MINUTES FROM I II I EPA SCIENCE ADVISORY BOARD
Perflourooctanoic Acid (PFOA) Draft Risk Assessment Review Panel
Face-to-Face Meeting Conference Meeting
February 22-23, 2005

PURPOSE: The Perflourooctanoic Acid (PFOA) Draft Risk Assessment Review Panel of the
EPA Science Advisory Board (SAB) met on February 22-23, 2005 to review the Agency's Draft
Risk Assessment of Potential Human Health Effects Associated with PFOA and Its Salts. The
purpose of the teleconference meeting was for the SAB PFOA Review Panel to consider
available advisory and background materials, to hear oral presentations by members of the public
and other interested parties, and to discuss and deliberate on the draft charge questions to the
SAB. Attachment A is the Federal Register notice announcing the teleconference (70 FR 8,
January 12, 2005). A meeting agenda is included as Attachment B.

LOCATION: SAB Conference Center, Woodies Building, 3rd floor, 1025 F St., NW,
Washington, DC

DATE AND TIME: February 22-23, 2005, 8:30 - 5:00 pm Eastern Time.

PARTICIPANTS: The following individuals participated in this meeting: PFOA Review
Panel Members - Drs. Deborah Cory-Slechta (Chair), James Kehrer, Norman Drinkwater, James
Klaunig, Ron Melnick, Ernest Abel, Thomas Zoeller, Steve Roberts, Mathew Longnecker,
Michael Kamrin, Melvin Andersen, William Hayton, Frank Mink, George Corcoran, David
Ozonoff, and Anne Sweeney. Dr. Buck-Louis was unable to participate in this panel. The
PFOA Review Panel roster is included as Attachment C and a set of biographical sketches is
included in Attachment D. SAB Staff - Dr. Vanessa Vu, SAB Staff Office Director, Mr. Tom
Miller and Dr. Sue Shallal, Designated Federal Officers (DFO); EPA Staff Presenters - Dr.
Jennifer Seed of the EPA Office of Pollution Prevention and Toxics and Dr. Hugh Barton of the
EPA Office of Research and Development (ORD); Other Participants - Approximately 30 other
EPA Staff and members of the public listened in (Attachment E)

MEETING SUMMARY: The meeting followed the agenda (Attachment B). A summary of
the meeting follows.

Convene the Meeting and Introductory Remarks - Dr. Suhair Shallal, Designated
Federal Officer (DFO), opened the meeting at 8:30 pm and gave an overview of SAB procedures
for panel formation and then outlined the purpose of this meeting, namely to review the draft
PFOA RA document and respond to the charge questions. Dr. Shallal also explained the next
steps regarding the development and approval of the SAB Panel's report. Dr. Shallal then
introduced Dr. Cory-Slechta, Chair of the PFOA Review Panel, to proceed with the agenda.

Dr. Cory-Slechta gave Dr. Vu of the SAB Staff Office an opportunity to speak. Dr. Vu
thanked Dr. Cory-Slechta for her willingness to serve as Chair of PFOA Review Panel and
thanked the other Panel Members for taking time from their busy schedules to participate in this
review. Dr. Vu reiterated that the report would take several weeks to complete and must be

1


-------
approved by the Members of the Board before it is transmitted to the Administrator.

Dr. Cory-Slechta then reviewed the agenda and asked panel members to introduce themselves.
She then introduced the first speaker Mr. Jim Willis.

Overview of the draft PFOA Risk Assessment

Mr. Willis referred participants to the powerpoint presentation that he was using as a
guide for his talk (Attachment F). He explained that the presentation was written to highlight the
background and the Agency's ongoing activities to develop a risk assessment for PFOA and its
salts.

He noted that interest in perfluorinated compounds began in the late 1990's. PFOA has
unique chemical properties that make it a commercially valuable chemical with numerous uses
(elastomers, flame retardants, lubricants, architectural coatings, etc.). He informed the panel
members of ongoing activities to gather more data on PFOA that include: biomonitoring by
CDC, toxicology data being generated by EPA ORD and NTP, as well as, activities to try to
understand the pathways of exposure to PFOA.

