Chen Guo, Zhang Tiansheng, Li Feng, Cui Chi, Huang Zhiyong, Gou Xin, Song Yajun, Li Yang
Department of Urology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, 400042, China.
Department of Urology, Mianyang Central Hospital, University of Electronic Science and Technology of China, Sichuan Province, 621099, China.
J Cancer. 2024 Apr 29;15(11):3481-3494. doi: 10.7150/jca.94685. eCollection 2024.
Tumor angiogenesis is closely related to the progression of clear cell renal cell carcinoma (ccRCC). Long non-coding RNAs (lncRNAs) regulating angiogenesis could be potential biomarkers for predicting ccRCC prognosis. With this study, we aimed to construct a prognostic model based on lncRNAs and explore its underlying mechanisms. RNA data and clinical information were obtained from The Cancer Genome Atlas (TCGA) database. Angiogenesis-related genes (ARGs) were extracted from the Molecular Signatures database. Pearson correlation and LASSO and COX regression analyses were performed to identify survival-related AR-lncRNAs (sAR-lncRNAs) and construct a prognostic model. The predictive power of the prognostic model was verified according to Kaplan‒Meier curve, receiver operating characteristic (ROC) curve and nomogram analyses. The correlation between the prognostic model and clinicopathological characteristics was assessed via univariate and multivariate analyses. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was subsequently performed to elucidate the mechanisms of the sAR-lncRNAs. In vitro qPCR, immunohistochemistry, migration and invasion assays were conducted to confirm the angiogenic function of sAR-lncRNAs. Three sAR-lncRNAs were used to construct a prognostic model. The model was moderately accurate in predicting 1- , 3- and 5-year ccRCC prognosis, and the risk score according to this model was closely related to clinicopathological characteristics such as T grade and T stage. A nomogram was constructed to precisely estimate the overall survival of ccRCC patients. KEGG enrichment analysis indicated that the MAPK and Notch pathways were highly enriched in high-risk patients. Additionally, we found that the expression of the lncRNAs AC005324.4 and AC104964.4 in the prognostic model was lower in ccRCC cell lines and cancer tissues than in the HK-2 cell line and paracancerous tissues, while the expression of the lncRNA AC087482.1 showed the opposite trend. In a coculture model, knockdown of lncRNA AC005324.4 and lncRNA AC104964.4 significantly promoted the migration and invasion of human umbilical vein endothelial cells (HUVECs), but siR-AC087482.1 transfection alleviated these effects. We constructed a prognostic model based on 3 sAR-lncRNAs and validated its value in clinicopathological characteristics and prognostic prediction of ccRCC patients, providing a new perspective for ccRCC treatment decision making.
肿瘤血管生成与透明细胞肾细胞癌（ccRCC）的进展密切相关。调控血管生成的长链非编码RNA（lncRNAs）可能是预测ccRCC预后的潜在生物标志物。通过本研究，我们旨在构建基于lncRNAs的预后模型并探索其潜在机制。从癌症基因组图谱（TCGA）数据库获取RNA数据和临床信息。从分子特征数据库中提取血管生成相关基因（ARGs）。进行Pearson相关性分析、LASSO分析和COX回归分析以鉴定与生存相关的AR-lncRNAs（sAR-lncRNAs）并构建预后模型。根据Kaplan-Meier曲线、受试者工作特征（ROC）曲线和列线图分析验证预后模型的预测能力。通过单因素和多因素分析评估预后模型与临床病理特征之间的相关性。随后进行京都基因与基因组百科全书（KEGG）富集分析以阐明sAR-lncRNAs的机制。进行体外qPCR、免疫组织化学、迁移和侵袭实验以证实sAR-lncRNAs的血管生成功能。使用三个sAR-lncRNAs构建预后模型。该模型在预测1年、3年和5年ccRCC预后方面具有中等准确性，并且根据该模型的风险评分与T分级和T分期等临床病理特征密切相关。构建列线图以精确估计ccRCC患者的总生存期。KEGG富集分析表明，丝裂原活化蛋白激酶（MAPK）和Notch信号通路在高危患者中高度富集。此外，我们发现预后模型中的lncRNAs AC005324.4和AC104964.4在ccRCC细胞系和癌组织中的表达低于HK-2细胞系和癌旁组织，而lncRNA AC087482.1的表达呈现相反趋势。在共培养模型中，敲低lncRNA AC005324.4和lncRNA AC104964.4显著促进人脐静脉内皮细胞（HUVECs）的迁移和侵袭，但转染siR-AC087482.1可减轻这些作用。我们基于3个sAR-lncRNAs构建了预后模型，并验证了其在ccRCC患者临床病理特征和预后预测中的价值，为ccRCC治疗决策提供了新的视角。
