The majority of breast cancers are diagnosed as breast cancer of no specific type (or ductal carcinoma), a heterogeneous category of tumors that includes curable neoplasms as well as highly aggressive cancers. This classification is inadequate for clinical management of breast cancer patients as well as for designing laboratory and clinical research protocols. Although there are clearly a number of different tumor phenotypes within this diagnostic category, traditional morphology and even current molecular markers of prognosis are incapable of clearly discriminating different subtypes of ductal cancer from one another. In this project, we propose to distinguish biologically different groups of ductal breast cancer from one another using gene expression profiles. This is based on the concept that gene expression profiles represent an objective and quantitative measure of a neoplasm's phenotype. We have in preliminary experiments identified a number of genes that are differentially expressed among different ductal cancers, consistent with our hypothesis that will use a custom breast cancer array that we developed to represent the genes most capable of differentiating breast cancers, and then use this array to analyze gene expression profiles in microdissected breast cancer samples. Groups of tumors that share expression profiles identified through cluster analysis of data will then be examined for pathological and clinical similarities to develop more specific hypothesis of ductal cancer classification. Finally, we will design case-control studies to test and refine these classification hypotheses. Through this stepwise development and testing of classification hypotheses, we expect to identify biologically distinctive categories of breast cancer. A sound molecular classification of this nature will greatly benefit clinical management of breast cancer patients and will provide a useful reference for the design of breast cancer research.
Showing the most recent 10 out of 282 publications
