Overview of Glucagon-Like Peptide-1 Receptor Agonists for the Treatment of Patients with Type 2 Diabetes

Abstract

Background

It is estimated that 29.1 million people or 9.3% of the US population have diabetes, which contributes to considerable medical and financial burden. Type 2 diabetes mellitus is characterized by insulin resistance and insulin secretion impairment leading to hyperglycemia. The presence of insulin resistance is strongly correlated with obesity.

Objective

This article reviews the available glucagon-like peptide-1 (GLP-1) receptor agonists and their role in the management of patients with diabetes, to help guide the selection of the most suitable agent for the individualized treatment of patients with type 2 diabetes.

Discussion

This article reviews the evidence from phase 3 clinical trials for each of the 5 GLP-1 receptor agonists by comparing them against one another and with other existing therapies, including metformin, dipeptidyl peptidase-4 (DPP-4) inhibitors, and sulfonylureas. Incretin-based therapies have emerged as attractive agents for the treatment of type 2 diabetes. They target the GLP-1 hormone, which is partly responsible for insulin release and for attenuating hyperglycemia during meals (ie, the incretin effect). The 2 classes of incretin-based therapy currently available are GLP-1 receptor agonists and DPP-4 inhibitors, which prevent the breakdown of GLP-1. Both classes are attractive options, given their glucose-lowering effects without the adverse effects of hypoglycemia and weight gain. The different mechanisms of action of these therapies result in generally greater efficacy with GLP-1 receptor agonists, albeit at the expense of slightly increased gastrointestinal symptoms. These agents exert their effects by improving glucose-dependent insulin release, suppressing glucagon release, suppressing hepatic glucose output, and decreasing the rate of gastric emptying, thereby reducing appetite. Currently, 5 GLP-1 receptor agonists are available, including exenatide, liraglutide, albiglutide, dulaglutide, and lixisenatide; semaglutide may soon become available as the newest agent. With the exception of the investigational oral semaglutide, which has shown promising results, the other 5 agents are administered as subcutaneous injections, at different dosing intervals.

Conclusion

Currently, 5 GLP-1 receptor agonists are available for use in the United States. Although they are all in the same drug class, some significant differences exist among the various GLP-1 receptor agonists. The choice of a specific GLP-1 receptor agonist will depend on the patient preferences, potential adverse effects, and cost.

It is estimated that 29.1 million people or 9.3% of the US population have diabetes, which contributes to considerable medical and financial burden.1 Type 2 diabetes mellitus is characterized by insulin resistance, and by some impairment in insulin secretion leading to hyperglycemia. The presence of insulin resistance is strongly correlated with obesity.1

A significant challenge in the treatment of diabetes is avoiding the development of hypoglycemia, particularly with sulfonylureas and insulin. Complications of hypoglycemia include unconsciousness, brain damage, and even death if untreated.1 Another adverse effect associated with the treatment of diabetes is weight gain, which occurs with most antidiabetes agents, including sulfonylurea, insulin, and thiazolidinediones.2 Because obesity is closely linked to diabetes, these agents' efficacy in treating diabetes become partly limited because of their link to weight gain.2

Cost is also an important consideration when selecting among the many antidiabetes medications. Table 1 compares the costs of diabetic agents. Glucagon-like peptide (GLP)-1 receptor agonists are generally the most expensive agents. Of note, the cost of Soliqua 100/33 (insulin glargine and lixisenatide injection), which is a combination of insulin glargine and a GLP-1 receptor agonist, is comparable to other GLP-1 receptor agonists that are given as monotherapy. The cost of individual antidiabetes agents may vary depending on insurance coverage, although coupons are often available for a significant cost reduction. Although the cost of diabetes medications (and associated supplies) is significant (12% of the overall cost of treating diagnosed diabetes), the costs of treating the complications of diabetes (18%) and of diabetes-related inpatient care (43%) are even greater.3 Therefore, it is more cost-effective for patients when their diabetes is appropriately controlled with medications, as necessary.

Table 1.

Costs of Diabetes Medications, by Class

Drug/drug class	Cost of 30-day supply, range, $
Metformin	5–9
Insulin	145–650
Sulfonylureas	9–15
Pioglitazone	12–17
DPP-4 inhibitors	173–397
SGLT-2 inhibitors	432–443
GLP-1 receptor agonists	492–684

DPP-4 indicates dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1; SGLT-2, sodium-glucose cotransporter-2.

Source: Cost obtained from GoodRx based on 30-day supply.

The Rationale for GLP-1 Receptor Agonists

The pathology of type 2 diabetes involves inherited traits and environmental factors. The vast majority of patients with type 2 diabetes have a genetic risk for insulin resistance; however, the risk for diabetes also worsens with increasing age and weight.2 Obese patients have more adipocytes, which release leptin, adiponectin, tumor necrosis factor–alpha, and resistin, and these hormones are thought to further contribute to insulin resistance.

During periods of hyperglycemia, there is an increase in glucose transport into beta-cells of the pancreas, which leads to insulin secretion. It is well-recognized that continued poor control of hyperglycemia leads to a decline in beta-cell function, which is likely a result of decreased insulin gene expression and decreased production of insulin. Therefore, it is important that lifestyle changes and treatments are implemented to maintain euglycemia. Uncontrolled diabetes will eventually lead to complications, such as microvascular disease (ie, retinopathy, nephropathy, and neuropathy), and cardiovascular (CV) events and hypertension.

