GM Feed Toxic, New Meta-Analysis Confirms

A meta-analysis on 19 studies confirms kidney and liver toxicity in rats and mice fed on GM soybean and maize, representing more than 80 percent of all commercially available GM food; it also exposes gross inadequacies of current risk assessment Dr Eva Sirinathsinghji

A team of independent scientists led by Gilles-Eric Séralini at Caen University in France carried out a meta-analysis combining the results of 19 previous studies [1], and their report concluded: “From the regulatory tests performed today, it is unacceptable to submit 500 million Europeans and several billions of consumers worldwide to the new pesticide GM-derived foods or feed, this being done without more controls (if any) than the only 3-month-long toxicological tests and using only one mammalian species, especially since there is growing evidence of concern.”

Multiple organ abnormalities revealed on re-analysis

The nineteen feeding studies performed to date were performed by both industry and independent scientists on either Bt maize or RR soybean, which constitute 83 percent of commercially available GM food. The Bt maize varieties all contain a specific pesticidal protein from the soil bacterium Bt (Bacillus thuringiensis), one variety was also glufosinate herbicide tolerant; the RR soybean is tolerant to Roundup Ready (glyphosate) herbicide. The data were re-analysed with new biological and statistical methods, including the meta-analysis. Meta-analyses allow a more objective appraisal of the evidence and provide a more precise estimate of a treatment effect, giving greater statistical power, and reducing the significance of false-positive or false-negative results.

Although none of the findings are new, the meta-analysis gives further strength to the previous evidence. Importantly, it found that nine percent of tested parameters were disrupted, which is almost double the five percent that could be obtained by chance alone.

Forty-three percent of significant abnormalities were found in the kidneys of males. The liver was more affected in females and the kidney was more affected in males.

Kidney pathology in animals fed RR soybean included significant ionic disturbances resulting from renal leakage. This is consistent with results from cell cultures treated with glyphosate [2] (see [3] Death by multiple poisoning, glyphosate and Roundup, SiS 42), suggesting that the presence of the herbicide in the GM food was responsible.

Rats fed Bt maize had significantly decreased kidney size as well as focal inflammation. This was acknowledged by the European Food Safety Authority (EFSA) even though they went on to approve the products.

Liver pathology of animals fed RR soybean included the development of irregular hepatocyte nuclei, more nuclear pores, numerous small fibrillar centres, and abundant dense fibrillar components, indicating increased metabolic rates. These features were consistent with previous findings in cell cultures treated with herbicides [4].

GM maize-fed animals showed significant abnormal blood protein levels, indicative of disturbed liver metabolism. Histopathological features were also found in some cases. Again, this was acknowledged by the EFSA.

EFSA risk assessment totally inadequate

Current risk assessment of GM foods is based on the ‘substantial equivalence’ concept, where the genetically modified food is deemed equivalent to other products already consumed with regards to components such as fats and oils, carbohydrates and proteins, in which the GMO can be compared, not to the non-GM variety from which it was created, but to an arbitrary combination of conventional varieties or produce. This practice has been thoroughly criticised since the beginning (see [5] The Principle of Substantial equivalence is Unscientific and Arbitary, I-SIS scientific publication).

On that already shaky basis, animal feeding experiments are not always required for regulatory tests, and those that have been performed have been analysed with very dubious methods.

Feeding experiments by Monsanto deeply flawed

While studies carried out by independent scientists all reported significant effects due to GM-feeding, those carried out by Monsanto on MON863, MON810 (both Bt maize lines), and NK603 (glyphosate-tolerant soybean line) reported no evidence of toxicity. The results were kept confidential by Monsanto and the EFSA, until Séralini and his colleagues gained access to the raw data through court action, and found the experiments deeply flawed at every stage, from experimental design to data analysis and interpretation.

Statistical power was greatly reduced by the small number of animals in GM-treated groups, while unmatched groups on 7 different diets were inappropriately included as controls. The findings lack generality as only one species (rat) was used, and tests were performed just once for each GM line. The trials lasted at most 90 days, which is insufficient to pick up chronic effects. No developmental, carcinogenic, reproductive, multi-generational or endocrine parameters were tested. Only two doses of GM foods were used, making it inadequate for detecting dose-dependent effects.

The statistical methods for analysing the data were inadequate, and EFSA had suggested a revision of the methods, highlighting current inadequacies in risk assessment. Statistical comparisons of GM-fed animals to ‘historical’ control groups from previous unreferenced studies were sometimes used instead of control groups from the same study, thereby introducing large variations that hide actual treatment effects. Séralini’s team re-analysed the data comparing treated groups to the closest control group, as is standard scientific practice.

Significant effects were often ignored by Monsanto, and were only taken into account if seen across both sexes. This is unjustified as sex differences are expected for certain pathologies including endocrine-related disturbances, carcinogenesis and liver and kidney abnormalities. As is the case with non-diabetic renal disease, females show protective effects compared with males [6]. Monsanto dismissed differences that were not dose-dependent, but their experimental design precluded the detection of such effects.

