29 / 2012, 01.10.2012
[Séralini et al. scientific publication reviewed by BfR ]
The thesis
that rats, their lives received genetically modified corn die earlier
than animals that are fed conventional
corn is experimentally
not sufficiently documented. That is the
conclusion of a review that "long term toxicity of
a Roundup
herbicide and a Roundup-tolerant genetically modified maize" by Gilles-Eric
Séralini and others in the journal "Food and Chemical Toxicilogy" carried out by the
Federal Institute for risk assessment (BfR ) after the publication. "The study
has both weaknesses both
in the design and statistical reporting, so that
the conclusions of the authors
are incomprehensible",
says Professor Dr. Reiner
Wittkowski, Vice President of the Federal
Institute. Also the statement
that Roundup will lead the
time lapse of the
glyphosate product to serious damage to health
and earlier death, is not
sufficiently documented. Many long term studies exist
to glyphosate as herbicide m active substance. Cancer, a higher mortality rate, or effects on the endocrine system of animals
as they the authors in the publication report have not been
observed in these studies.
Mid September
2012 a scientist team led by
Gilles-Eric Séralini, at the
University of Caen in France published the results
of a longitudinal study with rats, glyphosate tolerant, genetically modified corn was been
administered. A part of
genetically modified maize had been
thus treated with a glyphosathaltigen plant protection product (Roundup), another part was untreated.
The corn was each
given in three doses. In addition, Roundup were given other animals were fed conventional
feed, via the drinking water at three
doses. A control group
was fed with
non genetically modified corn. The authors
report that the animals insome of the
test groups of early tumors and other organ damage would have developed and died earlier than
in the control group. The results could be caused
by hormone effects of Roundup as well as ingredients of genetically modified corn.
The Federal
Institute for risk assessment (BfR) has evaluated
the study with regard to the
relevance for the health risk assessment of
genetically modified, glyphosate tolerant corn, and also for the assessment
of the health risk of glyphosate
as the active
substance. Based on the publication, the BfR comes
to the conclusion that the key
messages of the publication are experimentally not sufficiently documented. Main conclusions of the authors
are also incomprehensible due to the inadequacies
of the study
design and the type of presentation and interpretation of the
data.
In particular the too small
number of animals per group, which corresponds to the
recommendations
international qualification recognised
standards for studies
on the carcinogenic effect is criticised
in the study. Used rat tribe has a relatively high spontaneous tumour
rate especially for breast and Hypophsen tumoren, and the number
of animals used is not
sufficient to evaluate the differences
between the test and control groups, alleged by the
authors. The thesis of the
authors who observed effects could be based
on endocrine adverse effects,
are not sufficiently covered in the information collected. The BfR
still criticized that no provision of the
dose was the investigations with the glyphosat
haltigen plant protection product Roundup. Also, the data collected are only imperfectly set forth.
The BfR
has asked the authors
due to these weaknesses of the
study to make the full study
report including the individual animal data available and in addition specific questions asked to allow
a further evaluation of the reported effects.
About the
BfR
The Federal
Institute for risk assessment (BfR) is a scientific
institution in the
Division of the Ministry of food,
agriculture and consumer protection (BMELV).
It advises the Federal
Government and the FederalStates on questions of food,
chemicals and product safety. The BfR
operates its own research
to topics that are closely related with his assessment
tasks.
No evidence of cancer and tumour due to GM corn and/or Herbicide, see the Opinion of the German Agency.
http://www.bfr.bund.de/cm/343/
http://www.bfr.bund.de/de/pres
Updates
See later GMO Pundit posts:
Feeding study in rats with genetically modified NK603 maize and with a glyphosate containing formulation (Roundup) published by Séralini et al. (2012)
BfR-Opinion 037/2012 of 1 October 2012
In mid-September 2012, a scientific team headed by Gilles-Eric Séralini at the University of Caen in France published the results of a long-term study with rats which had been fed ge-netically modified glyphosate-tolerant NK603 maize. One part of the maize had been treated with a glyphosate-containing plant protection product (Roundup) during cultivation, whereas another part was untreated. In each case the maize was administered in three doses. In ad-dition, other animals fed with conventional feed received Roundup via the drinking water, also in three doses. The only control group was fed a non-genetically modified maize. The authors reported that the animals in some of the test groups developed increased incidences of several tumours and other non-neoplastic lesions and died earlier than animals in the con-trol group. The effects could have been caused by hormonal effects of Roundup and specific constituents of the genetically modified maize, respectively. The Federal Institute for Risk Assessment (BfR) has evaluated the study in terms of its relevance for the evaluation of the health risk of genetically modified glyphosate-tolerant maize NK603 and also with regard to the evaluation of the health risk of the glyphosate-containing formulation. On the basis of the publication, the BfR has come to the conclusion that the authors’ main statements are not sufficiently corroborated by experimental evidence, due to deficiencies in the study design and in the presentation and interpretation of the study results. Therefore, the main conclu-sions of the authors are not supported by the presented incomplete data. The study does not comply with internationally recognised standards for long-term carcinogenicity studies. The rat strain used shows a relatively high spontaneous tumour rate, especially for mammary and pituitary tumours, and the number of animals used was too small and insufficient for assess-ing the claimed differences between the test groups and the control group. Also the authors’ hypothesis that the observed effects could result from adverse effects on the endocrine sys-tem is not sufficiently supported by the data presented. Furthermore, the BfR criticises that the glyphosate dose administered was not determined in the studies with the glyphosate-containing plant protection product Roundup. Due to these deficiencies the BfR has asked the authors to provide the complete study report including the individual animal data. More-over, it has asked specific questions in order to allow for a further evaluation of the reported effects.
