protective of sensitive populations.

Uncertainty Factors

In deriving the toxicity values for PFAS, the selected points of departure are divided by uncertainty

factors. Uncertainty factors are applied in order to account for:

1. Variation in susceptibility among the human population (intraspecies uncertainty);

2. Uncertainty in extrapolating animal data to humans (interspecies uncertainty);

3. Uncertainty in extrapolating from data obtained from a study with a less-than-lifetime

exposure (subchronic to chronic uncertainty);

4. Uncertainty in extrapolating from a lowest observed adverse effect level (LOAEL) as

opposed to a no observed adverse effect level (NOAEL); and

5. Uncertainty associated with an incomplete toxicity database. Uncertainty factors assigned

for each of these five categories are typically 1x, 3x (10

0.5

x), or 10x with the default value

being 10x, which represents greater uncertainty.

For both interspecies and intraspecies uncertainty factors, the variability in response to a toxicant

may result from differences in toxicokinetics and/or toxicodynamics. Toxicokinetics refers to the

absorption, distribution, biotransformation and excretion of the toxicant following exposure.

Toxicodynamics refers to the molecular, biochemical and physiological effects of the toxicant or

its metabolites leading to the toxic response. Therefore, the interspecies and intraspecies

uncertainty factors are divided into subparts representing the toxicokinetic factor and the

toxicodynamic factor. In evaluating the interspecies uncertainty for the selected PFAS, in each

case the toxicokinetic subfactor was able to be reduced to 1x on account of adjustments based

on serum half-lives or allometric scaling. Due to lack of data to depart from the default the

toxicodynamic subfactor 3x (10

0.5

x), the resulting interspecies uncertainty factor is 3x (10

0.5

x).

When considering the subchronic to chronic uncertainty, the relevant consideration is whether the

selected point of departure may differ if the duration of exposure were to be increased. For PFAS,

a weight of evidence approach was used to assess the subchronic to chronic uncertainty factor,

including, but not limited to, duration of the key study, potential impact of duration on the selected

point of departure, as well as availability of chronic repeat-dose toxicity data.

For the NOAEL to LOAEL uncertainty factor, use of a NOAEL (or lower confidence limit on the

benchmark dose [BMDL]) allows for an uncertainty factor of 1x. If the point of departure is based

on a LOAEL, the uncertainty factor is either 3x (10

0.5

x) or 10x depending on the severity and/or

reversibility of the critical effect.

The database uncertainty factor is based on the ability of the existing data to support a scientific

judgment of the likely critical effect from exposure to the compound. In assessing the database

completeness, the types of toxicity data (e.g., human, animal, mode of action) as well as data

gaps that may have improved the derived risk values should be emphasized. This approach

should take into consideration issues such as the types of endpoints evaluated, life-stages

evaluated, duration, timing, route of exposure, and the potential for latent effects and/or

reversibility of effects (USEPA, 2002). For the selected PFAS, each database was unique;

however, common concerns were lack of appropriate characterization of immune, endocrine or

neurodevelopmental effects.

Relative Source Contribution

Relative source contribution (RSC) is the percentage of a person’s exposure to a chemical that

comes from drinking water. For example, an RSC of 20 percent assumes that the other 80

percent of a person’s exposure to a chemical comes from non-drinking water sources. The

USEPA (2000) provides guidance on the selection of an RSC value using an exposure decision

tree that takes into account specific populations of concern, whether these populations are

experiencing exposure from multiple sources, and whether levels of exposure or other

circumstances make apportionment of the toxicity value or POD/UF desirable. The most

conservative RSC is established at 20 percent, and the RSC can reach a ceiling of 80 percent

as more information is available about exposure pathways and the source of exposure.

Drinking Water Health-Based Value Derivation

The traditional risk assessment approach using simple equations based on body weight, water

intake rate and RSC to calculate drinking water HBVs is not adequate to address the

bioaccumulative nature and known or presumed developmental toxicity of PFAS. These

traditional equations do not consider the PFAS body-burden at birth or any transfer of maternal

PFAS through breastmilk. To better address these concerns, and to also account for higher early-

life intake rates, the Goeden et al. (2019) simple one-compartment toxicokinetic model was used

where the data were available for the individual PFAS. The resulting drinking water HBVs are

considered protective for an infant exclusively breast-fed for 12 months, followed by drinking

contaminated water through life. Additionally, these drinking water HBVs also protective for

formula-fed infants. Where data were not available to derive drinking water HBVs using the model,

traditional equations were used.

Confidence Statement

Following USEPA guidance (2002), risk assessments may contain a narrative description of the

overall confidence in the derived health-effects based values. Confidence in the risk assessment

would be low if there is a high degree of scientific uncertainty and would be high if there is a low

degree of scientific uncertainty. Major elements of scientific uncertainty may be considered to

include, but not limited to, the following; database completeness, quality of key study(ies), severity

and relevance of the critical effect, quality of the dose-response analysis and consideration of

sensitive subpopulations. (NRC, 2009; Beck et al., 2016).

For the selected PFAS for which quantitative values were derived there remains significant

scientific uncertainty. Health outcomes due to PFAS exposure that warrant additional study

include, but are not limited to, endocrine disruption, immunological and neurodevelopmental

effects as well as cancer. Further information is needed on the mode of action as well as the

cumulative risk of exposure to multiple PFAS. Overall, the present evaluation of the selected

PFAS is based on sound science and current practices in risk assessment; however, the

Workgroup recognizes that the science of PFAS is constantly evolving and new information may

come to light that requires a re-evaluation of the drinking water HBVs established herein.

PFAS Chemical Summary Sheets

Chemical Summary for PFNA

Decision Point

Rationale/justification

Critical

study

Das KP, Grey BE, Rosen MB, et al. 2015. Developmental toxicity

of perfluorononanoic acid in mice. Reproductive Toxicology 51:133-

144.

The Workgroup reviewed the available evaluations and

focused on the assessments by ATSDR and New Jersey.

Das et al. (2015) was selected by both ATSDR (2018)

and NJDEP (2015).

Description

of the critical

study

Timed-pregnant CD-1 mice were administered 0, 1, 3, 5 or 10 mg/kg

PFNA by daily oral gavage from gestational day (GD) 1 to 17. Maternal

toxicity and reproductive outcomes were investigated. Postnatal

toxicity, liver gene expression and developmental effects were

evaluated in mouse offspring.

Body weight endpoints – Decreased body weight gain in mouse pups

Developmental endpoints – Delayed eye opening, preputial separation,

and vaginal opening in mouse pups

The Workgroup reviewed the health endpoints

investigated in Das et al. (2015) and identified the

developmental endpoints as more relevant than liver

endpoints.

Point of

Departure

(POD)

A NOAEL of 1 mg/kg/day was identified for developmental effects. The

average serum concentration for NOAEL (1 mg/kg/day) was estimated

(6.8 mg/L) in dams using an empirical clearance model (Wambaugh et

al., 2013). The estimated time-weighted average serum concentration

corresponding to the NOAEL was 6.8 mg/L.

The Workgroup decided that serum-based points of

departure were appropriate for PFAS.

Human

equivalent

dose (HED)

The time-weighted average serum concentration of 6.8 mg/L was

converted to the HED using the below equation.

NOAEL

HED

= (TWA serum x k

x V

) = 0.000665 mg/kg/day

Ke = 0.000489165 (4.8 x 10

-4

) based on a human serum half-life of

1417 days (calculated from Zhang et al. [2013] as described above)

Vd = 0.2 L/kg (ATSDR [2018]; Ohmori et al. [2003])

The Workgroup discussed the human serum half-lives

available from Zhang et al. (2013), which were

an arithmetic mean of 2.5 years (913 days) for 50 year old

or younger females and 4.3 years (1570 days)

for females older than 50 years old and all males. An

average of 3.9 years (1417 days) was calculated based on

those averages. The Workgroup selected the calculated

average as it would better represent the entire

population.

