GLP-1 is cleaved from the proglucagon molecule by the gut specific prohormone convertase enzymes 1 and 3.29-31 Two forms of GLP-1 are secreted, GLP-1(7-37) and GLP-1(7-36)amide. 32 The majority of circulating active GLP-1 appears to be GLP-1(7-36)amide.33,34 GLP-1 is stored in the L-cells of the ileum and colon, and released at mealtime in response to neurohormonal signals and the presence of food in the gut.35-38 GLP-1 exerts its effect on postprandial glucose concentrations through several mechanisms, including enhancing insulin secretion and suppressing postprandial glucagon secretion in a glucose-dependent manner.39,40 In addition GLP-1 slows the rate of gastric emptying, which is often paradoxically accelerated in patients with diabetes.39,41 GLP-1 also acts as a postprandial satiety signal through neurohormonal networks that signal the brain to suppress appetite and food intake.34,42-44 Furthermore, GLP-1 also has direct effects on the β-cells, as shown in studies done in animal models and cell lines, promoting cell proliferation and neogenesis, while preventing β-cell apoptosis.45 Additionally, GLP-1 can promote transformation of noninsulin-producing pancreatic cells into cells capable of synthesizing and secreting insulin.46-49
GLP-1 reportedly has effects on the liver, skeletal muscle as well as the adipose tissue, resulting in increased glycogen synthesis and peripheral insulin-stimulated glucose uptake.50-52 However, this effect is considered controversial, as there have been contradictory observations as well.53,54 GLP-1 receptors are also located in the heart, and have been associated with cardiovascular response to stress.54-57 Unfortunately in spite of all the potentially beneficial effects of GLP-1 on glucose homeostasis, the therapeutic potential of naturally occurring GLP-1 is limited, due to its extremely short half-life of less than 2 minutes largely due to the degradation by the enzyme dipeptidyl peptidase IV (DPP-IV).35,58-60
Administration of exogenous GLP-1, as a continuous subcutaneous or intravenous infusion, has been shown to increase insulin secretion, and normalize both fasting and postprandial blood glucose concentrations. In one of the earliest studies with GLP-1, Gutniak et al. showed that a continuous intravenous infusion of GLP-1 resulted in a reduction in the amount of insulin needed to maintain isoglycemia in patients with either type 1 or type 2 diabetes.61 Few years later, Zander et al., in a non-randomized parallel group study, showed that six weeks of continuous subcutaneous infusion of GLP-1 not only reduced fasting and postprandial plasma glucose concentrations, but also lowered glycosylated hemoglobin (A1C) by 1.3%, and induced up to a 2 kg reduction in body weight.39 Only a few randomized clinical trials have been reported with continuous infusion of GLP-1, and results from four of these randomized studies are summarized in table 1.62-65
Juntti-Berggren et al. reported the first randomized trial with the addition of subcutaneous infusion of GLP-1 to regular insulin therapy, and compared it with insulin therapy alone in 12 patients with type 2 diabetes.62 After initial intensive insulin therapy for 1 week, 8 patients were randomized to receive a subcutaneous infusion of GLP-1 along with injections of regular insulin with meals, and NPH insulin at bedtime, whereas 4 patients were randomized to regular insulin with meals and NPH insulin at bedtime. A significant reduction in postprandial glucose concentration was observed after one week of treatment in patients receiving subcutaneous infusion of GLP-1 added to insulin therapy, compared to insulin therapy alone. Furthermore, an additive lipid-lowering effect was also observed in the GLP-1 infused patients.62
In 2001, Larsen et al. reported a randomized trial with continuous intravenous infusion of GLP-1 versus placebo infusion, in 40 hospitalized patients with type 2 diabetes, poorly controlled with sulfonylurea treatment.63 Patients received an infusion of GLP-1 at 4 or 8 ng/kg/min for 16 or 24 hours, versus placebo, for 7 days. Patients infused with GLP-1 showed a significant, dose-dependent, reduction in mean 24-hour plasma glucose concentrations compared with placebo. The fasting and nocturnal plasma glucose concentration was higher in the 16-h GLP-1 infused patients compared to the 24-h GLP-1 infused patients. GLP-1 infusion was also associated with a dose related increase in nausea, headache, and vomiting.63
