Molecular subtyping, prognostic evaluation, and tumor immune microenvironment heterogeneity in intrahepatic cholangiocarcinoma based on epithelial-mesenchymal transition and mitophagy-related genes
Year: 2026
Type: article
Abstract: Background: Intrahepatic cholangiocarcinoma (ICC) is a highly aggressive malignancy with a dismal prognosis, yet effective prognostic biomarkers and therapeutic targets remain scarce. Both epithelial-mesenchymal transition (EMT) and mitophagy are known to drive tumor progression and therapy resistance, but their synergistic regulatory network in ICC has not been systematically characterized. This study aims to integrate EMT- and mitophagy-related genes to establish a molecular subtyping and prognostic evaluation system for ICC. Methods: Transcriptomic data were obtained from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) repositories. Differentially expressed genes (DEGs) associated with prognosis, EMT, and mitophagy were screened using bioinformatics approaches. Consensus clustering was applied to stratify ICC patients into molecular subtypes. Protein-protein interaction (PPI) network analysis was performed to identify hub genes, followed by Cox regression validation for independent prognostic predictors. Functional enrichment, immune infiltration, and drug-sensitivity analyses were conducted to explore underlying biological mechanisms and subtype-specific therapeutic vulnerabilities. Results: A total of 18 prognosis-related DEGs linked to EMT and mitophagy were identified. Consensus clustering divided ICC patients into two distinct subtypes (Cluster 1 and Cluster 2), with Cluster 1 showing significantly poorer overall survival (OS) (P=0.009). PPI network analysis revealed nine core hub genes, among which PYGB , IGF1 , NRG1 , and PLOD2 were validated as independent prognostic predictors. Functional enrichment highlighted microRNA-mediated regulation, ErbB signaling, and endocrine resistance. Immune infiltration uncovered marked heterogeneity between subtypes: in Cluster 1, activated mast cells correlated most strongly with resting natural killer cells (r=0.78), whereas in Cluster 2, activated CD4 + memory T cells showed the strongest correlation with activated dendritic cells (r=0.46). Drug-sensitivity predictions indicated that Cluster 1 was more sensitive to UMI-77 and staurosporine. Conclusions: This integrative analysis establishes a novel molecular subtyping and prognostic framework for ICC based on EMT- and mitophagy-related genes. The identified hub genes and subtype-specific immune/drug profiles provide new theoretical insights and candidate targets for understanding ICC pathogenesis and developing personalized therapeutic strategies.
Source: Translational Cancer Research
Authors: Yecheng Wang, Dongbin Liu, Kuo Liang, Zhenshun Wang
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7224089547