1 Subject
This opinion refers to the study “Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize” published by Séralini and co-authors in the journal Food and Chemical Toxicology on 19 September 2012 (Séralini et al., 2012).
2 Result
After having reviewed the publication, the German Federal Institute for Risk Assessment (BfR) is of the opinion that the experimental data do not support the main statements in the publication. Further, due to shortcomings in the study design as well as in the presentation and interpretation of the data, relevant conclusions drawn by the authors are not comprehen-sible.
For further assessment, the BfR has asked the authors to provide the complete study report including the individual animal data and has also put forward specific questions. This request has not yet been answered.
3 Justification
Based on scientific opinions of the European Food Safety Authority (EFSA) (EFSA, 2003a; EFSA, 2003b), the NK603 maize that was used in the study was authorised for feed use in accordance with Directive 2001/18/EC on 19 July 2004 and for food use in accordance with Regulation (EC) No 258/97 on 10 October 2004. An application for renewal of these authori-sations in accordance with Regulation (EC) No 1829/2003 was already assessed by EFSA (EFSA, 2009).
The active ingredient in the Roundup formulation used in the study was glyphosate which was included in Annex I of Directive 91/414/EEC in 2002 for a ten-year period based on Di-rective 2001/99/EG from 20.11.2001 (European Commission, 2002). The submitted data were evaluated in the monograph (Draft Assessment Report, DAR) by Germany as Rappor-teur Member State (RMS) with the involvement of the BgVV, the predecessor of the BfR, in 1998. This comprehensive assessment in the DAR has been supplemented several times. Following a decision of the European Commission, the inclusion of glyphosate in Annex I was prolonged until 31 December 2015 with Directive 2010/77/EC from 10 November 2010. Currently, a renewal of the assessment of glyphosate is ongoing within the AIR2 programme based on Regulation (EC) No 1141/2010. Germany acts again as RMS and will, with in-volvement of the BfR (responsible for drafting the chapters on toxicology, residues and ana-lytical methods), establish a new DAR that will be discussed in the framework of the Com-munity centralised procedure led by EFSA. Numerous glyphosate containing plant protection products have been authorised in EU countries (currently 75 herbicides in Germany including 13 different Roundup formulations).
Beside studies on potential health effects from genetically modified NK603 maize the re-searchers led by Gilles-Eric Séralini had published a series of papers on effects of
glyphosate containing plant protection products. Some of them (Richard et al., 2005; Bena-chour et al., 2007; Bellé et al., 2007; Gasnier et al., 2009; Benachour and Séralini, 2009) had been already commented by the BfR.
The aim of the now published study by Séralini and co-authors was to examine potential ef-fects of the genetically modified glyphosate-tolerant NK603 maize and of one glyphosate containing formulation (Roundup) administered to rats over two years. Groups of 10 male and 10 female Sprague Dawley rats (Harlan) were fed diets containing 11, 22 or 33 per cent of NK603 maize (Monsanto Corporation, USA), which had been treated or not treated with Roundup during cultivation. The diet for the control group contained 33 per cent of a non-genetically modified maize line. The animals of another test series received drinking water containing 1.1 x 10-8, 0,09 or 0,5 per cent Roundup .
The authors concluded from the results of the study that the mortality of female animals in all treated groups as well as male animals in three of the groups that had received NK603 maize was higher than in the control group and deaths occurred earlier. According to the authors, all results were hormone and sex dependent, and the pathological profiles were comparable. It was postulated that females developed more frequently large tumours of the mammary gland, and the pituitary was the second most affected organ. In treated males, pathologic lesions in the liver (congestions and necrosis) and kidney (severe nephropathies) were more frequent, the latter was confirmed by biochemical data. The authors further indicated that these results could be explained by non-linear endocrine-disrupting effects of Roundup, but also by overexpression of the transgene in the NK603 maize and its metabolic consequences.
BfR noticed with interest that for the first time a long-term feeding study was performed with a glyphosate containing formulation. Long-term studies were not yet available because for regulatory purposes such studies are worldwide requested only with the active substances. Glyphosate itself has been comprehensively tested. Numerous long-term studies in rats and mice showed no indications of either a carcinogenic potential or increased mortality or any effects on the endocrine system, as reported by Séralini and co-workers in their publication.
However, the BfR is aware of certain co-formulants, in particular surfactants from the group of polyethoxylated alkyl amines (POEA, often designated as tallow amines), that might affect the toxicity of glyphosate containing herbicides. The toxic effects are in some cases more severe compared to studies with the active ingredient. Therefore, the results of the study performed by Séralini's group could provide an experimental contribution to the elucidation of the possible influence of formulants on long-term effects of plant protection products.
While the performance of a long-term study in the case of the glyphosate containing formulation is in principle appreciated, it needs to be mentioned that the published study shows significant shortcomings in the study design and further shortcomings due to incomplete and unclear presentation of the collected data. Furthermore, the main statements were not supported by the experimental data. As outlined in detail below, it is therefore impossible to comprehend the main conclusions of the authors.
