(A) Schematic of the timeline of the clinical study design, showing omalizumab injections from weeks 1–16 and OIT from weeks 8–30. Blood samples were drawn at baseline and week 30. (B) Mass cytometric gating strategy for NK cell subsets. (C) Total NK cells (left), CD56^dimCD16⁺ NK cells (middle) and CD56^brightCD16^dim/– NK cells (right) as a percentage of total PBMCs cultured with or without peanut protein at baseline and week 30 after OIT from 23 PA participants. (D) Representative flow cytometry plots gated on NK cells showing the effect of OIT on CD69⁺NK, NKp30⁺NK, and CD107a⁺NK for 1 PA participant. (E) Percentage of CD69⁺NK in NK cells shown for 6 PA participants before and during OIT. (F) The degree of increase in the CD69⁺ NK cells in response to peanut stimulation was calculated by subtracting the percentage of this population in the unstimulated sample from that in the corresponding stimulated sample. The degree of peanut-induced increase in CD69⁺ NK cells was examined before and during OIT for 6 PA participants. (G) Percentage of NKp30⁺NK (left) and CD107a⁺NK (right) in NK cells shown for 6 PA participants before and during OIT. Since the PBMC sample from 1 of the 6 PA participants at the 300 mg peanut OIT was not available, we only have 5 readouts at this timepoint. Box plots indicate the IQR and median; whiskers extend to the farthest data point within a maximum of 1.5 × IQR. Paired sample sets were analyzed using a 2-sided Wilcoxon signed rank test. Each pair of points connected by a line represents 1 participant. Unpaired sample sets were analyzed using a 2-sided Wilcoxon rank sum test. The stars indicate the P values, **P < 0.01.