TECHNICAL ABSTRACT Ductal carcinoma in situ (DCIS) is one of the most common human precancers, diagnosed in over 50,000 women in the US annually. Despite the highly variable risk of progression to invasion, 98% of DCIS is currently treated with surgical excision, often combined with radiation, to prevent invasive cancer. For those patients with high risk of progression, aggressive treatment is warranted. However, for DCIS at low risk of progression to invasive cancer, there may be little or no clinical benefit to surgery. Since current guidelines recommend that all patients with DCIS have surgery, virtually all prior analyses, including our own, have been limited to patients who had surgical excision, with recurrence of disease serving as an imperfect proxy for invasive progression. Epidemiologic evidence suggests that only 20-30% of DCIS progress. Thus, some patients may derive little benefit from surgery, and therefore could be managed with active surveillance (i.e. close monitoring without surgery), similar to an approach now broadly accepted for early-stage prostate cancer. However, there are currently NO predictors to determine which patients may safely avoid surgery. Recently, four international clinical trials have been launched to evaluate the feasibility of active surveillance for DCIS, and both data and clinical samples of patients undergoing active surveillance are now starting to emerge. Here, we propose to determine the biologic behavior of DCIS during the course of active surveillance, with the goal of eliminating the need for surgery in patients with the lowest biologic risk for invasive progression. We will investigate alterations in both DCIS tumors and their cell neighborhoods to define the molecular and cellular events conferring the highest risk of invasion in the absence of surgical excision. Our Aims are the following: Aim 1: Assess the molecular variation of DCIS with respect to the biologic pathways and evolutionary dynamics that distinguish progressors from non-progressors over time in a unique cohort of patients with DCIS undergoing active surveillance with repeated sampling. Aim 2: Identify the molecular features at single cell resolution that characterize a pro-tumorigenic cellular neighborhood in the transition between DCIS and invasive cancer. Aim 3: Construct and validate a molecular predictor of invasive progression based on two multicenter clinical trials of active surveillance for DCIS. We will leverage our team’s clinical, genomic, spatial imaging, and computational expertise combined with access to unique clinical trial samples. The deliverables from this project will be timely and relevant to inform near-term implementation of active surveillance strategies for low-risk breast precancers.