Novel endoplasmic reticulum stress-related gene signature unveils CDKN3 as a prognosticator in neuroblastoma.

Abstract

Background: Neuroblastoma (NB) is the most common extracranial solid tumor in children and a major cause of pediatric cancer mortality. This study aims to develop an endoplasmic reticulum (ER) stress-based risk model to evaluate patient prognosis, identify novel therapeutic targets, and predict immunotherapy responses.

Methods: NB cases with transcriptome and clinical data were obtained from Gene Expression Omnibus (GEO) and ArrayExpress databases. ER stress-related genes were extracted from GeneCards. Differentially expressed genes (DEGs) were identified to construct an ER stress-related gene signature for prognosis prediction. The predictive ability was assessed using survival analysis, receiver operating characteristic (ROC) curves, and statistical tools. The relationship between the ER stress risk score and clinicopathological features, immune infiltration, and drug sensitivity was evaluated. A predictive nomogram was developed for prognostic accuracy. Immunohistochemistry (IHC) validated the hub gene using NB clinical specimens.

Results: A five-gene signature (PINK1, IL7, CDKN3, C1S, MMP9) was established, effectively stratifying patients into high- and low-risk groups with significant differences in overall survival. The model demonstrated robust predictive performance in training and testing datasets. High-risk NB patients exhibited poorer clinicopathological characteristics and a higher likelihood of being unresponsive to immunotherapy. Specific targeted agents were identified for high-risk patients. A nomogram integrating the gene signature and clinical variables enhanced prognostic accuracy. IHC analysis of CDKN3 supported its role as a biomarker for poor prognosis in NB.

Conclusions: This five-gene model linked to ER stress can independently forecast NB prognosis and correlates with immune and antitumor agent susceptibility, providing a basis for personalized treatment strategies.