Editorial: New Commission Implementing Regulation on Risk Assessment of GM plant applications: novel elements and challenges


EFSA Journal
EFSA Journal 2013;11(12):e11121 [5 pp.].
Panel members at the time of adoption
Waigmann E, Gomes A, Lanzoni A and Perry JN
11 December 2013
European Food Safety Authority (EFSA), Parma, Italy
Download Article (109.46 KB)

No abstract available


On 8 December 2013 the new Implementing Regulation (IR) (EU) No 503/2013[2] on applications for EU market authorisation of genetically modified (GM) food and feed submitted in the frame of Regulation (EC) 1829/2003[3]became fully effective. Therefore it is now mandatory to submit applications concerning GM plants and their derived food and feed products in accordance with its requirements. The new IR is the outcome of a lengthy discussion process between the Member States and the European Commission with the aim of incorporating the existing EFSA Guidance for the risk assessment of food and feed from GM plants (EFSA Panel on Genetically Modified Organisms (GMO Panel), 2011; referred to hereafter as ‘2011 EFSA Food/Feed GD’) into a legal text. The new IR was adopted by the European Commission after having received a positive opinion of the Standing Committee on the Food Chain and Animal Health in April 2013.

The 2011 EFSA Food/Feed Guidance Document (GD), a revision of an earlier document (EFSA, 2006), provided more specific guidance to applicants, particularly concerning the conduct, design and analysis of field trials for compositional, agronomic and phenotypic comparison, which now have sufficient statistical power to detect unintended effects if they exist. In particular, it introduced a new methodology, bioequivalence, to enable differences between the GMO event and its conventional counterpart to be placed into context and assessed for their biological relevance. The solution that EFSA identified was to gather field data for the GM, its conventional counterpart and commercial reference varieties, collected concurrently within the same trials. This allows the concept of the history of safe use of food items on which the comparative analysis is based to be quantified. Bioequivalence is adapted from existing practice in the field of medicine (van der Voet et al., 2011). The distinction between statistical significance and biological relevance was also recognised by EFSA’s Scientific Committee (SC) (2011a).

Whilst the risk assessment approach and methodology detailed in the EFSA 2011 Food/Feed GD were largely maintained in the IR, there were several important elements newly introduced by the Member States and the European Commission. These elements were in part prompted by the views of certain Member States and the desire to improve consumer confidence in GM plants (see e.g. Preamble 11 of IR (EC) No 503/2013).

In the following overview the most relevant differences in the risk assessment requirements between the EFSA 2011 Food/Feed GD and the new IR are discussed.

The IR has several requirements related to the molecular characterisation of GM plants with stacked events (i.e. two or more GM events are present in a GM plant) that differ from those presented in the EFSA 2011 Food/Feed GD. For example, the new IR requests the re-sequencing of DNA inserts and their flanking regions in GM plants containing stacked events and their comparison with the nucleotide sequence of the respective single events.

GM plants containing stacked events (‘stacks’) are usually produced by conventional crossing of parental GM plant lines containing one or more single events without the involvement of additional genetic transformations. As outlined in the EFSA 2011 Food/Feed GD, the strategy for the risk assessment of such stacks relies on a preceding full risk assessment of the single events and then focuses on the identification of potential interactions between the respective single events.

The new requirement to re-sequence DNA inserts and their flanking regions in GM stacks is designed to ensure the integrity of the insert in the stack. Previously, during the assessment of stacks, the GMO Panel has accepted data produced by Southern analysis to ensure insert integrity. Several Member States commented that this technique confirms only the macro structure of the inserts but does not inform on small changes, e.g. point mutations that may have occurred. However, it has been reported that point mutations leading to single-nucleotide polymorphisms (SNPs), as well as other types of small mutations, are naturally occurring genomic changes contributing to the plasticity of plant genomes (for review see Weber et al., 2012). SNP frequencies in plant genomes are highly variable depending on species, genetic diversity of the cultivars assessed, and whether coding or noncoding regions are considered (Maughan et al., 2010). A study in maize MON 810, the only GM maize event authorised to date for cultivation in the EU, concluded that the mutation rates of the insert and its flanking regions were not different from those observed in other maize sequences (La Paz et al., 2010). Mutations can occur spontaneously in GM as well as in non-GM crop plants during their growth in the field, but these affect single plants, with a huge dilution effect at harvest, minimising the likelihood of any adverse impact on human health or the environment. Since there is no evidence from the scientific literature indicating that small sequence changes should be analysed for the safety evaluation of a stacked event, the added value of this new requirement is unclear.

The toxicological assessment of GM plants intended for food/feed production is impacted by a major change in the new IR, which now requires a mandatory 90-day rodent feeding study for single transformation events.

The toxicological evaluation of GM plants and derived products as laid down in the EFSA 2011 Food/Feed GD, in line with Regulation (EC) No 1829/2003, is performed by comparing the GM plant with an appropriate non-GM control. It is primarily informed by results from the preceding molecular characterization and the comparative compositional, agronomic and phenotypic analysis of the GM plant which, on a case-by-case basis, may trigger additional, appropriate toxicological studies, for example on constituents newly expressed in the GM plant. Typical examples of case-by-case requirements are the 28-day oral toxicity study in rodents on newly expressed protein(s), and the 90-day study in rodents when driven by a specific hypothesis identified in the preceding evaluation. The new IR introduces a significant shift in the approach to the 90-day study in rodents: from a study needed on a case-by-case basis to a mandatory requirement. The EFSA Guidance on conducting a repeated dose 90-day toxicity study in rodents on whole food/feed (EFSA SC, 2011b) is specifically quoted in the IR as the reference guidance document for the study design and statistical approach. However, the study is now intended to fulfil two separate roles: to support the identification of potential adverse effects in the whole genetically modified food/feed (representing an exploratory study) and to address remaining uncertainties (representing a confirmatory, hypothesis-driven study). Given the now mandatory nature of the 90-day study, EFSA is faced with the challenge to determine how such an exploratory study should be interpreted within toxicological risk assessment.

