Challenge-verified thresholds for allergens mandatory for labeling: How little is too much for the most sensitive patient?

Background: It is mandatory to label food products with the 14 main allergens in the EU. Reasonable allergen labeling requires knowledge of population-based thresholds derived from food challenges. The aim of this study was to evaluate the threshold-distribution in clinically verified food allergic patients for allergens mandatory for labeling. Methods: All positive open oral food challenges and double-blind placebo-controlled food challenges (DBPCFC) performed at the Allergy Center, Odense University Hospital, Denmark (2000–2022) were included. For each included challenge, the cumulative threshold (LOAEL) was obtained and NOAEL estimated. Data were modelled as an interval censored log-normal distribution. Results: Overall, 38 of all 2612 challenges (1.5%) in 1229 patients (717 male, 986 children) reacted to <5 mg protein. The majority of the most sensitive patients reacted with a Sampson severity score of 2–3. Using interval censored log-normal models only five groups (hens´ egg, fish, peanut, milk, tree-nuts) elicited reactions after ingestion of 0.5 mg protein and in low frequencies of the population. Hen's egg was the most potent allergen, with reactivity to <0.5 mg protein in 0.24% [0.13–0.44%] of egg allergic patients while the estimated fraction of allergic patients reacting to a eliciting dose on 0.5 mg protein for most other allergens were below 0.04%. Conclusion: Our data demonstrates that the majority of food allergic patients as expected tolerating traces of allergenic foods without developing severe allergic symptoms and signs. Hen's egg appears to be the food most likely to elicit reactions in the most sensitive individuals at very low doses.