BACKGROUND: Glioma exhibits significant molecular heterogeneity, necessitating improved understanding of molecular subtypes for personalized treatment strategies. METHODS: This study integrated bulk RNA sequencing (GSE35169) to derive transcriptome-defined groups and single-cell RNA sequencing (GSE131928) to characterize cellular heterogeneity and EGFR-associated malignant cell programs. Differential gene expression analysis identified subtype-specific biomarkers. Single-cell analysis included cell-type annotation, pseudotime trajectory, and cell-cell communication inference, with malignant cells stratified by EGFR expression to interrogate EGFR-associated programs. Key findings were validated by qRT-PCR in glioma cell lines (LN229, U251) and normal astrocytes (NHA). RESULTS: Two transcriptome-defined groups were identified from bulk RNA-seq. In scRNA-seq, the EGFR-high malignant cell state showed enrichment of ECM-related genes (IGFBP2, COL1A1) and enhanced PI3K-AKT, focal adhesion, and angiogenic signaling, with predominant stromal-to-tumor communication (EGF/AREG→EGFR, TGFB1→TGFBR2). The EGFR-low malignant cell state exhibited higher expression of immune-related genes (CXCL10, IL6, STAT1). Functional enrichment analysis based on differentially expressed genes (FDR < 0.05) indicated significant enrichment of interferon signaling and JAK-STAT pathway-related gene sets in the EGFR-low group. Pseudotime analysis revealed gradual transcriptional transitions between proliferation, hypoxia, and immune programs. qRT-PCR validation confirmed elevated EGFR, IGFBP2, and COL1A1 in U251 cells, and higher CXCL10, IL6, and STAT1 in LN229 cells. CONCLUSION: This study identifies bulk transcriptome-defined groups and reveals EGFR-associated malignant cell programs linked to distinct microenvironmental interaction patterns. The EGFR-high malignant cell state features proliferative and angiogenic programs, while the EGFR-low malignant cell state shows enhanced immune infiltration. These findings provide molecular evidence for precision classification and subtype-specific therapeutic strategies.
Wang et al. (Fri,) studied this question.