In prospective cohort studies an association between fat consumption and breast and colon cancer was not confirmed. Alcoholic beverages enhance the risk of cancer of the oral cavity, pharynx, larynx and oesophagus, while the relation is less well established for cancer in the stomach, pancreas, colon, rectum and breast. Consumption of vegetables and fruits reduces the risk of lung, oesophagus, stomach and pancreatic cancer. Which food components are responsible for this observation is unclear, but beta-carotene does not appear to be responsible. Clinical depletion in cancer patients is a result of metabolic disturbances as a consequence of the cancer and has negative effects on the course of the disease and on the treatment. Altered eating behaviour, food aversion, decreased food intake and loss of body weight may indicate the presence of clinical depletion. Positive effects of (artificial) nutrition as an adjunct to chemoor radiation therapy are very limited, larger but still limited as an adjunct to surgical therapy. It is not justified to allow insufficient food intake to continue during treatment of cancer; early (artificial) nutritional intervention is warranted in those cases. Food requirements of cancer patients and healthy persons apparently do not differ. A well balanced normal diet providing adequate amounts of energy, proteins, vegetables and fruits is therefore advised. Alternative diets, such as the Moerman or Houtsmuller diet, may provide the necessary energy and other specific food components as long as animal proteins (milk, meat, fish) are not omitted completely. There is no evidence that these diets will alter the disease course in a positive way; patients may benefit from the idea that they are contributing to their own health, however. If the physician doubts whether the alternative diet is used voluntarily, or if there are negative effects (social isolation, financial burden) he or she should advise against its continuation.