The diagnosis of IgE-mediated wheat allergy was made if they have one of the following criteria

1. a convincing clinical history of the reactions within 4 hours after wheat ingestion during the past 12 months combined with positive skin prick test (SPT) and/or the level of specific IgE (sIgE) to wheat or
2. a positive oral food challenge (OFC) result to wheat during the past 12 months

Patients with wheat tolerant confirmed by negative oral food challenge (OFC) result to wheat during the past 12 months

Patients with IgE-mediated wheat allergy and underwent wheat oral immunotherapy for at least 6 months or at the maintenance phase of the treatment

Blood drawn will be done once using 15 mL of blood volume (Serum or plasma samples will be collected during the routine follow up of level of sIgE to wheat or during the oral food challenge test which intravenous insertion routinely prepared in case of the emergency reaction occurred). The specimen will be transferred to the Icahn School of Medicine at Mount Sinai, New York, USA for Laboratory processing (Luminex-based peptide assay)

For Luminex data, the binding intensity of antibody to epitope specific beads was calculated from the florescence signal of each bead. An index score of binding intensity was generated from the log2 transformation of the signal-to-background ratio and each peptide element, which was close to a normal distribution. A plate effect was observed and adjusted for using linear models. Antibody binding intensities (Luminex) from different groups were compared. To assess the changes in epitope-binding profiles, a linear model was used with group as a factor. Comparisons of interests were tested using t tests and resultant P-values were adjusted for multiple hypotheses using the Benjamini-Hochberg approach, which controls the false discovery rate (FDR) across epitopes. Epitopes were defined as differentially binding epitopes (DBE) if the FDR < 0.05 and fold changes (FCH) > 1.5.

For Luminex data, the binding intensity of antibody to epitope specific beads was calculated from the florescence signal of each bead. An index score of binding intensity was generated from the log2 transformation of the signal-to-background ratio and each peptide element, which was close to a normal distribution. A plate effect was observed and adjusted for using linear models. Antibody binding intensities (Luminex) from different groups were compared. To assess the changes in epitope-binding profiles, a linear model was used with group as a factor. Comparisons of interests were tested using t tests and resultant P-values were adjusted for multiple hypotheses using the Benjamini-Hochberg approach, which controls the false discovery rate (FDR) across epitopes. Epitopes were defined as differentially binding epitopes (DBE) if the FDR < 0.05 and fold changes (FCH) > 1.5.