September 30, 2013

Animal feeding studies using feed from genetically engineered crops*

image001

Alison Van Eenennaam, Department of Animal Science, UC Davis

Genetically engineered (GE), also known as GM or biotech, crops first came to market in 1996 and since then billions of animals on commercial livestock operations throughout the world have been uneventfully consuming feed increasingly derived from GE crop varieties. However, over the past couple of years, several sensational small-scale animal feeding studies have been published suggesting that GE feed is resulting in deleterious health effects in animals. One study with rats[1] and one with hogs[2] reported adverse effects of feeding glyphosate resistant soybean meal. Both suggested that these crops were not adequately studied and suggested that there need to be longer studies with more animals. As an animal scientist familiar with the substantial GE feed literature, such results are perplexing given the wealth of peer-reviewed studies that have shown no effect of GE crops in animal feeding studies.

Prior to commercialization, GE crops must go through an extensive safety evaluation. The Organization for Economic Co-operation and Development (OECD) has established safety assessment processes based on the principle of “substantial equivalence” to assure that foods derived from GE crops are as safe and nutritious as those from plants derived through conventional breeding. The concept is based on the principle that “if a new food is found to be substantially equivalent in composition and nutritional characteristics to an existing food, it can be regarded as being as safe as the conventional food”. For GE crops, this comparison entails an extensive chemical analysis of key macronutrients, micronutrients, antinutrients and toxins. Some countries also require whole food animal feeding studies, typically with rats using a 90-day rodent toxicity study design.  Most conventionally-bred crops that are on the market have not ever been tested for their safety in animals, but they are known to be safe based on their history of safe use. Likewise, few whole foods have been subject to toxicological testing.

There are hundreds of animal GE feeding studies that have been published in peer-reviewed journals[3]. The Federation of Animal Science Societies (FASS) maintains a bibliography that currently consists of more than 400 animal GE feeding studies. Despite suggestions to the contrary, independent long-term and multigenerational feeding studies have been conducted in a variety of species.  Published, long-term feeding studies using a GE-based diet range from 110 d to 728 d. The longest published multigenerational study involved ten generations of quail fed up to 50% GE corn. It has been estimated that over 70-90% of harvested GE biomass is fed to food-producing animals3, making the world’s livestock populations the largest consumers of the current generation of GE crops. Crops that are produced using GE are likely to become even more important to animal agriculture as the global livestock population grows in response to increased demand for animal protein products.

In a comprehensive review of the health effects of GE plants, Snell et al.[4] focused on 12 long-term and 12 multigenerational feeding trials with GE crops that also had a 90-day rodent GE feeding toxicity study comparator.  It is important to note that these studies were financially supported by public funds. The question they specifically asked was, “Do long-term and multigenerational GE feeding studies provide any new evidence indicative of some adverse effect(s) that were not previously identified in the 90-day rat study?” The authors concluded that none of the long-term or multigenerational studies they evaluated revealed any new effect that had not been found in the 90-day rodent toxicology study. The authors suggested that while a more standardized protocol for long-term and multigenerational studies would be useful for exploratory fundamental research projects, such studies should be conducted on a case-by-case basis for GE food safety only if some reasonable doubt remained after a 90-day rodent feeding trial.

Another 2013 review[5] examined 60 high-throughput “-omics” comparisons between GE and non-GE crop lines, including 17 long-term and 16 multigenerational animal feeding studies, to determine if these additional tests raised new safety concerns. High-throughput “-omics” – transcriptomics, proteomics, and metabolomics – methods have been suggested as a nontargeted approach to detect unintended effects in GE plants.  Long-term studies included those carried out in rats, mice, salmon, beef cattle, dairy cows, macaques, pigs, and quail. Multigenerational studies included rats, mice, pigs, bulls, dairy cows, goats, sheep, broilers, laying hens, and quail. These powerful studies consistently revealed that GE had fewer unintended effects than conventional breeding techniques. Even when GE crops were designed to intentionally have altered metabolic traits, “-omics” expression profiling technologies revealed few unintended effects.  The authors concluded that “none of the “-omics” comparisons has raised new safety concerns about (marketed) GE varieties; neither did the long-term and multigenerational studies on animals.” They further proposed that the data collected to date suggest that the risk assessment should actually be lowered for GE crops.

So why is there discordance between the data and public perception, and why does it matter? One reason for this disconnect is that studies purporting that GE feed is dangerous tend to attract considerable media attention, whereas those that show no effect are rarely mentioned. A critical review of studies showing dramatic deleterious effects from consuming GE feed reveal significant experimental design flaws and failure to cite and address the body of contradictory results that are reported in the published literature. Criticisms of these studies by independent scientists and the global regulatory community include over interpretation of differences that lie within the normal range of variation and hence are not biologically relevant, poor toxicological interpretation of the data, erroneous use of correlations to imply causation, inappropriate or absence of any statistical analyses of data, and/or the use of control feedstuffs that was not derived from near isogenic lines[6]. This emphasizes the importance of following standard protocols[7] when designing animal feeding studies. The lack of compliance with internationally-recognized protocols by some research groups and the highly sensational presentation of their results in public settings have exacerbated the continued controversy associated with the safety of GE food and feed. Independent animal scientists have an obligation to ensure that animal feeding studies are carried out according to standard protocols to ensure data can be appropriately analyzed and unambiguously interpreted in the absence of confounding factors.

