Spinal Ependymoma (SP-EP) is the most commonly occurring tumor affecting the spinal cord. Prompt diagnosis and treatment can significantly enhance prognostic outcomes for patients. In this study, we conducted a comprehensive analysis of RNA sequencing data, along with associated clinical information, from patients diagnosed with SP-EP. The aim was to identify key genes that are characteristic of the disease and develop a survival-related nomogram.
We first accessed the Gene Expression Integrated Database (GEO) to acquire the microarray dataset pertaining to SP-EP. This dataset was then processed to identify differentially expressed genes (DEGs) between SP-EP samples and normal controls. Furthermore, machine learning techniques and the CIBERSORT algorithm were employed to extract immune characteristic genes specific to SP-EP patients, thereby enhancing the characterization of target genes. Next, we retrieved comprehensive information on patients diagnosed with SP-EP between 2000 and 2020 from the Surveillance, Epidemiology, and End Results Database (SEER). Using this data, we screened for predictive factors that have a significant impact on patient outcomes. A nomogram was constructed to visualize the predicted overall survival (OS) rates of these patients at 3, 5, and 8 years post-diagnosis. Finally, to assess the reliability and clinical utility of our predictive model, we evaluated it using various metrics including the consistency index (C-index), time-dependent receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration curves, and decision curve analysis (DCA).
A total of 5,151 DEGs were identified between the SP-EP sample and the normal sample. Analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways revealed that these DEGs were primarily involved in cellular processes, including cell cycle regulation and cell sensitivity mechanisms. Furthermore, immune infiltration analysis was utilized to identify the core gene
Through bioinformatics analysis of microarray datasets, we identified