3.1 Comments on the study design
Long-term studies are highly complex and elaborate as rats spontaneously develop tumours and other age-related alterations. The published study was not conducted in accordance with internationally accepted standards, such as OECD Test Guidelines No. 451 or Nr. 453 (OECD, 2009a; OECD, 2009b). Instead, the authors have chosen a study design (OECD Test Guideline Nr. 408) that was developed for 90 day (subchronic toxicity) studies (OECD, 1998). Therefore, only 10 animals per sex instead of 50 have been assigned to each group.
However, subchronic studies show a substantially lower variation of age-related pathological changes between animals within a group while those changes are inevitable in long-term studies. As the published study has confirmed, the two-year duration of the study is of the order of the expected life span in rats including the Sprague Dawley strain that was used in the study. This strain, provided by the breeder Harlan, is known to develop spontaneous tumours, particularly mammary and pituitary tumours, at relatively high rates compared to other strains (Brix et al., 2005; Dinse et al., 2010). Therefore, it can be expected that a significant number of animals develop age-related illnesses or die for diverse reasons already during conduct of the study. The distribution of the cases of death between groups can be random, and a number of 10 animals per sex and group is too low to confirm a trend or an effect. Fur-thermore, no statements on statistically significant dose-response-relationships can be made. Larger sample sizes, as recommended for carcinogenicity studies in OECD Test Guidelines No. 451 or No. 453, would be required in order to allow precise statements with respect to the findings.
Regarding the design of the study, another point of criticism is that the mean levels of the daily applied doses of Roundup have not been determined. It should also be noted that the glyphosate containing formulation (Weather-MAX) used for the treatment of NK603 maize during cultivation was different from the formulation (GT Plus) used in the test series with Roundup. Further details on the composition of the applied formulations are lacking.
The publication does not inform whether the diets of all groups contained a total of 33 per cent maize, i.e. whether the diets with 11 and 22 per cent have been supplemented with non-genetically modified maize. The only information given by the authors is that balanced diets were fed and that these diets were considered "substantially equivalent" except for the newly introduced gene. However, detailed information on the composition of the diets is lacking. Moreover, data on feed and water consumption as well as body weight development are missing. The question therefore is, whether balanced diets really had been administered. There are also no further details on the identity of the control maize line that is referred to as "nearest isogenic non-transgenic control". Furthermore, it has to be critically stated that the maize varieties used in the study were not analysed for the presence of mycotoxins.
3.2 Comments on the presentation of results
The first part of the study considers mortality, tumour incidences and other pathological changes and contains descriptive data while statistical analyses are lacking. The presentation of the data in percentage terms or as “x times more”, suggest more impressive results compared to absolute figures.
The BfR is of the opinion that the treatment-related increase in mortality as reported by the authors is not confirmed by the published data. The two cases of death caused by Wilms’ tumours (nephroblastoma) in male animals of two not clearly indicated test groups fed with Roundup treated NK603 maize are not chemically induced and are correctly not claimed to be treatment-induced. Therefore, they should not be used as evidence for a higher mortality compared to the non treated control group. Likewise, no effects of Roundup on the mortality of male rats can be detected.
In female rats mammary tumours are indicated as the main cause of mortality. However, this type of tumour occurs rather frequently particularly in Sprague Dawley rats and if feed is offered ad libitum. In the current study this type of tumour also occurred in approximately 50 per cent of the animals in the control group. As outlined above, the number of animals is not sufficient for an assessment of the difference to the treated animals (60 to 100 per cent with-out a clear dose dependence). The reported comparison with historical control data published in 1992 is not acceptable.
The incomplete and undifferentiated presentation of the data makes evaluation very difficult. For example, it is absolutely insufficient to mention only findings in liver and the digestive tract, as done in table 2, without characterising them from a differential diagnostic standpoint and assessing the grade of severity. Further, the graphs demonstrating mortality and tumours, respectively, are not always in agreement with the statements in the text or can not be followed, as in the case of the observed deaths caused by Wilms' tumours.
A statistical analysis was only performed for the biochemical parameters. This was done with a special kind of principal component analysis (OPLS-DA = Orthogonal Partial Least Squares Discriminant Analysis), but results were only presented for one group (females that had received feed with 33 per cent NK603 maize compared to the control group). In addition, figure 5, presenting biochemical parameters, is difficult to understand. Their assessment would require data of all measurement time points.
3.3 Comments on the mechanisms suspected by the authors
One hypothesis of the authors was that specific compound(s) present in the genetically modi-fied NK603 maize and in the applied glyphosate containing formulation, respectively, could account for the observed increased tumour incidences, particularly in female test animals, by affecting the endocrine system. However, the BfR is of the opinion that no convincing arguments have been provided to support this.
The following points are discussed.
The authors indicate that most of the observed effects show a non-monotonic dose- response-relationship and show a threshold. They consider this as a clear indication that the endocrine system is adversely affected. The authors refer to a recent review published by Vandenberg et al. (2012). However, a detailed look into this paper reveals that its content is not correctly reflected by Séralini et al.. Vandenberg et al. explicitly question the existence of a threshold for adverse effects induced by endocrine disruptors. Thus the cited literature is not suitable to support the authors’ claims. Furthermore, the presence of a non-monotonic dose-response- relationship does not mean that the effects are caused by an impairment of the endocrine system. Non-monotonic dose- response-relationships have also been described for other substances. For example some essential minerals show a non-monotonic dose-response-relationship (Stern, 2010; Calabrese, 2008) yet without affecting the endo-crine system. Considering a non-monotonic dose-response-relationship, a quantitative relationship between the dose and the effect is observed which, however does not proceed in a monotonic manner over the examined dose range. Instead of a non-monotonic dose-response-relationship, the data presented does not allow the identification of any obvious relationship between the observed adverse effects and the applied dose. Rather, the datasets consisting of 3 dose levels and the control group with animal numbers less than 10 per group and sex show no statistically significant relationship between the observed effects and the applied dose.
