Three-dimensional organoid models to study breast cancer progression

Approximately 20% of breast cancers detected through mammography are pre-invasive Ductal Carcinoma in situ (DCIS). If left untreated, approximately 20-50% of DCIS will progress to more deadly Invasive Ductal Carcinoma (IDC). No prognostic biomarkers can reliably predict the risk of progression from DCIS to IDC. Similar genomic profiles of matched pre-invasive DCIS and IDC suggests that the progression is not driven by genetic aberrations in DCIS cells, but microenvironmental factors, such as hypoxia and metabolic stress prevalent in DCIS, may drive the transition. We need innovative models to investigate how to halt steps of DCIS progression to invasive phenotypes and subsequent metastasis from the primary site. This proposal directly addresses this unmet need by developing a novel three-dimensional in vitro organoid model that recapitulates key hallmarks of DCIS to IDC progression: tumor-size induced hypoxia and metabolic stress, tumor heterogeneity and spontaneous emergence of migratory phenotype in the same parent cells without any additional stimulus. A tangible advantage of the proposed organoid models is the ability to precisely and reproducibly study how the hypoxic microenvironment induces tumor migration in real time and in isolation from non-tumor cells present in vivo, providing unique opportunity to define tumor-intrinsic mechanisms of DCIS to IDC progression. During July 2018-Feb 2022 ESI MERIT Award period, we have shown that inhibition of tumor-secreted factors effectively halts organoid migration, while inhibition of hypoxia is effective only within a time window and is compromised by tumor-to-tumor variation, supporting our notion that hypoxia initiates migratory phenotypes but does not sustain it. We have also analyzed secretome from metastatic breast cancer pleural effusion showing significantly higher levels of CCL2/MCP1, CXCL10/IP10, IL-6, IL-8, regulatory IL-10, and IL-7 and IL-15. Strategies to neutralize these key cytokines may generate anti-tumor responses in the pleural environment. Microarray analysis of hypoxia-induced migration and secretome-induced migration suggested role of Rho GTPase and PI3K/AKT signaling pathways in maintaining migration. Our results show that hypoxic organoid models exhibit partial EMT signatures as early as day 1, which is maintained as these non-migratory organoids transition to migratory phenotypes. During the two-year extension period, we will continue 1) to optimize our DCIS models incorporating ductal structure and other components from DCIS microenvironments; 2) to test new mechanisms linking tumor-intrinsic hypoxia, partial/hybrid EMT and collective migration; 3) to inhibit signaling mechanisms to halt emergence of migratory phenotypes. The successful completion of the proposed work will provide answers to two fundamental questions in the progression of invasive breast cancer: 1) What causes some DCIS cells to become migratory and develop into invasive tumors? 2) How and where does the migratory phenotype (IDC) emerge? The mechanistic understanding gained from these studies will improve diagnosis, lead to the development of treatment strategies to arrest invasion at the pre-malignant stage, and thus prevent patient overtreatment.