There has been considerable media interest in an announcement by CSIRO that it has discontinued the development of a particular type of genetically modified (GM) pea after the publication of a paper reporting an immune response in mice that were fed the peas under laboratory conditions compared with those given non-GM peas.  

Between 1997 and 2002, CSIRO conducted research and development field trials of genetically modified (GM) peas that were protected against pea weevils as a result of introducing an insecticidal protein (alpha-amylase inhibitor) derived from bean plants. These field trials enable data to be collected progressively under controlled conditions to enable researchers to decide whether results justify proceeding with further development. The trials were conducted under strict containment conditions – none of the GM material has been permitted to enter the human food chain.

FSANZ has not conducted any safety assessment of the peas since they are still in research and development, nor has any data been submitted to Food Standards Australia New Zealand (FSANZ) for assessment. However, CSIRO approached FSANZ several years ago to obtain advice on the type of data they would need to support an application to FSANZ to approve the peas for human consumption. FSANZ gave advice to conduct studies to, amongst other things, fully characterise the novel protein and in particular to determine its potential for toxicity and allergenicity in line with internationally accepted guidelines.

CSIRO has subsequently reported that these protein characterisation studies revealed that a modified form of the alpha-amylase inhibitor protein had unexpectedly been produced in the GM peas. In further studies, seed meal from the GM crops was fed to mice to determine if the modified form of the alpha-amylase inhibitor protein would also have a modified immune response.   The results show that, under the specified experimental conditions, the modified protein has an altered immune response.   However, the results do not show that the mice became allergic to the modified protein in food, nor does the study make any conclusions in relation to its relevance in humans.    

The current approach internationally to the assessment of potential allergenicity is to use ‘an integrated, stepwise, case-by-case approach’.   As no single definitive test can be relied upon to predict allergenic response in humans to a novel protein, a variety of information is considered in the assessment.   At present, the information considered most relevant to the assessment is:

• the source of the novel protein;
• the physicochemical characteristics of the protein;
• similarity of the novel protein to known allergenic proteins; and
• susceptibility of the novel protein to enzymatic degradation in conditions that mimic normal digestion.

By using this ‘weight of evidence’ approach, a conclusion can be made as to the likelihood of the novel protein being a food allergen.

This internationally accepted approach has been elaborated by the Codex Alimentarius Commission, the Food and Agricultural Organisation of the United Nations and the World Health Organisation (WHO), and is followed by FSANZ.   It   uses a variety of data and information, which when considered together can be used to reach a conclusion about potential allergenicity of a new protein. The various animal models that are available are not considered to be sufficiently well developed or validated to use at the present time for this assessment. It is recognized however that, once developed and validated, they will form an important component of the weight of evidence approach.  

As the modified form of the alpha-amylase inhibitor protein has not been subjected to this standard assessment for potential allergenicity, it is not possible to make any conclusions about its potential to be a food allergen in humans.   Furthermore, the animal model used by the study authors has not been validated to predict human immune or allergic responses and the authors make no such predictions.   It is therefore not clear what relevance (if any) the findings have in relation to human food allergy.   Amongst other factors, the genetic composition of the host influences the outcome of exposure to the foreign protein.

While the significance of the research results for human allergenicity is not clear, the CSIRO has decided to end the research program. This type of situation is not unique to the development of GMOs - the development of conventionally bred, non-GM plants have also been terminated when unexpected or adverse effects have been detected.  

FSANZ continues to monitor new scientific information from all sources and internationally agreed methodologies as they becomes available and will take this into account and modify safety assessment requirements where appropriate.