基于TCPA数据库的结直肠癌预后相关蛋白预后模型的建立与应用

Establishment and application of protein prognosis model for colorectal cancer based on TCPA database

目的利用TCPA数据库建立预测结直肠癌预后的蛋白预后模型,分析蛋白预后模型与结直肠癌各临床病理特征之间的关系。方法通过R语言软件筛选出TCPA数据库中结直肠癌样本及正常样本中差异表达的蛋白质,进一步利用COX比例风险回归模型筛选和建立多蛋白预后模型,计算预后评分。根据预后评分的中位数将患者分为高风险组和低风险组,利用Kaplan-Meier生存曲线分析高风险组及低风险组生存差异。将临床病理因素和预后评分因素纳入COX回归模型分析结直肠癌患者的生存影响因素。根据临床病理因素与结直肠癌相关蛋白预后模型预后评分的相关性,验证多蛋白模型对结直肠癌患者中的预后价值。结果共筛选出7个差异蛋白(BIM,SRC,BID,CYCLIND,SRC_pY527,IGFBP2,INPP4B)建立结直肠癌相关蛋白预后模型。Kaplan-Meier生存分析证实,在结直肠癌患者中,预后评分低风险的患者总体生存率明显升高,差异有统计学意义(P<0.001)。COX回归分析显示,结直肠癌预后相关蛋白预后模型是结直肠癌患者预后的独立影响因素(P<0.01)。结论结直肠癌预后相关蛋白预后模型可用于预测结直肠癌患者的预后,有利于进一步指导临床治疗。

Objective To establish a protein prognosis model for predicting the prognosis of colorectal cancer using TCPA database,and to analyze the relationship between protein prognosis model and clinicopathological features of colorectal cancer.Methods R language software was used to screen differentially expressed proteins in colorectal cancer samples and normal samples in TCPA database.COX proportional hazard regression model was used to screen and establish multi-protein prognosis model,and the prognosis score was calculated.The patients were divided into the high-risk group and the low-risk group according to the median of prognosis score.The survival difference between the high-risk group and the low-risk group was analyzed by Kaplan-Meier survival curve.COX regression model was used to analyze the influencing factors for the survival of patients with colorectal cancer.According to the correlation between clinicopathological factors and prognosis score of the protein prognosis model related to colorectal cancer,the prognosis value of multi-protein model in patients with colorectal cancer was verified.Results Seven differential proteins(BIM,SRC,BID,CYCLIND,SRC_pY527,IGFBP2,and INPP4B)were screened to establish the prognosis model of colorectal cancer related proteins.Kaplan Meier survival analysis confirmed that in the overall patients with colorectal cancer,the overall survival rate of patients with low risk of prognosis score was significantly higher,and the difference was statistically significant(P<0.001).COX regression analysis showed that the prognosis related protein model was an independent prognostic factor(P<0.01).Conclusion The prognosis related protein model for colorectal cancer can be used to predict the prognosis of colorectal cancer patients,which is helpful to further guide the clinical treatment.