A multi-type programmed cell death gene signature predicts survival and reveals therapeutic targets in osteosarcoma

• Published in: Discover. Oncology Vol. 16; no. 1; pp. 1118 - 16
• Main Author: Wu, Chunyang
• Format: Journal Article
• Language: English
• Published: New York Springer US 16.06.2025; Springer Nature B.V; Springer
• Subjects: Analysis; Apoptosis; Autophagy; Bone cancer; Cancer Research; Cancer therapies; Cell cycle; Cell death; COL13A1; Cytokines; Extracellular matrix; Gene expression; Genomics; HAVCR2; Internal Medicine; Medical prognosis; Medicine; Medicine & Public Health; Molecular Medicine; Oncology; Overall survival; Parameter estimation; Programmed cell death; Radiotherapy; Regression analysis; Regularization methods; Software; Surgical Oncology; Survival analysis; Tumorigenesis; Programmed cell death; OS; COL13A1; Overall survival; HAVCR2
• ISSN: 2730-6011; 2730-6011
• DOI: 10.1007/s12672-025-02944-y

Background Osteosarcoma (OS) is an aggressive bone tumor with poor outcomes in advanced stages. Programmed cell death (PCD) plays a crucial role in tumor biology, but the prognostic value of integrating multiple PCD-related genes in OS remains underexplored. Methods Transcriptomic data from GEO and clinical data from TARGET were analyzed. Functional enrichment, N6-methyladenosine (m6A)-related genes analysis and immune checkpoint analyses were performed. Genes related to 18 types of PCD were intersected with OS-specific differentially expressed genes. Prognostic genes were identified by Kaplan-Meier and Cox regression. A LASSO model was developed to construct a survival prediction signature. Single-cell RNA-seq data were used to explore gene expression across cell types. Results There are 781 differentially expressed genes totally. M6A-related genes including ALKBH3, CBLL1 and immune checkpoints including HAVCR2, PDCD1 showed differential expression in OS. Twelve PCD-related genes were significantly associated with OS survival. A four-gene (FUCA1, GM2A, MAN2B1, COL13A1) prognostic model was established, showing strong predictive performance (3-year AUC = 0.801; 5-year AUC = 0.842). Lysosome-dependent cell death, apoptosis and anoikis emerged as key PCD pathways in OS. Single-cell analysis revealed COL13A1 expression in malignant cells while GM2A and FUCA1 were enriched in macrophages. Conclusion This study identifies a robust PCD-related prognostic model for OS and highlights key genes and pathways involved in tumor progression. The present findings determine potential biomarkers and therapeutic targets to improve OS prognosis and treatment.