Integrating Bulk and Single-cell RNA-seq to Construct a Macrophage-related Prognostic Model for Prognostic Stratification in Triple-negative Breast Cancer

Background: Triple-negative breast cancer (TNBC) is a poor prognostic subtype of breast cancer due to limited treatment. Macrophage plays a critical role in tumor growth and survival. Our study intends to explore the heterogeneity of macrophage in TNBC and establish a macrophage-related prognostic model for TNBC prognostic stratification. Materials and Methods: Seurat package was conducted to analyze the single-cell RNA expression profilers. The cell types were identified by the markers derived from public research and online database. The cell-cell interactions were calculated by the CellChat package. Monocle package was used to visualize the cell trajectory of macrophages. The prognostic model was constructed by six macrophage-related genes after a series of selections. The expression of six genes were validated in normal and TNBC tissues. And several potential agents for high-risk TNBC patients were analyzed by Connectivity Map analysis. Results: Nine cell types were identified, and the macrophages were highly enriched in TNBC samples. five distinct subgroups of macrophage were identified. Notably, SPP1+ tumor-associated macrophages exhibited a poor prognosis. The prognostic model was constructed by HSPA6, LPL, IDO1, ALDH2, TK1, and QPCT with good predictive accuracy at 3-, 5- years overall survival for TNBC patients in both training and external test cohorts. Finally, several drugs were identified for the high-risk TNBC patients decided by model. Conclusion: Our study provides a valuable source for clarifying macrophage heterogeneity in TNBC, and a promising tool for prognostic risk stratification of TNBC.