To further support their thesis the authors refer to their results obtained by measurement of testosterone and estradiol levels (figure 5B). The figure presents the data for hormone levels of the single female animals of the control group and the group, which had received a diet with 33 per cent NK603 maize, 15 months after the commencement of the study. A balanced scientific discussion should include a critical discussion of specific points by the authors. For example, statistically significant differences in hormone levels might easily be assessed on the basis of mean values plus standard deviation. However, figure 5B does not provide a clear basis to perform a statistical evaluation with sufficient accuracy. In addition, the respective data for male animals were not shown. Furthermore, the natural variation in hormone levels caused by the circadian rhythmic and during the estrous cycle was not acknowledged by the authors as a possible cause for the results given in figure 5. It is also known that Sprague Dawley rats develop estrous cycle abnormalities relatively early (from 4-6 months of age; OECD, 2009). The differences observed between treated and control animals 15 months after study begin could thus also be due to variations in hormone levels inde-pendent of the applied substances. If the authors were right in stating that the particularly higher incidence of mammary tumours could be related to the estradiol level, one would expect a statistically significant difference in the estradiol level of the female animals in the group, which had received a diet with 33 per cent NK603 maize, when compared to the control animals. However, this is not identifiable on the basis of the data presented.
The authors also hypothesise that NK603 maize and Roundup could cause hormonal disturbances via an impact on the estrogen system. In this respect, the authors re-gard lower contents of specific organic acids (caffeic and ferulic acid) present in NK603 maize as being responsible for the observed effects. These acids are claimed to exert protective effects in the experimental animals and to impact on the estrogen metabolism. However, significant differences in the estrogen levels of female animals in the group fed with 33 per cent NK603 maize can not be identified on the basis of the data presented. Additional factors, for example a possible modulation of the ER-receptor expression have not been addressed experimentally. Furthermore, the discussion of possible protective effects by plant constituents on tumour development does not reflect the current state of scientific knowledge. With regard to effects induced by the glyphosate containing formulation, the authors discuss the possibility of aromatase inhibition as well as an interaction with cellular estrogen or androgen receptors. However, these anticipated mechanisms have not been tested experimentally in this work. They are based on results from in vitro studies, which have been questioned by the BfR in previous opinions. The thesis of the authors that the observed effects could result from adverse effects on the endocrine system, exerted by the genetically modified NK603 maize and Roundup, respectively, are therefore not sufficiently supported by the experimental data presented in the publication.
4 References
Bellé, R. et al. (2007). Sea urchin embryo, DNA-damaged cell cycle checkpoint and the mechanisms initiating cancer development. Journal de la Societé de Biologie 201 (3), 317-327.
Benachour, N. et al. (2007). Time- and dose-dependent effects of Roundup on human em-bryonic and placental cells. Archives of Environmental Contamination and Toxicology 53 (1), 126-133.
Benachour, N., Séralini, G.-E. (2009). Glyphosate formulations induce apoptosis and necro-sis in human umbilical, embryonic, and placental cells. Chemical Research in Toxicology 22 (1), 97-105.
Seite
Brix, A. E. et al. (2005). Incidences of selected lesions in control female Harlan Sprague-Dawley rats from two-year studies performed by the National Toxicology Program. Toxi-cologic Pathology 33 (4), 477-483.
Calabrese, E. J. (2008). Hormesis: Why it is important to toxicology and toxicologists. Envi-ron Toxicol Chem. 27:1451-1474
Dinse, D. E. et al. (2010). Comparison of NTP historical control tumor incidence rates in fe-male Harlan Sprague Dawley and Fischer 344/N rats. Toxicologic Pathology 38 (5), 765-775.
European Commission (2002). Review report for the active substance glyphosate.
6511/VI/99-final, 21 January 2002. Finalised in the Standing Committee on Plant Health at its meeting on 29 June 2001 in view of the inclusion of glyphosate in Annex I of Directive 91/414/EEC.
EFSA (2003a). Opinion of the Scientific Panel on Genetically Modified Organisms on a re-quest from the Commission related to the Notification (Reference CE/ES/00/01) for the plac-ing on the market of herbicide-tolerant genetically modified maize NK603, for import and processing, under Part C of Directive 2001/18/EC from Monsanto (Question No EFSA-Q-2003-003). The EFSA Journal 10, 1-13.
EFSA (2003b). Opinion of the Scientific Panel on Genetically Modified Organisms on a re-quest from the Commission related to the safety of foods and food ingredients derived from herbicide-tolerant genetically modified maize NK603, for which a request for placing on the market was submitted under Article 4 of the Novel Food Regulation (EC) No 258/97 by Mon-santo (Question No EFSA-Q- 2003-002). The EFSA Journal 9, 1-14.