He reminded panel members that the current draft RA considers information available as
of June 2004 and that it is not intended to provide final risk estimates. He stated that OPPTS is
seeking advice on their conclusions with regards to the carcinogenic potential of PFOA, the
appropriateness of the toxicological endpoints that were selected, the pharmacokinetic modeling
and internal dose metrics approaches that were used, cross-species extrapolation and robustness
of the biomonitoring database.

Dr. Cory-Slechta thanked Mr. Willis for his presentation and allowed panel members to
ask questions.

Dr. Ozonoff inquired as to what is meant by "robust" database. Dr. Seed responded that
it meant how reliable it is. She also noted that more data will be generated through the CDC
NHANES study that Mr. Willis spoke of during his presentation.

There were no further questions and Dr. Cory Slechta proceeded to calling the public
commenters to make their presentations.

Presentation by Public Commenters

There were 8 individuals registered to speak (Attachment G). Ms. Deborah Cochran was
absent and therefore the next commenter was called, Dr. John Moore.

Dr. Moore focused his presentation (Attachment H) on the endpoints used in the draft
PFOA RA, suggested other data sets that should have been used, and advocated the use of a
Benchmark Dose approach. He agreed that the male rat body weight is an important endpoint,
but added that EPA's prenatal endpoint for male body weight gain between weaning and sexual
maturity reflects effects from direct dosing, not from prenatal exposure as evidenced by the York
(2002) study. Additionally, he contended that the Sibinski (1987) data on adult female body
weight effects was not suitable for establishing a NOAEL.

2


-------
Dr. Joseph Rodrick was the next presenter (Attachment I). He discussed the Agency's
MOE approach, internal dose comparisons, Benchmark Dose and uncertainty factors for
interspecies comparisons. He agreed with the use of the MOE approach comparing animal and
human serum levels. Furthermore, he stated that the data are adequate for internal dose (AUC)
comparisons for all endpoints. He also recommended the use of Benchmark Dose approach
instead of a NOAEL since it normalizes the data across the varying dose rates. He also believed
that the uncertainty factor for interspecies extrapolation can be reduced by using the data for
internal dose (serum levels).

Dr. Abel questioned why liver weight to brain weight ratio was used? Dr. Rodrick
explained that the brain weight is not as readily affected by insults and remains relatively stable.

Dr. Melnick asked about the BMD analysis and wondered if it would be useful to
characterize the lower dose range since 10 is within the range of experimental data. Dr.

Roderick stated that this was a rare database which has information about this many different
populations, and is thought to be reliable.

Dr. Geary 01 sen

Dr. Olsen showed (Attachment J) that human PFOA Biomonitoring Data is available
(e.g., Children study, American Red Cross adult study, Elderly study). He asserted that a re-
analysis of data from Kannan et al (YEAR?) showed a lower serum level than previously
reported. He also presented data from a half-life of serum elimination study using retired 3M
workers that showed a half-life of 3.8 years for PFOA. He concluded that PFOA biomonitoring
data was consistent in the general population, PFOA serum levels are at near steady state
condition, and that these were adequate for use in calculating a MOE.

Dr. Cory Slechta enquired about the stability of PFOA. Dr. Olsen stated that it was
stable through several freeze-thaw cycles and when kept at -20°C. When Dr. Olsen was asked
about the rationale for using retirees, he stated that "new" (non-occupational exposures) were
minimal, in the range of 0.07 -5.0 ug/ml. Dr. Melnick questioned the accuracy of the conclusion
that the retirees had reached a steady state condition since children have the same blood levels.
Dr. Melnick also wondered if children could be at steady state when the half-life was up to 4.5
years.

