Saturday, December 19, 2009

Help us understand the new study that shows GMOs are toxic

Tom Philpott and Tom Laskawy are reporting on a new study by French researchers, which re-analyzes some Monsanto data and concludes that GMO varieties may be toxic in rats.

Philpott explains the history.
The researchers analyzed data from tests done on rats by Monsanto and another biotech firm, Covance Laboratories, submitted to European government in 2000 and 2001. The firms conducted the tests to prove that their products were safe to eat; scrutinizing the same data, the researchers arrived at a different conclusion.

The three products in question are still quite relevant: one strain of Roundup Ready corn, engineered to withstand Monsanto’s flagship herbicide; and two strands of Bt corn, engineered to contain the insect-killing gene from the BT bacteria. Roundup Ready and Bt products are ubiquitous in the U.S. seed supply, often “stacked” into the same seed.
Laskawy says the earlier Monsanto study fudged the analysis.
Firstly, let's be clear -- industry scientists got bad results, fudged the analysis and then figured no one would notice. Well, it took almost a decade, but these enterprising French scientists did notice.
To see if this is true, I read the new study. The abstract is clear enough, and it does indeed say the GMO varieties were toxic to rats:
Our analysis clearly reveals for the 3 GMOs new side effects linked with GM maize consumption, which were sex- and often dose-dependent. Effects were mostly associated with the kidney and liver, the dietary detoxifying organs, although different between the 3 GMOs. Other effects were also noticed in the heart, adrenal glands, spleen and haematopoietic system. We conclude that these data highlight signs of hepatorenal toxicity, possibly due to the new pesticides specific to each GM corn.
But, I wanted to read the actual empirical results that back this claim. There are dozens of results reported in the study, and I could not understand many of them.

Here is my request for help: Can you point to some particular table entries that support the claim in the abstract?

If you have a particular result to suggest, we can read it closely together, with one dose of open-mindedness and one dose of skepticism. Then, we can decide whether Laskawy is right that the Monsanto researchers have committed a scientific crime.

You might look for such a result in Table 1 and Table 2, which report selected parameters showing the difference between GMO-fed rats and conventional-fed rats. From the abstract, the important parameters might relate to liver or kidney outcomes. All I need is for somebody to pick and explain one or two of those parameters. Personally, I couldn't understand these tables because they seemed to lack clear information about the mean values for the GMO-fed and conventional-fed rats, the column headings were confusing to me, and the accompanying discussion did not include clear interpretation sentences.

Alternatively, you might look for such a result in Tables F or G of the appendix. Here, too, my understanding is confused, because I could not tell if these are raw results, or statistically-adjusted results based on Tables 1 and 2.

Philpott and Laskawy and the French researchers make a striking claim about the dangers of GMO crops. I'd like to understand the empirical evidence that supports this claim.


The Almond Doctor said...

You posted the answer to your concern when you cited the abstract: "We conclude that these data highlight signs of hepatorenal toxicity, possibly due to the new PESTICIDES specific to each GM corn."

Nothing in this paper proves that GMOs are toxic. This is evident by the data - they show the toxic effects are due to the pesticides that are used with these GMO technologies. This is not surprising - any chemical that is used to kill a pest is toxic when the duration of exposure is long enough. Even "organic" alternatives such as copper sulfate can kill a mouse a high enough dose/duration.

Your answer to your confusion is there - you don't even need to read the paper - just read the abstract. GMOS WERE NOT PROVEN TO BE TOXIC IN THIS STUDY.

Matt DiLeo said...

Hmm, so it sounds like they got ahold of some of Monsanto's raw data and re-ran it with "proper" statistical methods. I can't really judge whether this was neccessary without seeing the data (and without a background in how toxicity studies are supposed to be done), but their statistical quibbles seem logical.

The important thing to keep in mind with statistics is it's really a subjective science and it's super easy to make your results come out as either significant/nonsignificant based on the specific methods you employ.

Aside from the nitty gritty of statistics, I'd like to point out two things.

First, glyphosate (Roundup) is less toxic to humans than either caffeine or aspirin. So it's a little silly to worry about trace residues of glyphosate on your veggies if you don't worry about drinking a dozen cups of coffee/tea/soda/chocolate every week.

