Practical Use of Glucagon-Like Peptide-1 Receptor Agonist Therapy in Primary Care

The GLP-1 receptor agonist class has five important actions for patients with type 2 diabetes. The first is an increase in glucose-mediated insulin production by pancreatic β-cells.^2–6  “Glucose-mediated” is an important nuance because insulin production and release remains under the control of the glucose-sensing mechanisms of β-cells and only occurs during hyperglycemia. As a consequence, there is a low incidence of hypoglycemia. In rat and mouse models, there is a slowing of β-cell death.^7–10  Conflicting data in humans involving a variety of measures of β-cell function make it unclear whether this is also a phenomenon in humans.^11–15  Other actions include a decrease or no change in fasting endogenous glucose release via a reduction in glycogenolysis but not gluconeogenesis and a reduction in glucagon secretion.^3,16,17  The gastric emptying rate is also slowed, thereby slowing the absorption of carbohydrate, leading to a lower rise in plasma glucose.^3,18  GLP-1 receptor agonists also act via the central nervous system, resulting in a sensation of satiety and reduced food intake.^19–21  This effect explains, in part, the added benefit that many patients lose weight when taking a GLP-1 receptor agonist.^{11,15,22–26} 

By comparison, DPP-4 inhibitors, which also act on the incretin system, do not slow the gastric emptying rate, promote satiety, reduce food intake, or promote weight loss.²⁷  These differences between DPP-4 inhibitors and GLP-1 receptor agonists are thought to result from differences in how the two classes exert their actions on the incretin system. DPP-4 inhibitors work indirectly by inhibiting the metabolism of native GLP-1 produced in the gut, thereby raising the level of endogenous GLP-1 to ~ 10 pmol/L.²⁸  By comparison, GLP-1 receptor agonists act directly on the GLP-1 receptor, providing a level of GLP-1 activity of ≈ 60 pmol/L of GLP-1.²⁹  From this, it is clear that DPP-4 inhibitors should not be considered an oral form of GLP-1 receptor agonist.

Beyond the low associated incidence of hypoglycemia and weight loss effects, GLP-1 receptor agonists offer several advantages that may be useful in individualizing therapy.

A reduction in A1C of 0.5–1.5% has been reported with GLP-1 receptor agonists as monotherapy.^{11,15,22–24,30}  Head-to-head clinical trials show significantly greater lowering of A1C with exenatide QW compared to exenatide BID,^31,32  1.8 mg liraglutide compared to exenatide BID,^14,33  and 1.8 mg liraglutide compared to exenatide QW³⁴  (Table 1).

Although A1C reduction results from a lowering of both fasting blood glucose (FBG) and postprandial glucose (PPG), there are differences among the three GLP-1 receptor agonists. A 30-week comparison showed a significantly greater reduction in FBG with exenatide QW compared to exenatide BID (reduction of 41 vs. 25 mg/dl, respectively, P < 0.0001). Both treatments resulted in significant improvements in seven-point self-monitoring of blood glucose (SMBG) profiles.³²  Another investigation showed that exenatide BID resulted in a significantly greater reduction than liraglutide in PPG after breakfast and supper but not after lunch, whereas liraglutide resulted in a significantly greater reduction than exenatide BID in FBG.¹⁴  A 26-week comparison of exenatide QW and 1.8 mg liraglutide once daily showed reductions in A1C of 1.28 and 1.48% (95% CI 0.08–0.33), respectively.³⁵ 

Another benefit of GLP-1 receptor agonists is their impact on cardiovascular risk factors and biomarkers. GLP-1 receptor agonists cause a reduction of 1–7 mmHg in systolic blood pressure but have no significant effect on diastolic blood pressure.^{11,14,15,23,33,36–42}  Improvements in lipid profile are also observed, notably a reduction in triglycerides of 12–40 mg/dl.^{11,14,23,25,36,38–42}  It is unclear whether these changes will have a beneficial outcome on cardiovascular event risk reduction. Long-term cardiovascular outcome trials are in progress with most GLP-1 receptor agonists.

As with all medications used to treat type 2 diabetes, GLP-1 receptor agonists have some limitations that should be discussed with patients when considering treatment options. Although hypoglycemia is common with many glucose-lowering agents, the occurrence of hypoglycemia with GLP-1 receptor agonists is generally low (Table 2). Focusing on the limitations that patients are more likely to experience (e.g., transient gastrointestinal [GI] adverse events) and how these limitations will be addressed is helpful in minimizing patients' concerns. Providing patient education materials and involving other members of the diabetes health care team can further reassure patients that the limitations of GLP-1 receptor agonists will be addressed as needed to improve self-management.