EFSA (2009). Scientific Opinion of the Panel on Genetically Modified Organisms on applica-tions (EFSA-GMO-NL-2005-22 and EFSA-GMO-RX-NK603) for the placing on the market of the genetically modified glyphosate tolerant maize NK603 for cultivation, food and feed uses and import and processing, and for renewal of the authorisation of maize NK603 as existing product. The EFSA Journal 1137, 1-50.
Gasnier, C. et al. (2009). Glyphosate-based herbicides are toxic and endocrine disruptors in human cell lines. Toxicology 262 (3), 184-191.
OECD (1998). Repeated Dose 90-Day Oral Toxicity Study in Rodents. Test Guideline No. 408. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009a). Carcinogenicity Studies. Test Guideline No. 451. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009b). Combined Chronic Toxicity/Carcinogenicity Studies. Test Guideline No. 453. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009c). Guidance document for histologic evaluation of endocrine and reproductive tests in rodents, part 3: Female reproductive system. OECD Environment, Health and Safety Publications, Series on Testing and Assessment, No. 106, OECD, Paris.
Richard, S. et al. (2005). Differential effects of glyphosate and Roundup on human placental cells and aromatase. Environmental Health Perspectives 113 (6), 716-720.
Séralini, G.-E. et al. (2012). Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Food and Chemical Toxicology, Article in Press.
Stern, B. R. (2010). Essentiality and toxicity in copper health risk assessment: Overview, update and regulatory considerations. Journal of Toxicology and Enviromental Health, Part A, 73:114-127.
Vandenberg, L. N. et al. (2012). Hormones and endocrine-disrupting chemicals: Low-dose effects and nonmonotonic dose responses. Endocrine Reviews 33 (3), 378-455.
No evidence of cancer and tumour due to GM corn and/or Herbicide, see the Opinion of the German Agency.
Two documents in German:
http://www.bfr.bund.de/cm/343/
http://www.bfr.bund.de/de/pres
Updates
See later GMO Pundit posts:
Official English translation of main report:
Feeding study in rats with genetically modified NK603 maize and with a glyphosate containing formulation (Roundup) published by Séralini et al. (2012)
BfR-Opinion 037/2012 of 1 October 2012
In mid-September 2012, a scientific team headed by Gilles-Eric Séralini at the University of Caen in France published the results of a long-term study with rats which had been fed ge-netically modified glyphosate-tolerant NK603 maize. One part of the maize had been treated with a glyphosate-containing plant protection product (Roundup) during cultivation, whereas another part was untreated. In each case the maize was administered in three doses. In ad-dition, other animals fed with conventional feed received Roundup via the drinking water, also in three doses. The only control group was fed a non-genetically modified maize. The authors reported that the animals in some of the test groups developed increased incidences of several tumours and other non-neoplastic lesions and died earlier than animals in the con-trol group. The effects could have been caused by hormonal effects of Roundup and specific constituents of the genetically modified maize, respectively. The Federal Institute for Risk Assessment (BfR) has evaluated the study in terms of its relevance for the evaluation of the health risk of genetically modified glyphosate-tolerant maize NK603 and also with regard to the evaluation of the health risk of the glyphosate-containing formulation. On the basis of the publication, the BfR has come to the conclusion that the authors’ main statements are not sufficiently corroborated by experimental evidence, due to deficiencies in the study design and in the presentation and interpretation of the study results. Therefore, the main conclu-sions of the authors are not supported by the presented incomplete data. The study does not comply with internationally recognised standards for long-term carcinogenicity studies. The rat strain used shows a relatively high spontaneous tumour rate, especially for mammary and pituitary tumours, and the number of animals used was too small and insufficient for assess-ing the claimed differences between the test groups and the control group. Also the authors’ hypothesis that the observed effects could result from adverse effects on the endocrine sys-tem is not sufficiently supported by the data presented. Furthermore, the BfR criticises that the glyphosate dose administered was not determined in the studies with the glyphosate-containing plant protection product Roundup. Due to these deficiencies the BfR has asked the authors to provide the complete study report including the individual animal data. More-over, it has asked specific questions in order to allow for a further evaluation of the reported effects.
1 Subject
This opinion refers to the study “Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize” published by Séralini and co-authors in the journal Food and Chemical Toxicology on 19 September 2012 (Séralini et al., 2012).
2 Result
After having reviewed the publication, the German Federal Institute for Risk Assessment (BfR) is of the opinion that the experimental data do not support the main statements in the publication. Further, due to shortcomings in the study design as well as in the presentation and interpretation of the data, relevant conclusions drawn by the authors are not comprehen-sible.
For further assessment, the BfR has asked the authors to provide the complete study report including the individual animal data and has also put forward specific questions. This request has not yet been answered.
3 Justification
Based on scientific opinions of the European Food Safety Authority (EFSA) (EFSA, 2003a; EFSA, 2003b), the NK603 maize that was used in the study was authorised for feed use in accordance with Directive 2001/18/EC on 19 July 2004 and for food use in accordance with Regulation (EC) No 258/97 on 10 October 2004. An application for renewal of these authori-sations in accordance with Regulation (EC) No 1829/2003 was already assessed by EFSA (EFSA, 2009).