Insulin secretion occurs in 2 phases. The first phase occurs after a meal, manifested as an immediate rise in insulin lasting approximately 10 minutes. This is followed by a second phase, in which insulin is released more slowly for a prolonged period. Patients with type 2 diabetes have markedly reduced first-phase insulin secretion, which likely explains why the majority have persistently elevated postprandial glucose concentrations despite relatively normal fasting glucose levels.4,5 The beta-cells in the pancreas respond to this by increasing second-phase insulin response.6 However, prolonged elevation of insulin from persistent hyperglycemia leads to beta-cell toxicity and ultimately contributes to insulin resistance.7 Interventions that mimic normal first-phase insulin secretion, rather than the second phase, have been correlated with improved glucose tolerance.8

KEY POINTS

• ▸This article reviews the available glucagon-like peptide-1 (GLP-1) receptor agonists and their role in the management of patients with diabetes.
• ▸Clinical trials demonstrate the superiority of GLP-1 receptor agonists to other antidiabetes drugs in HbA_1c reduction, blood pressure reduction, and weight loss, without hypoglycemia risk.
• ▸The 5 GLP-1 receptor agonists available include exenatide, liraglutide, albiglutide, dulaglutide, and lixisenatide.
• ▸A new, oral agent, semaglutide, is currently under FDA review and may soon become available as the newest GLP-1 receptor agonist.
• ▸The GLP-1 receptor agonists are valuable options for the treatment of type 2 diabetes as adjunctive therapy or as monotherapy.

GLP-1 is a naturally occurring hormone responsible for the incretin effect. The incretin effect is a response to release more insulin because of high glucose levels after a meal. Studies suggest that patients with type 2 diabetes have an attenuated incretin effect, possibly because of reduced levels of active GLP-1.9 Evidence shows that GLP-1 regulates the expression of beta-cell genes by inhibiting beta-cell apoptosis, preventing beta-cell glucolipotoxicity, and improving beta-cell function.10 GLP-1 has been shown to suppress glucagon release and hepatic glucose output.10 GLP-1 also decreases the rate of gastric emptying and acid secretion, thereby reducing appetite and contributing to weight loss. GLP-1 is degraded by dipeptidyl peptidase (DPP)-4, resulting in a shorter half-life, as shown in patients with type 2 diabetes and in healthy volunteers.11 This has led to the development of DPP-4 inhibitors, which inhibit the degradation of GLP-1. GLP-1 had been considered a treatment modality, but it has a very short half-life and would require continuous infusions.11 This has led to the development of GLP-1 receptor agonists, which are structurally similar to the natural hormone to provide beneficial effects but differ structurally to prevent breakdown by DPP-4.

This article reviews the evidence available for current GLP-1 receptor agonists.

Exenatide

Exenatide (Byetta) is a synthetic derivative of exendin-4 (isolated from salivary secretions of the Gila monster lizard) with a 53% amino acid sequence overlap.12 In 2005, it became the first GLP-1 receptor agonist to receive approval by the US Food and Drug Administration (FDA) for the treatment of type 2 diabetes. As an agonist of pancreatic beta-cells and resistance from DPP-4 inactivation, exenatide has a longer duration of action than GLP-1 and more than 1000-fold potency for lowering glucose than GLP-1.12 Exenatide has been shown to stimulate insulin production in response to blood glucose concentration, inhibit postprandial glucagon release, slow the rate of gastric emptying, slow the rate of nutrient absorption in the bloodstream, and reduce appetite.12 It is also found to promote the proliferation of beta-cells and islet-cell neogenesis from precursor cells.12

Exenatide was first introduced as a twice-daily injection of 5 mcg for 1 month followed by 5 mcg or 10 mcg. Pharmacokinetics demonstrated a plasma level reaching peak concentrations at 2 to 3 hours after administration with levels remaining detectable for 6 hours after administration. Patients with type 2 diabetes who were inadequately controlled with a sulfonylurea and/or metformin were given 0.08-mcg/kg subcutaneous injections of exenatide, which showed significant reductions in postprandial plasma glucose (PPG) and glycated hemoglobin (HbA_1c).12

Exenatide was studied in the phase 3 clinical trials AMIGO I, II, and III.12,13 In all 3 trials, the continuation of previous therapy (with metformin alone, sulfonylurea alone, or the combination of both) was compared between the addition of exenatide and placebo. The exenatide treatment group demonstrated a significant reduction in PPG concentrations and HbA_1c compared with the placebo group. Nausea was the most common adverse effect, with an increased rate of nausea in the exenatide groups versus the placebo groups. The rates of hypoglycemia in AMIGO I, which included patients who had received metformin, were equal between the exenatide and the placebo groups; however, in the AMIGO III study, which included patients who had received sulfonylurea and metformin combination therapy, patients receiving 10-mg exenatide had increased hypoglycemia (28% vs 13% in the placebo group). No changes in heart rate, blood pressure, and electrocardiograms were noted. The small increase in cortisol levels normalized by day 28.12,13

Buse and colleagues compared exenatide 5 μg twice daily for 4 weeks and then 10 μg twice daily thereafter with placebo in patients receiving insulin glargine.14 Insulin glargine was titrated to achieve a fasting glucose of <100 mg/dL on the basis of the Treat-to-Target Trial algorithm. The study showed an HbA_1c reduction of 1.74% with exenatide versus 1.04% with placebo. No significant increase in hypoglycemia or weight gain occurred. Similar to the AMIGO trials, exenatide was associated with more events of nausea (41% vs 8%, respectively) and vomiting (18% vs 4%, respectively) than placebo.14

Exenatide ER

A new formulation of exenatide, exenatide extended-release (ER; Bydureon) 2-mg once-weekly injection was approved by the FDA in 2012 as an adjunct therapy or monotherapy in patients with type 2 diabetes.15 Exenatide ER reaches therapeutic levels after 2 weeks, and after 6 weeks the drug attains a maximum concentration higher than that attained by a single injection of exenatide 10 mcg.15 Six weeks after stopping treatment, the serum concentration of exenatide once weekly declines to insignificant levels.

