Séralini damage done despite retraction
A paper claiming genetically modified corn caused cancer in rats leaves serious questions over the journal's peer-review process, Elizabeth Finkel writes.
There’s nothing worse than a half-hearted apology. But what emerged from the Journal of Food and Toxicology last Friday, 29 November, was barely even that. A year after publishing a controversial paper that claimed genetically modified (GM) corn and the herbicide Roundup caused cancer in rats, the journal announced they were retracting the paper because of methodological flaws.
“A more in-depth look at the raw data revealed that no definitive conclusions can be reached with this small sample size,” the journal’s publisher, Elsevier, explained in a press release. But the company made a point to say that the editor-in-chief “found no evidence of fraud or intentional misrepresentation of the data.”
The publisher also went out of its way to repeatedly applaud the lead author, Gilles Eric Séralini, for his cooperation during the post-publication review of his paper. “The editor-in-chief again commends the corresponding author for his willingness and openness in participating in this dialogue. The retraction is only on the inconclusiveness of this one paper.”
The mealy-mouthed wording of the retraction leaves many scientists unsatisfied.
The mealy-mouthed wording of the retraction leaves many scientists unsatisfied. “The journal has abrogated their responsibility to do something,” says Rick Roush, Dean of Land and Environment at the University of Melbourne. “There’s been tremendous damage from this paper,” he says alluding to the fact that it has been used to support anti-GM campaigns ranging from food labelling to an attempted ban on imports by the European Union.
“It’s disappointing from my view. It’s quite clear the authors tried to hide some of their results,” says Christopher Preston, a weed expert at the University of Adelaide.
“It was clear from even a superficial reading that this paper was not fit for publication, and in this instance the peer review process did not work properly,” wrote David Spiegelhalter, Winton Professor of the Public Understanding of Risk at the University of Cambridge.
Even before the study was published in November 2012, it smacked of irregular processes. Journalists received advance notice of the paper but in order to get a preview they were required to sign a confidentiality agreement preventing them from sourcing second opinions. As a result the sensational findings of the paper strutted in the limelight for 24 hours before independent scientific opinions started rolling in. When they did, they were scathing. The world’s expert toxicologists and statisticians uniformly declared the paper to contain poorly designed experiments with ridiculous conclusions.
One obvious problem was that the background rate of cancer over the lifetime of the rats was around 80%. Detecting a signal above that noise would have required an enormous number of rats and expertly applied statistics. The authors did neither. Instead, they cherry-picked the data. For instance when the data showed that female rats actually had fewer cancers on a diet of 33% GM corn than on a diet of 11% GM corn, they argued for an interesting dosage effect. The fact that males showed no differences between control and treated, suggested a sex-hormone effect.
“The authors were trying to make the data fit a conclusion,” says Preston.
The immense scepticism towards the paper’s findings snowballed. A petition signed by 700 scientists called for Séralini to surrender all the raw data for his experiments or for there to be a retraction. (He would eventually comply with that request from the journal’s editor-in-chief.) The journal also published 16 critical letters to the editor.
Additionally, the world’s major regulatory bodies conducted reviews and repudiated the findings of the paper. One Australian scientist, who asked to remain anonymous, even extracted the data from one of the figures on the graph, and carried out his own statistical analysis. Finding it did not support the author’s conclusions, he submitted his own findings as a paper for review to the same journal, which it declined to publish.
Séralini’s study reverberated around the world with serious consequences. The Russians responded by banning imports of GM corn. The French Prime Minister, Jean-Marc Ayrault, announced he might try to have GM corn banned throughout the European Union, reinforcing France’s record of anti-GM politics.According to some, this was part of a trade-off with the Greens in return for a truce on nuclear power.
Despite the wide-ranging barrage of criticism against the paper, until last week it was never clear that the journal intended to act. The press release explained: “This retraction comes after a thorough and time-consuming analysis of the published article and the data it reports, along with an investigation into the peer-review behind the article.” And they are self-congratulatory. “The peer-review process is not perfect, but it does work.” The only note of contrition is right at the end. “The editorial board will continue to use this case as a reminder to be as diligent as possible in the peer-review process.”
The retraction has brought a sigh of relief to some scientists but the tortuous rationale for it still leaves many looking askance at the journal and its process of peer review.
“Many of us are wondering what else is going on,” says Roush. There are suggestions of law suits. As far as intentional misrepresentation, Roush says, “it looks pretty much like that to most of us, they really had to go through statistical contortions to draw the conclusions they did”.