Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

Cancer Technologies

Cancer Technologies - Poster Session E

Poster S8 - An engineered model of metastatic colonization of human bone marrow reveals breast cancer cell remodeling of the hematopoietic niche

Saturday, October 26, 2024
10:00 AM – 11:00 AM EST
Location: Exhibit Hall E, F & G

Introduction:  There are an estimated ~2 million new cases and 600,000 cancer deaths per year in the United States. In breast cancer (BC), the majority of deaths (90%) are due to metastatic disease, where disseminated tumor cells can spread even before detection of the primary tumor. The human bone marrow (BM) is a major regulator of homeostasis and disease, and as the residing site of hematopoietic stem and progenitor cells (HSPCs), can respond to acute and systemic cues from cancer cells. Importantly, HSPCs are uniquely controlled by cellular components in the BM microenvironment, including osteoblasts, endothelial cells, and mesenchymal stem/stromal cells (MSCs). In metastatic disease, such as in triple-negative breast cancer (TNBC), tumor cells travel to the bone and BM, hide within the perivascular BM niche, and redirect the lineage commitment of HSPCs, generating myeloid-derived suppressor cells (MDSCs) that promote tumor progression. Following the treatment of metastatic disease, lingering cells often remain within the BM niche, establishing a minimal residual disease that stays dormant until recurrence. The cellular mechanisms which produce pro-survival signals and their impact on the hematopoietic niche are still poorly understood. Bioengineered human tissue models of metastatic sites capable of recreating the physiologically relevant milieu of metastatic colonization may strengthen our grasp of cancer progression and contribute to the development of effective therapeutic strategies. We report the use of an engineered tissue model of human bone marrow (eBM) to identify microenvironmental cues regulating cancer cell proliferation and to investigate how triple-negative breast cancer (TNBC) cell lines influence hematopoiesis.

Materials and

Methods:  All engineered human bone marrow (eBM) tissues were generated by first seeding BM-derived mesenchymal stem/stromal cells (MSCs) onto custom decellularized bone scaffolds and exposing cultures to osteogenic media for 3+ weeks. After maturation, additional BM-MSCs, human umbilical vein endothelial cells, and cord blood-derived hematopoietic stem/progenitor cells (HSPCs) were seeded within a fibrin hydrogel on top of the matured osteogenic tissues, with or without an additional tumor cell population (i.e., MDA-MB-231 cancer cell line, T47D HR+ cancer cell line, or primary breast cancer-derived tumor organoids). eBM tissues colonized with cancer cell lines were maintained with SFEM II basal media supplemented with 5% heat-inactivated FBS, 1% P/S, and 50 ng/mL each of SCF, TPO, and FLT-3L, for up to 14 days. For experiments characterizing tumor cell cycling and proliferation, a dual tdTomato and mVenus-p27K reporter in MDA-MB-231 cells was used in each ‘N-1’ combination of eBM tissues (i.e., OBs +/- ECs, MSCs, and HSPCs), and experiments were maintained for 10 days post-tissue maturation. For all experiments, subsets of eBM tissues were digested with Collagenase II and Nattokinase for flow cytometry analysis or RNA sequencing, in addition to a subset of tissues being fixed in 4% paraformaldehyde for histological analysis.

Results, Conclusions, and Discussions: Individual stromal components of the BM niche (osteoblasts, endothelial cells, MSCs) were each critical for regulating tumor cell quiescence and proliferation in the 3D eBM niche. In particular, endothelial cells and MSCs regulated increased tumor cell quiescence as compared to osteoblast-only tissues. We found that HSPCs impacted TNBC cell growth and responded to cancer cell presence with a shift of hematopoietic progenitors (CD34+CD38-) to downstream myeloid lineages (CD11b+CD14+) over two weeks. To account for tumor heterogeneity and demonstrate proof-of-concept ability for patient-specific studies, we demonstrated that patient-derived tumor organoids survive within the eBM tissues, maintaining both quiescent (p27K+) and proliferative (Ki-67+) patient-derived tumor cells over one week. These primary tumor cells within eBM tissues resulted in distinct shifts in myelopoiesis that are similar to those observed for aggressively metastatic cell lines, with a notably additional increase in granulocytes not found in integrated MDA-MB-231 cultures.

Over the last decade, several studies have documented the critical role of the microenvironment in modulating cancer cell proliferation, both in vitro and in vivo. The BM niche is known to be growth restrictive, and a multitude of mechanisms driving cellular dormancy of DTCs has been identified. Although animal models have unlocked a realm of therapeutic interventions, human tissue models are critical for evaluating the contributions of stromal cell populations to cancer cell behavior. We show that the human eBM model provides versatile cellular compositions in its microenvironment with a concerted impact on cancer cell cycling that can be parsed out by selective inclusion of constituent stromal cells using both transcriptomic analysis and reporters tailored to monitoring cell quiescence.

In summary, we report a novel approach to examining changes in the BM niche upon its metastatic colonization by BC cell lines and patient-derived tumor cells. We envision that this model will enable studies of the tumor-niche interactions that control metastatic homing in patient-specific settings, contributing to identification of therapeutic targets that can halt metastatic progression. In addition, patient-derived metastatic cells within engineered tissue models allow for the incorporation of patient variability into experimental design to identify commonalities in the resistance or responses to therapy.

Acknowledgements (Optional):  We gratefully acknowledge the funding support of this work by the National Institutes of Health (Grants 5R01CA249799, 3UH3EB025765 and P41 EB027062 to G.V.-N.; NRSA 1F31CA275733 to D.N.T.), National Science Foundation (Grant GRFP 1644869 to D.N.T.), and the Columbia Biomedical Engineering Kaganov Fellowship (to I.B.).