Key points are not available for this paper at this time.
Abstract Ovarian carcinosarcoma (OCS) is a rare and extremely lethal gynecological cancer, accounting for less than 1% of all ovarian cancers, and harboring an unfavorable prognosis compared to the common subtype. There are no efficient treatments available, highlighting the need for a reliable pre-clinical model as a tool to study this disease. Our goal was to generate a genetically engineered mouse model (GEMM) of OCS, carrying relevant genetic alterations, in order to study OCS pathogenesis. To generate the model, we used a Cre-Lox system for targeted knockout of three relevant tumor suppressor genes: Trp53, Brca2 and Pten. Mice in our model also carry a Lox-stop-Lox-TdTomato transgene, which results in fluorescent TdTomato protein expression in all recombined cells. Gene knockout in our model was achieved via injections of adenoviral particles coding for CRE recombinase into the ovarian bursa. Female mice in our model developed tumors arising from the ovary with an admixture of malignant epithelial and stromal elements, successfully mimicking human OCS. Tumors often metastasized to the abdominal cavity and at times generated ascites. The tumors showed patchy staining for PAX8, ER, Cytokeratin and Vimentin, further confirming their resemblance to human OCS. Mice harboring homozygous deletions of all three tumor suppressor genes developed tumors rapidly, first imaged 7 weeks after induction, and euthanized at a humane end point 11-14 weeks after CRE injection (average 11. 4±0. 95 weeks). This rapid timeline, together with high Ki-67 expression, suggest that tumor development in our model mimics the highly aggressive nature of the human disease. Interestingly, homozygous deletion of Trp53 and Pten and heterozygous deletion of Brca2 did not significantly alter time to tumor formation or tumor morphology, compared to the original model, suggesting that homozygous deletion of Brca2 is not essential or OCS development in our model. On the other hand, heterozygous deletion of either Trp53 or Pten, maintaining homozygous deletion of the other alleles, significantly delayed tumor formation and reduced tumor penetration, suggesting that Trp53 and Pten play a significant role in this model. Finally, harvesting mouse ovaries at very early time points, prior to tumor development, showed early pre-malignant lesions emerging from the ovarian surface epithelium, implicating ovarian surface epithelial cells as a potential OCS cell of origin of in our model. In conclusion, we have developed a reproducible model of OCS arising from the ovarian surface epithelium, correctly mimicking human disease in terms of clinical course, morphology, and tumor markers. This novel pre-clinical model, together with the mouse OCS cell lines we have generated from the model, will allow studies of disease pathogenesis, development of early detection and prevention tools, and studies of new OCS treatments. Citation Format: Einav Bangiev-Girsh, Amir Basis, Paul Zannou, Inna Naroditsky, Asaf Aizic, Ami Aronheim, Liron Berger, Ruth Perets. A novel genetically engineered mouse model of ovarian carcinosarcoma abstract. In: Proceedings of the AACR Special Conference on Ovarian Cancer; 2023 Oct 5-7; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84 (5 Suppl₂): Abstract nr A067.
Bangiev-Girsh et al. (Mon,) studied this question.