Identification and Validation of a Novel Immune-related Five-lncRNA Signature for Osteosarcoma

Abstract Background: Osteosarcoma (OS) is the most common malignant tumor in primary bone cancer. It often occurs in children or adolescents. It has the characteristics of strong invasiveness, early blood metastasis, and poor prognosis. Recent studies have shown that the tumor microenvironment (TME) and immune-related lncRNAs (IR-lncRNAs) play an important role in tumor prognosis. Therefore, exploring prognostic biomarkers that reflect the bioheterogeneity of OS can provide better interventions for patients.Methods: The gene expression data and corresponding clinical data of OS patients were downloaded from the GEO (n=102) and TARGET (n=101) databases. We applied single-sample gene set enrichment analysis (ssGSEA) to the transcriptomes of these samples, generating the three immune subgroups. The ESTIMATE algorithm, CIBERSORT, and Kaplan–Meier analysis were used to explore the TME and tumor-infiltrating immune cells (TIICs) among the 3 immune subgroups. The immune gene set in the ImmPort database was used to screen out immune-related mRNAs (IR-mRNAs) in the two databases, and IR-lncRNAs were selected through coexpression with IR-mRNAs. Based on these groupings, we identified differentially expressed immune-related lncRNAs (IRD-lncRNAs) between the high- and low-immune subgroups. Then, an IRD-lncRNA prognostic signature was established by least absolute shrinkage and selection operator (LASSO) regression.Results: We found that as the immune status of the immune subgroups increased, the prognosis significantly improved (P<0.05). Moreover, the stromal, immune, and ESTIMATE scores increased, while tumor purity decreased (P<0.01). The TIIC results showed that CD8+ T cells had a positive correlation with the prognosis of OS (P<0.01). After using the ImmPort database and coexpression analysis, we screened 3767 IR-lncRNAs and identified 117 IRD-lncRNAs between the immune-H and immune-L subgroups. Finally, we identified five IRD-lncRNAs as a prognostic signature for OS. Kaplan–Meier, ROC curve and univariate and multivariate Cox regression analyses further confirmed that the prognostic signature was an innovative and significant independent prognostic factor.Conclusion: The study results may help clarify the impact of the TME on OS and provide a method to predict the prognosis of OS patients using the five-IR-lncRNA signature as a potential biomarker.