I just read through the
paper. While certainly eyebrow raising, I see some issues in the study design. I don't think they used proper controls.
Basically, they made rodent chow with either 33% regular corn, 33% GMO corn, 22% GMO corn, or 11% GMO corn. I think they should have either supplemented the rat chow with regular corn to add up to 33% total corn, or used rat chow with 22% and 11% regular corn as additional controls. The way it is in the present study, there's a bit of apples to oranges comparison with 33% regular corn versus 22% and 11% GMO corn. So if it's something in all corn causing an effect, you won't be able to tell if it's the GMO or not.
Looking at the results (Fig 1), this seems to be a problem, as the LEAST amount of GMO (11%) in the males got the slowest mortality rate - less than the control group. If it's really something in the GMO, you would expect to see a dose dependent response. Looking at the death curves, I'm not convinced.
Also, 43% of these rats get tumors in the control / regular corn group (Table 2). In the GMO fed rats, it only goes to 66%. In the GMO+Roundup rats it goes to 62%. That's a difference,but not huge. Edit: D'oh. I take that back. I can't make that analysis, because the table is broken down to # of tumors, with how many affected rats. There could be rats that have several tumors in different organs, and they'd be counted at least twice.
A previous study this group did in 2009 looked at other Monsanto GMO corns - including 2 that produce their own insecticide (Bt toxin based). I'd like to see the results of those in a similar study. Unlike sprayed on insecticides, I don't think you can wash this stuff off.
I'd like to see this type of study reproduced with better controls. And definitely done in non-human primates, as this effect may very well be rodent specific, or even rat strain specific.