六种血清肿瘤标志物对结肠癌的诊断价值研究
发布时间：2019-06-21 08:44
【摘要】：目的:探讨血清肿瘤标志物CEA、CA199、CA724、CA125、CA242和Tu M2-PK单项及联合检测对结肠癌的诊断价值,筛选出最优的单项和联合诊断结肠癌肿瘤的标志物,并采用代谢组学的策略分别建立基于血清6种标志物的主成分分析(PCA)、经典判别分析、Logistic回归分析和偏最小二乘-判别分析(PLS-DA)模型,评估各模型对结肠癌的诊断和预测价值,找出对结肠癌早期诊断最有帮助的模型。方法:收集结肠癌患者、结肠良性病变和健康对照组外周血,采用电化学发光法测定血清中CEA、CA199、CA724、CA125和CA242含量,酶联免疫吸附试验(ELISA)检测血清中Tu M2-PK的含量。组内比较采用单因素方差分析,组间比较采用(Least-significant Difference)LSD检验,并以受试者工作特征曲线(ROC)下面积(AUC)的比较单个指标的诊断效能,以Logistic回归-ROC评估联合指标的诊断效能。采用PCA模型分析结肠癌、结肠良性疾病和健康对照组血清肿瘤标志物的代谢谱,以经典判别分析、Logistic回归分析和偏最小二乘-判别分析(PLS-DA)模型评估结肠癌的诊断和预测价值。结果:结肠癌患者血清中CEA、CA199、CA125、CA242和Tu M2-PK水平均显著高于良性疾病和健康对照,差异具有统计学意义(P0.05)。良性肿瘤组血清CEA、CA199、CA724、CA242和Tu M2-PK显著高于健康对照组(P0.05)。六种血清肿瘤标志物中CEA、CA199、CA724和Tu M2-PK的诊断效能较高(AUC≥0.70)。其中CA199具有最高的诊断效能,在临界值为69.5U/L时,具有64.1%的灵敏度和89.7%的特异度。单项标志物中Tu M2-PK诊断结肠癌的灵敏度与CEA和CA199相当。联合检测两项标志物,(CA199+Tu M2-PK)的诊断效能最高(AUC=0.87),具有77.5%的灵敏度和90.0%的特异度。联合检测三项血清肿瘤标志物,(CEA+CA199+Tu M2-PK)的诊断效能最高,具有82.3%的灵敏度和91.5%的特异性。联合检测四项标志物(CEA+CA72-4+Tu M2-PK+CA19-9),具有87.5%的灵敏度、83.0%的特异度和84.5%的准确度。PCA模型表明,结肠癌患者血清6种肿瘤标志物水平明显紊乱,结肠癌患者个体分散,正常个体和结肠良性病患者个体变异小,三组个体有分离的趋势。以AUC最高三项肿瘤标志物(CEA+CA199+Tu M2-PK)建立Fisher经典判别分析模型,该模型对结肠癌和非结肠癌(结肠良性疾病和正常对照组)的诊断正确率分别为(162/231)70.1%和(227/271)81.9%。以AUC最高三项肿瘤标志物(CEA+CA199+Tu M2-PK)建立PLS-DA诊断模型,该模型对结肠癌和非结肠癌(结肠良性病组和对照组)诊断的准确率分别为(179/231)77.5%和(259/271)95.6%。结论:六种血清肿瘤标志物当中,以CA199对结肠癌的早期诊断最有价值。而联合检测血清肿瘤标志物CEA、CA199和Tu M2-PK是较为经济、有效的组合标志物,有助于结肠癌的早期诊断。基于血清肿瘤标志物CEA、CA199和Tu M2-PK的Logistic回归和偏最小二乘判别分析(PLS-DA)诊断模型为结肠癌的诊断提供了可行有效的新模式,有助于结肠癌的早期诊断。
[Abstract]:Objective: to investigate the diagnostic value of serum tumor markers CEA,CA199,CA724,CA125,CA242 and Tu M2-PK in the diagnosis of colon cancer, to screen out the optimal markers for the diagnosis of colon cancer tumors, and to establish the classical discriminant analysis of (PCA), Logistic regression analysis and partial least squares discriminant analysis (PLS-DA) models based on the strategy of metabonomics. To evaluate the value of each model in the diagnosis and prediction of colon cancer, and to find the most helpful model for the early diagnosis of colon cancer. Methods: the peripheral blood of patients with colon cancer, benign lesions of colon and healthy control group were collected. The contents of CEA,CA199,CA724,CA125 and CA242 in serum were determined by electrochemiluminescence method, and the content of Tu M2-PK in serum was detected by enzyme-linked immunosorbent assay (ELISA). One factor analysis of variance was used in the group, (Least-significant Difference) LSD test), and the diagnostic efficacy of the single index was compared with the area (AUC) under the working characteristic curve (ROC) of the subjects, and the diagnostic effectiveness of the combined index was evaluated by Logistic regression ROC. PCA model was used to analyze the metabolic spectrum of serum tumor markers in colon cancer, benign colon diseases and healthy controls. Classical discriminant analysis, Logistic regression analysis and partial least squares discriminant analysis (PLS-DA) model were used to evaluate the diagnostic and predictive value of colon cancer. Results: the levels of CEA,CA199,CA125,CA242 and Tu M2-PK in serum of patients with colon cancer were significantly higher than those of benign diseases and healthy controls (P 0.05). Serum CEA,CA199,CA724,CA242 and Tu M2-PK in benign tumor group were significantly higher than those in healthy control group (P 0.05). Among the six serum tumor markers, CEA,CA199,CA724 and Tu M2-PK were more effective (AUC 鈮，
本文编号：2503944
[Abstract]:Objective: to investigate the diagnostic value of serum tumor markers CEA,CA199,CA724,CA125,CA242 and Tu M2-PK in the diagnosis of colon cancer, to screen out the optimal markers for the diagnosis of colon cancer tumors, and to establish the classical discriminant analysis of (PCA), Logistic regression analysis and partial least squares discriminant analysis (PLS-DA) models based on the strategy of metabonomics. To evaluate the value of each model in the diagnosis and prediction of colon cancer, and to find the most helpful model for the early diagnosis of colon cancer. Methods: the peripheral blood of patients with colon cancer, benign lesions of colon and healthy control group were collected. The contents of CEA,CA199,CA724,CA125 and CA242 in serum were determined by electrochemiluminescence method, and the content of Tu M2-PK in serum was detected by enzyme-linked immunosorbent assay (ELISA). One factor analysis of variance was used in the group, (Least-significant Difference) LSD test), and the diagnostic efficacy of the single index was compared with the area (AUC) under the working characteristic curve (ROC) of the subjects, and the diagnostic effectiveness of the combined index was evaluated by Logistic regression ROC. PCA model was used to analyze the metabolic spectrum of serum tumor markers in colon cancer, benign colon diseases and healthy controls. Classical discriminant analysis, Logistic regression analysis and partial least squares discriminant analysis (PLS-DA) model were used to evaluate the diagnostic and predictive value of colon cancer. Results: the levels of CEA,CA199,CA125,CA242 and Tu M2-PK in serum of patients with colon cancer were significantly higher than those of benign diseases and healthy controls (P 0.05). Serum CEA,CA199,CA724,CA242 and Tu M2-PK in benign tumor group were significantly higher than those in healthy control group (P 0.05). Among the six serum tumor markers, CEA,CA199,CA724 and Tu M2-PK were more effective (AUC 鈮，
本文编号：2503944
本文链接：https://www.wllwen.com/yixuelunwen/zlx/2503944.html