Dr. James Popp

Dr. James Popp was the next presenter (Attachment K) and began by stating that the
PFOA RA was a well written document covering a vast amount of data. He devoted his
comments to the Agency's final characterization of the carcinogenic potential of PFOA:

. suggestive evidence of carcinogenicity, but not sufficient to assess human carcinogenic
potential". He felt that it overstates the level of concern. He argued that PFOA probably does
not cause a human carcinogenic response due to its mode of action and internal exposure
differential between humans and rodents. Further, he explained that this conclusion was
consistent with other reviews (i.e., Hepatic tumors PPAR alpha agonists . .Not likely to occur

3


-------
in humans.." ILSI review and Non hepatic tumors ".. .probably do not represent a significant
cancer hazard for humans..EPA draft pages 8 and 84).

Dr. Popp explained that in the case of mammary tumors, there was no evidence of tumor
progression since the high dose showed a lower tumor incidence than the lower doses. The
incidence of tumors falls within historical control data. He stated that rat liver tumors are known
to occur via a PPAR alpha MOA which is not relevant to humans.

Dr. Ozonoff asked if there was any information regarding PPAR gamma agonist activity
or immunotoxicity issues with regards to PFOA. Dr. Popp responded that PPAR alpha receptor
binding occurs with PFOA. Cell proliferation occurs and is then followed by peroxisome
proliferation.

Dr. Corcoran added that the timing of the necrosis analysis is important. In the original
study, necrosis was described but no liver tumors were found. In the second study, tumors were
found but necrosis was not evaluated in addition no cell proliferation was found.

Dr. Cory Slechta indicated that PPARalpha receptors are found in the brain. What are
the effects of PFOA on the brain? There is no information on brain effects.

Dr. Drinkwater noted that the mutant mouse data show an increase liver weight through a
receptor-independent MOA. There is no histopathalogy associated with the increase in weight
and appears to be a degenerative response. There appears to be another pathway other than
PPAR alpha at work in mice.

Dr. Jane Houlihan and Dr. Timothy Kropp

Dr. Houlihan began the presentation (Attachment L) by stating that there are eight health
effects from PFOA that have failed to identify a NOAEL, including ovary, pituitary, kidney,
hormonal systems, immune system, spleen, male reproductive system (seminal vesicles, Leydig
cell morphology), and developmental effects (growth). She further stated that PFOA causes
adverse health effects through five basic mechanisms (modes of action), At least four of these
are relevant to humans, mitochondrial toxicity, inhibition of gap junction , intercellular
communication, thyroid hormone, serum estradiol changes (via liver aromatase), and
peroxisome proliferation

Dr. Kropp suggested that the Science Advisory Board's PFOA Panel urge EPA to the
relevance to humans of mammary, testicular and pancreatic cancer findings, to consider the
relevance of tumors to infants and children, to adopt the standard risk assessment methodology
we have presented here, and to err on the side of human health protection when technical choices
arise.

Dr. Kropp continued to suggest that PFOA be classified as a likely human carcinogen
since tumors are seen in more than one species and in more than one sex. For non-cancer risk,
he stated that EWG recommended a BMDL approach.

4


-------
Mr. Robert Griffin (Attachment M)

Mr. Griffin identified himself as a civil engineer and is the general manager of a non-
profit rural water system located in Southeast Ohio, The Little Hocking Water Association,
which serves approximately 12,000 people. Mr Griffin stated that their water wells are located
along the Ohio River directly across from Dupont's Washington works plant in Wood County,
West Virginia where C8 is used in the fluoropolymer manufacturing process and where telomers
are also produced.

Mr. Griffin reminded the panel that PFOA is an industrial chemical for which there is
very limited scientific data to determine the acceptable level of that chemical in drinking water.
As a consequence, he urged that the panel to put public health first by applying the precautionary
principle.

Another public commenter, Ms. Debra Cochran, had not yet arrived.

Mr. Robert Billot of Taft, Stettinius & Hollister LLP who had been inadvertently left off
the public commenter list was given 2 minutes to address the panel. He reminded the panel
members of the blood data that was included in public comments (Attachment N) provided to the
panel prior to the face-to-face meeting. These data included non-occupational blood levels for
individuals living near manufacturing facilities and on landfills. The duration of the exposures
were unknown, however the water testing occurred in 1984. There is a cancer survey being
conducted which is in press.

