The amylin analog pramlintide delays gastric emptying in type I diabetics. The effects of multiple doses of pramlintide and the mechanism of action in non-amylin-deficient humans are unknown. We investigated the effects of pramlintide on gastrointestinal and colonic transit and on the plasma pancreatic polypeptide response to the meal in a parallel-group dose-response study with subjects randomized to placebo, or 30 or 60 microg (tid, sc) of pramlintide. Pramlintide delayed gastric emptying [half-time (t(1/2)): 112 min (SE 8.7 min), 169 min (SE 12 min), or 177 min (SE 25 min) after placebo or 30- or 60-microg pramlintide treatment, respectively; P = 0.033]. Pramlintide did not significantly affect small bowel or colonic transit. Pancreatic polypeptide concentrations in the first postprandial hour were lower with pramlintide than with placebo (P<0.01 for drug effect). An inverse correlation was observed between mean pancreatic polypeptide concentrations in the first postprandial hour and gastric emptying t(1/2) [Spearman correlation coefficient (R(s)) = 0.48; P = 0.044]. Pramlintide at 30 and 60 microg delays gastric emptying in healthy humans without affecting small bowel or colonic transit. Vagal inhibition is a potential mechanism of the effects of pramlintide on gastric emptying.