Fig. 4: Spatial characterization of the tumor-immune cell interactions (using ML-based metrics) in the triple-negative subtype in patients that achieved pCR versus non-pCR. | npj Breast Cancer

Fig. 4: Spatial characterization of the tumor-immune cell interactions (using ML-based metrics) in the triple-negative subtype in patients that achieved pCR versus non-pCR.

From: Machine learning-based spatial characterization of tumor-immune microenvironment in the EORTC 10994/BIG 1-00 early breast cancer trial

Fig. 4

A Representative images depicting the spatial phenotyping and interactions among different immune cell subpopulations with tumor cells in a pCR (left) and a non-pCR (right) patient; Radar plots depicting normalized mixing score (NMS) values for different increasing radii for the interaction of overall immune-tumor cells (B) and the specific immune cell subpopulations in patients achieving pCR (green) versus non-pCR (red) TN (C). D Entropy gradient slope depicting attraction-like (red lines) and repulsion-like (blue lines) patterns for the interaction between immune and tumor cells in non-pCR (left panel) versus pCR (right panel) patients. Figure 3B was created with BioRender.com.

Back to article page