Safety assessment of GM foods
How does FSANZ ensure GM foods are safe?
FSANZ has established a rigorous and transparent process for assessing the safety of GM foods. The safety assessment is undertaken in accordance with internationally established scientific principles and guidelines developed through the work of the Organisation for Economic Cooperation and Development (OECD), Food and Agriculture Organization (FAO) of the United Nations, World Health Organization (WHO) and the Codex Alimentarius Commission.
The safety assessment process used by FSANZ is described in detail in the Guidance Document “Safety Assessment of Genetically Modified Foods”, which is available from the FSANZ website. FSANZ conducts a thorough safety assessment of all GM foods before they are allowed in the food supply. This assessment ensures that any approved GM foods are as safe and nutritious as comparable conventional foods already in the Australian and New Zealand food supply.
What's involved in a safety assessment?
The safety assessment of a GM food is conducted within the established risk assessment framework used by FSANZ. In the case of GM food, the primary purpose is: (i) to identify new or altered hazards associated with the food as a result of the genetic modification; (ii) to assess whether there is any risk associated with any identified hazards under the intended conditions of use; and (iii) to determine if any new conditions of use are needed to enable safe use of the food.
The safety assessment is characterised by:
- Case-by-case consideration of GM foods;
Case-by-case assessment is necessary because the key issues requiring consideration in a safety assessment will often depend on the type of food being evaluated and the nature of the genetic modification.
- Consideration of both the intended and unintended effects of the genetic modification;
In addition to the intended effect (e.g. a new trait, such as insect protection), there may also be other effects associated with the genetic modification that were unintended (e.g. compositional changes to the food) and which may impact on the health and safety of the population. Therefore it is important that both the intended and any unintended effects are evaluated.
- Comparisons with conventional foods having an acceptable standard of safety.
Such a comparative approach focuses on: (i) the identification of similarities and differences between the GM food and an appropriate comparator; and (ii) a characterisation of any of the identified differences in order to determine if they may raise potential safety and nutritional issues.
The goal of the safety assessment is not to establish the absolute safety of the GM food but rather to consider whether the GM food is comparable to the conventional counterpart food, i.e., that the GM food has all the benefits and risks normally associated with the conventional food.
Is company data used during the assessment of a GM food?
The responsibility for demonstrating the safety of any new food product on the market lies with the developer of that product. This is also the case for new chemicals and drugs. When an applicant seeks approval for a new GM food, they must provide FSANZ with the evidence that supports the safety of the product. It is a requirement that this data be generated according to quality assurance guidelines that are based on internationally accepted protocols (i.e. validated methodology and procedures that are consistent with Good Laboratory Practice (GLP)) and stand up to external scrutiny (i.e. independent audits and documentation trails). To achieve this, the applicant submits to FSANZ a comprehensive dossier of quality-assured raw experimental data for each GM food. This enables FSANZ to independently assess the data and reach a conclusion about the safety of the food.
FSANZ also complements the data package provided by the applicant with information from the scientific literature, other applications, other government agencies and the public.
Why does FSANZ not do its own independent testing of GM foods?
Paper reviews are a standard scientific method of evaluation used by regulators around the world, to evaluate the health and safety of a variety of products including food, drugs and agricultural and veterinary chemicals. The methods and approach used by FSANZ are wholly consistent with international guidelines developed according to scientific advice provided by the WHO, FAO, and OECD.
Companies involved in the development of GM foods spend millions of dollars rigorously testing their products according to these requirements, which include detailed documentation of testing. Thorough analysis is conducted of the data and of the protocol used to ensure the validity of results. If FSANZ determines that the data are not sufficient, additional information and testing may be required. FSANZ may also supplement the information provided by the Applicant with any published data in Australia and New Zealand or internationally that is relevant to the product in question.
Are there potential long-term risks associated with the consumption of GM foods?
Many of the food safety issues raised by GM foods are equally applicable to foods produced by conventional means. GM foods are however subjected to a safety assessment before they are permitted in the food supply. The safety assessment includes extensive analyses of the composition of the food, a full consideration of the safety of any new substances that have been introduced into the food (e.g. proteins), as well as a thorough characterisation of the genetic changes that have been introduced into the organism from which the food is derived. This ensures that any GM food that is approved is as safe as food already in the food supply, including in the long term.
