A Multi-Gene Myofibroblast Framework Stratifies Prognostic and Immune Landscapes in HCC
Li Bai, Qingmiao Shi, Leiya FuIntroduction:
Myofibroblasts play a critical role in the progression of hepatocellular carcinoma (HCC), a major subtype of primary liver cancer.
Methods:
Bulk RNA-seq data were analyzed to identify core genes of relevant cell subpopulations. Next, cell clustering was performed based on public scRNA-seq data of HCC. Using the R package CellChat, receptor-ligand communication networks between myofibroblasts and other cell subtypes were characterized. Hub genes within HCC myofibroblast subpopulations were screened via hdWGCNA. Subsequently, differentially expressed genes (DEGs) from the bulk analysis were intersected with these hub genes. Machine learning algorithms were employed to select key genes to construct a nomogram. Finally, correlations in immune cell infiltration were analyzed.
Results:
Six major cell subpopulations were identified from the scRNA-seq data, with prominent crosstalk observed between myofibroblasts and hepatocytes. Using hdWGCNA, 10 myofibroblastassociated co-expression modules were obtained, five of which were identified as functionally key modules. By intersecting the module hub genes with HCC-related DEGs, 15 overlapping genes were obtained. From these, four key genes were ultimately selected by machine learning algorithms: plasmalemma vesicle-associated protein (PLVAP), retinol binding protein 7 (RBP7), NADH dehydrogenase (Ubiquinone) 1 Alpha subcomplex subunit 4-like 2 (NDUFA4L2), and tropomyosin 2 (Beta) (TPM2). A nomogram integrating 6 imaging-histological features and PLVAP expression exhibited robust prediction capacity. All gene expressions were positively correlated with the infiltration of regulatory T cells (Tregs) and macrophage M0 and negatively correlated with the infiltration of neutrophils and monocytes.
Discussion:
Myofibroblasts participate in extensive intercellular crosstalk in the HCC microenvironment, suggesting the potential value of myofibroblast-related biomarkers in HCC therapy.
Conclusion:
In this study, hub genes associated with myofibroblast programs in HCC were identified using scRNA-seq and hdWGCNA.