DEQ-CFW_00001396

Perfluorooctanoic Acid (PFOA or C8) and Ammonium Perfluorooctanoate (APFO) Prepare_ d by Dr. Luanne K.
Williams, February 26, 2007 For information purposes only for Luanne
DWQ's December 7, 2006 Interim Maximum Allowable Concentration (IMAC) of 2 ug/L not Scientifically Valid
Against recommendations (see September 27 email) from Luanne and Bill, the NC Division of Water Quality established
an interim maximum allowable concentration or IMAC of 2.0 ug/L for PFOA or C8 on December 7, 2006. This is based on
a Lowest Observed Adverse Effect Level of 1 mg/kg-day seen in rats from chronic exposure where decreased body
weight was observed and safety factor of 3000 was applied (10 for pharmacokinetic and pharmacodynamic differences
between animals and humans, 10 for differences among humans, 10 for Lowest Observed Adverse Effect Level to No
Observed Adverse Effect Level, and 3 for database gaps) to derive a 0.0003 mg/kg-day reference dose. The IMAC will
be used like a groundwater quality standard in setting permitting limits for the Fayetteville DuPont plant's compliance
boundary and will also significantly reduce the amount of groundwater monitoring for DuPont. Also, if PFOA is found in
groundwater at other sites in NC, then this IMAC of 2 ug/L would guide remediation efforts or cleanup. The safety factor
of 3 used by DWQ to account for pharmacokinetic differences between rats and humans underestimates the known
pharmacokinetic differences between rats and humans. For instance, according to EPA's January 2005 Draft Risk
Assessment of the Potential Human Health Effects Associated with Exposure to Perfluorooctanoic acid on pages 6 and 7,
the half-life or time that it takes to eliminate 50% of PFOA in adult female rats is 2.8 to 16 hours, for adult male rats is 138
to 202 hours, and for humans 38,281 hours (or 4.37 years). The pharmacokinetic differences between rats and humans
ranges from 189 (38,281 hours/202 hours) to 13,671 (38,281 hours/2.8 hours) or the half-life in humans is 189 to 13,671
times longer than in rats. The actual pharmacokinetic differences of 189 to 13,671 are much higher than what was
accounted in the calculation of 3. Since humans retain PFOA longer than rats, the blood level from a given dose would be
expected to be higher for humans compared to rats. Using a safety factor of only three, seriously underestimates the
equivalent human dose. Because of these differences, the Federal Science Advisory Board in May 2006
(http://www.epa.gov/sab/ndf/sab 06 006.1)df on page 11) stated "compartmental modeling of serum concentrations provides
a sound approach for estimating internal dosimetry without exceeding the limits of the available data ..... PBPK modeling is
perhaps the ideal approach for addressing PFOA for purposes of cross -species extrapolation." In other words, the SAB
recommends the use of pharmacologically -based pharmacokinetic modeling to correlate the animal internal dose/animal
blood effect level to estimated human internal dose/human blood effect level given the differences in the pharmacokinetics
between rats and humans. The SAB further states on page 23 "the emphasis is on having data based on the internal dose
relationships (i.e., serum PFOA levels) in some of the animal studies so that interspecies differences in metabolism and
clearance are taken into account." Instead of using PBPK modeling as recommended by the Federal SAB and animal
internal dose/animal blood effect levels, the Division of Water Quality used animal administered dose and safety factors to
derive the chronic oral reference dose of 0.0003 mg/kg-day. In addition to the federal SAB, the EPA toxicologist that
presented at the SAB meeting on October 19, 2006 as well as by the DuPont's toxicologist that also presented to the SAB
meeting on September 28, 2006 stated that the administered animal effect doses should not be used to estimate human
effect doses or health -protective doses. Instead, modeling should be done to correlative animal internal dose/blood levels
to human internal dose/blood levels. The EPA toxicologist further stated that because of the differences in the half-lives
between rats and humans, PFOA is the poster child for not using an animal effect dose for calculating a health -protective
dose for humans (see his email to me dated October 20, 2006). The approach used by the Division of Water Quality to
derive the IMAC of 2 ug/L is not recommended by the federal SAB, both the EPA and DuPont toxicologists that presented
at recent NC SAB meetings and myself;/This approach is not scientifically valid and may not be health -protective.
