The study on GM corn that rouen refers to has serious flaws (something rouen also points out), including the fact that they released the study only to journalists who signed a confidentiality agreement, an agreement which prohibited them from asking other scientists about it until the article was officially released. The problem with this is that the study makes news (GM Corn Causes Cancer OMG) and the rebuttal, which can't happen till other qualified people can read the article, does not.
You can read about the study's flaws
here, read about the
secrecy hubbub here, and
listen to a story about it here. [That last link is to On The Media, an NPR program which probes topics in the media (journalism, media manipulation, etc.) and is generally pretty eye-opening and interesting.]
Anyway the crux of the criticism is that the study had too few participants (10 rats per group) and the rats used were of a strain known to spontaneously develop cancer! Some of the 'control' (untreated) rats died during the study, too. So while GM corn may contribute to cancer, no one knows that because this study is so flawed as to give no good information. Also, it is super fishy that the research group required a confidentiality agreement -- it looks like they knew there was a stink to their study, and preferred to manipulate the story rather than do proper research. This is an important topic and one that we'd love to have more info on, so to waste people's time and money this way is not cool. IMO.