Date of this Version
Minimal eliciting doses for objective allergic reactions were found for 13 priority allergens and over 1800 individuals from published clinical literature or unpublished clinical data. Allergic populations did not vary when analyzed by age or geographic region. Results of this study show there are sufficient clinical data from food allergic individuals to use for risk assessment purposes for several allergenic foods.
Of 186 food products bearing advisory statements regarding peanut or 16 products that had peanut listed as a minor ingredient, 8.6% and 37.5% contained detectable levels of peanut (>2.5 ppm whole peanut). An additional market survey of 215 nutrition bars with peanuts as a minor ingredient and/or an advisory statement for peanuts found 24.6% tested positive for peanut compared to 4% of products with no mention of peanuts on the label. Probabilistic risk assessment showed the risk of reaction among peanut allergic consumers from advisory labeled nutrition bars was significant but brand-dependent. The probabilistic approach provides the food industry with a quantitative method to assist with determining when advisory labeling is most appropriate.
Agricultural commodity cross-contamination of soybean was detected in 62.8% of samples representing all forms of wheat flour. Conservative probabilistic risk assessments predict a risk of allergic reaction occurring in the most sensitive soy-allergic individuals. Experimental milling and stream separation with spiked soy in wheat samples did not produce a soy-free wheat flour stream. Additional cleaning measures will be needed to remove soy before wheat milling begins.
LC-MS/MS identified fourteen known allergens in industry representative soy product samples, including all subunits of Gly m 5 (β-conglycinin) and Gly m 6 (glycinin), the Kunitz trypsin inhibitor, Gly m Bd 28K, and Gly m Bd 30K. Method refinements or different techniques might be necessary to detect low abundance proteins such as Gly m 3 and 4. The relative amount of an allergen in a sample correlated positively to the intensity of IgE binding at the expected molecular weight using sera from soy-allergic individuals.
Advisers: Stephen L. Taylor and Joseph L. Baumert