Figure 3.

Comparison of T cells infiltrates between CRC groups. (A) IMC analysis workflow using SIMPLI.⁸ For each region, images of the markers used (Supplementary Table 5) were preprocessed to extract pixel intensities. Masks for tumor and stroma were derived and used for the pixel analysis. Each region was segmented into single cells that were assigned to tumor or stroma, phenotypically identified through expression of representative markers, and used for single cell clustering. (B) Comparison of normalized CD3⁺ areas and (C) CD3⁺ cells between DB and nDB-CRCs in the discovery cohort. Benjamini-Hochberg false discovery rate (FDR) correction was applied for testing over 5 immune populations. (D) Comparison of normalized CD3⁺ areas and (E) CD3⁺ cells between DB- and nDB-CRCs in the validation cohort. (F) Uniform Manifold Approximation and Projection (UMAP) map of 20,890 T cells in 38 regions from 16 CRCs of the discovery cohort. Cells were grouped in 13 clusters based on the expression of 12 phenotypic markers using Seurat⁹ (Supplementary Table 8) and colored according to the mean intensities of representative markers. The circles indicate the 2 clusters enriched in hypermutated CRCs. (G) Proportions of cluster 1 (CD8⁺GzB⁺ cells) and cluster 2 (CD8⁺Ki67⁺ cells) over the total T cells in hypermutated and non-hypermutated CRCs. Distributions were compared using the 2-sided Wilcoxon’s rank sum test. Benjamini-Hochberg FDR correction was applied for testing over 13 clusters. (H) IMC-derived images of tumor-associated markers (E-cadherin and pan-keratin) and CD8 and GzB or CD8 and Ki67 in 2 representative samples. Scale bar = 100 μm. (I) Comparisons of normalized CD8⁺GzB⁺ and CD8⁺Ki67⁺ areas between hypermutated and non-hypermutated CRCs. Distributions were compared using the 2-sided Wilcoxon’s rank sum test. Benjamini-Hochberg FDR correction was applied for testing over 25 combinations of T-cell markers (Supplementary Table 6). The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the vertical lines mark minimum and maximum of all the data.