AUTHOR=Pan Jiahui , Zhang Xinyue , Fang Xuedong , Xin Zhuoyuan 

TITLE=Construction on of a Ferroptosis-Related lncRNA-Based Model to Improve the Prognostic Evaluation of Gastric Cancer Patients Based on Bioinformatics

JOURNAL=Frontiers in Genetics

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.739470

DOI=10.3389/fgene.2021.739470

ISSN=1664-8021

ABSTRACT=Background: Gastric cancer is one of the most serious gastrointestinal malignancies with bad prognosis. Ferroptosis is an iron-dependent form of programmed cell death, which may affect the prognosis of gastric cancer patients. Long non-coding RNAs (lncRNAs) can affect the prognosis of cancer through regulating ferroptosis process, which could be potential overall survival prediction factors for gastric cancer.
Methods: Ferroptosis-related lncRNA expression profiles, clinicopathological and survival data were collected from The Cancer Genome Atlas and FerrDB database, which DESeq2 method. Through co-expression and functional enrichment analysis, we identified the potential effects of ferroptosis-related lncRNAs towards prognosis. Using COX regression analysis with LASSO method, we constructed a 17 ferroptosis-related lncRNA prognostic signature model. And we also evaluated the prognostic power with Kaplan–Meier (K-M) survival curve, receiver operating characteristic (ROC) curve, as well as Decision Curve Analysis (DCA) methods.
Results: Ferroptosis-related lncRNA-mRNA co-expression network were constructed. And functional enrichment analysis revealed the “FoxO signaling pathway” and “HIF-1 signaling pathway” were dys-regulated, which might control the prognosis of gastric cancer patients. Then, a ferroptosis-related gastric cancer prognostic signature model, including 17 lncRNAs was constructed. Based on the RiskScore calculated using this model, the patients were divided into High-Risk and Low-Risk group. And K-M survival curve analysis revealed the higher RiskScore is, the worse prognosis obtained. ROC curve analysis shown the AUC of our model is 0.751, which was better than other published models. And the multi-variate COX regression analysis demonstrated the lncRNA signature is an independent risk factor for the overall survival rate. Finally, based on nomogram and DCA, we observed a better clinical practicality potential for gastric cancer patient prognostic prediction.
Conclusions: Our 17 ferroptosis-related lncRNAs prognostic signature model may improve the overall survival prediction in gastric cancer.