GM Free Cymru

More GM Trickery from EFSA. Environmental risk assessment guidelines are deeply flawed

Comments from GM-Free Cymru

The new EFSA Guidance on the environmental risk assessment of GM crops (1) has been subjected to careful scrutiny by Testbiotech -- and Christoph Then, the author of the report, has warned that EFSA is still maintaining its strategy of "softening the burden of regulation" for the GM industry. We would agree with that. The document is complex and difficult to read, but it is strong on theory and very weak on the practical or experimental protocols which are expected of applicants for GM approvals. Over and again we see signs that EFSA is prepared to accept the applicants assurances on various ERA matters -- rather than insisting on hard experimental or field evidence that demonstrates unequivocal lack of harm. And there is very little on the indirect or "unintended" effects of growing GM crops -- in spite of the insistence in the relevant EU Directives that these effects are important and must be considered. This is a crucial point now that abundant evidence has shown that the management of RR crops involves massive environmental damage associated with glyphosate / Roundup spraying procedures.

In the new Guidance there is clear evidence of a new strategy to dump the term "substantial equivalence" (which of course has been relentlessly criticised as a scientifically nonsensical term) and to replace it with equally nonsensical terms involving "the concept of familiarity" and "the concept of comparative assessment." The underlying EFSA strategy is to pretend that since the biology of the traditionally cultivated plants from which GM plants have been derived is well known, it can be assumed that there will be nothing new or surprising in the GM varieties -- which can therefore be assumed to be safe. That of course is a scam, as well as being intellectually feeble. It is probably also illegal, since the whole thrust of EU legislation on GM plants is that they are biologically DIFFERENT -- and indeed the developers of GM plants have to show that their varieties are distinctive, uniform and stable. Most of the GM varieties have patents attached to them, which also indicates that in the minds of the breeders like Monsanto and Syngenta, they are DIFFERENT from the varieties from which they were bred. As Christoph says, thew use of the "comparative assessment" concept by EFSA will substantially narrow the process of hazard identification and hazard characterization -- and that will substantially increase the risk to the environment.

Christoph also shows how EFSA has adopted the belief system of the industry-dominated International Life Sciences Institute (ILSI). The concept of comparative assessment as now used by EFSA was developed by industry and ILSI between 2001 and 2003. During this period, Harry Kuiper and Gijes Kleter (both members of the EFSA GMO Panel ) were active within the ILSI Task Force as experts and as authors of the relevant scientific publications. It is a scandal that Kuiper has remained as Chair of EFSA's GMO Panel since 2003, and that he is still Chair in spite of the massive criticism directed at the Panel from NGOs and even from the Commission and EU member states.

On the matter of the gene flow and the persistence and invasiveness of GM plants either in cultivated situations or in the wild, Christoph accuses EFSA of demonstrating no concern about these major problems. He says that the EFSA position "is not acceptable from the perspective of the precautionary principle." We would go further, and say that EFSA has long since abandoned any use of the Precautionary Principle, and is now fully signed up to the Anti-Precautionary Principle" by which the onus is placed on those who have concerns about harm to demonstrate unequivocally that GM products are not safe.

EFSA also demonstrates (not for the first time) that it has an extraordinarily complacent attitude towards stacked events. As Christoph points out, the text of the EFSA document is a confusing mish-mash of points probably designed to mislead and confuse. The bottom line is that EFSA has no intention of accepting that stacked events (even where there may be six or seven events built into a single variety) may lead to plant behaviour which is unpredictable and possibly harmful. The bottom line appears to be that EFSA will not insist on practical or field environmental assessments of new varieties containing stacked events if it thinks that it has adequate knowledge of those events individually as expressed in earlier plant varieties. It appears to be prepared to accept assurances or presumptions on the part of the plant breeder. It will not look for synergistic or other effects (such as signs of stress, genetic instability or gene scrambling) because it appears to have decided in advance that it will not find them.

