Cancer cells undergo numerous adaptive processes to sustain rapid growth and survival. One notable mechanism is by rewiring their metabolism, most prominently through a phenomenon known as the Warburg Effect (WE). The WE is defined as an increase in glucose consumption and lactate secretion in the presence or absence of oxygen. Although the WE has been extensively studied, efforts to therapeutically target it have been largely unsuccessful due to the lack of obvious metabolic biomarkers and difficulties achieving full enzyme inhibition without inducing toxicity in normal tissue. Although targeted cancer therapies that use genetics have been successful, principles for selectively targeting tumor metabolism that also depend on environmental factors remain unknown. This limitation prompted the investigation to determine whether differential control in metabolism can be exploited for therapy. In this dissertation, I first determined whether therapeutic targeting of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), an enzyme that differentially regulates the Warburg Effect in cancer, can result in anti-tumor efficacy. Using comparative metabolomics, integrated pharmacogenomics, and systems biology, I found that koningic acid (KA), a natural product produced by the Trichoderma species, is a highly specific inhibitor of GAPDH. Notably, I determined that the quantitative extent of the Warburg Effect predicts response to KA in both cancer cells and tumors. Secondly, given the efficacy of KA in cancers specifically undergoing the Warburg Effect, I next used KA as a tool to determine whether there is a biological distinction between the Warburg Effect and glycolysis. I developed an evolved resistance model to KA and confirmed using metabolomics, stable isotope tracing, and a set of pharmacological approaches that glucose metabolism can exist in multiple states in the cell with distinct metabolic outputs. Lastly, I used therapeutic and genetic interventions to target 3-phosphoglycerate dehydrogenase (PHGDH), which diverts glucose flux into serine biosynthesis, and showed that this disrupts cancer cell growth through inhibition of de novo serine synthesis and downstream serine metabolism. Together, my findings contribute to a shifting paradigm in the current understanding of metabolic cancer therapy and show the potential use of metabolic factors as predictors and important determinants of therapeutic response.