Integrated Analysis Construct a Tumor-Associated Macrophage Novel Signature with Promising Implications in Predicting the Prognosis and Immunotherapeutic Response of Gastric Cancer Patients.

ABSTRACT

BACKGROUND: Gastric cancer (GC) remains one of the most prevalent malignant tumors worldwide. At present, tumor-associated macrophages (TAMs) are essential in the progression, metastasis, and drug resistance of tumors. Therefore, TAMs can be a crucial target for tumor treatment. AIMS: We intended to investigate the TAM characteristics in GC and develop a risk signature based on TAM to predict the prognosis of GC patients. METHODS: The single-cell RNA sequencing (scRNA-seq) and bulk RNA-seq data were acquired from a publicly available database. We utilized the Seurat pipeline to process the scRNA-seq data and determine TAM cell types using marker genes. Univariate Cox regression analysis was utilized to examine TAM-related prognostic genes, and then we employed Lasso-Cox regression analysis, and Multivariate Cox regression analysis established a novel risk profile to forecast the clinical value of the model with a new nomogram combining risk profiles and clinicopathological characteristics. RESULTS: The current study employed scRNA-seq data to identify five TAM clusters in GC, among which four were significantly associated with GC prognosis. Accordingly, we further developed a TAM-related risk signature utilizing nine genes. After evaluation, our model accurately predicted the prognosis of gastric cancer. Generally, GC patients with low TAMS scores exhibited a more favorable prognosis, greater benefits from immunotherapy, and higher levels of immune cell infiltration. CONCLUSIONS: The prognosis of GC can be effectively predicted by TAM-based risk signatures, and the signature may provide a new perspective for comprehensively guiding clinical diagnosis, prediction, and immunotherapy for gastric cancer.