A meeting took place at EPA offices on August 16, 2007 between EPA staff
and members of the Perchlorate Study Group (PSG) relating to PSG
comments on perchlorate in the May 1, 2007 Regulatory Determination
Federal Register Notice.

Present at the meeting from EPA was:  

Pam Barr, Phil Oshida, Ann Johnson, Bruce Schillo, Eric Burneson

Present from PSG was: 

Jonathan Whitehill, Troy Hiller, James Rollins

Jonathan Koplos and Susan Bjork from The Cadmus Group were present via
phone and took notes.

PSG has been representing companies with an interest in perchlorate
since 2002.  These companies include:  American Pacific Corporation;
Tronox (formerly Kerr McGee); ATK Tech; and Aerojet.  PSG believes that
their work with these companies has given them access to very recent,
sound science.  This science has led them to the conclusion that the
neonate and fetus are true sensitive subpopulations affected by exposure
to perchlorate.  PBPK (Physiologically Based Pharmacokinetic) modeling
has shown perchlorate is not stored in the body and its behavior is
predictable.  Human and rat data are available to support this
conclusion.  

A report was published in March regarding children showing that they are
exposed to more perchlorate (ingest more perchlorate) than any other
population group.  Food data (90th percentile values) show that children
receive three times the perchlorate dose that an adult would, but
children have lower blood serum levels of perchlorate than do adults,
perhaps due to higher metabolic rates (see Tables 4 and 5 of the March
report).  The low blood serum levels indicate that children are the
least sensitive sub-population.  Regarding the effect on the body,
children have only a 0.3% inhibition of iodine uptake, which was three
or four times lower than the fetus and neonate.  The study showed that
90% of perchlorate exposure comes from food.  PSG concluded that the
current RfD for perchlorate was sufficiently protective of children. 
The unknown is the effect on the neonate and fetus; however, if 90% of
exposure is from food then the neonate would be protected because they
don’t eat solid food, and the fetus would be protected in utero
(higher perchlorate doses are mediated through the mother’s breast
milk).

PSG cited a study by Blount, and some work done by TERA to extrapolate
what the effects would be on an entire population.  The extrapolation
showed that it would take a dose of 1,000-5,000 ug/L to push a woman
with low iodine levels to borderline hypothyroidism.  PSG reiterated
that the RfD is conservative enough to be protective.  (PSG apparently
left a paper with EPA that shows how TERA did these extrapolations.)  

Tonacherra’s lab work has established the relative iodine uptake
inhibition  affects of nitrate and thiocyanate.  Several studies have
been done on pregnant women in Northern European countries where the
population typically has low iodine levels.  The study was done on
thiocyanate in blood (from smoking), which was used as a surrogate for
perchlorate (based on the iodine uptake 

inhibition/IUI potential, serum equivalent of perchlorate).  Results
showed a dose of 2,000-3,000 ug/L would be needed to see any changes
toward hypothyroidism.  PSG again stated they believe the RfD for
perchlorate is protective.

The discussion then turned to occurrence of perchlorate in ambient and
drinking water.  PSG stated that occurrence of perchlorate has gone down
in California and cited Metropolitan Water District (MWD) monthly
sampling results showing mean perchlorate levels in a MWD intake from
the Colorado River had decreased from 6.2 ug/L to non-detections from
approximately 2000 to 2006.  EPA asked why this was, and PSG replied
that there has been considerable progress in the clean-up of the source
of the perchlorate in the Colorado River.  

PSG recommended that EPA not estimate the population potentially exposed
to perchlorate based on entire water systems’ populations-served (if
that system has a perchlorate detection), but rather base it on the
percent of entry points at a system with perchlorate detections and
apply that fraction to the system’s entire service population to
determine potential exposure.  EPA replied that they had in fact done
that.  PSG also said that water systems and entry points with a single
perchlorate detection might overestimate occurrence.  More than one
detection should be considered.  EPA has made similar decisions in the
past on organics. 

PSG commended EPA’s use of biomonitoring to determine exposure given
the problems and challenges of traditional estimations of food and water
exposure.  They said based on information in the NAS biomonitoring
handbook, perchlorate is the perfect candidate for using the
biomonitoring approach and is the best possible option available.  They
believe that there are enough data out there that models do not need to
be used, and that actual data can be collected that would support an MCL
in the high teens or low twenties.  They estimate that all but 0.002% 
of the population is exposed at levels below the RfD with all food and
water exposure.

PSG does not think that it is worth EPA spending the money to write a
rule since there are apparently so few at risk from perchlorate exposure
(very few would benefit).  PSG urged EPA to make their decision swiftly,
and offered to help in any way they could.

NOTE: PSG submitted talking points to EPA for use during this meeting on
August 16, 2007 

PSG and EPA Meeting – August 16, 2007

Page   PAGE  1  of   NUMPAGES  2 

