Pre-clinical and clinical studies have investigated the role of a dysregulated metabolism in the sustainability of tumor initiation and progression. One of the most familiar metabolic alterations encountered in several types of cancers is the upregulation of glycolysis, which is also maintained in conditions of normal oxygen tension (aerobic glycolysis, Warburg effect) while oxidative phosphorylation is apparently reduced. As a result, cancer cells convert most incoming glucose to lactate. Although more rapid, adenosine triphosphate (ATP) production by glycolysis is less efficient in terms of ATP generated per unit of glucose consumed than oxidative phosphorylation. The consequence is that tumor cells require an abnormally higher rate of glucose compared to the normal counterpart. New evidence shows that other metabolic substrates such as glutamine may also have an important role in cancer metabolism. Ketogenic diet (KD) replaces all but non-starchy vegetable carbohydrates with low to moderate amounts of proteins and high amounts of monounsaturated and polyunsaturated fats. The rationale of KD is valid both because it lowers carbohydrate uptake possibly leading to cancer cell starvation and apoptosis and, at the same time, increases the levels of ketone bodies available for energy production in normal cells but not in cancer cells which have an allegedly downregulated oxidative phosphorylation. For this reason, several authors speculate on the possibility to evaluate KD as a novel approach in the treatment of cancer. In this review we will assess the data supporting the use of such alimentary regimen and its impact on tumor development and progression.