Uncertainty

factors

A total uncertainty factor of 300:

• 1 for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability

• 1 for subchronic to chronic

• 10 for database deficiencies was used.

The Workgroup discussed the uncertainty factors selected

by ATSDR (2018) and agreed that those selected were

appropriate.

Toxicity

value

2.2 ng/kg/day (2.2 x 10

-6

mg/kg/day) which corresponds to a serum

concentration of 0.023 mg/L

Serum levels used in development of these toxicity levels are not

meant to indicate a level where health effects are likely. These serum

levels are calculated to be at a point where no or minimal risk exists for

people drinking water with a certain PFAS.

Human equivalent dose or serum level divided by the total

uncertainty factors = toxicity value

Exposure

parameters

for drinking

water

screening

HBVs

Breast-fed infant, which is also protective of a formula-fed infant

Placental transfer of 69% (MDHHS 2019)

Breastmilk transfer of 3.2% (MDHHS 2019)

Half-life = 1417 days (3.9 years) (calculated from Zhang et al. [2013] as

described above)

Volume of distribution = 0.2 L/kg (ATSDR [2018]; Ohmori et al. [2003])

percentile drinking water intake, consumers only, from birth to more

than 21 years old (Goeden et al. [2019])

Upper percentile (mean plus two standard deviations) breast milk

intake rate (Goeden et al. [2019])

Time-weighted average water ingestion rate from birth to 30-35 years

of age (to calculate maternal serum concentration at delivery)

(Goeden et al. [2019])

Relative Source Contribution of 50% (0.5)

Based on NHANES 95

percentiles for 3-11 (2013-2014) and over 12

years old (2015-2016) participants (CDC 2019)

The Workgroup discussed the Goeden et al. (2019) model

which considered full life stage exposure, from fetal

exposure, to infant exposure through breastfeeding, and

into adulthood. While the model was also developed for a

formula-fed infant, the breastfed infant scenario is

protective of a formula-fed infant. The Workgroup selected

this model for developing drinking water HBVs when the

needed inputs were available.

Drinking

water HBV

6 ng/L (ppt)

Numeric HBV derived and justified using the above

information

Chemical Summary for PFOA

Decision point

Rationale/justification

Critical

study

Onishchenko N, Fischer C, Wan Ibrahim WN, Negri S, Spulber S,

Cottica D, Ceccatelli S. 2011. Prenatal exposure to PFOS or PFOA

alters motor function in mice in a sex-related manner. Neurotox. Res.

19(3):452-61.

Koskela A, Finnilä MA, Korkalainen M, Spulber S, Koponen J, Håkanss

on H, Tuukkanen J, Viluksela M. 2016. Effects of developmental

exposure to perfluorooctanoic acid (PFOA) on long bone morphology

and bone cell differentiation. Toxicol. Appl. Pharmacol. 301:14-21.

The Workgroup reviewed the available evaluation and

selected the ATSDR (2018) critical studies. The

Workgroup concluded that the ATSDR

position was defensible with respect to range and

sensitivity of health endpoints identified and considered in

ATSDR (2018).

Description

of the critical

study

Onishchenko et al.: Pregnant C57BL/6 mice were exposed to 0 or 0.3

mg PFOA/kg/day throughout pregnancy. The critical effects considered

were Neurobehavioral effects (decreased number of inactive periods,

altered novelty induced activity) at 5-8 weeks of age.

Koskela et al.: Pregnant C57BL/6 mice were exposed to PFOA mixed

with food at the dose of 0 or 0.3 mg PFOA/kg/day throughout

pregnancy. Group of five offspring (female) were sacrificed at either 13

or 17 months of age. The critical effects considered were skeletal

alteration such as bone morphology and bone cell differentiation in the

femurs and tibias.

The Workgroup selected these

developmental delays as most appropriate health

endpoint as the mammary gland effects may represent a

delay that may not be considered adverse. However, the

mammary gland effects may be representative of

endocrine effects at doses below the selected POD.

Point of

Departure

The average serum concentration was estimated in the mice (8.29

mg/L) using a three-compartment pharmacokinetic model (Wambaugh

et al. 2013) using animal species-, strain-, sex-specific parameters.

The Workgroup decided that serum-based points of

departure were appropriate for PFAS.

Human

equivalent

dose

The time-weighted average serum concentration of 8.29 mg/L was

converted to the HED using the below equation.

LOAEL

HED

= (TWA serum x k

x V

) = 0.001163 mg/kg/day

Ke = 0.000825175 (8.2 x 10

-4

) based on a human serum half-life of 840

days (Bartell et al. 2010)

Vd = 0.17 L/kg (Thompson et al. 2010)

The Workgroup selected the PFOA serum half-life of 840

days (2.3 years) as more relevant for exposure to the

general population as this half-life corresponds to data

from Bartell et al. (2010) in which 200 individuals (100

men, 100 women) were exposed by drinking PFOA-

contaminated water.

The Workgroup selected the volume of distribution based

on human data, when available.

Uncertainty

factors

A total uncertainty factor of 300:

• 3 (10

0.5

) for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability

• 1 for subchronic to chronic

• 3 (10

0.5

) for database deficiencies (endocrine effects)

The Workgroup discussed the use of an uncertainty factor

of 3 for use of a LOAEL. They noted that a NOAEL for

immune effects was similar to the LOAEL selected and

that the selected LOAEL represented less severe effects.

The Workgroup concluded that use of the 3 (10

0.5

) would

be sufficiently protective.

The Workgroup added a database uncertainty factor of

3 (10

0.5

) for deficiencies the database regarding endocrine

effects. The Workgroup noted that the mammary gland

effects may signal a concern for other low dose endocrine

effects.

Toxicity

value

3.9 ng/kg/day (3.9 x 10

-6

mg/kg/day) which corresponds to a serum

concentration of 0.028 mg/L

Serum levels used in development of these toxicity levels are not

meant to indicate a level where health effects are likely. These serum

levels are calculated to be at a point where no or minimal risk exists for

people drinking water with a certain PFAS.

Human equivalent dose or serum level divided by the total

uncertainty factors = toxicity value

Exposure

parameters

for drinking

water HBVs

Breast-fed infant, which is also protective of a formula-fed infant

Placental transfer of 87% (MDH 2017)

Breastmilk transfer of 5.2% (MDH 2017)

Human Serum half-life of 840 days (Bartell et al. 2010)

Volume of distribution of 0.17 L/kg (Thompson et al. [2010])

percentile drinking water intake, consumers only, from birth to

more than 21 years old (Goeden et al. [2019])

Upper percentile (mean plus two standard deviations) breast milk

intake rate (Goeden et al. [2019])

Time-weighted average water ingestion rate from birth to 30-35 years

of age (to calculate maternal serum concentration at delivery) (Goeden

et al. [2019])

Relative Source Contribution of 50% (0.5)

Based on NHANES 95

percentiles for 3-11 (2013-2014) and over 12

years old (2015-2016) participants (CDC 2019)

The Workgroup discussed the Goeden et al. (2019) model

which considered full life stage exposure, from fetal

exposure, to infant exposure through breastfeeding, and

into adulthood. While the model was also developed for a

formula-fed infant, the breastfed infant scenario is

protective of a formula-fed infant. The Workgroup selected

this model for developing drinking water HBVs when the

needed inputs were available.

Drinking

water HBV

8 ng/L (ppt)

Numeric HBV derived and justified using the above

information

Chemical Summary for PFHxA

Decision point 

Rationale/justification

Critical

study

Klaunig, J.E., Shinohara, M., Iwai, H., Chengelis, C.P., Kirkpatrick, J.B.,

Wang, Z., Bruner, R.H., 2015. Evaluation of the chronic toxicity and

carcinogenicity of perfluorohexanoic acid (PFHxA) in Sprague-Dawley

rats. Toxicol. Pathol. 43 (2), 209–220.

