DEQ-CFW_00000042

ONE HUNDRED -AND -FORTIETH MEETING OF THE
SCIENCE ADVISORY BOARD (NCSAB) ON TOXIC AIR POLLUTANTS
Proceedings of the March 25, 2009 Teleconference
Dr. Starr called the meeting to order at 2:11 PM. NCSAB Members Drs. Thomas Starr, Woodhall Stopford,
Wayne Spoo, James Gibson and Man -Sung Yim were in attendance. Dr. Robert Tardiff, The Sapphire
Group, Mike Johnson, DuPont, and Sandra Moore, DWQ were also in attendance. Dr. Elaina Kenyon
was absent.
Presentation: PFOA Risk Assessment
Robert G. Tardiff, Ph.D., A.T.S.
The Sapphire Group
3 Bethesda Metro Center
Suite 830
Bethesda, MD 20814
Dr. Tardiff, in his presentation:
• Reviewed the current literature on PFOA
• Described in the methodology for WOE
• Discussed sensitive organs
• Revealed ranges for non -cancer and cancer drinking water equivalent levels for PFOA
Dr. Tardiff began his presentation by indicating that he and Dr. Mel Andersen of the Hamner Institutes
had been asked by DuPont to examine, as comprehensively as possible, all of the health and safety
information related to PFOA in terms of human health consequences. They were also tasked to define
safe levels of exposure, to the extent possible, using a framework that would be fairly understandable to
the regulators. This culminated in a manuscript that was submitted to Food and Chemical Toxicology in
March, 2009.
The research included current epidemiology, developmental toxicology, and experimental animal studies.
Dr. Tardiff did not discuss MOA, deferring instead to the presentation by Dr. Rickard in the February 25th
2009 NCSAB meeting.
The animal studies were integrated with human studies to develop conclusions about the WOE
associated with subchronic toxicity, chronic toxicity, immunotoxicity, and genotoxicity - all of which were
used in the evaluation of drinking water -equivalent levels. The methodology employed followed that used
by EPA, starting with baseline information, developing a reference dose and then developing a value
applicable to drinking water, as a proportion of exposure from multiple sources, some of which are not
regulated by EPA. The value that is determined becomes the precursor to the MCLG (maximum
contaminant level goal). This level, by law, is supposed to be a value that is exclusively restricted to safe
levels of exposure to chemicals in water. The MCLG is achieved by considering treatment, economics,
and so forth. Their work stopped short of the last step because the purpose of this study was not of risk
management but risk assessment.
Dr. Tardiff stated that some of the MOAs for PFOA in rodents were only of limited value in predicting
toxicity in humans because the studies were conducted using high doses that either exceeded a
maximum tolerated dose or were so dominated by MOAs, like PPAR-a, that they were of limited or of no
value in terms of predicting an outcome. It was found, however, that monkey studies, particularly
involving the Cynomolgus monkey, provided sufficient evidence with regards to effects on liver toxicity.
Liver hypertrophy was the dominant effect. Liver/brain weight ratios and liver/body weight ratios were
investigated as well. After looking closely at developmental toxicity, thinking that it was going to be the
most sensitive outcome, it was determined that other non -cancer toxicities might be more relevant to
humans, particularly liver toxicity.
The data indicated that when adverse effects were found in animals there were very few concordant
findings in occupationally exposed humans. There were a few exceptions - such as changes in
DEQ-CFW 00000042

