Fig. 7

Detailed characterization of DCs in HPSCC. a UMAP plot of the DCs colored by cell (up) and sample (down) types. b Heatmap showing the signature DEGs among four distinct DC subsets. c Violin plots showing the expression of marker genes in the DC subsets. d Violin plot showing the expression of immune-suppressive genes in four distinct DC subsets. e Dot plot representative of the activation, migration, and tolerogenic signatures of DCs [Z-score normalized log₂ (count + 1)]. f Pseudotime trajectory analysis of DCs. Each dot represents one single cell, colored according to its cluster label. The inlet plot shows each cell with a pseudotime score from dark blue to light blue, indicating an early and terminal state, respectively. g RNA velocities are visualized on the UMAP projection of DCs using Gaussian smoothing on a regular grid. h Violin plot showing the expression of the lymphocyte recycling chemokines CD274 and PDCD1LG2 in the major HPSCC cell types. i Kaplan–Meier curve of the OS in the TCGA-HNSC cohort stratified by the optimal cut-off point for the LAMP3⁺ DC infiltration level. P-values were calculated using the two-sided log-rank test. j Expression of PD-L1, CCL19, and CD83 on LAMP3⁻DCs or LMAP⁺DCs in tumor tissues (n = 7) was analyzed using flow cytometry. k Flow cytometry of LAMP3⁺ DCs infiltrated in the tumor and corresponding adjacent tissues (n = 4). l Correlation analysis of the PD-L1⁺LAMP3⁺ DCs and PD-1⁺CD8 T cells, LAMP3⁺ DCs, and Tregs infiltrated in the HNSCC tumor tissue (n = 7) using Spearman rank. R: correlation coefficient. Bars and errors are represented as mean ± SEM; data were analyzed using unpaired or paired t-tests (*P < 0.05; **P < 0.01, ***P < 0.001)