Clinical management of anorexia and cachexia in patients with advanced cancer.

Cachexia occurs in the majority of cancer patients before death. It is the result of major metabolic changes produced by tumor-released substances as well as by cytokines and some endogenous peptides. The most significant clinical manifestation is profound anorexia. Aggressive parenteral nutrition has not been able to increase patient survival or produce any significant symptomatic improvement. Recent research, therefore, has focused on drugs that might result in symptomatic improvement, even if no significant nutritional changes are detected. Corticosteroids have been shown to increase appetite for a brief period of time, but they do not appear to improve caloric intake or nutritional status. In addition to appetite stimulation, corticosteroids also improve a number of other symptoms transiently. Progestational drugs have been found in a number of studies to increase appetite, caloric intake, and nutritional status. The most effective type and dose of progestational drugs have not been clearly established. Cyproheptadine, hydrazine sulfate, and cannabinoids have all been suggested to have beneficial effects on appetite; their effectiveness, however, needs to be confirmed in prospective, controlled trials. Some of these trials are currently under way. Current data suggest that megestrol acetate or other progestational agents could be useful--because of effects on not only appetite but also overall nutritional status--in patients who have profound anorexia as the main manifestation of cachexia, provided expected survival can be measured in weeks or months. In patients with shorter expected survival or those who have problems tolerating progestational drugs, a brief course of corticosteroids may provide short-term symptomatic effects. Future studies should focus on (1) improving understanding of both the pathophysiology of cancer cachexia and the interaction of some of the major syndromes of terminal cancer--e.g., pain, cachexia, and cognitive failure--and (2) characterizing the symptomatic effects of different drugs more completely.