were used across all endpoints? Dr. Tardiff replied that there is a detailed narrative in the Discussion
section of the submitted manuscript on the rationale and analysis of each outcome and associated UFs.
Dr. Starr continued by saying that at the BMC10 there would be a 10% incidence of testicular tumors in the
rats predicted by the dose -response modeling. If the internal dose was decreased by a factor of 30 what
risk is calculated using the model that was fit to the higher dose data? Would it be a 30-fold reduction in
risk? Dr. Tardiff said yes, noting that the probability is well above the background rate of testicular tumors
in the US. If PFOA were to cause this effect, it should be detectable, because the incidence in humans in
this country is between 1 in 10,000 and 1 in 50,000. So if the probability is 1 in 3,000 and there is a large
population, in WV or MN for instance, then the epidemiology studies would indicate this. Dr. Starr asked
if a power calculation was performed and Dr. Tardiff responded that the epidemiological investigators had
performed these calculations, but not all the data were from published literature, some were from industry
reports.
Dr. Starr asked Mr. Mike Johnson (DuPont) if Dr. Robert Rickard (DuPont) had investigated any studies
that had performed power calculations. Dr. Starr continued by saying that epidemiological studies of
highly exposed workers should have had a sample size large enough to see the effect that is being
predicted by the rat testicular tumor model. Therefore, there seems to be a qualitative disconnect
between the human evidence in absence of risk and the prediction from the animal data that the risk
should have been three times that of the background if it was there. So the argument that could be made
is that there is a discrepancy between what is not observed in humans and what is being predicted based
on the rat data.
Dr. Starr asked if the mammary tumors were dismissed based on the report from the pathology working
group and if the report was available? Dr. Tardiff said yes. Dr. Starr also asked about the unpublished
Pathology Working Group (PWG) report regarding the pancreatic acinar tumors. Dr. Tardiff responded
that all of the reports should be available as they referenced everything.
J,Lo7.1
Dr. Starr asked why the PPAR-a MOA was not considered dominant in this study. Dr. Tardiff answered
that these were rat studies, and there was not enough data to support the PPAR-a MCA. If mouse
studies had been used in the analysis, the PPAR-a MOA would have been the driving factor. In the
mouse PPAR-a seems to be such a dominant factor that at least 90% of the MOA for the liver and kidney
are related to PPAR-a. That is not the case in the rat. In the rat, PPAR-a does not seem to be a dominant
factor. Also, the tumors observed in the rats are not liver tumors; if there had been liver tumors an
uncertainty factor related to PPAR-a would have been included.
Dr. Starr asked if any other MOAs had been identified. Dr. Tardiff responded that there were none for the
testes. All the other MCA information relates largely to the liver and somewhat less to the immune
system and about the same amount for developmental toxicity. In the developmental toxicity studies, the
NOAELs and LOAELs are low. In one study reporting delays in ossification, there is no NOAEL, just a
LOAEL. In this study, one group was allowed to survive, and the developmental delays observed were
reversed in a few weeks. Dr. Milly Christian, consultant, indicated that humans undergo a similar reversal
of developmental delays without any consequent clinical manifestations later in life. Therefore,
developmental toxicity described in the manuscript is a mild transient reversible consequence for those
persons concerned with this outcome.
NC Division of Water Quality (DWQ) Interim Level
The NC DWQ has set the Interim Maximum Allowable Concentration (IMAC) for PFOA at 2 ppb, Sandra
Moore (DWQ) stated. The IMAC was calculated using the York (2002) and Butenoff (2004) studies. She
stated that an RfD of 0.0003 mg/kg-day was derived and used in the EPA drinking water standard formula
together with a relative source contribution of 20%, 70 kg body weight, and 2L drinking water per day per
person consumption.3 Dr. Starr asked Dr. Tardiff if he thought this IMAC would be protective of all
adverse health effects associated with PFOA exposure to humans. Dr. Tardiff responded that it would be
protective, even for cancer. Dr. Starr asked if water consumption levels had changed in the EPA
Exposure Handbook since the body weight value had been changed. Dr. Tardiff said that it had not
changed, it was still 2L/day for adult, and 1 L/day for 10 kg child.
DEQ-CFW 00000043

Agenda for the Next NCSAB Meeting
The NCSAB will review the Tardiff et al. paper when it becomes available. The NCSAB will also review a
summary of studies by Butenhoff et. a/. and EPA focusing on the main points, the final standards
suggested, and the differences between the studies.
The next meeting of the NCSAB will be held at 2:OO13M on WEDNESDAY, May 27", 2009 by
teleconference. The call -in number is [phone].
The meeting was adjourned at 3:19PM.
Respectfully submitted,
Reginald C. Jordan, Ph.D., CIH
Liaison, Science Advisory Board
These minutes were accepted as written at the 141sr SAB meeting, June 24, 2009.
Biegel, L.B, Hurtt, M.E., Frame, S.R., O'Connor, J.C., and Cook, J.C. (2001). Mechanisms of
extrahepatic tumor induction by peroxisome proliferators in male CD rats. Toxicol. Sci. 60:
44-55.
2 Sibinski, L.J. (1998). Final Report of a two year oral (diet) toxicity and carcinogenicity study of
fluorochemical FC-143 (perfluorooctanane ammonium carboxylate) in rats. Vol. 1-4, 3M Company/RIKER
Exp. No. 0281 CR0012; 8EHQ-1087-0394, October 16, 1987.
3 http://h2o.enr.state.nc.us/csu/documents/IMACBasisC8.pdf
DEQ-CFW 00000044