Great discrepancies exist in the reported prevalence of altered energy metabolism (hypo- or hypermetabolism) in cancer patients, which is likely due to the vast array of phenomena that can affect energy expenditure in these patients. The purpose of this review was to critically evaluate key determinants of energy expenditure in cancer and the relevance for clinical practice. Resting energy expenditure (REE) is the largest and most commonly measured component of total energy expenditure. In addition to the energetic demand of the tumor itself, REE may be increased due to changes in inflammation, body composition and brown adipose tissue activation. Energy expenditure from physical activity is often lower in cancer compared with healthy populations, and there is evidence to suggest that the thermic effect of food might also be blunted and affected by cancer therapy. Although accurate assessment of energy metabolism is a cornerstone of adequate nutritional therapy, prediction methods often do not capture the true energy expenditure of most cancer patients. In fact, limits of agreement of prediction equations may range from 40% below to 30% above measured REE. Such variability highlights the need for a more comprehensive understanding of energy expenditure in cancer and the value of accurately assessing the energy needs of these patients.European Journal of Clinical Nutrition advance online publication, 8 June 2016; doi:10.1038/ejcn.2016.96.