EXPRESSION BASED MARKERS FOR BREAST CANCER DETECTION

该提案将尝试确定基于表达的乳腺癌标志物，并开发使用这些标志物检测循环乳腺癌细胞的检测方法。我们的应用程序是由杜克乳腺癌项目和雅培诊断乳腺癌风险集团合作开发的，并将利用每个机构的相当大的优势。标记识别的基本方法利用了正在进行的（和单独资助的）使用强大的新基因组学工具的努力。在杜克大学，SAGE（基因表达序列分析）乳腺癌和正常乳腺上皮的文库正在构建中，作为CGAP的一部分。同样在杜克大学，一系列20例原发性乳腺癌将使用高密度Affymetrix表达芯片进行分析（将取样3万个基因/ESTs）。最后，在雅培诊断公司，已经使用Incyte公司的表达文库鉴定了标记物。这三种表达数据的来源将通过电子方式进行比较，并确定在组织或癌症特定方式中高水平表达的最有希望的标记。在计算机分析之后，先导标记将通过验证算法，包括northern blotting、多组织RNA点blots、RT-PCR和原位细胞杂交。一组乳腺细胞系和一系列50例原发性乳腺癌将被用来确定这些标记物在靶组织中的普遍表达程度。在这些初步验证步骤之后，将进行一系列重构实验，通过将已知数量的乳腺癌细胞（来自细胞系和恶性积液）注入正常全血中。上皮细胞将使用磁珠/抗体技术纯化，纯化后的样品将通过流式细胞术、组织学和定量RT-PCR进行分析。为了提高灵敏度，一个单独的平行筛查将集中在乳腺癌中高度过度表达的标记物。PCR多路复用方法将在此背景下进行探索，还将包括使用端粒酶基因（TERT）表达作为癌症特异性标志物与其他主要标志物相结合。针对最有希望的候选药物的抗体将被开发出来，特别是针对可能分泌或呈递在细胞表面的基因产物。最后，将从杜克大学新诊断为乳腺癌的妇女身上采集全血和血清样本。该银行将构成已被证明是最稳健的标记和方法的主要验证集。

This proposal will attempt to identify expression-based markers for breast cancer and develop assays using these markers to detect circulating breast cancer cells. Our application has been developed as a collaborative effort between the Duke Breast Cancer Program and Abbott Diagnostic Breast Cancer Venture Group and will utilize the considerable strengths of each institution. The basic approach for marker identification takes advantage of ongoing (and separately funded) efforts using powerful new genomics tools. At Duke, SAGE (serial analysis of gene expression) libraries of breast cancer and normal breast epithelium are being constructed as part of CGAP. Also at Duke, a series of 20 primary breast cancers will be analyzed using high-density Affymetrix expression chips (30,000 genes/ESTs will be sampled). Finally, at Abbott Diagnostics, markers have already been identified by using the Incyte Corporation expression libraries. These three sources of expression data will be compared electronically and the most promising markers that are expressed at high levels in a tissue or cancer specific manner will be identified. After the in silico analysis, the lead markers will be put through a validation algorithm that will include northern blotting, multiple tissue RNA dot blots, RT-PCR, and in situ cytohybridization. A panel of breast cell lines and a series of 50 primary breast cancers will be used to determine how commonly expressed these markers are in the target tissue. After these initial validation steps, a series of reconstitution experiments will be performed by spiking normal whole blood with known numbers of breast cancer cells, derived from cell lines and malignant effusions. Epithelial cells will be purified using magnetic bead/antibody technology and the resulting purified samples will be analyzed by flow cytometry, histology, and quantitative RT-PCR. Towards increasing sensitivity, a separate and parallel screen will focus on markers that are highly overexpressed in a subset of breast cancers. PCR multiplexing approaches will be explored in this context and will also include using telomerase gene (TERT) expression as a cancer specific marker in conjunction with other lead markers. Antibodies to the most promising candidates will be developed, particularly for gene products that may be secreted or presented on the cell surface. Finally, a collection of whole blood and serum samples will be obtained from women newly diagnosed with breast cancer at Duke. This bank will constitute a primary validation set for markers and approaches that have proven to be the most robust.