(335) Impact of CAR T-Cell Therapy on Ovarian Follicles: Insights from an ALL Mouse Model

Natalie Hanby^1*; Raghu Kavarthapu^1*; Alka Dwivedi^2*; Mary E. Soliman¹; Nirali N. Shah²; Naomi Taylor²; Veronica Gomez-Lobo²

*Equal contribution

1. Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Department of Pediatric and Adolescent Gynecology, Bethesda, United States

2. National Cancer Institute (NCI), Pediatric Oncology Branch, Bethesda, United States

Abstract Text:

Approximately 80% of B-lineage acute lymphocytic leukemia (B-ALL) cases are diagnosed in children, making it the most common form of cancer amongst pediatric patients. Standard treatment options for B-ALL include chemotherapy, radiation, and stem cell transplant, resulting in a 90% long-term survival rate for children. Although effective, these standard treatment options are gonadotoxic. This can have detrimental impacts on the ovarian follicle pool, often resulting in premature ovarian insufficiency. Chimeric antigen receptor T-cell (CAR T-cell) therapy is a targeted immunotherapy that has shown promising results for blood cancers such as B-ALL. As CAR T-cell therapy continues to evolve, there is a need to understand its implications on ovarian health, especially in young patients. To fill this gap, we have investigated the effects of CAR T-cell therapy on ovarian follicular development in a humanized mouse model of B-ALL. Immunodeficient NOD-SCID gamma (NSG) mice were engrafted with human GFP⁺luciferase⁺ NALM6 B-ALL (n=5).  Mice were then treated with T-cells engineered to express a CD19-targeting CAR bearing either a CD28 (n=5) or 4-1BB (n=4) costimulatory domain. Mice treated with mock-transduced T-cells (n=5) were used as a negative control. A non-cancerous sham group injected with saline was also included (n=5). After 17 days, mice were sacrificed and ovaries were harvested and processed for histology, immunohistochemistry, and follicle counting. Bioluminescent imaging revealed a rapid progression of leukemia in untreated and mock-transduced T-cell treated mice whereas leukemia was rapidly eradicated by adoptive transfer of either CD19-CD28 or CD19-41BB CAR T-cells. In mice with progressive disease, leukemic cells were detected in the ovaries as indicated by GFP staining. Additionally, this group showed a high level of inflammation in the ovaries, as demonstrated by the presence of TNF-alpha. Consistent with bioluminescent imaging, tumor infiltration was not detected in mice treated with CAR T-cells and the absence of leukemia was associated with the recruitment of adoptively transferred CD8⁺ CAR T-cells. Oocyte specific protein DDX4 was used to quantify follicles, revealing a significantly decreased number of primordial and primary follicle counts in untreated, mock-transduced, and CD19-41BB CAR T-cell treated mice. However, follicle counts of CD19-CD28 CAR T-cell treated mice had no significant differences compared to control. Leukemia progression was also associated with significantly increased numbers of apoptotic follicles, as assessed by CASP3 expression. Notably, treatment with both CD19 CAR T-cell therapies reduced apoptosis to levels similar to that found in control mice. In conclusion, human B-ALL infiltration into mouse ovaries was associated with reduced numbers of primordial and primary follicle numbers, increased apoptosis, and inflammation. Treatment with either CD19-28 or CD19-41BB CAR T-cells eradicated leukemia, reduced ovarian inflammation, and resulted in lower levels of apoptotic follicles. While CD19-41BB CAR T-cell therapy effectively cleared leukemia, it did not prevent follicle loss. In contrast, CD19-CD28 CAR T-cells not only cleared leukemia but also significantly preserved follicle numbers, suggesting a unique protective effect on ovarian follicle reserve. Together, these findings highlight the detrimental impact of leukemia on follicle health, reveal potential differences between CAR T-cell therapies in preserving follicle loss, and show promising outcomes for ovarian health compared to standard of care. This work was supported by NIH Grant ZIA HD009005 and ZIA BC 011923 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the NCI, respectively.