The phase 3 clinical trials of exenatide ER included the DURATION series, and are summarized in Table 2.16–22 DURATION-1 and -5 compared exenatide twice daily versus exenatide ER, showing that exenatide ER had a greater HbA_1c reduction and better glucose control compared with the twice-daily formulation. DURATION-2 and -4 compared exenatide ER with other diabetic oral medications, including pioglitazone, sitagliptin, and metformin, which demonstrated comparable efficacy in reducing HbA_1c and significantly reducing weight.16–21

Table 2.

Phase 3 DURATION Trials with Exenatide ER16–22

Trial	Study drug	Exenatide ER outcomes vs comparator drugs
DURATION-1	Exenatide ER 2 mg vs exenatide 10 mcg twice daily	Greater HbA1c reduction: −1.9% vs −1.5% Greater reduction in lipid profile, total cholesterol, triglycerides Better glucose control, body weight reduction, systolic blood pressure reduction Reduced nausea
DURATION-2	Exenatide ER vs pioglitazone vs sitagliptin; all agents taken with metformin	Greater HbA1c reduction w/ exenatide ER: −1.5% vs −0.9% vs −1.2% Greater weight loss: −2.3 kg vs −0.8 kg vs +2.8 kg Less nausea (5% vs 10.8% vs 9.6%) No hypoglycemia w/ exenatide ER
DURATION-3	Exenatide ER vs insulin glargine, titrated to goal <100 mg/dL	Greater HbA1c reduction w/ exenatide ER: −1.5% vs −1.3% 3 × lower hypoglycemia rate w/ exenatide ER
DURATION-4	Exenatide ER vs metformin vs pioglitazone vs sitagliptin; all in treatment-naïve patients	HbA1c reduction: −1.53% vs 1.48% vs 1.63% vs 1.15% Weight loss: −2.0 kg vs −2.0 kg vs +1.5 kg vs −0.8 kg Nausea & diarrhea: 11.3% and 10.9% w/ exenatide ER No major hypoglycemia occurred
DURATION-5	Exenatide ER vs exenatide; this is similar to DURATION-1	At 24 weeks, greater HbA1c reduction: −1.6% vs −0.9% Greater fasting glucose reduction: −35 mg/dL vs −12 mg/dL Similar weight reduction, adverse effects
DURATION-6	Exenatide ER vs liraglutide	Greater HbA1c reduction w/ liraglutide: −1.48% vs −1.28% More patients reached goal w/ liraglutide: 60% vs 53% Greater weight loss w/ liraglutide

ER indicates extended-release; HbA_1c, glycated hemoglobin.

Exenatide was associated with an increase in gastrointestinal (GI) adverse effects, including nausea, vomiting, and diarrhea,16–21 as is expected of the GLP-1 class. Nausea was most notable during the first few weeks of therapy and was minimized by gradual dose titration. In DURATION-2 and -4, no significant differences were reported in the rates of hypoglycemia between exenatide ER and metformin, pioglitazone, or sitagliptin.18,20 DURATION-3 compared exenatide ER with insulin glargine, showing 3 times fewer hypoglycemic events with the GLP-1 inhibitor than in the insulin glargine group.19

Mild injection-site pruritus was observed more often with exenatide ER, but it resolved with treatment continuation.17 Despite concerns for a possible association of exenatide and the other GLP-1 receptor agonists with increased risk for pancreatitis, this was not observed in the DURATION trials.15

Liraglutide

Liraglutide (Victoza) is an acylated analog of GLP-1 that has 97% amino acid sequence identity to the endogenous GLP-1 analog. In 2009, it was the second GLP-1 agonist to be approved by the FDA for the treatment of type 2 diabetes. Liraglutide is a long-acting GLP-1 receptor agonist that is administered once daily as a subcutaneous injection in contrast to twice-daily injections of the first exenatide formulation.23 Liraglutide has been reported to increase beta-cell mass in animal models via increased beta-cell replication and reduced apoptosis.24 In a study with normal-weight and obese rats, liraglutide was associated with a reduction in food intake, resulting in weight loss of approximately 15%.25 Preclinical studies showed improvement in first- and second-phase insulin secretion, implying that liraglutide leads to improved biphasic insulin secretion in response to hyperglycemia.26,27

The Liraglutide Effect and Action in Diabetes (LEAD) program is comprised of 6 phase 3 clinical trials, which are summarized in Table 3.28–33 Liraglutide, given as adjunct therapy and as monotherapy, was associated with significant reductions in HbA_1c levels, blood pressure, fasting plasma glucose (FPG), and PPG levels.28–33 Liraglutide is superior to insulin glargine and to twice-daily exenatide in HbA_1c reduction. Weight loss was similar between the liraglutide and the exenatide groups, but greater weight loss was seen with liraglutide compared with insulin glargine.28–33

Table 3.

