Recently an article was published that caused an uproar around the world and in Israel in which it was claimed that genetically modified corn and the herbicide Roundup are carcinogenic. Why do many researchers cast serious doubts on it and how does it relate to politics and sales promotion? The explanations are inside
By: Noam Levithan and Yonat Ashchar

A few months ago, rumors about the harms of genetic engineering began again in the world, and in Israel as well. This time the blame was placed on genetically modified corn due to a study that claimed that corn supposedly causes the appearance of cancerous tumors. Some of the world's countries, such as Russia and France, were shocked and immediately banned the import and use of genetically modified corn. They did this without checking the quality of the research and what is behind its publication.
Genetic engineering, and especially genetically modified food (GMO), are controversial issues in the world. On the one hand, transgenic plants - that is, plants whose properties have been changed by introducing genes from another source or due to the modification of existing genes - make it possible to increase the yield in the fields and the nutritional value of the food.
A famous example is "golden rice", rice that has been injected with genes responsible for the production of beta-carotene from narcissus and a soil bacterium. Beta-carotene, which gives golden rice its color, turns into vitamin A in our bodies, and thus the genetically engineered rice is able to easily (and cheaply) provide vitamin A to the millions of people living in poor areas and suffering from severe vitamin deficiency.
The fear of genetically modified food
The opponents of genetically modified food claim that it may harm the health of consumers and cause environmental damage. Part of the resistance stems from unfounded fear, stemming from a lack of knowledge or irrational arguments that food is "unnatural," and that genetic engineering is man's defiance of God. However, some concerns are reasonable. One of the concerns is that transgenic plants that themselves, following the Hindus, produce insecticides, will also harm innocent insects that are not harmful, or that plants that have been infused with various resistant properties will breed offspring with wild species and pass on these properties to them, and thus the environment will be damaged and the ecological balance will be violated. But studies in which these issues were examined did not find evidence of more environmental damage than that caused by growing "normal" plants.
Opponents also fear, and rightly so, that following the introduction of foreign genes some people will have an allergic reaction to food that normally does not cause them to have this reaction. However, since any food - genetically modified or normal - must pass allergen tests before being marketed, genetically modified food that causes an increased allergic reaction will not be approved and will not endanger consumers.
Another concern is that eating genetically modified food is dangerous in any case and causes various diseases, but many studies - most of which were done by independent scientists, who are not funded by companies that market genetically modified plants - have shown that there is no truth in this. In these studies, they examined the long-term effect, even over several generations, of a diet based on genetically modified plants, such as corn, soy or rice, on the health of various animals.
No significant difference was found between them and animals fed "normal" food. It should be noted that genetically modified foods are different from each other and are not one piece. There is no biological logic in thinking that a plant will be dangerous just because it is genetically modified, and each case should be checked, and indeed checked.
Despite the abundant evidence for the safety of genetically modified food, many green organizations, led by Greenpeace, are fighting a vicious war against the distribution and use of such food, apparently out of ignorance (or out of excessive and unreasonable fear). As a result of the activities of these organizations, the golden rice - which can save hundreds of thousands of children from blindness and death - is not distributed or grown in the countries that need it, as in the past the technology of the Green Revolution was prevented from being brought to fight hunger in Africa.
Trial and error
But, maybe the green organizations are right after all? Gilles-Eric Serlini (Séralini) from the University of Caen (Caen) in France and his colleagues claim, in a study published in September 2012 in the journal Food and Chemical Toxicology, that in rats that ate throughout their lives (about two years) genetically modified corn from the Monsanto company or were exposed to the herbicide " Roundup" (Roundup - Monsanto's trade name for Glyphosate) Benign tumors were discovered and cancerous.
The researchers divided rats into several groups. Three groups were fed food containing different concentrations of the genetically engineered corn NK603, Monsanto's corn that has been given resistance to Roundup: one group received food that contained 11% of the genetically modified corn, a second group received a higher dose of the corn (22%) and 33% From the food of the third group was this corn. Since it is customary to spray the field with Roundup when growing corn, Serlini and his colleagues fed three additional groups of rats with different doses of the corn, but the corn that was added to the food was sprayed with the herbicide.
Three other groups were fed regular rat chow, but the researchers added different doses of Roundup to their drinking water. Another group served as a control group. The rats in this group drank regular water and were fed rat food, 33% of which was regular non-GM corn. All the groups were separated into males and females so that finally 20 groups were obtained - two control groups and 18 experimental groups.
