Prognostic Analysis of a Hypoxia-Associated lncRNA Signature in Glioblastoma and its Pan-Cancer Landscape.

Abstract

Background:  Hypoxia is an important clinical feature of glioblastoma (GBM), which regulates a variety of tumor processes and is inseparable from radiotherapy. Accumulating evidence suggests that long noncoding RNAs (lncRNAs) are strongly associated with survival outcomes in GBM patients and modulate hypoxia-induced tumor processes. Therefore, the aim of this study was to establish a hypoxia-associated lncRNAs (HALs) prognostic model to predict survival outcomes in GBM patients.

Methods:  LncRNAs in GBM samples were extracted from The Cancer Genome Atlas database. Hypoxia-related genes were downloaded from the Molecular Signature Database. Co-expression analysis of differentially expressed lncRNAs and hypoxia-related genes in GBM samples was performed to determine HALs. Six optimal lncRNAs were selected for building HALs models by univariate Cox regression analysis.

Results:  The prediction model has a good predictive effect on the prognosis of GBM patients. Meanwhile, LINC00957 among the six lncRNAs was selected and subjected to pan-cancer landscape analysis.

Conclusion:  Taken together, our findings suggest that the HALs assessment model can be used to predict the prognosis of GBM patients. In addition, LINC00957 included in the model may be a useful target to study the mechanism of cancer development and design individualized treatment strategies.