The Workgroup reviewed the Luz et al. (2019) compiled

information and development of a toxicity value. The

Workgroup was in agreement with Luz et al. (2019) on

selection of the chronic study (Klaunig et al. 2015) for

toxicity value development.

Description

of the critical

study

PFHxA was administered to male and female Crl:CD rats (n=60-

70/sex/dose) via daily oral gavage for up to 104 weeks. Males: 0, 2.5,

15, and 100 mg/kg/day. Females: 0, 5, 30, and 200 mg/kg/day.

Functional observational battery, locomotor activity, ophthalmic,

hematology, serum chemistry, and tissue and organ histopathology

endpoints were evaluated.

The Workgroup also considered the developmental effects

observed in Loveless et al. (2009) one generation

reproductive assay. Pup body weight was significantly

reduced in the 500 mg/kg/day, resulting in NOAEL of 100

mg/kg/day. Data were not available for Benchmark Dose

Modeling for further evaluation.

Point of

Departure

Critical effect renal tubular degeneration and renal papillary necrosis in

female rats – BMDL

90.4 mg/kg/day (Luz et al., 2019).

The Workgroup noted that the Benchmark Dose approach

is preferred over the use of a NOAEL/LOAEL. 

Human

equivalent

dose

Therefore, the BMD was adjusted by (80kg/0.45 kg)

= 3.65. The

resulting POD

HED

(90.4 mg/kg/day divided by 3.65) = 24.8 mg/kg/day.

(Luz et al., 2019).

The Workgroup discussed the description of the

Benchmark Dose modeling conducted by Luz et al. (2019)

and concluded the modeling was adequate for use. The

Workgroup did not conduct their own Benchmark Dose

modeling.

The Workgroup took into consideration the available

serum half-life data presented in Russell et al. (2013) and

concluded that, unlike most PFAS, allometric scaling could

be supported.

Uncertainty

factors

Total uncertainty factor of 300:

• 1 for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability

• 1 for subchronic to chronic

• 10 for database deficiencies – lack of additional chronic toxicity

studies and no additional developmental data in a second species,

and immune and thyroid endpoints

The Workgroup discussed the uncertainty factors and

selected an uncertainty factor of 10 for database

deficiencies. Several items noted were that the available

studies were largely in one species, with no mouse or

non-human primate data, and that there was insufficient

information addressing immune or thyroid endpoints.

Toxicity

value

83,000 ng/kg/day (8.3 mg/kg/day)

Human equivalent dose divided by the total uncertainty

factor = toxicity value

Exposure

parameters

for drinking

water HBVs

95th percentile of water intake for consumers only (direct and indirect

consumption) for adults (>21 years old) of 3.353 L/day, per Table 3-1,

USEPA Exposure Factors Handbook, 2019.

An adult body weight of 80 kilograms was used (Table 8-1, USEPA

2011b).

A default Relative Source Contribution of 20% was included.

The Workgroup discussed the use of an upper percentile

water intake. The 95

percentile for consumers only was

selected as it would protect those drinking larger amounts

of water.

As no human serum data were available to assess the

population’s exposure to PFHxA from sources other than

drinking water, a default Relative Source Contribution of

20% was selected consistent with USEPA (2000)

guidance.

The Workgroup evaluated the protectiveness of the renal

tubular degeneration and renal papillary necrosis in

relation to the reduced pup weights observed in Loveless

et al. (2009).

Available data did not support Benchmark Dose Modeling

for further evaluation of Loveless et al. (2009) data.

Drinking

water HBV

400,000 ng/L (ppt) (400 micrograms per Liter or parts per billion)

Numeric HBV derived and justified using the above

information in the following equation:

 =

 ×   ×  ℎ

 

Chemical Summary for PFOS

Decision point 

Rationale/justification

Critical

study

Dong GH, Zhang YH, Zheng L, Liu W, Jin YH, He QC. (2009). Chronic

effects of perfluorooctanesulfonate exposure on immunotoxicity in adult

male C57BL/6 mice. Arch Toxicol. 83(9):805-815.

The Workgroup discussed the available evaluations,

particularly MDH (2019) and New Jersey Department of

Environmental Protection (NJDEP) (2018), and selected a

critical study with an immune system functional assay

rather than observational data.

Description

of the critical

study

Adult male C57BL/6 mice were exposed to PFOS daily via oral gavage

for 60 days with 0, 0.5, 5, 25, 50 or 125 mg/kg total administered dose,

equivalent to 0 or approximately 0.008, 0.08, 0.4, 0.8 or 2.1 mg/kg/day.

The NOAEL for suppression of plaque forming cell response and

increase in liver mass was 0.5 mg/kg total administered dose which

corresponded to a serum concentration of 0.674 mg/L.

The Workgroup acknowledged that immune effects in

mice were seen at lower doses in Peden-Adams et al.

(2008). Serum concentrations from Peden-Adams et al.

(2008) were well below both the NOAEL and LOAEL

serum concentrations measured from several other

studies as described by Pachkowski et al. (2019) and may

be an outlier in the database.

Point of

Departure

The NOAEL for suppression of plaque forming cell response and

increase in liver mass was 0.5 mg/kg total administered dose which

corresponded to a serum concentration of 0.674 mg/L.

The Workgroup decided that serum-based points of

departure were appropriate for PFAS.

Human

equivalent

dose

The serum concentration of 0.674 mg/L was converted to the HED

using the below equation (based on ATSDR 2018).

NOAEL

HED

= (TWA serum x k

x V

) = 0.0000866 mg/kg/day

Ke = 0.000558539 (5.5 x 10

-4

) based on a human serum half-life of

1241 days (Li et al. 2018)

Vd = 0.23 L/kg (Thompson et al. 2010)

The Workgroup selected the serum half-life from a non-

occupationally exposed population as it is closer to the

general population’s exposure. The Workgroup selected

volume of distributions based on human data,

when available.

Uncertainty

factors

A total uncertainty factor of 30:

• 1 for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human difference (toxicodynamics)

• 1 for subchronic to chronic

• 1 for database deficiencies

The Workgroup reviewed the uncertainty factors selected

by MDH (2019) and adjusted the database uncertainty

factor to 1 based on the critical study selection. With

consideration of the selected immunotoxicity endpoint, the

database uncertainty factor of 1 was supported by the

assessments by USEPA (2016), NJDEP (2018), ATSDR

(2018) and New Hampshire (2019).

Toxicity

value

2.89 ng/kg/day (2.89 x 10

-6

mg/kg/day) which corresponds to a serum

concentration of 0.022 µg/ml

Serum levels used in development of these toxicity levels are not

meant to indicate a level where health effects are likely. These serum

levels are calculated to be at a point where no or minimal risk exists

for people drinking water with a certain PFAS.

Human equivalent dose or serum level divided by the total

uncertainty and modifying factors = toxicity value

Exposure

parameters

for drinking

water HBV

Breast-fed infant, which is also protective of a formula-fed infant

Placental transfer of 43% (MDHHS 2019)

Breastmilk transfer of 1.3% (MDHHS 2019)

Human serum half-life of 1241 days (3.2 years) (Li et al. 2018)

Volume of distribution of 0.23 L/kg (Thompson et al. 2010)

95th percentile drinking water intake, consumers only, from birth to

more than 21 years old (Goeden et al. [2019])

Upper percentile (mean plus two standard deviations) breast milk

intake rate (Goeden et al. [2019])

Time-weighted average water ingestion rate from birth to 30-35 years

of age (to calculate maternal serum concentration at delivery)

(Goeden et al. [2019])

Relative Source Contribution of 50%

Based on NHANES 95th percentiles for 3-11 (2013-2014) and over 12

years old (2015-2016) participants (CDC 2019)

The Workgroup discussed the Goeden et al. (2019) model

which considered full life stage exposure, from fetal

exposure, to infant exposure through breastfeeding, and

into adulthood. While the model was also developed for a

formula-fed infant, the breastfed infant scenario is

protective of a formula-fed infant. The Workgroup selected

this model for developing drinking water HBVs when the

needed inputs were available.

