Three-Dimensional Modelling of Ovarian Cancer: From Cell Lines to Organoids for Discovery and Personalized Medicine

Abstract

Ovarian cancer has the highest mortality of all of the gynecological malignancies. There are several distinct histotypes of this malignancy characterized by specific molecular events and clinical behavior. These histotypes have differing responses to platinum-based drugs that have been the mainstay of therapy for ovarian cancer for decades. For histotypes that initially respond to a chemotherapeutic regime of carboplatin and paclitaxel such as high-grade serous ovarian cancer, the development of chemoresistance is common and underpins incurable disease. Recent discoveries have led to the clinical use of PARP (poly ADP ribose polymerase) inhibitors for ovarian cancers defective in homologous recombination repair, as well as the anti-angiogenic bevacizumab. While predictive molecular testing involving identification of a genomic scar and/or the presence of germline or somatic BRCA1 or BRCA2 mutation are in clinical use to inform the likely success of a PARP inhibitor, no similar tests are available to identify women likely to respond to bevacizumab. Functional tests to predict patient response to any drug are, in fact, essentially absent from clinical care. New drugs are needed to treat ovarian cancer. In this review, we discuss applications to address the currently unmet need of developing physiologically relevant in vitro and ex vivo models of ovarian cancer for fundamental discovery science, and personalized medicine approaches. Traditional two-dimensional (2D) in vitro cell culture of ovarian cancer lacks critical cell-to-cell interactions afforded by culture in three-dimensions. Additionally, modelling interactions with the tumor microenvironment, including the surface of organs in the peritoneal cavity that support metastatic growth of ovarian cancer, will improve the power of these models. Being able to reliably grow primary tumoroid cultures of ovarian cancer will improve the ability to recapitulate tumor heterogeneity. Three-dimensional (3D) modelling systems, from cell lines to organoid or tumoroid cultures, represent enhanced starting points from which improved translational outcomes for women with ovarian cancer will emerge.

Keywords: 3D bio-printing; 3D cell culture; drug screening; organoids; ovarian cancer; personalized medicine; tumoroid.

FIGURE 1

Ovarian cancer histotypes and gene mutations. Epithelial ovarian cancers constitute approximately 90% of all malignant ovarian tumors and are made up of different histotypes: high-grade serous ovarian cancer (HGSOC), endometrioid ovarian cancer (EnOC), ovarian clear cell carcinoma (OCCC), low-grade serous ovarian cancer (LGSOC) and mucinous ovarian cancer (MOC). Ovarian carcinosarcomas (OCS)/malignant mixed mullerian tumors (MMMT) have epithelial and mesenchymal components. Stromal cell tumors include granulosa cell tumors (GCT, adult and juvenile) as well as Sertoli-Leydig cell tumors (SLCTs). Small cell carcinoma of the ovary, hypercalcemic type (SCCOHT) are a rare histotype. Gene mutations, copy number amplifications, methylation and other epigenetic silencing are noted against histotypes.

FIGURE 2

Techniques to create scaffold-free 3D in vitro cancer models. Creation of 3D cell models in the absence of scaffolds promotes cell-cell interactions in three dimensions that mediate cell behavior and drug response when compared to (A) 2D monolayers. Use of (B) liquid overlay techniques with i) flat or ii) round-bottomed ULA plates (C), hanging drop techniques and (D) rotating bioreactors such as i) spinner flasks and ii) horizontal rotating vessels have been used as time and cost-effective spheroid creation methods or to investigate drug response and other factors that my influence ovarian cancer progression, such as fluid shear stress and hypoxia. Created with Biorender.com.

FIGURE 3

Techniques to create 3D in vitro cancer models using scaffolds. Addition of extracellular matrix (ECM) as scaffolds for 3D cell cultures enables both cell-cell and cell-ECM interactions for a more physiologically relevant 3D cancer cell model. Methods include (A) ECM/hydrogels with cancer cells i) on top of, or ii) encapsulated within an ECM, iii) organotypic omental co-culture model and iv) organoid propagation. (B) 3D bioprinting techniques such as i) extrusion-based bioprinting enables creation of 3D cell-laden models in hydrogels in a layer-by-layer manner, and ii) droplet-based bioprinting enabling high-throughput creation of 3D cell models in hydrogels with higher spatial control for more complex co-culture. (C) Tumor-on-a-chip microfluidic devices have been used to model the effects of fluid shear stress, as well as simulating nutrient, gas and drug gradients, for ascites metastasis modelling. Created with Biorender.com.

FIGURE 4

A bench-to-bedside approach using 3D cell cultures to fast track personalized therapies for ovarian cancers. Utilization of (A) samples from multiple patient tumor sites, (B) isolation of cancer cells ex vivo for (C) molecular profiling and (D) propagation as 3D cell cultures can identify clues regarding a patient’s unique tumor phenotype. Based on these findings, (E) a high-throughput drug screen of molecularly relevant drugs in 3D cell cultures can be employed to predict drug efficacy and utilized to guide a personalized medicine approach. Created with Biorender.com.