The authors of the article report that in some of the experimental groups, that is, in some of the rats that ate the genetically engineered corn and/or were exposed to the herbicide, more hormonal problems, tumors and other damage to the organs appeared than in the control group. Also, up to 50% of the males and up to 70% of the females died prematurely in some of the experimental groups, compared to 30% and 20% in the control group.
On the face of it, the results of the study (which contradict the results of previous studies on the safety of Roundup and genetically modified food) should greatly increase the fear of genetically modified food (or at least of the particular corn tested) and turn on red lights regarding the marketing and growing of such food. Indeed, with the publication of the results, many panic-inducing articles about the study appeared in the newspapers and across the web, only a few hours later a flood of reactions and criticisms of the article and the initial articles began.
Problems in research
Why was the initial media coverage biased in favor of the researchers and not critical of the study? Usually, journalists interested in reporting on new scientific research receive the article in their hands even before it is published publicly. In this way, they are given the opportunity to present the article to other scientists, who were not involved in its writing, and to receive different opinions on its value and the strength of the conclusions presented in it. But Serlini and his colleagues refused to allow the journalists to see the article (before its publication) without signing a confidentiality agreement preventing them from talking about the study and before showing it to researchers who are not involved in the study itself. In this way the researchers ensured the creation of media noise and an initial report presenting only their position.
Serlini has been active against genetically modified food since 1997. He worked for years with Greenpeace and other green organizations in an attempt to prevent the cultivation and marketing of genetically modified food and even founded a research institute (CRIIGEN) to help opponents. The research institute, of which Serlini is the chairman of the scientific committee, even financed part of the current research. Despite this, in the part of the scientific article where it is customary to write about any possible conflict of interests of the researchers, such conflicts are not mentioned at all! But even if Serlini appears to be biased, and is trying to use the study to influence French government policy, the results of the study could be correct.
Various researchers who examined the study after its publication, as well as the European Food Safety Authority (EFSA), claimed that it had many problems. The first problem is the number of rats that participated in the experiment. Each of the 18 experimental groups and the two control groups numbered only 10 rats, meaning a total of 200 rats. This is a very small amount of rats, which does not allow to draw conclusions.
Is the death of 30% of the males in the control group, that is, of three rats out of ten, significantly different from the death of five out of ten males in one of the experimental groups? Studies of this type must be based on a much larger number of rats in each group. According to the guidelines of the Organization for Economic Cooperation and Development (OECD), for example, studies in which substances suspected of being carcinogenic are tested, such as this study, require groups of 50 rats each - five times the number of rats tested in Serlini and his colleagues' experiment. Furthermore, the organization recommends that in experiments with the rat strain used in the study, Sprague-Dawley, the number of individuals per group should be even greater (at least 65 rats per group).
Tendency to develop tumors spontaneously
Why was the Sprague-Dawley rat breed given special attention in the OECD guidelines? The strain chosen for Serlini and his colleagues' experiment is a special strain, which tends to die earlier than other strains and also tends to develop tumors spontaneously. This strain is suitable for monitoring the spontaneous development of a tumor or for research unrelated to tumors, such as toxicity testing, but in studies examining suspected carcinogens, its spontaneous tendency to develop tumors (and the type of tumors that develop in it) must be taken into account.
From previous studies it became clear that 72% of females and 86% of males of this species develop tumors spontaneously during their lifetime, even without exposure to a carcinogenic substance or special treatment. Furthermore, Tom Sanders (Sanders), head of the research department in nutritional sciences at King's College in London, pointed out that the diet of the rats affects their tendency to develop tumors: they are more affected by tumors when the food is rich in fat or when its quantity is unlimited. The problem with using rats that have the greatest tendency to develop tumors spontaneously, divided into tiny groups, is that random differences between the groups can easily arise. For example, all ten rats in one group may develop tumors at random.
The agronomist Andrew Kniss demonstrated, with the help of a computer program, how easy it is to get results similar to the results of the article when choosing several small groups of ten "rats" each, and randomly deciding - like flipping a coin - which of the rats will have crops (who got a "tree"). The difference is that in tossing a coin, as mentioned, the chance of getting a "tree" is 50%, while the chance of the rats getting sick spontaneously is 72%.
It seems that the condition of the rats in the control group is better than the rats in most of the other groups, although there seems to be no relationship between the dose to which the rats were exposed and the strength of the reaction in the experimental groups, and that the condition of some of the experimental groups (for example, males exposed to a very high concentration of Roundup) is much better than the control group.
