School of Dentistry and Dental Research Institute, University of California Los Angeles, Los Angeles, California, United States of America.
PLoS One. 2010 Dec 31;5(12):e15573. doi: 10.1371/journal.pone.0015573.
A sensitive assay to identify biomarkers using non-invasively collected clinical specimens is ideal for breast cancer detection. While there are other studies showing disease biomarkers in saliva for breast cancer, our study tests the hypothesis that there are breast cancer discriminatory biomarkers in saliva using de novo discovery and validation approaches. This is the first study of this kind and no other study has engaged a de novo biomarker discovery approach in saliva for breast cancer detection. In this study, a case-control discovery and independent preclinical validations were conducted to evaluate the performance and translational utilities of salivary transcriptomic and proteomic biomarkers for breast cancer detection.
METHODOLOGY/PRINCIPAL FINDINGS: Salivary transcriptomes and proteomes of 10 breast cancer patients and 10 matched controls were profiled using Affymetrix HG-U133-Plus-2.0 Array and two-dimensional difference gel electrophoresis (2D-DIGE), respectively. Preclinical validations were performed to evaluate the discovered biomarkers in an independent sample cohort of 30 breast cancer patients and 63 controls using RT-qPCR (transcriptomic biomarkers) and quantitative protein immunoblot (proteomic biomarkers). Transcriptomic and proteomic profiling revealed significant variations in salivary molecular biomarkers between breast cancer patients and matched controls. Eight mRNA biomarkers and one protein biomarker, which were not affected by the confounding factors, were pre-validated, yielding an accuracy of 92% (83% sensitive, 97% specific) on the preclinical validation sample set.
Our findings support that transcriptomic and proteomic signatures in saliva can serve as biomarkers for the non-invasive detection of breast cancer. The salivary biomarkers possess discriminatory power for the detection of breast cancer, with high specificity and sensitivity, which paves the way for prediction model validation study followed by pivotal clinical validation.
使用非侵入性采集的临床标本来鉴定生物标志物的灵敏检测方法是乳腺癌检测的理想方法。虽然有其他研究显示唾液中存在乳腺癌的疾病生物标志物，但我们的研究通过全新发现和验证方法来检验唾液中存在乳腺癌鉴别生物标志物的假设。这是此类研究中的首例，尚无其他研究在唾液中采用全新的生物标志物发现方法来检测乳腺癌。在这项研究中，进行了病例对照发现和独立的临床前验证，以评估唾液转录组和蛋白质组生物标志物用于乳腺癌检测的性能和转化实用性。
方法/主要发现：使用 Affymetrix HG-U133-Plus-2.0 阵列和二维差异凝胶电泳 (2D-DIGE) 分别对 10 名乳腺癌患者和 10 名匹配对照者的唾液转录组和蛋白质组进行了分析。通过 RT-qPCR（转录组生物标志物）和定量蛋白质免疫印迹（蛋白质组生物标志物），在 30 名乳腺癌患者和 63 名对照者的独立样本队列中对发现的生物标志物进行了临床前验证。转录组和蛋白质组分析揭示了乳腺癌患者和匹配对照者之间唾液分子生物标志物的显著差异。8 个 mRNA 生物标志物和 1 个蛋白质生物标志物不受混杂因素的影响，在临床前验证样本集中，其准确率为 92%（83%的敏感性，97%的特异性）。
我们的研究结果支持唾液中的转录组和蛋白质组特征可以作为乳腺癌非侵入性检测的生物标志物。唾液生物标志物具有用于乳腺癌检测的鉴别能力，具有高特异性和敏感性，为后续的预测模型验证研究和关键的临床验证铺平了道路。