Finally, Christoph identifies that EFSA wishes to concentrate its environmental assessment work on target species (such as the corn borer or the "weeds" taken out in the management of herbicide tolerant GM crops) and to disregard any effects that there might be on non- target species and species higher in the food chain -- including vertebrates. It mentions these effects, but seems to think that they are theoretical rather than real. Again, this is unbelievably complacent, given that there is abundant evidence from all over the world of major damage done by GM crops and management procedures to livestock and human beings. As Christoph says: "There are certain risks for wildlife such as birds that are not assessed under food and feed, and secondly, wild animals can contribute to the dissemination of viable seeds and plant material". These are major ecological / environmental issues, and EFSA is being negligent (yet again) in failing to address them.

The EFSA document also has relevance for health and safety -- which of course is at the centre of another battle being waged between EFSA and the European Parliament and NGOs. To quote Christoph again: "At present, if comparative assessment does not reveal any significant findings, EFSA does not even require feeding studies with genetically engineered plants to investigate any health risks for humans or livestock. This kind of risk assessment concept is not acceptable and inadequate to provide necessary consumer and environmental protection." That is very diplomatically put! As far as we are concerned, this document demonstrates yet again that EFSA's GMO Panel is unfit for purpose, and that its prime strategy at the moment is to loosen and soften up the regulatory framework for GMOs, in order to facilitate GMO approvals and speed up the authorisation process. That august body appears to have no concerns at all that in the process of doing what the GM industry wants, it might be endangering both the environment of Europe but also the health of millions of European consumers.

(1) http://www.efsa.europa.eu/en/scdocs/doc/1879.pdf

EFSA´s standards for environmental risk assessment not sufficient

TESTBIOTECH Background 21-12-2010

Testbiotech analysis of EFSA Guidance on the environmental risk assessment of genetically modified plants http://www.testbiotech.de/sites/default/files/Testbiotech_comment_on_EFSA_%20Guidance_ERA.pdf

Christoph Then

Summary

EFSA Guidance on environmental risk assessment (ERA) is inadequate to assess the risks of genetically engineered plants. It is based upon comparative assessment – a concept that is far too narrow and biased. It does not define any cut off criteria even for plants that are persistent and might become invasive. It is not sufficiently clear on the risk assessment of stacked events and does not integrate all relevant levels of the food web. For example, wildlife vertebrates are completely omitted. Further deficiencies are listed in previous documents prepared by Testbiotech.

Introduction:

In its recent comment on the Draft Guidance document, Testbiotech (2010) proposed a list of ten points essential for the risk assessment of genetically engineered plants (link). These recommendations are based on a more detailed analysis (Then, 2010a) – and are as follows:

In the opinion of Testbiotech, EFSA Guidance on environmental risk assessment (EFSA 2010a) still mostly lacks these crucial elements and the ten points remain valid. Some changes were introduced into the final EFSA Guidance and therefore the following points are addressed in this background paper:

1. Substantial equivalence, familiarity and comparative assessment

In its previous draft, EFSA reverted to the concept of substantial equivalence as the underlying paradigm of risk assessment. In the final Guidance (EFSA 2010a), the term substantial equivalence is deleted from the text. Comparative safety assessment and the concept of familiarity are, nevertheless, still used to describe the very basic ERA concept. The substance of the Guidance has not been changed. EFSA Guidance still relies on the concept of substantial equivalence as proposed by the OECD (1993). The terms comparative assessment, familiarity and substantial equivalence are simply used as synonyms for the same concept: “The ERA of GM plants involves generating, collecting and assessing information on a GM plant in order to determine its potential adverse impact relative to its non-GM plant comparator, and thus assessing its comparative safety. The underlying assumption of the comparative assessment for GM plants is that the biology of traditionally cultivated plants from which the GM plants have been derived, and the appropriate comparators is well known. To this end the concept of familiarity was developed by the OECD (OECD, 1993).” (page 11). As a recent Testbiotech background paper shows (Then & Bauer-Panskus, 2010), the concept of comparative assessment as used by EFSA was substantially developed by industry and the International Life Sciences Institute (ILSI) between 2001 and 2003. During this period, Harry Kuiper and Gijes Kleter (both members of the EFSA GMO Panel ) were active within the ILSI Task Force as experts and as authors of the relevant scientific publications. In 2004, the concept was adopted in the first GMO Panel (EFSA 2004) food and feed Guidance document. The main problem with comparative assessment as proposed by EFSA is that genetically engineered plants are not seen as basically different from conventionally bred plants. Therefore, genetically engineered plants are not assessed as technical products inheriting specific risks and technical qualities – to the opposite they are assessed by comparing them with plants derived from conventional breeding. This has huge impact on the overall process of risk assessment. According to EFSA Guidance, comparative assessment will largely influence and substantially narrow the outcome of hazard identification and hazard characterization (see for example Figure 1 of EFSA´s Guidance). As Testbiotech explained (Then & Bauer-Panskus, 2010; Then & Potthof, 2009), the comparison of genetically engineered plants and their conventional counterparts can be seen as an important tool, but should not be used as starting point or general concept of risk assessment. Instead of using comparative assessment, a broad range of non-biased and specific technical data should be generated by subjecting the genetically engineered plants to a range of standardized conditions. These data should, for example, cover genetic stability, interactivity between the genome and the environment, potential impact of climate change and also include reactions to specific abiotic and biotic stressors. Metabolic profiling, measurement of gene activity and determining the content of decisive components (such as Bt toxins or metabolities from application of herbicides) are crucial elements when investigating the impact that defined environmental conditions have on genetically engineered plants, and for performing proper hazard identification at the beginning of risk assessment. In 2009, Testbiotech presented this overall concept as a “stress test” or “crash test” (Then & Potthof, 2009). These more specific investigations are decisive for risk assessment because genetic engineering in plants is the only technology in the sector of plant breeding that does not rely on the plants´ own genome regulation, but on technically enforced gene activity, and in many cases also involves the insertion of additional genetic information from other species. EFSA Guidance fails to provide an adequate scientific concept on how to generate and how to assess the relevant data prerequisite for hazard identification and hazard characterization.

2. Persistence and invasiveness

EFSA Guidance (EFSA, 2010a) does not consider gene flow as such to be a problem. Even if plants are likely to persist in the environment and are able to exchange their genetic information with wild species, EFSA will not regard this as a risk per se (e.g. see decision- making tree, Figure 4 of EFSA Guidance). This position of EFSA is not acceptable from the perspective of the precautionary principle. The long-term impact of gene flow in the fields and in the surrounding environment (as is the case with rape seed or poplar trees) can scarcely be assessed by existing scientific methods. For example, under current climate change, genetically engineered plants might show unintended effects that cannot be taken into account during risk assessment because they are triggered only by certain environmental conditions. There are several publications that show unintended effects in genetically engineered plants that cannot be predicted (see list of publications in Then, 2010). For example, higher fitness can emerge via hybridization with wild relatives as observed in rice (Lu&Yang, 2009). In this context, non-knowledge is a significant factor that must be taken into account in risk assessment. Thus, unequivocal cut off criteria should not be designed to favour the authorization of genetically engineered plants that cannot be recalled from the fields and / or the environment. This approach also is necessary in regard to the general requirements of the EU regulation (Dir 2001/18), that foresees reevaluation of genetically engineered plants every ten years, and allows the withdrawal of products from the market if new risks or technical failures become apparent. Current EU regulations would be contradicted if genetically engineered plants could not be controlled in their distribution and persistence.

3. Stacked events

The wording of EFSA Guidance (EFSA 2010a) on the risk assessment of stacked events is not clear and in some ways ambiguous. For example, in the summary it says: “Further, GM plants containing stacked events are considered with respect to specific areas of risk.” The relevant chapters of EFSA Guidance contain a confusing mixture of criteria and measures that can be applied to stacked events. It leaves out any clarification that the risk assessment of stacked events cannot rely on the assessment of single traits. Undoubtedly, stacked events should undergo full risk assessment and be treated as new applications. The general need for empirical investigation into the combinatorial effects of stacked events cannot be denied. Currently, six different and technically modified Bt-toxins are being combined in a single plant (see http://www.testbiotech.org/en/node/423). In addition, these plants even incorporate gene constructs for conferring herbicide tolerance. It is not possible to deduce possible interactions within these plants from the properties of single traits.