The panel reconvened at 1:15 PM after a one hour break.

The panel began discussing their assigned charge questions. Dr. Cory-Slechta asked Dr.
Drinkwater to summarize the findings for the Issue 1 group.

Issue 1

Dr. Drinkwater was the lead discussant for the Issue 1 group; its members include Dr.
Drinkwater, Dr. Kehrer, and Dr. Klaunig. Their charge question was: Please comment on the
weight of evidence and adequacy of the data available to identify the key events for the PPAR
alpha agonist induced rodent liver toxicity and hepatocarcinogenesis for PFOA. Discuss
whether the uncertainties and limitations of these data have been adequately characterized.

Dr. Drinkwater summarized the group's comments as follows:

The sequence of four key events that defines the mode of action by which PPAR-alpha agonists
induce liver tumors includes (1) activation of PPAR-alpha (2) increased cell proliferation and/or
decreased cell death. (3) clonal expansion of the preneoplastic lesions, (4) development of
hepatocellular neoplasms.

The issue 1 group agreed that the weight of evidence strongly supports the conclusion
that hepatotoxicity and liver tumor induction in rats by PFOA results from a mode of action

5


-------
involving PPAR-alpha agonism. However, studies of PPAR-alpha mutant mice by Yang el al.
(2002), which are cited in the report in the context of the receptor dependence of PFOA
immunotoxicity, did indicate an increased liver weight in PFOA treated mutant mice. A more
critical issue is whether arguments related to the relevance to humans of this mode of action for
induction of liver tumors in adults may be extended to exposed infants and children.

Additionally, there are no data available on the effects of peroxisome proliferators in
human Kupffer cells. The role of Kupffer cell activation in the induction of DNA synthesis and
subsequent neoplastic development by PPARa agonists has not been characterized and thus
results from the PPARa null mouse are not directly applicable to the human situation in which
PPARa is present and can be activated.

Much of the discussion that ensued focused on the Yang et al study using the PPAR
alpha null mouse and the role of Kupffer cells.

Issue 2: Descriptor for Carcinogenic Potential

Dr. Melnick was the lead discussant for the Issue 2 group; the members included Dr.
Melnick, Dr. Klaunig and Dr. Drinkwater. Their charge question was as follows: OPPThas
proposed that the PFOA cancer data may be best described as providing "suggestive evidence
of carcinogenicity, but not sufficient to assess human carcinogenic potential" under the interim
1999 EPA Guidelines for Carcinogen Risk Assessment, as well as the 2003 draft EPA Guidelines
for Carcinogen Risk Assessment. The determination of an appropriate descriptor for
carcinogenic potential requires an examination of the available carcinogenicity data, an
evaluation of mechanistic or mode-of-action (MOA) data, and guidance on how EPA applies
various descriptors for summarizing weight of evidence.

Dr. Melnick suggested that in order to respond to the charge question he would first
discuss the cancer studies for PFOA, the Mode of action for PFOA and the application of cancer
descriptors by the Agency.

Dr. Melnick concluded that the selection of "likely to be carcinogenic to humans" as
presented in the 2003 EPA cancer guidelines appears to be the best descriptor of carcinogenicity.
This was based on the fact that PFOA tested positive at multiple sites in Sprague-Dawley rats
(males: liver, testis, pancreas; females: mammary gland), some of the tumor responses were
highly significant and included relatively uncommon sites, and mechanistic data do not justify
the exclusion of these responses as irrelevant to humans.

A review of the cancer descriptors that are found in both the 1999 and 2003 draft cancer
guidelines ensued. Some panel members expressed some uneasiness in labeling PFOA a "likely
human carcinogen" when there is uncertainty that exists. They were not able to assign a
probability associated with the likelihood of being carcinogenic to humans.

Issue 3- Question 3

Dr. Roberts was the lead discussant for the Issue 3 group; the members included Dr.

