No one has ever been made ill by a GM food
There is no scientific evidence to support this claim
GM proponents claim that people have been eating GM foods in the United States for 16 years without ill effects. But this is an anecdotal, scientifically untenable assertion, as no epidemiological studies to look at GM food effects on the general population have ever been conducted.
Furthermore, there are signs that all is not well with the US food supply. Reports show that food-related illnesses increased two- to ten-fold in the years between 1994 (just before GM food was commercialized) and 1999.66,67 No one knows if there is a link with GM foods because they are not labelled in the US and consumers are not monitored for health effects.
Under the conditions existing in the US, any health effects from a GM food would have to meet very specific and unusual conditions before they would be noticed. They would have to:
- Occur soon after eating a food that was known to be GM – in spite of its not being labelled – so that the consumer could establish a causal correlation between consumption and the harmful effect. Increases in diseases like cancer, which has a long latency period, would not be traceable to a GM food.
- Cause symptoms that are different from common diseases. If GM foods caused a rise in common diseases like allergies or cancer, nobody would know what caused the rise.
- Be dramatic and obvious to the naked eye or to the consumer of the GMO. No one examines a person’s body tissues with a microscope for harm after they eat a GM food. But just this type of examination is needed to give early warning of problems such as pre-cancerous changes.
In addition, health effects would have to be recorded and reported by a centralized body that the public knew about and that could collate data as it came in and identify correlations. Currently, there is no such monitoring body in place anywhere.
Moderate or slow-onset health effects of GM foods could take decades to become apparent through epidemiological studies, just as it took decades for the damaging effects of trans fats (another type of artificial food) to be recognised. Slow-poison effects from trans fats have caused millions of premature deaths across the world.68 To detect important but subtle effects on health, or effects that take time to appear (chronic effects), long-term controlled studies on large populations would be needed.
3.5.1. Two outbreaks of illness linked to GM foods
Two high-profile cases have emerged in which a GM food was suspected of causing illness in people. In both cases, industry and regulators denied that genetic engineering was the cause, but an examination of the evidence gives no such reassurance.
In 1989 in the US, a food supplement, L-tryptophan, produced using GM bacteria, was found to be toxic, killing 37 people and permanently disabling over 1500 others.69,70,71 The resulting disease was named eosinophilia myalgia syndrome (EMS). Symptoms included an overproduction of white blood cells called eosinophils, severe myalgia (muscle pain), and in some cases, paralysis.
The L-tryptophan that affected people was traced back to a single source, a Japanese company called Showa Denko. In July 1990, a study published in the Journal of the American Medical Association mentioned that Showa Denko had introduced a new genetically engineered bacterium, called Strain V, in December 1988, a few months before the main epidemic hit.71
There is an ongoing debate about whether the toxin’s presence in the L-tryptophan was due to genetic engineering or to Showa Denko’s sloppy manufacturing processes. The company had made changes to its carbon filtration purification process before the toxic contaminant was discovered.
However, the authors of a 1990 study on the outbreak published in the New England Journal of Medicine (NEJM) pointed out that blaming a failure in the carbon filtration process leaves unanswered the question of how the toxin got into the product in the first place.72 This was a novel toxin that was not found in other companies’ L-tryptophan products. The authors of the study, which was sponsored by the US Centers for Disease Control, noted that the new GM bacterial strain introduced by the manufacturer before the outbreak “may have produced larger quantities” of the toxin than earlier strains.72
One of the study’s co-authors, Dr Michael Osterholm, an epidemiologist at the Minnesota Department of Health, commented in a press article of August 1990 that the new bacterial strain “was cranked up to make more L-tryptophan and something went wrong. This obviously leads to that whole debate about genetic engineering.”73
Following Osterholm’s comment, a number of press articles appeared voicing doubts about the safety of genetic engineering. The FDA took on the role of exonerating genetic engineering from blame for the EMS epidemic. An article in Science magazine quoted FDA official Sam Page as saying that Osterholm was “propagating hysteria”. Tellingly, Page added, “The whole question: Is there any relation to genetic engineering? is premature – especially given the impact on the industry”74 (our emphasis).
