by Marion Nestle
Jan 13 2010

GM corn causes organ problems in rats?

French investigators have published a reinterpretation of some feeding studies in small samples of rats.  The studies were done originally by Monsanto to test three varieties of the company’s genetically modified corn.  These investigators obtained the data from the feeding trials as the result of a court case in Europe, which Monsanto lost.   They analyzed the data using their own statistical methods.

I found the paper extremely difficult to read, in part because it is written in exceptionally dense and opaque language, and in part because it presents the data in especially complicated tables and figures.  I must confess to giving up trying to make sense of it and will simply present its conclusion:

our data strongly suggests that these GM maize varieties induce a state of hepatorenal toxicity. This can be due to the new pesticides (herbicide or insecticide) present specifically in each type of GM maize, although unintended metabolic effects due to the mutagenic properties of the GM transformation process cannot be excluded…All three GM maize varieties contain a distinctly different pesticide residue associated with their particular GM event (glyphosate and AMPA in NK 603, modified Cry1Ab in MON 810, modified Cry3Bb1 in MON 863). These substances have never before been an integral part of the human or animal diet and therefore their health consequences for those who consume them, especially over long time periods are currently unknown. Furthermore, any side effect linked to the GM event will be unique in each case as the site of transgene insertion and the spectrum of genome wide mutations will differ between the three modified maize types.

And here is Monsanto’s response.  I would be most intererested to hear the opinion of animal toxicologists on these studies.

  • Pingback: uberVU - social comments()

  • I’d also like to see some unbiased feedback from not only animal toxicologists, but statisticians. In my quick read, the strongest repudiation of the Monsanto tests looks to me to be the poor sample design (low sample size, confounding controls). Regarding the widely reported health effects, I read elsewhere that Monsanto contends (among other things) that statistical significance doesn’t equal biological significance. That is very true, and some of the differences between GMO and control rats seem very small to me, especially given the small sample size. The authors themselves state that the statistically significant results are “signs of toxicity rather than proofs of toxicity” due to various shortcomings of the original study design.

    Larger, longer feeding trials by independent researchers surely seem in order here, which was also the conclusion of the authors. I’m no fan of Monsanto, but the various headlines stating that this paper concludes that GMO corn causes organ failure are inaccurate — the authors do not make that statement.

  • Pingback: Tweets that mention Food Politics » GM corn causes organ problems in rats? --

  • Anthro

    Thank you Julie; you don’t mention your field of study, but you certainly sound informed. So much hysteria has prevailed over GM food. Instinctively, I don’t like the idea, and I find many of Montsanto’s practices abhorrent, but I’m afraid this study will only increase the hysteria, rather than add any clarity. Some people will never believe that cell phones don’t cause brain cancer or that vaccinations don’t cause autism, and I don’t want to join that crowd, but those of us who are scientifically literate would like to know the truth about GM food and whether or not it is harmful to us or the biosphere.

  • I’m an ornithologist/ecologist, and one of my main concerns with this sort of thing is that it warps science. GMO is just one of many very important biological and ecological issues, and the public needs good, accurate information. Media outlets without solid scientific literacy pass on misinformation which just confuses the public and fuels mistrust in science. In cases like this, the science isn’t at fault, but the misinterpretation.

    Kudos to Marion Nestle for requesting clarification and not merely regurgitating (pun intended) the off-base parsing of the paper.

  • I’m no animal toxicologist, although I have a little experience in that area as my first lab was essentially a mouse e-tox lab. But at first glance the paper is confusing, and a little self-contradictory. After aruing that the data isn’t good enough to base conclusions off of, it tries to base conclusions off of it. (?)

    They are choosing a few data points out of hundreds. In one case, the data they used had 500 measurements, and at the 95% significance level, you will have about 25 that come up as significant, on average. That’s approximately how many they had. The distribution of them is pretty random, too.

    And when they apply a more stringent statistical test, called an FDR test (for false discovery rate) many of these ‘significant’ effects become insignificant. In fact for one of the three GE corn varieties, this necessary test eliminates ALL of the significant effects. To quote the paper on the MON 810 section: “However, p-values adjusted for FDR are not significant.”

    I plan on writing a long post about it at, and I will post a link when it is up. Suffice to say, the reports that “organ damage” and “organ failure” (and even Cancer??) have been found are exaggerations of a problematic paper.

  • Janet

    Thanks for pointing out this interesting article. I can’t really comment on the toxicology side, but I do use statistics in my work and do think the article conveniently picks and chooses which statistical method to use. I thought the Monsanto rebuttal had a great analogy with the coin tossing example. It reminds me of the quote, “lies, damn lies, and statistics”. Small differences can be exploited by various techniques to “show” a difference.

