Food sensitization and allergy: Defining phenotypes and their risk factors and biomarkers

Symptoms of food allergy, caused by IgE antibodies against food proteins, can vary from mild to life-threatening, affecting multiple organs. The gold standard for food allergy diagnosis is the oral food challenge, but these are burdensome and potentially dangerous. Diagnosis can be supported by detection of serum IgE against food extracts, but IgE sensitization can occur in the absence of clinical food allergy. It is even more challenging to predict the risk for severe reactions. Strategies to better estimate this risk, without needing provocations, are the main topic of this thesis. The studies herein mostly used data on sensitization patterns and clinical history of patients with hazelnut, walnut and peanut allergy, collected in EuroPrevall, the largest European survey on food allergy. Observed geographical variations in molecular sensitization profiles were strongly related to birch pollen exposure which was high in Northern and Central Europe and low in Southern Europe. This translated into sensitization to birch pollen cross-reactive allergens in the former and to primary food allergens in the latter areas. Statistical methods were used to develop models to predict severity of food allergic reactions, based on combinations of patient history and IgE sensitization profiles against allergen molecules and extracts. Models significantly improved the diagnostic performance compared to IgE tests alone. Interestingly, skin related factors from clinical history contributed to the models for all three foods. Although there still remains quite some overlap between patients with mild and severe symptoms, the models are a good starting point for further development and validation.