Phase 3 LEAD Trials with Liraglutide28–33

Trial	Study drug	Liraglutide outcomes vs comparator drugs
LEAD-1	Liraglutide 1.2 mg & 1.8 mg once daily vs rosiglitazone 4 mg once daily; all concurrently taking sulfonylurea	Significant HbA1c reduction w/ liraglutide 1.2 mg & 1.8 mg: 1.1% vs −0.4% w/ rosiglitazone 4 mg Significant decrease in FPG & PPG w/ liraglutide vs rosiglitazone Minor hypoglycemia, <10%; nausea, <11%; vomiting, <5%; diarrhea, <8%
LEAD-2	Liraglutide 1.2 mg & 1.8 mg vs glimepiride 4 mg; all concurrently taking metformin	Noninferior HbA1c reduction in liraglutide groups: mean decrease, −1% Body weight −2.8 kg w/ 1.8-mg liraglutide vs +1.0 kg w/ glimepiride Less hypoglycemic events in liraglutide groups: 3% vs 17% w/ glimepiride Increased nausea in liraglutide groups
LEAD-3	Liraglutide 1.2 mg & 1.8 mg once daily vs glimepiride 8 mg once daily	HbA1c reductions: −0.84% & −1.23% w/ liraglutide 1.2 mg & 1.8 mg vs 0.51% w/ glimepiride 8 mg No major hypoglycemic events Significantly less minor hypoglycemia: 8% & 12% vs 24%
LEAD-4	Liraglutide 1.2 mg & 1.8 mg vs placebo; all concurrently taking metformin and rosiglitazone	HbA1c reduction: −1.5% vs −0.5% Significant FPG and PPG reductions w/ 1.2-mg & 1.8-mg liraglutide Body weight reductions: −1.0 kg & −2.0 kg w/ liraglutide 1.2 mg & 1.8 mg vs +0.6-kg weight gain w/ placebo Systolic BP reductions: −6.7 mm Hg & −5.6 mm Hg w/ liraglutide 1.2 mg & 1.8 mg vs −1.1 mm Hg w/ placebo Minor hypoglycemia: 7.9% & 9% vs 5.1% No major hypoglycemic events
LEAD-5	Liraglutide 1.8 mg vs insulin glargine; all concurrently taking metformin and glimepiride	Significantly greater HbA1c reduction: −1.33% vs −1.09% Significantly greater weight loss w/ liraglutide: −1.39 kg vs +3.43 kg Systolic BP reduction: −4 mm Hg vs +0.5 mm Hg Major & minor hypoglycemia rates: 0.06 & 1.2 vs 0 & 1.3 events/patient annually
LEAD-6	Liraglutide 1.8 mg vs exenatide 10 μg twice daily, all concurrently taking metformin and sulfonylurea	Significant HbA1c reduction: −1.12% vs −0.79% Greater FPG reduction vs exenatide Weight loss: 3.24 kg vs 2.87 kg (difference not significant) Significantly less minor hypoglycemia w/ liraglutide: 25.5% vs 33.6% 2 patients taking exenatide & sulfonylurea had major hypoglycemia Less nausea w/ liraglutide

BP indicates blood pressure; FPG, fasting plasma glucose; HbA_1c, glycated hemoglobin; PPG, postprandial glucose.

The LEAD trials showed that the risk for hypoglycemia is low with liraglutide and is significantly lower than with a sulfonylurea or twice-daily exenatide.28–33 Like exenatide, liraglutide was associated with increased GI side effects, including nausea and vomiting, which were generally mild and transient. A total of 3.4% of the patients receiving liraglutide in the phase 3 trial withdrew because of nausea.30 In general, the GI adverse effects can be managed by starting at lower doses of liraglutide and then gradually increasing the dose. Liraglutide was associated with a lower antibody formation than exenatide, likely because of the greater (97%) amino acid sequence identity than human GLP-1.34 Exenatide has a lower sequence identity than liraglutide, which may explain the incidence of anti-exenatide antibody formation in up to 43% of exenatide-treated patients.35

There have been few case reports of liraglutide-associated pancreatitis. Studies in rodents have shown that liraglutide induces C-cell proliferation and medullary thyroid adenomas and carcinomas via GLP-1 receptor agonist activation and calcitonin release, but this pattern was not seen in humans. Follow-up studies have been inconclusive to definitively define a cause-and-effect relationship between liraglutide and pancreatitis, because patients with type 2 diabetes already have a 3-fold increased risk for pancreatitis.36 In the LEADER trial, liraglutide taken for 3.5 years was associated with a 23% reduction in CV events, a 22% reduction in CV mortality, and a 15% reduction in all-cause mortality.37

Albiglutide

Albiglutide (Tanzeum) is a GLP-1 agonist that was approved by the FDA in 2014 as an adjunct treatment for diabetes; it is administered as a weekly injection.38 Albiglutide has 97% homology to the amino acid sequence of GLP-1. A single amino acid substitution (alanine to glycine) renders albiglutide resistant to DPP-4–mediated protein degradation, resulting in a longer half-life. After subcutaneous injection of a single 30-mg dose, patients with type 2 diabetes achieved mean maximum plasma concentration 3 to 5 days after administration. Plasma concentrations reach steady state within 3 to 5 weeks of repeated once-weekly administrations. Albiglutide is currently available as a 30-mg and a 50-mg once-weekly injection.38

Albiglutide was tested in the HARMONY phase 3 clinical trials, which comprised 8 studies (Table 4).39–46 HARMONY-2 demonstrated the superiority of albiglutide monotherapy to diet and exercise in glycemic control.40 In HARMONY-3, once-weekly albiglutide add-on therapy was noninferior to once-daily sitagliptin and once-daily glimepiride at reducing HbA_1c levels in patients inadequately controlled with metformin alone,41 whereas HARMONY-4 and -6 demonstrated that albiglutide was noninferior to insulin therapy in patients inadequately controlled with oral antidiabetes therapy.42,44 However, in HARMONY-5, albiglutide was found to be inferior to pioglitazone in HbA_1c reduction.43 HARMONY-8 revealed that albiglutide was superior to sitagliptin in patients with and without renal impairment.46

Table 4.