The active ingredient in the Roundup formulation used in the study was glyphosate which was included in Annex I of Directive 91/414/EEC in 2002 for a ten-year period based on Di-rective 2001/99/EG from 20.11.2001 (European Commission, 2002). The submitted data were evaluated in the monograph (Draft Assessment Report, DAR) by Germany as Rappor-teur Member State (RMS) with the involvement of the BgVV, the predecessor of the BfR, in 1998. This comprehensive assessment in the DAR has been supplemented several times. Following a decision of the European Commission, the inclusion of glyphosate in Annex I was prolonged until 31 December 2015 with Directive 2010/77/EC from 10 November 2010. Currently, a renewal of the assessment of glyphosate is ongoing within the AIR2 programme based on Regulation (EC) No 1141/2010. Germany acts again as RMS and will, with in-volvement of the BfR (responsible for drafting the chapters on toxicology, residues and ana-lytical methods), establish a new DAR that will be discussed in the framework of the Com-munity centralised procedure led by EFSA. Numerous glyphosate containing plant protection products have been authorised in EU countries (currently 75 herbicides in Germany including 13 different Roundup formulations).
Beside studies on potential health effects from genetically modified NK603 maize the re-searchers led by Gilles-Eric Séralini had published a series of papers on effects of
glyphosate containing plant protection products. Some of them (Richard et al., 2005; Bena-chour et al., 2007; Bellé et al., 2007; Gasnier et al., 2009; Benachour and Séralini, 2009) had been already commented by the BfR.
The aim of the now published study by Séralini and co-authors was to examine potential ef-fects of the genetically modified glyphosate-tolerant NK603 maize and of one glyphosate containing formulation (Roundup) administered to rats over two years. Groups of 10 male and 10 female Sprague Dawley rats (Harlan) were fed diets containing 11, 22 or 33 per cent of NK603 maize (Monsanto Corporation, USA), which had been treated or not treated with Roundup during cultivation. The diet for the control group contained 33 per cent of a non-genetically modified maize line. The animals of another test series received drinking water containing 1.1 x 10-8, 0,09 or 0,5 per cent Roundup .
The authors concluded from the results of the study that the mortality of female animals in all treated groups as well as male animals in three of the groups that had received NK603 maize was higher than in the control group and deaths occurred earlier. According to the authors, all results were hormone and sex dependent, and the pathological profiles were comparable. It was postulated that females developed more frequently large tumours of the mammary gland, and the pituitary was the second most affected organ. In treated males, pathologic lesions in the liver (congestions and necrosis) and kidney (severe nephropathies) were more frequent, the latter was confirmed by biochemical data. The authors further indicated that these results could be explained by non-linear endocrine-disrupting effects of Roundup, but also by overexpression of the transgene in the NK603 maize and its metabolic consequences.
BfR noticed with interest that for the first time a long-term feeding study was performed with a glyphosate containing formulation. Long-term studies were not yet available because for regulatory purposes such studies are worldwide requested only with the active substances. Glyphosate itself has been comprehensively tested. Numerous long-term studies in rats and mice showed no indications of either a carcinogenic potential or increased mortality or any effects on the endocrine system, as reported by Séralini and co-workers in their publication.
However, the BfR is aware of certain co-formulants, in particular surfactants from the group of polyethoxylated alkyl amines (POEA, often designated as tallow amines), that might affect the toxicity of glyphosate containing herbicides. The toxic effects are in some cases more severe compared to studies with the active ingredient. Therefore, the results of the study performed by Séralini's group could provide an experimental contribution to the elucidation of the possible influence of formulants on long-term effects of plant protection products.
While the performance of a long-term study in the case of the glyphosate containing formulation is in principle appreciated, it needs to be mentioned that the published study shows significant shortcomings in the study design and further shortcomings due to incomplete and unclear presentation of the collected data. Furthermore, the main statements were not supported by the experimental data. As outlined in detail below, it is therefore impossible to comprehend the main conclusions of the authors.
3.1 Comments on the study design
Long-term studies are highly complex and elaborate as rats spontaneously develop tumours and other age-related alterations. The published study was not conducted in accordance with internationally accepted standards, such as OECD Test Guidelines No. 451 or Nr. 453 (OECD, 2009a; OECD, 2009b). Instead, the authors have chosen a study design (OECD Test Guideline Nr. 408) that was developed for 90 day (subchronic toxicity) studies (OECD, 1998). Therefore, only 10 animals per sex instead of 50 have been assigned to each group.
However, subchronic studies show a substantially lower variation of age-related pathological changes between animals within a group while those changes are inevitable in long-term studies. As the published study has confirmed, the two-year duration of the study is of the order of the expected life span in rats including the Sprague Dawley strain that was used in the study. This strain, provided by the breeder Harlan, is known to develop spontaneous tumours, particularly mammary and pituitary tumours, at relatively high rates compared to other strains (Brix et al., 2005; Dinse et al., 2010). Therefore, it can be expected that a significant number of animals develop age-related illnesses or die for diverse reasons already during conduct of the study. The distribution of the cases of death between groups can be random, and a number of 10 animals per sex and group is too low to confirm a trend or an effect. Fur-thermore, no statements on statistically significant dose-response-relationships can be made. Larger sample sizes, as recommended for carcinogenicity studies in OECD Test Guidelines No. 451 or No. 453, would be required in order to allow precise statements with respect to the findings.
Regarding the design of the study, another point of criticism is that the mean levels of the daily applied doses of Roundup have not been determined. It should also be noted that the glyphosate containing formulation (Weather-MAX) used for the treatment of NK603 maize during cultivation was different from the formulation (GT Plus) used in the test series with Roundup. Further details on the composition of the applied formulations are lacking.
