Objective: To explore the alteration of plasma metabolomic profiles, screen the new serum markers of multidrug resistant epithelial ovarian cancer (EOC), and investigate the mechanism. Methods: The serum of 132 cases with cisplatin-resistant EOC, cisplatin-sensitive EOC, benign ovarian cyst and healthy donors were collected. Differentially plasma metabolic profiles were identified by liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS). The significantly different metabolites of each group were screened by using principal component analysis. Then compounds that played a key role in cisplatin resistance were identified by using nuclear magnetic resonance (NMR). The relationships between these compounds and clinical characteristics and prognosis were analyzed. Results: LC-MS/MS identified 25 800 metabolic compounds. According to the descending dimension algorithm by principal component analysis, six compounds which were the biggest contributor to grouping were identified. The identified results of NMR showed that the serum level of C16 Sphinganine was lower while Dodemorph was higher in the EOC than those of the normal control. Compared to the cisplatin sensitive group, cisplatin resistant group exhibited a specific metabolic trait characterized by upregulation of 1-Monopalmitin, Ricinoleic acid methyl ester, Polyoxyethylene (600) mono-ricinoleate/Glycidyl stearate and downregulation of Calycanthidine. The four components were all associated with fatty acid metabolism, and the combinational diagnostic sensitivity of these biomarkers for cisplatin-resistance was 86.50% and the specificity was 81.80%, the area of receiver operating characteristic (ROC) curve was 0.93. Conclusions: The metabolic signatures of normal control, benign ovarian cyst, cisplatin sensitivity and cisplatin resistance can be clearly separated from each other by LC-MS/MS technology.The combinational four biomarkers including Calycanthidine, 1-Monopalmitin, Ricinoleic acid methl ester and Polyoxyethylene (600) mono-ricinoleate/Glycidyl stearate are more sensitive and specific for the diagnosis of cisplatin resistant EOC, and may provide the potentially predict markers of chemotherapeutic response in metabolic level. The fatty acid metabolism may participate in the cisplatin resistant progression of EOC.
目的： 应用代谢组学技术研究上皮性卵巢癌多药耐药患者血清代谢指纹谱，筛选卵巢上皮癌多药耐药相关的诊断标志物，并从代谢水平探索卵巢上皮癌多药耐药的发生机制。 方法： 收集卵巢上皮癌铂类耐药患者、铂类敏感患者、卵巢良性囊肿及正常对照者的血清共132例，采用液质联用正离子模式检测4组血清样品的代谢指纹谱，通过主成分分析方法筛选出对分组意义重大的差异代谢物，进一步采用核磁共振技术对其进行鉴定，并分析其与临床特征和预后的关系。 结果： 液质联用共鉴定到25 800个代谢化合物，使用主成分分析对数据降维后，得到6个对分组贡献最大的差异代谢物。核磁共振技术的进一步鉴定显示，相对于正常对照组，Dodemorph在卵巢上皮癌组血清中高表达，C16 Sphinganine则表达下调。相对于卵巢癌铂类敏感组，Calycanthidine在铂类耐药组中表达下调，而1－Monopalmitin、Ricinoleic acid methl ester、Polyoxyethylene (600) mono－ricinoleate/Glycidyl stearate在铂类耐药组中表达上调，四者均与脂肪酸代谢相关，且联合诊断铂类耐药的敏感度为86.50%，特异度为81.80%，受试者工作特征曲线下面积(AUC)为0.93。 结论： 采用液质联用技术能较好地区分卵巢上皮癌铂类耐药患者、铂类敏感患者、卵巢良性囊肿及正常对照者，其中差异物Calycanthidine、1－Monopalmitin、Ricinoleic acid methl ester和Polyoxyethylene (600) mono－ricinoleate/Glycidyl stearate联合诊断卵巢上皮癌铂类耐药具有较高的敏感度和特异度，可作为潜在的诊断标志物。脂肪酸的代谢可能参与卵巢上皮癌铂类耐药的生物过程。.
Keywords: Biomarkers; Chemo-resistance; Metabolomics; Ovarian neoplasms.