  • Emily

    I would like to know what, exactly, hepatorenal toxicity is. Something to do with the kidneys, obviously, but what?

  • Jeff

    After looking at the paper, I have to say that the statistical techniques used here are in fact very, very basic tests. Simple ANOVA with appropriate tests for normality and homoscedacity. Welch test in the case of heteroscadacity, and KW where normality assumptions are violated. Additionally, paired t-tests with an appropriate effects/comparison test depending on violations of assumptions. This is very basic stuff.

    I am not drawing any strong conclusions from this study, but I would refute what others have said above that the statistics were picked or chose arbitrarily. I have performed similar analyses and it’s pretty cut and dried. People have to realize (and scientists should be well aware of this) that statistics are a tool and we are expected to use them honestly. But there is still a large human/subjective component at work, in experimental design, parameter definition, etc., which is just inevitable. We only gain knowledge about the things we measure, but unfortunately cannot gain knowledge about the things we do not measure. This is why our scientific knowledge is dynamic and evolving.

    I do not think this study “distorts science”. It seems like an ok study to me. It’s very simple, and just as good as many others I see. It is unfortunate that some will take this as “proof” of something, when it is far from definitive, but it has merit in my mind. I think the scientists were diligent here in acknowledging the low power they have to work with, and in doing so sufficiently qualified the results.

  • A more important question for me rather than pointing fingers at the possible deficieinces to be found in IJBS’s study, is how did Monsanto get permission to sell these GMO seeds in the US and Europe in the first place, w/o any of those the governments demanding the more exacting and rigorous independent testing before they granted their approval? You know, the kind of testing and rigor that everyone seems to be calling for now?

    For me, the FDA has long since become useless as anyone can see it is little more than a revolving door to ease into prior to those retirement years settling-in for the scientists and management of the corporations they’re supposed to be overseeing. I’ve come to depend upon the Europeans for food safety issues, and now it looks as though I can’t trust them either. Oy……

  • Pingback: The Progressive Mind » Monsanto GM Corn Causing Organ Failure In Rats Study: Everything You Need To Know()

  • Pingback: Monsanto GM Corn Causing Organ Failure In Rats Study: Everything You Need To Know()

  • Pingback: Monsanto GM Corn Causing Organ Failure In Rats Study « Green

  • thanks for addressing this topic! a colleague and I have provided our own response to the GM corn debate on our work blog at Johns Hopkins Center for a Livable Future:

  • Jean-Pierre Zryd

    The above mentioned French investigators had their study financed by Greenpeace and Carrefour (a leading supermarket chain). This study is a part of the strategy of European anti-GMO groups to maintain fear and suspicion among consumers. It works by focusing on the so called “insufficient data” from Monsanto.
    The question is why they didn’t do toxicology experiments using their own rats or mice and feeding them with GMO containing food ! Money was not missing. The answer is probably that they knew from their own experience or results published by other groups that these studies have always produced negative or inconclusive results (that is: no proof of toxicity).
    Extensive discussions on the subject are available on European sites dedicated to public information. It might be interesting for US citizen to have a look on such sites.
    And the European Food Security Agency Rebuttal:

  • Kimball

    I received my doctorate in the field of pesticide toxicity and treatment. I have been a college professor in environmental health, and also have an undergraduate bachelors degree in Natural Resource Management. I just read the previous mentioned study twice. First, the statistical analysis in the study is far from complex and uncommon. Seems rather basic and straightforward to me. I am surprised that a scientist(DR. Nestle) found this study difficult to read. Studies are supposed to be technical and dense. This study does not prove GMO’s are toxic. What it does do is show that in this case liver and kidney organs of laboratory animals were irritated(overtaxed) by these foods. Longer, more comprehensive studies need to be done. Besides looking at organ detoxification and elimination of these foods I would like to see if GMO’s alter telomere integrity in humans. In the future perhaps independent researchers, if granted complete access to Monsanto’s data, will provide more clarity to these issues. What I do take away from this study is that Monsanto is hiding data from interested scientists. This is troubling to say the least. The manufacturer of GMO’s should provide comprehensive indepentent scintific data proving that GMO’s are safe. So far I am unconvinced that they have done that. At this point I would argue that GMO foods be so labeled to allow the consumer to accept or reject these products as they wish.

  • Your Monsanto link does not work.

  • Marion

    @Michelle: they must have taken it down. Sorry.

  • Pingback: Study Links GM Corn to Organ Damage - Food Safety News()