For allergenicity, both the EFSA 2011 Food/Feed GD and the IR follow the same strategy, considering the allergenicity of the newly expressed proteins and the endogenous allergenicity of the whole GM plant. While allergenicity assessment of the newly expressed protein requires similar data sets in both documents, assessment of the endogenous allergenicity of the whole plant contains a significant change: building on an EFSA recommendation, the quantitative measurement of known plant allergens as part of the compositional analysis is now a mandatory requirement in the IR.

The whole GM plant is assessed for its endogenous allergenicity in case the recipient plant is known to be allergenic. For example in soybean the allergen repertoire of the GM plant has historically been compared with that of its non-GM comparator(s), with the use of sera from individuals allergic to soybean. This approach has limitations due to the intrinsic variability between allergic individuals and the difficulty of obtaining well-characterised sera. Consequently, the EFSA 2011 Food/Feed GD recommends that, in order to minimise the use of human sera, relevant known endogenous allergens could be included in the comparative compositional analysis, which would imply a quantification of the relevant allergens of a crop (see also EFSA GMO Panel, 2010). Several technologies (e.g. mass spectrometry), which do not require human sera, have proven useful in the quantification of endogenous allergens (Fernandez et al., 2013). The IR takes this EFSA recommendation forward by requesting the mandatory inclusion of endogenous allergens of a crop plant, as referred to in relevant OECD (Organisation for Economic Co-operation and Development) documents (e.g. OECD, 2012), within the compositional analysis. This is expected to provide robust, quantifiable and reproducible information for risk assessment.

The new elements in the IR have triggered several activities within EFSA. One is an update of the EFSA submission guidance (EFSA, 2011) which advises and supports applicants in the preparation and presentation of a dossier. This update incorporates the new elements in the IR and has been published on the EFSA website[1].

In addition, EFSA has set up a Task Force[5] to provide clarity concerning the objectives of the mandatory 90-day rodent feeding study in GM risk assessment and more detailed instructions on how to apply the existing EFSA Guidance (EFSA SC, 2011b). The Task Force, consisting of EFSA scientists with relevant expertise, aims to deliver a Statement in April 2014 for publication in the EFSA Journal. To ensure consistency and scientific quality, external reviewers with relevant expertise will be appointed and the relevant EFSA scientific Panels (GMO Panel, NDA Panel and the Scientific Committee) will be consulted.

In summary, whilst the new IR reflects largely the content of the EFSA 2011 Food/Feed GD, several important new elements have been added by the Member States and the European Commission. Some of the new elements are in line with EFSA recommendations, for example regarding allergenicity. Other elements are a significant modification of the current EFSA requirements, amongst which the mandatory requirement for the 90-day rodent feeding study to support the toxicological assessment of single transformation events seems the most controversial and challenging. It should be noted that a review of the requirement for a mandatory 90-day study is clearly foreseen by the legislator by 2016. The Commission will perform this review based on new scientific information such as the outcome of the EU seventh Framework project GRACE (GMO Risk Assessment and Communication of Evidence) (see Art. 12, IR 503/2013). The key research objectives of GRACE[6] include to ‘test various types of animal feeding trials [explanation added by authors: including 90-day rodent feeding studies with GM plant derived feed], as well as alternative in vitro methods in order to determine how suitable they are and what useful scientific information they provide for health risk assessments of GM food and feed’.

Looking into the mid-term future, the new IR and the EFSA 2011 Food/Feed GD will co-exist, since application dossiers compiled in line with the EFSA GD and submitted before 8 December 2013 will continue to be assessed under the EFSA GD. Moreover, notwithstanding the coming into force of the IR, EFSA will fulfil its ongoing mission to develop new or additional guidance for the risk assessment of GM plants, taking into account new scientific developments and methodologies. For example, an additional guidance on the agronomic and phenotypic characteristics of a GM plant is currently being developed by a dedicated Working Group of the EFSA GMO Panel[7]. The IR explicitly foresees future reviews incorporating new EFSA guidance (see Art. 12, IR 503/2013).

The safety and usefulness of GM plants is subject to an intense political and societal debate, characterised by widely diverging positions in different EU Member States. Considering that the IR was endorsed by the Member States with a qualified majority, one expectation is that risk assessment requirements outlined in the IR will satisfy the majority of Member States, and allow them to support safety conclusions on GM Plants, where appropriate. The future will show whether the IR will fulfil this expectation and contribute to a convergence of Member States views on the safety of GMOs.

[2] Commission Implementing Regulation (EU) No 503/2013 on applications for authorisation of genetically modified food and feed in accordance with Regulation (EC) No 1829/2003 of the European Parliament and of the Council and amending Commission Regulations (EC) No 641/2004 and (EC) No 1981/2006. OJ L 157, 8.6.2013, p. 1-48.
[3] Regulation (EC) No 1829/2003 of the European Parliament and of the Council of 22 September 2003 on genetically modified food and feed. OJ L 268, 18.10.2003, p. 1-23.
[4] EFSA guidance on the submission of applications for authorisation of genetically modified plants under Regulation (EC) No 1829/2003
[5] http://registerofquestions.efsa.europa.eu/roqFrontend/questionLoader?question=EFSA-Q-2013-00718
[6] http://www.grace-fp7.eu/content/grace-brief
[7] http://registerofquestions.efsa.europa.eu/roqFrontend/questionLoader?question=EFSA-Q-2013-00606

Print on demand
Number of Pages