The study of dormancy and reactivation of disseminated tumor cells (DTCs) in the distant organs are critical in in metastatic cancer research. During early stages of the disease, breast cancer cells have high chance to metastasize and settle in other organs such as bone. It is important to understand the spatial distribution of tumor cells under distinct microenvironments and how these niches determine the fates of tumor cells in homeostatic and pathological settings. In this work, we proposed a deep imaging systems biology (DISB) framework by integrating multiple components, including in vivo spontaneous tissue metastases models, tissue clearing, deep imaging, digital reconstruction, image quantitation, and graphic-theoretic analysis, to study the spatial distribution and pathophysiology of DTCs and related stroma cells in organs over time. Within the DISB framework, genetically engineered fluorescent reporter murine strains, which highlight specific bone marrow lineages, were used as the dimensional maps of diversified perivascular and endosteal niches. Metastases asynchronously occur in immune competent mice after resection of syngeneic primary tumors with or without adjuvant therapies, mimicking the natural course of metastatic spreading in clinics. Long bones carrying metastatic tumor cells were collected and subjected to the BABB (Benzoic Acid Benzyl Benzoate) clearing methodology. The spatial distribution of cancer and bone cells in the tumor tissue was captured by whole slide confocal imaging and digitally reconstructed. We developed deep learning image computing model to segment, quantitate, and obtain the spatial distribution and correlations among different components in the niche, such as cancer cells, vessels, and normal tissue cells over different bone specimens at various time points. We applied the DISB framework to image and trace specific bone marrow lineages (Olympus FluoView FV1000), and the results indicated that the spatial distribution of tumor cells in such spontaneous metastases models is significantly different from those in experimental metastases models. DTCs are mainly in close proximity to the endothelium, but such distribution is likely a random process. Of interest, the proliferating micrometastases were observed in close association with the endosteal vessels, suggesting the potential correlation of osteogenesis and metastatic growth in bone. In addition, adjuvant chemotherapies likely did not shift tumor cells across different bone marrow niches. Combined image-omics and graph-theoretic analysis on various cell types and vessels provide a new powerful new approach to understand the mechanism leading to diverse behaviors of tumor and various types of stroma cells in homeostatic and pathological settings in various bone marrow microenvironments under different perturbations or developmental stages.

Citation Format: Tiancheng He, Weijie Zhang, Jiasong Li, Jianting Sheng, Xiang Zhang, Stephen T. Wong. Mapping spatial cellular dynamics of bone marrow niches in metastatic breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 128.