Fig. 4

Specialized tumor-infiltrating NK cell subsets associate to distinct immune contextures in human breast tumors. Unsorted multicellular suspensions derived from fresh breast tumor specimens were cultured in complete medium with IL2 in the presence or absence of trastuzumab (Tz; 210 ng/ml) for 24 h. A Dot plots showing the NK cell gate (CD56⁺CD3^-) used for the analysis of the expression of CD16 and CD103 in TI-NK cells of three representative samples. B-C Proportions of CD137⁺ cells in CD16⁺, CD16^-CD103⁺ and CD16^-CD103^- subpopulations in fresh breast tumor-derived multicellular cultures treated or not with trastuzumab. D-G Fresh breast tumor specimens (n = 84) were processed and stained with a combination of antibodies specific for CD45, CD3, CD56, CD8, CD4, CD16, CD103, NKG2C and PD1 and analysed by flow cytometry. D tSNE of major alive lymphocyte subsets (DAPI^-CD45⁺) in breast tumor immune infiltrates generated with data from 18 tumors. Red, blue and green circles indicate major NK, CD4 and CD8 T cell subsets, respectively. E Percentage of CD16⁺ , CD16^- CD103⁺ and CD16^-CD103^- TI-NK cell subpopulations in treatment naïve, fresh tumor samples. F, G Heat maps showing Spearman’s correlation coefficients between the indicated lymphocyte subsets in breast tumors (n = 73–84) and HER2 tumors (n = 20). Asterisks label significant correlations (**** p < 0.0001; *** p < 0.001; ** p < 0.01; * p < 0.05