Objective To identify cuproptosis-related lncRNAs (CRLs) and use them to construct models to predict survival in osteosarcoma (OS) patients.

Methods RNA-seq data of OS patients were downloaded from the TARGET database along with relevant clinical information. Cuproptosis-related gene sets were obtained from related studies. (CRLs) associated with OS survival were screened using co-expression analysis as well as univariate Cox regression. The OS prognostic models were constructed using LASSO-Cox regression, and the model efficacy was assessed by receiver operating characteristic curve (ROC) and Kaplan-Meier (KM) survival analysis. Single-sample gene set enrichment analysis (ssGSEA) was utilized to explore the relationship between CRLs model scores and signaling pathways in OS. Functional and pathway enrichment analyses were performed by Gene Ontology (GO) and Kyoto Encyclopedia of Genomes (KEGG) in OS patients of different risk groups. The level of immune cell infiltration in tumor samples from OS patients was inferred by the ESTIMATE algorithm. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was used to verify the expression of the four CRLs in different cell lines.

Results A total of 374 lncRNAs co-expressed with 10 cuproptosis-related RNAs were identified in 85 OS patients. Based on the results of Cox regression analysis, 62 CRLs associated with the prognosis of OS patients were identified. Consensus clustering analysis revealed that OS patients had two CRLs expression patterns, and there was a significant difference in the prognosis of patients with 2 CRLs expression patterns. LASSO- Cox regression analysis was used to construct a prognostic model for CRLs. The tROC curves were used to assess the model' s efficacy, and the results showed AUC values of 0.78, 0.83, and 0.85 for years 1, 3, and 5, respectively, and the results of the KM survival analyses differed significantly between the high- and low-risk groups (P＜0.05). GSEA functional enrichment analysis found antigen receptor-mediated signaling pathway, B lymphocyte activation, and positive T lymphocytes were enriched in the high-risk group. GO and KEGG enrichment analyses revealed that the tumor-related pathways were presenting differences in different risk groups. The expression levels of CRLs in prognostic models were verified in three OS cell lines.

Conclusion CRLs are closely associated with the prognosis of OS patients. A prognostic model constructed based on CRLs accurately predicts the prognosis of OS patients, and further in-depth study of the role of CRLs in OS may contribute to the development of more reliable and personalized therapeutic regimens.