Aims/hypothesis The cAMP-degrading phosphodiesterase 4 (PDE4) enzyme has recently been implicated in the regulation of glucagon-like peptide-1 (GLP-1), an incretin hormone with glucose-lowering properties. We investigated whether the PDE4 inhibitor roflumilast elevates GLP-1 levels in diabetic db/db mice and whether this elevation is accompanied by glucose-lowering effects. Methods Plasma GLP-1 was determined in db/db mice after single oral administration of roflumilast or its active metabolite roflumilast-N-oxide. Diabetes-relevant variables including HbA 1c , blood glucose, serum insulin, body weight, food and water intake, and pancreas morphology were determined in db/db mice treated daily for 28 days with roflumilast or roflumilast-N-oxide. Pharmacokinetic/pharmacodynamic analysis clarified the contribution of roflumilast vs its metabolite. In addition, the effect of roflumilast-N-oxide on insulin release was investigated in primary mouse islets.Results Single treatment of db/db mice with 10 mg/kg roflumilast or roflumilast-N-oxide enhanced plasma GLP-1 2.5-and fourfold, respectively. Chronic treatment of db/db mice with roflumilast or roflumilast-N-oxide at 3 mg/kg showed prevention of disease progression. Roflumilast-N-oxide abolished the increase in blood glucose, reduced the increment in HbA 1c by 50% and doubled fasted serum insulin compared with vehicle, concomitant with preservation of pancreatic islet morphology. Furthermore, roflumilast-N-oxide amplified forskolininduced insulin release in primary islets. Roflumilast-N-oxide showed stronger glucose-lowering effects than its parent compound, consistent with its greater effect on GLP-1 secretion and explainable by pharmacokinetic/pharmacodynamic modelling. Conclusions/interpretation Our results suggest that roflumilast and roflumilast-N-oxide delay the progression of diabetes in db/db mice through protection of pancreatic islet physiology potentially involving GLP-1 and insulin activities.

A robust population PK model for INFS was developed. The model enhances the understanding of fentanyl PK after INFS dosing in cancer patients with BTP, a population for whom real-life data would be very hard to obtain.

No abstract