I've actually been really surprised at the number of papers there are in pubmed when I've look up references and sources. The GENERA project looks at peer-reviewed studies to assess the risk of GMOs and their database has 600 papers, 1/3 of which are independently funded. A quick pubmed search revealed that there are over 130 papers on MON810 (that's Monsanto's Bt corn); of course, not all of them were about health: a large number of papers were environmental studies, looking at the impact to soil, bugs, etc, and a fair number were methods papers describing ways to identify the GMO based on DNA or protein.
In trying to learn about the long-term health impact of GMOs, I didn't know where to start, so I decided to read a 2012 literature review that examined long-term and multigenerational impact of GMOs to health. If you have access to the paper, I suggest you take a look. The review has a helpful table, where each study is listed together with the source of funding, # of animals in the study, parameters that were examined, conclusions, and criticisms, among many other things. The long term studies, defined as any study greater than 96 days, ranged from 26 weeks to 104 weeks, used various model organisms (including cows) and the authors examined 12 of them. Some studies did not meet international standards (ironically, two of these were reviewed last week where I stated that they were indeed horrible papers). The most common criticism for several of these studies is that they didn't use isogenic lines as controls. An isogenic plant is the genetic equivalent to the GM variety, but without the GM trait. Why is this important? Well, take the corn that popcorn kernels are derived from vs the sweet delicious corn that you eat with butter in the middle of summer. As learned in my "Plants as Human Resources" course in my sophomore year, the two have different amounts of starch and sugar. So if you were to make a GM-variety out of the sweet corn, but use popcorn as your control, then the study will not be accurate. You won't know if your results are a consequence of the GM trait or if it's because of nutritional differences.
Anyway, getting back to the review: papers that met International Standards generally concluded that feeding GMOs to animals was safe and there were no long term effects. After looking at all the long-term studies, the authors conclude "the available long-term studies do not yield new safety concerns and confirm that the studied GM varieties ... are nutritionally equivalent to their non-GM conventional counterparts)."
Moving on to the multigenerational studies, these also used a variety of different model organisms, and the authors again looked at 12 papers. These ranged from 2 to 10 generations, and goal of all the papers was to determine if feeding GM plants to one generation would have an effect on the next. Several papers suffered from the same issues regarding controls. Two papers observed unexplained differences (one saw a difference in lactic dehydrogenase enzyme activity, which may affect metabolism; a second found changes in the immune system). The review concludes "overall, the multigenerational studies on animals fed GM plants do not reveal signs of toxicity or other macroscopic effects on health ... The relevance of the observed differences in some of the parameters is not known and may reflect some natural variations. The authors suggest that additional multigenerational studies should be done in order to study the reproducibility of these results and to try to find the true causes of the detected changes." The authors go on to repeat that several studies had serious flaws so "the data cannot be interpreted in terms of toxicological effects".
The authors' discussion had a few interesting points:
- They point out that there are norms and standards for feeding studies, and that all these studies (which were publicly funded for the most part) just have to stick to the rules in order to be successful.
- They point out that long term studies didn't discover anything new. They suggest that long term studies should be carried out only if there's something odd or alarming in standard 90 days feeding studies (I'm not sure I agree with this, but it's an interesting argument. I think that studies longer than 90 days add value to the debate regarding GMOs and should be conducted at this point in time).
- The authors point out that their critical examination found that studies where changes in some parameter were identified did not follow the required standard protocols.
- That due to "recurrent lack of compliance with international standards of many studies", the private sector is now unwilling to provide plant material for studies (it's no wonder after scenarios like the whole Seralini publication disaster). They highlight that without some level of cooperation, scientists may never be able to get the appropriate controls for their studies.
- The authors suggest that protocols and methods should be harmonized and standardized. They highlight that no two studies in their review were conducted the same way. Each study used a different organism, different duration for the study, different parameters, etc. Having a standard experimental design would help control for variability. I couldn't agree more.
But let's get back to the crux of the matter. The review highlights the fact that, other than the poorly designed studies, no paper has found significant differences between GM and non-GM. So... when is the evidence sufficient enough to be conclusive? How long does a study have to be or how many generations need to be examined? That's a great question. There seem to be quite a few papers published where no difference was found between GM and non-GM food. In fact, if I were running my own lab, I'd be pretty unwilling to perform another study because a) animal studies are expensive, particularly large mammals and b) odds are, I wouldn't find anything and the absence of interesting data generally does not lead to high impact/sexy publications. So given the fact that getting grant money is pretty difficult and using private sector funding would label my study as being biased, what options are there? I think collaboration between public and private sectors may be needed, and perhaps oversight is required to ensure that studies remain impartial.
In investigating this topic, I think that the statement "long-term studies have not been performed" or "we're the guinea-pigs for the GMO industry" are false and misleading. However, I do think that norms and standards need to be universally adopted for these studies, and once a well-designed experiment is performed, it's findings need to be accepted. For example, there is a study that examined the impact on milk when feeding Bt corn to cows for 100 weeks. According to this review (note: I haven't read the paper myself), there are no criticisms of the study and it concludes that there are no differences in milk yield and composition when feeding Bt corn to cows. So, do we need another study? Or is it time to start to break out the celebratory chocolate milkshakes made with milk derived from Bt-corn-fed cows? Given the number of 90 day studies that exist examining Bt-corn and the testing standards adopted by the biotech industry, I vote for the latter.