The statistics do not say
When you want to compare different groups, it is customary to perform a statistical analysis of the results to determine whether the differences between the control group and the other groups are statistically significant. That is, we check what is the chance of getting the observed result, or a more different result, at random. The smaller the chance, the more convincing the evidence that the results obtained are real, and not obtained by chance.
The researchers present in the article some complicated (and frankly unnecessary) statistical methods, which are unacceptable to use in experiments of this type, but do not present even one statistical analysis of the differences in mortality or the incidence of tumors between the groups. They do not show at all whether the difference between the control groups and the other groups is significant, and as we have seen, when the groups are small it is very likely to find differences that were created by chance.
David Spiegelhalter, a statistician from the University of Cambridge, claims that the working methods and statistical methods in the study are so flawed that he is surprised the article was accepted for publication. The researchers did not use enough control groups. In a properly planned experiment, there should be a control group corresponding to each of the tested conditions - one for the experiment with the genetically modified corn, a second for the sprayed genetically modified corn, and a third for the experiment with the spraying only. Therefore, Serlini and his colleagues' experiment should have included 1,560 rats (six control groups and 18 experimental groups with 65 Sprague-Dawley rats in each group) and not 200 - to be meaningful.
The researchers compared all the different experimental groups to the same control group, all the males to the ten males of the control group and all the females to the ten control females, in which most of the rats also had tumors. When making so many comparisons there is a chance that differences between the groups will be found only due to the hand of chance. This is similar to a person claiming to be able to hit the center of the target board in a game of darts while blindfolded. If he throws arrows at the board enough times, sooner or later he will hit the center, maybe even more than once, even if he is not skilled at all - and thus we may inadvertently conclude that his claim is true.
There are statistical methods designed to reduce the chance of drawing incorrect conclusions following many comparisons, but they are not mentioned in the article by Serlini and his colleagues. In any case, as stated above, it seems that the researchers did not statistically analyze the differences between the groups of rats.
Not all results were displayed
When there are so many groups and comparisons to choose from, another problem may arise. The authors of the study may consciously or unconsciously choose a comparison whose results best fit their world view (a way of incorrect inference known as cherry picking, or the fallacy of incomplete evidence). Serlini and his colleagues acknowledge in the paper that they chose not to present all the results.
They claim that "it is impossible to show all the information in one scientific report, [and] only the most relevant is presented here", although in all scientific journals an unlimited amount of data can be attached as online supplementary data to the article (Supplementary data) and the researchers are even required to publish all the information to make it accessible to other scientists.
A common way to prevent the researchers' unconscious influence on the results is a study using the blinded experiment method, that is, a study in which the researchers do not know which of the groups received which treatment and which is the control group until after the collection of the results, their analysis and the conclusion have been completed. The present study was performed without any blinding.
Carl Zimmer, a writer and science reporter for the New York Times, points out another problem in his blog - the scientific article is edited in a biased way. It shows many shocking photos of rats that ate transgenic corn and/or were exposed to Roundup and were exposed to crops, but also a picture of the control rats that were exposed to crops. These are the same photos that were distributed to the press and it seems that their entire purpose is to create panic among the readers!
Caution: Bad studies

In conclusion, it is not clear why eating genetically modified corn with resistance to Roundup or exposure to the herbicide itself, two completely different factors, would cause the same effects in rats. Especially when similar phenomena were also observed in the control rats. As some critics of the study wrote: All that can be concluded from the results is that rats of a strain prone to developing tumors will develop tumors over time, regardless of the presence of Roundup or genetically modified corn in their food.
Mark Tester, from the Australian Center for Functional Genomics in Plants at the University of Adelaide, mentions that genetically modified food has been eaten for many years. If indeed this corn, or other genetically modified food, is so dangerous, why have epidemiological studies that are constantly being done on large populations of humans not revealed a huge increase in tumors in people and animals that feed on it?
It is true that every invention must be investigated, tested and monitored, but do not panic. Serlini and his colleagues published the current article as part of sales promotion for the book and the film, and also with the aim of scaring the public and thus exerting pressure on the politicians to ban the cultivation and marketing of genetically modified food. They partially succeeded in this in France. Following the media uproar surrounding the study, the French government called for an examination of the results and if it is necessary to ban the marketing of genetically modified corn or genetically modified food in general. The study did not innovate anything about the safety of genetically modified foods and Roundup, but it certainly showed that one should beware of bad research done with political motives - and this is true on both sides of the fence.
