Zhang Xiaoting, Su Yue, Fu Xian, Xiao Jing, Qin Guicheng, Yu Mengli, Li Xiaofeng, Chen Guihong
Shenzhen Bao'an District Songgang People's Hospital, Shenzhen, China.
School of Pharmaceutical Sciences, Guangzhou Medical University, Guangzhou, China.
J Oncol. 2022 Jan 13;2022:9273628. doi: 10.1155/2022/9273628. eCollection 2022.
Lung squamous cell carcinoma (LUSC) is the most common type of lung cancer accounting for 40% to 51%. Long noncoding RNAs (lncRNAs) have been reported to play a significant role in the invasion, migration, and proliferation of lung cancer tissue cells. However, systematic identification of lncRNA signatures and evaluation of the prognostic value for LUSC are still an urgent problem. In this work, LUSC RNA-seq data were collected from TCGA database, and the limma R package was used to screen differentially expressed lncRNAs (DElncRNAs). In total, 216 DElncRNAs were identified between the LUSC and normal samples. lncRNAs associated with prognosis were calculated using univariate Cox regression analysis. The overall survival (OS) prognostic model containing 10 lncRNAs and the disease-free survival (DFS) prognostic model consisting of 11 lncRNAs were constructed using a machine learning-based algorithm, systematic LASSO-Cox regression analysis. We found that the survival rate of samples in the high-risk group was lower than that in the low-risk group. Results of ROC curves showed that both the OS and DFS risk score had better prognostic effects than the clinical characteristics, including age, stage, gender, and TNM. Two lncRNAs (LINC00519 and FAM83A-AS1) that were commonly identified as prognostic factors in both models could be further investigated for their clinical significance and therapeutic value. In conclusion, we constructed lncRNA prognostic models with considerable prognostic effect for both OS and DFS of LUSC.
肺鳞状细胞癌（LUSC）是最常见的肺癌类型，占比40%至51%。据报道，长链非编码RNA（lncRNAs）在肺癌组织细胞的侵袭、迁移和增殖中发挥着重要作用。然而，lncRNA特征的系统鉴定以及对LUSC预后价值的评估仍然是一个紧迫的问题。在这项工作中，从TCGA数据库收集了LUSC的RNA测序数据，并使用limma R包筛选差异表达的lncRNAs（DElncRNAs）。总共在LUSC样本和正常样本之间鉴定出216个DElncRNAs。使用单变量Cox回归分析计算与预后相关的lncRNAs。使用基于机器学习的算法——系统LASSO-Cox回归分析，构建了包含10个lncRNAs的总生存期（OS）预后模型和由11个lncRNAs组成的无病生存期（DFS）预后模型。我们发现，高风险组样本的生存率低于低风险组。ROC曲线结果显示，OS和DFS风险评分的预后效果均优于包括年龄、分期、性别和TNM在内的临床特征。在两个模型中均被共同鉴定为预后因素的两个lncRNAs（LINC00519和FAM83A-AS1），其临床意义和治疗价值可进一步研究。总之，我们构建了对LUSC的OS和DFS均具有显著预后效果的lncRNA预后模型。