6


-------
Roberts, Dr. Abel, Dr. Zoeller and Dr. Longnecker. Their charge question was as follows:
Please comment on the selection of these toxicity endpoints for the risk assessment.

Dr. Roberts stated that the toxicity endpoints chosen are reasonable and that the
presentation of results from multiple endpoints is valuable. He continued that organ and body
weights are likely to be among the least sensitive endpoints for toxicants that exert specific
effects on physiological or developmental systems. Nevertheless, in the absence of information
leading one to propose that PFOA exerts a specific effect on non-cancer endpoints, the proposed
endpoints are likely to be the most revealing.

Additional endpoints that were suggested by the issue 3 group members included:

1)	neoplastic responses observed in rats (liver, testis, pancreas, and mammary gland)

2)	ataxia in the female rat 2-year feed study (4% controls, 18% at 30 ppm, and 30% at 300 ppm)

3)	reductions in thyroxine and triodothyronine in the 6-month oral study in monkeys

4)	increased severity and incidence of ovarian stromal hyperplasia in the 2-year feed study in
female rats

5)	decreased pituitary weights in F1 female rats

The discussion that followed Dr. Roberts' summary further elaborated on the issue of the
lack of sensitivity of body weight and organ weight as an indicator of toxicity. Other
confounding factors that need to taken into account include maternal toxicity (e.g., malaise)
which may affect offspring development.

Issue 3- Question 4

Given the available data to date, please comment on the most appropriate
lifestage/gender/species for assessing human risk.

Dr. Roberts continued to explain the viewpoints of issue 3 group members regarding this
question. There are those who prefer a more "conservative" approach since there is already
substantial, widespread environmental contamination with this persistent, bioaccumulative
compound and the prospect of increasing levels of contamination in the future are possible.
Others prefer to use the adult non-human primate data as the most appropriate model for
assessing human risk. These test animals are most similar biologically to humans.

Panel members discussed the appropriateness of rat vs monkey data. Panel members
noted that the half-life of PFOA in rats appeared to ready a steady state, i.e., as the dose
increased, serum levels remained the same. The half-life of PFOA in monkey was very short.
There were additional limitations with the monkey data since it did not present any life-stage
data for comparison of effects in adults vs. infants and children.

Issue 3- Question 5

Please comment on the appropriateness of the available animal models. Please comment on
whether additional animal models should be investigated, and if so, what information would
better enable us to ascertain potential human risks.

7


-------
Dr. Roberts indicated that group members thought the available animal models are
appropriate. Wildlife may serve as sentinels. He added that for DDT it was effects in wildlife
that drove the policy to reduce use. He explained that continued monitoring of human
populations, especially exposed workers was important, with an obvious need for more
information about how humans are being exposed to this compound.

After Dr. Roberts completed his summary, panel members reiterated the need for further
studies with PPAR-alpha knockout mice and rats. Some panel members suggested that the
animal model to be used is dependent upon which endpoint was being evaluated.

Issue 4a

Dr. Mel Andersen was lead discussant for the Issue 4a group. The members of this group
included Dr. Andersen, Dr. Hayton, and Dr. Kamrin. Dr. Andersen began by stating he wished
to address question 7 before addressing question 6. Dr. Andersen provided a summary of his
group's response, as follows:

Issue 4a Question 7: Please comment on the use of the AUC as a measure of internal dose for
rats and humans for calculation of the MOE.

Dr. Andersen continued to respond to charge question 7. He stated that all three
reviewers provided a guarded endorsement of the use of AUC as a better measure of dose than
provided by a daily intake. Each of the three also discussed other measures of dose related to
pharmacodynamics that might also be considered superior to an average AUC (or average daily
concentration). There may not be sufficient information to discern the appropriate dose measure;
however, some discussion of the intentions in using the AUC and in potential alternative dose
measures appears essential. In addition, it would be extremely useful to attempt in the document
to compare, to the extent possible, the MOEs derived from alternate dose measures with the
currently estimated MOEs. For instance, if maximum daily concentrations were used, how
would the MOE compare to the ones in the present document in at least a qualitative manner.

