论文部分内容阅读
目的使用SELDI-TOF蛋白质质谱,分析和寻找影响三阴性乳腺癌预后的新指标。方法采集浙江省肿瘤医院提供的三阴性乳腺癌样本51例,每例样本均包含病理及质谱数据。将病例数据分为死亡或复发组与无复发组,并统计检验两组间病理因素有无显著差异。首先统计分析两组样本的蛋白质质谱数据,检验样本中具有显著差异的蛋白质峰;然后对具有显著差异的蛋白质峰所在的特征位点使用SVM分类器分析三阴性乳腺癌的预后。结果统计结果显示,病理因素很难作为精确预测三阴性乳腺癌预后的因素。死亡或复发样本组与无复发组蛋白质质谱存在显著差异。在51例三阴性乳腺癌蛋白质质谱中,发现了7个蛋白质峰P值小于0.01,联合7个峰分类,分类率达90.20%,敏感性达82.35%,特异性达94.12%。结论 SELDI-TOF蛋白质质谱数据分析能够较好地区分三阴性乳腺癌死亡或复发样本与无复发样本,挑选出的蛋白质峰可以为三阴性乳腺癌的预后判断提供新的有价值的标志物。
Objective To analyze and find new markers that influence the prognosis of triple negative breast cancer using SELDI-TOF protein mass spectrometry. Methods 51 cases of triple negative breast cancer samples collected from Zhejiang Tumor Hospital were collected, and pathology and mass spectrometry data were included in each sample. The case data were divided into death or recurrent group and non-recurrent group, and statistically test the pathological factors between the two groups was no significant difference. First of all, the two groups of samples were statistically analyzed for protein mass spectrometry data to test samples with significant differences in protein peaks; and then with significant differences in the protein peak locus using SVM classifier analysis of triple negative breast cancer prognosis. Results The statistical results show that pathological factors are difficult to predict accurately prognosis of triple negative breast cancer. There was a significant difference in protein mass spectrum between death or recurrent samples and non-recurrent ones. In 51 cases of triple negative breast cancer protein spectrum, seven proteins were found to have P values less than 0.01, combined with seven peak classification, the classification rate was 90.20%, the sensitivity was 82.35% and the specificity was 94.12%. Conclusion SELDI-TOF protein mass spectrometry data analysis can distinguish between triple negative breast cancer death or recurrence samples and no recurrence samples. The selected protein peaks can provide new valuable markers for the prognosis of triple negative breast cancer.