Second, Bt toxin is routinely sprayed on organic crops as an organic pesticide, and I've never heard any suggestion that this is dangerous.

Knowing these two things, I'm inclined to attribute these results to the authors pushing a little too hard on their statistical tests. And if it's true that the sample sized used IS too small, than no amount of statistical monkeying is going to give a reliable answer.

I'm not personally worried about these GM products because, while no technology is 100.000% safe 100.000% of the time, they are replacing really dangerous pesticides such as atrazine and organophosphates.

If the environmental toxicity scientific community has a problem with the methods used in this study, then the whole experiment should be re-run. If the study was conducted according to proven methods, then I don't see a problem.

Anonymous said...

The inclusion of glyphosate-resistant corn and Bt corn is problematic here. Quoting from the cited study

"One corn (NK 603) has been genetically engineered to tolerate the broad spectrum herbicide Roundup and thus contains residues of this formulation. The two other types of GM maize studied produce two different new insecticides namely modified versions of Cry1Ab (MON 810) and Cry3Bb1 (MON 863) Bacillus thuringiensis-derived proteins. Therefore, all these three GM maize contain novel pesticide residues that will be present in food and feed.

The two Bt hybrids (MON810 and MON863) are engineered to contain a pesticide (Bt) which is available in other forms (eg. to be sprayed on a crop). The other hybrid (NK603) resistant to glyphosate (Roundup). Contrary to the quote above, it does NOT contain "novel pesticide residues," as the trait is for resistance, not production.

The Almond Doctor said...

@ Anon:
You have a good point - the release of the Bt protein into the environment would provide the opportunity to create a pesticide

It just seemed a bit misleading as the title of the blog suggests that all GMOs are toxic. It should limit the claims only to mentioned technologies that produce a the Bt protein.

namnezia said...

OK, look at table 1. The column labeled week is the number of weeks the rats were fed a given diet. Each of the other columns represents four different experimental groups - whether the rats were male or female and whether the GM diet represented 11 or 33% of their daily food intake. The rows are the different parameters measured. The numbers are the percent change of a given measure, relative to a control group fed the wildtype strain of corn in equivalent proportions. If there is a star, then that measure was supposedly statistically different from the control group. Notice that, for example in male rats, levels of urine creatine or sodium and potassium are altered by the GM diet, indicating abnormal kidney function. Likewise the liver may be slightly enlarged.

However, one problem with this is that the data is all over the place. In some cases the effect was significant with 11% but not 33%, or the GM diet has opposite effects depending on dose.

The reality is that the number of rats Monsanto used for their original studies (10 per group) is way too small, which forced the authors to use more exotic statistical measures. And the authors acknowledge this. In my view, I'm not sure that this study makes a particularly compelling case for toxicity. But more importantly, if this is the kind of limited data Monsanto is collecting in order to show that a given GM crop product is safe, then it also casts serious doubt on any claims they are making about food safety.

I don't understand why they simply don't make their seeds available so that independent biomedical researchers (as opposed to subcontractors) can actually test and research the true safety profiles of these products, and give Monsanto the opportunity to improve them.

Anonymous said...
This comment has been removed by a blog administrator.
usfoodpolicy said...

I agree with the comments. Namnezia, in particular, takes up my specific challenge in the post. To summarize, I think the authors should have gone with the story line "Monsanto studies had insufficient sample size to demonstrate the safety of GMO crops" rather than the story line Laskawy and Philpott picked up, which is "Monsanto data shows GMOs are unsafe."

Michelle Clay said...

About MAT’s comments on glyphosate being "less toxic to humans than either caffeine or aspirin": aspirin and caffeine are both almost entirely avoidable, being both reasonably well-labeled and limited to certain products. And as it should be; certain segments of the population are warned by their doctors not to take either. GM corn and soybeans are staples, however, and are in most packaged foodstuffs. They would be nearly unavoidable even if they were labeled.

Is it really the case that Monsanto used only 10 mice per test? That sounds woefully inadequate.

Thanks for the great discussion everyone.