An innovative prognostic model based on autophagy-related long noncoding RNA signature for low-grade glioma

Autophagy is a highly conserved lysosomal degradation process essential in tumorigenesis. However, the involvement of autophagy-related long noncoding RNAs (lncRNAs) in low-grade glioma (LGG) remains unclear. In this study, we established an autophagy-related lncRNA prognostic signature for patients with LGG and assess its underlying functions. We used univariate Cox, least absolute shrinkage and selection operator and multivariate Cox regression models to establish an autophagy-related lncRNA prognostic signature. Kaplan-Meier survival analysis, receiver operating characteristic curve, nomogram, C-index, calibration curve and clinical decision-making curve were used to assess the predictive capability of the identified signature. A signature comprising nine autophagy-related lncRNAs (AL136964.1, ARHGEF26-AS1, PCED1B-AS1, AS104072.1, PRKCQ-AS1, LINC00957, AS125616.1, PSMB8-AS1 and AC087741.1) was identified as a prognostic model. Patients with LGG were divided into the high- and low-risk cohorts based on the median model-based risk score. The survival analysis revealed a 10-year survival rate of 9.3% (95% CI 1.91-45.3%) and 13.48% (95% CI 4.52-40.2%) in high-risk patients in the training and validation sets, respectively, and 48.4% (95% CI 24.7-95.0%) and 48.4% (95% CI 28.04-83.4%) in low-risk patients in the training and validation sets, respectively. This finding suggested a relatively low survival in high-risk patients. In addition, the lncRNA signature was independently prognostic and potentially associated with the progression of LGG. Therefore, the 9-autophagy-related-lncRNA signature may play a crucial role in the diagnosis and treatment of LGG, which may offer new avenues for tumour-targeted therapy.