Single-cell analysis reveals heterogeneity in the molecular profile of the tumor microenvironment of biliary tract cancers

Background: Biliary tract cancers (BTCs) are heterogeneous malignancies with diverse tumor microenvironment (TME) cells that contribute to tumor progression. However, the characteristics of the TME among different subtypes of BTCs remain unclear. Methods We integrated single-cell RNA sequencing data from BTC patients (iCCA, GBC, and eCCA) and RNA sequencing data from cholangiocarcinoma samples to construct a cellular molecular atlas of BTCs. Through enrichment analysis, we explored signaling pathways in tumor and TME cells, identified developmental trajectories and characterized transcription factors regulating different cell subpopulations using pseudotime analysis and gene regulatory network analysis. Moreover, we analyzed the intercellular communication between tumor cells and TME cells in different BTCs and developed a prognostic model. Results Malignant cells in different BTCs displayed significant heterogeneity, with variations in immune cell composition, biological functions, and metabolic pathways. Notably, the iCCA subpopulation demonstrated enhanced invasive potential, while the eCCA and GBC subpopulations showed significant activation of energy metabolism-related pathways. The carcinoma-associated fibroblast subpopulations in eCCA and GBC may regulate tumor cell metabolism through metabolic reprogramming, thereby facilitating tumor proliferation. GBC specifically exhibited enrichment of CXCL13 + tumor-reactive CD8 + T cells, indicating a potentially favorable response to immune checkpoint blockade therapy. Additionally, as tumor progression occurred, macrophage subpopulations in all three BTCs displayed M1/M2 dichotomous polarization, and the antigen-presenting ability of monocyte-derived dendritic cells gradually decreased. Conclusion This study analyzed single-cell landscape heterogeneity across BTC subtypes, revealing complex intercellular communication within the TME. A prognostic model was developed, providing insights for personalized precision medicine.