A seven-lncRNA signature for predicting Ewing’s sarcoma

• DOI: 10.7717/peerj.11599
• Publication Date: 2021-06-17T08:43:48Z
• Authors: Zhihui Chen; Xinyu Wang; Guozhu Wang; Bin Xiao; Zhe Ma; Hongliang Huo; Weiwei Li
• Keywords: Nomogram; Lasso; Univariate

Long non-coding RNAs (lncRNAs) are a class of with unique characteristics. These RNA can regulate cancer cells' survival, proliferation, invasion, metastasis, and angiogenesis potential diagnostic prognostic markers. We identified seven-lncRNA signature related to the overall survival (OS) patients Ewing's sarcoma (EWS).We used an expression profile from Gene Expression Omnibus (GEO) database as training cohort screen out OS-associated lncRNAs in EWS further established using univariate Cox regression, least absolute shrinkage, selection operator (LASSO) regression analysis. The lncRNA was validated external dataset International Cancer Genome Consortium (ICGC) validation cohort.We obtained 10 survival-related Kaplan-Meier ROC curve analysis (log-rank test P < 0.05; AUC >0.6). Univariate LASSO analyses confirmed seven key we predict prognosis. were categorized into low-risk group or high-risk based on their median risk score. group's time significantly shorter than group's. This by cohort. area under (AUC) for this up 0.905 0.697 3-year group. nomogram's calibration curves demonstrated that probability two cohorts consistent between nomogram prediction actual observation.We screened patients' Our findings provide new reference current evaluation direction diagnosis treatment EWS.