Department of Pathology, Stavanger University Hospital, Stavanger, Norway.
Mod Pathol. 2010 Dec;23(12):1567-76. doi: 10.1038/modpathol.2010.177. Epub 2010 Sep 3.
Breast cancer is a heterogeneous disease. Different subgroups can be recognized on the basis of the steroid receptors, HER-2, cytokeratin expression and proliferation patterns. As a result of mRNA-profiling studies, five major groups can be recognized, of which the triple-negative and basal-like tumors have the worst prognosis. Many of these tumors have a high proliferation that has the strongest prognostic value in node negative breast cancer. In the current study we analyzed the microRNA pattern in 103 lymph node negative breast cancers and compared these profiles with different biological characteristics and clinicopathological features. Unsupervised hierarchical cluster analysis divides the patients into four main groups, of which the basal-like/triple-negative group is the most prominent (11% of all cases), the luminal A cancers containing the Her2 negative and estrogen receptor/progesterone receptor-positive tumors is the largest group (57%), and the group of luminal B (32%) is more heterogeneous and contains the Her2 positive/estrogen receptor-negative patients as well. The highest overall classification values by analysis of variance followed by cross validation (leave one sample out and reselect genes) were found for cytokeratin 5 and 6, triple-negative and estrogen receptor, with 97, 90 and 90% accuracy, respectively. MiR-106b gene is prominent in all of these signatures and correlates strongest with high proliferation. Other interesting observations are the presence of several microRNAs (miR532-5p, miR-500, miR362-5p, and miR502-3p) located at Xp11.23 in cancers with a triple-negative signature, and the upregulation of several miR-17 cluster members in estrogen receptor-negative tumors. The current study shows that estrogen receptor negativity and cytokeratin 5 and 6 expression are important, and specific biological processes in lymph node negative breast cancer, as microRNA signatures are strongest in these subgroups.
乳腺癌是一种异质性疾病。根据甾体受体、HER-2、细胞角蛋白表达和增殖模式，可以识别不同的亚组。由于 mRNA 谱研究，可识别出五个主要亚组，其中三阴性和基底样肿瘤的预后最差。这些肿瘤中有许多具有较高的增殖性，在淋巴结阴性乳腺癌中具有最强的预后价值。在本研究中，我们分析了 103 例淋巴结阴性乳腺癌的 microRNA 图谱，并将这些图谱与不同的生物学特征和临床病理特征进行了比较。无监督层次聚类分析将患者分为四个主要亚组，其中基底样/三阴性组最为突出（占所有病例的 11%），包含 Her2 阴性和雌激素受体/孕激素受体阳性肿瘤的 luminal A 癌是最大的组（57%），而 luminal B 组（32%）更为异质，包含 Her2 阳性/雌激素受体阴性的患者。通过方差分析和交叉验证（留一个样本并重新选择基因）进行的总体分类值最高的是细胞角蛋白 5 和 6、三阴性和雌激素受体，准确率分别为 97%、90%和 90%。miR-106b 基因在所有这些特征中都很突出，与高增殖性相关性最强。其他有趣的观察结果是，在具有三阴性特征的癌症中，存在几个位于 Xp11.23 上的 microRNAs（miR532-5p、miR-500、miR362-5p 和 miR502-3p），以及雌激素受体阴性肿瘤中 miR-17 簇成员的上调。本研究表明，雌激素受体阴性和细胞角蛋白 5 和 6 的表达是重要的，并且在淋巴结阴性乳腺癌中存在特定的生物学过程，因为 microRNA 特征在这些亚组中最强。
