Objective: To assess the value of cyclin D1 immunocytochemistry combined with a small panel molecular analysis in indeterminate cytological diagnosis of Bethesda category Ⅲ-Ⅴ. Methods: A consecutive cohort of 96 thyroid FNA specimens with indeterminate diagnosis (TBSRTC category Ⅲ-Ⅴ) and available histopathologic follow-up data were collected between December 2018 and December 2021 in Department of Pathology, Beijing Hospital. The cases were evaluated by cyclin D1 immunocytochemistry and molecular testing of BRAFV600E or a small panel of markers (BRAF, N-RAS, H-RAS, K-RAS and TERT) in the FNA specimens. The identification of the optimal cut-off point of cyclin D1 for the diagnosis of malignancy was evaluated using the receiver operating characteristic (ROC) curves and the assessment of the area under the ROC curve (AUC). The specificity, sensitivity, positive predictive value (PPV) and negative predictive value (NPV) of all these markers were evaluated with the crosstabs and significance was calculated. Results: Ninty-six patients with 96 thyroid nodules were enrolled, including 42 cases of TBSRTC-III, 10 cases of TBSRTC-IV and 44 cases of TBSRTC-V. There were 79 females and 17 males with a median age of 47 years (range, 25 to 75 years). A 7.5% cut-off value for positive cyclin D1 nuclear immunostaining in thyroid cells demonstrated 100% PPV, 57.1% NPV, 81.0% sensitivity and 100% specificity for thyroid malignancy diagnosis. The sensitivity of the BRAFV600E mutation test or combined with a small panel test alone for thyroid malignancy diagnosis were 65.5% and 69.0% respectively. The sensitivity for thyroid malignancy diagnosis increased to 94.0% and 95.2% respectively when combining the cyclin D1 immunocytochemistry with the molecular test, and the specificities remained 100% and 91.7% respectively.The accuracy of cyclin D1 immunocytochemistry combined with a small panel of molecular test in detecting thyroid malignancy increased to 94.8% compared to using these markers alone. Conclusions: The addition of cyclin D1 immunocytochemistry and a small panel of molecular testing to FNA cytology can increase the sensitivity and NPV of cytology in indeterminate categories, and this supplementary approach provides a simple, accurate and convenient diagnostic method for reducing unnecessary thyroidectomies.
目的： 评估细胞周期蛋白D1（cyclin D1）免疫细胞化学染色联合小组合的分子检测，在Bethesda分类（TBSRTC）不确定类别（Ⅲ~Ⅴ级）细胞学诊断中的价值。 方法： 收集2018年12月至2021年12月北京医院病理科96例有组织病理学随访结果、术前穿刺细胞学诊断为TBSRTC Ⅲ、Ⅳ及Ⅴ级的病例。细胞学涂片经退染cyclin D1免疫细胞化学染色，及刮取涂片行BRAFV600E单基因或5基因（BRAF、N-RAS、H-RAS、K-RAS和TERT）检测。利用受试者工作特征曲线（ROC）和ROC曲线下面积（AUC）评估cyclin D1诊断甲状腺恶性肿瘤的最佳临界值。通过交叉列联表评价各指标的特异度、灵敏度、阳性预测值（PPV）和阴性预测值（NPV），并评价其诊断的准确性。 结果： 入组的96例甲状腺结节中，细胞学诊断TBSRTC Ⅲ级42例，Ⅳ级10例，Ⅴ级44例。其中女性79例，男性17例，中位年龄47岁（范围25~75岁）。甲状腺细胞cyclin D1核染色阳性阈值设定为7.5%，其诊断甲状腺恶性肿瘤的PPV为100%，NPV为57.1%，灵敏度为81.0%，特异度为100%。BRAFV600E突变检测或联合5基因检测诊断甲状腺恶性肿瘤的灵敏度分别为65.5%和69.0%，而联合cyclin D1免疫细胞化学与上述分子检测，其诊断甲状腺恶性肿瘤的灵敏度提高到94.0%和95.2%，差异具有统计学意义（P<0.01）。但特异度并未降低，仍为100%和91.7%。与单独使用这些标志物相比，cyclin D1免疫细胞化学联合单基因或5基因检测，辅助FNA诊断甲状腺恶性肿瘤的准确性提高至94.8%。 结论： 对FNA细胞学诊断不确定的病例，加上cyclin D1免疫细胞化学和小组合的分子检测，可以提高细胞学在诊断甲状腺恶性肿瘤中的灵敏度和阴性预测值，为细胞病理学医师提供一个简单、准确和方便的诊断方法，同时减少了外科不必要的甲状腺切除手术。.