To date, gene technology has not been shown to introduce any new or altered hazards into the food supply, therefore the potential for long term risks associated with GM foods is considered to be no different to that for conventional foods already in the food supply. As a consequence, FSANZ does not consider that long term studies are generally needed to ensure the safety of GM foods.
Should developments in the technology result in modifications that potentially introduce new or altered hazards into the food, such foods may require additional considerations to address the potential for long term health effects. In such cases, long term studies may be a valid approach to include in the assessment of the overall safety of such foods.
Why doesn’t FSANZ require that GM foods be tested in animals?
FSANZ considers that a scientifically-informed comparative assessment of GM foods with their conventional counterparts can generally identify any potential adverse health effects or differences requiring further evaluation. In the majority of circumstances, animal toxicity studies with whole foods are not likely to contribute any further useful information to the safety assessment and are therefore not warranted. As a consequence, FSANZ does not require that animal toxicity studies with whole GM foods be undertaken on a routine basis.
FSANZ acknowledges there may be future GM applications, particularly for foods with intentional modifications to composition, where the results of animal toxicity studies may be informative. FSANZ therefore continues to assess the need for animal studies on a case-by-case basis, taking into account the nature of the genetic modification and the results of the comparative assessment. While FSANZ does not routinely require animal toxicity studies to be undertaken, where such studies already exist, Applicants are expected to provide these to FSANZ to evaluate as additional supporting information.
As part of a continual review of FSANZ’s scientific approach to the safety assessment of GM foods, FSANZ convened an expert panel in June 2007 to develop guidance and recommendations on the role animal feeding studies can play in the safety assessment of GM foods.
The expert panel recommended that FSANZ should continue with its case-by-case assessment of GM foods on the basis of best available science. The panel noted that whole-food animal feeding studies may be informative in some limited circumstances, but concurred with FSANZ that any potential adverse health effects can generally be identified by a scientifically informed comparative assessment of the GM food against its conventional counterpart.
The report from the expert panel is available from: www.foodstandards.gov.au/_srcfiles/Workshop%20Report%20FINAL.pdf
Is the composition of GM foods different to the composition of non-GM foods?
Sometimes the purpose of the genetic modification may be to deliberately introduce a compositional change to the food, e.g. to increase the level of a particular nutrient. Examples of GM foods with deliberate changes to composition that have been assessed by FSANZ include high oleic acid soybeans and high lysine corn. In most cases however other characteristics are changed in the organism, without affecting the composition of derived food, e.g. herbicide tolerance.
Compositional analyses are primarily done to determine if any unexpected changes in composition have occurred to the food and to establish its nutritional adequacy. Compositional analyses are particularly important where there has been a deliberate change to the composition of food. Analysis of the composition of the food in these circumstances helps to confirm that the new or altered characteristic is being expressed appropriately, and also helps to quantify the magnitude of the change, which may be important for assessing safety.
Sometimes, even when there has been no intent to alter the composition of a food, the analysis will indicate that the composition of the GM food differs slightly from the non-GM control line. Where this occurs, FSANZ will analyse such differences to determine their biological relevance. This is done by comparing the values for the altered constituent to those of other commercial varieties and to the range of natural variation (obtained from the published literature). The use of published ranges and historical control data as comparators in safety assessment studies is standard for interpreting biological and analytical variation.
To date, the significant differences between test and control lines tend to be small and they are usually within the range that would be expected for other commercial varieties. In these cases, FSANZ concludes that the observed difference is not biologically meaningful and certainly does not pose a risk to consumers. Sometimes, differences can be explained by differences in locations or seasons. Generally, the differences do not occur consistently in all locations and are therefore not considered to indicate a trend.
Does genetically modifying a food decrease its nutritional value?
There is no evidence to suggest that genetic modification per sereduces the nutritional value of food. To assess this, FSANZ evaluates the major constituents of the food (fat, protein, carbohydrate, fibre, ash and moisture) as well as the key nutrients (amino acids, vitamins, minerals, fatty acids). This approach enables both the intentional effects and any unintentional compositional changes in the food to be assessed. To date, none of the GM foods assessed by FSANZ have been shown to have any unintentional changes to their nutritional value.