Drinking

water HBV

16 ng/L (ppt)

Numeric HBV derived and justified using the above

information

Chemical Summary for PFHxS

Decision point 

Rationale/justification

Critical

study

NTP 2018 TOX-96: Toxicity Report Tables and Curves for Short-term

Studies: Perfluorinated Compounds: Sulfonates and personal

communication between MDH and NTP project manager Dr.

Chad Blystone (as cited in the HRA Toxicology Review Worksheet

for PFHxS, last revised 3/8/2019)

The Workgroup reviewed available evaluations and focused

on the ones from Minnesota Department of

Health (2019) and ATSDR (2018). In both evaluations,

thyroid endpoints were selected.

The Workgroup discussed Chang et al. (2018) and

concluded that the health outcome (reduction in litter size)

was a marginal effect.

Description

of the critical

study

28-day oral toxicity study in Sprague Dawley rats (NTP,

2018). PFHxS was administered via daily gavage at the following

doses for 28 continuous days:

Male rats: 0, 0.625, 1.25, 2.5, 5 or 10 mg/kg/day

Male rats mean measured plasma levels: 0.102, 66.76, 92.08, 129.0,

161.7, and 198.3 µg/ml

Female rats: 0, 3.12, 6.25, 12.5, 25, 50 mg/kg/day

Female rats mean measured plasma levels: 0.1754, 37.03, 50.41, 63.82,

83.82, and 95.51 µg/ml

n=10/sex/dose

Critical effect: decreased serum free thyroxin (T

) levels was

observed in adult male rats at the lowest PFHxS dose administered

(0.625 mg/kg/day)

Co-critical effects: decreased free and total T

, triiodothyronine (T

and changes in cholesterol levels and increased hepatic focal

necrosis 

The Workgroup selected this thyroid endpoint as it was a

measure of a clinical or functional effect rather

than observational.

Point of

Departure

POD of 32.4 mg/L serum concentration for male rats based on

BMDL

. A BMR of 20% was used in the BMD modeling based on clinical

and toxicological knowledge regarding adverse outcomes associated with

decreases in circulating thyroid hormones. MDH stated that 20% provided

a more statistically reliable and biologically significant BMR. (MDH

conducted Benchmark Dose modeling and provided modeling run data in

the HRA Toxicology Review Worksheet for PFHxS, last revised

3/8/2019.

The Workgroup decided that serum-based points of

departure were appropriate for PFAS.

Although the Workgroup concluded that the Chang et al.

(2018) health outcome was marginal, they did note that the

serum concentration at the NOAEL for Chang et al. (2018)

was equivalent to the serum concentration at the selected

POD.

Human

equivalent

dose

The POD (32.4 mg/L) was multiplied by a toxicokinetic adjustment

based on the chemical’s specific clearance rate of 0.000090 L/kg-d

(Vd = 0.25 L/kg [Sundstrom et al. [2012], half-life = 1935 days [Li et al.

2018]) for a human equivalent dose of 0.00292 mg/kg/day.

The Workgroup selected the human serum half-life from Li

et al. (2018) as it was a non-occupational population

drinking water with elevated PFAS.

Uncertainty

factors

Total Uncertainty Factor of 300

• 1 for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability (toxicodynamic

differences)

• 1 for subchronic to chronic

• 10 for database deficiencies - to address concerns for early life

sensitivity and lack of 2-generation or immunotoxicity studies

The Workgroup reviewed the uncertainty factors used by

MDH (2019) and concluded that the database uncertainty

factor of 10 was very defensible in this situation, especially

for the lack of information on early-life sensitivity.

Toxicity

value

9.7 ng/kg/day (9.7 x 10

-6

mg/kg/day) which corresponds to a serum

concentration of 0.11 µg/ml

Serum levels used in development of these toxicity levels are not

meant to indicate a level where health effects are likely. These serum

levels are calculated to be at a point where no or minimal risk exists

for people drinking water with a certain PFAS.

Human equivalent dose or serum level divided by the total

uncertainty factors = toxicity value

Exposure

parameters

for drinking

water HBV

Breast-fed infant, which is also protective of a formula-fed infant

Placental transfer of 80% (MDHHS 2019)

Breastmilk transfer of 1.2% (MDHHS 2019)

Human serum half-life of 1935 days (Li et al. [2018])

Volume of distribution of 0.25 L/kg (MDH [2019] based on

Sundstrom et al. [2012])

percentile drinking water intake, consumers only, from birth to more

than 21 years old (Goeden et al. [2019])

Upper percentile (mean plus two standard deviations) breast milk

intake rate (Goeden et al. [2019])

Time-weighted average water ingestion rate from birth to 30-35 years

of age (to calculate maternal serum concentration at

delivery) (Goeden et al. [2019])

Relative Source Contribution of 50% (0.5)

Based on NHANES 95

percentiles for 3-11 (2013-2014) and over 12

years old (2015-2016) participants (CDC 2019)

The Workgroup discussed the Goeden et al.

(2019) model which considered full life stage exposure, from

fetal exposure, to infant exposure through breastfeeding,

and into adulthood. While the model was also developed for

a formula-fed infant, the breastfed infant scenario is

protective of a formula-fed infant. The Workgroup selected

this model for developing drinking water HBVs when

the needed inputs were available.

Drinking

water HBV

51 ng/L (ppt)

Numeric HBV derived and justified using the above

information

Chemical Summary for PFBS

Decision point 

Rationale/justification

Critical

study

Feng, X; Cao, X; Zhao, S; Wang, X; Hua, X; Chen, L; Chen, L. (2017).

Exposure of pregnant mice to perfluorobutanesulfonate causes

hypothyroxinemia and developmental abnormalities in female

offspring. Toxicol Sci 155: 409-419.

The Workgroup evaluated available agency decision

documents and selected the study associated with the draft

USEPA (2018) PFBS toxicity value based on thyroid effects.

The kidney effects identified in the draft USEPA (2018)

toxicity assessment were identified as a potentially

compensatory response. The thyroid effects were

identified as having greater functional significance.

Description

of the critical

study

PFBS was orally administered to pregnant ICR mice (n=30/dose) at

doses of 0, 50, 200, and 500 mg/kg/day from gestational day (GD) 1 to

GD20. Dams (F0) and female offspring (F1) from each dose

group were subsequently evaluated for 1) growth and development, 2)

hormone levels, and 3) serum PFBS levels. The critical effect is

decreased serum total thyroxine (T

) in newborn (PND 1) mice.

Selection of total T

as the critical effect is based on a several key

considerations that account for cross-species correlations in thyroid

physiology and hormone dynamics particularly within the context of a

developmental life stage.

Point of

Departure

A POD of 28.19 mg/kg/day (BMDL

) for decreased serum total T

newborn (PND 1) mice was selected

The Workgroup noted that a Benchmark Dose approach is

preferable to a NOAEL/LOAEL.

The Workgroup noted that the thyroid point of departure

would be protective of the kidney effects as well.

The draft USEPA (2018) toxicity assessment contained

administered doses from the individual studies converted to

HED doses using study-specific Dosimetric Adjustment

Factors (DAF; not reported for each dosing group) derived

using allometric scaling (BW

3/4

) prior to BMD model

analysis.

An example DAF calculation was provided in Table 8 of the

draft USEPA (2018) toxicity assessment: dose x DAF = 200

x 0.149 = 29.9 mg/kg/day, where DAF equals

(BW

animal

1/4

)/(BW

human

1/4

) = 0.0399

1/4

÷ 80

1/4

= 0.149

The POD

HED

= 4.2 mg/kg/day for decreased serum total T

newborn (PND 1) mice (USEPA 2018).

