Prediction of neuroblastoma prognosis with a novel T-cell exhaustion-related gene signature

Neuroblastoma (NB) is the most common type of pediatric extra-cranial tumor that arises in the sympathetic nervous system. The heterogeneity of T-cell exhaustion (TEX) has been linked to the determination of distinct clinical outcomes and the effectiveness of immunotherapy in numerous adult malignancies. Therefore, studying the heterogeneous TEX landscape in NB as well as its impact on clinical outcomes is meaningful. The gene expression and clinical datasets of the Sequencing Quality Control (SEQC), E-MTAB-8248, and Therapeutically Applicable Research to Generate Effective Treatments (TARGET) cohorts were downloaded from publicly available databases. Two TEX-related clusters for neuroblastoma were identified in the SEQC cohort. Patients in TEX-C1 exhibited superior overall survival (OS) and event-free survival (EFS) rates compared with those in TEX-C2. And TEX-C1 had more immune cells infiltrating, as well as higher expression of immune checkpoint genes. A total of 1984 genes were differentially expressed between these two clusters, of which 1712 were associated with OS. A gene signature consisting of ten TEX-related genes was developed, and a risk score was computed for each patient. Based on the risk score, SEQC patients were split into high- and low-risk groups with significantly different survival rates. The risk score was an independent risk factor predicting survival and showed superior prediction power for 3, 5, and 10-year survival compared to individual clinical parameters. The signature was further confirmed in the TARGET and E-MTAB-8248 cohorts. This study has successfully constructed a risk score model for NB prognosis, utilizing TEX as its foundation. The model provides risk classification and survival evaluation, which can further guide treatment.