Identification and verification of a polyamine metabolism-related gene signature for predicting prognosis and immune infiltration in osteosarcoma.

BACKGROUND: Although an established correlation exists between tumor cell proliferation and elevated polyamine levels, research on polyamine metabolism in osteosarcoma (OS) remains limited. This study aimed to identify polyamine metabolism-related genes (PMRGs) associated with OS prognosis and develop a prognostic model, thereby offering novel insights into targeted therapies for patients with OS. METHODS: Datasets related to OS and PMRGs were sourced from publicly accessible databases. Candidate genes were initially identified through differential expression and weighted gene co-expression network analyses. Subsequently, prognostic genes were screened using univariate Cox and least absolute shrinkage and selection operator (LASSO) regression analyses, leading to the development of a risk model. Furthermore, a nomogram model was developed using variables selected through univariate Cox regression analysis. The relationship between the signature and immune landscape was also analyzed. Following the pre-processing of single-cell RNA sequencing data, a cell communication analysis was conducted based on the identified cell types. Finally, the expression levels of prognostic genes in clinical samples were verified using reverse transcription quantitative polymerase chain reaction, western blotting and immunohistochemistry. RESULTS: Ninety-six candidate genes were selected for univariate Cox and LASSO regression analyses, leading to the identification of eight prognostic genes: FAM162A, SIGMAR1, SQLE, PYCR1, DDI1, PAQR6, GRIA1, and TNFRSF12A. The risk model constructed from these genes demonstrated strong predictive accuracy and classified patients into two risk groups based on the median cut-off. A nomogram model was developed, incorporating the risk score as an independent prognostic factor. The high-risk cohort exhibited lower single-sample gene set enrichment analysis scores for 17 immune cell types and reduced expression levels of seven immune checkpoint-related genes. Furthermore, eight cell types were identified, among which endothelial cells, cancer-associated fibroblasts, osteoclasts, myeloid cells, and osteoblast OS cells showed significant interactions with NK/T, B, and plasma cells. Eight prognostic genes were confirmed to be overexpressed in OS tissues. CONCLUSION: The identification of FAM162A, SIGMAR1, SQLE, PYCR1, DDI1, PAQR6, GRIA1, and TNFRSF12A as prognostic genes associated with PMRGs in OS provides valuable references for prognostic assessment and personalized treatment in patients with OS.