The USEPA POD

HED

of 4.2 was divided by 0.149 (USEPA

example DAF) to obtain a BMDL

of 28.19 mg/kg/day.

Human

equivalent

dose

The BMDL

-HED is 0.0892 mg/kg/day.

The BMDL

of 28.19 mg/kg/day was divided by the Dose Adjustment

Factor of 316 (human serum half-life/female mouse serum half-life =

665 hours/2.1 hours = 316) (MDH, 2017).

The Workgroup evaluated the half-life based Dose

Adjustment Factor used by the Minnesota Department of

Health (MDH) (2017). As that allowed conversion of the

point of departure to a human equivalent dose using

chemical-specific information, the Workgroup selected this

approach over the allometric scaling used in the draft

USEPA (2018) PFBS toxicity assessment.

Uncertainty

factors

The total uncertainty factor is 300.

• 1 for LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability

• 1 for subchronic to chronic

• 10 for database deficiencies, for the lack of

neurodevelopmental, immunotoxicological, and chronic studies

The Workgroup discussed the uncertainty factors selected

in the draft USEPA (2018) toxicity assessment and

supported their use.

Toxicity

value

300 ng/kg/day (0.0003 mg/kg/day)

Human equivalent dose or serum level divided by the total

uncertainty factors = toxicity value

Exposure

parameters

for drinking

water HBV

percentile of water intake for consumers only (direct and indirect

consumption) for infants (birth to <1 year old) of 1.106 L/day, per

Table 3-1, USEPA Exposure Factors Handbook, 2019.

An infant body weight of 7.8 kilograms was used and represents a

time-weighted average for birth to 1 year old (Table 8-1, USEPA

2011).

A default Relative Source Contribution of 20% was included.

The Workgroup discussed the use of an upper percentile

water intake. The 95

percentile for consumers only was

selected as it would protect those drinking larger amounts of

water.

As insufficient human serum data was available to assess

the population’s exposure to PFBS from sources other than

drinking water, a default Relative Source Contribution of

20% was selected consistent with USEPA (2000) guidance.

Drinking

water HBV

420 ng/L (ppt)

Numeric HBV derived and justified using the above

information in the following equation:

 =

 ×   ×  ℎ

 

Chemical Summary for GenX

Decision point 

Rationale/justification

Critical

study

Oral (Gavage) Reproduction/ Developmental Toxicity Study in Mice

(OECD TG 421; modified according to the Consent Order) DuPont-

18405-1037 (2010) (also contains 90-day toxicity study information

and outcomes - that information is not described here)

The Workgroup evaluated the North Carolina Department

of Health and Human Services (2017) and draft USEPA

(2018) information. The draft USEPA (2018) evaluation

was identified as providing a more in-depth and robust

analysis and approach.

Description

of the critical

study

In a combined oral gavage reproductive/developmental toxicity study

in mice with HFPO dimer acid ammonium salt, the test compound was

administered by oral gavage to Crl:CD1(ICR) mice (25/sex/group) at

doses of 0, 0.1, 0.5, or 5 mg/kg/day, according to a modified OECD

TG 421. Parental F0 males were dosed 70 days prior to mating and

throughout mating through 1 day prior to scheduled termination.

Parental F0 females were dosed for 2 weeks prior to pairing and were

dosed through LD 20. F1 animals (offspring) were dosed daily

beginning on PND 21 through PND 40.

At 0.5 mg/kg/day, liver effects (increased absolute and relative weight

and histopathologic findings) were reported in both males and

females.

At 5 mg/kg/day, male and female F1 pups exhibited lower mean BWs

at PNDs 4, 7, 14, 21, and 28. Male F1 pups continued to exhibit lower

mean BWs at PNDs 35 and 40. The USEPA (2018) identified

additional developmental effects (delays in balanopreputial separation

and vaginal patency) that occurred at the same dose level, but the

biological significance of these effects are equivocal as described.

NOAEL (F0) = 0.1; LOAEL (F0) = 0.5 for liver effects (single-cell

necrosis in males, and increased relative liver weight in both sexes).

NOAEL (F1) = 0.5 for developmental effects (decreased pup

weights).

The Workgroup noted that while primarily industry-funded

studies are the only ones available, they followed

recognized testing guidelines and/or were published

following external peer-review. These studies appear to be

sufficient for developing values.

Point of

Departure

BMDL

= 0.15 mg/kg/day for liver single cell necrosis in parental

males (DuPont-18405-1037, 2010).

The Workgroup noted that the Benchmark Dose approach

is preferred over the use of a NOAEL/LOAEL.

USEPA (2018) evaluated the relevance of this endpoint in

humans and noted that, per the Hall criteria (Hall et al.,

2012) liver effects accompanied by effects such as

necrosis or inflammation, among others, are indicative of

liver tissue damage (USEPA, 2018).

While some liver effects in rodents are mediated through

PPARα and may be less relevant to humans, available

information indicates that liver single cell necrosis may be

mediated by a number of processes and pathways.

In PPARα-mediated rodent hepatocarcinogenesis, liver

necrosis is not a key event. (DeWitt and Belcher, 2018)

Human

equivalent

dose

A candidate POD

HED

was derived from the BMDL

for liver

effects using a BW

3/4

allometric scaling approach. A BW

0.0372

kg was identified as the mean BW of the F0 male mouse controls.

A BW

of 80 kg for humans was selected. The resulting DAF for

the allometric scaling of doses from mice to humans is 0.15. Using

the BMDL

of 0.15 mg/kg/day to complete the calculation results

in a POD

HED

for single-cell necrosis of the liver from DuPont-

18405-1037 (2010) of 0.023 mg/kg/day (USEPA 2018).

The Workgroup noted that a toxicokinetic adjustment from

the point of departure to human equivalent dose would

provide a chemical-specific conversion. However, no

chemical-specific data on human serum half-life was

available that would allow this conversion. Allometric

scaling, per USEPA (2011a) guidance, was used.

Uncertainty

factors

Total Uncertainty Factor of 300

• 1 for use of a LOAEL to NOAEL

• 10 for human variability

• 3 (10

0.5

) for animal to human variability

• 3 (10

0.5

) for subchronic-to-chronic

• 3 (10

0.5

) for database deficiencies, including lack of

epidemiological, and developmental

and immunotoxicological studies in laboratory animals

The Workgroup evaluated the uncertainty factors selected

by USEPA (2018). Given the deficiencies in the database,

including a lack of epidemiological studies and

developmental and immunotoxicological in laboratory

animals, a database uncertainty factor of 3 was retained.

In conjunction with the deficiencies covered by the

database uncertainty factor, the subchronic to chronic

uncertainty factor of 3 was identified as sufficient.

Toxicity

value

77 ng/kg/day (7.7 x10-5 mg/kg/day)

Human equivalent dose or serum level divided by the total

uncertainty = toxicity value

Exposure

parameters

for drinking

water HBV

percentile of water intake for consumers only (direct and

indirect consumption) for adults (>21 years old) of 3.353 L/day

, per

Table 3-1, USEPA Exposure Factors Handbook, 2019.

An adult body weight of 80 kilograms was used (Table 8-1,

USEPA 2011b).

A default Relative Source Contribution (RSC) of 20% was

included.

The Workgroup discussed the use of an upper percentile water

intake. The 95

percentile for consumers only was selected as it

would protect those drinking larger amounts of water.

As no human serum data was available to assess the population’s

exposure to GenX from sources other than drinking water, a

default Relative Source Contribution of 20% was

selected consistent with USEPA (2000) guidance.

The Workgroup evaluated the protectiveness of adult exposure in

combination with the point of departure. The NOAEL for

developmental effects described above was at a dose five times

higher than the NOAEL for liver necrosis effects. As a drinking

water value based on the developmental NOAEL would be higher

than the level presented below, the Workgroup decided that the

drinking water HBV below based on liver effects would be

sufficiently conservative to be protective of infant exposure.

