Comprehensive analysis of a glycolysis and cholesterol synthesis-related genes signature for predicting prognosis and immune landscape in osteosarcoma

Abstract

BackgroundGlycolysis and cholesterol synthesis are crucial in cancer metabolic reprogramming. The aim of this study was to identify a glycolysis and cholesterol synthesis-related genes (GCSRGs) signature for effective prognostic assessments of osteosarcoma patients.MethodsGene expression data and clinical information were obtained from GSE21257 and TARGET-OS datasets. Consistent clustering method was used to identify the GCSRGs-related subtypes. Univariate Cox regression and LASSO Cox regression analyses were used to construct the GCSRGs signature. The ssGSEA method was used to analyze the differences in immune cells infiltration. The pRRophetic R package was utilized to assess the drug sensitivity of different groups. Western blotting, cell viability assay, scratch assay and Transwell assay were used to perform cytological validation.ResultsThrough bioinformatics analysis, patients diagnosed with osteosarcoma were classified into one of 4 subtypes (quiescent, glycolysis, cholesterol, and mixed subtypes), which differed significantly in terms of prognosis and tumor microenvironment. Weighted gene co-expression network analysis revealed that the modules strongly correlated with glycolysis and cholesterol synthesis were the midnight blue and the yellow modules, respectively. Both univariate and LASSO Cox regression analyses were conducted on screened module genes to identify 5 GCSRGs (RPS28, MCAM, EN1, TRAM2, and VEGFA) constituting a prognostic signature for osteosarcoma patients. The signature was an effective prognostic predictor, independent of clinical characteristics, as verified further via Kaplan-Meier analysis, ROC curve analysis, univariate and multivariate Cox regression analysis. Additionally, GCSRGs signature had strong correlation with drug sensitivity, immune checkpoints and immune cells infiltration. In cytological experiments, we selected TRAM2 as a representative gene to validate the validity of GCSRGs signature, which found that TRAM2 promoted the progression of osteosarcoma cells. Finally, at the pan-cancer level, TRAM2 had been correlated with overall survival, progression free survival, disease specific survival, tumor mutational burden, microsatellite instability, immune checkpoints and immune cells infiltration.ConclusionTherefore, we constructed a GCSRGs signature that efficiently predicted osteosarcoma patient prognosis and guided therapy.

Identifying squalene epoxidase as a metabolic vulnerability in high‐risk osteosarcoma using an artificial intelligence‐derived prognostic index

“…However, because of gene naming alterations and gene absences in microarray datasets, only 53 published signatures and the AIDPI were tested in at least one cohort, and the results were visualized in a heat map (Figure 1N). Strikingly, this heat map showed that only two signatures consistently exhibited statistical significance across all cohorts: our AIDPI and Xu et al.’s signature 12 . A direct C‐index comparison underscored AIDPI's definitive superiority over Xu et al.’s signature within three cohorts (Figure 1O).…”

“…Strikingly, this heat map showed that only two signatures consistently exhibited statistical significance across all cohorts: our AIDPI and Xu et al.’s signature. 12 A direct C‐index comparison underscored AIDPI's definitive superiority over Xu et al.’s signature within three cohorts (Figure 1O ). These findings demonstrate that the AIDPI can predict OSA patients’ outcomes and outperforms previously established signatures.…”

“…However, the high costs of collecting transcriptomic, genomic, and epigenomic profiles could hinder the clinical application of this strategy. Conversely, emerging studies suggest that simplified gene expression signatures could predict OSA outcomes cost‐effectively 11,12 . However, many of these signatures are based on specific gene sets or small datasets, resulting in doubts regarding their potential for application in clinical settings.…”

“…Conversely, emerging studies suggest that simplified gene expression signatures could predict OSA outcomes cost‐effectively. 11 , 12 However, many of these signatures are based on specific gene sets or small datasets, resulting in doubts regarding their potential for application in clinical settings.…”

The Effect of Cholesterol in MCF7 Human Breast Cancer Cells

“…Hypercholesterolemia facilitated the entry of Chol into cancer cells [ 10 ] due to the overexpression of the LDL receptor in these cells [ 11 ]. Moreover, metabolic reprogramming in transformed cells was described [ 12 ]. Thus, the upregulation of enzymes for Chol synthesis was demonstrated in cancer cells [ 13 , 14 ].…”