基于铜死亡相关lncRNAs的胰腺癌预后模型构建与验证

Construction and validation of a prognostic signature for patients with pancreatic adenocarcinoma based on cuproptosis-associated lncRNAs

目的研究铜死亡相关lncRNAs在胰腺癌(PAAD)患者中的预后预测价值,并进一步构建预后预测模型。方法从TCGA数据库中下载胰腺癌患者的转录组测序数据和相应临床信息,通过Pearson相关性分析筛选与预后相关的铜死亡相关lncRNAs,先后利用单因素Cox回归和Lasso回归分析并进一步构建预后模型。根据模型的风险评分中位数,将所有患者分为高风险组和低风险组。通过Kaplan-Meier生存分析、亚组分析、ROC曲线分析及一致性指数分析评估模型的预后预测价值,并利用单因素和多因素回归分析验证模型的独立性。对高、低风险组的差异表达基因进行GO及KEGG功能富集分析,并对高、低风险组患者进行肿瘤突变负荷(TMB)分析、免疫治疗反应预测以及药物敏感性分析。结果通过Pearson相关性分析,确定了127个铜死亡相关的lncRNAs,先后利用单因素Cox回归分析及Lasso回归分析构建了一个基于6个铜死亡相关lncRNAs的预后预测模型。根据模型计算结果将PAAD患者队列分成高风险组和低风险组,Kaplan-Meier生存分析表明低风险组患者的生存时间要长于高风险组(P<0.05)。ROC曲线证明了该模型对胰腺癌患者预后的预测性能良好:1、3、5年ROC曲线下面积分别为0.687、0.753、0.771;基因功能富集分析表明,高、低风险组差异表达基因主要富集于免疫相关通路。此外,高风险组患者的TMB值明显大于低风险组,而TIDE评分明显低于低风险组。最后,通过药物敏感性分析发现不同组的胰腺癌患者对特定药物的敏感性存在统计学差异,对临床用药具有一定的指导意义。结论本研究基于铜死亡相关lncRNAs成功构建了一个PAAD患者预后模型,可精准预测PAAD患者的预后,并为患者的临床药物治疗选择提供个性化指导。

Objective To explore the prognostic value of cuproptosis-associated lncRNAs in pancreatic cancer patients and develop a prognostic prediction model.Methods Transcriptome sequencing data and corresponding clinical information of pancreatic cancer patients were obtained from the TCGA database.Prognosis-associated cuproptosis-associated lncRNAs were identified through Pearson correlation analysis.A prognostic model was then constructed using univariate Cox regression and Lasso regression analysis.Patients were divided into high and low-risk groups based on the median risk score from the model.The prognostic predictive value of the model was evaluated using Kaplan-Meier survival analysis,subgroup analysis,ROC curve analysis,and concordance index analysis.The model’s independence of was confirmed through univariate and multivariate regression analysis.Differential genes expression between high and low-risk groups was analyzed using GO and KEGG functional enrichment analyses.Furthermore,tumor mutation burden(TMB)analysis,immune therapy response prediction,and drug sensitivity analysis were performed for patients in both risk groups.Results Through Pearson correlation analysis,127 cuproptosis-associated lncRNAs were identified.Subsequently,a prognostic prediction model based on six cuproptosis-associated lncRNAs was constructed using univariate Cox regression analysis and Lasso regression analysis.Stratification of PAAD patients into high-risk and low-risk groups was performed based on the model’s calculations.Kaplan-Meier survival analysis indicated that patients in the lowrisk group had significantly longer survival time compared to those in the high-risk group(P<0.05).ROC curve analysis showed good predictive performance of the model for pancreatic cancer patient prognosis,with AUCs of 0.687(1 year),0.753(3 years),and 0.771(5 years),respectively.Gene function enrichment analysis revealed that differentially expressed genes between the high and low-risk groups were mainly enriched in immunerelated pathways.Moreover,patients in the high-risk group exhibited significantly higher TMB values but lower TIDE scores compared to those in the low-risk group.Furthermore,drug sensitivity analysis revealed significant differences in the sensitivity of pancreatic cancer patients to specific drugs among the different risk groups,providing valuable guidance for clinical treatment decisions.Conclusion This study successfully constructed a prognostic signature for pancreatic cancer patients based on cuproptosis-associated lncRNAs,providing accurate prognosis predicitions and personalized guidance for clinical drug treatment selection.