The Royal Commission on Genetic Modification - submissions

ANZFA — Responses to Criticisms Raised at Royal Commission

Strong criticism has been levelled at the Australia New Zealand Food Authority (ANZFA) in some submissions presented to the Commission. (See for example submissions from the National Nutritional Foods Association of New Zealand (NNFA), the Pesticide Action Network of New Zealand (PANNZ) and GE Free New Zealand (RAGE) in Food and Environment Incorporated (GE Free New Zealand. In response to these criticisms ANZFA submitted the following rebuttal evidence.

That ANZFA has refused to acknowledge it received advice based on false evidence in relation to royal jelly.

ANZFA contended that events overtook consideration of an Expert Committee’s report (P161) as it related to royal jelly, bee pollen and propolis because a separate proposal (P154) was raised in June 1997 after release of coronial findings that implicated royal jelly in the death of an asthmatic. Even though ANZFA’s emergency powers under s37 were used to strengthen the warning, the decision to include the word ‘fatalities’ was not based on the committee’s findings, but on consideration of the coronial findings, which had recommended a ban of sale of royal jelly.

That ANZFA has failed to rescind the ‘harshest warning statement in the world’ in relation to royal jelly.

ANZFA stated that they continue to have Proposal (P154) under consideration because the matter of the appropriate wording of a warning statement for royal jelly is not yet resolved. The process is on hold pending the outcome of the current review of evidence of harm of royal jelly by the Therapeutic Goods Administration’s Adverse Drug Reaction Advisory Committee. These latest developments are about developing a label statement commensurate with a broadly accepted level of risk of royal jelly sold as therapeutic goods or foods, rather than the fear of losing face.

That ANZFA’s Safety Assessment Process for GM Foods is not scientifically comprehensive and consistent, and relies on data submitted by the applicants that is not published in peer reviewed journals (NNFA, PANNZ, and GE Free New Zealand).

ANZFA contended that the approval process under Food Standard A18 - Food Produced using Gene Technology of the Food Standards Code is open, consultative and based on a scientific risk assessment undertaken by ANZFA. Two full rounds of public comment are invited during the assessment process. The Standard was widely published at the time of its gazettal in August 1998 and the requirement for safety assessment applies to all foods on the Australian market, including imported foods. ANZFA then highlighted that applications were subject to both internal and external peer review and the names and qualifications of the panel were listed by ANZFA. ANZFA stated that they receive raw experimental data from applicants not a summary of conclusions. This enables a more in depth analysis of the evidence as submitted by the applicant. The assessment focuses on the new gene product(s), its properties including potential allergenicity, compositional differences in the food and its history of use as a food or food product. Applicants are required to submit data that defines:

How the food crop was developed, including the molecular biological data which characterizes the genetic change;
Composition of the novel food compared to non-modified counterpart foods;
Nutritional information for the novel food compared to non-modified counterparts;
Potential for new toxins; and
Potential for causing allergic reactions.

Additional research or testing can be required if ANZFA scientists are not satisfied at any stage in the safety assessment process. Only if all of ANZFA’s stringent criteria are met is a novel food allowed access to the Australian and New Zealand markets. In addition to its assessment of the data provided by applicant companies, ANZFA considers all relevant data available from other food regulatory authorities. It also maintains close links with international scientific and clinical bodies to gain access to research, trial and other reports relevant to the safety of GM foods. ANZFA maintains a public register of all the scientific data provided (other than a very small amount of data classified commercial-in-confidence). Furthermore, ANZFA invites two rounds of public comment to obtain the views of interested parties and carefully considers them when making a recommendation to Health Ministers.

That toxicity tests performed on GM food are inadequate and that there is a need for long-term toxicity studies on genetically modified foods to ensure their safety (NNFA, PANNZ).

