【摘要】 目的 筛选可用于乳腺癌早期诊断的特异性肿瘤标志物，为乳腺癌的早期诊断提供新的方法。方法 采集41例乳腺癌女性患者和20例健康人血清样本，应用气相色谱-质谱联用技术开展乳腺癌血清代谢组学研究，建立乳腺癌血清代谢指纹图谱。结果 随机森林算法(RF)能有效区分乳腺癌Ⅱ期组、乳腺癌Ⅲ期组及健康对照组。在乳腺癌组样本和健康组样本中共筛选出4种具有明显差异的代谢物：肉豆蔻酸、棕榈酸、硬脂酸、1-棕榈酸单甘油酯。结论 肉豆蔻酸、棕榈酸、硬脂酸、1-棕榈酸单甘油酯的水平变化可能反映了乳腺癌在发展过程中发生的潜在代谢变化，可引起脂肪酸代谢紊乱，有可能成为潜在的乳腺癌标志物。更多还原

【Abstract】 Objective To screen specific tumor markers for early diagnosis of breast cancer, and provide a new method for early diagnosis of breast cancer.Methods A metabonomic study of breast cancer was carried out by gas chromatography-mass spectrometry(GC-MS).20 cases of breast cancer and 41 healthy women were collected.Results Breast cancer Ⅱ phase, breast cancer Ⅲ phase and healthy control group could be effectively separated by random forests algorithm.Four distinct metabolites, myristic acid, palmitic acid, stearic acid and 1-palmitic acid monoglyceride, were selected from breast cancer group and healthy group.Conclusion The level of four metabolites of myristic acid, palmitic acid, stearic acid and 1-palmitic acid monoglyceride may reflect the potential metabolic changes in the development of breast cancer, resulting in the disorder of fatty acid metabolism, and may become a potential marker of breast cancer.更多还原