Abstract 418: Mitochondrial perturbations as a novel approach to personalized medicine

Proceedings: AACR 107th Annual Meeting 2016; April 16-20, 2016; New Orleans, LA Recent progress in cancer research has been marked by the continued development of exciting drugs that selectively target vulnerabilities present only in cancer cells, so-called targeted therapies. Too often, however, we are lacking the proper predictive biomarkers to identify which patients will benefit most from individual targeted therapies. Since most chemotherapeutic agents ultimately function by inducing the mitochondrial apoptotic pathway, we hypothesized that a tool that measures mitochondrial sensitivity may serve as a broadly predictable biomarker. Hence to address the unmet need for predictive biomarkers, we have developed a tool called Dynamic BH3 Profiling (DBP). In this approach, we briefly treat cancer cells (14h) from patients with targeted drugs and measure the apoptotic threshold of a cell “priming” by treatment with peptides derived from BH3 domains of pro-apoptotic proteins. Previous studies from our laboratory demonstrated that when a patient's cells exhibit robust death signaling in the DBP assay in response to a particular drug, it predicts a good clinical response to that drug in patients. Using xenotransplantation of primary AML samples into NSG mice (PDX models), we validated the utility of DBP to personalize acute myeloid leukemia therapy. We established 12 independent AML PDX models and determined the mitochondrial response of splenic myeloblasts from each PDX to 25 targeted agents. Ex-vivo DBP measurement demonstrated heterogeneous responses; while some drugs induced priming in all PDXs, others increased priming in only selective PDX models. For instance, JQ-1 (BRD-4 inhibitor) and birinapant (SMAC mimetic) demonstrated inverse correlation of their activity among 3 different PDXs. These suggested that mitochondrial priming can stratify AML PDX models according to predicted sensitivity to targeted agents. When we compared DBP results with cell death assay, we found that myeloblasts from most of the PDXs failed to sustain long-term culture (72h), a requirement for cytotoxicity measurements. However, myeloblasts from 3 of the PDXs with high starting viability showed strong correlation between mitochondrial priming and cell death, implying that cells with increased mitochondrial response are destined to commit apoptosis. By testing functional state of the apoptotic pathway in myeloblasts from spleen, bone-marrow and peripheral blood with BH3 profiling, we found that splenic blasts were relatively more primed for apoptosis in 8 out of 10 PDXs. Correlatively, drug induced priming also varied between myeloblasts from these 3 different sites. Finally, we are directly testing the performance of DBP as a predictive biomarker by treating a variety of PDX models with drugs, including drugs predicated to be active and drugs predicated to be inactive. Collectively, mitochondria based measurements can predict good and bad response to individual targeted agents and may serve as a powerful biomarker to choose patient therapy. Citation Format: shruti bhatt, david weinstock, Anthony Letai. Mitochondrial perturbations as a novel approach to personalized medicine. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 418.