Phase 3 HARMONY Trials with Albiglutide39–46

Trial	Study drug	Albiglutide outcomes vs comparator drugs
HARMONY-1	Albiglutide 30 mg vs placebo	HbA1c: −0.8% vs −0.1% Hyperglycemia events: 24.4% vs 47.7% No significant differences in weight change All GI events: 31.3% vs 29.8% Diarrhea: 11.3% vs 8.0% Nausea: 10.7% vs 11.3% Vomiting: 4% vs 4%
HARMONY-2	Albiglutide 30 mg vs albiglutide 50 mg vs placebo	HbA1c: −0.84% vs −1.04% No significant changes in weight w/ 2 albiglutide doses Similar nausea, diarrhea, vomiting, hypoglycemia rate in all groups, including placebo
HARMONY-3	Albiglutide 30 mg vs sitagliptin 100 mg vs glimepiride 2 mg vs placebo; all concurrently taking metformin	HbA1c: −0.9% vs −0.4% vs −0.3% (vs placebo) Weight change: −1.21 kg vs −0.86 kg vs +1.17 kg vs −1.0 kg Hyperglycemia rates: 25.8% vs 36.4% vs 32.7% vs 59.2% Diarrhea: 12.9% vs 8.6% vs 10.9% (vs placebo) Nausea: 10.3% vs 6.2% vs 10.9% (vs placebo)
HARMONY-4	Albiglutide vs insulin glargine titrated to fasting plasma glucose goal of 100 mg/dL	HbA1c: −0.7% vs −0.8% Weight change: −1.0 kg vs +1.5 kg Hypoglycemia: 17.5% vs 27.4%
HARMONY-5	Albiglutide 30 mg titrated up to 50 mg vs pioglitazone 30 mg titrated up to 50 mg; all concurrently taking metformin ± glimepiride 4 mg	HbA1c reduction: −0.87% vs placebo HbA1c +0.25 vs pioglitazone: not meeting noninferiority criteria Hypoglycemia: 14% vs 25% vs 14% Weight change: −0.42 kg vs +4.4 kg vs −0.4 kg
HARMONY-6	Albiglutide 30 mg titrated up to 50 mg vs insulin lispro 3 × daily adjusted per glucose level	HbA1c: −0.82% vs −0.66% Weight change: −7.3 kg vs +0.81 kg Severe hypoglycemia: 0 vs 2 events Nausea: 11.2% vs 1.4% Vomiting: 6.7% vs 1.4% Injection-site reaction: 9.5% vs 5.3%
HARMONY-7	Albiglutide 30 mg titrated up to 50 mg vs liraglutide 0.6 mg titrated up to 1.8 mg; all concurrently taking metformin ± sulfonylurea ± thiazolidinedione	HbA1c: −0.78% vs −0.99% Injection-site reaction: 12.9% vs 5.4% GI adverse effects: 35.9% vs 49%
HARMONY-8	Albiglutide vs sitagliptin with GFR >60 mL/min, GFR 30–59 mL/min, GFR 15–29 mL/min; all ± oral diabetes drugs	HbA1c: −0.83% vs −0.52% Time to hyperglycemic rescue longer w/ albiglutide All adverse events: 51.7% vs 25.2% Diarrhea: 10% vs 6.5% Nausea: 4.8% vs 3.3% Vomiting: 1.6% vs 1.2% Hypoglycemia: 24.1% vs 15.9% (sulfonylurea: 22.5% vs 14.2%; no sulfonylurea: 4% vs 4%) Weight change: −0.79 kg vs −0.19 kg

GFR indicates glomerular filtration rate; GI, gastrointestinal; HbA_1c, glycated hemoglobin.

Albiglutide demonstrated greater weight loss in all studies compared with sitagliptin, glimepiride, pioglitazone, and insulin therapy, although more GI adverse effects were reported with albiglutide compared with other agents.39–46 All trials demonstrated no significant differences in rates of hypoglycemia, except in patients with impaired renal disease who used albiglutide and a sulfonylurea.39–46

Dulaglutide

Dulaglutide (Trulicity) is a once-weekly subcutaneously administered GLP-1 receptor agonist approved by the FDA in 2014 as an adjunct to diet and exercise to improve glycemic control in patients with type 2 diabetes.47 The initial dosage is 0.75 mg administered subcutaneously once weekly, which may be increased to 1.5 mg once weekly for additional glycemic control. Dulaglutide is comprised of 2 identical GLP-1 analog peptide chains (approximately 90% homologous to native human GLP-1) linked to an immunoglobulin (Ig) G4 heavy chain. The alteration of the GLP-1 analog provides protection against degradation by DPP-4, improved solubility, and reduced immunogenicity. The addition of IgG4 increases the size of the protein, which helps decrease the rate of renal clearance, and the Fc fragment of IgG4 prevents antibody formation to further reduce the potential for immunologic cytotoxicity.47

Dulaglutide has been studied in comparison with other antidiabetes agents and with placebo in the phase 3 AWARD trials (Table 5).48–54 These trials demonstrate that once-weekly dosing of 1.5-mg dulaglutide was superior to metformin, insulin glargine, and sitagliptin in reducing HbA_1c levels; dosing with dulaglutide 0.75 mg was noninferior to these agents. Patients in these trials experienced greater loss with 1.5-mg and with 0.75-mg dosing of dulaglutide compared with other agents. In patients taking dulaglutide and insulin concomitantly, there was either attenuation of the weight gain or overall weight loss compared with patients receiving placebo.48–54 In AWARD-6, patients receiving liraglutide 1.8 mg experienced greater weight loss than those receiving dulaglutide 1.5 mg.53

Table 5.

