THIRD WORLD NETWORK BIOSAFETY INFORMATION SERVICE
Dear Friends and Colleagues
EU Risk Assessments of GM Food and Feed Found Inadequate
Currently, 55 different genetically engineered plants have been authorised in the EU for use in food and feed. There is a controversy in the EU now as to whether feeding trials with GE plants have to be conducted before the plants are granted market authorization, with industry demanding that this obligation be abandoned.
In response, Testbiotech has published a new report providing an overview of the risks of GE plants for humans and animals. The report is based on an analysis of the EFSA opinions published in the last few years as well as the outcomes of recent and ongoing EU research projects that Testbiotech says have failed, as yet, to develop sufficiently reliable and robust methods to replace the feeding studies.
Testbiotech shows how current EU regulations are inadequate and makes an urgent call to raise the standards of risk assessment for GE plants before they are granted market authorisation. Among its recommendations are:
- Broaden risk assessment to include additional methods, compounds and plant characteristics.
- Assess the impact on the immune system and reproduction.
- Take into account long-term and accumulated effects.
- Assess the residues from spraying with herbicides the plants were made resistant to.
- Any decision on market authorisation must prioritise the Precautionary Principle, and reflect the limits of current knowledge and resulting uncertainties.
The Summary, Conclusions and Recommendations of the report are reproduced below.
With best wishes,
Third World Network
131 Jalan Macalister
10400 Penang
Malaysia
Website: https://biosafety-info.net/ and http://www.twn.my/
To subscribe to other TWN information services: www.twnnews.net
RISKS AND SIDE EFFECTS FOR HUMANS AND ANIMALS: WHAT REALLY GOES WRONG IN THE REGULATION OF GENETICALLY ENGINEERED PLANTS
– Risk Assessment of Genetically Engineered Plants Used for Food And Feed – and Flaws in the Work of the European Food Safety Authority EFSA
by Christoph Then & Andreas Bauer-Panskus
http://www.testbiotech.org/sites/default/files/Risk%20assessment%20Food%20%26%20Feed_Testbiotech.pdf
SUMMARY
This report provides an overview of deficiencies in current European Food Safety Authority EFSA risk assessment in regard to genetically engineered plants for import and usage in food and feed.
The report is based on an analysis of EFSA opinions published within the last few years. It further takes in the outcomes of the EU research projects GRACE and MARLON as well as the EU Commission announcement that it will reconsider the Implementation Regulation (503/2013). This regulation sets the standards for the risk assessment of genetically engineered plants for use in food and feed. Amongst others, it includes a request that feeding studies with the plants are conducted in order to assess health risks. Industry is demanding that this request be abandoned.
As Testbiotech shows, the EU projects mentioned above only partially fulfilled their objectives. The GRACE project failed to develop sufficiently reliable and robust methods to replace the feeding studies. The MARLON project was unable to define suitable methods to monitor potential health effects after a market authorisation has been issued.
In the light of these findings and after a detailed analysis of EFSA opinion, Testbiotech recommends that the standards for risk assessment are substantially raised. Some of the relevant topics are:
›› The requirements on the data to be provided for the first steps of risk assessment (which is the comparison between the genetically engineered events and mostly isogenic lines) need to be broadened to include more detailed methods and data on further substances (such as small RNAs) and on plant characteristics.
›› The functional stability of the additional DNA and its interaction with the environment have to be given priority. The events should be exposed to a broad range of defined stress conditions. Data on metabolomics should be provided.
›› Stacked events need to be examined just as carefully as the single plants. In field trials, the parental plants have to be included as comparators.
›› Independently of the outcome of the first stage of risk assessment, further investigations must be performed to assess the impact on human and animal health from the consumption of whole food and feed. These investigations should also include potential impacts on the immune system and reproduction.
›› Long-term effects have to be given priority. To assess these effects, feeding trials must take into account the whole lifetime of the animals, including their offspring.
›› Residues from spraying with complementary herbicides must be assessed in detail.
