This study developed a 3D pentaculture model incorporating five human cell types, successfully simulating the tumor microenvironment of high-grade serous ovarian cancer (HGSOC), and revealed that tumor cells regulate macrophage polarization through genomic and transcriptomic heterogeneity, providing a novel platform for studying tumor-immune interactions and targeted therapies.
Literature Overview
The article "3D pentaculture model unveils malignant cell-driven macrophage polarization in high-grade serous ovarian cancer," published in Nature Communications, reviews and summarizes the complex immunosuppressive tumor microenvironment (TME) of high-grade serous ovarian cancer (HGSOC) and its impact on therapy resistance. The research team developed a 3D pentaculture model integrating malignant HGSOC cells, primary fibroblasts, mesothelial cells, adipocytes, and monocytes, successfully recapitulating the structural and functional features of the human TME in vitro. This model induces monocyte differentiation into macrophages without exogenous cytokines and, through bulk RNA sequencing and deconvolution analysis of single-cell RNA-seq data, confirms that the resulting macrophage clusters closely resemble those found in patient metastatic lesions. Additionally, the study evaluated the effects of monoclonal antibodies targeting the "don't eat me" signals CD47 and CD24 on macrophage activity, demonstrating the model’s potential for drug screening. The study emphasizes the dominant role of tumor cells in regulating myeloid cell behavior within the TME, offering new insights into HGSOC immune escape mechanisms.Background Knowledge
High-grade serous ovarian cancer (HGSOC) is the most common and lethal subtype of ovarian cancer, with most patients diagnosed at an advanced stage, making treatment challenging. Although initially responsive to platinum-based chemotherapy, over 80% of advanced patients experience recurrence and develop resistance. In recent years, anti-angiogenic and targeted therapies have improved outcomes for some patients, but the five-year survival rate remains low. HGSOC exhibits high genomic heterogeneity, with nearly ubiquitous TP53 mutations and BRCA1/2 or homologous recombination repair deficiency (HRD) in some cases—these intrinsic features influence treatment response. Single-cell and spatial transcriptomic studies have further revealed high heterogeneity among immune and stromal cells in the TME, suggesting that tumor cells can reprogram the microenvironment via secreted signaling molecules to promote immune escape. Macrophages are among the most abundant immune cells in the HGSOC TME, typically exhibiting an M2-like immunosuppressive phenotype that supports tumor progression. Strategies targeting macrophages include inhibiting their recruitment (e.g., CCL2/CSF1R pathway) or reprogramming their function (e.g., blocking "don't eat me" signals such as CD47-SIRPα and CD24-SIGLEC10), but clinical translation faces challenges, partly due to the lack of in vitro models that accurately simulate the human TME. Traditional 2D cultures poorly maintain cell-cell interactions and polarization states, while animal models are time-consuming and costly, especially limiting for evaluating human-specific antibody efficacy. Therefore, developing more clinically relevant in vitro TME models—particularly complex co-culture systems containing functional myeloid cells—has become a current research focus and challenge. This study addresses this need by integrating multiple primary and tumor cells to establish a 3D model that dynamically simulates the HGSOC metastatic microenvironment, filling a critical gap in existing models regarding macrophage polarization and function.
Research Methods and Experiments
The research team built upon a previously established four-cell model by incorporating CD14+ monocytes derived from healthy donors to create a 3D pentaculture system consisting of malignant HGSOC cells (AOCS1, G164, OvCAR3), primary fibroblasts, mesothelial cells, adipocytes, and monocytes. An adipocyte gel layer, together with fibroblasts and mesothelial cells, formed a stromal environment mimicking the omentum, followed by the addition of tumor cells and monocytes, with cultures maintained for 7 days. Cell phenotypes and spatial distributions were analyzed using histological staining, flow cytometry, and immunofluorescence. Bulk RNA sequencing combined with scRNA-seq data deconvolution (CIBERSORTx) was used to assess the proportions of cell subpopulations and macrophage polarization states within the TME. Expression levels of the "don't eat me" signals CD47 and CD24 and their receptors SIRPα and SIGLEC10 were measured across different models. Tumor cell phagocytosis by macrophages was evaluated using red-labeled tumor cells and live imaging, and the effects of anti-CD47, anti-CD24 monoclonal antibodies, and a TGFβ receptor inhibitor on phagocytosis and tumor cell survival were tested.Key Conclusions and Perspectives
Research Significance and Prospects
The pentaculture model provides a highly physiologically relevant in vitro platform for studying the HGSOC tumor microenvironment, overcoming the limitations of traditional 2D models in capturing complex cellular interactions and the shortcomings of animal models in evaluating human-specific antibody efficacy. By recapitulating macrophage polarization features observed in patient tumors, the model demonstrates strong utility for mechanistic studies and drug screening.
The study reveals the profound impact of tumor cell genomic and transcriptomic heterogeneity on macrophage function, underscoring the importance of personalized therapeutic strategies. Notably, the heterogeneous expression of "don't eat me" signals suggests that targeted therapies should be guided by biomarkers to avoid ineffective treatments.
The model can be further used to explore signaling networks between tumor and myeloid cells, identify novel immune regulatory targets, and test combination therapies (e.g., TGFβ inhibitors with phagocytosis checkpoint blockade). In the future, it could be expanded to include T cells or other immune cells, enabling the construction of more comprehensive immune microenvironment models and accelerating immunotherapy development.
Conclusion
This study established an innovative 3D pentaculture model capable of effectively simulating the tumor microenvironment of high-grade serous ovarian cancer, particularly recapitulating macrophage polarization and dynamic interactions with tumor cells. The model induces monocyte-to-macrophage differentiation without exogenous factors, and the resulting macrophage subpopulation distribution closely resembles that of patient metastases, highlighting the dominant role of tumor cells in shaping the immune microenvironment. The study further revealed that different tumor cell lines regulate macrophage phagocytic function through differential expression of "don't eat me" signals such as CD47 and CD24, and validated that targeting these pathways with monoclonal antibodies can enhance anti-tumor immune responses in specific contexts. This platform not only deepens our understanding of HGSOC immune escape mechanisms but also provides a reliable in vitro tool for developing personalized immunotherapies. In the future, this model holds promise for high-throughput drug screening, optimization of combination therapies, and biomarker discovery, advancing precision medicine for ovarian cancer.
