Glioblastoma is a fatal primary malignant brain tumor. Despite therapies involving surgical resection, chemotherapy, and radiation therapy, the average survival for glioblastoma patients remains at approximately 15 months. MicroRNAs (miRNAs) are short noncoding RNA molecules that regulate the expression of the majority of human genes. Numerous genes are concurrently deregulated in glioblastoma. Consequently, molecular monotherapies have failed to achieve improvements in clinical outcomes. Several lines of evidence suggest that simultaneous targeting of several deregulated molecules is required to achieve better therapies. However, the simultaneous targeting of several deregulated oncogenic drivers is severely limited by the fact that the drugs needed to target many deregulated molecules do not currently exist, and because combining several drugs in a clinical setting leads to an exponential increase in toxicity. We hypothesized that we can develop and use miRNA to simultaneously inhibit multiple deregulated genes for more efficacious glioblastoma therapies. The goal of this study was therefore to identify master regulatory microRNAs (miRNAs) and use them to simultaneously target multiple deregulated molecules for GBM therapy. We defined master regulatory miRNAs as those that target several deregulated genes in glioblastoma. To find master regulatory miRNAs, we first used PAR-CLIP screenings to identify all targets of all miRNAs in glioblastoma cells. We then analyzed TCGA tumor data to determine which of these targets are deregulated in human tumors. We developed and used an algorithm to rank these targets for significance in glioblastoma malignancy based on their magnitude of deregulation, frequency of deregulation, and correlation with patient survival. We then ranked the miRNAs for their capacity of targeting multiple glioblastoma-deregulated genes and therefore the potential to exhibit strong anti-tumor effects when delivered as therapy. Using this strategy, we selected two tumor suppressor master regulatory miRNAs, miR-340, miR-382 and an oncogenic master regulatory miRNA, miR-17. We validated the target genes of the miRNAs and showed that they form part of important glioblastoma regulatory pathways. We then showed that the miRNAs (miR-340 and miR-582) or the miR-17 inhibitor have strong inhibitory effects on glioblastoma cell growth, survival, invasion, stemness and in vivo tumor growth. Ultimately, we developed and successfully tested a new therapeutic approach to delivery miR-340 using MRI guided focused ultrasound and microbubbles (FUS-MB) and special brain penetrating nanoparticles (BPN). This approach resulted in a substantial reduction in tumor volume and prolongation of the survival of glioblastoma-bearing mice and can be translated into clinical trials. We therefore developed and successfully tested a novel strategy to discover and deliver miRNAs for glioblastoma and cancer therapy.