铜死亡相关差异长链非编码RNA评分公式在肾透明细胞癌中的临床价值

Clinical value of a cuproptosis-related differential long non-coding RNA scoring formula in clear cell renal cell carcinoma

目的探讨基于生物信息学方法构建的肾透明细胞癌(ccRCC)预后相关的铜死亡相关差异长链非编码RNA(lncRNA)评分公式在患者临床诊断、预后预测以及治疗抉择中的价值。方法从癌症基因组图谱(TCGA)数据库获取ccRCC患者的基因矩阵和临床数据(数据更新至2022年3月29日)。基因矩阵收录539例ccRCC组织和72例癌旁正常组织表达数据;临床数据收录530例ccRCC患者资料。采用Pearson相关性分析、Wilcoxon符合秩检验和单因素Cox比例风险模型分析筛选预后相关的铜死亡相关差异lncRNA。采用R软件将530例包含生存资料的ccRCC患者以近似1∶1的比例随机抽样分组,分为训练集(266例)和验证集(264例)。采用LASSO回归构建铜死亡相关差异lncRNA评分公式并进行交叉验证。采用受试者工作特征(ROC)曲线评估铜死亡相关差异lncRNA评分公式的特异度和灵敏度,计算曲线下面积(AUC)。根据中位风险值将ccRCC患者分为低风险组和高风险组,采用Kaplan-Meier法分析高、低风险组患者总生存(OS)的差异。采用t检验分析不同临床病理特征患者的风险值差异。采用R包rms构建可以预测ccRCC患者1、3、5年OS率的列线图。采用R包pRRophetic预测ccRCC患者临床治疗中常见的靶向药物(如索拉菲尼、舒尼替尼等)的半数抑制浓度(IC50),采用Wilcoxon符号秩检验比较高、低风险组患者IC50值。收集山西医科大学第一医院2021年6月至12月行根治性肾切除术患者经病理检测确诊的20例ccRCC组织及其配对癌旁正常组织。实时荧光定量聚合酶链反应(qRT-PCR)检测ccRCC组织关键lncRNA的表达。结果基于TCGA数据库中ccRCC患者10个铜死亡基因(FDX1、LIAS、LIPT1、DLD、DLAT、PDHA1、PDHB、MTF1、GLS、CDKN2A)的表达矩阵,筛选出153个与预后相关的铜死亡相关差异lncRNA。LASSO回归分析,最终得到包含4个铜死亡相关差异lncRNA的评分公式:风险值=0.020×AC015912.3+0.011×AC026401.3+0.063×AC103706.1+(-0.076)×EPB41L4A-DT。根据中位风险值(0.76)将患者分为高风险组(≥0.76)和低风险组(<0.76)。ROC曲线结果显示,评分公式对于患者1、3、5年OS率的预测准确性较好。训练集、验证集、总队列中高风险组均较低风险组患者OS更差(均P<0.001)。年龄、病理分级、肿瘤分期、根据铜死亡相关差异lncRNA评分公式计算的风险值是OS的独立影响因素(均P<0.001)。铜死亡相关差异lncRNA评分公式计算得到的风险值在不同病理分级、肿瘤分期、T分期、N分期、M分期患者间差异均有统计学意义(均P<0.01),但在不同性别、年龄患者间差异均无统计学意义(均P>0.05)。构建的列线图对于患者1、3、5年OS率具有较好的预后准确性。高风险组应用舒尼替尼、西罗莫司敏感性更高,低风险组应用阿昔替尼、索拉非尼、帕唑帕尼敏感性更高。qRT-PCR检测结果表明,ccRCC肿瘤组织与癌旁组织比较,AC015912.3基因相对表达量为1.00±0.04比0.68±0.24(t=6.37,P<0.01),AC026401.3基因相对表达量为1.00±0.05比0.64±0.22(t=7.29,P<0.01),AC103706.1基因相对表达值为1.00±0.04比0.64±0.20(t=7.49,P<0.01),EPB41L4A-DT基因相对表达量为1.00±0.06、0.73±0.10(t=10.68,P<0.01)。结论基于TCGA数据库构建的铜死亡相关差异lncRNA评分公式可作为ccRCC患者临床诊断、预后预测的新型指标,帮助指导患者的临床药物治疗,有助于精准诊疗。

