Incorporating stromal cells into a 3D patient-derived model for immuno-oncology drug testing in ovarian cancer

Abstract

High-grade serous ovarian cancer (HGSC) is the most aggressive form of ovarian cancer, with limited treatment options and frequent relapse. The tumour microenvironment (TME), particularly cancer-associated fibroblasts (CAFs), plays a crucial role in disease progression and therapy resistance. However, current HGSC in vitro models lack a comprehensive representation of TME, limiting their predictive value. To enable the development of a more TME-representative HGSC model, this thesis presents an optimized pipeline for isolating highly pure and viable CAFs from freshly resected HGSC tumours. Using 14 patient-derived samples from the ONCOSYS-OVA clinical trial, I refined dissociation and enrichment protocols as well as confirmed the successful maintenance of CAFs under various 3D culture conditions. I analysed cell composition using multicolour flow cytometry to evaluate dissociation protocols and compare enrichment strategies. When comparing dissociation protocols, sequential digestion using dispase followed by collagenase and hyaluronidase mix yielded the highest number of CAFs while preserving other key TME components. I used magnetic-activated cell sorting for CAF enrichment, and a negative selection strategy to deplete cancer and immune cells maximized CAF recovery with minimal contamination. I then cultured the enriched CAFs under various 3D conditions, including basement membrane extract, omentum-derived matrix, and suspension culture, to assess their growth potential and suitability for future incorporation into 3D HGSC models. While all tested conditions supported CAF spheroid formation, matrix selection influenced cellular morphology. These results will contribute to the development of a patient-derived multicellular HGSC model to support future therapeutic research.