The publication does not inform whether the diets of all groups contained a total of 33 per cent maize, i.e. whether the diets with 11 and 22 per cent have been supplemented with non-genetically modified maize. The only information given by the authors is that balanced diets were fed and that these diets were considered "substantially equivalent" except for the newly introduced gene. However, detailed information on the composition of the diets is lacking. Moreover, data on feed and water consumption as well as body weight development are missing. The question therefore is, whether balanced diets really had been administered. There are also no further details on the identity of the control maize line that is referred to as "nearest isogenic non-transgenic control". Furthermore, it has to be critically stated that the maize varieties used in the study were not analysed for the presence of mycotoxins.
3.2 Comments on the presentation of results
The first part of the study considers mortality, tumour incidences and other pathological changes and contains descriptive data while statistical analyses are lacking. The presentation of the data in percentage terms or as “x times more”, suggest more impressive results compared to absolute figures.
The BfR is of the opinion that the treatment-related increase in mortality as reported by the authors is not confirmed by the published data. The two cases of death caused by Wilms’ tumours (nephroblastoma) in male animals of two not clearly indicated test groups fed with Roundup treated NK603 maize are not chemically induced and are correctly not claimed to be treatment-induced. Therefore, they should not be used as evidence for a higher mortality compared to the non treated control group. Likewise, no effects of Roundup on the mortality of male rats can be detected.
In female rats mammary tumours are indicated as the main cause of mortality. However, this type of tumour occurs rather frequently particularly in Sprague Dawley rats and if feed is offered ad libitum. In the current study this type of tumour also occurred in approximately 50 per cent of the animals in the control group. As outlined above, the number of animals is not sufficient for an assessment of the difference to the treated animals (60 to 100 per cent with-out a clear dose dependence). The reported comparison with historical control data published in 1992 is not acceptable.
The incomplete and undifferentiated presentation of the data makes evaluation very difficult. For example, it is absolutely insufficient to mention only findings in liver and the digestive tract, as done in table 2, without characterising them from a differential diagnostic standpoint and assessing the grade of severity. Further, the graphs demonstrating mortality and tumours, respectively, are not always in agreement with the statements in the text or can not be followed, as in the case of the observed deaths caused by Wilms' tumours.
A statistical analysis was only performed for the biochemical parameters. This was done with a special kind of principal component analysis (OPLS-DA = Orthogonal Partial Least Squares Discriminant Analysis), but results were only presented for one group (females that had received feed with 33 per cent NK603 maize compared to the control group). In addition, figure 5, presenting biochemical parameters, is difficult to understand. Their assessment would require data of all measurement time points.
3.3 Comments on the mechanisms suspected by the authors
One hypothesis of the authors was that specific compound(s) present in the genetically modi-fied NK603 maize and in the applied glyphosate containing formulation, respectively, could account for the observed increased tumour incidences, particularly in female test animals, by affecting the endocrine system. However, the BfR is of the opinion that no convincing arguments have been provided to support this.
The following points are discussed.
The authors indicate that most of the observed effects show a non-monotonic dose- response-relationship and show a threshold. They consider this as a clear indication that the endocrine system is adversely affected. The authors refer to a recent review published by Vandenberg et al. (2012). However, a detailed look into this paper reveals that its content is not correctly reflected by Séralini et al.. Vandenberg et al. explicitly question the existence of a threshold for adverse effects induced by endocrine disruptors. Thus the cited literature is not suitable to support the authors’ claims. Furthermore, the presence of a non-monotonic dose-response- relationship does not mean that the effects are caused by an impairment of the endocrine system. Non-monotonic dose- response-relationships have also been described for other substances. For example some essential minerals show a non-monotonic dose-response-relationship (Stern, 2010; Calabrese, 2008) yet without affecting the endo-crine system. Considering a non-monotonic dose-response-relationship, a quantitative relationship between the dose and the effect is observed which, however does not proceed in a monotonic manner over the examined dose range. Instead of a non-monotonic dose-response-relationship, the data presented does not allow the identification of any obvious relationship between the observed adverse effects and the applied dose. Rather, the datasets consisting of 3 dose levels and the control group with animal numbers less than 10 per group and sex show no statistically significant relationship between the observed effects and the applied dose.
To further support their thesis the authors refer to their results obtained by measurement of testosterone and estradiol levels (figure 5B). The figure presents the data for hormone levels of the single female animals of the control group and the group, which had received a diet with 33 per cent NK603 maize, 15 months after the commencement of the study. A balanced scientific discussion should include a critical discussion of specific points by the authors. For example, statistically significant differences in hormone levels might easily be assessed on the basis of mean values plus standard deviation. However, figure 5B does not provide a clear basis to perform a statistical evaluation with sufficient accuracy. In addition, the respective data for male animals were not shown. Furthermore, the natural variation in hormone levels caused by the circadian rhythmic and during the estrous cycle was not acknowledged by the authors as a possible cause for the results given in figure 5. It is also known that Sprague Dawley rats develop estrous cycle abnormalities relatively early (from 4-6 months of age; OECD, 2009). The differences observed between treated and control animals 15 months after study begin could thus also be due to variations in hormone levels inde-pendent of the applied substances. If the authors were right in stating that the particularly higher incidence of mammary tumours could be related to the estradiol level, one would expect a statistically significant difference in the estradiol level of the female animals in the group, which had received a diet with 33 per cent NK603 maize, when compared to the control animals. However, this is not identifiable on the basis of the data presented.
