EXTH-64. MECHANISMS OF MITOCHONDRIAL COMPLEX-I SENSITIVITY IN GBM

Abstract The biguanide metformin that is widely used for the management of Type 2 diabetes is being evaluated as an anti-neoplastic agent in cancer trials including GBM (NCT02780024, NCT03243851). The mechanisms of action of metformin as an anti-neoplastic agent is still being investigated, but recent studies have unequivocally demonstrated that its therapeutic effects in tumor cell growth and signaling are dependent on inhibition of mitochondrial complex I. Based on the success of metformin in combination therapy in some cancers, new generation of mitochondrial complex I inhibitors (MCI-i) such as IM156, IACS-010759, are currently under investigation. Diffusion of metformin across the plasma membrane occurs slowly. Its entry is facilitated by two transporters OCT1 and OCT2 that are not well expressed outside the liver and kidney. This presents a problem due to its lack of accumulation in other tissues at therapeutic concentrations. Therefore, identification of tumor subtypes that respond better to lower concentrations of metformin may be leveraged for targeted metformin therapy. Metformin GBM trials may be unsatisfactory since there are no molecular markers to distinguish metformin responders from non-responders. Identification of molecular markers may greatly improve metformin and other MCI-i-based therapy in GBM. By examining TCGA data and experimental analysis, we have identified that PDHA, that catalyzes pyruvate to acetyl Co-A, is a predictive molecular marker for MCI-i sensitivity in GBM and potentially beyond GBM. PDHA expression varies widely in patient tumors, potentially due to differential methylation. We are performing joint pathway analysis by examining differentially expressed genes and stable isotope tracing studies to determine distinct metabolic signatures in the two subsets of GBM. We are also performing Classification And Regression Tree (CART) analysis on a cohort of sixty primary GBM lines and tissues to predict if PDHA expression provides a binary outcome for MCI-i sensitivity.