ANZFA noted that foods are complex mixtures of compounds characterised by wide variations in composition and nutritional value. Due to their bulk, they can usually be fed to animals only at low multiples of the amounts that might be present in the human diet. Therefore, in most cases, it is not possible to conduct dose-response experiments for foods in the same way that these experiments are conducted for chemicals. The nutritional balance also needs to be maintained in order. A diet that is poorly balanced will compromise the interpretation of any feeding study, since the effects observed would override any small adverse effect that may be related to a component or components of the food. Identifying any potentially adverse effects and relating these to an individual component or characteristic of a food can, therefore, be extremely difficult. Animal welfare must also be considered especially if the tests unlikely to produce meaningful information. It is also more appropriate to study the safety of a newly expressed protein in a genetically modified (GM) food on its own rather than as part of a whole food. Furthermore, the few proteins in nature that exert toxic activity do so in an acute time frame. In addition, protein toxicity is dependant on integrity of the three-dimensional structure. Consequently, susceptibility of the protein to heat and other processing conditions that the food may undergo, and its digestibility in the gastrointestinal tract are required. ANZFA also noted that acute toxicity tests or short-term feeding studies are not meant to substitute for long-term toxicity tests. ANZFA then stated that should developments in the technology result in modifications that provide significantly different nutrient combinations or other novel food characteristics not previously encountered in the food supply, such foods may require additional considerations to permit comprehensive assessment. The guidelines for the safety assessment of GM foods provide this flexibility of approach. If ANZFA scientists recommended that longer-term studies were required, the food will not be approved and the particular applicant concerned will be required to undertake the longer-term studies before their application is considered further.

That the use of the concept of substantial equivalence as a final decision making tool on the safety of GM foods is inappropriate. (NNFA, PANNZ)

ANZFA submitted that the concept has been internationally recognised and embraced as a valuable tool in the safety assessment of foods produced using gene technology. ANZFA then outlined the process that is used in the assessment of new food products. Each food needs to be evaluated on an individual basis with regard to the significance of any changes in relation to its composition or to its properties. If a difference is identified in comparison with the control line it then has to be evaluated for its biological or food safety significance. If the difference exceeds natural variation then further assessment would be required. If the difference does not exceed natural variation then further assessment would usually not be required. It needs to be emphasised, however, that identification of a difference does not necessarily equate to an adverse food safety outcome. Many differences are neutral with respect to food safety and are consistent with the natural variation that occurs in all food.

That the sample sizes used by applicants to undertake statistical comparisons in for example, compositional analyses and animal studies, are inadequate to ensure the safety of the food (PANNZ).

ANZFA submitted that perceived problems with the sample size taken by applicants in the course of studies conducted have been misinterpreted. All procedures are run to internationally accepted standards.

In the case of the herbicide-tolerant canola GT73 (Application A363), that the data submitted on laboratory rats that were fed GM canola mash derived from GT73 canola is flawed and the study should be repeated (NNFA, PAANZ).

ANZFA pointed out that canola meal is not considered to be a human food fraction due to the presence of the natural toxicants, erucic acid and glucosinolates, and was evaluated to compare levels of major components to determine any potential unintended effects. Canola meal, whether GM or not, is not regarded as a food fraction and the genetic modification does not change this pattern of consumption. Although liver weights were enlarged in rats fed GM canola meal, this difference was not considered biologically meaningful and these livers appeared normal when grossly examined. Based on the data submitted, the slight increase in liver weight was possibly attributable to a slightly higher level of the natural toxicant glucosinolate in the GM canola meal. ANZFA concluded that there were no human health and safety concerns in relation to the feeding studies in rats fed on canola meal. A third study submitted to ANZFA included an assessment of glyphosate-tolerant canola meal, control lines from around the world and rat chow as a negative control. Duplicate samples were prepared for all lines tested to determine the variation that may arise due to processing of the canola to canola meal. There were no mortalities and no treatment-related adverse clinical signs that were considered treatment related in any group. There were no significant differences in body weight, cumulative weight gain, terminal body weights or food consumption for animals fed GM canola meal compared to the non-GM control canola meal. There were also no significant differences in absolute or relative liver or kidney weights between animals fed the GM canola meal compared to the non-GM canola meal or the population of canola varieties. Glucosinolate levels were within the acceptable normal range for all groups and there was no difference in glucosinolate levels between the GM and non-GM canola meals. This third study confirms the conclusion reached by ANZFA for the previous two studies.

