Therapeutic efficacy and underlying transcriptional mechanisms of the BRAF inhibitor. (A) Pie charts showing the response of 22 LCH patients after 1 (i) and 3 months (ii) of treatment with dabrafenib. Briefly, treatment response was categorized as complete resolution (nonactive disease/NAD), regression (active disease/AD-better), lesion stable or regression with new involvement (AD-intermediate), and progression (AD-worse). (B) Line graph showing plasma cfBRAF^V600E load kinetics across time during dabrafenib treatment, with the median load of cfBRAF^V600E at each time point shown (upper right). The number of samples are shown below the corresponding time points. Wilcoxon rank sum test is used to test the significance of difference, and P value is indicated for the comparison. The plasma cfBRAF^V600E load significantly decreased after 1 month of treatment (therapy onset vs 1 month), whereas there is no significant difference in cfBRAF^V600E load between any subsequent time points across individuals. (C) tSNE visualization of pre- and posttherapy samples with clusters projected (n = 1261 cells total). Red and blue dots represent cells of pre- and posttherapy samples, respectively. (D) Volcano diagrams showing the DEGs in CD14⁺ Mo (i) and CD16⁺⁺ Mo (ii) of pre- and posttherapy samples. Orange and blue dots represent upregulated DEGs in pre- and posttherapy samples, respectively. Red dots represent DEGs involved in the RAS-MAPK-ERK pathway. Having investigated the expression of interested genes across individual patient, only the ones with significantly differentially expressed in at least 2 patients before and after 1-month treatment are selected to visualization. (E) GSEA for DCs and monocytes of pretherapy samples compared with posttherapy samples. Pathways enriched in pretherapy samples with nominal (NOM) P < .05 and false discovery rates (FDR) q < 0.20 are selected for visualization. Links represent different enrichment pathways and the numbers correspond to the pathways listed in the boxes on the right.