Abstract Ovarian cancer is the most lethal gynecological cancer with a five-year survival rate of 49.7%. Almost 1 in 6 patients dies within the first three months of diagnosis. Targeted therapies such as poly (ADP-ribose) polymerase (PARP) inhibitors have gained popularity due to their reduced toxicities. However, only 41% of patients respond to PARP inhibitors and most patients develop resistance. Here, we explored mechanisms of PARP inhibitor resistance in ovarian cancer at single-cell resolution to characterize potentially transient transcriptomic phenotypes associated with resistance. Parental (sensitive) and PARP inhibitor-persistent populations from two ovarian cancer cell lines (OVCAR4 and OVCA3) were sequenced by single-cell RNA sequencing (10X Genomics). Count matrices were log-normalized and scaled to remove variation due to read counts. Dimension reduction was performed using principal component analysis (PCA); cells were clustered phenotypically with graph-based clustering (Louvian algorithm) and visualized with UMAP projections. Downstream analysis includes characterization of the cell cycle phase, calculation of single-cell metabolic activity scores across 37 KEGG metabolic pathways, and pseudotime evolutionary trajectory analysis was performed to study the transition from the sensitive to the persistent phenotype. Across sensitive populations, most cells were in G1-phase; however, after developing persistence to PARP inhibitors, the OVCAR4 cells arrested at G2M phase, while the OVCA3 cells continued to arrest in the G1 phase. Our metabolic single cell RNA seq analysis pipelines revealed that OVCA3 sensitive population expressed high glycolysis metabolic scores compared to their persistent population, while the OVCAR4 persistent population showed a shift toward oxidative phosphorylation compared to the sensitive population implying the heterogenous metabolic adaptation across cancer cells. Persistent cells have high glutamine metabolism scores, suggesting that this pathway is upregulated in these populations compared to the parental ancestors. These findings were experimentally validated in cell lines treated with glutamine deprived media treated with different concentrations of PARP inhibitor. We used Palantir to characterize the evolutionary trajectories of the cells based on predictions of cell fate probabilities. OVCAR4 cells were predicted to develop into two distinct transcriptomic phenotypes, which could be related with sensitivity or tolerance to PARP inhibitors. However, OVCA3 cell did not have distinct, clear differentiation trajectories suggesting that transcriptomic information alone is not sufficient to explain the development of resistance in this cell line. These approaches further our understanding PARP inhibitor resistance in ovarian cancer, which is needed to improve current treatment strategies. Citation Format: Adriana Del Pino Herrera, Meghan Ferrall-Fairbanks, Mehdi Damaghi, Joon-Hyun Song, Snigdha Kanadibhotla. Glutamine metabolism drives some mechanisms of PARP inhibitor persistence in ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 870.
Read full abstract