Are there risks associated with the possible introduction of new toxins and allergens in GM foods?
Typically the main outcome of genetic modification is the expression of a new protein or proteins in the organism. These new proteins may subsequently be present in derived food products. While a large and diverse range of proteins are ingested as part of the normal human diet without any adverse effects, a small number have the potential to impair health, e.g. because they are allergens or may have toxic or anti-nutritional effects. As a consequence, one of the main parts of the safety assessment of any GM food is a consideration of the potential toxicity and allergenicity of any new proteins introduced into the food.
Assessment of potential toxicity and allergenicity is done using a weight of evidence approach, which means a variety of evidence, drawn from a number of different studies, is used to reach an overall conclusion. Further details on the assessment of potential toxicity and allergenicity can be obtained from the following sources:
- For a detailed technical description of the assessment process, refer to the Guidance Document “Safety Assessment of Genetically Modified Foods”;
- For a lay persons description of the assessment process, refer to the bookletGM Foods – Safety Assessment of Genetically Modified Foods, www.foodstandards.gov.au/_srcfiles/GM%20Foods_text_pp_final.pdf
In addition to these considerations, the GM food safety assessment also considers the issue of naturally occurring toxins (e.g. solanine in potatoes) and allergens in the food and whether any changes in the amounts of any pre-existing natural toxins and/or allergens have occurred as a result of the modification.
While the introduction of new toxins and allergens is a recognised risk associated with GM foods, it may also be possible to use gene technology to develop foods specifically where such compounds are significantly reduced or eliminated. For example, research is underway on certain proteins in peanuts that are known to be the cause of major allergies in susceptible individuals.
Is there potential for an increase in antibiotic resistance in humans resulting from the use of antibiotic resistance marker genes in GM foods?
Antibiotic resistance marker genes are sometimes used in the development of GM plants to facilitate the selection of cells that have been transformed with the gene of interest. The antibiotic resistance gene provides a selective advantage to the transformed cell enabling it to grow and divide in the presence of the antibiotic. Cells that have not been transformed will not survive.
It has been suggested that ingestion of food containing copies of antibiotic resistance genes could facilitate the transfer of the gene to bacteria inhabiting the gut of animals and humans. It is argued that these genes may then be transferred to disease causing bacteria and that this could compromise the therapeutic use of antibiotics.
This is one of the many issues that FSANZ considers during the food safety assessment.
Although theoretically possible, horizontal DNA transfer of antibiotic resistance genes from food products to gut microorganisms is regarded as a rare possibility because of the many complex and unlikely events that would need to occur consecutively. Furthermore, the transfer of antibiotic resistance genes from GM food to bacteria has not been observed under natural conditions. As horizontal DNA transfer cannot be completely ruled out, the potential human health impact, should such a transfer occur, is also considered by FSANZ during the safety assessment.
The human health impact will largely depend on the nature of each particular antibiotic resistance gene and must be assessed on a case-by-case basis. However, the commonly used kanamycin, ampicillin and streptomycin resistance genes, are not considered to pose any significant health concerns and have been safely used for a number of years. As a precaution, the use of marker genes encoding resistance to antibiotics with significant public health uses (e.g. vancomycin) are avoided.
Not all GM foods contain antibiotic resistance genes. In some cases, these genes are removed after successful transformation. In other cases alternative selectable marker genes may be used (e.g. herbicide tolerance genes are frequently used for selection purposes in plants). With time, it is anticipated that the presence of antibiotic resistance genes in GM foods will become less commonplace. If alternative marker genes are used, they also need to be evaluated for their safety in the same way as for any other novel gene.
Is post market monitoring of GM foods undertaken?
Post-market monitoring is often suggested as a means of demonstrating whether long-term adverse health effects may be associated with the consumption of GM foods.
It has been recognised internationally that the use of pre-market safety assessment already provides assurance that a GM food is comparable to its conventional counterpart in relation to health risks and benefits, therefore the likelihood of identifying long-term effects specifically attributable to GM foods would be very low. Moreover, the practicality of using post-market monitoring to assess the long-term human health impacts of consuming GM foods has not been established.
