Page 227 - Environmental Nanotechnology Applications and Impacts of Nanomaterials

212 Principles and Methods

could act as biopersistent fibers. A fullerene could also be considered for
the development as a carbon-based reference standard NM. No doubt
as we undertake a more deliberate exploration of NM properties that
lead to adverse biological events, we may be able to develop reference
materials with predictive toxicity. The establishment of benchmark
standards will rapidly expand our knowledge of NM toxicity.

Overall Considerations in the Assessment
of NM Toxicity

A rational approach to NM toxicity begins with an assessment of the
physico-chemical properties of the materials that allow them to interact
with and possibly damage biological systems. Another set of tests
involves in vitro cellular assays that reflect the response of a number
of different cell types that may be targeted at the portal of entry or sys-
temic sites after NM uptake and spread. Finally, any test strategy
should include appropriate use of animal models, potentially the most
difficult assay to consider. While some experts regard animal testing as
obligatory, it is important to understand that this mode of experimen-
tation is costly and time-consuming. In our opinion, animal testing
should be done judiciously and possibly reserved for those materials
that show a toxic potential in the in vitro studies. Moreover, where such
testing is used, it is likely to be more informative if used in conjunction
with a known mechanism of injury rather than a descriptive approach.
This aspect will be discussed later.
The decision about an appropriate test strategy for NM should con-
sider cost and time. A sobering thought is that among the 80,000 chem-
icals that are currently registered for commercial use in the US, only
slightly more than 500 have undergone long-term and only 70 short-term
testing by the National Toxicology Program (NTP). Moreover, the
resource-intensive nature of these studies puts the cost of each bioas-
say at $2–4 million, and tests generally take more than three years to
complete. Thus, the test strategy must be pragmatic. While it may be
optimal to conduct comprehensive in vitro and in vivo toxicological
assessment in some instances, it is probably more practical to develop
in vitro decision matrices that avoid unnecessary in vivo testing, such
as when in vitro testing fails to show any evidence of toxicity.
Accordingly, the construction of a database that is premised on an
in vitro predictive test result could help to determine under what cir-
cumstances in vivo testing should be performed.
Predictive assays should ideally be premised on a mechanism of
injury that is directly related to in vivo toxicological or pathological out-
comes [11]. Elucidation of NM properties that are responsible for ROS
generation, with the possibility of generating oxidative stress responses