Abstract
BACKGROUND: Neuroblastoma (NB) is a prevalent extracranial solid tumor in pediatric patients. Of these, the MYCN-amplified type has a poor treatment response and prognosis. To enhance therapeutic efficacy and prognostic outcomes, numerous research teams have undertaken extensive investigations through various pathways and directions. Among these, ferroptosis has recently emerged as a significant area of research focus.Ferroptosis, a type of iron-dependent cell death, is primarily caused by lipid peroxides. This study intends to develop a prognosis model based on MYCN-amplified NB and ferroptosis-related genes (FGs). METHODS: Data for this study were sourced from the TARGET and FerrDb databases. Lasso regression algorithms and univariate COX analysis were leveraged to determine feature genes; multivariate COX analysis was employed to develop a prediction model and risk scores; and receiver operating characteristic (ROC) curves and Kaplan-Meier analysis were utilized to assess the predictive ability of the model. Furthermore, discrepancies in immune cell infiltration (ICI) between the high-risk (HR) and low-risk (LR) populations were assessed via CIBERSORT analysis. Finally, experiments were conducted on MYCN-amplified and MYCN non-amplified cells so as to validate the differential expression of the gene. RESULTS: A prediction model was constructed and risk scores were calculated based on 4 genes (LIFR, TP53, NRAS, and OSBPL9). The HR group, which was stratified by the median score, had a lower overall survival rate than the LR group.The differences in expression of each gene between MYCN-amplified and MYCN non-amplified cells were further confirmed through cell experiments and qPCR. CONCLUSION: The prediction model in this study can be employed to forecast the prognosis of MYCN-amplified NB. These genes may represent promising new ferroptosis-related intervention targets (FITs) in treating MYCN-amplified NB, with the potential to improve patient outcomes.