摘要
目的探讨乳腺癌动态增强MRI的影像学征象与细胞分子表型之间的相关性。方法回顾性分析85例经手术病理确诊的乳腺癌患者术前MRI资料,采用统计学方法分析MRI影像学征象与乳腺癌分子表型之间的相关性。免疫组织化学检测采用二步法观察癌组织中雌激素受体(ER)、孕激素受体(PR)和人类表皮生长因子受体2(Cerb B-2)的表达水平。结果动态增强MRI诊断乳腺癌病理类型的敏感性和特异性分别为94.0%和80.0%,诊断其分化程度的敏感性和特异性分别94.8%和71.4%。肿瘤的形态和边缘与其病理类型和组织学分级相关(χ2值分别为6.21和5.88,P均<0.05),肿瘤形态和环形强化均与ER和PR的阴性表达相关(P<0.05),环形强化和肿瘤内部坏死与Cerb B-2的阳性表达相关(P<0.05),而肿瘤边缘与ER、PR和Cerb B-2的表达均无相关性(P>0.05)。肿瘤不规则形态可以预测ER的阴性表达,而环形强化是ER、PR阴性预测因子(P<0.05),肿瘤内部坏死是Cerb B-2阳性表达的预测因子(P<0.05)。结论乳腺癌MRI的影像学特征对判断肿瘤类型和组织学分级具有一定的指导意义,可预测乳腺癌的分子表型表达情况,评估预后。
Objective To investigate the correlation between MRI imaging characteristics and Cell molecular phenotype of breast cancers. Methods The preoperative MRI materials of 85 patients with pathologically-proved breast cancer were retrospectively analyzed. The correlation between MRI imaging characteristics and cell molecular phenotype was analyzed by using statistical analysis. ER, PR and CerbB-2 expressions in breast cancer tissue were determined with immunohistochem- ical method. Results The sensitivity and specificity of dynamic MRI in judging the pathological types of breast cancer were 94.0% and 80.0% respectively, and in judging the differentiation degree of the breast cancer were 94.8% and 71.4% respectively. The morphology and the border of the breast cancer demonstrated on MR scans carried a close rela- tionship with the pathological type and histological classification （ X2 = 6.21 and X2 = 5.88 respectively, P 〈 0.05 ）. The morphology and circular enhancement were well related to the negative expression of ER and PR （ P 〈 0.05 ） , while the cir- cular enhancement bore a definite relationship with the necrosis within tumor as well as with the positive expression of CerbB-2 （P 〈 0.05 ）. No statistically significant correlation existed between the shape of tumor＇ s margin and the expres- sions of ER and PR and CerbB-2 （ P 〉 0.05 ）. The irregular shape of the tumor could predict the negative expression of ER, while the circular enhancement was a reliable predictor for the negative expression of ER and PR （ P 〈 0.05 ） and the necrosis within tumor was also a reliable predictor for the positive expression of CerbB-2 （ P 〈 0.05 ）. Conclusion Breast cancers have certain MRI features, which are very helpful in judging the pathological type and histological classifica- tion, and also in predicting the molecular phenotypic expression as well as in evaluating the prognosis.
出处
《临床放射学杂志》
CSCD
北大核心
2015年第1期36-40,共5页
Journal of Clinical Radiology
关键词
乳腺癌
磁共振成像
病理学
分子生物学
Breast cancer Magnetic resonance imaging Pathology Molecular biology
