A Comprehensive Examination of Ductal Carcinoma In Situ to Predict Disease Outcome

Abstract

Ductal carcinoma in situ (DCIS) is a non-obligate precursor of invasive ductal carcinoma (IDC). For most patients, the current standard of care is surgery and radiation. Since not all lesions progress to IDC, there is a critical imperative to better predict DCIS outcome to allow for de-escalated treatment. Uncovering the cellular interactions and molecular mechanisms contributing to DCIS progression has provided optimism for the search for prognostic biomarkers that can accurately predict the risk of progression in the clinical setting. This thesis employed a range of methodologies to contribute to this work. Several active surveillance trials are underway as an alternative management approach for patients with low-risk DCIS. Surgical upstaging rates were investigated among a cohort of patients with biopsy diagnosed DCIS, in which 28% of patients were surgically upstaged to either invasive (82%) or microinvasive (18%) breast cancer. Histological and clinical features collected during routine care were insufficient to accurately predict concurrent IDC, imploring the need to better define risk. To investigate biomarkers of DCIS progression, the genomic impacts of coculturing breast cancer and normal breast organoids in matrix-embedded culture were assessed via single-cell transcriptomics. Several functional changes were seen in multiple normal breast cell types in response to cancer cells and the presence of normal organoids also influenced tumouroid size and transcriptomes. Conventional and highly multiplexed spatial proteomics were employed to investigate all cell types within pure and mixed DCIS tissue. This revealed extensive changes to almost all cell types contained within the DCIS ecosystem. Moreover, phenotypic changes to marker expression, cellular composition and tissue organisation were specific to disease state, suggesting that these changes are predictive of progression along the continuum to IDC.