Ductal carcinoma in situ (DCIS) is a preinvasive lesion of the breast that makes up almost 30% of all mammographically detected malignancies [1,2]. In DCIS, as for many pre-malignant lesions, the main clinical challenge is predicting which lesions are likely to progress to invasive and metastatic disease. In the absence of reliable prognostic tools, all DCIS is treated as if it would progress, resulting in harm for those patients whose DCIS lacks potential for progression. Thus accurate and clinically actionable prognostic markers for DCIS are critically needed. Neoplasms progress through a process of random mutations and clonal expansions, leading to widespread heterogeneity both between and within neoplasms  that makes it challenging to predict prognosis on the basis of specific markers. Moreover, the clonal diversity within a neoplasm is overlaid upon a background of heterogeneous microenvironments, which can accelerate the evolutionary process [5,6] by imposing different selective pressures on different meta-populations in different tumor regions . Genetic diversity in a population is the fuel for natural selection and is a key determinant of the rate of evolution [8,9]. The more genetic and microenvironmental diversity, the more opportunities for selection to drive clonal expansions and for the neoplasm to adapt to new selective pressures. Our solution to the problem of predicting progression given its stochastic nature is not to measure the products of somatic evolution (e.g., presence/absence of a mutation), but to measure the process of somatic evolution (e.g., genetic diversity).
The increased diagnosis and treatment of DCIS has had an unclear impact on breast cancer specific survival;thus, there has been growing concern that the current management of pre-invasive cancers results in more harm than benefit for patients. In this project, we propose to collect genomic, phenotypic, and radiographic measures of tumor cell diversity in DCIS and the tumor microenvironment, and test whether these diversity measures can identify which patients are most likely to develop metastatic disease. Deliverables from the proposal have high potential for rapid integration into clinical trials for active surveillance of DCIS and in addition, could have universal relevance for management of other solid tumors.