Neuroblastoma (NB), the most common extracranial solid tumor in children, is characterized by marked clinical heterogeneity and extremely poor outcomes in high-risk patients, and remains a leading cause of pediatric cancer-related mortality. Although lactylation has recently emerged as a metabolic–epigenetic modification implicated in tumor biology, its relevance to NB remains largely unexplored.

This study aimed to explore the prognostic value of lactylation-associated transcriptional patterns in NB, their relationship with the tumor immune microenvironment, and potential therapeutic implications. Transcriptomic and clinical data were obtained from the Gene Expression Omnibus (GEO) and ArrayExpress databases. Lactylation-related genes(LRGs) were curated from published literature. Differential expression analysis and survival modeling were used to construct a prognostic gene signature. Model performance was evaluated using Kaplan-Meier (K-M) analysis, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA). Immune infiltration and predicted drug sensitivity were assessed using established computational algorithms. In vitro functional assays and immunohistochemical (IHC) analysis of clinical NB specimens were performed to examine the relevance of the core gene RPS6KA6.

A five-gene prognostic signature derived from LRG-defined clusters (GRIN3A, TERT, E2F3, RPS6KA6, and CAMTA1) was established and validated. Patients in the high-risk group exhibited significantly poorer overall survival compared with those in the low-risk group. The low-risk group showed higher immune cell infiltration and was predicted to be more responsive to immunotherapy. Through Connectivity Map analysis, Entinostat (MS-275), a class I histone deacetylase inhibitor, was identified as a candidate compound associated with the high-risk transcriptional state. In vitro assays indicated that RPS6KA6 contributes to NB cell proliferation, migration, and invasion, and IHC analysis demonstrated that elevated RPS6KA6 expression in tumor tissues is associated with unfavorable clinical outcomes.

We developed a five-gene lactylation-associated, pattern-derived prognostic signature derived from LRG-defined subtypes that effectively stratified neuroblastoma patients by risk and was associated with distinct immune features and predicted drug sensitivity. RPS6KA6 was further characterized by in vitro functional assays and immunohistochemical evaluation in clinical specimens, supporting its candidacy as a prognostic biomarker. These findings provide preliminary insights into metabolism-associated epigenetic programs in NB and their relevance to personalized treatment strategies.