Bad science in the paper “long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize”

By: Myles Power Edited by: Hannah

A good friend of mine recently sent me a French paper entitled, “Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize”. The paper investigated the health effects of a Roundup-tolerant genetically modified maize, cultivated with Roundup weed killer on rats over 2 years. The paper came to the conclusion that rats fed on the GM maize sprayed with Roundup were much more likely to die at an earlier age, in addition to other health problems, including large mammary tumors and severe liver and kidney damage. Even though the paper has only been published for a short amount of time, the internet is rife with anti-GMO articles claiming this paper is the proof that GMOs are dangerous. The paper and the majority of articles also include graphic images of rats with painful looking tumours, allegedly caused by the maize and Roundup.

The paper was peer reviewed and published in Food and Chemical Toxicology on September 19th. So where is the problem? Firstly, you should know that just because something is peer reviewed does not mean that the experiment, or the paper, is perfect. This is an example of a paper that has a poorly designed experiment.

The experiment involved 100 male and 100 female albino Sprague-Dawley rats, who were divided into groups of 10. For each sex, a control group was fed on plain water and standard maize. 6 groups were fed with 11%, 22% and 33% of GM maize, either treated with Roundup or not. The final 3 groups were fed with the control maize, but had access to water contaminated with 1.1×10-8% (the contaminating level of some regular tap waters), 0.09% (concentrations found in some GM feed) and 0.5% (half of the minimal agricultural working dilution) of Roundup. The results apparently showed that 50% of males and 70% of females died prematurely, compared with only 30% and 20% in the control group.

There are two main problems with this experiment; the first being the number of rats. Having only 20 rats in each group cannot lead to statistically significant results. The control group, with only 10 rats of each sex, was also vastly lacking in size. Furthermore, I find it confusing why an experiment with so few rats would investigate the effects both Roundup and GM maize in one paper.

The second problem is the strain of rat used. The albino Sprague-Dawley rat is known to have a high likelihood of spontaneously growing tumours. The paper “Spontaneous Tumours in Sprague-Dawley Rats and Swiss Mice” found that 45% of the Sprague-Dawley rats (179 males and 181 females – a larger group than the French paper) developed tumours in an 18-month period (a shorter time than French paper). It also found that the percentage of female rats with tumors was almost double that of males. This difference was accounted for chiefly by the high incidence of mammary tumours in the females. It is also noteworthy that this paper did not find it necessary to publish pictures of the rats with tumours, which would have been identical to the ones published in the French paper.

There have been hundreds of studies comparing GM food and non-GM food in multiple species, and almost every study showed no difference in health between animals fed on either GM and non-GM food. So why would a paper suddenly, out of the blue, claim that there is a risk? It might have something to do with the lead author (Gilles-Eric Séralini) being the President of the Scientific Board of Criigen an anti-GM lobbying group. Funny, you would think something like that would have shown up in the conflict of interest section…

So, in conclusion, it is hard to draw any conclusions from this paper; a paper that used the wrong type of rat and had a control group of just 20; a paper whose author mislead the reader into thinking that the tumours dont occur in the control group and who may have had ulterior motives for producing this work, apart from just to further our knowledge of science.