Table_3_Using Immune-Related Long Non-coding Ribonucleic Acids to Develop a Novel Prognosis Signature and Predict the Immune Landscape of Colon Cancer.XLS

Purpose: This study aimed to construct a novel signature to predict the survival of patients with colon cancer and the associated immune landscape, based on immune-related long noncoding ribonucleic acids (irlncRNAs). Methods: Expression profiles of irlncRNAs in 457 patients with colon cancer were retrieved from the TCGA database (https://portal.gdc.cancer.gov). Differentially expressed (DE) irlncRNAs were identified and irlncRNA pairs were recognized using Lasso regression and Cox regression analyses. Akaike information criterion (AIC) values of receiver operating characteristic (ROC) curve were calculated to identify the ideal cut-off point for dividing patients into two groups and constructing the prognosis signature. Quantitative real-time polymerase chain reaction (qRT-PCR) was performed to validate the expression of LINC02195 and SCARNA9 in colon cancer. Results: We identified 22 irlncRNA pairs and patients were divided into high-risk and low-risk groups based on the calculated risk score using these 22 irlncRNA pairs. The irlncRNA pairs were significantly related to patient survival. Low-risk patients had a significantly longer survival time than high-risk patients (p < 0.001). The area under the curve of the signature to predict 5-year survival was 0.951. The risk score correlated with tumor stage, infiltration depth, lymph node metastasis, and distant metastasis. The risk score remained significant after univariate and multivariate Cox regression analyses. A nomogram model to predict patient survival was developed based on the results of Cox regression analysis. Immune cell infiltration status, expression of some immune checkpoint genes, and sensitivity to chemotherapeutics were also related to the risk score. The results of qRT-PCR revealed that LINC02195 and SCARNA9 were significantly upregulated in colon cancer tissues. Conclusion: The constructed prognosis signature showed remarkable efficiency in predicting patient survival, immune cell infiltration status, expression of immune checkpoint genes, and sensitivity to chemotherapeutics.研究目的：本研究旨在基于免疫相关长链非编码RNA（immune-related long noncoding ribonucleic acids，irlncRNAs），构建一种新型特征模型以预测结肠癌患者的生存情况及其相关免疫特征图谱。 研究方法：从癌症基因组图谱（The Cancer Genome Atlas，TCGA）数据库（https://portal.gdc.cancer.gov）中获取457例结肠癌患者的irlncRNA表达谱数据。通过Lasso回归与Cox回归分析，筛选出差异表达（differentially expressed，DE）的irlncRNA，并识别出irlncRNA对；计算受试者工作特征（receiver operating characteristic，ROC）曲线的赤池信息准则（Akaike information criterion，AIC）值，以确定将患者分为两组的最佳截断值，进而构建预后特征模型。采用实时荧光定量聚合酶链反应（quantitative real-time polymerase chain reaction，qRT-PCR）验证LINC02195与SCARNA9在结肠癌组织中的表达水平。 研究结果：本研究共筛选出22对irlncRNA，并基于该22对irlncRNA计算得到的风险评分将患者分为高风险组与低风险组。上述irlncRNA对与患者生存情况显著相关，低风险组患者的生存时长显著长于高风险组（p < 0.001）。该特征模型预测5年生存率的曲线下面积为0.951。风险评分与肿瘤分期、浸润深度、淋巴结转移及远处转移均显著相关。经单因素与多因素Cox回归分析后，风险评分仍具有统计学显著性。基于Cox回归分析结果，构建了用于预测患者生存情况的列线图模型。免疫细胞浸润状态、部分免疫检查点基因的表达水平以及化疗药物敏感性同样与风险评分相关。qRT-PCR结果显示，LINC02195与SCARNA9在结肠癌组织中显著上调表达。 研究结论：本研究构建的预后特征模型在预测患者生存情况、免疫细胞浸润状态、免疫检查点基因表达水平以及化疗药物敏感性方面均表现出优异的效能。