ABSTRACT: High-grade serous ovarian cancer (HGSOC) is the most lethal gynaecological malignancy, largely due late diagnosis and tumor heterogeneity. Accurate preclinical modelling of HGSOC biological complexity is essential to improve translational research outcomes and therapeutic development; however, traditional 2D models have limited translational relevance. We therefore compared the morphological, proliferative, and molecular characteristics of four HGSOC cell lines (OVCAR-4, OVSAHO, COV362, and Kuramochi) in 2D versus 3D culture environments. To support systematic interrogation of 3D models, we developed reproducible protocols for 3D spheroid generation and optimized a dissociation method for downstream analysis. We observed significant differences in cellular behaviour, morphology, cell cycle regulation, and gene expression. Kuramochi cells showed upregulated EMT markers WNT11B and VIM, suggesting an increased invasive phenotype, i.e., enhanced epithelial–mesenchymal transition and signalling pathways. In OVSAHO cells, MMP2 was significantly downregulated while VEGFA was upregulated, highlighting alterations in matrix remodelling and angiogenesis, characterizing this cell line as proliferative rather than invasive. COV362 cells showed an initial decrease followed by an increase in KI67 expression. Downregulation of BRCA1 and KI67 indicates reduced cell proliferation in 3D environments. Proliferation kinetics were also altered in 3D environments: Kuramochi and OVCAR-4 adopted a more quiescent phenotype, while COV362 remained more proliferative with significant KI67 overexpression. These proliferation dynamics were confirmed with immunocytochemistry. Alignment of expression data showed that different cell types align with distinct HGSOC subtypes: OVASHO with immunoreactive, OVCAR4 with mesenchymal and differentiated, COV362 with differentiated and proliferative, and Kuramochi with proliferative subtype. The described 3D models can serve as high-throughput platforms for drug testing and functional studies, as well as for studying the complex molecular dynamics of ovarian cancer. Moreover, co-culturing systems can enhance their translational potential for the development of more effective therapeutic strategies.
