Abstract 666: Developing an ascitic fluid-derived organoid model for transcriptomic and therapeutic investigations in ovarian cancer.

Abstract Introduction: Epithelial ovarian cancer (EOC) spreads freely within the intraperitoneal cavity, enabling tumor dissemination, abdominal organ adhesion, and ascites formation through lymphatic obstruction. Because solid-tumor samples are scarce after debulking surgery, ascites provide a crucial window into disease evolution. Its rich cellular and acellular components fuel tumor survival, immune evasion, and chemoresistance but remain underused in preclinical models. We therefore aim to establish and characterize an ascites-derived organoid (AsO) platform that captures this heterogeneous tumor microenvironment (TME) and improves drug-response profiling. Methods: Malignant ascites and tumor tissues were collected from consenting Hispanic patients under an approved IRB protocol. Ascitic fluid was centrifuged and repeatedly lysed to remove red blood cells. The whole ascitic cells (WhAs) were divided for organoid culture and biobanking. Part of WhAs was embedded in extracellular matrix and overlaid with optimized organoid culture media. Organoids were passaged every 8-13 days and cryopreserved in freezing media. We documented morphology with brightfield imaging, stained FFPE sections with H&E, and immuno-stained whole-mount organoids. Bulk RNA-seq was performed on matched primary tumors (WhTu), WhAs and AsO, and we quantified expression using standard bioinformatics pipelines followed by correlation and pathway analyses. Drug sensitivity in organoids was measured by live/dead assay (CC50) and luminescence assay (IC50). Results: Organoid culture media optimization markedly improved organoid formation, passaging efficiency, and long-term expansion. A major achievement was the successful cryopreservation of both WhAs and AsO, which resumed growth after resuscitation in nearly all cases. H&E staining revealed nuclear atypia, a hallmark of EOC, and immunofluorescence confirmed expression of key biomarkers including PAX8, p53, CK8, and Ki67. Pearson correlation coefficients demonstrated strong global transcriptional similarity (R2 = 0.87) between AsO and WhAs. To identify ascites-specific genes within the ovarian cancer (OC) TME, we also examined transcriptional differences (DEGs) between AsO and WhTu. We identified 22 overlapping genes when comparing between the top 50 highly expressed genes from AsO and OC tumor-derived organoid datasets from Gene Expression Omnibus (GEO). Patient-specific drug responses to Carboplatin, Paclitaxel, Lenvatinib, and Saruparib have been observed. Conclusion: The comprehensive characterization of ascites-derived organoids at histological, protein marker, and transcriptomic levels, along with systematic drug sensitivity comparisons across multiple experimental platforms, will significantly enhance their clinical and translational relevance. Citation Format: Md Mynul Hassan, Zulfikar Azam, Taslim A. Al-Hilal, Sourav Roy. Developing an ascitic fluid-derived organoid model for transcriptomic and therapeutic investigations in ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 666.