Abstract: Glioblastoma multiforme (GBM), a grade IV glioma, is the most malignant form of an astrocytoma and is the most common malignant brain tumor in adults. The cause of this primary and highly aggressive cancer is unclear. The current treatment for glioblastoma is limited to maximal safe surgical resection, followed by chemotherapy and radiation therapy. Unfortunately, virtually all patients will have tumor recurrence and die of this disease. While survival without treatment is approximately three months, survival following treatment is only 12 to 15 months. Less than 5% of people survive longer than five years. A cardinal metabolic characteristic of tumorigenesis is a metabolic shift in which glycolysis, even in the presence of adequate tissue oxygen, increases disproportionately relative to oxidative phosphorylation (OXPHOS), a phenomenon known as the Warburg effect. This glycolytic shift occurs in GBM and is mechanistically associated with post-translational inhibition of the mitochondrial pyruvate dehydrogenase complex (PDC), which normally catalyzes the rate-limiting step in the aerobic oxidation of glucose-derived pyruvate and lactate. PDC inhibition is due to transcriptional upregulation of one or more of four pyruvate dehydrogenase kinase isoforms (PDK 1-4) that inhibit PDC by reversible phosphorylation. Dichloroacetate (DCA), the prototypic PDK inhibitor, readily crosses the blood-brain barrier and represents an entirely new class of small molecule metabolic modulators that act in mitochondria to reset cellular homeostasis in various congenital and acquired metabolic disorders. Indeed, pharmacological inhibition of PDK in cancer cells by DCA restores PDC activity, reverses the Warburg effect and induces a caspase-mediated selective apoptosis of tumors. Extensive pre-clinical research and early clinical trials in patients with recurrent GBM and other brain tumors indicate that DCA may be a safe and uniquely effective metabolic therapy for GBM. DCA inhibits its own metabolism and its only clinically limiting toxicity is reversible peripheral neuropathy. To mitigate this adverse effect, we developed and validated a genotyping method for genetics-based dosing of DCA that dichotomizes subjects into fast and slow drug metabolizers, leading to safe, personalized DCA dosing.