The authors also hypothesise that NK603 maize and Roundup could cause hormonal disturbances via an impact on the estrogen system. In this respect, the authors re-gard lower contents of specific organic acids (caffeic and ferulic acid) present in NK603 maize as being responsible for the observed effects. These acids are claimed to exert protective effects in the experimental animals and to impact on the estrogen metabolism. However, significant differences in the estrogen levels of female animals in the group fed with 33 per cent NK603 maize can not be identified on the basis of the data presented. Additional factors, for example a possible modulation of the ER-receptor expression have not been addressed experimentally. Furthermore, the discussion of possible protective effects by plant constituents on tumour development does not reflect the current state of scientific knowledge. With regard to effects induced by the glyphosate containing formulation, the authors discuss the possibility of aromatase inhibition as well as an interaction with cellular estrogen or androgen receptors. However, these anticipated mechanisms have not been tested experimentally in this work. They are based on results from in vitro studies, which have been questioned by the BfR in previous opinions. The thesis of the authors that the observed effects could result from adverse effects on the endocrine system, exerted by the genetically modified NK603 maize and Roundup, respectively, are therefore not sufficiently supported by the experimental data presented in the publication.
Bellé, R. et al. (2007). Sea urchin embryo, DNA-damaged cell cycle checkpoint and the mechanisms initiating cancer development. Journal de la Societé de Biologie 201 (3), 317-327.
Benachour, N. et al. (2007). Time- and dose-dependent effects of Roundup on human em-bryonic and placental cells. Archives of Environmental Contamination and Toxicology 53 (1), 126-133.
Benachour, N., Séralini, G.-E. (2009). Glyphosate formulations induce apoptosis and necro-sis in human umbilical, embryonic, and placental cells. Chemical Research in Toxicology 22 (1), 97-105.
Seite
Brix, A. E. et al. (2005). Incidences of selected lesions in control female Harlan Sprague-Dawley rats from two-year studies performed by the National Toxicology Program. Toxi-cologic Pathology 33 (4), 477-483.
Calabrese, E. J. (2008). Hormesis: Why it is important to toxicology and toxicologists. Envi-ron Toxicol Chem. 27:1451-1474
Dinse, D. E. et al. (2010). Comparison of NTP historical control tumor incidence rates in fe-male Harlan Sprague Dawley and Fischer 344/N rats. Toxicologic Pathology 38 (5), 765-775.
European Commission (2002). Review report for the active substance glyphosate.
6511/VI/99-final, 21 January 2002. Finalised in the Standing Committee on Plant Health at its meeting on 29 June 2001 in view of the inclusion of glyphosate in Annex I of Directive 91/414/EEC.
EFSA (2003a). Opinion of the Scientific Panel on Genetically Modified Organisms on a re-quest from the Commission related to the Notification (Reference CE/ES/00/01) for the plac-ing on the market of herbicide-tolerant genetically modified maize NK603, for import and processing, under Part C of Directive 2001/18/EC from Monsanto (Question No EFSA-Q-2003-003). The EFSA Journal 10, 1-13.
EFSA (2003b). Opinion of the Scientific Panel on Genetically Modified Organisms on a re-quest from the Commission related to the safety of foods and food ingredients derived from herbicide-tolerant genetically modified maize NK603, for which a request for placing on the market was submitted under Article 4 of the Novel Food Regulation (EC) No 258/97 by Mon-santo (Question No EFSA-Q- 2003-002). The EFSA Journal 9, 1-14.
EFSA (2009). Scientific Opinion of the Panel on Genetically Modified Organisms on applica-tions (EFSA-GMO-NL-2005-22 and EFSA-GMO-RX-NK603) for the placing on the market of the genetically modified glyphosate tolerant maize NK603 for cultivation, food and feed uses and import and processing, and for renewal of the authorisation of maize NK603 as existing product. The EFSA Journal 1137, 1-50.
Gasnier, C. et al. (2009). Glyphosate-based herbicides are toxic and endocrine disruptors in human cell lines. Toxicology 262 (3), 184-191.
OECD (1998). Repeated Dose 90-Day Oral Toxicity Study in Rodents. Test Guideline No. 408. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009a). Carcinogenicity Studies. Test Guideline No. 451. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009b). Combined Chronic Toxicity/Carcinogenicity Studies. Test Guideline No. 453. OECD Guidelines for the Testing of Chemicals, OECD, Paris.
OECD (2009c). Guidance document for histologic evaluation of endocrine and reproductive tests in rodents, part 3: Female reproductive system. OECD Environment, Health and Safety Publications, Series on Testing and Assessment, No. 106, OECD, Paris.
Richard, S. et al. (2005). Differential effects of glyphosate and Roundup on human placental cells and aromatase. Environmental Health Perspectives 113 (6), 716-720.
Séralini, G.-E. et al. (2012). Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Food and Chemical Toxicology, Article in Press.
Stern, B. R. (2010). Essentiality and toxicity in copper health risk assessment: Overview, update and regulatory considerations. Journal of Toxicology and Enviromental Health, Part A, 73:114-127.
Vandenberg, L. N. et al. (2012). Hormones and endocrine-disrupting chemicals: Low-dose effects and nonmonotonic dose responses. Endocrine Reviews 33 (3), 378-455.
Pundits thoughts:
Much that's not a surprise, and plenty more to be said..
Much that's not a surprise, and plenty more to be said..
No comments:
Post a Comment