Phase 3 AWARD Trials with Dulaglutide48–54

Trial	Study drug	Dulaglutide outcomes vs comparator drugs
AWARD-1	Dulaglutide 1.5 mg once weekly vs dulaglutide 0.75 mg vs exenatide 10 μg twice daily; all concurrently taking metformin and pioglitazone (26-week study)	HbA1c: −1.51% vs −1.3% vs −0.99% Body weight: −1.3 kg vs +0.2 kg vs −1.07 kg No severe hypoglycemia w/ dulaglutide Very similar rates of nausea, vomiting, & diarrhea with dulaglutide 1.5 mg & exenatide Less adverse effects w/ dulaglutide 0.75 mg
AWARD-2	Dulaglutide 1.5 mg vs dulaglutide 0.75 mg vs insulin glargine; all concurrently taking metformin and glimepiride (52-week study)	HbA1c: −1.08% vs −0.76% vs −0.63% Superiority met w/ dulaglutide 1.5 & noninferiority w/ 0.75-mg dose Hypoglycemia rates lower w/ dulaglutide More nausea and diarrhea w/ dulaglutide than glargine
AWARD-3	Dulaglutide 1.5 mg vs dulaglutide 0.75 mg vs metformin titrated up to 2000 mg/day (26-week study)	HbA1c: −0.78% vs −0.71% vs −0.51% Dulaglutide 1.5 mg & 0.75 mg met superiority w/ HbA1c reduction Similar decreases in weight between all 3 groups Nausea, diarrhea, & vomiting were similar between dulaglutide & metformin
AWARD-4	Dulaglutide 1.5 mg vs dulaglutide 0.75 mg vs insulin glargine; all concurrently receiving prandial insulin lispro (52-week study)	HbA1c: −1.64% vs −1.59% vs −1.41% Noninferiority of dulaglutide 1.5 mg vs glargine w/ HbA1c reduction Weight change: −2.88 kg vs −2.39 kg vs −1.75 kg; significance difference w/ dulaglutide 1.5 mg vs sitagliptin More nausea, diarrhea, & vomiting w/ dulaglutide than glargine
AWARD-5	Dulaglutide 1.5 mg vs dulaglutide 0.75 mg vs sitagliptin 100 mg; all concurrently taking metformin (52-week study)	HbA1c: −1.10% vs −0.87% vs −0.39% Both doses met superiority to sitagliptin More nausea, diarrhea, & vomiting w/ dulaglutide
AWARD-6	Dulaglutide 1.5 mg vs liraglutide 1.8 mg; all concurrently taking metformin (26-week study)	HbA1c: −1.42% vs −1.36%; met noninferiority criteria Greater weight loss w/ liraglutide (–2.9 kg vs −3.61 kg) Nausea: 20% vs 18% Diarrhea: 12% vs 12% Dyspepsia: 8% vs 6% Vomiting: 7% vs 8%
AWARD-8	Dulaglutide 1.5 mg vs placebo; all concurrently taking glimepiride	HbA1c difference: −1.3%; superior to placebo Fasting plasma glucose difference: −33.54 mg/dL; superior to placebo Weight loss was significant from baseline w/ dulaglutide, but between-groups difference not significant Hypoglycemia higher w/ dulaglutide (2.37 vs 0.07 events/participant annually) No severe hypoglycemic events

HbA_1c indicates glycated hemoglobin.

Similar to other GLP-1 receptor agonists, the most frequently reported adverse events with dulaglutide were GI in nature, including nausea, vomiting, and diarrhea.48–54 These events were generally mild to moderate, peaked at 2 weeks, and rapidly declined over the next 4 weeks. The majority of adverse events were reported during the first 2 to 3 days after receiving the initial dose and declined with subsequent doses. Hypoglycemic events were not common in patients taking dulaglutide, and occurred less frequently compared with patients receiving insulin therapy, as was shown in AWARD-2 and -449,51; however, significantly more hypoglycemic events were reported with a sulfonylurea as background therapy compared with placebo as demonstrated in AWARD-8.48–54

Given its mechanism of action, dulaglutide was evaluated for pancreatic safety. Throughout the AWARD trials, 4 events were reported in patients taking dulaglutide (3 patients receiving dulaglutide 1.5 mg and 1 receiving the 0.75-mg dose). Laboratory studies of pancreatic amylase and lipase in these trials revealed a mean 14% to 20% increase in amylase and lipase levels in patients receiving dulaglutide; however, these events were not predictive of acute pancreatitis. Given the association of GLP-1 analogs with medullary thyroid carcinoma, thyroid safety was assessed as well. In the AWARD trials, only 1 case of medullary thyroid carcinoma in AWARD-5 was reported, although this case was determined to be preexisting.52

Lixisenatide

Lixisenatide (Adlyxin) is a once-daily subcutaneous GLP-1 receptor agonist that was approved by the FDA in July 2016 for the treatment of type 2 diabetes in adults.55 Lixisenatide is designed as C-terminal modification with 6 lysine residues and deletion of 1 proline, allowing it to withstand physiologic degradation by DPP-4. Lixisenatide is renally excreted, with a half-life of 2 to 4 hours. Despite its short half-life, lixisenatide is intended for once-daily dosing as a result of its strong binding affinity to the GLP-1 receptor. No clinically relevant difference was found in the rate of absorption if lixisenatide is injected into the abdomen, thigh, or arm. In a dose-dependent manner, lixisenatide tested at 5-mcg, 10-mcg, and 20-mcg doses reached peak concentrations between 1 and 2 hours.55 Preclinical trials have also shown that the addition of a GLP-1 receptor agonist to insulin analog–like glargine demonstrated a protective effect on beta-cells, suggesting that the combination of these medications may preserve beta-cell mass in patients with type 2 diabetes.56 Thus, in November 2016, the FDA approved the combination of lixisenatide with insulin glargine (Soliqua 100/33).

