胰腺癌核磁共振血清代谢组学分析  

Analysis of nuclear magnetic resonance-based metabonomics of pancreatic cancer

作　　者：林贤超[1] 占博涵 文实[1] 李志水[2] 冯江华[2] 黄鹤光[1] 

机构地区：[1]福建医科大学附属协和医院基本外科,福州350001 [2]厦门大学电子科学系,福建省等离子体与磁共振研究重点实验室,361005

出　　处：《中华消化外科杂志》2016年第6期574-578,共5页Chinese Journal of Digestive Surgery

基　　金：基金项目：国家自然科学基金（81272581）;福建省卫生教育联合攻关计划项目（WKJ-FJ-10）;国家级临床重点专科建设项目及福建省临床重点专科建设项目[闽财指（2012）966号]

摘　　要：目的：探讨核磁共振代谢组学分析胰腺癌患者血清代谢变化的临床价值。 方法：采用回顾性病例对照研究方法。收集2013年12月至2014年12月福建医科大学附属协和医院收治的23例胰腺癌患者（设为胰腺癌组）及同期16例健康体检者（设为对照组）的临床资料。运用核磁共振方法检测胰腺癌组患者和对照组者血清，结合主成分分析（PCA）、偏最小二乘判别分析（PLSDA）、正交偏最小方差判别分析（OPLSDA）的多变量统计分析，筛选胰腺癌组患者和对照组者的差异代谢物。观察指标：（1）核磁共振氢谱数据的多变量统计分析：PCA结果、PLSDA结果和OPLSDA结果。（2）筛选差异代谢物。正态分布的计量资料以±s表示，组间比较采用t检验；计数资料组间比较采用χ2检验。 结果：（1）核磁共振氢谱数据的多变量统计分析：PCA结果显示：第一主成分（PC1）和第二主成分（PC2）对原始资料的解释率分别为54.9%、23.5%，两者的累计贡献率达78.4%。PLSDA结果显示：胰腺癌组和对照组样本得分点显示出较好的分离趋势。该模型的X矩阵解释率（R2X）、Y矩阵解释率（R2Y）和预测率（Q2）分别为0.254、0.816和0.385。OPLSDA结果显示：胰腺癌组和对照组者样本对比的差异进一步凸显，两组得分点达到很好的区分效果，以此为基础筛选两组间的差异代谢物。该模型的R2X、R2Y与Q2分别为0.254、0.816和0.433。（2）筛选差异代谢物：胰腺癌组患者和对照组者血清共归属出35种代谢物。与对照组比较，胰腺癌组患者血清中3羟基丁酸、柠檬酸、甲酸、谷氨酸、异亮氨酸、蛋氨酸、苯丙氨酸表达量升高（r=0.524，0.511，0.656，0.566，0.503，0.498，0.648，P〈0.05）；而3甲基组氨酸、丙氨酸、谷氨酰胺、低密度脂蛋白、极低密度脂蛋白表达量降低（r=-0.607，-0.508，-0.560，-0.568，-0.559，P〈0.05）。 Objective：To investigate the clinical value of serum metabonomic profile of pancreatic cancer using nuclear magnetic resonance （NMR）based metabonomics. Methods：The retrospective casecontrol study was adopted. The clinical data of 23 patients with pancreatic cancer （PC group） and 16 healthy volunteers （control group） who were admitted to the Fujian Medical University Union Hospital between December 2013 and December 2014 were collected. The serum of the 2 groups was measured by 1H NMR spectroscopy. Multivariate statistical analyses were performed to identify the characteristic metabolites in the 2 groups, including principal component analysis （PCA）, partial least squares discriminant analysis（PLSDA） and orthogonal partial least squares discriminant analysis （OPLSDA）.Observation indicators included： （1） multivariate statistical analysis of serum metabonomic profile, results of PCA, PLSDA and OPLSDA, （2） screening of metabolites. Measurement data with normal distribution were presented as ±s. The comparison between groups was evaluated with the t test. The count data were analyzed using the chisquare test. Results： （1） The multivariate statistical analysis of serum metabonomic profile： results of PCA showed that expression rates of principal component 1 （PC1） and principal component 2 （PC2） to original data were 54.9% and 23.5%, with both cumulative contribution rate of 78.4%. Results of PLSDA showed that the separative trend between PC group and control group was appeared,and variance of X and Y matrixes and predictive value were 0.254, 0.816 and 0.385. Results of OPLSDA showed that the differences of samples between the 2 groups were further increased, and differential metabolites were screened according to the distinction of scores between the 2 groups, value of R2X, R2Y and Q2 was 0.254, 0.816 and 0.433. （2） Screening of metabolites： 35 serum metabolites were detected in the 2 groups. Compared with the control group, levels of 3hydroxybuyarate, ci

关 键 词：胰腺肿瘤 核磁共振 代谢组学 诊断 血清标志物 

分 类 号：R735.9[医药卫生—肿瘤]