Unveiling a NovelGlioblastoma Deep Molecular Profiling:Insight into the Cancer Cell Differentiation-Related Mechanisms

Background: the sophisticated cellular heterogeneity of cell populations in glioblastoma (GBM) has been a key factor influencing tumor progression and response to therapy. The lack of more precise stratification based on cellular differentiation status poses a great challenge to therapeutic strategies. Materials and Methods: harnessing the bulk multiomics and single-nucleus RNA sequencing data available from the National Center for Biotechnology Information (NCBI) and The Cancer Genome Atlas (TCGA) Program repositories, we developed a novel and accurate GBM risk classification using an ensemble consensus clustering approach based on the junction of prognosis and trajectory analysis. Comprehensive cluster labeling and multiomics data characterization were also performed. Results: a novel GBM stratification model was constructed using 45 malignant cell fate genes: (a) energy metabolism-enhanced-type GBM; (b) invasion-enhanced-type GBM; (c) invasion-attenuated-type GBM; and (d) glycolysis-dominant energy metabolism-enhanced-type GBM. The biological plausibility of the model was verified through a range of comprehensive analyses of multiomics data, showing that cases with invasion-attenuated-type were the best prognosis and energy metabolism-enhanced-type the poorest. Conclusions: the study has uncovered GBM complex cellular heterogeneity and a differentiated hierarchy of cell populations underlying tumorigenesis. This precise stratification system provided implications for further studies of individual therapies.