Each of the GLP-1 receptor agonists, to varying degrees, can cause transient nausea and diarrhea. As monotherapy, nausea occurs in 8, 11, and 28% of patients treated with exenatide BID, exenatide QW, and liraglutide, respectively, whereas diarrhea is experienced by < 2, 11, and 17%, respectively.^43–45  Similar GI adverse effects can occur with metformin; nausea/vomiting occurs in 26%, whereas diarrhea has been reported to lead to discontinuation in 6%.⁴⁶ 

GI adverse events are usually self-limiting. Nausea is usually mild and peaks within 8 weeks of starting exenatide BID and 4–8 weeks of starting liraglutide. Nausea resolves in all but ~ 10% within 28 weeks with exenatide BID and in 8 weeks with liraglutide.^23,47,48  For exenatide QW, nausea peaks soon after initiation and resolves within 10 weeks in nearly all patients.¹⁵ 

These possible GI adverse events are important to discuss with patients because patients who are not expecting such situations may stop the medication believing that it is a permanent problem. In addition, protracted vomiting, should it occur, may lead to pre-renal azotemia.

The most common approach to minimizing the risk and severity of nausea includes using a dose escalation strategy for exenatide BID and liraglutide; dose escalation is not needed for exenatide QW. Exenatide BID should be initiated at a dose of 5 μg twice daily and increased to 10 μg twice daily after 1 month based on clinical response. Exenatide BID should be taken within 60 minutes before the morning and evening meals. Liraglutide should be initiated at a dose of 0.6 mg once daily for 1 week and then increased to 1.2 mg once daily. If the 1.2-mg dose does not result in acceptable glycemic control, the dose can be increased to 1.8 mg once daily. One modification of this dose escalation strategy that is not described in the approved package inserts but that the author has found useful is to lengthen the time period during which dose escalation occurs.

Other strategies to help minimize the risk of nausea not described in the approved package insert include self-administering exenatide BID < 60 minutes before mealtime, temporarily reducing the dose, or stopping eating when patients feel full.^49–51  Patients should be educated to avoid administering a GLP-1 receptor agonist close to a large or high-fat meal because doing so is likely to cause nausea. An advantage of exenatide QW and liraglutide is that they can be administered without regard to meal times.

Another possible limitation is the local irritation and nodule formation around the injection site that occurs most frequently with exenatide QW. The nodules, which are generally not visible but can be felt, typically last only a few weeks. Local irritation or nodule formation is usually a minor issue that can be prevented or managed by rotating injection sites in the abdomen, thigh, and upper arms.

As with other medications, infrequent adverse events are possible with glucose-lowering medications, including the GLP-1 receptor agonists. Although these adverse events may not be routinely discussed with patients, if questions arise, they should be discussed at a depth and in a manner appropriate for the patients based on their questions and previous discussions and as part of a risk-benefit discussion. Patients should also be referred to the medication guide for each product for additional information.

One possible but infrequent adverse event is pancreatitis, which has been reported in post-marketing surveillance for all marketed GLP-1 receptor agonists and DPP-4 inhibitors. In clinical trials of liraglutide, 13 cases of pancreatitis (9 acute and 4 chronic) were reported in patients treated with liraglutide and 1 case was reported in a patient treated with glimepiride (2.7 vs. 0.5 cases/1,000 patient-years).⁴⁵ 

Analyses of two insurance claims databases show rates of pancreatitis that are similar among the glucose-lowering agents examined.^52,53  In one analysis (n = 786,656), the incidence rates (cases per 1,000 patient-years) were 5.7 for exenatide BID, 5.6 for sitagliptin, and 5.6 for metformin, sulfonylureas, or thiazolidinediones.⁵²  Similar rates of acute pancreatitis over 1 year were observed in a second analysis among patients treated with exenatide BID, sitagliptin, metformin, or glyburide.⁵³ 

In contrast, an association between GLP-1 receptor agonist therapy and an increased risk of pancreatitis was found in two other analyses. In one, which examined the FDA database of reported adverse events from 2004 to 2009, the use of exenatide BID or sitagliptin increased the odds ratio for reported pancreatitis sixfold compared to other glucose-lowering agents.⁵⁴  In the other, Singh et al.⁵⁵  found an increased risk of acute pancreatitis with current or recent (within 2 years) use of a GLP-1 receptor agonist. These findings were based on the health records of 1,269 patients hospitalized with acute pancreatitis and matched controls.

