Assessing the role of tumour-associated macrophage subsets in breast cancer subtypes using digital image analysis

Purpose The number of M1-like and M2-like tumour-associated macrophages (TAMs) and their ratio can play a role in breast cancer development and progression. Early clinical trials using macrophage targeting compounds are currently ongoing. However, the most optimal detection method of M1-like and M2-like macrophage subsets and their clinical relevance in breast cancer is still unclear. We aimed to optimize the assessment of TAM subsets in different breast cancer subtypes, and therefore related TAM subset numbers and ratio to clinicopathological characteristics and clinical outcome. Methods Tissue microarrays of 347 consecutive primary Luminal-A, Luminal-B, HER2-positive and triple-negative tumours of patients with early-stage breast cancer were serially sectioned and immunohistochemically stained for the pan-macrophage marker CD68 and the M2-like macrophage markers CD163, CSF-1R and CD206. TAM numbers were quantified using a digital image analysis algorithm. M1-like macrophage numbers were calculated by subtracting M2-like TAM numbers from the total TAM number. Results M2-like markers CD163 and CSF-1R showed a moderate positive association with each other and with CD68 ( r ≥ 0.47), but only weakly with CD206 ( r ≤ 0.06). CD68 + , CD163 + and CSF-1R + macrophages correlated with tumour grade in Luminal-B tumours ( P