Osterholm countered: “Anyone who looks at the data comes to the same conclusion [that there may be a link with genetic engineering]… I think FDA doesn’t want it to be so because of the implications for the agency.”74
James Maryanski, FDA biotech policy coordinator, blamed the EMS epidemic on Showa Denko’s changes to the purification process.75 Maryanski also said that genetic engineering could not have been solely or even chiefly responsible for EMS because cases of the illness had been reported for several years before Showa Denko introduced its genetically engineered bacterial Strain V in December 1988.76
However, a study published in 1994 shows that this argument is misleading. Showa Denko had named its bacterial strain “V” because there had been four previous strains of the bacterium. Over a period of years, Showa Denko had progressively introduced more genetic modifications into the bacteria used in its manufacturing process. It began using Strain V in December 1988, shortly before the EMS main outbreak in 1989.69 But it had begun using its first genetically modified strain, Strain II, in 1984, according to lawyers who took on the cases of EMS sufferers.77 This timescale means that Showa Denko’s genetically engineered bacteria could have been responsible for the EMS epidemic.
The FDA responded to the crisis by claiming that all L-tryptophan was dangerous and temporarily banning all L-tryptophan from sale.78 But a study sponsored by the Centers for Disease Control said if that were true, then “all tryptophan products of equal dose produced from different companies should have had the same [effect]”. The study concluded that this was not the case, since out of six manufacturers of L-tryptophan, only Showa Denko’s product was clearly associated with illness.79
If Showa Denko’s L-tryptophan were produced today, it would have to be assessed for safety, since it was derived from GM bacteria. However, since this L-tryptophan was greater than 99% pure and devoid of DNA, it would be passed as substantially equivalent to the same substance obtained from non-GM organisms. In other words, the tests that would be required to detect novel toxins of this type would be seen as unnecessary and no labelling would be required. So the same tragedy would result.80
In 2000 in the US, people reported allergic reactions, some of them severe, to maize (corn) products. A GM Bt maize called StarLink was found to have contaminated the food supply. Regulators had allowed StarLink to be grown for animal feed and industrial use but had not approved it for human food because of suspicions that the Bt insecticidal protein it contained, known as Cry9C, might cause allergic reactions.
The number of people who reported allergic reactions to maize products is not known because there was no centralized reporting system. The US Food and Drug Administration (FDA) analyzed reports that had reached it and asked the US Centers for Disease Control (CDC) to investigate just 28 cases that met its criteria. CDC carried out tests on blood serum taken from these people but concluded that the findings did not provide evidence that the allergic reactions were associated with the Cry9C protein.81
However, there were problems with the CDC investigation, many of which were identified by the researchers themselves. For example, the control group of serum was obtained from blood samples taken before the 1996 release of StarLink. Yet this serum showed a more dramatic allergic response to Cry9C than the serum from people who had reported allergic reactions to StarLink.81 The researchers stated that this is common in samples that have been frozen and stored, as the control samples had been. But they expressed no concern that this would skew the results towards a false conclusion of no effect from StarLink. Neither did they replace the problem control samples with more reliable ones – for example, samples freshly taken from people who were unlikely to have been exposed to StarLink.
CDC’s test and findings were reviewed by a panel convened by the US Environmental Protection Agency (EPA) – which criticised them on several grounds. The panel pointed out that the CDC researchers had isolated the Cry9C protein from E. coli bacteria rather than from StarLink maize. So the protein tested would have been different from the Cry9C protein suspected of causing allergic reactions.82 Specifically, the Cry9C protein from E. coli bacteria would have lacked sugar molecules, which would have been attached through a process called glycosylation to the same protein derived from maize. Glycosylation can be crucial in eliciting an allergic reaction. CDC’s use of the incorrect protein invalidates its analysis and conclusions.
The seriousness of CDC’s error in using E. coli- rather than maize-derived Cry9C protein is graphically illustrated by the study on GM peas containing an insecticidal protein from beans (see 3.1.1).4 The study found marked changes in the pattern of sugar molecules on the insecticidal protein expressed in the GM peas, as compared with its native form in beans. The authors concluded that this change in the nature and structure of the sugar molecules was the reason why the GM insecticidal protein caused immune and allergic-type inflammation reactions in mice.
This case shows that it is necessary to derive the GM protein being studied from the GM crop rather than an unrelated source, as sugar molecule patterns will differ and the potential to cause immune and allergic reactions could vary significantly between the two.
Furthermore, the EPA panel criticised the CDC’s test for its lack of proper controls. It also questioned the methodology and sensitivity of the test used. The EPA panel concluded, “The test, as conducted, does not eliminate StarLink Cry9C protein as a potential cause of allergic symptoms”. The panel’s verdict was that there is a “medium likelihood” that the Cry9C protein is an allergen.82
Claims that no one has been made ill by a GM crop or food are scientifically unjustifiable, since no epidemiological studies have been carried out. However, the cases of L-trypophan produced with GM bacteria and StarLink maize give cause for concern.