In another randomized study published in 2003 (Meneilly et al.), 16 elderly, insulin-naïve patients with diabetes were divided in two groups of 8 patients each, and given either a continuous subcutaneous infusion of GLP-1 at an initial dose of 100 pmol/kg/h, titrated up to a maximum dose of 120 pmol/kg/h for 12 weeks, compared to controls treated with conventional therapy.64 Concomitant treatment with glucose lowering agents was discontinued for 1 week prior to study treatment in the group of patients infused with GLP-1, while the control group continued with their usual glucose lowering therapy without any dose changes. GLP-1 infusion was well tolerated, without significant side effects. Hypoglycemia occurred only once in the GLP-1 treatment group, while hypoglycemic events were frequent in the control group (87 events). No significant changes in A1C were observed, despite a discontinuation of oral hypoglycaemic medications in the GLP-1 treated group. There was an enhanced glucose-induced insulin secretion, as well an in insulin-mediated glucose disposal, in the GLP-1 infused group compared to controls.64
In 2004, Zander et al. reported a randomized cross-over trial that evaluated the additive effect of continuous infusion of GLP-1 and a TZD, pioglitazone, in patients with type 2 diabetes.65 Eight patients with type 2 diabetes were given a saline infusion, continuous infusion of GLP-1 (4.8 pmol/kg/min), pioglitazone treatment (30 mg), and a combination treatment with GLP-1 infusion and pioglitazone, in a random order. During the studies with either saline or GLP-1 only infusions, the patients discontinued their normal medications for three weeks, with a 2-day washout period between the saline or GLP-1 infusions. During the studies with pioglitazone with or without GLP-1 infusion, all patients received 30 mg of pioglitazone for 12 weeks and then were randomly assigned to a 48-hour monotherapy period with either pioglitazone or a combination of GLP-1 with pioglitazone. A significant drop in fasting plasma glucose (FPG) concentrations in the GLP-1 only, pioglitazone only, and the combination of pioglitazone with GLP-1 group were observed compared to saline infusion group. FPG was lower in combination therapy compared to monotherapy with either agent. Mean insulin concentrations were significantly higher with GLP-1 compared with pioglitazone. Glucagon concentrations were reduced in both GLP-1 treated arms compared to saline and pioglitazone treatments, and a sensation of appetite reduction was also noted in the GLP-1 treatment arms. 65
All these studies demonstrated the clinical potential of GLP-1 therapy in patients with type 2 diabetes. Unfortunately, for reasons of practicality, intravenous or subcutaneous continuous infusion of GLP-1 is not practical, as a long-term treatment of diabetes.
Two new classes of agents are being developed to take advantage of the unique glucoregulatory potential of GLP-1:
-
The incretin mimetics, such as exenatide (synthetic exendin-4, a peptide resistant to proteolytic cleavage by DPP-IV) or derivatives of GLP-1 (created by chemical modification of the native hormone, to render them resistant to rapid proteolytic cleavage by DPP-IV), targeting the pharmacological effects, and insufficient secretion of GLP-1 seen in type 2 diabetes; and
-
The DPP-IV inhibitors, compounds that increase the concentration of endogenous incretins, including GLP-1, by limiting the proteolytic cleavage by DPP-IV.
Incretin mimetics are compounds that mimic several of the physiologic effects of GLP-1, such as enhancement of glucose-dependent insulin secretion, suppression of postprandial glucagon secretion, slowing of gastric emptying, reduction of food intake and body weight in human studies. Incretin mimetics have also been shown to promote β-cell proliferation and neogenesis in cell lines and animal models.
