EU regulators unimpressed by genetically modified crop study
Ars Technica » Scientific Method 2012-10-23
In late September, the University of Caen's Gilles-Eric Séralini released a paper that claimed to show an increased incidence of tumors in rats fed either an herbicide or corn engineered to resist the herbicide. At the time, he used an unusual agreement to prevent outside experts from commenting on his work. Once they saw it, the study was generally panned.
Now, both the European Food Safety Authority and Germany’s Federal Institute for Risk Assessment have weighed in on the matter, and both find the study inadequate. Both cite the use of tumor-prone rats in a long-term study that focused on tumor incidence, and the insufficient size of the experimental and control groups. The EU group also highlights the lack of standard statistical approaches to analyzing the results. The Germans aren't convinced by the authors' explanation for the fact that the supposed toxins showed no dose effect—or increased susceptibility to tumors with increased GMO-consumption—concluding "this hypothesis is not sufficiently supported by the data presented."
They're unlikely to find out whether Séralini's data, presented or otherwise, support it. That's because Séralini is refusing to release any of his original data until those organizations release all their data on their original approval of the crop.