TFPI and ROCK1 serve as the key genes in the vasculogenic mimicry-related prognostic nomogram for glioblastoma

Glioblastoma (GBM) represents the predominant malignant brain tumor, characterized by unfavorable prognoses. The identification of novel molecular markers plays a pivotal role in advancing clinical prognosis. Vasculogenic mimicry (VM) has been reported to serve a crucial role in angiogenesis within glioblastoma (GBM). Thus, we extracted data from The Cancer Genome Atlas Program (TCGA) and the Chinese Glioma Genome Atlas (CGGA) databases. Subsequently, we used Kaplan-Meier (K-M) survival analysis combined with univariate and multivariate COX regression analysis to identify meaningful VM-related genes (VRGs). Based on these analyses, the VM index model and risk score model were built. The reliability of the model was then tested using CGGA data, western blotting, and tube formation assay experiments. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) demonstrated the accuracy of these models. The correlation of TFPI and ROCK1 with tumor characteristics and immune infiltration was explored with the help of TIMER. Vitro experiments revealed that reducing the expression of ROCK1 and TFPI tended to decrease the formation of vascular-like tubes on Matrix-Gel™ compared to the control siRNA groups. Our findings suggest that ROCK1 and TFPI contribute to GBM metastasis through vasculogenic mimicry.