Metabolic reprogramming plays an essential role in the initiation and aggressiveness of malignant tumors. This study aims to establish a prognostic signature for Wilms tumor (WT) based on metabolism-related genes (MRGs) and to explore potential molecular mechanisms. RNA sequencing data of WT samples and MRGs were sourced from GEO, TCGA and KEGG, respectively. Prognosis-associated differentially expressed MRGs (DE-MRGs) and Lasso regression were employed to create a nine-gene prognostic signature. Internal validation was conducted through bootstrap resampling. Prognostic performance was assessed through Kaplan-Meier curves, ROC curves, the C-index, and calibration curves. Additional analyses encompassed signaling pathways and chemotherapy response prediction. The expression levels and biological functions of NNMT were experimentally validated. A nine-gene signature comprising B3GAT2, COLGALT2, CYP27C1, GAD2, GSTM5, LPIN3, NNMT, ST6GALNAC1 and TCIRG1 was established. The risk score derived from this signature was shown to be an independent prognostic predictor and significantly associated with immune function and autophagy. NNMT expression levels were validated in both cells and tissues. Further experiments in vivo and in vitro indicated that NNMT might influence cholesterol efflux via PPARG-LXRα pathway, thereby enhancing cell proliferation and migration. This study established a metabolism-related gene signature to predict the prognosis of patients with WT. The findings may provide a promising tool for personalized diagnosis and treatment.
Keywords: NNMT; PPARG‐LXRα pathway; Wilms tumor; metabolism‐related gene; prognostic signature.
© 2025 John Wiley & Sons Ltd.