Abstract IA021: Treatment of advanced ovarian cancer using CAR-NK cells

Abstract Recurrent ovarian cancer (OC) remains a major clinical challenge, with limited therapeutic success from novel immunotherapies such as immune checkpoint inhibitors and chimeric antigen receptor (CAR) T cells, which show response rates below 20%. The median 5-year survival of OC patients also remains dismal at 20%, highlighting the need for novel therapies. To this end, our group is developing a natural killer (NK) cell-based immunotherapy, which offers advantages over T cell therapies, including minimal graft versus host disease (GVHD) and cytokine release syndrome (CRS), and rapid off-the-shelf deployment. Members of our team pioneered clinical trials using NK cells for OC, but success has been limited, likely due to poor NK persistence and the immunosuppressive tumor microenvironment. To overcome these challenges, we developed nonviral transposon engineering methods to generate soluble IL15 (sIL15) cytokine-armored, mesothelin targeting CAR-NK cells (CAR-NK15). CAR-NK15 cells exhibit enhanced persistence, metabolic fitness, cytokine production and cytolytic activity against OC in vitro and in vivo but fail to clear OC tumors in vivo. Provocatively, recent studies demonstrated that the addition of TGFβ during NK cell expansion transcriptionally reprograms NK cells (‘TGFβ imprinting’; TGFβi) with novel characteristics, including TGFβ resistance, improved metabolic fitness, enhanced homing, hypersecretion of cytokines, and improved killing kinetics. Thus, we deployed TGFβi in CAR-NK15 cells to generate TGFβi CAR-NK15 (CAR-NK15i) cells. Our studies demonstrate that CAR-NK15i demonstrates further improvement in metabolic fitness, serial killing capabilities (up to 20 rounds), and clearance of xenografted OC cell lines in NSG mice within 2 weeks. However, subsequent to tumor clearance we observe significant unchecked in vivo expansion of the CAR-NK15i cells that leads to animal death. Preclinical IND enabling studies demonstrate that CAR-NK15i remain cytokine (IL2) dependent in vitro and are not cytotoxic to normal human cells. We thus hypothesized that the unchecked in vivo expansion would not be observed in patients due to an intact immune system that would act as a cytokine sink for sIL15 and mount rapid elimination of the allogeneic CAR-NK15i cells. Indeed, using humanized NSG animals xenografted with OC cell lines prevented unchecked in vivo expansion of CAR-NK15i cells and CAR-NK15i cells were eliminated in less than 2 weeks while still reducing tumor growth and extending animal survival, albeit without complete tumor clearance. Based on these studies, we are translating CAR-NK15i cells through an NIH SPORE grant (CA136393; in yr 4 of 5) initiating in early 2026. Ongoing research is focused on multi-dosing strategies of CAR-NK15i cells in humanized animal models and further enhancement of CAR-NK15i cells through engineering to express and secrete bi-specific T-cell engager (BiTEs) to engage the host T cells with the hope of instigating a secondary immune response mediated by neoantigen reactive T cells for long term durable cures. Citation Format: Branden Moriarity, Beau R. Webber, Kylie Buchanan, Timothy Folsom, Melissa Geller, Joseph Skeate, Joshua B. Krueger, Jae Woong Chang. Treatment of advanced ovarian cancer using CAR-NK cells abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Ovarian Cancer Research; 2025 Sep 19-21; Denver, CO. Philadelphia (PA): AACR; Cancer Res 2025;85 (18Suppl): Abstract nr IA021.