To investigate the incretin axis in people with cystic fibrosis.Adults with cystic fibrosis-related diabetes, cystic fibrosis without diabetes, and controls (adults without cystic fibrosis and without diabetes) underwent an oral glucose tolerance test and then a closely matched isoglycaemic i.v. glucose infusion. On each occasion, glucose, insulin, C-peptide, total and active glucagon-like peptide-1 and gastric inhibitory polypeptide responses were recorded and incremental areas under curves were calculated for 60 and 240 min.Five adults with cystic fibrosis-related diabetes, six with cystic fibrosis without diabetes and six controls, matched for age and BMI, completed the study. Glucose during oral glucose tolerance test closely matched those during isoglycaemic i.v. glucose infusion. The calculated incretin effect was similar in the control group and the cystic fibrosis without diabetes group (28% and 29%, respectively), but was lost in the cystic fibrosis-related diabetes group (cystic fibrosis-related diabetes vs control group: -6% vs 28%; p=0.03). No hyposecretion of glucagon-like peptide-1 or gastric inhibitory polypeptide was observed; conversely, 60-min incremental area under the curve for total glucagon-like peptide-1 was significantly higher in the cystic fibrosis-related diabetes group than in the control group [1070.4 (254.7) vs 694.97 (308.1); p=0.03].The incretin effect was lost in cystic fibrosis-related diabetes despite adequate secretion of the incretin hormones. These data support the concept that reduced incretin hormone insulinotropic activity contributes significantly to postprandial hyperglycaemia in cystic fibrosis-related diabetes.