Table_7_Pyroptosis-Related lncRNA Prognostic Model for Renal Cancer Contributes to Immunodiagnosis and Immunotherapy.docx

BackgroundRenal clear cell cancer (ccRCC) is one of the most common cancers in humans. Thus, we aimed to construct a risk model to predict the prognosis of ccRCC effectively.
MethodsWe downloaded RNA sequencing (RNA-seq) data and clinical information of 539 kidney renal clear cell carcinoma (KIRC) patients and 72 normal humans from The Cancer Genome Atlas (TCGA) database and divided the data into training and testing groups randomly. Pyroptosis-related lncRNAs (PRLs) were obtained through Pearson correlation between pyroptosis genes and all lncRNAs (p < 0.05, coeff > 0.3). Univariate and multivariate Cox regression analyses were then performed to select suitable lncRNAs. Next, a novel signature was constructed and evaluated by survival analysis and ROC analysis. The same observation applies to the testing group to validate the value of the signature. By gene set enrichment analysis (GSEA), we predicted the underlying signaling pathway. Furthermore, we calculated immune cell infiltration, immune checkpoint, the T-cell receptor/B-cell receptor (TCR/BCR), SNV, and Tumor Immune Dysfunction and Exclusion (TIDE) scores in TCGA database. We also validated our model with an immunotherapy cohort. Finally, the expression of PRLs was validated by quantitative PCR (qPCR).
ResultsWe constructed a prognostic signature composed of six key lncRNAs (U62317.1, MIR193BHG, LINC02027, AC121338.2, AC005785.1, AC156455.1), which significantly predict different overall survival (OS) rates. The efficiency was demonstrated using the receiver operating characteristic (ROC) curve. The signature was observed to be an independent prognostic factor in cohorts. In addition, we found the PRLs promote the tumor progression via immune-related pathways revealed in GSEA. Furthermore, the TCR, BCR, and SNV data were retrieved to screen immune features, and immune cell scores were calculated to measure the effect of the immune microenvironment on the risk model, indicating that high- and low-risk scores have different immune statuses. The TIDE algorithm was then used to predict the immune checkpoint blockade (ICB) response of our model, and subclass mapping was used to verify our model in another immunotherapy cohort data. Finally, qPCR validates the PRLs in cell lines.
ConclusionThis study provided a new risk model to evaluate ccRCC and may be pyroptosis-related therapeutic targets in the clinic.

研究背景
肾透明细胞癌（renal clear cell cancer, ccRCC）是人类最常见的恶性肿瘤之一，因此本研究旨在构建一款预后风险模型，以有效预测ccRCC的预后。
研究方法
本研究从癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库下载了539例肾透明细胞癌（kidney renal clear cell carcinoma, KIRC）患者与72例正常人群的RNA测序（RNA-seq）数据及临床信息，并将数据随机划分为训练集与测试集。通过焦亡基因与所有长链非编码RNA（long non-coding RNA, lncRNA）的皮尔逊相关分析（p < 0.05，相关系数>0.3），筛选得到焦亡相关lncRNA（pyroptosis-related lncRNAs, PRLs）。随后开展单因素与多因素Cox回归分析以筛选合适的lncRNA。接下来构建了一款新型风险标签，并通过生存分析与受试者工作特征（receiver operating characteristic, ROC）分析对其进行评估；同时在测试集中开展相同分析以验证该标签的临床价值。通过基因集富集分析（gene set enrichment analysis, GSEA），本研究预测了潜在的信号通路。此外，本研究还计算了TCGA数据库中的免疫细胞浸润评分、免疫检查点、T细胞受体/B细胞受体（T-cell receptor/B-cell receptor, TCR/BCR）、单核苷酸变异（single nucleotide variant, SNV）以及肿瘤免疫功能异常与排斥（Tumor Immune Dysfunction and Exclusion, TIDE）评分。本研究同时利用免疫治疗队列对模型进行了验证。最后，通过定量聚合酶链式反应（quantitative PCR, qPCR）验证了PRLs的表达水平。
研究结果
本研究构建了由6个关键lncRNA（U62317.1、MIR193BHG、LINC02027、AC121338.2、AC005785.1、AC156455.1）组成的预后风险标签，该标签可显著区分不同的总生存期（overall survival, OS）。受试者工作特征曲线证实了该标签的预测效能，且该标签在各队列中均为独立的预后因素。此外，基因集富集分析结果显示，PRLs通过免疫相关通路促进肿瘤进展。进一步通过检索TCR、BCR与SNV数据筛选免疫特征，并计算免疫细胞评分以评估免疫微环境对风险模型的影响，结果显示高、低风险评分组具有不同的免疫状态。本研究随后采用TIDE算法预测模型的免疫检查点阻断（immune checkpoint blockade, ICB）应答情况，并通过亚类映射（subclass mapping）在另一免疫治疗队列数据中验证了本模型。最后，通过qPCR在细胞系中验证了PRLs的表达。
研究结论
本研究构建了一款可用于评估ccRCC预后的新型风险模型，其或可成为临床中与焦亡相关的治疗靶点。