甲状腺恶性肿瘤的疾病多基因诊断及范围检测

Diagnostic and extent of disease multigene assay for malignant thyroid neoplasms.

作者信息

Kebebew Electron, Peng Miao, Reiff Emily, McMillan Alex

机构信息

Endocrine Surgery and Oncology Program, San Francisco Comprehensive Cancer Center, University of California-San Francisco, San Francisco, California 94143-1674, USA.

出版信息

Cancer. 2006 Jun 15;106(12):2592-7. doi: 10.1002/cncr.21922.

DOI:10.1002/cncr.21922

PMID:16688775

Abstract

BACKGROUND

Approximately 30% of fine-needle aspiration (FNA) biopsies of thyroid nodules are indeterminate, nondiagnostic, or suspicious. The purpose of the current study was to determine the accuracy of novel candidate diagnostic markers to distinguish benign from malignant thyroid neoplasms, and to predict the extent of disease.

METHODS

A real-time quantitative reverse-transcriptase polymerase chain reaction (RT-PCR) assay of 6 novel candidate diagnostic and extent of disease marker genes (extracellular matrix protein 1 [ECM1]; transmembrane protease, serine 4 [TMPRSS4]; angiopoietin 2 [ANGPT2]; TIMP metallopeptidase inhibitor 1 [TIMP1]; ephrin-B2 [EFNB2], and epidermal growth factor receptor [EGFR]) was used in 126 thyroid tissues. To evaluate the performance of the scoring model for the diagnostic markers in combination, the area under the receiver operating characteristic (ROC) curve (AUC) was determined.

RESULTS

The levels of ECM1, TMPRSS4, ANGPT2, and TIMP1 mRNA expression were found to be independent diagnostic markers of malignant thyroid neoplasms. The AUC for the 4 diagnostic genes in combination was 0.993 with a sensitivity of 100%, a specificity of 94.6%, a positive predictive value of 96.5%, and a negative predictive value of 100%. In 31 thyroid nodule FNA biopsy samples, the scoring model had a sensitivity of 91.0%, a specificity of 95.0%, a positive predictive value of 92.9%, and a negative predictive value of 92.3%. The multigene assay correctly classified 93% of tumors into the correct risk group (low-risk vs. high-risk) with a sensitivity of 78.9% (true positive in high-risk tumors), specificity of 92% (true negative in low-risk tumors), positive predictive value of 87.5%, and negative predictive value of 92%. In 11 malignant thyroid nodule FNA samples, the extent of disease scoring model correctly identified 3 of 4 high-risk differentiated thyroid cancers and 7 of 7 low-risk differentiated thyroid cancers.

CONCLUSIONS

This novel multigene assay is an excellent diagnostic and extent of disease marker for differentiated thyroid cancer and would be a helpful adjunct to FNA biopsy of thyroid nodules.

摘要

背景

甲状腺结节细针穿刺（FNA）活检中约30%的结果为不确定、无法诊断或可疑。本研究的目的是确定新型候选诊断标志物区分甲状腺良性和恶性肿瘤以及预测疾病范围的准确性。

方法

对126份甲状腺组织进行6种新型候选诊断和疾病范围标志物基因（细胞外基质蛋白1[ECM1]；跨膜蛋白酶，丝氨酸4[TMPRSS4]；血管生成素2[ANGPT2]；金属蛋白酶组织抑制因子1[TIMP1]；ephrin-B2[EFNB2]和表皮生长因子受体[EGFR]）的实时定量逆转录聚合酶链反应（RT-PCR）检测。为评估诊断标志物联合评分模型的性能，确定了受试者操作特征（ROC）曲线下面积（AUC）。

结果

发现ECM1、TMPRSS4、ANGPT2和TIMP1 mRNA表达水平是甲状腺恶性肿瘤的独立诊断标志物。4种诊断基因联合的AUC为0.993，敏感性为100%，特异性为94.6%，阳性预测值为96.5%，阴性预测值为100%。在31份甲状腺结节FNA活检样本中，评分模型的敏感性为91.0%，特异性为95.0%，阳性预测值为92.9%，阴性预测值为92.3%。多基因检测将93%的肿瘤正确分类到正确的风险组（低风险与高风险），敏感性为78.9%（高风险肿瘤中的真阳性），特异性为92%（低风险肿瘤中的真阴性），阳性预测值为87.5%，阴性预测值为92%。在11份甲状腺恶性结节FNA样本中，疾病范围评分模型正确识别出4例高风险分化型甲状腺癌中的3例和7例低风险分化型甲状腺癌中的7例。