A bad day at the lab for GM reserach

During this summer, I recall reading in the Sunday Times that the environmental NGOs are beginning to re-think their strategy on GM foods. I see some evidence that this is the case since I cannot find any mention of the following paper on any of their websites: Séralini et al (2012) “Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize”[1]. The publication of this paper has led to the greatest backlash by the scientific community that I have seen in 4 decades in this business. Essentially, Séralini published a paper showing that rats exposed to a GM food (maize) and a herbicide (which is used with the resistant GM crop) developed breast tumors significantly faster and to a greater extent than controls rats over 104 weeks (2 years). The most significant critic is the European Food Safety Authority (EFSA), which is the independent body charged with protecting consumer health in the EU and which is the judge, on the consumers’ behalf, of all scientific publications on food safety including those on GM technology.

The authors used Sprague-Dawley rats that will naturally develop tumors over their lifetime irrespective of any dietary or other treatment and the authors did not discuss the implications of this natural tendency to tumor development for their study. They also used 10 rats per treatment, which according to OECD protocols is adequate for standard 90-day toxicity studies. Monsanto’s submission to EFSA on the GM maize (NK603) used only 10 rats per treatment, but it was for a 90-day toxicity study. However, Séralini’s study was over a “lifetime” and the OECD guidelines recommend the numbers now be increased to 20 per treatment for chemical toxicity tests but that for carcinogenicity studies, this should be increased to 50 per treatment. In an article on this topic, Nature contacted Harlan Laboratories who supplied the rats and were told that for this strain of rat, only 33% of males and 50% of females live to 2 years. According to the OECD protocols, if a study is to last 104 weeks, then the survival rate should be 50% at least and that then 130 rats (half male half female) should be used per treatment.

The lead author apparently agrees that more rats per treatment would have boosted his statistical power but according to Nature[2], he argues that he did not design the study to find tumors.  If at this stage you are confused, then you’re normal!!!

Further criticism from EFSA includes the fact that no information is given on the composition of the rat diets and that no data is given on how much of the herbicide was consumed through its route, drinking water. No data are given on lesions that were found which were not tumors or dropout rates and reasons for dropouts. In addition, the EFSA working group state that the statistical techniques used were not “commonly-used statistical methods” and that the authors do not state whether the unusual statistical techniques they used were, in fact, the a priori choice and if so, why so? Finally EFSA requested the basic data from the authors to examine these shortcomings and they were refused access. Trust is hard won but easily lost. 

If all that wasn’t bad enough, Nature reports on a very sinister dimension to this saga, which has not received widespread attention. According to their correspondent, Declan Butler, the author orchestrated a very tight media offensive that included a film and his new book (Tous Cobayes: OGM, Pesticides, Produits Chimique: All Guinea Pigs, GMOs, pesticides and chemicals) on the work. A select group of journalists were invited (not from Nature) to preview the paper and were asked to sign a confidentiality agreement demanding total secrecy until formal publication. A breach of the terms of the confidentiality agreement would require, according to Nature, the following: “A refund of the cost of the study of several million euros would be considered damages if the premature disclosure questioned the release of the study”. I’m in the wrong business I believe!!!!

The Ethics Committee of one of France’s most august academic bodies Centre National de la Recherche Scientifique described the PR exercise as “inappropriate”.

Who’d like to be the first to write a review of his new book on Amazon[3]? Well although I would, it would be so slanderous that I could not ever afford the libel fee I’d have to pay.

---

[1] Séralini GE, Clair E, Mesnage R, Gress S, Defarge N, Malatesta M, Hennequin D, de Vendômois JS. Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Food Chem Toxicol. 2012 Nov;50(11):4221-31

[2] http://www.nature.com.eproxy.ucd.ie/news/hyped-gm-maize-study-faces-growing-scrutiny-1.11566

[3] http://www.amazon.co.uk/Tous-cobayes-pesticides-produits-chimiques/dp/2081262363/ref=sr_1_1?s=books&ie=UTF8&qid=1350944840&sr=1-1