Issue 4a Question 6: Please comment on the use of the one compartment pharmacokinetic
model.

The one compartment model used to calculate AUC for rats based on daily intake of
PFOA in rats provided an adequate approximation for the purposes of the calculations of AUC
in the rat. The Palazzo 13-week study could be used directly to estimate the rat steady-state
AUC. The underlying assumptions in using this PK modeling approach should be clarified in an
appendix to the document. All three reviewers noted areas where multi-compartment PK models
with more mechanistic detail might be developed to provide an improved analysis of the rat and
the monkey pharmacokinetic data and a better understanding of the biological processes that
control the kinetics of PFOA in test animals and humans.

8


-------
Other panel members appeared to agree with the Issue 4a group and no further discussion
took place.

Issue 4b

Question 8 - Please comment on the need to use or modify the default value of 10 for cross
species extrapolation given the pharmacokinetic analysis.

Dr. Mink was the lead discussant for the Issue 4b group. The group members included
Drs. Mink, Corcoran, and Roberts (Drs Melnick and Kamrin had also provided some input to
this group's response. Dr. Mink summarized the group's response, as follows:

He stated that the use of an uncertainty/safety factor to account for pharmacodynamic
equivalence was generally accepted (with a caveat by Dr. Kamrin based on his assessment of the
data's statistical robustness). He continued that a modification could be made given the
complexity of the data sets- with an uncertainty factor ranging of 3 to 200 proposed. Some
questioned the use of LOAEL/NOAEL methodology for systemic effects versus using a Bench
Mark Dose approach. Issues of newer data, steady state serum equilibrium assumptions, life
stage, and cumulative effects of the C-8 class were also mentioned in regard to the overall
complexity of the dynamics analysis.

The subsequent discussions focused on the complexity of the PFOA pharmacokinetic
data and the uncertainty that exists. Dr. Andersen explained that a 10X factor is usually added to
account for interspecies (3X) and intra-species (3X) difference. Although the use of internal
dose metrics may eliminate the need for the interspecies uncertainty factor of 3X, panelists were
generally not willing to remove or reduce it due to the obvious gender differences seen in the rat
studies.

Issue 4c

Question 9 Please comment on the adequacy of the human exposure data for use in calculating
aMOE.

Dr. Anne Sweeney was the lead discussant for the issue 4c group. The members of this
group were Dr. Sweeney and Dr. Ozonoff; both had submitted preliminary comments and Dr.
Sweeney had submitted integrated comments prior to the meeting date. Dr. Sweeney was,
however, unable to attend this meeting and Dr. Cory-Slechta asked Dr Ozonoff to summarize the
group's response. Dr. Cory Slechta also asked Dr. Longnecker to provide assistance during the
meeting to Dr. Ozonoff as he revised the group's response.

In his summary, Dr. Ozonoff questioned the appropriateness of using a MOE approach.
He also questioned the exclusion of some epidemiological data, including the worker
biomonitoring data. He continued to explain that the utilization of the correct summary
measures for biomonitoring data is important, i.e., using the 90th, 95th percentile or maximum

9


-------
figures for any MOE calculation is the preferred method. He also stated that the data were
poorly depicted.

Dr. Cory Slechta reviewed the next steps. She told panel members that they will be
revising their responses to the charge questions during the subsequent day's meeting. She
explained that panel members will be asked to be present at 8:30 AM when we will reconvene
the meeting and discuss the format of the meeting.

The meeting adjourned at 5:45PM.

Wednesday February 23, 2005

Dr Shallal opened the meeting at 8:40 AM when all panel members were present. She
made some brief opening remarks and then turned the agenda over to Dr. Cory Slechta.

Dr. Cory-Slechta informed the panel members to get into their issue groups and attempt
to find consensus. She said they should also determine the most important issues which they
would like to include in the executive summary and letter to the Administrator.

Dr. Shallal directed the panelists to their meeting locations. The panel separated into their
writing groups according to their assigned issue group at 8:50 AM.

