History and development of incretin therapy
The incretin-based therapies were developed as a consequence of painstaking investigation of two unsolved problems in diabetes. The first was the nature of the incretin effect, whereby oral glucose stimulates a greater insulin response than intravenous glucose, and the second was the function of various glucagon-like molecules secreted by cells lining the intestine. This led to the discovery that the gut peptides GLP-1 and GIP are released in response to oral nutrients and promote insulin secretion by the pancreas, thus accounting for the incretin effect. GLP-1 levels are deficient in type 2 diabetes, suggesting therapeutic possibilities; the main obstacle was that GLP-1 has a very short (1-2 minute) survival in the circulation. Two strategies have been used to overcome this. The first is development of GLP-1 receptor agonists which are resistant to degradation, and the second is inhibition of DPP4, thereby enhancing delivery of the naturally-produced hormone. Both GLP-1 agonists and DPP4 inhibitors are in wide used, but concerns have been raised as to the pleiotropic effects of GLP-1, whose actions on the exocrine pancreas might for example explain why there have been safety alerts for acute pancreatitis for all agents in this class.
The "incretin effect" describes a long-recognised phenomenon whereby oral glucose provokes a greater insulin secretory response than the same amount of insulin injected into a vein. This strongly suggested a mediating factor(s) which augments insulin secretion, but the nature of the stimulus was unknown.
It was also known that administration of an extract from the upper intestine could produce a fall in blood glucose, and In 1932 La Barre proposed the name "incretin" for the presumed hormone. Here matters rested until the identification of the gut peptides in the 1960s. Glucagon itself stimulates insulin secretion, but the role of a family of immunoreactive glucagon-related peptides in cells lining the intestine remained conjectural.
This family of glucagon-related molecules includes the 69 amino-acid glicentin and a truncated form known as oxyntomodulin. Gastric inhibitory peptide, identified in 1970, was renamed glucose-dependent insulinotrophic peptide (also GIP) when its ability to stimulate insulin was demonstrated, but evidence emerged that GIP alone could not be wholly responsible for the incretin effect.
Proglucagon was cloned and sequenced in 1983, with the surprising information that the molecule contained two additional glucagon-like sequences, coding for two glucagon-like peptides, subsequently known as GLP-1 and GLP-2. GLP-2 is an important gastrointestinal growth factor, and the 7-37 sequence within GLP-1 was found to be the most powerful insulinotropic agent yet known, thus explaining the incretin effect.
GLP-1 as a therapy
Several investigators then undertook the experiment of infusing GLP-1 into volunteers. Zander et al. carried out a clinical study in which native GLP-1 or placebo (saline) was administered as a continuous subcutaneous infusion (using insulin pumps) for 6 weeks to a group of patients with type 2 diabetes. The patients were evaluated before, after 1 week and after 6 weeks of treatment. No changes were observed in the placebo group, whereas in the GLP-1-treated group, fasting and average plasma glucose concentrations were lowered by approximately 4-6 mM, HbA1c decreased by 1.3% and the patients had a weight loss of approximately 2 kg after 6 weeks. There were very few side effects and no differences between saline and GLP-1 treated patients in this respect.
High infusion rates induced nausea and vomiting, but lower rates reduced glucose effectively with minimal side effects for up to 6 weeks of therapy. Studies published in 1992 confirmed efficacy in volunteer patients with type 2 diabetes . Long term IV administration is clearly impractical, but these studies provided proof of concept for the development of stable analogues for clinical use.
Development of GLP-1 receptor agonists
The short-half-life of GLP-1 in the circulation (1-2 minutes) is due to its speedy inactivation by the enzyme Dipeptidyl Peptidase 4 (DPP4) resulting in a plasma clearance amounting to 2-3 times cardiac output. Amino-acid substitution overcame this problem but did not greatly prolong the half-life of some analogues since native GLP-1 is actively eliminated by the kidney.
A number of biologically active peptides were discovered by analysis of substances secreted by amphibians in their skin or saliva. These included peptides closely related to GLP-1, notably Exendin-4, which is present in the saliva of the Gila monster, Heloderma suspectum. Exendin-4 is resistant to degradation by DPP4 and is not actively excreted by the kidney. This, under the name exenatide (Byetta), was approved by the FDA in April 2005, and became the first clinically useful GLP-1 receptor agonist. Exenatide has a plasma half-life of 30 minutes, and can therefore be administered by twice daily injection. A very long-acting variant (Bydureon) permitting once-weekly injection was subsequently developed.
A second GLP-1 receptor agonist named liraglutide (Victoza) is resistant to DPP4 inactivation and renal elimination by virtue of a short fatty acid chain attached to the molecule, and has a plasma half-life of 12 hours, allowing once daily injection. Other GLP-1 analogues are in various stages of development.
The DPP4 Inhibitors
By 1993, in vitro studies had shown that both GIP and GLP-1 are N-terminally degraded to an inactive metabolite by DPP4, and the same phenomenon was soon documented in vivo. DPP4 is widely distributed in the capillary endothelium, renal tubules and hepatocytes, and its location in the endothelium of capillaries adjacent to the GLP-1 containing L cells of the intestine showed that a proportion of the secreted peptide is degraded before it even reaches the circulation. By 1995 DPP4 inhibition was already under consideration for the treatment of diabetes.
Clinical proof-of concept was provided in 2001 by Ahren et al. when they demonstrated significant glucose-lowering effect and reduction of fasting blood glucose and HbA1c upon oral administration of an early inhibitor from Novartis in a 4-week study in patients with type 2 diabetes. Subsequent studies with a more long-acting inhibitor (vildagliptin) documented sustained effects on HbA1c (about 1% reduction compared to placebo) for 52 weeks when the inhibitor was added to existing metformin treatment.
^ Holst JJ. Glucagon-like peptide 1: from extract to agent. The Claude Bernard Lecture 2005. Diabetologia 2006;
^ Zander M, Madsbad S, Madsen JL, Holst JJ. Effect of 6-week course of glucagon-like peptide 1 on glycaemic control, insulin sensitivity, and beta-cell function in type 2 a parallel-group study. Lancet. 2002 Mar 9;359(9309):824–30.
^ Gutniak M et al. Antidiabetogenic effect of glucagon-like peptide -1 (7-37) amide in normal subjects and patients with diabetes mellitus. N Engl J Med 1992;326:1316-22
^ Nathan DM et al. Insulinotropic action of glucagonlike peptide-1 (7-37) in diabetic and non-diabetic subjects. Diabetes Care 1992;15:270-6
^ Vilsbøll T, Holst JJ. Incretins, insulin secretion and Type 2 diabetes mellitus. Diabetologia. 2004 Mar;47(3):357–66.
^ Holst JJ. The physiology of glucagon-like peptide 1. Physiol. Rev. 2007 Oct;87(4):1409–39.
^ Deacon CF, Holst JJ. Dipeptidyl peptidase IV inhibition as an approach to the treatment and prevention of type 2 diabetes: a historical perspective. BBRC 2002;294:1-4