9218 Highly Multiplexed Mapping of the Ductal Carcinoma In Situ Ecosystem to Predict Disease Outcome

Abstract Disclosure: G.M. Wilson: None. T.B. Doan: None. B.J. Guild: None. N. Pathmanathan: None. J.D. Graham: None. Mammographic screening programs have resulted in remarkable improvements in breast cancer survival due to early-stage detection. However, this has resulted in a dramatic increase in the detection of non-invasive ductal carcinoma in situ (DCIS), a non-obligate precursor of invasive ductal carcinoma (IDC), which is confined to the ducts and surrounded by a normal myoepithelial cell layer. Many DCIS cases will not progress to IDC within the lifetime of the patient. However, there are limited tools to accurately predict progression, thus virtually all DCIS is surgically removed, and most patients receive adjuvant radiotherapy as it halves recurrence rates. This likely represents overtreatment for a majority of patients. Emerging evidence has revealed that the adjacent microenvironment becomes progressively altered in DCIS and IDC compared to normal tissue, and this is correlated with disease outcome. The objective of this study was to investigate changes to the DCIS microenvironment during disease progression and identify disease states that predict patient outcome. We used immunohistochemistry and imaging mass cytometry with a well-annotated cohort of DCIS, with or without adjacent IDC, to measure a panel of 40 phenotypic and functional markers that allowed us to distinguish the diverse cell populations present in the tissues. We observed striking changes to almost all cell types within the DCIS tissue when comparing pure DCIS cases to DCIS adjacent to IDC. In particular, there were considerable changes to tissue-resident immune populations, with differences observed in the type, location and population size of a variety of immune cell subsets, in IDC-associated DCIS compared with DCIS-only cases. Computational mapping of cell type communities in pure and IDC-adjacent DCIS tissues revealed specific differences that could be used to predict the emergence of invasive disease, informing clinical management of this good prognosis malignancy and potentially avoiding over-treatment. Presentation: 6/3/2024