Integration of single-cell and bulk RNA sequencing to establish a prognostic signature based on tumor-associated macrophages in colorectal cancer.

Several studies have shown significant involvement of tumor-associated macrophages (TAMs) in the tumor microenvironment and cancer progression. However, no data on reliable TAM-related biomarkers are available for predicting the prognosis of patients with colorectal cancer (CRC). We analyzed the clinical data and gene expression profiles of patients with CRC from databases. The single-cell transcriptomic data was applied to identify M2-like TAM-related differentially expressed genes. Univariate Cox and least absolute shrinkage and selection operator regression analyses were used to determine the prognostic signature genes. Then, seven key genes were screened to develop the prognostic signature. In the training and external validation cohorts, the overall survival (OS) of patients in the high-risk group was significantly shorter compared to the low-risk group. Consequently, we created a nomogram that could accurately and reliably predict the prognosis of patient with CRC. A significant correlation was observed between the patient's prognosis, clinical features, sensitivity to anticancer drugs, TME, and risk scores. Moreover, risk score was strongly related to the response to immunotherapy in patients from GSE91061, GSE78220, and GSE60331 cohorts. Finally, high expression of HSPA1A, SERPINA1, CXCL1, and low expression of DNASE1L3 were found in human CRC tissue and normal tissue by using qRT-PCR. In conclusion, the M2-like TAM-related prognostic signature could predict the survival, prognosis, and response of patients with CRC to immunotherapy, which sheds light on the role of TAMs in CRCs and enhances our understanding of TAMs.