In 2012, a long-term rat feeding study was published investigating the toxicity of Roundup herbicide and a Roundup-tolerant GM maize, 'NK603' [1,2]. Its results indicated adverse effects on the kidney and liver (the organs of detoxification), and early death. Routine examination of the condition of the animals during the course of the experiment unexpectedly revealed an excess of palpable tumours. When presented chronologically, the emergence of tumours and premature death were clearly accelerated in both the NK603- and Roundup-fed rats. It was also noted that all the results were hormone- and sex-dependent.
The French team of scientists which carried out this study concluded that two-year feeding trials need to be conducted to evaluate thoroughly the safety of GM foods and pesticides in their full formulations (not just the isolated active ingredient, such as the glyphosate in Roundup).
These results and conclusion raised concerns which were very inconvenient for the biotech industry and regulators, so much so that the European Commission (EC) funded a four-year research project to carry out "at least one rat feeding trial" with NK603, and to provide recommendations regarding the "added value" of long-term GM feeding trials. In the event, the 'GM Plant Two Year Safety Testing" study (G-TwYST) turned into two 90-day trials (such as are obligatory for GMO crop approval in the EU), and a two-year combined toxicity/cancer study with toxicity testing at 12 and 24 months, and cancer testing at 24 months.
The EC's G-TwYST project wasn't a repeat of the earlier study, but used different materials and methods in some key areas.
A major difference was that, the earlier study had been careful to use the same strain of rat as Monsanto's 90-day feeding trial in order to derive comparable results to those used in the regulatory process. The 'Sprague-Dawley' rats it used are known to be prone to mammary and pituitary tumours. Reasoning that this rat strain might have led to false positives in the tumours recorded, the G-TwYST scientists used 'Wistar Han RCC', another common strain of laboratory rat which is also prone to mammary and pituitary tumours, but is known to "show the lowest incidence of spontaneous tumours in most organs". This is, no doubt, scientifically sound reasoning, but no parallel study of the two strains was made to check its validity in the face of the unavoidable differences in, for example, the test diets and laboratory facilities used for the two experiments.
Although the published G-TwYST report is entitled "Lack of adverse effects in subchronic and chronic toxicity/carcinogenicity studies on the glyphosate-resistant genetically modified maize NK603 in Wistar Han RCC rats", it reported that 60% of the male rats fed the highest dose of Roundup-sprayed NK603 died prematurely, mainly due to pituitary gland growths. This compared to 36% in the controls. This adverse effect isn't mentioned in the Abstract.
The finding was explained away by the G-TwYST authors as an effect of the high-calorie experimental diet leading to obesity and spontaneous tumours. Why the (presumably) equally high-calorie (but less GM) diet of all the other groups didn't make them equally fat and dead isn't explained.
Other confounding factors, not present in the French experiment, were noted in the G-TwYST rat diets: all the G-TwYST feed, both test and control, was contaminated with glyphosate and with other GMOs. This has major implications.
The G-TwYST team put a lot of effort into 'proving' that historical controls from the same laboratory provided a valid comparison in evaluating the significance of feeding study results. However, besides previous control animals having been fed test diets grown in a different place and at a different time from the current rats, it's clear that the levels of confounding contaminants (pesticides, environmental toxins and GMOs) could have varied wildly between these different experiments. In fact, this finding of compromised feed purity also casts serious doubts about the claimed record of so-called 'spontaneous' tumours in any laboratory animals. (For more on this see .)
No attempt was made to duplicate the Roundup-feeding part of the French experiment. This was needed to separate out the adverse effects caused by the GM NK603 from those of the herbicide formulation sprayed on it and from potential changes in the NK603 itself induced by Roundup exposure. The G-TwYST scientists themselves queried whether their test crop had actually been sprayed with realistic quantities of Roundup.
It seems inescapable that the G-TwYST findings actually tie in rather well with the French study with regard to early death and hormone disruption: the actual presentation may be different only because a different strain of rat with different inherent susceptibilities was used.
Does all this reassure you that NK603 Roundup-tolerant GM maize is safe to eat?
You might like to help GM Freeze with its objections to UK trials of GM crops. Check out https://www.gmfreeze.org/current-actions/
 GM MAIZE IS NOT SAFE TO EAT - October 2012
 GM PROBLEM FEEDING STUDYRE-PUBLISHED - August 2014
 POISONED LAB-RATS ARE NORMAL - August 2015
- Giles-Eric Séralini, et al., 2014, Republished study: long-term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize, Environmental Sciences Europe 26:14
- Pablo Steinberg, et al., February 2019, Lack of adverse effects in subchronic and chronic toxicity/carcinogenicity studies on the glyphosate-resistant genetically modified maize NK603 in Wistar Han RCC rats, Archives of Toxicology
- Feeding study with genetically engineered maize NK603 does not provide evidence of adverse effects on the health of rats, Testbiotech, 17.04.18
- G-TwYST GM Plants Two Year Safety Testing, Conclusions and Recommendations, 29.04.18
- Eating GM maize sprayed with Roundup increased death rate - study, GM Watch 16.02.19
- Sacks G., et al., 2018, How food companies influence evidence and opinion-straight from the horse's mouth, Critical Public Health, 28(2)