Multiplex Immunophenotyping of Tumor Samples

Tissue slides were imaged at 20× magnification and, at least, three regions of interest containing cancer cells were analyzed per sample (1.34 mm²/region of interest) with the Vectra 3.0 Automated Quantitative Pathology Imaging System (Perkin Elmer). An analysis algorithm was trained for tissue and cell segmentation as well as immunophenotyping of cells. DAPI and keratin were employed for segmenting images into tumour, stroma, and ‘no tissue’ areas. Next, cellular segmentation was performed using a counterstain-based approach with DAPI to segment nuclei and membrane markers (CD8, CD3, CD163) to detect cell contours. The following phenotypes were identified: CD3⁺CD8⁻FOXP3⁻ T cells (which were mainly comprised of CD4⁺ T helper cells), CD3⁺CD8⁻FOXP3⁺ T cells (corresponding to Tregs) CD3⁺CD8⁺ T cells (corresponding to CTLs), and CD163⁺ myeloid cells to pinpoint tumor-associated macrophages. All images were visually inspected to confirm the correct attribution and quantification of phenotypes. For each case, cell counts were normalized by tissue area (number of cells/mm²) [37 (link)].