EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Omics Identification of Vasculogenic Mimicry-Associated Molecular Subtypes in Hepatocellular Carcinoma for Prognostic Stratification and Therapeutic Response Prediction

Yuting Tao, Shuzhen Liao, Tao Liu, Ruyi Lai, Chao Feng, Qiuyan Wang

Cancers · 02.08.2026 · 18 (15) · 2482

DOI: 10.3390/cancers18152482 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cancers, online erschienen am 02.08.2026. Verfasst von Yuting Tao und 5 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Life Sciences, Guangxi Medical University, Nanning 530021, China. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Guangxi Natural Science Foundation Program, Guangxi Science and Technology Department.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Vasculogenic mimicry (VM), characterized by the de novo formation of microvascular-like channels derived from aggressive tumor cells without involving traditional endothelial cells, is a pivotal pathological hallmark driving extreme invasiveness and dismal prognosis in hepatocellular carcinoma (HCC). This study aimed to establish a VM-based molecular subtyping system and systematically characterize its associated biological features, thereby providing a potential framework for individualized prognostic assessment and treatment decision-making in HCC.

Methods

We integrated curated VM-associated gene sets with single-cell RNA sequencing data to identify malignant epithelial cell-enriched VM-associated candidate genes. Subsequently, univariate Cox regression, LASSO-Cox regression, and multivariate Cox regression were sequentially performed to identify six prognostic VM-related genes: HSPA9, TGFA, MAD2L1, PROM1, AGXT, and GCGR. HCC patients were stratified into VM, Mixed-VM, and Non-VM subtypes according to VM scores. Kaplan–Meier survival analysis, time-dependent ROC analysis, and Cox regression were used to assess prognostic performance. The biological features of the classification system were evaluated using bulk transcriptomic cohorts, spatial transcriptomics, Cytometry by Time-of-Flight (CyTOF), metabolomics, lipidomics, somatic mutation and copy number alteration analyses, and treatment-related HCC cohorts.

Results

The VM score-based classification stratified HCC patients into three molecular subtypes with distinct prognostic and biological characteristics. Patients classified as the VM subtype had significantly poorer overall survival than those classified as Mixed-VM or Non-VM subtypes, and this prognostic pattern was validated across independent cohorts. Multi-omics analyses showed that the VM subtype was associated with YAP-TAZ-TEAD-related transcriptional programs, stemness/proliferation-related features, immunoregulatory and exhaustion-like tumor microenvironmental characteristics, and distinct metabolic and lipidomic alterations involving modified nucleosides, keto acid-related metabolites, cholesteryl esters, and sphingolipid-related species. Spatial transcriptomics revealed focal enrichment of VM-score-high regions and their association with YAP-TAZ-TEAD and immune checkpoint-related signatures. In orthotopic HCC mouse models, YAP1 overexpression increased PAS+/CD34− VM-like structures, whereas verteporfin treatment reduced these structures. In two treatment-related cohorts, the Non-VM subtype showed higher response rates to sorafenib and transarterial chemoembolization (TACE) than the VM subtype. Connectivity Map (CMap)-based computational drug prioritization and molecular docking analysis prioritized ivermectin as a candidate compound; however, its antitumor activity requires further experimental validation.

Conclusions

This study establishes a VM score-based molecular classification framework for HCC and identifies VM-subtype-associated prognostic, spatial, immune, metabolic, genomic, and therapeutic features. These findings provide a candidate framework for molecular risk stratification and subtype-guided therapeutic exploration in HCC.

Schlagwörter

Stichwörter der Autor:innen

hepatocellular carcinomamolecular subtypingmulti-omicsvasculogenic mimicry

Autor:innen und Institutionen

  • Institute of Life Sciences, Guangxi Medical University, Nanning 530021, China.
  • Guangxi Key Laboratory for Genomic and Personalized Medicine, Guangxi Medical University, Nanning 530021, China.
  • Center for Genomic and Personalized Medicine, Guangxi Medical University, Nanning 530021, China.
  • The Academician Workstation on Metabolism and Health, The First Affiliated Hospital of Guangxi Medical University, Nanning 530021, China.
  • School of Information and Management, Guangxi Medical University, Nanning 530021, China.

Förderung

  • National Natural Science Foundation of China
  • Guangxi Natural Science Foundation Program
  • Guangxi Science and Technology Department
  • First-class discipline innovation-driven talent program of Guangxi Medical University
  • Guangxi science and technology development project

Zitieren

APA

Tao, Y., Liao, S., Liu, T., Lai, R., Feng, C., & Wang, Q. (2026). Multi-Omics Identification of Vasculogenic Mimicry-Associated Molecular Subtypes in Hepatocellular Carcinoma for Prognostic Stratification and Therapeutic Response Prediction. Cancers, 18(15), 2482. https://doi.org/10.3390/cancers18152482

Vancouver

Tao Y, Liao S, Liu T, Lai R, Feng C, Wang Q. Multi-Omics Identification of Vasculogenic Mimicry-Associated Molecular Subtypes in Hepatocellular Carcinoma for Prognostic Stratification and Therapeutic Response Prediction. Cancers. 2026;18(15):2482. doi:10.3390/cancers18152482

BibTeX
@article{Tao2026,
  author = {Tao, Yuting and Liao, Shuzhen and Liu, Tao and Lai, Ruyi and Feng, Chao and Wang, Qiuyan},
  title = {Multi-Omics Identification of Vasculogenic Mimicry-Associated Molecular Subtypes in Hepatocellular Carcinoma for Prognostic Stratification and Therapeutic Response Prediction},
  journal = {Cancers},
  year = {2026},
  volume = {18},
  number = {15},
  pages = {2482},
  doi = {10.3390/cancers18152482}
}
Kostenlos starten →