New Jersev Department of Environmental Protection Develops Recommended Health -based PFOA Dose
In 2007 New Jersey Dep03 artment of 6 ronmental Protection calculated target human blood levels (ug/L) based on one in
a million cancer risk of 5.7 ug/L (calculated from 10% testicular, liver, and pancreatic cancer incident rat dose of 13.6
mg/kg-day) and recommended noncancer human blood levels of 18, 42, 26, 35, 92, and 130 ug/L. The recommended
noncancer human blood levels were calculated by dividing rat and primate NOAEL blood levels by 100 to 3000 fold
uncertainty factors and then models were used to estimate equivalent target human blood levels. The noncancer health -
based drinking water concentrations were calculated by dividing the target human blood levels by 100 which assumes the
blood levels would be 100 times the drinking water level based on studies in the literature and multiplying by a20%
relative source contribution which corresponded to 0.04, 0.08, 0.05, 0.07, 0.18, 0.26 (all based on noncancer) and 0.06
ug/L (based on one in a million cancer risk only, no 20% relative source contribution used). New Jersey chose 0.04 ug/L
as the recommended health -based guidance level. This is protective against noncancer and cancer effects (one in a
million). Using 0.04 uq/L, one can back calculate a reference dose as follows: 0.04 ua/L = x ua/ka-day x 70 ka x 1 efav/21
x 0.20 relative source contribution x = 0.006 ug/kg-day = 0.000006 mg/kg-day which is 50 times lower than the DWQ
reference dose of 0.0003 mg/kg-day.
1 oa d lCL9_e1L)4 012 .
lod
bl�cs2 In velCs C" sue, aS l � -�7 13 c� µ3 l L,3
aoo,-, I 6too
DEQ-CFW 00001396

Minnesota Department of Health Develop Recommended Health -based PFOA Doses
In 2006 Minnesota changed their health -based drinking water value from 7 ug/L to 1 ug/L . Currently, Minnesota is
reviewing data to determine if the health -based value should be lower or 0.5 ppb or less based on reference doses of
0.00006, 0.00007 mg/kg-day, and 0.00014 mg/kg-day (which are two to five times lower than the reference dose used
by DWQ of 0.0003 mg/kg-day). They used a BMDL10 based on liver effects for rats of 0.4 and 0.6 mg/kg-day and divided
by a factor for different kinetics of 200 and additional safety factor of 30 to get 0.00006 and 0.00007 mg/kg-day and for
monkeys 3 mg/kg-day divided by a factor for different kinetics of 70 and additional safety factor of 300 to get 0.00014
mg/kg-day.
CIIT Develops Model to Correlate Monkey PFOA Levels to Human Levels and Derives a Reference Dose
CIIT at RTP used a pharmacokinetic model to derive a reference dose of 0.00009 mg/kg-day based on liver toxicity
LOAEL serum level of 23,000 ug/L in monkeys, a safety factor of 30 (3 animal to human and 10 for human variability), and
a target human blood level of 770 ug/L. Using criteria specified in 15A NCAC 2L for calculating groundwater quality
standards, a recommended groundwater quality standard can be calculated using this dose as follows: 0.00009 mg/kg-
day x 70 kg x 1 day/2L x 0.20 relative source contribution = 0.00063 mg/L = 0.6 ug/L compared to the 2 ug/L level
calculated by DWQ.
