Model Integrated Risk for Allergy, Bayesian Estimation for Life quality (MIRABEL)
This third-party funded project is conducted in the framework of the BfR research program for functional analytic and early risk detection.
The MIRABEL project consisted of four distinct but integrated scientific tasks. The first task corresponded to a field survey to collect data on the peanut allergic consumers’ food consumption habits, their attitudes toward actual precautionary labelling and willingness to pay (WTP) relating to labels depending on allergen concentration limits, and their threshold of allergic reaction. The survey was designed and implemented by partners 1 (ANSES/DSA), 2 (Inra) and 3 (Allergo-Vigilance Network). The second task corresponded to a field survey to acquire data on peanut contamination of food. Analysed products were selected in regards of the products declared consumed by peanut allergic sufferers in the survey realized in task 1 and analytical feasibility. The design of the survey involved partner 1 and was implemented by service providers. Task 3 consisted of the developments of a probabilistic Bayesian model in order to quantify the risk due to adventitious peanut allergen in food. The model was applied on the data collected during tasks 1 and 2 and was used to assess the impact on the risk of different management options. Task 4 consisted of the cost-benefit analysis for the different stakeholders (food industry, allergic individuals and regulators). The analysis was developed with results of the allergic consumers’ behaviour acquired during task 1.
Part of the BfR
BfR analyzed 900 food samples for the presence of peanut allergen. Samples represented a broad range of products including bread and bakery goods, chocolate and sweets, snacks and crackers, ice cream, nut spread and breakfast cereals. In a stepwise procedure all samples were screened using a sensitive lateral flow assay with a limit of detection (LOD) of 2 ppm total peanut, respectively 0.5 ppm peanut protein. Positive as well as suspect samples (140/900) were confirmed by a real-time PCR method of comparable sensitivity. Finally, two different commercial ELISA quantified positives in both approaches. 1 % (10/900) out of all samples did contain measurable peanut DNA and protein traces above the LOD of the methods applied. Six samples had a content of total peanut protein < 5 mg/kg, two samples between 8 – 10 mg/kg and one sample about 20 mg/kg. The highest contamination was measured for a cheese cracker. An excellent correlation was found between Ct-values obtained by PCR and ppm peanut calculated by ELISA. The reliable detection down to 2 ppm total roasted peanut in the samples was possible with both methods.
- Crepet et al. 2014. Regulatory Toxicology and Pharmacology. pii: S0273-2300(14)00299-2. DOI: 10.1016/j.yrtph.2014.12.006.