Skip to main content
Fig. 1 | BMC Medicine

Fig. 1

From: Tissue-infiltrating lymphocytes signature predicts survival in patients with early/intermediate stage hepatocellular carcinoma

Fig. 1

Characteristics of the immune microenvironment and selection of immune features by LASSO analysis in liver cancer. a Relative fractions of 22 leukocyte subsets across 8 datasets estimated by CIBERSORT. b Comparison of immune cells between neoplastic and adjacent tissues in the GSE14520 and TCGA datasets. *, **, ***, and **** denote P < .05, < 0.01, < 0.001, and < 0.0001, respectively. NS denotes no significance (Student’s t test). c, d Correlation matrix followed by unsupervised hierarchical clustering in immune cell fractions of TCGA (c) and 28 immune features of HCC tissues (d). Pearson correlation coefficients (R) were calculated. Correlation coefficients were plotted with negative correlation (green), positive correlation (red), and R = 0 (black). e, f Five immune features selected by LASSO Cox regression analysis. Left panel: the two dotted vertical lines are drawn at the optimal values by minimum criteria (e) and 1-s.e. criteria (f). Right panel: LASSO coefficient profiles of the 28 immune features. g TRIS distribution of the 5-immune features in the training and validation cohorts. Upper panel: TRIS distribution of the five-immune-feature-based classifier and patient survival status. Lower panel: heatmap presenting density of the five immune features in HCC patients. h IHC expression pattern of selected immune features, including CD3T, CD27T, CD103T, PD1T, and CD68P in 4 different patients. Bar, 20 μm. T is for tumor specimens and P for peritumor specimens

Back to article page