Objective To investigate the value of a cuproptosis-related differential long non-coding RNA(lncRNA)scoring formula related to the prognosis of clear cell renal cell carcinoma(ccRCC)patients in the clinical diagnosis,prognosis prediction and treatment options based on bioinformatics.Methods Gene matrix and clinical data of ccRCC patients were obtained from the Cancer Genome Atlas(TCGA)database(update to 29 March,2022).The expression data of 539 ccRCC tissues and 72 paracancerous normal tissues were collected from gene matrix;the data of 530 ccRCC were collected from clinical data.Pearson correlation analysis,Wilcoxon signed rank test and univariate Cox proportional risk model were used to analyze the screened cuproptosis-related differential lncRNA related to the prognosis.R software was used to randomly divide 530 ccRCC patients with survival data into training set(266 cases)and validation set(264 cases)according to approximate 1∶1 ratio.LASSO regression analysis was used to construct a cuproptosis-related differential lncRNA scoring formula and cross-validation was performed.Receiver operating characteristic(ROC)curve analysis was used to evaluate the specificity and sensitivity of cuproptosis-related differential lncRNA scoring formula,and the area of the curve(AUC)was calculated.According to the median risk value,all patients were divided into the low-risk group and high-risk group;Kaplan-Meier method was used to analyze the difference in the overall survival(OS)of patients in the low-risk group and high-risk group.T test was used to detect the differences in the risk value of patients with different clinicopathological characteristics.R package rms was used to construct the nomogram for predicting 1-year,3-year,5-year OS rates of ccRCC patients,R package pRRophetic was used to predict the half-inhibitory concentration(IC50)of common targeted drugs such as sorafenib and sunitinib in clinical treatment of ccRCC patients,and IC50 value of patients in low-risk group and high-risk group was compared by using Wilcoxon signed rank test.Tissue samples of 20 ccRCC patients who underwent radical nephrectomy and were diagnosed with pathology and the matched paracancerous normal tissues were collected from the First Hospital of Shanxi Medical University between June 2021 and December 2021.Real-time fluorescence quantitative polymerase chain reaction(qRT-PCR)was used to detect the expression of key lncRNA in ccRCC tissues.Results Based on the expression matrix of 10 cuproptosis genes(FDX1,LIAS,LIPT1,DLD,DLAT,PDHA1,PDHB,MTF1,GLS,CDKN2A)of ccRCC patients in TCGA database,153 cuproptosis-related differential lncRNA related to the prognosis were identified.According to LASSO regression analysis,a scoring formula of 4 cuproptosis-related differential lncRNA related to the prognosis was obtained,risk value was calculated as 0.020×AC015912.3+0.011×AC026401.3+0.063×AC103706.1+(-0.076)×EPB41L4A-DT.All patients were divided into high-risk group(≥0.76)and low-risk group(<0.76)based on the median value(0.76).ROC curve analysis showed that the scoring formula had good prediction accuracy in 1-year,3-year,5-year OS rates.In training set,validation set,the total cohort,the OS of patients in the high-risk group was worse than that in the low-risk group(all P<0.001).The age,pathological degree,tumor staging,risk value calculated by cuproptosis-related differential lncRNA were independent influencing factors of OS(all P<0.001).There were statistically significant differences in the risk value calculated by cuproptosis-related differential lncRNA scoring formula among patients with different pathological degree,tumor staging,T staging,N staging,M staging(all P<0.01),while there were no statistically significant differences among patients with different gender and age(all P>0.05).The established nomogram had good prediction accuracy in the 1-year,3-year,5-year OS rates.Sunitinib and sirolimus showed higher sensitivity in the high-risk group;axitinib,sorafenib and pazopanib showed higher sensitivity in the low-risk group.qRT-PCR results showed that relative expression level of AC015912.3 in ccRCC tissues was up-regulated compared with paracancerous tissues(1.00±0.04 vs.0.68±0.24,t=6.37,P<0.01);the relative expression level of AC026401.3 in ccRCC tissues was up-regulated compared with paracancerous tissues(1.00±0.05 vs.0.64±0.22,t=7.29,P<0.01);the relative expression level of AC103706.1 in ccRCC tissues was up-regulated compared with paracancerous tissues(1.00±0.04 vs.0.64±0.21,t=7.49,P<0.01);the relative expression level of EPB41L4A-DT in ccRCC tissues was up-regulated compared with paracancerous tissues(1.00±0.06 vs.0.73±0.10,t=10.68,P<0.01).Conclusions Cuproptosis-related differential lncRNA scoring formula based on TCGA database can be used as a new marker for clinical diagnosis and prognosis prediction of ccRCC patients,which can help guide the clinical drug treatment of patients and facilitate accurate diagnosis and treatment.