Fig. 5: Integrating spatial proteomics with multiplexed imaging uncovers unique TME interactions.
A Scatter plots show Pearson’s correlation between log2 protein expression values and blood vessel percentages in the P-ROI and the surrounding ring area. Each dot represents a valid data point. The solid line represents the linear regression fit, with the grey shaded area representing the 95% confidence interval. Pearson’s r, two-sided p-value from a t-test on the correlation coefficient, and FDR-adjusted p-values are shown. B Representative immunofluorescent images after cell segmentation and cell type classification show the spatial distribution of CD3+ T cells (cyan), CD68+ macrophages (magenta), panCK+ cancer cells (green), and other cell populations (red). Among 47 tumor slides from 11 patients, patient BC1 is shown as an example. The immune phenotype classification of each region is indicated. Scale = 200 µm. C Dot plot example of immune phenotypes of BC1 based on percentages of CD3+ T cells and CD68+ macrophages within the P-ROIs and surrounding rings. D Heatmap of 414 significantly changing proteins associated with immune phenotypes, as determined by a three-way ANOVA, not confounded by Grade and molecular subtype. Color scale denotes Tukey’s HSD post-hoc score values Colors represent Tukey’s HSD post-hoc scores. Visualization was generated using the “ComplexHeatmap” (v2.22.0) R package. E Heatmap of 658 significantly changing proteins between immune phenotypes in TNBC P-ROIs. Color scale denotes Tukey’s HSD post-hoc score. F Heatmap of 2822 significantly changing proteins between immune phenotypes in HR+ P-ROIs. Color scale denotes Tukey’s HSD post-hoc score values. Protein network representation of the Kynurenine pathway, Aryl Hydrocarbon Receptor (AHR), and Prostaglandins in TNBC (G) and HR+ (H) subtypes. Each node is divided into four immune-phenotype quadrants, color-coded by Tukey’s HSD post-hoc scores. Protein-protein interactions were obtained from the STRING database, and visualized in Cytoscape (v3.10.0) using the Omics Visualizer app63.
