Development and validation of apoptosis-related signature and molecular subtype to improve prognosis prediction in osteosarcoma patients

Machine learning-based individualized survival prediction model for prognosis in osteosarcoma: Data from the SEER database

Patient outcomes of osteosarcoma vary because of tumor heterogeneity and treatment strategies. This study aimed to compare the performance of multiple machine learning (ML) models with the traditional Cox proportional hazards (CoxPH) model in predicting prognosis and explored the potential of ML models in clinical decision-making. From 2000 to 2018, 1243 patients with osteosarcoma were collected from the Surveillance, Epidemiology, and End Results (SEER) database. Three ML methods were chosen for model development (DeepSurv, neural multi-task logistic regression [NMTLR]) and random survival forest [RSF]) and compared them with the traditional CoxPH model and TNM staging systems. 871 samples were used for model training, and the rest were used for model validation. The models' overall performance and predictive accuracy for 3- and 5-year survival were assessed by several metrics, including the concordance index (C-index), the Integrated Brier Score (IBS), receiver operating characteristic curves (ROC), area under the ROC curves (AUC), calibration curves, and decision curve analysis. The efficacy of personalized recommendations by ML models was evaluated by the survival curves. The performance was highest in the DeepSurv model (C-index, 0.77; IBS, 0.14; 3-year AUC, 0.80; 5-year AUC, 0.78) compared with other methods (C-index, 0.73-0.74; IBS, 0.16-0.17; 3-year AUC, 0.73-0.78; 5-year AUC, 0.72-0.78). There are also significant differences in survival outcomes between patients who align with the treatment option recommended by the DeepSurv model and those who do not (hazard ratio, 1.88; P < .05). The DeepSurv model is available in an approachable web app format at https://survivalofosteosarcoma.streamlit.app/. We developed ML models capable of accurately predicting the survival of osteosarcoma, which can provide useful information for decision-making regarding the appropriate treatment.

A Novel Defined Necroptosis-Related Genes Prognostic Signature for Predicting Prognosis and Treatment of Osteosarcoma

Osteosarcoma (OS) is a frequent primary malignant bone tumor, with a poor prognosis. Necroptosis is strongly correlated with OS and may be an influential target for treating OS. This study's objective was to establish a necroptosis-related gene (NRG) prognostic signature that could predict OS prognosis and guide OS treatment. First, we identified 20 NRGs associated with OS survival based on the TARGET database. We then derived a 7 NRG prognostic signature. Our findings revealed that the 7 NRG prognostic signature performed well in predicting the survival of OS patients. We next analyzed differences in immunological status and immune cell infiltration. In addition, we examined the relationship between chemo/immunotherapeutic response and the 7-NRG prognostic signature. In addition, to probe the mechanisms underlying the NRG prognostic signature, we performed functional enrichment assays including GO and KEGG. Finally, CHMP4C was selected for functional experiments. Silencing CHMP4C prevented OS cells from proliferating, migrating, and invading. This 7-NRG prognostic signature seems to be an excellent predictor that can provide a fresh direction for OS treatment.

Construction of molecular subtype model of osteosarcoma based on endoplasmic reticulum stress and tumor metastasis-related genes

Introduction

Osteosarcoma, the prevailing primary bone malignancy among children and adolescents, is frequently associated with treatment failure primarily due to its pronounced metastatic nature.

Methods

This study aimed to establish potential associations between hub genes and subtypes for the treatment of metastatic osteosarcoma. Differentially expressed genes were extracted from patients diagnosed with metastatic osteosarcoma and a control group of non-metastatic patients, using the publicly available gene expression profile (GSE21257). The intersection of these gene sets was determined by focusing on endoplasmic reticulum (ER) stress-related genes sourced from the GeneCards database. We conducted various analytical techniques, including functional and pathway enrichment analysis, WGCNA analysis, protein-protein interaction (PPI) network construction, and assessment of immune cell infiltration, using the intersecting genes. Through this analysis, we identified potential hub genes.

Results

Osteosarcoma subtype models were developed using molecular consensus clustering analysis, followed by an examination of the associations between each subtype and hub genes. A total of 138 potential differentially expressed genes related to endoplasmic reticulum (ER) stress were identified. These genes were further investigated using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA) pathways. Additionally, the PPI interaction network revealed 38 interaction relationships among the top ten hub genes. The findings of the analysis revealed a strong correlation between the extent of immune cell infiltration and both osteosarcoma metastasis and the expression of hub genes. Notably, the differential expression of the top ten hub genes was observed in osteosarcoma clusters 1 and 4, signifying their significant association with the disease.