The meeting reconvened at 10:00 AM

Issue 1

Dr. Drinkwater presented the findings for the Issue 1 group (Attachment O). The group
suggested that a closer look at the relevance to infants and children should be addressed. They
also suggested a closer look at the immunotoxic effects seen in the Yang et al study.

Issue 2

Dr. Melnick presented the response for the Issue 2 group (Attachment P). He discussed
the carcinogenicity evidence that is available. He also summarized the difference between the
cancer descriptors, suggestive vs. likely, as presented in the 1999 and 2003 cancer guidelines.

A discussion regarding the cancer classification continued and an alternative response
was crafted by Dr Mink which tried to capture the variety of points of view (Attachment Q). It
was agreed that this new language would be incorporated into the Issue 2 response.

Issue 3

Dr. Roberts presented the revised response for the Issue 3 group (Attachment R). Further
discussion focused on the immunotoxicology and pituitary endpoints. There was also a

10


-------
suggestion that the limitations of using a particular endpoint for calculating an MOE be clearly
articulated. It was agreed that additional information regarding the epidemiology data will be
provided by the Issue 4c group.

Issue 4a

Dr. Andersen presented the revised integrated comments for the issue 4a group
(Attachment Y). A discussion on using the average serum concentration for calculating AUC
concluded with an agreement that a statement regarding its appropriateness or that of another
measure would be added to the final response.

Issue 4b

Dr. Mink discussed the revised response for the Issue 4b group (Attachment S). He
stated that there needs to be a discussion of what is currently done and how uncertainty is
dealt with. There continued to discussion regarding the use of uncertainty factor when a
pharmacokinetic model is being used. Some questioned the utility of developing such
models if they will not impact the use of uncertainty factors. All panel members agreed
that there was not enough data available to be able to modify the default value of 10 for
cross species extrapolation given the pharmacokinetic analysis.

Issue 4c

Dr. Ozonoff presented the revised response (Attachment T) for the Issue 4c group with
input from Dr. Longnecker. He stated that this data set is as good as others. Depiction of
biomonitoring data needs to be enhanced. More information regarding children's exposure
should also be included.

After a short lunch break, the Panel next began discussing the major points that they
wanted to include in the Executive Summary and letter to the Administrator.

The following points were discussed:

Carcinogenic potential
Handling uncertainty factors
Additional evaluation of non-cancer endpoints
Improve understanding of sources of exposure
Improve understanding of kinetics
Endorse use of internal dose metrics

Overall assessment is reasonably comprehensive and well-written

The Panel agreed that these are the important issues that should be included in the
executive summary.

The panel took a short break and returned to continue their discussion. They now
focused on the key data gaps.

PFOA distribution

11


-------
PFOA speciation in blood and how this is related to the effects seen
Hepatocyte proliferation in the monkey-role of PPAR alpha
Kinetics
Cumulative risk

Since more refinement of the executive summary was needed, it was decided that Dr.
Cory-Slechta would draft the executive summary and panelists would be given an opportunity to
provide comments.

Dr. Cory-Slechta then asked the panel members to revised their comments and submit
them to Dr. Shallal by March 4, 2005. She will then provide them to Dr. Cory-Slechta so that
she made draft the review report. The panel will likely meet via teleconference in order to
finalize their draft review report. The exact time will be determined at a later date.

Finally, as a reminder, Dr. Shallal instructed all panel members to conduct all their
discussion and deliberations regarding the PFOA risk assessment in the public domain. All
correspondence between panel members in reference to this review should also include Dr.
Shallal as a recipient.

Meeting Adjournment - Dr. Shallal stated that she would send an e-mail to all panel
members reminding them of the various deliverables and their due dates as agreed. The meeting
was adjourned at 2:25 pm.

Respectfully Submitted:

Dr. Suhair Shallal
Designated Federal Officer,
EPA SAB PFOA Review Panel

I certify that these minutes are accurate to the best of my knowledge:

Dr. Deborah Cory Slechta
Chair,

EPA SAB PFOA Review Panel

12


-------