Prompted by the recent Elle anti-GMO article fiasco that was based on zero science and only one person’s unproven ideas, I looked up the paper where scientists “proved” that GMO corn and Roundup herbicide causes cancer. For those who have heard about this and believe it, I invite you to read the paper, along with the accompanied Letters to the Editor by other scientists.
Seralini et al., Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize, Food and Chemical Toxicology, Volume 50, Issue 11, November 2012, Pages 4221-4231.
In case it’s all too long and dry for you, and you trust my judgment instead 😉 here are my 10 reasons why this paper is bullshit, plus other important points from the Letters to the Editor:
- Zero statistics/no significance testing of mortality (death rate) or tumour formation in the rats. ‘Nuff said!! – one author of a Letter ran a significance test on the mortality results and p was wayyyyy over 0.05 (= absolutely non-significant), a total opposite of what was claimed. How does anyone get away with publishing data without statistics these days?!
- Sample size of only 10 rats per treatment (so a difference of 30% vs 50% = 2 rats). 10 is basically insufficient to make any claims. Instead of 10 different treatments (control + 3x Roundup concentrations + 3x different GMO “doses” + 3x GMO + Roundup doses ), they should have gone for 1 concentration of herbicide (the highest) and used 25 rats per treatment (no GMO/R, +GMO, +R, +GMO + R).
- The sort of rats used (Normal Sprague-Dawley) have a tendency to form tumours even in normal conditions — 57-80% of female rats develop tumours in normal conditions (males not so much). (http://cancerres.aacrjournals.org/content/16/3/194.abstract) Tumour formation is also highly linked to diet. So, the tumour formation of rats fed GMO/herbicide seen in this study is actually within the normal values… except that the study controls tumour rates are much lower than other people’s large-scale experiments…. Hmmm… They must look after their rats really well (or fed them differently?)
- Male rats that were fed GMO corn and herbicide-laced water actually LIVED LONGER than the rats that ate the “healthy non-GMO corn”! In fact, the MORE herbicide, the better! Funny they didn’t mention that at all! But check out the graphs of Figure 1, it’s there, I’m not lying! Time to start adding Roundup to our water perhaps!!! Another crazy conclusion I can make (up) based on looking at the graphs of Figure 1 is that you should never serve GMO corn alone, but always add Roundup to it – this will reduce tumour formation. Yup yup.
- To be honest, I really did not understand their tumour graph (Figure 2). Are they showing the total number of tumours? In that case, control rats have just as much/more tumours in total than the GMO/herbicide-fed rats. But who cares about total number of tumours, shouldn’t they have shown when the rats started getting their first tumour instead? Or number of tumours per rat? Or maybe they’re actually showing the size of the tumour? I don’t know because I don’t think they really described it! And what happens when a rat dies? Is he still included in the tumour count or removed completely? This could change the numbers quite a bit… Or I might just be totally missing something? Confusing for sure. And not just to me (many other scientists wrote in saying they couldn’t make out what the graphs were saying).
- French Society of Toxicologic Pathology calls bullshit on “The authors declare that there are no conflicts of interest” — “Pr Gilles-Eric Séralini being President of the Scientific Board of the CRIIGEN, and the CRIIGEN having been a “major support” of the study, it seems to us that this should have been disclosed.” Don’t tell me the publishers did not know this?! And they let them get away with it? (Just in case you haven’t realized yet,
- The European Food Safety Authority and The German Federal Institute for Risk Assessment (BfR) say that the data does not support the statements of the paper, and find the conclusions not scientifically sound. (But you should also have reached the same conclusions if you read the paper).
- One anti-GMO professor (possibly good friends with the paper’s authors) wrote a Letter to the Editor saying that this paper should be accepted because other papers on GMO that were sponsored by Monsanto/DuPont also used 10 rats only. Shouldn’t it be the other way round? That none of these papers ought to be accepted if they are not statistically sound? Two wrongs don’t make a right! (He did not cite those papers properly so I have not found them to read yet).
- The authors were clearly up to no good when they gave activists and reporters the paper before it was published, but made them sign non-disclosure agreements such that they could not ask any other scientist for a comment. They clearly want to play by their own rules. Oh, and the lead author published an anti-GMO book right after this paper was published.
- On an animal right’s perspective, many scientists questioned the need for running such long experiments when spontaneous tumours already start to occur after a few months in this strain of rats (and you can no longer tell apart the effect of strain vs GMO/R). The longer the experiments, the larger the tumours get, which is just plain cruel (but good publicity for the anti-GMO people). Also, early tumours can be detected by imaging technologies – you don’t need to wait until they are humongous and causing the rats distress. Authors failed to mention how other tests were carried out on the rats (e.g. blood was drawn, but maybe there were hygiene issues leading to death? Cause of death in the rats were never mentioned anywhere!! Leading you to assume it was death by cancer!) The gruesome photos of tumours in GMO/R-fed rats are not scientific, they are merely to rile up people’s emotions. But they conveniently forgot to include photos of the control rats who would also look just as terrible after 2 years. Again, keeping the rats till they have tumours this size is unethical, cruel and merely a publicity stunt! 200 rats were sacrificed for a paper that only sought to support their anti-GMO stance and boost book sales. Can you support that?
OK, I need to make this a list of 11…. It’s probably the most important point, but I’ll keep it short. No discussion at all by the authors about the difference between “GMO causes cancer” and “Transgene X causes cancer”. They just threw a blanket-GMO-term on their study and called GMOs dangerous. Every transgene is different, each of them needs to be studied carefully. A scientist who does not differentiate GMO technology from the genes contained in the GMO are clearly not scientists.
In conclusion, these are the sort of analysis even a non-scientist/pathologist/rat-model/cancer researcher can and should make. Media-hype is usually just that – media hype. But, the media is powerful and can make you believe anything. Even publications in supposedly-reputable journals can make you believe anything, by merely writing whatever they wish, regardless of the data that they publish on the same page. It’s sometimes really easy to just read and believe. It’s better though, to read with a critical eye!