Conclusion

The identification of ten key genes linked to osteosarcoma metastasis and endoplasmic reticulum stress bears potential clinical significance. Additionally, exploring the molecular subtype of osteosarcoma has the capacity to guide clinical treatment decisions, necessitating further investigations and subsequent clinical validations.

A glycometabolic gene signature associating with immune infiltration and chemosensitivity and predicting the prognosis of patients with osteosarcoma

Background

Accumulating evidence has suggested that glycometabolism plays an important role in the pathogenesis of tumorigenesis. However, few studies have investigated the prognostic values of glycometabolic genes in patients with osteosarcoma (OS). This study aimed to recognize and establish a glycometabolic gene signature to forecast the prognosis, and provide therapeutic options for patients with OS.

Methods

Univariate and multivariate Cox regression, LASSO Cox regression, overall survival analysis, receiver operating characteristic curve, and nomogram were adopted to develop the glycometabolic gene signature, and further evaluate the prognostic values of this signature. Functional analyses including Gene Ontology (GO), kyoto encyclopedia of genes and genomes analyses (KEGG), gene set enrichment analysis, single-sample gene set enrichment analysis (ssGSEA), and competing endogenous RNA (ceRNA) network, were used to explore the molecular mechanisms of OS and the correlation between immune infiltration and gene signature. Moreover, these prognostic genes were further validated by immunohistochemical staining.

Results

A total of four genes including PRKACB, SEPHS2, GPX7, and PFKFB3 were identified for constructing a glycometabolic gene signature which had a favorable performance in predicting the prognosis of patients with OS. Univariate and multivariate Cox regression analyses revealed that the risk score was an independent prognostic factor. Functional analyses indicated that multiple immune associated biological processes and pathways were enriched in the low-risk group, while 26 immunocytes were down-regulated in the high-risk group. The patients in high-risk group showed elevated sensitivity to doxorubicin. Furthermore, these prognostic genes could directly or indirectly interact with other 50 genes. A ceRNA regulatory network based on these prognostic genes was also constructed. The results of immunohistochemical staining showed that SEPHS2, GPX7, and PFKFB3 were differentially expressed between OS tissues and adjacent normal tissues.

Conclusion

The preset study constructed and validated a novel glycometabolic gene signature which could predict the prognosis of patients with OS, identify the degree of immune infiltration in tumor microenvironment, and provide guidance for the selection of chemotherapeutic drugs. These findings may shed new light on the investigation of molecular mechanisms and comprehensive treatments for OS.

Identification and validation of an anoikis-related lncRNA signature to predict prognosis and immune landscape in osteosarcoma

Background

Anoikis is a specialized form of programmed apoptosis that occurs in two model epithelial cell lines and plays an important role in tumors. However, the prognostic value of anoikis-related lncRNA (ARLncs) in osteosarcoma (OS) has not been reported.

Methods

Based on GTEx and TARGET RNA sequencing data, we carried out a thorough bioinformatics analysis. The 27 anoikis-related genes were obtained from the Gene Set Enrichment Analysis (GSEA). Univariate Cox regression and least absolute shrinkage and selection operator (LASSO) analysis were successively used to screen for prognostic-related ARLncs. To create the prognostic signature of ARLncs, we performed multivariate Cox regression analysis. We calculated the risk score based on the risk coefficient, dividing OS patients into high- and low-risk subgroups. Additionally, the relationship between the OS immune microenvironment and risk prognostic models was investigated using function enrichment, including Gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), single-sample gene set enrichment analysis (ssGSEA), and GSEA analysis. Finally, the potential effective drugs in OS were found by immune checkpoint and drug sensitivity screening.

Results

A prognostic signature consisting of four ARLncs (AC079612.1, MEF2C-AS1, SNHG6, and TBX2-AS1) was constructed. To assess the regulation patterns of anoikis-related lncRNA genes, we created a risk score model. According to a survival analysis, high-risk patients have a poor prognosis as they progress. By using immune functional analysis, the lower-risk group demonstrated the opposite effects compared with the higher-risk group. GO and KEGG analysis showed that the ARLncs pathways and immune-related pathways were enriched. Immune checkpoints and drug sensitivity analysis might be used to determine the better effects of the higher group.

Conclusion

We identified a novel prognostic model based on a four-ARLncs signature that might serve as potential prognostic indicators that can be used to predict the prognosis of OS patients, and immunotherapy and drugs that may contribute to improving the overall survival of OS patients and advance our understanding of OS.