EPA and West Virginia Establish A Drinking Water Action Level or Level of Concern for PFOA at 0.5 uq/L on
November 20, 2006
On November 20, 2006 EPA established an action level of 0.5 ppb for the DuPont facility in West Virginia under a consent
order. This is based on a Lowest Observed Adverse Effect Level of 3 mg/kg-day in monkeys from subchronic exposure
where liver disease, gastrointestinal distress, and decreased body weight were observed anct safety factor of 13,500 was
applied (135 for pharmacokinetic and pharmacodynamic differences between animals and humans, 10 for differences
7among humans, 10 for Lowest Observed Adverse Effect Level to No Observed Adverse Effect Level) to derive a 0.00022
mg/kg-day reference dose (which is 1.36 times lower than the reference dose used by DWQ of 0.0003 mg/kg-day). A
child body weight of 10 kg, ingestion rate of 0.85 liters/day and relative source contribution of 20% was applied to the
0.00022 mg/kg-day to derive an action level of 0.5 ppb. With this order, DuPont will be required to offer alternative
drinking water or treatment for public or private water users living near the Washington Works plan if the level of C-8
detected in drinking water is equal to or greater than 0.5 ug/L. Dr. Chris Weis, an EPA toxicologist was responsible for
developing this level. It is based on 6 month subchronic study in monkeys where the critical effects were liver effects.
There was no No Observed Effect Level in this study, and the low dose (3 mg/kg/day) is a Lowest Observed Effect Level.
This assessment uses a body weight and drinking water ingestion rate of a child not an adult and was based on
subchronic exposure not chronic (2006 USEPA).
German Ministry of Health establishes Health -protective Values and Precautionary action values for PFOA and
PFOS July 13, 2006
The German Ministry of Health established health -protective values based on a tolerable daily intake of 0.1 ug/kg-day or
0.0001 mg/kg=day for both PFOA and PFOS (which is three times lower than the reference dose used by D'v"vQ of
0.0003 mg/kg-day). The minimal health -based total level for PFOA and PFOS for long-term exposure is 0.1 ug/L.
The maximum health -based total level for PFOA and PFOS for long-term exposure is 0.3 ug/L. The precautionary
action value or total level for PFOA and PFOS to protect infants is 0.5 ug/L (where drinking water with a total PFOA
and PFOS of greater than 0.5 ug/L should not be used for baby food). Precautionary action level for adults or the
level that would require immediate action to reduce intake for adults is a total PFOA and PFOS level of 5 ug/L.
For comparison, DWQ just pastffiecember 2006) their interim maximum allowable concentration level of 2`ug/L for PFOA
based on a dose of 0.0003 mg/kg-day. i
Uses of Ammonium perfluorooctanoate (APFO) and PFOA
APFO is the derivative of greatest concern and has the most wide spread use. APFO quickly dissociates in drinking
water to PFOA. APFO is used as a processing aid in the production of fluoropolymers which are used to promote fire
resistance as well as oil, stain, grease, water repellency and in non-stick cookware surfaces. APFO is also used to
manufacture surfactants in fire fighting foams, personal care and cleaning products, and soil, stain, grease, and water
repellent coatings on carpet, textiles, leather, and paper (US EPA 2005a)
Differences in Pharmacokinetics Between Rats and Humans
Half-life or time that it takes to eliminate 50% of PFOA in adult female rats is 2.8 to 16 hours, for adult male rats is 138 to
202 hours, and for humans 38,281 hours (or 4.37 years). The pharmacokinetic differences between rats and humans
ranges from 189 (38,281 hours/202 hours) to 13,671 (38,281 hours/2.8 hours). Since humans retain PFOA longer than
rats, the blood level from a given dose would be expected to be higher for humans compared to rats. The urine is the
major route of excretion. PFOA is absorbed following oral, inhalation and dermal exposure. Based on animal studies,
PFOA readily crosses the placenta and is present in breast milk (US EPA 2005b). Dr. Robert Rickard, a toxicologist for
DEQ-CFW 00001397

DuPont, stated at the September 28, 2006 North Carolina Scientific Advisory Board meeting that drinking water containing
1 ug/L could correlate to a blood level of 100 ug/L. Dr. Hugh Barton, an EPA toxicologist, stated at the October 19, 2006
North Carolina Scientific Advisory Board meeting that the 90th percentile blood levels in adults and children are 9.4 ug/L
and 8.5 ug/L, respectively. On December 7, the North Carolina Division of Water Quality approved an interim maximum
allowable concentration of 2 ug/L which could correlate to a blood level of 200 ug/L which would be significantly higher
than the 90th percentile blood levels for adults and children in the United States.
