Anoikis associated long noncoding RNA signature predicts prognosis and small molecule drug response in glioma across multiple databases
TL;DR
The established eight-ARlncRNA signature serves as a robust prognostic tool, effectively predicting patient survival and mirroring the heterogeneity of the tumor immune landscape, and offers potential guidance for personalizing therapeutic strategies in glioma.
Abstract
Anoikis resistance is a critical mechanism driving glioma invasion, recurrence, and treatment failure. While long non-coding RNAs (lncRNAs) play vital regulatory roles in cancer biology, the functions of anoikis-related lncRNAs (ARlncRNAs) in glioma remain largely uncharacterized. A total of 1719 glioma samples from The Cancer Genome Atlas (TCGA, n = 701) and Chinese Glioma Genome Atlas (CGGA, n = 1018) were analyzed to identify prognostic anoikis-related lncRNAs via co-expression and differential screening. A risk signature was constructed using LASSO-Cox regression and validated by survival and ROC analyses. Additionally, immune infiltration, molecular subtypes, and drug sensitivity were evaluated to explore therapeutic potential. The prognostic model incorporated eight ARlncRNAs (LINC00519, AC140481.1, LINC00928, HOXA-AS2, CRNDE, ACAP2-IT1, USP30-AS1, and TMPO-AS1). The risk score was an independent predictor of overall survival and showed high predictive accuracy in both TCGA and CGGA cohorts (AUC > 0.78). High-risk patients displayed activation of focal adhesion, ECM–receptor interaction, and immune-related pathways, with elevated immune/stromal scores and increased immune checkpoint expression. Consensus clustering identified two molecular subtypes with distinct survival outcomes, immune landscapes, and drug sensitivities. The established eight-ARlncRNA signature serves as a robust prognostic tool, effectively predicting patient survival and mirroring the heterogeneity of the tumor immune landscape. Furthermore, it offers potential guidance for personalizing therapeutic strategies in glioma.