›› The toxicity of each of the Bt toxins produced in the plants must be assessed in detail. Whereby special attention needs to be paid to combinatorial effects with other substances (especially with stress factors) and the impact on the immune system.
›› Independent control during the data generation is absolutely necessary during each step of risk assessment.
›› Cut-off criteria such as a prohibition of market authorisation must be established for genetically engineered organisms able to spread into native populations.
CONCLUSIONS AND RECOMMENDATIONS
Within the bounds of the current state of knowledge, the EU project GRACE was unable to find sufficiently reliable and robust methods that could replace animal feeding studies. Furthermore, the EU project MARLON did not succeed in identifying methods for monitoring the effects on human and animal health from the consumption of genetically engineered plants. In the light of these findings, and in awareness of substantial deficiencies in current EFSA risk assessment, there is an urgent need to raise the standards of risk assessment for genetically engineered plants before they are granted market authorisation. We recommend the following measures:
First steps in risk assessment:
›› The requirements for data that has to be provided for the first steps of risk assessment (which is the comparison between the genetically engineered events and mostly isogenic lines) must be broadened to include more detailed methods and data on further substances (such as small RNAs) and plant characteristics.
›› Priority must be given to the functional stability of the additional DNA and its interaction with the environment. All events should be exposed to a broad range of defined stress conditions. Data on metabolomics should also be provided.
›› Data must be provided for all parts of the plants and represent several stages of vegetation.
›› Stacked events must be examined at least as carefully as the single plants. In field trials, the parental plants must be included as comparators.
›› The genetically engineered varieties, which are actually used for cultivation also need to be assessed. Data stemming from these varieties could also be part of post-market monitoring.
›› Where there is some doubt, only data stemming from the comparison of the comparators (isogenic lines and parental plants) can be relied on for making a decision on market authorisation.
›› There must be a requirement to provide reliable methods allowing independent scientists to determine the expression of all newly produced substances in the plants such as enzymes or Bt toxins. The variations in the expression of the substances must be examined under defined stress conditions.
›› If no new substances are produced in the plants, but the plants natural composition is changed by, for example, RNAi, the changes in the concentration also have to be examined under defined stress conditions.
Second stage of risk assessment:
›› Independently of the outcome of the first stage of risk assessment, further investigations must be performed to assess the impact of the consumption of the whole food and feed on human and animal health. These investigations should also include potential impacts on the immune system and reproduction.
›› Long-term effects have to be given priority. To assess these effects, feeding trials must take into account the whole lifetime of the animals, including their offspring.
›› Data generated from feeding trials should include metabolomic data of the animals.
›› The potential impacts on the microbiome of humans and animals must be added in at this stage.
›› Accumulated effects, which, for example, can result from mixing genetically engineered plants in food and feed have to be assessed in detail.
›› Residues from spraying with complementary herbicides have to be assessed in detail. Not only the active ingredients, but also the formulations of the herbicides used for their cultivation have to be taken into account, in addition to the combinatorial effects if more than one complementary herbicide can be applied.
›› The toxicity of each of the Bt toxins produced in the plants must be assessed in detail, focussing, in particular, on the combinatorial effects with other substances (especially with stress factors) and the impact on the immune system. Simple assumptions derived from the mode of action of naturally occurring Bt toxins are not sufficient.
›› Special attention has to be given to the empirical investigation of combinatorial effects emerging in stacked events.
Further recommendations:
›› For each step of the risk assessment, independent control is absolutely essential during data generation.
›› Cut-off criteria such as a prohibition of market authorisation for genetically engineered organisms able to spread into native populations, have to be established.
›› If new relevant methods for data generation or data assessment become available or new risk related questions do emerge, these must be included within a short period of time. In this case, the events that have already been granted authorisation will have be to be re-assessed without delay.
›› Specific guidance needs to be established for genetically engineered plants that are changed in their nutritional composition, in particular, taking into account the long-term effects from consumption.
›› Any decision on market authorisation must prioritise the precautionary principle, and reflect the limits of current knowledge and resulting uncertainties.