Toxicity
PFOA has shown moderate to severe toxicity in short term studies involving primates. Exposure to monkeys for six
months showed liver toxicity in all dose groups tested. In a 13 week monkey study, toxicity including gastrointestinal
distress, reduction in body weight was observed at doses of 3 to 30 mg/kg-day and mortality at 30 mg/kg-day. PFOA
appears to be immunotoxic (effects on thymus and spleen) in animals with decreased function of the bone marrow and
spleen. PFOA is associated with developmental effects including skeletal effects following prenatal exposure to rabbits
and decreased body weight following prenatal exposure to rats. PFOA is carcinogenic in rodents by multiple mechanisms
and in multiple organ systems. Based on no adequate human studies and uncertain relevance of the tumors from rat
studies for PFOA, there is suggestive evidence of carcinogenicity according to EPA. Epidemiological studies on the
effects of PFOA in humans have been conducted on workers. A retrospective cohort mortality study demonstrated a
statistically significant association between prostate cancer mortality and employment duration in the chemical facility of a
plant that manufactures PFOA. However, in an update to this study in which more specific exposure measures were
used, a significant association for prostate cancer was not observed. Other mortality studies lacked adequate exposure
data which could be linked to health outcomes. Cholesterol and triglyceride levels in workers were positively associated
with PFOA exposures, which is inconsistent with the hypolipidemic effects observed in rat studies. A statistically
significant positive association was reported for PFOA and T3 thyroid hormone levels in workers but not for any other
thyroid hormones (US EPA 2005b; US EPA 2006).
REFERENCES
US EPA 2005a. Perfluorooctanoic Acid (PFOA) and Fluorinated Telomers Basic Information on PFOA website
http://www.epa.gov/opptintr/pfoa/pfoainfo.htm
US EPA 2005b. Draft Risk Assessment of the Potential Human Health Effects Associated with Exposure to
Perfluorooctanoic Acid and Its Salts. US Environmental Protection Agency Office of Pollution Prevention and Toxics Risk
Assessment Division, January 4, 2005. http://www.ewq.org/issues_content/PFCs/20050112/pfoarisk.pdf
2006 USEPA. Memorandum from Christopher Weis, Senior Toxicologist with US EPA Office of Criminal Enforcement,
Forensics and Training to Walker Smith, Office of Civil Enforcement, November 17, 2006.
January 2005 Draft Risk Assessment of the Potential Human Health Effects Associated with Exposure to
Perfluorooctanoic acid.
Presentations made by Dr. Robert Rickard with DuPont and Dr. Hugh Barton with US EPA to North Carolina Scientific
Advisory Board on September 28 and October 19, 2006.
Presentation made by Dr. Harvey Clewell with CIIT at the February 22, 2007 NC SAB meeting.
2007 New Jersey Department of Environmental Protection Guidance for PFOA in Drinking Water found at
http://www.ni.gov/dep/watersupr)l /pfoa dwquidance.pdf.
2006 Minnesota Department of Health Health Risk Limit (HRL) Rule Revision Staff Worksheet, November 15, 2006.
2006 German Ministry of Health Provisional Evaluation of PFT in Drinking Water with the Guide Substances with PFOA
and PFOS as Examples ttp://www.umweltbundesamt.de/uba-info-presse-e/hintergrund/pft-in-drinking-water.pdf.
DEQ-CFW 00001398