Following a request from the Food Safety Authority of Ireland, the EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA Panel) was asked to deliver a scientific opinion on the evaluation of allergenic foods and food ingredients for labelling purposes. In view of the request, the NDA Panel decided to update its previous opinions relative to food ingredients or substances with known allergenic potential listed in Annex IIIa of 2003/89/EC, as amended. These include cereals containing gluten, milk and dairy products, eggs, nuts, peanuts, soy, fish, crustaceans, molluscs, celery, lupin, sesame, mustard and sulphites. The opinion relates to immunoglobulin (Ig)E- and non-IgE-mediated food allergy, to coeliac disease and to adverse reactions to sulphites in food, and it does not address non-immune-mediated adverse reactions to food. It includes information on the prevalence of food allergy in unselected populations, proteins identified as food allergens, cross-reactivities, the effects of food processing on the allergenicity of foods and ingredients, methods for the detection of allergens and allergenic foods, doses observed to trigger adverse reactions in sensitive individuals and risk assessment methodologies that have been used to derive individual and population thresholds for selected allergenic foods.
© European Food Safety Authority, 2014
Following a request from the Food Safety Authority of Ireland, the EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA Panel) was asked to deliver a scientific opinion on the evaluation of allergenic foods and food ingredients for labelling purposes.
In view of the request, the NDA Panel decided to update its previous opinions relative to food ingredients or substances with known allergenic potential listed in Annex IIIa of Directive 2003/89/EC, as amended, which include cereals containing gluten, milk and dairy products, eggs, nuts, peanuts, soy, fish, crustaceans, molluscs, celery, lupin, sesame, mustard and sulphites. In this context, EFSA launched a procurement project (CT/EFSA/NDA/2012/02) to review published data on the prevalence of food allergy in Europe and to gather prevalence data on food allergy in the general (unselected) population.
The present opinion relates to immunoglobulin (Ig)E- and non-IgE-mediated food allergy, to coeliac disease and to adverse reactions to sulphites in food, and it does not address non-immune-mediated adverse reactions to food. For each food ingredient or substance listed in Annex IIIa, it includes information on the prevalence of food allergy in unselected populations, on proteins identified as food allergens, on cross-reactivities, on the effects of food processing on the allergenicity of foods and ingredients, on methods for the detection of allergens and allergenic foods, and on doses observed to trigger adverse reactions in sensitive individuals.
Immune-mediated adverse reactions to foods manifest with clinical signs and symptoms of variable severity and duration, which may affect different organs and systems. Anaphylactic reactions to food are IgE mediated and may occur at any age. Non-IgE-mediated food allergy includes a wide range of diseases, including protein-induced enterocolitis and eosinophilic oesophagitis.
A careful family and clinical history are the basis for diagnosis of food allergy. Food diaries, skin prick tests (SPTs), allergen-specific IgE measurements, food elimination diets and food challenges are part of the standard protocol for the diagnosis of food allergy. A positive SPT indicates sensitisation to the tested food, but it is not diagnostic of food allergy. Allergen-specific serum IgE antibodies similarly denote sensitisation to a particular food, but they are not diagnostic without a clinical history or food challenge. The use of atopy patch tests for the diagnosis of food allergy is controversial. Other available tests have no current role in the diagnosis of food allergy. Diagnosis is confirmed by exclusion of the suspected food and the subsequent amelioration of symptoms and by the recurrence of symptoms on re-introduction of the offending food, ideally in double-blind placebo-controlled food challenges, provided that the initial symptoms were not life threatening. The Panel notes that there is a need for standardisation of derived allergens for SPTs. The Panel also notes that guidelines aiming to standardise oral challenge protocols for the diagnosis of food allergy are now available. Dietary avoidance of specific allergenic foods in combination with nutritional advice is the mainstay of management in IgE- and non-IgE-mediated food allergy. Food-allergic individuals may occasionally outgrow their allergy later in life.
The prevalence of food allergies in developed countries is uncertain. The scarcity of studies available for some geographical areas and the use of different methodologies across studies to retrieve prevalence data are the main reasons for this uncertainty. Using food challenges as a criterion for diagnosis, the prevalence of food allergy has been estimated to be around 3 %, when considering data from Europe, the USA and Australia/New Zealand, and about 1 % when considering European studies only, in both adults and children. However, the heterogeneity among the studies used to estimate the prevalence of food allergy was high. There are insufficient objective data to conclude on time trends with respect to the prevalence of food allergy in Europe. About 75 % of allergic reactions among children are due to egg, peanut, cow’s milk, fish and various nuts. About 50 % of allergic reactions among adults are due to fruits of the latex group and of the Rosaceaefamily, vegetables of the Apiaceae family, and various nuts and peanuts.
Geographical variation in the prevalence of food allergy is due to differences in in environmental (e.g. pollen exposure or differences in food habits) and individual factors. Sex, age, family history of atopy and the presence of other allergic diseases are among the individual factors considered important in the development of food allergy. Extrapolations of prevalence data on specific food allergies from a single European country to the entire European population are of limited accuracy owing to differences in genetic background, exposure to the offending foods and eating habits.