Lixisenatide was studied in the 10 phase 3 GETGOAL clinical trials that assessed its efficacy and safety profile (Table 6).56–64 In these trials, the 20-mcg dose of lixisenatide was selected, because it had demonstrated in previous trials the best efficacy-to-tolerability ratio. The phase 3 studies assessed lixisenatide in a 1-step titration as a 10-mcg dose for 2 weeks, then a 20-mcg dose once-daily subcutaneously, and in a 2-step titration as a 10-mcg dose for 1 week, 15-mcg dose for 1 week, and then as a 20-mcg dose. In all these trials except GETGOAL-M,64 lixisenatide was administered in the morning. No significant differences were seen in efficacy and adverse events between the 1- and 2-step titration groups.56–64

Table 6.

Phase 3 GETGOAL Trials with Lixisenatide56–64

Trial	Study drug	Lixisenatide outcomes vs comparator drugs
GETGOAL-Mono	Lixisenatide 1-step AM vs 2-step AM vs placebo; 12-week study	HbA1c: −0.66% 1-step vs −0.54% 2-step Achieved HbA1c goal (<7%): 46.5% 1-step vs 52.2% 2-step Decrease in body weight: ∼2 kg in both groups Symptomatic hypoglycemia: 1.7% in lixisenatide groups vs 1.6% in placebo groups Significant improvements in HbA1c: 2-hr PPG, FPG vs placebo
GETGOAL-F1	Lixisenatide 1-step AM vs 2-step AM; all concurrently taking metformin; 24-week study	HbA1c: −0.9% vs −0.8% vs −0.4% Improved FPG: −0.5 vs −0.6 vs +0.1 mmol/L Body weight: −2.6 kg vs −2.7 kg vs −1.6 kg Symptomatic hypoglycemia: 1.9% vs 2.5% vs 0.6%
GETGOAL-S	Lixisenatide 2-step AM vs placebo; all concurrently taking sulfonylurea; 24-week study	HbA1c: −1.1% mean significant difference vs placebo Significant 2-hr postprandial glucose & FPG vs placebo Body weight: −1.12 kg vs −1.02 kg GI adverse effects: 52.6% vs 29.4% Symptomatic hypoglycemia: 17.1% vs 9.8% No cases of severe symptomatic hypoglycemia in either group
GETGOAL-L	Lixisenatide 2-step AM vs placebo; all concurrently taking basal insulin	HbA1c reduction: −0.4% difference vs placebo Symptomatic hypoglycemia: 28% vs 22% Severe hypoglycemia: 1.2% vs 0%
GETGOAL-P	Lixisenatide 2-step AM vs placebo; all concurrently taking pioglitazone; 24-week study	HbA1c reduction: −0.56% difference vs placebo Significantly improved FPG: −0.84 mmol/L Small decrease body weight w/ lixisenatide & small but insignificant increase w/ placebo Symptomatic hypoglycemia: 3.4% vs 1.2%; no severe episodes
GETGOAL-X	Lixisenatide 2-step AM vs exenatide 10 mcg twice daily; all concurrently taking metformin; 24-week study	Noninferiority in HbA1c reduction vs exenatide FPG reduction was comparable Weight loss: −2.8 kg vs −3.8 kg Serious adverse events: 2.8% vs 2.2% Significantly reduced symptomatic hypoglycemia: 2.5% vs 7.9% Significantly less nausea events: 24.5% vs 35.1%
GETGOAL-M	Lixisenatide 2-step AM or PM vs placebo; all concurrently taking metformin; 24-week study (similar to GETGOAL-F1)	Significant HbA1c reduction: −0.36% difference vs placebo Significant reduction in 2-hr PPG vs placebo No difference in weight loss Nausea: 16.3% vs 2.6% Symptomatic hypoglycemia: 5.6% vs 2.6% No severe symptomatic hypoglycemia
GETGOAL-Mono JAPAN LTS	Lixisenatide 1-step AM vs 2-step AM; 52-week study with primary end point on safety measures	Nausea: 59% vs 36.4% Symptomatic hypoglycemia: 0% vs 6.1% HbA1c, FPG, body weight reduced from baseline
GETGOAL-L-ASIA	Lixisenatide 2-step AM vs placebo; all concurrently taking basal insulin; 24-week study (similar to GETGOAL-S)	HbA1c reduction: −0.88% vs placebo Significant improvement in 2-hr postprandial glucose Nausea and vomiting: 18.2% vs 1.9% Symptomatic hypoglycemia: 42.9% vs 23.6% No severe hypoglycemia
GETGOAL-M-ASIA	Lixisenatide 2-step AM vs placebo; all concurrently taking metformin + sulfonylurea; 24-week study	HbA1c reduction: −0.57% (significant difference) Superior to placebo in lowering 2-hr postprandial glucose Body weight in lixisenatide group trended to decrease

AM indicates morning administration; FPG, fasting plasma glucose; GI, gastrointestinal; HbA_1c, glycated hemoglobin; PM, evening administration; PPG, postprandial plasma glucose.

