Transcriptional regulons of tumor-associated macrophages reveal divergent polarization states across cancer types

Abstract

Aim: To investigate transcription factor regulons that drive macrophage polarization across multiple cancer types using single-cell RNA sequencing data.

Method: Publicly available datasets from lung, ovarian, pancreatic, and head and neck cancers were retrieved. After standard quality control and normalization, gene regulatory networks were reconstructed using our bioinformatic workflow, which integrates co-expression analysis, motif enrichment, and regulon activity scoring. Transcription factors associated with M1 and M2 macrophage polarization were systematically examined.

Results: Analysis of regulon activity revealed that macrophages in lung, ovarian, and pancreatic cancers predominantly exhibited M1-associated transcriptional programs, suggesting anti-tumor properties. Key active transcription factors in these cancers included JUND, STAT1, NFκB, and IRF1. In contrast, head and neck cancers displayed a predominance of M2-associated transcriptional activity, with strong enrichment of SPI1, CEBPB, and IRF8, indicative of pro-tumor macrophage polarization. These findings highlight cancer type–specific heterogeneity in macrophage transcriptional regulation.

Conclusions: This study demonstrates that transcription factor regulon analysis can distinguish tumor-associated macrophage states across cancers and provides insights into their divergent roles in tumor progression. The identification of cancer type–specific transcriptional drivers of macrophage polarization may inform the development of macrophage-targeted immunotherapies and improve prognostic biomarker discovery.