Diagnosis and Endotyping
The gold standard for the diagnosis of FA remains the double-blind placebo-controlled food challenge (DBPCFC); however, they are time-intensive with high-risk of severe reaction, necessitating a need for alternative diagnostic techniques (Figure 2). Allergen-specific IgE (sIgE) assays are readily available in clinical settings but have a high rate of false positives as they cannot distinguish true FA from sensitivity without clinical allergy. The ImmunoCAP assay is a fluorescent method which is currently the standard for sIgE quantification. It is sensitive but requires a large amount of blood, which can be problematic when testing young children. A method that has shown to be comparable to ImmoCAP is IMMULITE, a chemiluminescent method.11 Recently, LuLISA, an bioluminescent method which requires 1 μL or less of plasma sample has been published.12 Additionally, a peanut bead-based epitope assay was developed using the LEAP cohort and validated in CoFAR-2 and POISED studies. It uses two sIgE antibodies in sequential fashion to diagnose PA and has demonstrated good sensitivity (92.3%), specificity (94.1%), and accuracy (93.4% concordance with DBPCFC). Although requiring less than 100 µl of plasma/serum and being easily adapted for high-throughput use in clinical labs, gaps in the molecular characterization of non- peanut allergens has limited its use.13
Beyond IgE characterization, the potential of basophil activation tests (BAT) has been increasingly recognized in recent years.14,15 However, the effectiveness of BAT differs significantly between allergens 15,16. Despite variation in sensitivity, BAT has demonstrated high levels of specificity enabling it to complement skin-prick and sIgE tests, which lack specificity but provide considerable sensitivity. Using BAT as a second-round diagnostic test after skin-prick and sIgE pre-screening was shown to reduce the number of required diagnostic OFC by 5-15% for peanut, sesame, and cashew17. The MONAS study found that a single BAT was efficacious in predicting clinical allergy status across peanut (AUROC 0.98), cashew (0.97), hazelnut (0.92), pistachio (0.95), and walnut (0.97), outperforming sIgE testing for peanut and hazelnut in a sub-analysis of sensitized patients undergoing OFC15. In addition to a potential role in the diagnosis of FA, BAT may also predict response to OIT.18 The POISED study found that patients who failed DBPCFCs after a period of desensitization followed by peanut avoidance had higher %CD63high basophils upon peanut stimulation and had significantly higher Ara h 1, Ara h 2, Ara h 3, and sIgE/total IgE than those who passed DBPCFCs19. Classification of patients into “non/low”, “intermediate”, or “high” basophil responders at baseline was additionally able to predict success of DBPCFCs following treatment. Sustained unresponsiveness (SU) to peanut was observed in patients with low basophil activation at baseline and those with a greater than 80% reduction in peanut-induced basophil activation after OIT20, supporting the utility of BAT in predicting and monitoring response to OIT. These new technologies including others using novel gating strategies with optimization of storage and automation of measurements and analysis may enable routine high throughput analysis in the future.21
Mast cell activation tests (MAT) is another in vitro diagnostic test, similar to BAT. The BAT uses whole blood whereas the MAT uses plasma or serum to sensitize mast cells. Expression of activation markers are measured on stimulation with allergen. The MAT has similar specificity in the diagnosis of PA but lower sensitivity.22Ongoing research and novel biomarkers in addition to IgE and basophil/mast activation biomarkers for diagnosis of FA are being developed.23-26
In addition to the inherent risks associated with DBPCFCs, there exists considerable variation across trial design, providers, and academic sites making DBPCFCs challenging to standardize. Several groups, including DeFASe27, Dribin et al.28, and CoFAR29,30, and others31 are attempting to more uniformly approach the characterization of reactions during food challenges through standardized grading scales for FA-related adverse events (AEs). As there is wide variability of AE severity, there is also the push to understand and develop tools for prediction of patient-specific response to DBPCFC32-35. These tools can assist in diagnostic and treatment strategies in those at risk for the most severe reactions.