Agriculture Reference
In-Depth Information
the GM crop and its conventional counter-
part) with a given variability in the
parameter being analysed and in the light of
the questions to be addressed by the study.
In the case of equivalence testing, the
null hypothesis is non-equivalence and, in
the approach devised by the EFSA guidance,
a comparison is made between the GM crop
and the various reference varieties.
Equivalence limits are established on the
basis of the variability of the values of the
reference varieties. If the range of the values
of the GM crop, expressed statistically as
90% coni dence limits, lies within these
boundaries, then the null hypothesis is
rejected and the GM crop can be considered
equivalent. In these cases, the type I error
that is controlled (at the 5% level) is the false
conclusion that the crops would be
equivalent when they are actually not, while
the type II error is the incorrect outcome
that the null hypothesis is true and that the
crops would therefore be non-equivalent.
h e EFSA guidance foresees a number of
scenarios for the comparison of the GM crop
with its non-GM comparator based on eight
possible combinations of outcomes; namely,
two possible outcomes for the dif erence
test (i.e. GM crop being statistically sig-
nii cantly dif erent or not from the
comparator) and four for the equivalence
test (i.e. GM crop being equivalent; more
likely to be equivalent than not; more likely
to be non-equivalent than equivalent; or
non-equivalent to its comparator) (EFSA,
2011). In the approach recommended by the
EFSA, the reference crops grown during the
same i eld trial have, in the i rst instance,
replaced the other sources of information on
background variability, such as the ILSI Crop
Composition Database and literature data,
providing an accurate indication of the
variability of the particular crop grown
under the same conditions as the GM crop
and its conventional non-GM counterpart.
ingredients are commonly employed by
producers to predict the nutritional value of
the produced feed formulation. Hence, it can
be argued that the data on the compositional
analysis of the GM crop and its counterparts
can provide insight into its possible
nutritional impact.
A working group of the EFSA's Panel on
Genetically Modii ed Organisms (GMOs)
reviewed the knowledge gathered through
nutritional studies on GM crops and their
counterparts, particularly with poultry, pigs
and ruminants (beef and dairy cattle), as
well as other species (EFSA, 2007). It was
concluded that if the compositional data for
the GM crop were comparable to those of
non-GM crops, nutritional equivalence
could also be established in these cases
(EFSA, 2007), which was also in line with
the conclusion by Flachowsky
et al
. (2012)
that i rst-generation GM crops with
agronomic traits did not inl uence the health
and performance of livestock animals, nor
the safety and quality of derived animal
products (see also Chapter 6).
h e EFSA's review also acknowledged
that if the genetic modii cation af ected the
bioavailability of ingredients, i.e.
af ecting
their uptake by the animal during digestion
of the GM-crop-containing feed, the impacts
might not be predicted solely with the aid of
the compositional analysis (EFSA, 2007). In
these cases, as well as for animal feeds with
improved nutritional characteristics owing
to either increased levels of nutrients or
decreased levels of anti-nutrients, studies
either in model laboratory animals or in
target livestock animals under controlled
conditions may be warranted (EFSA, 2007).
Such tests could be focused either on
assessing the digestibility of a certain
nutrient or on the increased performance
(related to improved nutrition) per se.
h e
review by Flachowsky
et al
. (2012) gives an
account of the requirements for such studies
in laboratory and animal species, including
the experimental design and physiological
end points to be measured, which logically
dif er from one species to another and,
conversely, also depend on the implicated
feed nutrient/anti-nutrient. Bioavailability
and other measures of nutritional impact
Nutrition
In the production of animal feeds for
livestock, data on the composition of