Drinking

water HBV

370 ng/L (ppt)

Numeric HBV derived and justified using the above information in

the following equation:

 =

 ×   ×  ℎ

 

Rationale for Individual HBVs

While there are on-going discussions regarding the grouping of multiple PFAS into one drinking

water value, there is no consensus from the scientific community on which PFAS should be

grouped or the basis of that grouping. Grouping methods that have been applied include

combining multiple PFAS into one number based on known or assumed toxicity, carbon chain

length, and/or biological half-life (simple addition) as well as the use of relative ability of the

grouped PFAS to lead to a comparable health endpoint (toxic equivalency); the latter approach

being similar to those used for dioxins, furans, and coplanar polychlorinated biphenyls.

There is, however, scientific agreement that the long-chain PFAS (eight carbons and above for

carboxylates and six carbons and above for sulfonates) have similar toxicity. Based on the

similarity in toxicity for the long-chain PFAS, the Workgroup recommends use of the HBV for

PFNA (6 ng/L [ppt]) as a screening level for all other long-chain PFAS included on the USEPA

Method 537.1 analyte list for which the Workgroup did not develop an individual HBV. This

screening level should not be used to evaluate the risk of developing health effects, but as a

screening tool for EGLE/public water supplies to use for decision making.

Adverse health effects of long chain (six-carbon perfluorosulfonic acids or eight-carbon

perfluorocarboxylic acids) have been established in epidemiological and laboratory animal model

studies. These adverse health effects include kidney and testicular cancer, elevated serum

cholesterol, endocrine effects, immune effects, and reproductive effects (ATSDR, 2018). These

effects are supported by studies of different human populations exposed to a few or to many

PFAS, including those from populations of high PFAS exposure and the general population and

demonstrate that many different long-chain PFAS can produce similar adverse health effects in

exposed humans. However, while not all long-chain PFAS have robust data available for the

development of a HBV, the totality of evidence indicates that long-chain PFAS in drinking water

may pose risks of adverse health effects.

While health concerns are based on the total exposure to PFAS across many sources, because

drinking water is the predominant source of exposure for many people consuming contaminated

water, it remains the focus for health-based regulation based on current knowledge. Therefore,

monitoring of drinking water should continue and be based on levels that will be protective for

exposure to all PFAS.

At this time, it is recommended that the proposed HBV for PFNA be used as a screening level for

the long chain PFAS included in USEPA Method 537.1 that may be found in drinking water that

are not covered by an individual PFAS HBVs as presented in the Summary Table of Drinking

Water HBVs.

Summary of Conclusions

Summary Table of Drinking Water HBVs

Specific PFAS

Drinking Water

Health-based

Value

Chemical Abstract

Services Registry

Number (CASRN)

PFNA

6 ng/L (ppt)

375-95-1

PFOA

8 ng/L (ppt)

335-67-1

PFHxA

400,000 ng/L (ppt)

307-24-4

PFOS

16 ng/L (ppt)

1763-23-1

PFHxS

51 ng/L (ppt)

355-46-4

PFBS

420 ng/L (ppt)

375-73-5

GenX

370 ng/L (ppt)

13252-13-6

For all other PFAS on the USEPA Method 537.1 analyte list, the Workgroup recommendation is

to use the lowest long-chain (eight carbons and above for carboxylates and six carbons and above

for sulfonates) HBV of 6 ppt, which is the HBV for PFNA. Those other long-chain PFAS included

in USEPA Method 537.1 are: NEtFOSAA (CASRN: 2991-50-6); NMeFOSAA (CASRN: 2355-31-

9); PFDA (CASRN: 335-76-2); PFDoA (CASRN: 307-55-1); PFTA (CASRN: 376-06-7); PFTrDA

(CASRN: 72629-94-8); and PFUnA (CASRN: 2058-94-8).

As shown in Figure 1 (below), the drinking water values for PFOS and PFOA have gone down

over time. This is a reflection of the evolving science, both the ever-increasing knowledge gained

from published toxicology and epidemiology studies and the risk assessments for development

of toxicity values and drinking water values. Information continues to become available on multiple

PFAS and as there are thousands of PFAS, new information will likely become available for many

years to come. It is quite possible that the same trend demonstrated in Figure 1 will be seen for

other PFAS, where drinking water values become lower over time and that new values could be

developed within a few years’ time. As described in the Challenges and Limitations section, along

with use of current scientific data, development of drinking water values includes a certain amount

of scientific judgement informed from the scientific knowledgebase. It is that combination of

scientific judgement and data that ultimately informs the development of drinking water values.

With emerging contaminants like PFAS, rapid availability of data drives public health protective

actions and drinking water values.

PFOS and PFOA

Figure 1: Screening Levels, Health-Based Values, and Regulatory Standards for PFOS and PFOA Over a 20-Year Timeframe.

The numbers in Figure 1 are the various screening levels, HBVs, and regulatory standards

developed by various agencies and states over time as of June 2019. It does not include the

agencies that include multiple PFAS into a single value. This should not be considered an

exhaustive list of all PFAS drinking water values available, and values may be updated, and

additional values will likely become available. The Michigan values included in Figure 1 are the

MPART Human Health Workgroup public health drinking water screening levels.

Concluding Remarks

The Workgroup would like to commend the State of Michigan for addressing PFAS concerns with

unusual rigor, openness, and reliance on independent scientific guidance. From the beginning of

the recognition of environmental and public health issues related to PFAS, the State of Michigan

has been at the forefront nationally in assessing the scope of the contamination, intervening to

mitigate exposure, and monitoring the evidence to guide policy. The statewide survey of drinking

water supplies was highly unusual if not unique relative to other areas, and the process of

developing Maximum Contaminant Levels as rigorous as any in the nation. By engaging experts

from outside the state agencies to complement the considerable expertise of the staff in the

Michigan Departments of Health and Human Services and Environment, Great Lakes, and

Energy, they have demonstrated their commitment to following the evidence through to

developing sound policy.

References

ATSDR. (2018). Agency for Toxic Substances and Disease Registry. Toxicological Profile for

Perfluoroalkyls. Draft for Public Comment. June 2018.

Bartell SM, Calafat AM, Lyu C, et al. 2010. Rate of decline in serum PFOA concentrations after

granular activated carbon filtration at two public water systems in Ohio and West Virginia.

Environ Health Perspect 118(2):222-228.

Beck, N.B., Becker, R.A., Erraguntla, N., Farland, W.H., Grant, Gray, G., Kirman, C., LaKind,

J.S., Lewis, R.J., Nance, P., Pottenger, L.H., Santos, S.L., Shirley, S., Simon, T., and Dourson,

M.L. (2016). Approaches for describing and communicating overall uncertainty in toxicity

characterizations: U.S. Environmental Protection Agency’s Integrated Risk Information System

(IRIS) as a case study. Environment International 89-90, 110-128.

Blystone, C. (2019). [Personal Communication. Use of NTP data tables and study protocol

(January 2019 email exchange).]., cited in Minnesota Department of Health (2019).

Toxicological Summary for Perfluorohexane sulfonate

. Accessed online June 2019 at:

CDC. (2019). (Center for Disease Control)

Fourth National Report on Human Exposure to

Environmental Chemicals, Updated Tables, January 2019, Volume One.

Chang, S., JL Butenhoff, GA Parker, PS Coder, JD Zitsow, RM Krisko, JA Bjork, KB Wallace,

JG Seed. (2018). Reproductive and developmental toxicity of potassium

perfluorohexanesulfonate in CD-1 mice. Reproductive Toxicology, 78, 150-168.

Das KP, Grey BE, Rosen MB, et al. 2015. Developmental toxicity of perfluorononanoic acid in

mice. Reproductive Toxicology 51:133-144.

DeWitt and Belcher (2018)

Memorandum from Jamie DeWitt and Scott Belcher to the members

of the North Carolina Secretaries’ Science Advisory Board. Accessed online June 2019.

