Open Letter to Testbiotech in Response to its Report and Press Release Dated 7-11-2014
Quedlinburg, 10 November 2014
This letter is sent to you in response to the report and press release dated Friday 7 November 2014 and issued by your organization on its website. In these documents, you criticize the scientific content of a recently published scientific article on 90-day feeding studies with diets containing genetically modified MON810-maize varieties and their comparators performed within the EU-funded GRACE project.
In the annex below, the arguments raised against the scientific content of the publication are discussed in more detail, which has not led us to revise our previous conclusions. In brief, it is concluded that:
- Testbiotech’s comments fail to distinguish between statistical significance and biological relevance.
- No differences were found that had not been detected by GRACE, while the comments fail to acknowledge the context provided by the additional statistical comparisons between control and conventional groups.
The research within GRACE is carried out according to established scientific standards and under conditions of well-documented quality control and good practices. Additionally, the GRACE consortium attaches great value to dialogue and transparency, among others by involving stakeholders during various stages of the research design, execution, and result interpretation. This has already included, for example, two on-site stakeholder events and rounds of written comments, during which also representatives of your organization had been able to submit comments and present their views. Data obtained during the experiments have been made publicly available, and allow for critical appraisal.
We would also like to draw your attention to the fact that Archives of Toxicology, the scientific journal in which the above mentioned article was published, offers a platform for scientific discussion of the GRACE results. All those interested are invited to submit comments on current and upcoming publications in the form of letters to the editor of Archives of Toxicology. These contributions will be published in Archives of Toxicology together with responses from the research project. We hope that our communications in future can be held in the spirit of dialogue and transparency fostered by the GRACE project.
The GRACE Consortium,
Signing on its behalf,
Prof. Dr. Joachim Schiemann
Response to the criticisms raised by Testbiotech regarding the interpretation of the results obtained by Zeljenková et al. (2014)
1) Testbiotech claims: “The GLP-controlled 90-day feeding study in Han Wistar RCC rats aimed at providing a safety evaluation of two varieties of MON810 as a dietary admixture of 11% (low dose) or 33% (high dose). The authors claim to have followed “the guidance for such studies published by the EFSA Scientific Committee in 2011 and the OECD Test Guideline 408”. Although this is generally true, it needs to be stated that it is not entirely true, because the histopathological assessment of macroscopic findings (in the low dose group) as required in Test Guideline 408 was obviously not followed.”
Response: In contrast to the statement of Testbiotech each gross lesion, as required by the OECD Test Guideline 408, was microscopically analyzed. The macroscopic as well as the corresponding histopathological examination findings were and are shown in the Electronic Supplementary Material Table 6 of the paper by Zeljenková et al. (2014).
2) Testbiotech states that Zeljenková et al. (2014) “dismiss the toxicological relevance of the observed lower levels of total protein in serum (TP)”.
Response: The TP level was significantly lower in the serum of male rats fed the 11% GMO and 33% GMO diet in the feeding trial A and in that of female rats fed the 33% GMO diet in the feeding trial B if compared to the corresponding control animals. However, a number of facts support the conclusion that the above-mentioned effects are not related to the feeding of the GMO-containing diets:
I) No such decreases were observed in the female rats fed the 11% GMO and 33% GMO diets in the feeding trial A as well as in the male rats fed the 11% GMO and 33% GMO diet in the feeding trial B. The changes in TP are therefore inconsistent across the various groups and treatments. Please note that in both trials A and B, genetically modified maize with the same GMO event (MON810) had been used.
II) A decrease of the total protein levels in serum was also observed in female rats of study B fed the conventional maize DKC6815 at a 33% level.
III) The magnitude of the differences between the groups (10-20%) was considered to be small, inconsistent and therefore not of any biological or toxicological relevance.
IV) If the decrease of the serum protein levels were as relevant as Testbiotech states, one would have expected that this decrease had an impact on the growth of the young animals used in these feeding trials. However, the body weight data of the rats fed the GMO diets were comparable to those of the rats fed the control diets.
V) Testbiotech argues that the decrease in the serum total protein levels described could be due to a nephrotic syndrome or to an impaired protein synthesis in the liver. If the rats were suffering from a nephrotic syndrome, oedema and histological alterations of the kidneys would become evident. In the experimental groups showing a decrease in serum total protein levels neither oedema nor histological alterations of the kidneys were observed in male rats fed the 33% GMO diet in feeding trial A and in female rats fed the 33% GMO diet in feeding trial B. Furthermore, if protein synthesis were impaired, a certain degree of liver parenchymal cell death would have become evident, but the histopathological analyses of the liver of male rats fed the 33% GMO diet in feeding trial A and in female rats fed the 33% GMO diet in feeding trial B revealed no histological alterations.
3) Testbiotech states “The second inappropriate dismissal involves weight changes of the pancreas and blood glucose levels. It is remarkable that the authors did not discuss these changes in conjunction, in spite of the well-known role of the pancreas in the regulation of blood glucose levels.”
Response: A number of facts also in this case support the conclusion that the decrease in pancreas weight changes and in the blood glucose levels are not related to the feeding of the GMO-containing diets:
I) The blood glucose levels were not altered in female rats fed the GMO diets in the feeding trial A as well as in male and female rats fed the GMO diets in the feeding trial B.
II) The measured blood glucose values in male rats fed the 11% and 33% GMO diet in the feeding trial A were within or close to the ranges of the groups of male rats fed the conventional maize varieties.
III) The increase in the blood glucose levels of male rats fed the 11% and 33% GMO diet in the feeding trial A was not dose-dependent.
IV) The decrease in the pancreas weight of male rats fed the 11% and 33% GMO diet in the feeding trial A was not dose-dependent.
V) The decrease in the pancreas weight was also observed with the two conventional maize varieties used in the feeding trial A, whereby the decrease only reached statistical significance in the case of the conventional variety 2.
VI) The conclusion reached by Zeljenková et al. (2014) in this case is also supported by the small change as compared to the presented standardized effect size and the inconsistency across groups and treatments, and therefore the changes cannot be considered biologically or toxicologically relevant.
GRACE involves stakeholders in research planning and drawing of conclusions. Read here for more information.
New Stakeholder Consultation Reports:
Consultation on Good Review Practice in GMO Impact Assessment
Three research projects, one question: Which methods are suitable for assessing the risk of genetically modified plants
Coordinators of GRACE, GMO90+ and G-TwYST explain their research goals and how the will co-operate by sharing methodologies, materials and data.