Exenatide, is currently the only incretin mimetic agent that has received regulatory approval in the U.S. as an adjunctive treatment for type 2 diabetes. It is a synthetic version of exendin-4, a peptide originally identified in the saliva of the Gila monster (Heloderma suspectum), and secreted upon ingestion of nutrients.66,67 While exendin-4 has many of the properties of GLP-1, the two peptides are unique and are transcribed from distinct genes in the Gila monster. In humans and animal studies, exenatide enhances glucose-dependent insulin secretion, suppresses the elevated postprandial glucagon levels seen in type 2 diabetes, and slows the rate of gastric emptying40,68,69 (which can be paradoxically accelerated in many people with diabetes).41 In addition, both exenatide and GLP-1 have been reported to promote β-cell proliferation and neogenesis from ductal precursor in animal models.34,46-49,70 Data obtained in animal models also indicate that exenatide reduces food intake, promotes weight loss, and partially due to this weight loss has an insulin sensitizing effect.71-73 Moreover, exenatide has been shown to reduce food intake in healthy humans.74 Exenatide does not appear to be significantly degraded in the circulation, and is primarily cleared by the kidneys.75 After a subcutaneous injection, exenatide plasma concentrations increase in a dose-dependent manner (tmax of approximately 2 hours) and decay in a linear manner (t1/2 of 3.3 to 4.0 hours).75,76
Earlier short term clinical trials in patients with type 2 diabetes demonstrated that exenatide acutely lowered both fasting and postprandial plasma glucose excursion. Furthermore, exenatide administration suppressed fasting and postprandial glucose largely due to decreased glucagon concentrations, and enhanced glucose-dependent insulin secretion.69,75 Exenatide infusion has also been demonstrated to fully restore acute insulin secretory response enhancing both 1st and 2nd phase insulin secretion in patients with type 2 diabetes.77 While exenatide is known to suppress glucagon secretion following food ingestion, it does not impair hypoglycemia-induced glucagon secretion.78
Three large randomized, placebo-controlled clinical trials with subcutaneous administration of exenatide (bid), conducted in subjects with type 2 diabetes who were unable to achieve glycemic control with sulfonylureas and/or metformin, demonstrated that exenatide treatment resulted in mean reductions in A1C from baseline of ~1% accompanied by an average weight loss of 2 to 3 kg in those treated with 10 mcg exenatide BID. 79-81
Mild-to-moderate nausea was the most commonly reported adverse event associated with exenatide, occurring with greatest frequency upon initiation, and generally subsiding with continued exposure to exenatide.79-81 Few new episodes of nausea were reported after 4 weeks of treatment. Furthermore, stepwise dose escalation was found to reduce the incidence of nausea and vomiting.82 There was no increase in the risk of hypoglycemia when exenatide was administered in combination with metformin, most likely due to the glucose-dependent actions of exenatide. However, the risk of hypoglycemia increased when exenatide was administered with a sulfonylurea, and this risk was greater when A1C was closer to normal, and the dose of sulfonylurea was not concomitantly reduced. Approximately 45% of patients in these studies developed anti-exenatide antibodies; however, the presence of antibodies, and/or the magnitude of the antibody titer, were not associated with an individual patient’s magnitude of glycemic improvement, nor was there an association with incidence of adverse events.79-81
In a phase 3 non-inferiority study, conducted in patients with type 2 diabetes inadequately controlled with metformin or sulfonylureas, comparing exenatide with basal insulin (insulin glargine), similar reductions in A1C were observed with exenatide (-1.0%) and insulin glargine(-1.1%).83 After 26 weeks, patients treated with exenatide reduced body weight (-2.3 kg), while patients treated with insulin glargine gained body weight (+1.8 kg). Comparison of self monitored blood glucose profiles showed improvements in postprandial glucose control and in 24-hour glucose profile with exenatide treatment, whereas treatment with insulin glargine mainly lowered fasting plasma glucose concentrations, but did not alter the magnitude of postprandial glucose excursions throughout the day. Results of this study indicate that exenatide may play a role as an alternative to basal insulin in patients with type 2 diabetes failing to achieve treatment targets on combination oral medications. 83
Liraglutide (NN2211, Novo Nordisk, Copenhagen, Denmark) is a GLP-1 analog acylated with a fatty acid, thereby allowing binding to circulating albumin and rendering it resistant to DPP-IV.86 Liraglutide has been reported to have a circulating half-life of 13 hours, thus facilitating once daily subcutaneous injection in patients with type 2 diabetes.86 Published early phase clinical trials suggest that liraglutide improves fasting and postprandial blood glucose, restores glucose-dependent insulin secretion, delays postprandial gastric emptying, and reduces plasma glucagon concentrations in patients with type 2 diabetes.87-92 Liraglutide has been also been reported to improve β-cell function as measured by HOMA-B analysis and proinsulin/insulin ratio.87,88 In a 12-week comparison study, comparing liraglutide (doses ranging from 0.045 mg/day to 0.75 mg/day) to metformin (1000 mg bid), it was reported that liraglutide at doses ≥0.45 mg/day achieved a similar level of glycemic control to metformin, measured by A1C and fasting plasma glucose.93 Modest reductions in weight have been reported in some studies with liraglutide, however, its full effects on body weight, as well as food intake, remain to be fully understood.89, 91-93 Treatment emergent adverse events reported with liraglutide were mostly gastrointestinal in nature, similar to those observed with exenatide and continuous infusion of GLP-1, suggesting that short term side-effects are a characteristic of exogenous treatment with incretins and/or the incretin mimetic class of agents.39,89,91-93