Dong GH, Zhang YH, Zheng L, Liu W, Jin YH, He QC. (2009). Chronic effects

of perfluorooctanesulfonate exposure on immunotoxicity in adult male C57BL/6 mice.

Arch Toxicol. 83(9):805-815.

DuPont-18405-1037: E.I. du Pont de Nemours and Company (2010) An Oral (Gavage)

Reproduction/Developmental Toxicity Screening Study of H-28548 in Mice. USEPA OPPTS

870.3550; OECD Test Guideline 421. Study conducted by WIL Research Laboratories, LLC

(Study Completion Date: December 29, 2010), Ashland, OH. (as cited in USEPA (2018)).

Feng, X; Cao, X; Zhao, S; Wang, X; Hua, X; Chen, L; Chen, L. (2017). Exposure of pregnant

mice to perfluorobutanesulfonate causes hypothyroxinemia and developmental abnormalities in

female offspring. Toxicol Sci 155: 409-419.

Goeden HM, Greene CW, Jacobus, JA (2019) A transgenerational toxicokinetic model and its

use in derivation of Minnesota PFOA water guidance. J. Exposure Sci. Env. Epidemiol. 29:183-

195.

Goeden, HM., CW Greene, JA Jacobus. (2019). A transgenerational toxicokinetic model and its

use in derivation of Minnesota PFOA water guidance. Journal of Exposure Science &

Environmental Epidemiology.

Hall, AP, Elcombe, CR, Foster, JR, et al. (2012) Liver Hypertrophy: A Review of Adaptive

(Adverse and Non-adverse) Changes—Conclusions from the 3rd International ESTP Expert

Workshop. Toxicologic Pathol. 40: 971-94.

Klaunig, JE, Shinohara, M, Iwai, H, Chengelis, CP, Kirkpatrick, JB, Wang, Z, Bruner, RH (2015)

Evaluation of the chronic toxicity and carcinogenicity of perfluorohexanoic acid (PFHxA) in

Sprague-Dawley rats. Toxicol. Pathol. 43 (2), 209–220.

Koskela A, Finnilä MA, Korkalainen M, Spulber S, Koponen J, Håkansson H, Tuukkanen J,

Viluksela M. (2016) Effects of developmental exposure to perfluorooctanoic acid (PFOA) on

long bone morphology and bone cell differentiation. Toxicol Appl Pharmacol. 301:14-21.

Li, Y., T Fletcher, D Mucs, K Scott, CH Lindh, P Tallving, K Jakobsson. (2018). Half-lives of

PFOS, PFHxS and PFOA after end of exposure to contaminated drinking water. Occupational

and Environmental Medicine, 75, 46-51.

Loveless, S.E., Slezak, B., Serex, T., Lewis, J., Mukerji, P., O'Connor, J.C., Donner, E.M.,

Frame, S.R., Korzeniowski, S.H., Buck, R.C., 2009. Toxicological evaluation of

sodium perfluorohexanoate. Toxicology 264 (1–2), 32–44.

Luz, AL, Anderson, JK, Goodrum, P, Durda, J. (2019) Perfluorohexanoic acid toxicity, part I:

Development of chronic human health toxicity value for use in risk assessment.

Reg. Toxicol. Pharmacol. 103: 41-55.

Minnesota Department of Health (2019).

Toxicological Summary for Perfluorooctane sulfonate

(PFOS). Accessed online June 2019.

Minnesota Department of Health (2017).

Toxicological Summary for: Perfluorobutane Sulfonate

(PFBS). Accessed online June 2019.

Minnesota Department of Health (2017). Background Document: Toxicokinetic Model for PFOS

and PFOA and Its Use in the Derivation of Human Health-based Water Guidance Values.

Minnesota Department of Health (2019). HRA Toxicology Review Worksheet for PFHxS, last

revised 3/8/2019.

North Carolina Department of Health and Human Services.

Questions and Answers Regarding

North Carolina Department of Health and Human Services Updated Risk Assessment for GenX

(Perfluoro-2-propoxypropanoic acid). July 14, 2017. Accessed online May 2019.

National Research Council (NRC). (2009). Science and Decisions: Advancing Risk Assessment.

Washington D.C.: National Academies Press.

NTP. (2018). National Toxicology Program. TOX-96: Toxicity Report Tables and Curves for

Short-term Studies: Perfluorinated Compounds: Sulfonates. Retrieved from

https://tools.niehs.nih.gov/cebs3/views/?action=main.dataReview&bin_id=3874., cited

in Minnesota Department of Health (2019). Toxicological Summary

for Perfluorohexane sulfonate. Accessed online June 2019 at:

https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfhxs.pdf.

New Jersey DEP DWQI. (2018).

Appendix A - Health-based Maximum Contaminant Level

Support Document: Perfluorooctane sulfonate. Trenton, NJ: New Jersey Drinking Water Quality

Institute.

New Jersey DEP DWQI. (2015).

Maximum contaminant level recommendation

for perfluorononanoic acid in drinking water. Trenton, NJ: New Jersey Drinking Water Quality

Institute.

Ohmori K, Kudo N, Katayama K, et al. 2003. Comparison of

the toxicokinetics between perfluorocarboxylic acids with different carbon chain length.

Toxicology 184:135-140.

Onishchenko N, Fischer C, Wan Ibrahim WN, Negri S, Spulber S, Cottica D, Ceccatelli S.

(2011) Prenatal exposure to PFOS or PFOA alters motor function in mice in a sex-related

manner. Neurotox Res. 19(3):452-61.

Pachkowski B, Post GB, Stern AH. (2019)

The derivation of a Reference Dose (RfD) for

perfluorooctane sulfonate (PFOS) based on immune suppression. Environ Res. 2019 171:452-

469.

Peden-Adams, MM, Keller, JM, EuDaly, JM, Berger, J, Gilkeson, GS, Keil, DE.

(2008) Suppression of Humoral Immunity in Mice following Exposure

to Perfluorooctane Sulfonate. Toxicol. Sci. 104(1): 144-154.

Russell, MH, Nilsson, H, Buck, RC (2013) Elimination kinetics of perfluorohexanoic acid in

humans and comparison with mouse, rat and monkey. Chemosphere. 93: 2419-2425.

Sundstrom, M., SC Chang, PE Noker, GS Gorman, JA Hart, DJ Ehresman, A Bergman,

JL Butenhoff. (2012). Comparative pharmacokinetics of perfluorohexanesulfonate (PFHxS) in

rats, mice, and monkeys. Reproductive Toxicology, 33, 441-451.

Thompson, J., M. Lorber, L.-M.L. Toms, K. Kato, A.M. Calafat, and J.F. Mueller. 2010. Use of

simple pharmacokinetic modeling to characterize exposure of Australians to perfluorooctanoic

acid and perfluorooctane sulfonic acid. Environment International 36:390–397.

United States Environmental Protection Agency (2000).

Methodology for Deriving Ambient

Water Quality Criteria for the Protection of Human Health. EPA-822-B-00-004. USEPA Office of

Water, Washington, DC. Accessed online June 2019.

United States Environmental Protection Agency. (2002).

A Review of the Reference Dose and

Reference Concentration Processes. Risk Assessment Forum. EPA/630/P-02/002F, December

2002.

United States Environmental Protection Agency. (2011)

Exposure Factor Handbook: 2011

Edition. EPA/600/R-09/052F. USEPA, National Center for Environmental Assessment, Office of

Research and Development, Washington, DC. Accessed online June 2019.

United States Environmental Protection Agency (2011a). Recommended Use of Body Weight

3/4

as the Default Method in Derivation of the Oral Reference Dose. EPA/100/R11/0001. USEPA,

Office of the Science Advisor, Risk Assessment Forum, Washington, DC. Accessed online May

2018.

United States Environmental Protection Agency (2011b).

Exposure Factors Handbook 2011

Edition. EPA/600/R-09/052F. USEPA, National Center for Environmental Assessment, Office of

Research and Development, Washington, DC. Accessed online May 2019.