Many chronic health problems have complex causes and it is unlikely that observational epidemiological studies could identify such effects specifically related to GM foods. The same also applies to the identification of potential long-term beneficial health effects.
In general, therefore, FSANZ does not believe post-market monitoring to be a practical, enforceable or effective risk management option. This is particularly the case where passive monitoring or general health surveillance, which does not address a specific hypothesis, is proposed.
Nevertheless we recognise that post-market monitoring may be appropriate in certain circumstances. Post-market monitoring may be useful in situations where a GM food has been developed specifically to produce a nutritional effect in the population. In these cases it may be desirable to monitor changes in nutrient intake levels in order to confirm assumptions made during the risk assessment and evaluate their potential effect on the nutritional and health status of the population.
The need for post-market monitoring following approval of a GM food will be considered by FSANZ on a case-by-case basis, taking into account the unique characteristics of the GM food and the feasibility of undertaking such a study.
There are currently no official mechanisms within Australia and New Zealand for monitoring the long-term impacts of GM foods. In Australia and New Zealand, as in most other countries, the responsibility for post-market monitoring is covered by an ongoing duty of care on the part of the developer. The developer is expected to monitor for existing and emerging risks that may be associated with its product and notify regulatory authorities whenever new information is uncovered.
Do herbicide-tolerant crops have higher chemical residues?
The use of herbicide tolerant crops is likely to result in a shift either in the type of herbicide that is used during the cultivation of the crop or the pattern of usage but will not necessarily result in higher chemical residues. Herbicide tolerant crops exist that have been developed using conventional plant breeding therefore the issue of herbicide tolerance is not specific to GM crops.
Residues of any agricultural chemicals, for example herbicides, can only legally be present in food if the residues comply with specific Maximum Residue Limits (MRLs) for agricultural and veterinary chemicals. The MRLs apply irrespective of whether the food is derived from conventional (non-GM) or GM crops.
Australia and New Zealand independently and separately develop MRLs for agricultural and veterinary chemicals in food. For New Zealand, MRLs for agricultural compounds are included in the New Zealand (Maximum Residue Limits of Agricultural Compounds) Food Standards, 2007(and subsequent amendments) issued under sections 11C and 11Z of theFood Act 1981. For Australia, MRLs are listed in Standard 1.4.2 of the Code.
Under the Trans Tasman Mutual Recognition Arrangement (TTMRA) between Australia and New Zealand, which commenced on 1 May 1998, the following applies:
- Food produced or imported into Australia that complies with Standard 1.4.2 of the Code can be legally sold in New Zealand.
- Food produced or imported into New Zealand that complies with the New Zealand (Maximum Residue Limits of Agricultural Compounds) Food Standards, 2007can be legally sold in Australia.
What about the safety of food derived from animals that have eaten GM stockfeed?
Concerns are occasionally expressed that the practice of feeding GM plant material to livestock may pose an indirect risk to humans, through consumption of meat, milk and eggs.
As many animal feeds are derived from the same GM crops that are used for human consumption (e.g. corn or soybean), they have typically already been assessed for their human food safety. Food derived from livestock consuming GM plant material already approved for human food use can therefore also be regarded as safe.
Scientific evidence published so far, including by the OECD and the European Food Safety Authority indicates that feeding GM plant material to animals does not affect the nutritional value or safety of the meat, milk and eggs derived from those animals. Minute traces of recombinant DNA fragments have occasionally been detected in animal tissues but there is no basis to suppose that these pose a hazard. This is because DNA is a natural component of the human diet, it being present to varying degrees in many plant and animal derived foods. There is no difference in terms of risk between small fragments of recombinant DNA and the DNA already present in our diet.
In the case of GM plants intended primarily for use as animal feed, it is now standard practice for these to also undergo food safety assessment and approval for human food use. This recognises it may be impossible to prevent inadvertent co-mingling of plant material during cultivation, transport and storage, and so ensures their use as feed will not pose indirect risks to humans.
FSANZ response to studies cited as evidence of adverse effects from GM Foods (2011) This table was last updated July 2011