Owing to the development of proteomics, spectroscopic methods and gene cloning, allergenic proteins can be well characterised. They have been classified into families based on their sequence and three-dimensional (3D) structure. However, although common structural features of proteins and their biological activity have been tentatively related to their immunogenicity, it is not possible to predict the allergenicity of a protein based on these two parameters only. Immunological and clinical data are also required to classify a protein as a food allergen.
Cross-reactivity occurs when IgE antibodies originally triggered against one antigen also bind a different antigen. Not all cross-reactivities identified in vitro are of clinical significance, and although most clinical cross-reactions are mediated by IgE antibodies, T cells may also be involved. However, in vitro cross-reactivity testing can help understanding allergenicity to multiple foods, as well as improving diagnosis and management of food allergy.
The allergenic activity of a food may decrease, remain unchanged, or even increase by food processing. Considering the multiplicity of the allergenic proteins contained in a whole food, and that different proteins may be differently affected by the same treatment, the impact of food processing on the structural and allergenic properties of allergenic foods/ingredients is difficult to predict. In addition, the extent to which allergenic proteins are modified during food processing depends on the type of process and its conditions, the structure of the proteins, and the composition of the matrix. Although the effects of different (technological and cooking) treatments on the IgE-binding capacity of several allergens have been investigated, less information is available on the effects of processing on clinical reactivity.
The majority of kits commercially available for routine food allergen analysis rely on immunological methods. Enzyme-linked immunosorbent assay (ELISA) methods are the most widely used because they are sensitive and specific for the detection of allergenic proteins and easy to use. However, commercial kits for quantitative analyses employ different extraction buffers and calibration procedures, they differ in the quality of the antibodies used and the results vary among commercial brands and batches. Major limitations include matrix effects, insufficient extraction of the protein, insufficient specificity due to cross-reactions and insufficient reproducibility of results. The use of incurred samples may help to improve the reliability of the method when analysing processed foods.
Mass spectrometry (MS), in combination with techniques such as 2D sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) or chromatography for the preliminary separation of proteins, and using allergen databases for their subsequent identification, is a reliable tool for the detection of known allergens and for the identification of new immunoreactive proteins. MS methods for quantitative analysis based on specific standard peptides or stable isotope labelling are not yet suitable for analyses of large numbers of samples but can confirm results obtained otherwise.
DNA methods allow detection of the allergenic food rather than of the allergenic protein and are complementary to immunological assays. DNA is generally more stable than proteins and thus suitable for analysis of processed foods. The extraction and amplification procedures are well established. Both end-point and real-time polymerase chain reaction (PCR) allow simultaneous multiple analyses. Whenever ELISA kits are not available or not specific for the analysis of a specific allergenic food/ingredient (e.g. celery), DNA analysis becomes the method of choice. Real-time PCR may provide quantitative results and allows multiplexed analysis. Commercial kits are available.
The main problem for the quantification of allergens by immunological or DNA-based methods is the unavailability of certified reference materials (CRMs). Reference materials (not certified) developed by different producers are commercially available for the most important food allergens, but the results obtained may not be comparable. To the Panel’s knowledge, a CRM for the detection of food allergens has been developed only for peanuts. For milk and egg, two reference materials are commonly used. Availability of CRMs for the quantification of food allergens is required.
The notion of determining concentrations of allergenic foods/ingredients in foodstuffs below which the majority of sensitised consumers are not at risk of developing severe allergic reactions has attracted much attention from regulatory bodies, consumer associations and industry throughout Europe. To that end, attempts have been made to define a framework for food allergen risk assessment, as it exists for the risk assessment of other food-related hazards (e.g. chemicals, microbiological agents). Three different approaches have been proposed for allergen risk assessment: (i) the traditional risk assessment using the no observed adverse effect level (NOAEL) and uncertainty factors; (ii) the Bench Mark Dose (BMD) and Margin of Exposure (MoE) approach; and (iii) probabilistic models. These approaches may be used to inform different risk management decisions for allergen labelling. The reliability of the risk estimates will depend on the type, quality and amount of data used, to estimate both population thresholds (or threshold distributions) and exposure to the allergenic food/ingredient. The purpose of the risk assessment (e.g. exemptions from labelling, labelling of allergens unintentionally present in food) and the level of risk that may be acceptable (e.g. the fraction of the allergic population that is intended to be protected and to what extent) are risk management decisions, which are outside EFSA’s remit.
Coeliac disease is a life-long autoimmune systemic disorder triggered by gluten and similar cereal storage proteins present in wheat, rye and barley. Its prevalence is estimated to be 0.5 to 1 %. A gluten-free diet is the conventional treatment. The limit values of 20 and 100 mg/kg of gluten in “gluten-free” and “very low gluten” foods, respectively, help in managing the diet of most patients with coeliac efficiently.
Labelling of foods containing sulphiting agents in concentrations > 10 mg/kg or 10 mg/L is mandatory in the EU; those levels were based on the limit of detection (LOD) of the analytical methods available at the time. Many very sensitive and reliable methods are now available for analysis of sulphites in foods, with LODs well below 10 mg/kg. However, minimum doses eliciting adverse reactions to sulphites have not been systematically assessed, and the lowest concentration of sulphites able to trigger a reaction in a sensitive person is unknown. (Source: http://www.efsa.europa.eu/)