Gastrointestinal symptoms such as abdominal fullness, nausea and vomiting commonly occur in the advanced cancer population and are a source of major morbidity. These are now recognized to occur even early in the course of disease, prior to, and independent of, chemotherapy administration. Several preliminary studies have suggested such symptoms may be related to a specific pathophysiologic defect, delayed gastric emptying. In the proposed study, quantitative measures of gastrointestinal symptoms will be determined to facilitate identification of future cancer control interventions in this area. Fifty male patients with unresectable carcinoma (non-small cell of the, dropharyngeal, and prostate) with no prior chemotherapy, not currently receiving narcotic analgesia or antiemetics will have three-day inpatient evaluations. Clinical complaints of abdominal fullness, nausea or vomiting will be elicited along with an expanded quantitative scale assessing these symptoms. Baseline determination of caloric intake, anthropometrics and peripheral blood betaendorphin levels will be performed. Patients will then randomly receive one I.V. injection of either placebo, beta-endendorphin or morphine sulphate on 3 consecutive days followed by quantitation of solid gastric emptying, satiety, pain, metabolic variables (glucose, insulin and glucagon) and food consumption. Patients with a combined symptom score for abdominal fullness, nausea and vomiting of 3 or more will then be given daily metoclopramide and undergo repeat inpatient evaluation after four weeks of therapy. As a result, the prevalence of quantitive abnormalities in gastric emptying in this population will be determined and correlated with satiety, pain, metabolic variables and dietary intake (amount and caloric density). In addition, the relationships among beta endorphin, morphine sulphate, metoclopramide, gastric emptying, and other study parameters will be explored. Taken together, this information will permit development of rational and potentially improved intervention strategies to reduce gastrointestinal morbidity in patients with unresectable cancer.