4. Non target organisms and animal feeding studies

Testbiotech is concerned that the tiered approach described by EFSA, will be influenced by linked working groups whose members are from the EFSA GMO panel (e.g. Detlev Bartsch and Jeremy Sweet) and industry (see Romeis et al, 2008). In the EFSA document on non-target organisms (EFSA 2010b), there are even some passages plagiarized from Romeis et al (2008) In general, combinatorial effects and additional stressors from the receiving environment will require empirical tests on all levels of the food web, regardless of whether effects are found on lower levels or not. This basic requirement is not incorporated in EFSA Guidance. By focussing on the tiered approach, the higher levels of the food web i.e. the vertebrate levels, are completely omitted. Two aspects are relevant for this food web level. Firstly, there are certain risks for wildlife such as birds that are not assessed under food and feed, and secondly, wild animals can contribute to the dissemination of viable seeds and plant material.

The conclusions from these observations are that risk assessment of non-target organisms should not be reduced to specific tiers. Investigations should be continued even if no risks are identified on the first tier. Furthermore, feeding studies to investigate health risks for wildlife vertebrates should be included. At present, if comparative assessment does not reveal any significant findings, EFSA does not even require feeding studies with genetically engineered plants to investigate any health risks for humans or livestock. This kind of risk assessment concept is not acceptable and inadequate to provide necessary consumer and environmental protection.

References

EFSA, 2004, Guidance document of the Scientific Panel on Genetically Modified Organisms for the risk assessment of genetically modified plants and derived food and feed, the EFSA Journal (2004) 99, 1–94
EFSA, 2010a, EFSA Panel on Genetically Modified Organisms (GMO); Guidance on the environmental risk assessment of genetically modified plants.
EFSA Journal 2010;8(11):1879. [111 pp.]. doi:10.2903/j.efsa.2010.1879. Available online: www.efsa.europa.eu/efsajournal.htm
EFSA, 2010b,EFSA Panel on Genetically Modified Organisms (GMO); Scientific Opinion on the assessment of potential impacts of genetically modified plants on non- target organisms. EFSA Journal 2010; 8(11):1877. [72 pp.] doi:10.2903/j.efsa.2010.1877. Available online: www.efsa.europa.eu/efsajournal.htm
Lu B-R, Yang C, 2009, Gene flow from genetically modified rice to its wild relatives: Assessing potential ecological consequences, Biotechnol Adv, doi:10.1016/ j.biotechadv.2009.05.018
OECD, 1993. Safety considerations for biotechnology: scale up of crop plants. Organisation for Economic Co-operation and Development, 1-43.
Romeis J, Bartsch D, Bigler F, Candolfi MP, Gielkens MMC, Hartley SE, Hellmich RL, Huesing JE, Jepson PC, Layton R, Quemada H, Raybould A, Rose RI, Schiemann J, Sears MK, Shelton AM, Sweet J, Vaituzis Z, Wolt JD, 2008. Assessment of risk of insect- resistant transgenic crops to nontarget arthropods. Nature Biotechnology 26, 203-208.
T estbiotech, 2010, Ten points for a better risk assessment of genetically engineered plants, a Testbiotech background, http://www.testbiotech.org/en/node/416
Then, C. & Potthof, C., 2009, risk reloaded – risk analysis of genetically engineered plants within the European Union, Testbiotech report, http://www.testbiotech.org/en/node/180
Then, C., 2010, AgroBiotechnology: Testbiotech opinion on EFSA’s draft guidance on the environmental risk assessment of genetically modified plants, Testbiotech report, http://www.testbiotech.org/en/node/400
Then C., & Bauer-Panskus, A., 2010, European Food Safety Authority: A playing field for the biotech industry, a Testbiotech background, http://www.testbiotech.org/en/node/431