Geoscience Reference
In-Depth Information
substrates, products, cofactors, and other small molecules (metabolomics) can
all be measured. But which of those signals, if any, are quantitatively predictive
of the ultimate adverse response of interest is the key. Changes in the profiles
are dynamic, tissue-specific, and dose-dependent, so the results may be drasti-
cally different depending on the tissue that was examined, the time when the
sample was taken, and the dose or concentration that was used. Sophisticated
bioinformatic analyses will be required to make biologic sense out of such mas-
sive amounts of data. Tremendous advances have been made in this field in the
last 5 years, and it is now possible to coalesce such information into pathway
analyses that may have utility in toxicity assessment. Indeed, EPA's ToxCast
program has begun to examine approaches discussed above to predictive in vitro
toxicity assessment (Judson et al. 2011).
Example of Using Emerging Science to Address Regulatory Issues
and Support Decision-Making: ToxCast Program
In 2006, EPA began a new computational toxicology program aimed at
developing new approaches to assess and predict toxicity in vitro (Judson et al.
2011). Agency scientists in the computational toxicology program have been
substantial contributors to the development of new approaches to toxicity test-
ing. They have collectively published over 130 peer reviewed articles since its
inception, including 38 publications from ToxCast (EPA 2012a). Although the
use of an array of high-throughput in vitro tests—focused on different putative
toxicity endpoints and pathways—to predict in vivo outcomes is attractive from
both a cost-savings and time-savings perspective, it entails many challenges,
including the following:
Chemical metabolism and disposition may differ between the in vitro
and in vivo situations.
A principle tenet of toxicology is that the concentration of
a toxicant at a specific target site is a key determinant of toxicity. If a metabolite
of a toxicant, not the parent molecule, is responsible for toxicity, the in vitro
systems must be able to form that metabolite—and other metabolites that might
modify the response (for example, alternate detoxification pathways)—in a ratio
similar to what occurs in vivo. If an in vitro system fails to form the toxicant or
if it forms one that does not occur in vivo, the test system will generate a false-
negative or false-positive response. The large amounts of data that can be gener-
ated from -omics experiments may be useful in identifying putative pathways of
toxicity, but the relevance of the pathways to human exposures depends on a
reasonably accurate simulation of the metabolic disposition of the substance that
would occur in vivo.
The time course of effects observed in vitro may be very different from
what occurrs in vivo.
Many chemical treatments of cells result in immediate
changes in gene expression, and the nature and magnitude of the changes are
highly dynamic. Initial responses may be largely adaptive in nature, and not
necessarily reflective of an ultimate toxic effect. Adaptive responses can indi-
Search WWH ::
Custom Search