P11.08 Systematic drug repurposing for faster cures of pediatric cancer identifies that Digoxin prolongs survival in a PDOX model of group 4 medulloblastoma

Abstract

Abstract Background: Medulloblastoma (MB) is the most common malignant brain tumor of childhood. While the current standard care for MB leads to long term survival in approximately 75 % of patients; recurrent and refractory MB continues to have dismal outcomes. In addition, long term survivors face significant treatment-related sequelae especially poor neurocognitive outcomes. Drug repositioning for known drugs is a promising potential strategy not only for drug discovery but also for accelerated translation into the clinical setting. To systematically explore thousands of known drugs available, we integrated computational biology and empirical biology methods to find old drugs for new indications in MB. Results: A non-parametric, bootstrapping based simulated annealing (NPBSA) algorithm was employed to identify driver signaling pathways for over 1,800 patients with Group 3 and Group 4 MB through an integrative analysis on their mRNA expression, DNA-copy number, DNA-methylation and DNA-seq profiles. Then, drug functional networks were constructed based on gene expression profiles under drug treatment as well as chemical structures and were clustered into drug modules with potential mechanisms of action. By evaluating targeted effects of 1,309 drugs from connectivity map database within each drug module in driver signaling pathways, we identified a group of known cardiac glycosides that top ranked among the total drug candidates for the Group 3 and 4 MB subtypes. In addition to traditional chemotherapeutic agents, members of the cardiac aminoglycoside family were repeatedly identified as potential therapeutic agents for MB. These findings were validated in multiple MB-derived cell lines which showed high rates of growth inhibition by cardiac aminoglycosides compared to controls. To evaluate if this growth inhibition in vitro correlated to prolonged survival in vivo, an extensively characterized patient-derived orthotopic-xenograft (PDOX) model of Group 4 MB (ICb-1078MB) was treated with digoxin (2 mg/kg i.p.) for 2 cycles of 14 days. Digoxin treatment significantly prolonged survival to 113 days from a median of 92 days in untreated controls (p=0.001). Histological evaluation of recurrent tumors following digoxin treatment demonstrated changes in the pattern of tumor spread, vascularity and necrosis compared to untreated controls. Conclusions: Leveraging big data in the domains of pharmacogenomics and the notion of drug functional networks and driver signaling pathways represents a powerful tool to repurpose known drugs for new indications in pediatric cancers. Using this integrative biology approach, we identified the cardiac aminoglycosides family generally and Digoxin, specifically, as potential novel agents in the treatment of pediatric medulloblastoma.