A recent patent in allergy & immunology: Biomarkers on allergen‐specific memory B cells to predict allergen immunotherapy outcome

Allergic diseases are amongst the most common chronic immune-mediated disorders, and can manifest with enormous diversity in clinical severity and range of symptoms that include anaphylaxis, asthma and urticaria. While allergen avoidance and pharmacotherapy (e.g., anti-histamines, adrenaline, corticosteroids) are effective, patients are still at risk of severe reactions from accidental exposure. The only disease-modifying treatment is allergen immunotherapy (AIT), which typically involves repeated exposure to low-doses of allergen over an extended period of time, often many years. Although this approach to redirect the existing pathogenic immune response has been practiced for more than 110 years, its mechanism of action remains poorly defined. Moreover, as patients need to be treated for at least several years for AIT to be effective and durable, and this can be accompanied by adverse reactions, treatment adherence is typically poor. Thus, there is an urgent need for a better mechanistic understanding of AIT, so that this can guide treatment optimization and lead to identification of early biomarkers that can predict treatment outcome. Our patent covers the detection of changes in specific surface proteins on allergen-specific B cells within several months of AIT initiation as biomarkers for successful treatment outcome.1 The acute allergic response is mediated by allergen-specific IgE, which is secreted by plasma cells, that is, terminally differentiated B cells. IgE is a potent activator of immune effector cells, such as mast cell and basophils, and thereby drives the acute allergic response. As IgE-expressing Bmem are very infrequent and short-lived,2 it is presumed that upon allergen exposure, IgG+ Bmem become activated and differentiate into IgE-secreting plasma cells.3 AIT is known to induce an allergen-specific IgG4 response, and in our previous work we reported that this response also included increased serum allergen-specific IgG2, and frequencies of IgG2+ and IgG4+ Bmem.4 Thus, we hypothesized that AIT acts through modulation of allergen-specific Bmem (Figure 1A). To test this, we applied our previously patented (PCT/AU2020/051116) pipeline for production of recombinant protein tetramers for detection and purification of Lol p 1+ Bmem from ryegrass pollen allergic patients before and after 4 months of daily sublingual AIT (SLIT) (Figure 1B).5 These cells were then subjected to single-cell RNA sequencing (scRNAseq) with the BD Rhapsody platform to uncover transcriptional changes. Disappointingly, we initially did not find substantial changes using traditional dimensionality reduction approaches (tSNE, UMAP). As we evaluated a highly homogeneous population of Bmem before and after treatment, we applied a pseudotime analysis as this arranges cells based on their progression through a differentiation process.1, 5 Importantly, pseudotime analysis showed a shift in transcriptional profiles with the majority of cells being in a less mature cluster before treatment towards two distinct clusters after treatment.1, 5 The cells in these latter clusters showed highly significant changes in multiple gene transcripts including 14 encoding surface proteins. The most promising were confirmed at the protein level by flow cytometry: FcεRII (CD23), Integrin β1 (CD29), and IL4Rα (CD124) (Figure 1C), making these candidate biomarkers.1 To further validate these markers, we applied flow cytometry to detect these markers on Bmem from bee venom allergic donors before and after 63 days of ultra-rush subcutaneous AIT (SCIT). In this population with a distinct type of allergy, and undergoing a different AIT procedure, we identified the same changes in surface markers.1 Thus, expression of CD23, CD29 and/or CD124 on allergen-specific Bmem can be a prognostic biomarker for AIT effects already after 2–4 months of treatment. This is valuable information for a treatment that typically requires years of adherence. Furthermore, these markers can be utilized in clinical trials for new treatments or treatment combinations to evaluate their effectiveness as these are objective laboratory markers, and can be evaluated within a shorter time frame than the classical clinical evaluations or exposure tests (typically applied after 1 year). At this stage, we have not evaluated expression of these biomarkers beyond 4 months of AIT. In addition to their application as biomarker, what would be the functional relevance of these proteins? Interestingly, CD23 and IL4Rα have recently been shown to mark a type 2 Bmem population poised to produce IgE, and thereby presumed to be pathogenic in allergy.3, 6 Our cluster 4 seems to represent this published type 2 Bmem population as it is enriched for CD23, IL4Rα, IGHG4 and IGHE transcripts.1, 5 It remains unclear why this presumed pathogenic population would be expanded after treatment.7 Intriguingly, we identified that another population (cluster 3) enriched for IGHG1 and IGHG3 was also expanded following SLIT.1, 5 Both clusters 3 and 4 have increased expression of CD29. As CD29 expression inhibits B-cell antigen receptor signalling,8 it might dampen the capacity of pathogenic, allergen-specific Bmem to respond upon exposure. Thus, our identified candidate biomarkers for evaluation of successful AIT are also new starting points to unravel the mechanism behind desensitization, a major unknown despite more than 110 year of practice. This work was supported by a Central Clinical School Early Career Fellowship to CIM, NHMRC Ideas grant 2000773 to MCvZ and REO'H, and in-kind support of reagents and equipment from BD Biosciences. Open access publishing facilitated by Monash University, as part of the Wiley - Monash University agreement via the Council of Australian University Librarians. This work was supported by a Central Clinical School Early Career Fellowship to CIM, NHMRC Ideas grant 2000773 to MCvZ and REO'H, and in-kind support of reagents and equipment from BD Biosciences. MCvZ, REO'H and CIM are inventors on the presented patent application. None.