The goal of this project is to identify the molecular alterations that occur in the progression from ductal carcinoma in situ (DCIS) to invasive breast cancer. This is of critical importance because it is only after the milestone of tissue invasion has been achieved that a cancer has the potential to metastasize and kill a patient. Invasive breast cancers harbor multiple molecular alterations that were sequentially acquired during progression from ADH to ductal carcinoma in situ (DCIS) to invasive breast cancer. We will identify the genetic alterations that may be specific to DCIS or may be cumulative with increasing malignant potential. 7Rationale: While the cure rate for ductal carcinoma in situ (DCIS) is over 90%, this comes at the cost of considerable treatment-related morbidity. We urgently need the prognostic markers that would identify key genetic alterations predictive of the long-term outcome of DCIS, since most DCIS never become invasive carcinomas, presumably because they lack critical genetic changes, the replicative lifespan, or simply the time necessary to do so. Our experimental design exploits recent technological developments that permit genome- wide testing of DNA samples for genetic and epigenetic alterations, even in the very limited material available from archival DCIS samples. DCIS may represent a stage at which various pathways of clonal evolution are attempted, until a particular combination of genetic and/or epigenetic changes allows tissue invasion. Indications are that genetic mutations and epigenetic silencing may both be operative in carcinogenesis, and that these alternatives may be used interchangeably, even in a complementary fashion in the same cancer cell. Therefore, a combined analysis of both types of alterations is necessary. 7Design: This study is designed as multicenter, longitudinal, case-control study of DCIS. We have identified unique cohorts of pure DCIS cases with known outcomes to enable a multicenter genome-wide screening study of 100 cases and 100 controls, which will be initially limited to patients of Caucasian origin in order to obtain a homogenous patient population and maximize the probability of successfully validating our hypothesis that prognostic markers exist in DCIS. Candidate markers of interest will be further characterized by quantitative PCR methods, with the goal of creating a single, prognostic multiplex PCR signature that can be used on the limited clinical material typically obtained in the diagnostic phase of the evaluation of DCIS. We will then validate the assay in an independent test cohort of 100 cases of DCIS with progressive disease vs. 100 cases of DCIS that remain disease free, obtained from the population-based repository provided by the Iowa SEER residual tissue repository (RTR). We will also use an additional population-based repository located in Hawaii, to test if the prognostic signature developed from Caucasian samples can be used successfully in a different ethnic group. This will guide future work, particularly regarding the necessity of exploring the predictive power of the multiplex assay in different populations.