Wang Gang, Liu Panhong, Li Jiangfeng, Jin Ke, Zheng Xiangyi, Xie Liping
Department of Urology, First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310003, China.
Department of Cardiology, The Second Affiliated Hospital Zhejiang University School of Medicine, Hangzhou 310009, China.
Vaccines (Basel). 2022 Jul 21;10(7):1161. doi: 10.3390/vaccines10071161.
Clear cell renal cell carcinoma (ccRCC) is the most common type of renal carcinoma. It is particularly important to accurately judge the prognosis of patients. Since most tumor prediction models depend on the specific expression level of related genes, a better model therefore needs to be constructed. To provide an immune-related lncRNA (irlncRNAs) tumor prognosis model that is independent of the specific gene expression levels, we first downloaded and sorted out the data on ccRCC in the TCGA database and screened irlncRNAs using co-expression analysis and then obtained the differently expressed irlncRNA (DEirlncRNA) pairs by means of univariate analysis. In addition, we modified LASSO penalized regression. Subsequently, the ROC curve was drawn, and we compared the area under the curve, calculated the Akaike information standard value of the 5-year receiver operating characteristic curve, and determined the cut-off point to establish the best model to distinguish the high- or low-disease-risk group of ccRCC. Subsequently, we reassessed the model from the perspectives of survival, clinic-pathological characteristics, tumor-infiltrating immune cells, chemotherapeutics efficacy, and immunosuppressed biomarkers. A total of 17 DEirlncRNAs pairs (AL031710.1|AC104984.5, AC020907.4|AC127-24.4,AC091185.1|AC005104.1, AL513218.1|AC079015.1, AC104564.3|HOXB-AS3, AC003070.1|LINC01355, SEMA6A-AS1|CR936218.1, AL513327.1|AS005785.1, AC084876.1|AC009704.2, IGFL2-AS1|PRDM16-DT, AC011462.4|MMP25-AS1, AL662844.3I|TGB2-AS1, ARHGAP27P1|AC116914.2, AC093788.1|AC007098.1, MCF2L-AS1|AC093001.1, SMIM25|AC008870.2, and AC027796.4|LINC00893) were identified, all of which were included in the Cox regression model. Using the cut-off point, we can better distinguish patients according to different factors, such as survival status, invasive clinic-pathological features, tumor immune infiltration, whether they are sensitive to chemotherapy or not, and expression of immunosuppressive biomarkers. We constructed the irlncRNA model by means of pairing, which can better eliminate the dependence on the expression level of the target genes. In other words, the signature established by pairing irlncRNA regardless of expression levels showed promising clinical prediction value.
透明细胞肾细胞癌（ccRCC）是最常见的肾癌类型。准确判断患者的预后尤为重要。由于大多数肿瘤预测模型依赖于相关基因的特定表达水平，因此需要构建更好的模型。为了提供一种独立于特定基因表达水平的免疫相关长链非编码RNA（irlncRNAs）肿瘤预后模型，我们首先下载并整理了TCGA数据库中ccRCC的数据，通过共表达分析筛选irlncRNAs，然后通过单因素分析获得差异表达的irlncRNA（DEirlncRNA）对。此外，我们对LASSO惩罚回归进行了修正。随后，绘制ROC曲线，比较曲线下面积，计算5年接受者操作特征曲线的赤池信息准则值，并确定截断点以建立区分ccRCC高疾病风险组或低疾病风险组的最佳模型。随后，我们从生存、临床病理特征、肿瘤浸润免疫细胞、化疗疗效和免疫抑制生物标志物等方面对该模型进行了重新评估。共鉴定出17对DEirlncRNAs（AL031710.1|AC104984.5、AC020907.4|AC127 - 24.4、AC091185.1|AC005104.1、AL513218.1|AC079015.1、AC104564.3|HOXB - AS3、AC003070.1|LINC01355、SEMA6A - AS1|CR936218.1、AL513327.1|AS005785.1、AC084876.1|AC009704.2、IGFL2 - AS1|PRDM16 - DT、AC011462.4|MMP25 - AS1、AL662844.3I|TGB2 - AS1、ARHGAP27P1|AC116914.2、AC093788.1|AC007098.1、MCF2L - AS1|AC093001.1、SMIM25|AC008870.2和AC027796.4|LINC00893），所有这些都纳入了Cox回归模型。使用截断点，我们可以根据不同因素，如生存状态、侵袭性临床病理特征、肿瘤免疫浸润、对化疗是否敏感以及免疫抑制生物标志物的表达，更好地区分患者。我们通过配对构建了irlncRNA模型，该模型可以更好地消除对靶基因表达水平的依赖。换句话说，无论表达水平如何，通过配对irlncRNA建立的特征显示出有前景的临床预测价值。
