Abstract PD5-07: Mapping the tumor and microenvironmental evolution underlying DCIS progression throughmultiplexed ion beam imaging

Abstract

Abstract Ductal Carcinoma in Situ (DCIS) is a pre-invasive lesion that accounts for nearly 20% of new breast cancer diagnoses. Of these cases, about half will progress to invasive breast cancer (IBC) within ten years. However, because diagnostic criteria for delineating low risk lesions from those likely to progress to IBC have not been identified, many patients are receiving unnecessary chemotherapy and surgery that can result in therapy-related morbidity and death. With this in mind, we used Multiplexed Ion Beam Imaging by time of flight (MIBI-TOF) and RNA-seq laser-capture microdissection (SMART-3SEQ) to construct a comprehensive spatial atlas describing the structure, function, and cellular composition of DCIS. MIBI-TOF and SMART-3SEQ were used to compare lesions from patients that later developed IBC with those from age- and history matched DCIS controls without recurrence. Using a 37-marker staining panel to interrogate 137 lesions, we identified 17 distinct cell populations of epithelial, stromal, and immune cells that were arranged in recurrent cellular microenvironments specific for DCIS or invasive disease. We observe a coordinated shift in the immune and stromal compartments as invasive disease arises, including a robust influx of lymphocytes and expansion reactive stromal phenotypes in synchronous DCIS + IBC, which was distinct from the macrophage-dominant, highly vascularized microenvironment of recurrent IBC. Further, single cell segmentation using a deep learning model was combined with pixel-level coexpression analysis to extensively evaluate how the thickness, continuity, and phenotype of ductal myoepithelium changes as tumors progress from a pre-invasive state. These data were incorporated into a comprehensive model which was subsequently used to identify a subset of features that correlate with disease-free survival following DCIS tumor resection. Taken together, these features represent important prognostic metrics that can be used to separate pre-invasive from indolent DCIS tumors, and allow for tailored therapy that improves patient outcomes and quality of life in this disease. Citation Format: Michael Angelo. Mapping the tumor and microenvironmental evolution underlying DCIS progression throughmultiplexed ion beam imaging [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD5-07.