It is important to address these issues because there are real policy decisions being made based on poorly-designed scientific studies that are sometimes being published in questionable open access journals (see Beall’s list for a useful list of “potential, possible, or probable predatory scholarly open-access publishers”), while the vast literature of carefully conducted studies carried out by independent scientists is effectively ignored. For example, in response to public concerns on GMOs fanned by sensational studies, the European Union adopted a regulation in June 2013 requiring, among different risk management measures, an obligatory 90-day whole food/feed rodent feeding study for regulatory approval of each GE crop event.  Depending on the outcome of that study, a 2-year long-term GE feeding study in rats may also be requested, on a case-by-case basis. This regulation passed despite the fact that the European Food Safety Authority (EFSA) determined that animal feeding studies do not add to the safety assessment if there are no compositional changes in the GE crop, i.e. the crop is substantially equivalent.

In 2007 it was estimated that the range of costs involved with animal performance and safety studies (typically a 90-day whole food/feed rodent feeding study) for approval of a GE crop ranged from $USD 300,000–845,000[8]. Presumably costs have increased since that time, and if longer studies are required costs would likewise be increased. A recent two-year rat feeding study involving 200 rats was purported to cost 3.2 million Euro (~ $USD 4 million)1. Calls for long-term or mutigenerational GE feeding studies on long-lived target species such as cattle would be orders of magnitude more expensive, assuming sufficient feed from the GE crop and its isogenic comparator was available to perform such work. Additionally, the cost of rendering the animals would need to be factored into regulatory evaluations as animals would not be saleable if fed an as-yet unapproved GE crop variety.

Given the weight of scientific evidence that already exists on the safety of commercialized GE crops, the decision to conduct an animal feeding study with a GE crop should be based on the need to answer a scientific question that cannot be addressed using in silico and in vitro methods.  A reasonable hypothesis-driven food safety concern should be the driver for the additional expense and use of experimental animals required for such studies. Mandating animal feeding studies based on the process used to derive a GE crop, rather than the unique traits and/or phenotype associated with the gene/crop combination is not justified based on the weight of evidence. Over 20 years of data has not revealed the process of GE to be risky, and in some cases feed derived from GE crops may pose fewer feed safety risks than are implied by the non-GE alternative[9]. Regulations triggered by how products are made are inconsistent with science-based risk assessment unless there is something inherently risky about the process, as compared to existing methods.

* Sections of this article are excerpted from a forthcoming publication:
Van Eenennaam, A. L., 2013. GMOs in Animal Agriculture: Time to Consider Both Costs and Benefits in Regulatory Evaluations. Journal of Animal Science and Biotechnology. 4:37.


[1]Séralini, G. E. et al. 2012. Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize. Food and Chemical Toxicology 50: 4221-4231.

[2] Carman, J. et al. 2013. A long-term toxicology study on pigs fed a  combined genetically modified (GM) soy and  GM maize diet. Journal of Organic Systems 8: 38-54

[3] Flachowsky G, Schafft H, Meyer U: Animal feeding studies for nutritional and safety assessments of feeds from genetically modified plants: a review. Journal fur Verbraucherschutz und Lebensmittelsicherheit (Journal of Consumer Protection and Food Safety) 2012, 7:179-194.

4 Snell, C. et al. 2012. Assessment of the health impact of GM plant diets in long-term andmultigenerational animal feeding trials: a literature review. Food and Chemical Toxicology 50: 1134-1148.

[5] Ricroch, A. E. 2013. Assessment of GE food safety using ‘-omics’ techniques and long-term animal feeding studies. New Biotechnology 30: 349-354

[7] Cromwell, G., G. Dana, and G. Hartnell. 2003. Best Practices for the Conduct of Animal Studies to Evaluate Crops Genetically Modified for Input Traits, International Life Sciences Institute, Washington, DC.

[8] Kalaitzandonakes N, Alston J, Bradford K: 2007. Compliance costs for regulatory approval of new

biotech crops. Nature Biotechnology 25:509-511.

[9] Hammond BG, Campbell KW, Pilcher CD, Degooyer TA, Robinson AE, McMillen BL, Spangler SM, Riordan SG, Rice

LG, Richard JL: Lower fumonisin mycotoxin levels in the grain of Bt corn grown in the United States in 2000-2002.

Journal of Agricultural and Food Chemistry 2004, 52:1390-1397.