Correlations between significant effects were required by Monsanto for accepting disturbances, even though many of them are not expected until long after the beginning of GM-feeding. For example, in the MON863 study, Monsanto noted anatomic signs of “chronic progressive nephropathy” on GM-fed male rats’ kidneys. However, they stated that this was normal in aging rats, even though they were only 5 months old, and these signs were not reported in similar aged rats used for the MON810 and NK603 studies. The animal tissues in question are not available for independent re-analysis as they belong to Monsanto.

There was no assessment of the chemical composition of food. The food was not analysed with regards to herbicides, pesticides or genetic modification, making it impossible to determine whether it was the pesticide/herbicide or the genetic modification that caused the toxicity.

Séralini and his colleagues also suggested that bias of interpretation could be expected as all the studies were performed by the very industry that was hoping to get their product onto the market.

Proposals to improve risk assessment studies for GM foods

With millions of people being subjected to GM foods, Séralini and his colleagues said, more thorough experiments are necessary.

One suggestion is the toxotest approach for chronic environmental, as well as reproductive and developmental effects. They last two years, are multigenerational, and include testing pregnant females, adding information on endocrine, carcinogenic, neural and hormonal dysfunctions. The existing 90-day trials on adult animals cannot match the sensitivity of these developmental tests on neonates. Developmental parameters such as disturbances in genomic imprinting, which determines whether maternal or paternal copy of the gene is expressed, may not be apparent until the second generation. Such abnormalities have been observed with endocrine disrupters such as bisphenol A and estrogenic compounds [7, 8] and are important unanswered questions with regards to GM foods.

Toxotests should also be performed on three animal species (same species as used in pesticide studies), with three doses of GM diet of 11, 22 and 33 percent. These should then be compared to control GM-free diets with equal sample size that are genetically identical, or as similar as possible to the GM lines. To deduce whether the toxic effects are due to herbicides or the genetic modification itself, it would be more informative to feed animals with GM foods both treated and untreated with herbicides. Lastly, statistical experimental design needs to be improved.

Post-market monitoring of GM diets on the human population should be employed to deduce unexpected effects such as allergenicity.  Blood samples could be banked and screened for antibodies against the transgene and its products. For such human studies, the labelling of all GM products is necessary.

Finally, all raw data from industry studies must be made publicly available so they can be objectively scrutinised and analysed. 

To conclude

Current risk-assessment studies are inadequate in detecting and acknowledging the toxicity of GM food consumption. Previous independent studies have clearly indicated the hazards of GM crops to human health, with widespread pathologies including birth defects and spontaneous abortions; (see for example [9] EU regulators Regulators and Monstanto Exposed for Hiding Glyphosate Toxicity, SiS 51) infertility, stunting and sudden deaths (see [10] GM Soya Fed Rats: Stunted, Dead, or Sterile, SiS 33); immune reactions and allergenicity, (see [11] More Illnesses Linked to Bt Crops, SiS  30), and as highlighted here, kidney and liver toxicity.  This review provides clear improvements to current study designs that need to be upheld by industry as well as the EU government.

Article first published 05/09/11

---

References

1. Séralini G-E, Mesnage R, Clair E, Gress S,Vendômois J, Cellier D. Genetically modified crops safety assessments: present limits and possible improvements. 2011. Environmental Sciences Europe, 23, 10-20
2. Benachour N and Séralini G-E. 2009. Glyphosate formulations Induce Apoptosis and Necrosis in Human Umbilical, Embryonic, and Placental Cells Chemical Research. Toxicology, 22, 97–105
3. Ho MW and Cherry B. Death by multiple poisoning, glyphosate and Roundup. Science in Society 42, 14, 2009
4. Malatesta M, Perdoni F, Santin G, Battistelli S, Muller S, Biggiogera M. 2008. Hepatoma tissue culture (HTC) cells as a model for investigating the effects of low concentrations of herbicide on cell structure and function.  Toxicology In Vitro, 22, 1853-1860
5. Ho MW and Steinbrecher R. Fatal flaws in food safety assessment: critique of the joint FAO/WHO biotechnology and food safety report. Environmental & Nutritional Interactions 1998, 2, 51-84.
6. Cherney DZ, Sochett EB and Miller JA. 2005. Gender differences in renal responses to hyperglycemia and angiotensin-converting enzyme inhibition in diabetes. Kidney International, 68, 1722–1728
7. Braun JM, Yolton K, Dietrich KN, Hornung R, Ye X, Calafat AM, Lanphear BP. 2009. Prenatal bisphenol A exposure and early childhood behavior. Environmental Health Perspectives, 117, 1945-1952
8. Anway MD, Cupp AS, Uzumcu M, Skinner MK. 2005. Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science, 308, 1466-1469
9. Sirinathsinghji E and Ho MW. EU regulators Regulators and Monstanto Exposed for Hiding Glyphosate Toxicity. Science in Society 51, 46-48, 2011
10. Ho MW. GM Soya Fed Rats: Stunted, Dead, or Sterile. Science in Society 33, 4-6, 2007
11. Ho MW More illnesses linked to GM crops. Science in Society 30, 8-10, 2006