9 תגובות
spring
You are against genetically modified food just because you are against it.
Be sure that genetically modified food goes through much more safety tests than normal food.
Critical scientific thinking must be taught as a compulsory subject in high school.
Today every child studies in middle school and at the beginning of high school a little chemistry, a little physics and a little biology.
Instead of focusing on small, not very important issues that in any case most children will forget (especially those who will not continue with a scientific major) you should focus on scientific criticality.
As a society we really need scientific knowledge for everyone. Absolutely not, the majority will not be scientists and that's fine
But it is very important that they know how to read research and understand if it is serious. This has real consequences for our ability to survive as a society.
It is not so important for a child to know the blood circulation and how many chambers there are in the heart or what elements sugar is made of. The one that will be taught by those who expand a scientific trend.
But critical scientific thinking, what is statistical significance, standard deviation and sampling error, ways to identify biased research,
This is important for anyone who expands literature to know.
The UN report lists several causes of world hunger, economic monopoly is not one of them and the source that does talk about such a monopoly is the unreliable website corpwatch.
What excuses? I have seen reasoned explanations.
Jonathan I read the lecture.
The argument that there is not enough food and therefore we need genetically modified food is not true,
There is no problem of food production in the world according to the findings of the UN Food Program:
http://www.wfp.org/hunger/causes
The reason is an economic monopoly on food:
We now believe that Monsanto has control over as much as 90 percent of (seed genetics). This level of control is almost unbelievable," said Neil Harl, agricultural economist at Iowa State University who has studied the seed industry for decades. "The upshot of that is that it`s tightening Monsanto`s control, and makes it possible for them to increase their prices long term. And we've seen this happening the last five years, and the end is not in sight."
At issue is how much power one company can have over seeds, the foundation of the world's food supply. Without stiff competition, Monsanto could raise its seed prices at will, which in turn could raise the cost of everything from animal feed to wheat bread and cookies.
http://www.corpwatch.org/article.php?id=15489
The excuses regarding the French study are the excuses of the tobacco industry at the time.
As soon as they extended the study to over 90 days and turned it into a cancer study instead of a toxicity study, it is no longer possible to claim that "the French study used exactly the same protocol as the Monsanto study." As written in the "doubtful" article, research examining whether any factor is cancerous requires a larger number of rats and it is recommended not of the kind that develop tumors spontaneously and easily.
In addition to the argument that humans and animals have been eating genetically modified foods for years, a number of experiments were conducted that tested the carcinogenic effects of eating genetically modified food throughout the life of the animal being tested and even over several generations. These studies did not reveal any danger.
There is no need for targeted suppression from the industry to dismiss Serlini's research, you just need to learn critical thinking, how to read scientific studies and be ready to change your mind according to the new information received (for example: http://www.marklynas.org/2013/01/lecture-to-oxford-farming-conference-3-january-2013/ )
A slightly different version of "Doubtful" with links to studies and sources of information:
http://lokissciphile.wordpress.com/2012/11/11/gmo-maize-causes-cancer-hype/
By the way, the corn in the picture is not genetically modified. These are cobs of different varieties of corn.
(sarcasm of course)
The corn in the picture is really appetizing.
For those interested in reading in detail the other side of the story
http://www.globalresearch.ca/stench-of-eu-corruption-in-monsanto-gmo-whitewash/5316294
The French study used exactly the same protocol as the Monsanto study relied upon by the authorities.
Here they extended the study from 90 days to two years and added more parameters for testing.
The argument that:
'Remembers that genetically modified food has been eaten for many years. If indeed this corn, or other genetically modified food, is so dangerous, why have epidemiological studies that are constantly being done on large populations of humans not revealed a huge increase in tumors in people and animals that feed on it?' Just absurd.
The lifespan of a rat is two years. (also the length of the study)
How many years have humans been eating genetically modified corn? 15 years maximum. We have not yet seen the results of genetic food in humans.
The farm animals that feed on it are not enough to reach a fifth of their natural age (20). Cows for slaughter are killed at the age of 2
and dairy cows at the age of 4
There is targeted thwarting from the industry here, just as they did to the Russian researcher and the research on genetically modified soybeans.
It is also better to mention the war in court to reveal the results of research on genetically modified potatoes (again cancer)