To the Editor,
Reliable early biomarkers are needed to monitor allergen immunotherapy (AIT) efficacy in allergic rhinitis (AR). Type 2 memory B cells, marked by IL‐4Rα and CD23, are implicated in antigen‐specific Immunoglobulin E (IgE) production and promote enhanced antigen recall responses in allergic disease [1, 2]. However, the precise dynamics of type 2 memory B (MBC2) cells during AIT remain poorly understood and their potential utility as candidate biomarkers for treatment monitoring has not been systematically evaluated.
We enrolled 35 house‐dust‐mite (HDM)‐sensitized AR patients receiving subcutaneous immunotherapy (SCIT) (Table S1). Peripheral blood was collected at baseline (V1), week 15 (end of up‐dosing, V2), and 1 year (V3). Responders were defined as those with > 50% reduction in daily combined symptom and medication score (CSMS) from baseline to 1 year [3, 4].
Type 2 switched memory B (sBmem) cells were markedly elevated in the AR versus healthy controls (p = 0.0013) but declined significantly after 2 years of AIT (p = 0.0036) (Figure 1A,B), indicating modulation of this subset by AIT. Within the AIT cohort, responders showed an early decline in type 2 sBmem cells (Figure 1C, p = 0.007; Figure S2A–C) and the reduction from V1 to V2 was significantly greater in responders than in non‐responders (Figure 1F, p = 0.0001). This underscores that monitoring the dynamic changes in memory B cell subsets during the early stages of AIT may better reflect the evolution of the immune response and associate with subsequent clinical outcomes than reliance on baseline characteristics alone.
No significant differences in Dermatophagoides pteronyssinus (Der p) or Dermatophagoides farinae (Der f) specific IgE or IgG4 were observed between responders and non‐responders (Figure S2D,E). By contrast, responders had a higher specific‐IgE/total‐IgE ratio at baseline and at V2 (Figure 1D,E) and showed a significant V2 increase in serum IL‐10 versus baseline (Figure 1G; p = 0.04). At V2, higher type 2 sBmem proportions correlated with worse clinical outcomes (higher CSMS), whereas type 2 sBmem frequencies showed an inverse correlation with IL‐10 levels, which in turn correlated inversely with CSMS (Figure 2A).
These inverse relationships align with previous reports [5] that AIT induces IL‐10‐producing, often IgG4‐expressing B cells associated with clinical benefit. Although our data are correlative, they suggest that an IL‐10 rich milieu contributes to contraction or functional reprogramming of the pathogenic MBC2 compartment during successful AIT. Type 2 sBmem levels correlated negatively with the specific‐IgE/total‐IgE ratio but showed a positive correlation with Der p‐specific IgE in responders (Figure 2B), potentially reflecting the characteristic early kinetics of allergen‐specific IgE during AIT.
Importantly, the V1 to V2 change in type 2 sBmem was also significantly associated with CSMS improvement (Figure 2C), indicating that this subset tracks both immunologic parameters and clinical responses. These data underscore the value of longitudinal immune readouts over static measurements. Using Generalized Estimating Equations (GEE) that incorporated time, we found that declining type 2 sBmem levels together with rising IL‐10 were significantly associated with improvements in CSMS (p = 0.005, Table S3).
Furthermore, logistic regression analysis demonstrated that the combination of type 2 sBmem frequency and the specific‐IgE/total‐IgE ratio was significantly associated with responder status (p = 0.032, Table S4).
Prior work reported reduced CD23 expression on sBmem after 12 months of HDM‐SCIT, correlating with remission [6]. However, von Borstel et al. reported an early expansion of phenotypically altered MBC2 during bee‐venom SCIT and ryegrass SLIT, consistent with venom SCIT data showing increased CD23 and CD29 expression, IgG2/IgG4 skewing, and epitope spreading in allergen‐specific memory B cells [7, 8]. These discrepancies may reflect differences in sampling time points, AIT regimens, allergens, and subset definitions.
A parsimonious explanation is that early AIT reshapes MBC2 phenotype, favoring cells that produce IgG4 rather than IgE, while simultaneously depleting the memory B‐cell subset most closely linked to IgE production. Consistent with this, a reduction in type 2 sBmem appears more closely associated with clinical benefit during HDM SCIT.
Given that Koenig et al. [1]. demonstrated that only approximately 50% of MBC2s express CD27, we performed a parallel analysis using a CD27‐independent gating strategy. The results were consistent with our primary analysis, confirming the robustness of our findings (Figure S3). A principal limitation of this study, however, is the relatively small sample size, which may limit the generalizability of our conclusions; validation in larger cohorts is therefore warranted.
Furthermore, future work should aim to quantify HDM‐specific memory B cells directly and rigorously investigate IL‐10‐mediated regulation of MBC2s using integrated multi‐omics approaches.
To our knowledge, this is the first study to longitudinally characterize type 2 sBmem dynamics in HDM‐sensitized AR patients over a 1‐year course of AIT. Our findings suggest that early changes in type 2 sBmem frequencies may serve as a clinically relevant biomarker for AIT efficacy. If validated in larger, multi‐center cohorts with harmonized flow cytometry protocols, type 2 sBmem measurement could be deployed as a routine immune monitoring tool to identify early non‐responders and guide treatment optimization.
Furthermore, compared to the detection of allergen‐specific memory B cells, type 2 sBmem assessment is technically more accessible and amenable to broad clinical implementation.
Author Contributions
H.Lo. and H.Li. designed the study. Y.M. and S.L. wrote the manuscript. L.Q., J.W., J.L., X.F., and W.Q. contributed to data collection.
Y.M., L.Y., S.L., and X.Y. performed the statistical analysis and interpretation of the results. All authors read and approved the final manuscript.
Funding
This study was supported by the National Natural Science Foundation of China (82271139, 82271138, 82000955, 82521001) and the National Key R&D Program of China (2022YFC2504100).
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information