Multiomics approaches disclose very-early molecular and cellular switches during insect-venom allergen-specific immunotherapy: an observational study

Allergen-specific immunotherapy (AIT) induces immune tolerance, showing the highest success rate (>95%) for insect venom while a much lower chance for pollen allergy. However, the molecular switches leading to successful durable tolerance restoration remain elusive. Here we applied a multilayer-omics approach to reveal dynamic peripheral immune landscapes during AIT-initiation phase in venom allergy patients (VAP) versus pollen-allergic and healthy controls. Already at baseline, VAP exhibited altered abundances of several cell types, including classical monocytes (cMono) and CD4/CD8 type 1-type 17 hybrids. At 8-24h following AIT launch in VAP, we identified a uniform AIT-elicited pulse of late-transitional/IL-10-producing B cells, IL-6 signaling within Th2 cells and non-inflammatory serum-IL-6 levels. Sequential induction of activation and survival protein markers also immediately occurred. A disequilibrium between serum IL-6 and cMono in VAP baseline was restored at day seven following AIT launch. Our longitudinal analysis discovers molecular switches during initiation-phase insect-venom AIT that secure long-term outcomes. Trial number: NCT02931955. Here we administered allergen-specific immunotherapy (AIT) in two groups of patients allergic to insect venom or pollen. For 18 Venome allergy patients (VAP), we implemented an ultra-rush (only within eight hours) up-dosing protocol while we administered a conventional (within six weeks) up-dosing AIT in 16 PAP. For the ultra-rush protocol, the physcians injected 14 times within 8 hours with the following dose (0.003, 0.006, 0.015, 0.030, 0.060, 0.150, 0.30, 0.60, 1.50, 3.00, 6.00, 15.00, 30.00, 55.00) ug. Although the exact sampling periods are different between VAP and pollen allergy patients (PAP), our sampling schemes allowed us to cover comparable periods, essentially the build-up phase, until the achievement of the expected cumulative maintenance dose for both types of AIT. We performed time-series sampling of peripheral blood during the first seven days (0h, 8h, 24h and D7) in VAP. As for PAP, we measured at 0h, 24h and extended our sampling period to two weeks (2W), six weeks (6W) and 12 weeks (12W). To benchmark with the natural immune fluctuations, we also recruited 10 healthy controls (HC), who did not receive any AIT or other immunotherapy, and sampled seven times (0h, 8h, 24h, D7, 2W, 6W and 12W) within 12 weeks to match various time points of both patient groups. For the samples collected at each time point, we performed a multi-layer cellular and molecular omics (multi-omics) analysis, including single-cell mass cytometry (CyTOF) for deep immunophenotyping of 77 immune subsets, FACS-sorting-enabled Th2-cell-type-specific whole-genome RNA sequencing (RNA-seq), kinome analysis of ~350 different kinases, multiplexing analysis of 10 serological cytokines and conventional 17-parameter whole-blood-count analysis. Our multi-omics analysis was accompanied by a detailed clinical workup in all the patients. For details of various immunological datasets from each participant at each time point, please visit our interactive resource website 'i3Dare' (https://public.tableau.com/app/profile/lihpublicdata/viz/i3Dare_SYSTACT_Database/SYSTACTHome) that allows investigators to explore the datasets and figures in an interactive and interlinked manner. While the age of the PAP was comparable to that of HC (median, ∼37 and ∼35 years, respectively), VAP were older (median, ∼49 years). In the study design, we did not consider the sex ratio as the recruitment criteria for this real-world observational study. While both patient groups included male and female participants at a comparable ratio (PAP: 8 males, 8 females; VAP: 11 males, 7 females), the HC group was predominantly female (9 of 10). For demographic details, please refer to the manuscript.