Nutrition in advanced cancer: anorexia as an outcome variable and target of therapy.

The endpoints used as outcome variables in clinical cancer treatment trials, including nutrition intervention studies, should contain items that are meaningful to the patient. Variables to consider are appetite, food intake, physical performance, psychological and social functioning, response to cancer therapies, survival time, nutrition status, associated morbidity, and costs. Ideally, the design and conduct of nutrition trials should be carried out by a multidisciplinary team comprising medical oncologists, physician specialists in nutrition, dietitians, and social scientists. Anorexia has not been a focus of nutrition support trials in the past partly because of the lack of effective strategies to reverse it. Anorexia is one important cause of cancer starvation, and it also causes patient discomfort. This paper describes outcome variables that include patient derived subjective factors such as anorexia, and outlines new strategies to reverse anorexia. Pharmacologic strategies tested to reverse anorexia include corticosteroids, anabolic steroids, cyproheptadine, hydrazine sulfate, cannabinoids, and megestrol acetate. Of these, only the latter has been consistently well tolerated and effective, with significant improvements in appetite and food intake demonstrated in large-scale, randomized, controlled trials involving more than 600 cancer patients. Dose-response studies have demonstrated increasing efficacy with increasing doses of megestrol acetate from 160 to 800 mg/day. Doses in excess of 800 mg/day are not currently recommended. The mechanisms of action of megestrol acetate involve both behavioral and metabolic effects, and its impact on energy expenditure, appetite, body composition, endocrine function, and lipid metabolism is the subject of ongoing research.(ABSTRACT TRUNCATED AT 250 WORDS)