Pan-cancer gene signature analysis of macrophage polarization and compound prediction for reprogramming tumor-associated macrophages toward M1-like macrophages.

Background: Resting tumor-associated macrophages (TAMs) are stimulated by the tumor microenvironment and can be primarily polarized into two subtypes: M1-and M2-like. M1-like TAMs promote inflammation and eradicate tumor cells, whereas M2-like TAMs suppress inflammation and facilitate tumor development. However, the mechanisms underlying phenotypic switching in these macrophages remain unclear. Therefore, we aimed to characterize the gene expression profiles of M1-like and M2-like TAMs in pan cancers. Methods: Three computational methods were used to estimate the infiltration score of TAMs in 9239 tumor samples across 31 solid cancer types, based on RNA sequencing databases. Tumor samples were divided into high- and low-score groups based on the median M1/M2 ratio. Furthermore, gene enrichment, protein interactions, and transcription factors were analyzed. Multiple pharmaco-omics profiles were used to identify potential drugs. Finally, binding between the compounds and drug targets was validated using molecular docking. Results: Of the top 100 dysregulated genes in each cancer type, 70 and 82 genes with upregulated and downregulated expression, respectively, were consistently differentially expressed. We identified candidate drugs targeting protein phosphatase 2A (PP2A), a core protein. These included efaproxiral, hesperidin, ezetimibe, calcitriol, and linopirdine. Conclusions: This study provides a pan-cancer characterization of the TAM polarization-related gene profile. Network pharmacology and molecular docking analyses revealed five promising therapeutic agents for TAM reprogramming. Thus, our findings provide valuable insights into the enhancement of immune responses to inhibit tumor immune escape and metastasis.