Agriculture Reference
In-Depth Information
and frequency of the proposed introduction.
Such an evaluation also considers the
potential direct and indirect, as well as the
immediate, delayed and cumulative long-
term adverse ef ects.
Key elements for the ERA are the
potential changes in the interactions
between the GM plant and the biotic and
abiotic factors, namely: changes in the
persistence and invasiveness of the GM
plant; potential for gene transfer; inter-
actions between the GM plant and target
organisms; interactions between the GM
plant and non-target organisms; ef ects on
biogeochemical processes and abiotic
environment; and impacts of specii c
cultivation, management and harvesting
techniques associated with the cultivation
of the GM plant. As for food and feed safety
assessment, the comparative approach is the
guiding principle for the ERA of GM plants
and it is applied to address each specii c area
of concern. Further details on the ERA
strategy are provided by the EFSA (2010a)
but are not addressed here since they are
outside the scope of this chapter.
and dietary habits of the dif erent consumer
groups. However, since reliable consumption
data may not always be available, data on
import and production quantities may be
helpful to assess worst-case scenarios.
Probabilistic methods can also be valuable
tools to determine plausible intake values.
In practice, intake estimates for any given
food product can dif er signii cantly
depending on the data source. h is is well
illustrated by the following example: the
intake of certain compounds has been
estimated either from European or US
population consumption data sets, or from
the WHO Global Environmental Monitoring
System - Food Contamination Monitoring
and Assessment Programme (GEMS/Food
Consumption Cluster Diets, accessible at
gems/en/index1.html)
that includes the
European population and the FAO food
balance sheet. h e food balance sheets are
based on the total amounts of food produced,
imported, exported and utilized for various
purposes, not on dietary surveys, and are
generally considered to overestimate average
consumption. h ese data sources are
intrinsically dif erent from each other and
inevitably lead to dif erent intake estimates,
which cannot be evenly compared. As a i rst
step to address this problem, the EFSA has
consolidated 27 national dietary surveys
into a so-called 'comprehensive European
database' (EFSA, 2012a). h e EFSA is
currently undertaking further work to
harmonize dietary data collections across
EU member states (EFSA, 2012b).
As already mentioned above, the aim of the
food and feed exposure assessment is to
estimate quantitatively the likelihood of
exposure of humans and animals to products
derived from GM plants. h is is done i rst by
determining the concentrations of the newly
expressed protein(s) and of the endogenous
constituents altered by genetic modii cation,
and second, by identifying and quantifying
any new constituents. In cases where the
genetic modii cation results in an altered
level of an endogenous constituent, or where
a new constituent occurs naturally in other
food and feed products, the anticipated
change in total intake of this constituent
must be assessed considering realistic
scenarios.
Ideally, intake levels should be
estimated from representative consumption
data and should take into account several
factors such as the inl uences of processing
and storage conditions and possible routes
of exposure, as well as the characteristics
market monitoring
h e risk characterization of GM plants and
derived products is based on data from
hazard identii cation, hazard character-
ization and exposure assessment. A sound
risk characterization considers all the
available evidence collected during the risk
assessment and highlights, whenever pos-
sible, any uncertainty identii ed at any stage
of the risk assessment (EFSA, 2007). When
addressing uncertainty, distinction should