乳腺癌患者与健康人血清蛋白质谱的差异分析

目的探索乳腺癌患者与健康人群的血清蛋白质谱差异，寻找能够帮助鉴别诊断乳腺癌的候选血清蛋白标志物。方法收集117例乳腺癌患者和56例健康人的血清标本，随机分为训练组（74例乳腺癌和36例健康人）与测试组（43例乳腺癌和20例健康对照）。采用表面增强激光解析离子化飞行时间质谱（SELDI—TOFMS）技术检测所有血清标本的蛋白质谱。用Biomarker Wizard统计软件比较训练组乳腺癌与健康对照间的蛋白质谱差异，再用Biomarker Pattern软件筛选出一组差异蛋白构建决策分类树模型以鉴别乳腺癌病例和健康人群，最后用测试组对分类模型进行验证。结果乳腺癌组和健康对照组的血清蛋白质谱存在14个差异显著的蛋白峰，以质荷比分别为3958、4288、4974、5902、8518、8930、9282和11360的8个差异蛋白构建决策树分类模型，鉴别乳腺癌与健康对照组的敏感性为82．43％（61／74），特异性为83．33％（30／36），准确性为82．73％（91／110），用测试组进行验证的敏感性为86．05％（37／43），特异性为65．00％（13／20），准确性为79．37％（50／63）。结论乳腺癌与健康人群的血清蛋白质谱存在差异，SELDI—TOFMS技术筛选出的血清差异蛋白有助于乳腺癌的鉴别诊断。

Analysis of the difference of Serum Proteome Profiles between breast cancer patients and heathy women

Objective To investigate the difference of serum proteome between breast cancer patients and healthy women, and seek for some potential biomarkers to distinguish breast caner patients from healthy women using a serum protein fingerprinting technique. Methods Total serum samples collected from 117 breast cancer patients and 56 healthy women were divided into a learn-group （74 breast cancer patients and 36 healthy） women and a test-group （43 breast cancer patients and 30 healthy women）. The samples were detected by surface enhanced laser desorption/ionization time-of-flight mass spectrometry （SELDI-TOF-MS）. Differences of mass spectrums between breast cancer patients and healthy women from the learn-group were analyzed by Biomarker Wizard software. Some different protein peaks were picked out and a classification tree was developed for identification of breast cancer patients using Biomarker Pattern software. The classification model was validated by an independent sample from the test-group. Results Fourteen protein peaks were significantly different between breast cancer and healthy women groups. Eight proteins mass/charge of 3958,4288,4974,5902,8518,8930, 9282 and 11 360 could identify breast cancer patients from healthy women with decision tree classification method. Sensitivity, specificity and correct ratio of the method were 82. 430/00 （61/74）,83.33% （30/36） and 82.73 % （91/110）,respectively. In validation test, the sensitivity, specificity and correct ratio of the method were 86.05% （37/43）,65.00% （13/20） and 79.37% （50/63）,respectively. Conclusions There are some different proteins in serum between breast cancer patients and healthy women. Candidate protein biomarkers in serum screened by SELDI-TOF MS could facilitate the diagnosis of breast cancer.