目的：观察分化型甲状腺癌（differentiated thyroid cancer，DTC）和良性甲状腺疾病的不同蛋白分子表型，从而确定能准确诊断甲状腺癌的分子标志物。方法：采用免疫组织化学的方法，对含有100个良性甲状腺病变和99个恶性DTC的组织芯片样本进行染色，检测了57种可以做为分子标志的蛋白。使用列联表和Mann-Whitney U（MU）检验来确定分子标志染色与肿瘤病理特征（DTC和良性疾病）之间的相关性，同时使用随机森林分类器算法来确定有用/重要的分子标志。结果：在多重检验校正后，发现57个诊断标志物中35个与DTC有显著相关性。与良性甲状腺癌疾病相比，在DTC中有8个标志物表达下调而有27个标志物上调(P<0.05)。在DTC诊断中最显著的标志物为Galectin-3、细胞角蛋白 19、血管表皮生长因子、雄激素受体、p16、Aurora-A与HBME-1。使用DTC分子标志芯片，通过随机森林分类算法对肿瘤良恶性进行诊断的预计灵敏度为879%、特异性为94.0%、准确性为91.0%。结论：研究甲状腺癌的分子表型在甲状腺疾病中的诊断具有重要的临床意义。

Objective:To identify molecular markers for accurate diagnosis of thyroid cancer the molecular phenotypes of varied proteiumark were evaluated in differentiated thyroid cancer (DTC) and benign thyroid lesions. Methods：Tissue microarrays containing 100 benign and 99 malignant thyroid lesions were stained for a panel of 57 molecular markers. Correlation between the marker staining and tumor pathology (DTC versus benign tumor) were determined using contingency table and Mann-Whitney U (MU) tests. A Random Forests classifier algorithm was also utilized to identify meaningful important molecular classifiers. Results： Of the 57 diagnostic markers evaluated, 35 (61%) were significantly associated with a DTC diagnosis after multiple testing correction. Among them, 8 markers were downregulated and 27 ones upregulated in DTC, when compared with benign thyroid tumor. The most significant markers for DTC diagnosis were Galectin-3, Cytokeratin 19, Vascular Endothelial Growth Factor, Androgen Receptor, p16, Aurora-A, and HBME-1. Using the entire molecular marker panel, a Random Forests algorithm was able to classify a neoplasm as DTC or benign tumor with an estimated sensitivity of 87.9%, specificity of 94.0%, and accuracy of 91.0%. Conclusion：Evaluating the DTC and benign thyroid tumor molecular phenotypes has allowed us to identify a diagnostic marker panel, which may help differentiate DTC with benign tumors and improve patient selection for thyroid surgery.