Transcriptomic profiling identifies a nucleotide metabolism-related signature with prognostic power in gliomas

Objective: Nucleotide metabolic reprogramming as a hallmark of cancer is closely related to the occurrence and progression of cancer. We aimed to comprehensively analyze the nucleotide metabolism-related gene set and clinical significance in gliomas.

Methods: The RNA sequencing data of 702 gliomas from the Cancer Genome Atlas (TCGA) dataset were included as the training set, and the RNA sequencing data from the other three datasets (CGGA, GSE16011, and Rembrandt) were used as independent validation sets. Survival curve, Cox regression analysis, time-dependent ROC curve and nomogram model were performed to evaluate prognostic power of signature. R language was the main tool for bioinformatic analysis and graphical work.

Results: Based on the expression profiles of nucleotide metabolism-related genes, consensus clustering identified two robust clusters with different prognosis. We then developed a nucleotide metabolism-related signature that was closely related to clinical, pathological, and genomic characteristics of gliomas. And ROC curve showed that our signature was a potential biomarker for mesenchymal subtype. Survival curve and Cox regression analysis revealed signature as an independent prognostic factor for gliomas. In addition, we constructed a nomogram model to predict individual survival. Finally, functional analysis showed that nucleotide metabolism not only affected cell division and cell cycle, but also was associated with immune response in gliomas.

Conclusion: We developed a nucleotide metabolism-related signature to predict prognosis and provided new insights into the role of nucleotide metabolism in gliomas.