science in ACTION
BUILDING A SCIENTIFIC FOUNDATION FOR SOUND ENVIRONMENTAL DECISIONS
oEPA
r
www.epa.gov
Concentration
I
Steady State Blood
Concentration (Css):
Upper 95^ Percentile Among — »
Healthy Individuals of Both
Sexes from 20 to 50 Yrs Old
Figure 1.
world exposures. This study
focuses on discovering what
level of human exposure is
required to result in the internal
concentrations that caused
effects in in vitro tests.
To provide insights into this
question, this study made
experimental measurements and
calculated relevant human
exposures for 239 of the 309
ToxCast Phase I chemicals. This
study indicates that
understanding relevant exposure
conditions is important when
using HTS in vitro data to
prioritize chemicals for further
testing and risk management.
Study Description
This study used a combination
of ToxCast data, public
exposure data, and new
Oral Doses Required to
Achieve the Upper 95m
Percentile Steady State Blood
Concentrations Across All In
Vitro Assays
(Represented as a Box Plot)
Reverse Dosimetry
Combining ToxCast™, Dosimetry and Human Exposure Research to Increase
the Relevance of Rapid Chemical Toxicity Testing Results
October 31, 2011
Impact Statement
Tens of thousands of chemicals
are currently in commerce, and
hundreds more are introduced
every year. Because there are so
many chemicals—and since
traditional chemical toxicity
tests using animals are
expensive and time
consuming—only a small
fraction of chemicals have been
fully assessed for potential risk.
In 2007, EPA scientists began
working on ToxCast, a research
project that identifies and
prioritizes potentially toxic
chemicals using rapid,
automated tests called high-
throughput screening (HTS)
assays. ToxCast is currently
assessing over 2,000 chemicals
from a broad range of sources,
including pesticides, industrial
and consumer products, food
additives, and failed drags that
were never released to the
market.
The technologies included in
ToxCast use non-animal tests
called in vitro assays to help
understand what might happen
when a human is exposed to a
chemical. However, it is
difficult to determine the
relevance of in vitro data when
predicting toxicity from real-
dosimetry data to identify
chemicals with the potential to
disturb cellular pathways at
relevant human exposure levels.
Below is a summary of how the
study combined and assessed the
data (See Figure 1):
•	239 chemicals taken from
ToxCast Phase I were tested
for possible toxicity using
over 500 HTS assays. The
amount of a chemical that
caused a 50% change in an
assay (Mm) was chosen as
an estimate of the blood
concentration needed for
bioactivity.
•	In order to relate chemical
exposure to blood
concentration, two
measurements were made
for each chemical - the

-------
ability of human plasma to
bind the chemical (plasma
protein binding) and the
ability of human liver cells
to metabolize the chemical
(hepatic clearance). These
two parameters were used
bioactivity at environmentally-
relevant concentrations. If
these chemicals had been
examined using only HTS
methods, they may have been
overlooked.
The pharmacokinetics
Fluroxypyi-meptyl
1000-
100

HftmiiliSH
c
~
rm
Compound
approaches in this study have
the potential to move beyond
simply identifying hazards,
toward the use of in vitro data
for realistic analysis of
enviromnental risks.
However, using molecular and
cellular in vitro assays for
predicting toxicity and adverse
outcomes in human biological
systems continues to be a
challenge requiring further
research. Activation of these in
vitro endpoints does not
necessarily represent an adverse
biological response, but should
be regarded as a measure of
potential biological changes
caused by a chemical.
Figure 2.
to convert in vitro
concentrations into
estimated real world
exposures. Chemicals can
then be prioritized from
most to least potent in terms
of relevant human
exposures (See Figure 2,
which displays the least
potent 25% of chemicals
tested). Finally, publically-
available exposure
estimates were compared
with the estimated
exposures necessary for in
vitro activity.
Conclusions
After examining data from the
HTS assays, 18 chemicals were
flagged as potentially causing
ToxCast analysis is underway to
develop predictive signatures
consisting of different in vitro
assays to predict adverse effects
in animal models and humans.
This analysis will help clarify
the difference between adverse
and adaptive responses.
Background
Most chemicals in commerce
have only undergone limited
safety testing, and conventional
toxicity testing is time-
consuming, complex, and
expensive. In an effort to
improve existing chemical
screening, U.S. EPA developed
ToxCast as a novel way to
efficiently screen chemicals and
prioritize limited testing
resources toward those that have
the potential to cause greatest
hazard to human health.
This research was performed by
the U.S. EPA National Center
for Computational Toxicology
in collaboration with The
Hamner Institutes for Health
Sciences (Durham, NC).
Reference
Wetmore, et al. "Integration of
Dosimetry, Exposure and High-
Throughput Screening Data in
Chemical Toxicity Assessment"
Toxicological Sciences (2011).
Accepted.
Contact:
Monica Linnenbrink
Office of Research &
Development
Chemical Safety for
Sustainability
(919)-541-1522
linnenbrink. monica Vvcpa. gov

-------