DataSheet1_Identification of a basement membrane-based risk scoring system for prognosis prediction and individualized therapy in clear cell renal cell carcinoma.ZIP

Clear cell renal cell carcinoma (ccRCC) belongs to one of the 10 most frequently diagnosed cancers worldwide and has a poor prognosis at the advanced stage. Although multiple therapeutic agents have been proven to be curative in ccRCC, their clinical application was limited due to the lack of reliable biomarkers. Considering the important role of basement membrane (BM) in tumor metastasis and TME regulation, we investigated the expression of BM-related genes in ccRCC and identified prognostic BM genes through differentially expression analysis and univariate cox regression analysis. Then, BM-related ccRCC subtypes were recognized through consensus non-negative matrix factorization based on the prognostic BM genes and evaluated with regard to clinical and TME features. Next, utilizing the differentially expressed genes between the BM-related subtypes, a risk scoring system BMRS was established after serial analysis of univariate cox regression analysis, lasso regression analysis, and multivariate cox regression analysis. Time-dependent ROC curve revealed the satisfactory prognosis predictive capacity of BMRS with internal, and external validation. Multivariate analysis proved the independent predictive ability of BMRS and a BMRS-based nomogram was constructed for clinical application. Some featured mutants were discovered through genomic analysis of the BMRS risk groups. Meanwhile, the BMRS groups were found to have distinct immune scores, immune cell infiltration levels, and immune-related functions. Moreover, with the help of data from The Cancer Immunome Atlas (TCIA) and Genomics of Drug Sensitivity in Cancer (GDSC), the potential of BMRS in predicting therapeutic response was evaluated and some possible therapeutic compounds were proposed through ConnectivityMap (CMap). For the practicability of BMRS, we validated the expression of BMRS-related genes in clinical samples. After all, we identified BM-related ccRCC subtypes with distinct clinical and TME features and constructed a risk scoring system for the prediction of prognosis, therapeutic responses, and potential therapeutic agents of ccRCC. As ccRCC systemic therapy continues to evolve, the risk scoring system BMRS we reported may assist in individualized medication administration.

透明细胞肾细胞癌（clear cell renal cell carcinoma, ccRCC）是全球十大高发恶性肿瘤之一，晚期患者预后较差。尽管已有多种治疗药物被证实对ccRCC具有治疗效果，但由于缺乏可靠的生物标志物，其临床应用受到限制。鉴于基底膜（basement membrane, BM）在肿瘤转移及肿瘤微环境（tumor microenvironment, TME）调控中的重要作用，本研究探讨了ccRCC中BM相关基因的表达情况，并通过差异表达分析与单因素Cox回归分析（univariate Cox regression analysis）筛选出预后相关BM基因。随后，基于上述预后相关BM基因，通过一致性非负矩阵分解（consensus non-negative matrix factorization）算法识别出ccRCC的BM相关亚型，并结合临床特征与TME特征对各亚型进行评估。接下来，利用BM相关亚型间的差异表达基因，经单因素Cox回归分析、套索回归分析（lasso regression analysis）及多因素Cox回归分析（multivariate Cox regression analysis）的系列分析，构建了风险评分系统BMRS。时间依赖性受试者工作特征曲线（time-dependent receiver operating characteristic curve, time-dependent ROC curve）分析显示，BMRS具有良好的预后预测能力，且经内部验证与外部验证证实。多因素分析证实BMRS具有独立预测价值，进而构建了基于BMRS的列线图（nomogram）以用于临床实践。通过对BMRS风险组进行基因组分析，本研究还发现了若干特征性突变。同时，不同BMRS风险组的免疫评分、免疫细胞浸润水平及免疫相关功能均存在显著差异。此外，借助癌症免疫组图谱（The Cancer Immunome Atlas, TCIA）与癌症药物敏感性基因组学（Genomics of Drug Sensitivity in Cancer, GDSC）数据集，本研究评估了BMRS预测治疗响应的潜力，并通过连接图谱（ConnectivityMap, CMap）筛选出若干潜在治疗化合物。为验证BMRS的临床实用性，本研究在临床样本中验证了BMRS相关基因的表达水平。综上，本研究成功识别出具有不同临床特征与TME特征的ccRCC BM相关亚型，并构建了可用于预测ccRCC患者预后、治疗响应及潜在治疗药物的风险评分系统。随着ccRCC系统治疗的不断发展，本研究报道的BMRS风险评分系统有望为患者的个体化用药提供辅助参考。