Development and External Validation of an M2 Macrophage-Related Prognostic Signature in Breast Cancer Using Multi-Omics Data and Machine Learning

Purpose

Tumor-associated macrophages (TAMs), particularly the M2-like phenotype, are central regulators of breast cancer progression; however, clinically applicable models capturing their functional heterogeneity and biological mechanisms remain limited. Patients and methods: We integrated single-cell RNA sequencing, spatial transcriptomics, and bulk RNA-seq data to construct an M2 macrophage-centered gene landscape. A machine learning framework based on 10-fold cross-validation was used to develop a prognostic signature in the TCGA-BRCA cohort (training set), with external validation in the independent GSE96058 cohort. Immune infiltration, genomic instability, pathway activity, and cell-cell communication were analyzed. KLF4 was examined in breast cancer cells and macrophages using qRT-PCR, Western blotting, siRNA-mediated knockdown, and a Transwell co-culture system.