In the case of the insect resistant Bt potato line Bt-06 (Application A382), that the study submitted on laboratory rats which were fed GM and non-GM potatoes should be repeated because the animals were ill. (PANNZ)

ANZFA submitted that in this study there were no effects of treatment on weight gain, no adverse clinical signs, no significant differences in absolute or relative organ weights of the kidney, liver or testes. Gross pathology revealed a number of sporadic pathological findings, which appeared equally in both the treated and control groups, that are common in these highly inbred colonies of rats. These findings could not be related to the test substance.

That GM animal feeds are fed to farm animals and then people will eat the products (e.g. milk, eggs and meat) and this has not always been considered nor assessed (PANNZ).

ANZFA stated that the human health consequences, if any, of the feeding of GM foods to animals should be assessed on a case-by-case basis, taking into account any potential hazards identified for the novel proteins present in the food and changes to the composition of the food combined with a consideration of the animal feeding practices used for the particular food/feed in question. If hazards are identified in an assessment of a GM food, then consideration should also be given to potential human exposure to that hazard through the use of the GM food as a feed for animals. Information available from the Federation of Animal Science Societies indicates that no DNA and/or protein can be detected in animals (products such as meat muscle, whole milk, poultry and eggs), fed a variety of GM commodities. In addition, nutrients in meat, milk and eggs from livestock fed GM feeds are the same as those from livestock fed conventional feeds. Because components of feeds are broken into smaller components during digestion by the animal, plant proteins have not been detected in milk, meat or eggs. Peer reviewed journal articles have recently investigated the fate of ingested recombinant DNA plant material in cattle and chickens being fed conventional maize or recombinant Bt maize. The results clearly showed no recombinant DNA Bt maize in cattle or poultry.

That ANZFA’s statements that foods such as corn are not generally regarded as food allergens have not been well researched and that the risk assessment process for potential allergenicity of the new proteins is flawed and inadequate (NNFA, PAANZ).

It was noted that virtually all food allergens are proteins, but only a small fraction of the many proteins found in food are allergenic. The safety assessment process applied to each new GM food on a case-by-case basis includes an evaluation of the likely allergenicity of any new proteins present in the food as a result of the genetic change. It should be noted that additional protein diversity could also be introduced into the food supply in other ways. For example, the recently introduced kiwi fruit, produced using conventional breeding techniques, has proven to be an additional source of food allergens.

That the methodology for assessing potential allergenicity of new proteins is theoretical and not science based (NNFA, PAANZ).

In ANZFA’s opinion, the prediction of allergenic potential of novel proteins is not a simple matter. They then outlined the following process used to assess potential allergenicity.

Source of the gene — particular caution should be exercised if the source organism of the gene contains known allergens, e.g. peanuts;

Amino acid sequence similarity of the newly introduced protein to known allergens — through an international data base

Stability to digestion - the ability of food allergens to reach and cross the intestinal mucosal barrier in immunologically intact form appears to be a prerequisite to allergenicity. Simulated gastric and intestinal digestive models of mammalian digestion have been used to assess the digestive stability of proteins;

Heat or processing stability;

Immunological reactivity of the newly introduced protein — if a novel protein is derived from a known allergenic source or if it has sequence similarity to a known allergen, then the reactivity of the novel protein with IgE from the blood serum of appropriate allergic individuals can be tested.

In cases where doubt exists about the potential allergenicity of a novel protein a cautious regulatory approach is warranted until such time as more evidence becomes available. This would entail further scientific scrutiny of additional information and risk management strategies that consider the levels of dietary exposure.

Our Royal Commission homepage