Abstract IA13: Organoids and model systems: Rare ovarian cancers

Abstract

Abstract Although the majority of morbidity and mortality from ovarian cancer is due to high-grade serous cancers, our lack of effective treatment for other subtypes represents a substantial knowledge gap and unmet medical need. Over the past several years a subtype-specific approach to ovarian cancer research including genomics has led to a reasonable understanding of key genomic features for several subtypes. However, this has not been translated affectively into new management approaches. A key obstacle has been and continues to be the lack of credentialed model systems to better understand the biologic characteristics of these tumors and to produce preclinical data. For some ovarian cancer subtypes, for instance mucinous and SCCOHT, uncertainty around cell of origin has delayed the development of model systems. For others, such as clear cell and endometrioid carcinoma, although a tissue of origin (endometriosis) is widely accepted, the biologic impact of key mutations and the specific role that the ovarian microenvironment plays in shaping oncogenic opportunity are not understood. Each subtype of ovarian cancer represents a distinct set of biologic questions, and the needs for and approaches to the development of model systems are discreet. In this presentation the state of play for development of models for several subtypes will be described, along with local efforts to move this field forward. Clear cell and endometrioid carcinomas arise from displaced endometrial epithelium. They have some genomic distinctions but no mutually exclusive molecular features. It is possible that they arise from discreet cells of origin within endometriosis, yet this is unproven. Murine models exist that reliably produce tumors akin to both endometrioid and clear cell carcinomas, yet these models are difficult to work with. We and others have developed systems to culture normal uterine epithelial cells and induce the mutations most commonly found in distinct subtypes; such experiments may shed light on the oncogenic journey these cancers represent. Low-grade serous carcinoma often arises directly from serous borderline tumors and, unlike high-grade serous carcinoma, is a genomically stable cancer with frequent mutations in the MAP kinase pathway. Beyond serous borderline tumors there is no globally accepted precursor lesion; it is uncertain whether these cancers come from displaced fallopian tube epithelium or through the metaplasia or epithelial cells within the ovary. We and others have attempted to model this using normal epithelial cells. As this cancer does not respond well to standard chemotherapy, there is a profound need for preclinical model systems to test alternative approaches. Even less is known about the origins of nonepithelial ovarian cancers; although these cancers are all rare, they are devastating for families and patients. Small-cell hypercalcemic ovarian cancer was completely nonunderstood until four years ago, when we and three other group described mutations in SMARCA4. Given that there are three reasonable cell lines to study in this rare cancer, great progress has been made in identifying potential vulnerabilities. These cell lines do grow as xenografts and patient-derived xenografts have been developed; however, the lack of a known precursor lesion has to this point precluded development of a transgenic mouse model to study the specific biologic impact of SMARCA4 in the appropriate cell context. Although models based on granulosa and germ cells have failed to produce tumors, we are attempting to model this cancer through mutation of mesenchymal stem cells, an approach that has worked for rhabdoid cancers. Granulosa cell tumor of the ovary, adult type, is assumed to be derived from granulosa cells and features a specific FOXL2 mutation in 95% of cases. Unfortunately, there is only a single cell line that reflects this cancer, and attempts to derive xenografts and other lines have been sadly unsuccessful. Our initial attempts to develop models from this culture of granulosa cells have failed, perhaps due to the differentiation state of the cell we have used or the requirement for additional genetic events. To move this forward, we have conducted a large genomic screen of this cancer and identified other key mutations and are working to access follicular phase granulosa cells, which may be more amenable to transformation. As adult-type granulosa cells of the ovary do not respond to treatment beyond recurrent surgery, there is an urgent need for new treatment approaches and therefore a need for a model system that accurately reflects the biology and behavior of this cancer. Citation Format: David G. Huntsman. Organoids and model systems: Rare ovarian cancers [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr IA13.