United States 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. USEPA, Office of Research and Development, Washington, DC. Accessed online

June 1, 2019.

United States Environmental Protection Agency (2018).

Human Health Toxicity Values for

Hexafluoropropylene Oxide (HFPO) Dimer Acid and Its Ammonium Salt (CASRN 13252-13-6

and CASRN 62037-80-3), Also Known as “GenX Chemicals”. Public Comment Draft. EPA-823-

P-18-001. USEPA Office of Water, Washington, DC. Accessed online May 2019.

United States Environmental Protection Agency (2019)

Exposure Factors Handbook Chapter 3

(Update): Ingestion of Water and Other Select Liquids. EPA/600/R-18/259F. USEPA Office of

Research and Development, Washington, DC. Accessed online June 2019.

Wambaugh JF, Setzer RW, Pitruzzello AM, et al. 2013. Dosimetric anchoring of in vivo and in

vitro studies for perfluorooctanoate and perfluorooctanesulfonate. Toxicol Sci 136(2):308-327.

Zhang Y, Beesoon S, Zhu L, et al. 2013. Biomonitoring of perfluoroalkyl acids in human urine

and estimates of biological half-life. Environ Sci Technol 47(18):10619-10627.

Appendix A: Acronym List

ATSDR Agency for Toxic Substances and Disease Registry

BMD benchmark dose

BMDL lower confidence limit on the benchmark dose

BMR benchmark response

BW body weight

BWa body weight animal

BWh body weight human

CDC Centers for Disease Control and Prevention

DAF dosimetric adjustment factor

EGLE Environment, Great Lakes, and Energy (Michigan Department of)

GD gestational day

GenX perfluoro-2-propoxypropanoic acid

HBV health-based value

HED human equivalent dose

HFPO hexafluoropropylene oxide

HRA health risk assessment

kg kilogram

L liter

LD lactation day

LHA lifetime health advisory

LOAEL lowest observed adverse effect level

MCL Maximum Contaminant Level

MDH Minnesota Department of Health

MDHHS Michigan Department of Health and Human Services

mg milligram

MI Michigan

ml milliliter

MPART Michigan PFAS Action Response Team

µg microgram

ng nanogram

NHANES National Health and Nutrition Examination Survey

NJDEP New Jersey Department of Environmental Protection

NOAEL no observed adverse effect level

OECD Organization for Economic Co-operation and Development

PFAS per- and polyfluoroalkyl substances

PFBS perfluorobutane sulfonic acid

PFHxA perfluorohexanoic acid

PFHxS perfluorohexane sulfonic acid

PFNA perfluorononanoic acid

PFOA perfluorooctanoic acid

PFOS perfluorooctane sulfonic acid

PND postnatal day

POD point of departure

POD

HED

point of departure human equivalent dose

PPAR peroxisome proliferator-activated receptor

ppt parts per trillion

RfD reference dose

RSC relative source contribution

TWA time weighted average

UF uncertainty factor

USEPA United States Environmental Protection Agency

Appendix B: MPART Motion for Creation of Science Advisory Workgroup,

April 4, 2019

Motion

Motion to establish a Science Advisory Workgroup with the Charge described below, comprised

of external members with expertise in toxicology, epidemiology, and risk assessment, and

further to authorize the chairperson of MPART to finalize the appointments in consultation with

MPART members.

Preamble

On March 26, 2019, Governor Whitmer directed the Michigan PFAS Action Response Team

(MPART) to further protect public health and the environment, by forming a Science Advisory

Workgroup to “review both existing and proposed health-based drinking water standards from

around the nation to inform the rule making process for appropriate Maximum Contaminant

Levels for Michigan...” Toward this objective, the Science Advisory Workgroup shall make

numeric recommendation(s) to MPART for those per- and polyfluoroalkyls substances (PFAS)

for which adequate information exists.

Charge

The Science Advisory Workgroup shall:

1. For the PFAS listed in USEPA Method 537.1, review all existing and proposed national-

and state-derived PFAS drinking water standards and identify the most scientifically

defensible non-cancer or cancer-based public health toxicity values available for each

individual PFAS chemical family member, or combination thereof, for which the Science

Advisory Workgroup determines that adequate information exists. Provide written

justification that shall include, but not be limited to, the basis for the selection of the

primary study, critical effect identification, point of departure determination, evaluation of

all uncertainty and/or modification factors applied, and the non-cancer or cancer-based

toxicity value derivation.

Review all existing and proposed national- and state-derived PFAS drinking water

standards and identify the most scientifically defensible exposure assessment and risk

evaluation methodology for each individual PFAS chemical family member, or

combination thereof, for which the Science Advisory Workgroup determines that

adequate information exists. Provide written justification that shall include, but not be

limited to, selection of the most appropriate receptor(s) and identification of all

appropriate exposure assumptions for the receptor(s).

Identify the most appropriate and scientifically defensible combination of each specific

PFAS toxicity value and exposure assessment and risk evaluation methodology,

including consideration of relative source contribution, from which to derive a health-

based drinking water value for each individual PFAS chemical family member, or

combination thereof, for which the Science Advisory Workgroup determines that

adequate information exists.

Provide to MPART no later than July 1, 2019, a report recommending scientifically-

defensible numeric health-based values to inform the rulemaking process for Maximum

Contaminant Levels for each individual PFAS chemical family member, or combination

thereof, with written justification for the calculation methodology and each input into used

in the methodology by the Science Advisory Workgroup.

End

Appendix C: USEPA Method 537.1 Analyte List

Analyte Name* Acronym

Fluorinated

Carbon Chain

Length

Chemical Abstract

Services Registry

Number (CASRN)

Perfluorotetradecanoic acid

PFTeA

376-06-7

Perfluorotridecanoic acid

PFTriA

72629-94-8

Perfluorododecanoic acid

PFDoA

307-55-1

Perfluoroundecanoic acid

PFUnA

2058-94-8

Perfluorodecanoic acid

PFDA

335-76-2

Perfluorononanoic acid

PFNA

375-95-1

Perfluorooctanoic acid

PFOA

335-67-1

Perfluoroheptanoic acid

PFHpA

375-85-9

Perfluorohexanoic acid

PFHxA

307-24-4

Perfluorooctanesulfonic acid

PFOS

1763-23-1

Perfluorohexanesulfonic acid

PFHxS

355-46-4

Perfluorobutanesulfonic acid

PFBS

375-73-5

2-(N-Ethylperfluorooctanesulfonamido)

acetic acid

N-EtFOSAA

2991-50-6

2-(N-Methylperfluorooctanesulfonamido)

acetic acid

N-MeFOSAA

2355-31-9

Hexafluoropropylene oxide dimer acid

HFPO-DA

(GenX)

13252-13-6

11-chloroeicosafluoro-3-oxaundecane-1-

sulfonic acid

11Cl-PF3OUdS

763051-92-9

9-chlorohexadecafluoro-3-oxanone-1-

sulfonic acid

9Cl-PF3ONS

756426-58-1

4,8-dioxa-3H-perfluorononanoic acid

ADONA

919005-14-4

HFPO-DA is one component of the GenX processing aid technology.

11Cl-PF3OUdS is available in salt form (e.g. CASRN of potassium salt is 83329-89-9).

9Cl-PF3ONS analyte is available in salt form (e.g. CASRN of potassium salt is 73606-19-6)

ADONA is available as the sodium salt (no CASRN) and the ammonium salt (CASRN is 958445-448).

* Some PFAS are commercially available as ammonium, sodium, and potassium salts. This method measures all

forms of the analytes as anions while the counterion is inconsequential. Analytes may be purchased as acids or as

any of the corresponding salts.

Appendix D: Timeline for the Science Advisory Workgroup’s Development of Drinking Water HBVs

Appendix E: Timeline of the Maximum Contaminant Level Development Process