富AT结合域蛋白1A在乳腺癌中的表达及其预后意义

Expression of AT-rich interactive domain-containing protein 1A in breast cancer and prognostic significance

目的旨在通过检测富AT结合域蛋白1A(ARID1A)在乳腺良恶性肿瘤中的表达差异,探讨其对患者预后的影响。方法用随机数字表法选取北京电力医院2009年1月至2015年5月乳腺浸润性导管癌和乳腺纤维腺瘤术后标本各50例,采用免疫组织化学EnVision法检测ARID1A的表达差异。分析ARID1A的表达与乳腺癌患者临床病理参数的关系。以电话及病案查阅的方式了解乳腺癌患者的预后情况,随访日期截止至2015年12月31日,计算出DFS和OS。ARID1A、HER-2、ER和P53间的相关性研究采用非参数Spearman检验。采用Kaplan-Meier法进行生存分析,Log-rank检验进行生存率比较,同时Cox比例风险回归模型解析影响预后的多种因素。结果ARID1A在50例乳腺浸润性导管癌中高表达率为46%(23/50),在50例纤维腺瘤中高表达率为72%(36/50),差异有统计学意义(字2=6.986,P=0.008)。在不同母乳喂养时间、组织学分级、P53表达、ER表达、PR表达、分子分型、HER-2表达的患者中ARID1A的表达差异亦有统计学意义(字2=4.177、4.726、4.177、5.469、14.661、5.469,P均<0.050)。ARID1A的表达与P53的表达呈正相关(r=0.289,P=0.042),与ER的表达呈正相关(r=0.331,P=0.019),而与HER-2的表达呈负相关(r=-0.372,P=0.008)。TNM玉、域期患者的3年DFS高于TNM芋、郁期(85.5%比42.4%,字2=9.657,P=0.002),肿瘤直径臆2cm的患者的5年OS高于肿瘤直径>2cm的患者(100%比71.3%,字2=5.598,P=0.018),P53阳性组的3年DFS和5年OS均高于P53阴性组(86.3%比45.5%,字2=5.077,P=0.024;95.2%比50.8%,字2=9.476,P=0.002),ARID1A高表达患者的3年DFS和5年OS均高于ARID1A低表达患者(81.8%比70.2%,字2=0.454,P=0.500;90.9%比76.1%,字2=0.090,P=0.765)。多变量Cox回归模型可见P53阳性表达为影响乳腺癌患者OS的独立保护性因素(P=0.028,OR=0.079,95%CI:0.008~0.761),ARID1A的表达不是影响OS的独立预后因素(P=0.194,OR=4.758,95%CI:0.453~49.957)。结论摇ARID1A在乳腺癌组织中的表达低于乳腺纤维腺瘤,可能对乳腺癌的发生发展起重要作用。

Objective To detect the expression difference of AT-rich interactive domain-containing protein1A(ARID1A)in malignant and benign breast tumor,and its impact on patients‘prognosis.Methods The specimens from50cases of breast invasive ductal carcinoma and50cases of breast fibroma were collected in Beijing Electric Power Hospital between January2009and May2015by random number table.The expression of ARID1A protein was detected by immunohistochemical EnVision method.The relationship between ARID1A expression and clinicopathological features was analyzed.The50breast cancer patients were followed up until December31,2015by phone or reviewing case history.DFS and OS were calculated.The correlations of ARID1A with HER-2,ER and P53were analyzed by nonparametric Spearman test.Kaplan Meier method was used for survival analysis and Log-rank test was used to compare the survival rate.Cox proportional hazards regression model was used to analyze the prognostic factors.