Understanding the Incretin Effect

Abstract

It is well known that the incretin effect contributes as much as half of the insulin secretory response to oral glucose load and that this effect is reduced along with worsening glucose tolerance in those with type 2 diabetes. Gastric inhibitory polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) are the two primary incretin hormones secreted from the intestine after the ingestion of glucose and other nutrients (1–3). In type 2 diabetes patients, the insulinotropic action of GIP is diminished, whereas that of GLP-1 is substantially preserved, although secretion of the latter appears to be diminished (4, 5). Thus, methods of enhancement of circulating concentrations of GLP-1 and/or GLP-1 receptor signaling have been established as therapeutic strategies in type 2 diabetes. GLP-1 receptor agonists and dipeptidyl-peptidase-4 (DPP-4) inhibitors are now widely and successfully used for this condition. DPP-4 inhibitors augment endogenous active GIP and GLP-1 concentrations. In this therapeutic strategy, DPP-4 inhibitors are expected to enhance the incretin effect by raising plasma concentrations of active incretin hormones. The study in this issue of JCEM by Vardarli et al. (6) assesses the incretin effect after treatment with the DPP-4 inhibitor vildagliptin in patients with type 2 diabetes. Subjects were recruited into a double-blind, two-way crossover study (order randomized) to evaluate the incretin effect under treatment with vildagliptin (100 mg once daily) or placebo. Under both treatments, an oral 75-g glucose load was performed on d 12, and on d 13 isoglycemic glucose infusion was administered to reproduce glucose excursions after oral glucose load. The glucose excursions after oral and iv glucose administration matched well. Vildagliptin reduced the integrated incremental glucose concentration during oral glucose load by 15.8%. Moreover, as expected, the ratio of integrated incremental responses of insulin, C-peptide, and insulin secretion rates over glycemic excursions was significantly enhanced under vildagliptin treatment. Surprisingly, however, the numerical incretin effect (based on insulin, C-peptide concentrations, and insulin secretion rates measured after oral and iv glucose testing) did not differ between vildagliptin and placebo. To further elucidate these findings, the authors show that the insulin secretory response to glucose is enhanced with vildagliptin treatment not only by oral load but also after iv load, contributing to the numerical similarity of the calculated incretin effects. This raises several questions regarding calculation of the incretin effect and its purported physiological basis. Oral glucose load was shown to produce a greater insulin response than iv injection of glucose in the 1960s (7, 8), and the incretin effect was defined as the difference in -cell secretory response to oral or iv glucose stimuli (9–11). The calculation was based on the assumption that after oral glucose administration, -cells are stimulated by the elevated plasma glucose as well as by the additional stimulation of incretin hormones not activated during iv glucose loading. Vardarli et al. (6) have now demonstrated that there is a minor rise in the intact GLP-1 concentration during iv glucose administrationafter vildagliptin exposure. Furthermore, this rise may underlie the enhanced insulin secretory response after iv glucose load that results in the substantial insulin response used in calculating the incretin effect. Vildagliptin was found not to change the numerical incretin effect compared with placebo. The authors discuss the relationship between the amount of iv glucose administration needed to obtain isoglycemia and the incretin effect calculated by insulin, C-peptide concentrations, and insulin secretion rates. The more that the incretin effect is reduced, the more glucose needs to be administered iv to reproduce the glucose ex-