The Séralini study - Scientific proof or spoof?

In 2012, a controversial study was published by Gilles-Eric Séralini and colleagues in the scientific journal, Food and Chemical Toxicology which caused mass media hype as well as critique from scientists globally. A two-year long feeding experiment using rats was set up to investigate whether consuming the genetically modified (GM) maize variety (NK603) or Roundup® herbicide was detrimental to the health of the mammals.

Genetic modification is a precise improvement technique in plant breeding; where a gene (or genes), each with a single beneficial trait, is inserted into the plant genome. The maize variety used in this study is genetically modified to express a protein from an Agrobacterium species, which renders it resistant to glyphosate herbicides, such as Roundup®1.

The principal investigator, Séralini, is a professor at the University of Caen in France and co-founder of the anti-GMO research group called CRIIGEN (the Committee for Research and Independent Information on Genetic Engineering). Previous studies consisted of 90-day rat feeding trials to study effects of GM maize on mammals and emphasized the need for a long-term multi-generational feeding trial2–5. Twelve previous long-term studies (reviewed by Snell et al), failed to show the same effects presented in Séralini’s study6. These studies concluded that GM maize was safe for animal and human consumption while Séralini and colleagues claim that GM maize had carcinogenic and toxic effects on rats. Different diets were used in the current study; containing differing levels of GM and non-GM maize, in conjunction with the herbicide Roundup®. Each diet was fed to 10 ‘Sprague-Dawley’ (the lab strain utilised) rats of each sex and food and water was freely accessible1.  

One of the most fundamental flaws of the study is the breed of rats used. Even though these rats have been used in numerous food safety studies (mainly toxicological studies), the laboratory strain of ‘Sprague-Dawley’ rats is known for spontaneous tumour development7,8; which is also dependent on the amount of food they are given8. Availability of food and water could have resulted in overfeeding and their propensity to develop tumours plays a bigger role in a 2-year trial versus a 90-day trial. Data from the study shows that more males with a smaller proportion of GM maize in their diets died than those fed a bigger proportion GM maize diet. It is also questionable whether the percentages of maize used in the diet constitute balanced diet for the rats? There was no non-maize control as part of the study design. Additional negative controls using more maize varieties and also a full diet using rat food only would have been useful to draw better conclusions.

Apart from the flaws in the research design, analyses of results were one-sided (death rate and tumour development) and some have pointed out that the study holds ethical violations. For these types of experiments, the sample size has to be at least 20 rodents per sex according to Food and Drug Administration guidelines9. The current sample size promotes confusion about whether tumours developed spontaneously or due to the specific diet. Scientific method requires clarity between control group and sample group. The proportion of control group to test group is unacceptable scientific method and there is no dose-response data (where increased exposure to a specific substance should show an increased incidence of cancer).

NK603 maize was granted a general release permit in South Africa in 200210 and has been used widely since. Various national food safety and regulatory agencies reviewed the paper and rejected its conclusions, however; Kenya banned GMOs. In response to Séralini’s paper, FSANZ (Food Standards Australia New Zealand) requested a copy of the original data for analysis to determine if the current approval of NK603 maize should be amended, but Séralini and co-authors did not respond. Toxicological studies in rodents were deemed unsuitable for testing whole GM foods. The European Food Safety Authority (EFSA), the German Federal Institute for Risk Assessment and Health Canada have reviewed the Séralini paper and also rejected its conclusions. EFSA considered independent assessments of the paper by organisations of six EU Member States11,12. Safety of NK603 maize is comparable to conventional maize and therefore placement of NK603 maize for food or feed or processing is unlikely to have an adverse effect on human or animal health or on the environment11,13.

The study was republished recently, in the journal Environmental Sciences Europe, a journal too young to have an impact factor, with admission that the design was not intended to be a carcinogenicity study14. The motive behind re-publication was to enable scientific transparency, but the fundamental flaws of the study were not addressed. There was great media hype about a book release and a movie based on what was labelled the “Séralini affair”. From a scientific point of view, the study does not substantiate whether NK603 maize or Roundup® is harmful or not. Conclusions from an experiment where the design, execution and analysis are fundamentally flawed, are disqualified.

The controversy behind GM warrants careful examination and interpretation of safety studies. Is the experiment scientifically sound? Funding sources and conflicts of interest of authors should also be questioned. Food safety is an important aspect of Biotechnology, and safety studies should be designed to give an indication of whether a particular GM product is safe for human consumption. The images of rats with tumours will circulate for a while, as shocking images cause a longer lasting reaction than scientific facts.

For further discussion on this study, visit the Online Panel.

Author: Azeeza Rangunwala, African Centre for Gene Technologies (ACGT)                                                                                    


  1. Séralini, G.E. et al. Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Food Chem. Toxicol. 50, 4221–31 (2012).
  2. De Vendômois, J. S., Roullier, F., Cellier, D. & Séralini, G.-E. A comparison of the effects of three GM corn varieties on mammalian health. Int. J. Biol. Sci. 5, 706–26 (2009).
  3. Kiliç, A. & Akay, M. T. A three generation study with genetically modified Bt corn in rats: Biochemical and histopathological investigation. Food Chem. Toxicol. 46, 1164–70 (2008).
  4. Knudsen, I. & Poulsen, M. Comparative safety testing of genetically modified foods in a 90-day rat feeding study design allowing the distinction between primary and secondary effects of the new genetic event. Regul. Toxicol. Pharmacol. 49, 53–62 (2007).
  5. He, X. Y. et al. A 90-day toxicology study of transgenic lysine-rich maize grain (Y642) in Sprague-Dawley rats. Food Chem. Toxicol. 47, 425–32 (2009).
  6. Snell, C. et al. Assessment of the health impact of GM plant diets in long-term and multigenerational animal feeding trials: a literature review. Food Chem. Toxicol. 50, 1134–48 (2012).
  7. Kaspareit, J. & Rittinghausen, S. Spontaneous neoplastic lesions in Harlan Sprague-Dawley rats. Exp. Toxicol. Pathol. 51, 105–7 (1999).
  8. Davis, R. K., Stevenson, G. T. & Busch, K. A. Tumor Incidence in Normal Sprague-Dawley Female Rats. Cancer Res. 16, 194–197 (1956).
  9. U.S. Food and Drug Administration. Redbook 2000: IV.B.1 General Guidelines for Designing and Conducting Toxicity Studies. (Center for Food Safety and Applied Nutrition, 2013).
  10. Department of Agriculture Forestry and Fisheries, South Africa. (2015).
  11. EFSA. WWW.EFSA.EUROPA.EU. (2015).
  12. Food Standards Australia. WWW.FOODSTANDARDS.GOV.AU. (2015).
  13. Devos, Y. et al. EFSA’s scientific activities and achievements on the risk assessment of genetically modified organisms (GMOs) during its first decade of existence: looking back and ahead. Transgenic Res. 23, 1–25 (2014).
  14. Séralini GE. et al Republished study: Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Environmental Sciences Europe. 26,14-31(2014).