Carboplatin resistance remains a major challenge in the treatment of high-grade serous ovarian cancer (HGSOC), the most lethal subtype of epithelial ovarian cancer. To uncover mechanisms driving chemoresistance, we combined patient-derived transcriptomic data with advanced 3D in vitro models. Analysis of TCGA datasets revealed significant upregulation of matrix remodelling gene MMP9. To mimic clinical response, 3D models of four HGSOC cell lines were compared with conventional 2D models. The IC ₅₀ values for carboplatin were substantially higher in 3D models, indicating increased chemoresistance, consistent with the results of other studies. This difference is mainly due to reduced drug penetration within the dense 3D spheroid structure, which limits the exposure of cells inside the spheroids to the chemotherapeutic agent. Functional assays showed that carboplatin induced G0/G1 cell cycle arrest and decreased Ki67 levels at IC ₅₀ concentrations, suggesting a cytostatic effect. qPCR confirmed geometry- and treatment-dependent significant upregulation of the genes MMP9, TWIST2, ZEB2 , and WNT11 in 3D models, that recapitulate the adaptive resistance patterns observed in patient tumours. These results identify MMP9, TWIST2, ZEB2 , and WNT11 as promising biomarkers for overcoming chemoresistance in HGSOC. Further functional validation of these genes is needed to confirm the biological significance of the observed correlations.