Lixisenatide demonstrated superiority in reducing HbA_1c, PPG, and FPG compared with placebo monotherapy or adjunct therapy. In GETGOAL-X, lixisenatide demonstrated noninferiority with HbA_1c reduction compared with exenatide 10 mcg twice daily. Weight loss was superior with lixisenatide treatment in all trials, except GETGOAL-M,64 compared with placebo56–64; however, in GETGOAL-X, lixisenatide treatment resulted in an average 2.8-kg weight loss compared with 3.8 kg in the exenatide group.63

As with other GLP-1 analogs, there was an increase in GI adverse effects with lixisenatide, including nausea and vomiting, as reported in GETGOAL-F1,58 GETGOAL-S,59 GETGOAL-L,60,62 GETGOAL-P,61 and GETGOAL-M64; however, there were fewer reports of nausea compared with exenatide. In all trials where lixisenatide was not combined with insulin, pioglitazone, or a sulfonylurea, no increase in hypoglycemic events was seen compared with placebo. If combined with these agents, the lixisenatide groups exhibited more hypoglycemic events. Compared with exenatide, fewer hypoglycemic events were reported in the patients receiving lixisenatide.56–64

CV outcomes were studied with lixisenatide in the separate phase 3 ELIXA trial.65 Patients who take lixisenatide do not have any increase in CV adverse effects after an acute coronary syndrome compared with placebo. In addition, no significant CV benefit was seen compared with placebo.65

Semaglutide

Semaglutide is an investigational agent that was developed as a once-weekly subcutaneous formulation, as well as the first oral GLP-1 analog formulation. The manufacturer applied for regulatory approval by the FDA of the injectable formulation in December 2016, after the phase 3 clinical trial SUSTAIN-6 showed promising results, including HbA_1c reduction, weight loss, and CV benefit.66 The oral formulation is still in phase 3 clinical trials.67

The SUSTAIN-6 trial showed the weekly subcutaneous formulation of semaglutide to have a significant HbA_1c reduction of 0.7% with the 0.5-mg dose, and 1% with the 1-mg dose, compared with placebo.68 Patients in the 0.5-mg group had a weight loss of 2.9 kg, and the 1-mg group had a 4.3-kg weight loss. Nonfatal myocardial infarction occurred in 2.9% of patients receiving semaglutide versus 3.9% in patients receiving placebo. Nonfatal stroke occurred in 1.6% and 2.7% of the patients, respectively. The rate of death from a CV cause was similar in both groups. The rate of new or worsening nephropathy was lower in the semaglutide group than in the placebo group, although the rate of retinopathy complications was significantly higher with semaglutide.68

Perhaps the most exciting development in the GLP-1 class is the oral formulation of semaglutide that has shown promising results in its phase 2 trial and is currently undergoing a phase 3 study.69 This oral formulation is combined with the absorption enhancer SNAC (sodium N-[8-(2-hydroxybenzoyl)amino] caprylate), which causes a localized increase in pH. This enables higher solubility and protects from enzymatic degradation. The patients in the phase 2 study experienced dose-dependent decreases in their HbA_1c and had similar results in their weight loss and other secondary outcomes.69 Mild-to-moderate GI side effects were the most frequently reported adverse events, which included nausea (13%-34%), vomiting (6%-22%), and diarrhea (7%-23%).69

Comparison of GLP-1 Receptor Agonists

Currently, 5 GLP-1 receptor agonists are FDA-approved in the United States for the treatment of patients with type 2 diabetes. Their formulations vary from the twice-daily injection of exenatide to once-weekly formulations of albiglutide, exenatide ER, and dulaglutide. Several head-to-head comparison studies have compared the GLP receptor agonists. A new drug is currently under FDA review.

Comparing exenatide twice daily with exenatide once weekly showed a significantly greater reduction of HbA_1c with exenatide ER (difference, 0.7%)17; the adverse effects were similar, but injection-site reactions were more common with exenatide ER. In DURATION-6, once-daily treatment with liraglutide 1.8 mg resulted in significantly greater reduction of HbA_1c (difference, 0.21%) and greater weight loss (difference, 0.90 kg) in comparison with once-weekly exenatide 2 mg, although GI adverse events occurred more often with patients taking liraglutide.22

Similarly, in LEAD-6, liraglutide 1.8 mg had a significantly greater HbA_1c reduction (difference, −0.33%) and less adverse effects, including hypoglycemia, than exenatide 10 mcg twice daily.33 HARMONY-7 compared once-weekly albiglutide 50 mg with liraglutide 1.8 mg and showed greater HbA_1c reduction with liraglutide (difference, 0.21%; noninferior).45 There were more injection-site reactions with albiglutide (difference, 7.5%), but more GI events with liraglutide (difference, 13.1%).45 AWARD-6 compared once-weekly dulaglutide 1.5 mg with liraglutide 1.8 mg, showing greater HbA_1c reduction with dulaglutide (difference, −0.06%; noninferior), although liraglutide had significantly greater weight loss (difference, 0.71 kg). No significant differences in the adverse-effect profile were noted in the study.53

GETGOAL-X compared lixisenatide 20 mcg with exenatide 10 mcg twice daily and showed similar HbA_1c reduction, although there was less hypoglycemia and nausea with lixisenatide.63 Lixisenatide is currently marketed mainly as a 5-mcg dose in combination with insulin glargine (Soliqua).

Conclusion

The GLP-1 receptor agonists are valuable options for the treatment of type 2 diabetes as adjunctive therapy or as monotherapy. There is robust evidence supporting the indication for the use of GLP-1 receptor agonists if patients are overweight or obese, have CV disease or renal disease, or are at high risk for hypoglycemia—common comorbidities of type 2 diabetes. Clinical trials demonstrate the superiority of GLP-1 receptor agonists to other antidiabetes drugs in HbA_1c reduction, blood pressure reduction, and weight loss, without hypoglycemia risk. Unlike metformin, there is no contraindication to giving patients with renal disease a GLP-1 receptor agonist. Although some significant differences exist among the agents in this class, the efficacy of the individual agents is generally comparable. Choosing among the available GLP-1 receptor agonists will likely depend on patient preferences, reaction to adverse effects, and cost.