Chung Liping, Moore Katrina, Phillips Leo, Boyle Frances M, Marsh Deborah J, Baxter Robert C
Breast Cancer Res. 2014 Jun 16;16(3):R63. doi: 10.1186/bcr3676.
Serum profiling using proteomic techniques has great potential to detect biomarkers that might improve diagnosis and predict outcome for breast cancer patients (BC). This study used surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectrometry (MS) to identify differentially expressed proteins in sera from BC and healthy volunteers (HV), with the goal of developing a new prognostic biomarker panel.
Training set serum samples from 99 BC and 51 HV subjects were applied to four adsorptive chip surfaces (anion-exchange, cation-exchange, hydrophobic, and metal affinity) and analyzed by time-of-flight MS. For validation, 100 independent BC serum samples and 70 HV samples were analyzed similarly. Cluster analysis of protein spectra was performed to identify protein patterns related to BC and HV groups. Univariate and multivariate statistical analyses were used to develop a protein panel to distinguish breast cancer sera from healthy sera, and its prognostic potential was evaluated.
From 51 protein peaks that were significantly up- or downregulated in BC patients by univariate analysis, binary logistic regression yielded five protein peaks that together classified BC and HV with a receiver operating characteristic (ROC) area-under-the-curve value of 0.961. Validation on an independent patient cohort confirmed the five-protein parameter (ROC value 0.939). The five-protein parameter showed positive association with large tumor size (P = 0.018) and lymph node involvement (P = 0.016). By matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS, immunoprecipitation and western blotting the proteins were identified as a fragment of apolipoprotein H (ApoH), ApoCI, complement C3a, transthyretin, and ApoAI. Kaplan-Meier analysis on 181 subjects after median follow-up of >5 years demonstrated that the panel significantly predicted disease-free survival (P = 0.005), its efficacy apparently greater in women with estrogen receptor (ER)-negative tumors (n = 50, P = 0.003) compared to ER-positive (n = 131, P = 0.161), although the influence of ER status needs to be confirmed after longer follow-up.
Protein mass profiling by MS has revealed five serum proteins which, in combination, can distinguish between serum from women with breast cancer and healthy control subjects with high sensitivity and specificity. The five-protein panel significantly predicts recurrence-free survival in women with ER-negative tumors and may have value in the management of these patients.
利用蛋白质组学技术进行血清分析，在检测可能改善乳腺癌（BC）患者诊断及预测预后的生物标志物方面具有巨大潜力。本研究采用表面增强激光解吸/电离飞行时间（SELDI-TOF）质谱（MS）技术，以识别BC患者和健康志愿者（HV）血清中差异表达的蛋白质，旨在开发一种新的预后生物标志物组合。
将来自99例BC患者和51例HV受试者的训练集血清样本应用于四种吸附芯片表面（阴离子交换、阳离子交换、疏水和金属亲和），并通过飞行时间质谱进行分析。为进行验证，对100例独立的BC患者血清样本和70例HV样本进行了类似分析。对蛋白质谱进行聚类分析，以识别与BC组和HV组相关的蛋白质模式。采用单变量和多变量统计分析方法开发一个蛋白质组合，用于区分乳腺癌血清和健康血清，并评估其预后潜力。
通过单变量分析，在BC患者中有51个蛋白质峰显著上调或下调，二元逻辑回归得出5个蛋白质峰，它们共同对BC组和HV组进行分类，受试者工作特征（ROC）曲线下面积值为0.961。在独立患者队列中进行的验证证实了这一五蛋白参数（ROC值为0.939）。该五蛋白参数与肿瘤体积较大（P = 0.018）和淋巴结受累（P = 0.016）呈正相关。通过基质辅助激光解吸/电离飞行时间（MALDI-TOF）质谱、免疫沉淀和蛋白质印迹法，鉴定出这些蛋白质为载脂蛋白H（ApoH）片段、ApoCI、补体C3a、转甲状腺素蛋白和ApoAI。对181名受试者进行中位随访>5年后的Kaplan-Meier分析表明，该组合显著预测了无病生存期（P = 0.005），与雌激素受体（ER）阳性患者（n = 131，P = 0.161）相比 在ER阴性肿瘤患者（n = 50，P = 0.003）中其疗效明显更高，不过ER状态的影响需要在更长时间随访后加以证实。
质谱蛋白质质量分析揭示了五种血清蛋白，它们联合起来能够以高灵敏度和特异性区分乳腺癌女性患者的血清与健康对照者的血清。该五蛋白组合显著预测了ER阴性肿瘤女性患者的无复发生存期，可能对这些患者的管理具有价值。
