Single-cell multi-omics analysis reveals drivers of intrahepatic cholangiocarcinoma metastasis

Intrahepatic cholangiocarcinoma (ICC) is a highly heterogeneous malignancy with poor prognosis, in which tumor metastasis represents a critical event driving disease progression and treatment failure. Understanding the cellular drivers of ICC metastasis and establishing effective risk assessment models are essential for improving patient outcomes. We analyzed the GSE201425 single-cell RNA sequencing dataset to explore the cellular origins and mechanisms of ICC metastasis. Copy number variation (CNV) profiling and clonal evolution analysis were performed to identify malignant epithelial cells. Pseudotime trajectory analysis revealed metastasis-associated epithelial cells (MAECs). Key biomarkers within MAECs were screened, and a Metastasis Index (Met-Index) based on one-class logistic regression was constructed to assess metastasis risk in bulk RNA-seq data from the TCGA-CHOL cohort. Multiplex immunofluorescence staining was performed in 34 clinical ICC specimens. Functional validation was conducted in HuCCT1 cells using siRNA-mediated gene silencing, followed by CCK-8 proliferation assays and wound healing assays. MAECs were identified as a distinct epithelial subpopulation enriched in metastatic ICC samples. Biomarkers including MMP7, FXYD2, and PTHLH were upregulated in MAECs and strongly correlated with metastatic status. The Met-Index effectively stratified patients into high- and low-risk groups, with the high Met-Index group exhibiting significantly worse overall survival (p = 0.005) and progression-free survival (p = 0.015). Immunofluorescence validation confirmed increased expression of MMP7, FXYD2, and PTHLH in metastatic tumors, which correlated with advanced clinicopathological features. Functional assays demonstrated that knockdown of these biomarkers inhibited HuCCT1 cell proliferation and suppressed cell migration capacity. This study delineates the cellular origin and molecular drivers of ICC metastasis at the single-cell level, identifies MAECs as a metastasis-related subpopulation, and establishes a clinically relevant Met-Index with prognostic value. These findings provide potential biomarkers and therapeutic targets for the early identification and management of ICC patients at high risk of metastasis.