The findings by Singh et al. were quickly challenged in a joint response from the ADA and the American Association of Clinical Endocrinologists (AACE),⁵⁶  which stated that this analysis “does not provide the basis for changing treatment in people with diabetes.” The joint statement by ADA and AACE noted that further clarity on this issue should come from nine prospective, controlled trials of GLP-1 receptor agonist therapy involving > 65,000 subjects. In the meantime, patients should be encouraged to speak to their providers to assess which treatments are best for them and to not stop therapy without consulting their provider.

The association of GLP-1 receptor agonists with pancreatitis is difficult to assess because of the nearly threefold greater risk of pancreatitis in people with type 2 diabetes compared to those without type 2 diabetes.⁵⁷  Until this issue is resolved, it is important to explain to patients the difference between the symptoms of pancreatitis and the minor, transient nausea described above. In addition, in patients with a history of pancreatitis, glucose-lowering agents other than exenatide BID⁴³  and exenatide QW⁴⁴  should be used, and liraglutide should be used with caution, according to the package inserts for each of these agents.⁴⁵ 

Although very rare in humans, another possible safety concern relates to the risk of thyroid C-cell neoplasms, which were found in preclinical studies with rats and mice but not with monkeys.^58,59  Thyroid C-cell tumors were also observed in rodents not exposed to a GLP-1 receptor agonist, and expression of GLP-1 receptors in the thyroid C-cell tissues of humans and monkeys is low.⁵⁹ 

One of the conditions required by the FDA for making exenatide QW and liraglutide commercially available was to create a national registry of patients with thyroid tumors. The registry will allow monitoring to identify increases in the incidence of this problem in humans. If patients have a history of medullary C-cell tumors or multiple endocrine neoplasia-2 (thyroid, parathyroid, and pheochromocytoma tumors), they should not be treated with a GLP-1 receptor agonist.^44,45 

There are several clinical situations in which patients with type 2 diabetes may benefit from treatment with a GLP-1 receptor agonist. This includes patients who are taking metformin, a sulfonylurea, or a thiazolidinedione (alone or in combination) but are not at their glycemic goal, as well as patients with an A1C of 7–9%. However, for patients with an A1C > 9%, the addition of basal insulin should be strongly considered until glycemic control has improved. At that time, a decision can be made as to whether to continue basal insulin. Of the three GLP-1 receptor agonists, liraglutide and exenatide BID are indicated for use in combination with basal insulin.^43,45 

Selection of a GLP-1 receptor agonist should be based on various clinical parameters. For example, if FBG is the primary target, exenatide QW or liraglutide are preferred over exenatide BID. Conversely, if PPG is the primary target, exenatide BID is preferred. Of course, other issues, including frequency of administration; side effects such as nausea, injection site reactions, and nodule formation; and patients' ability to use the administration devices are also important considerations.

It is important to realize that a small percentage of patients experience no or a minimal reduction in their blood glucose with a GLP-1 receptor agonist. Although the reason for this is unknown in most patients, in some it may be because they have been skipping doses or have reduced their dose. Because adverse events are a frequent reason for poor adherence, inquiring about medication difficulties such as adverse events at all follow-up visits can be helpful and provides an opportunity to find solutions that are acceptable to patients. It is most helpful to discuss with patients at the time of therapy initiation how to deal with common adverse events.

Other situations in which a GLP-1 receptor agonist might be a good choice include in older patients with type 2 diabetes, who are more likely to experience hypoglycemia unawareness.⁶⁰  In addition, limited data from clinical trials indicate that people ≥ 65 years of age experienced no difference in efficacy or safety compared to younger patients.^43–45  However, assessment of renal function before initiation of a GLP-1 receptor agonist is suggested because older patients may experience a reduction in renal function.

Patients with renal impairment or end-stage renal disease (creatinine clearance < 30 ml/min) should not use exenatide BID or exenatide QW. Liraglutide can be used without dosage reduction but should be used with caution because data are limited regarding its use in patients with various stages of renal impairment.

Another group for whom a GLP-1 receptor agonist might be a good choice includes patients who experience excessive hunger or weight gain. It is not uncommon for patients to describe a sensation of “always being hungry.” GLP-1 receptor agonists have been noted clinically by some providers to blunt that sensation.

Finally, although the U.S. Federal Aviation Administration does not allow insulin use for commercial pilots,⁶¹  and the U.S. Department of Transportation does not allow insulin use for commercial truck drivers,⁶²  GLP-1 receptor agonists are allowed for pilots and are not mentioned in the list of medications of concern for commercial truck drivers.

Patients generally should not be placed on concurrent treatment with a GLP-1 receptor agonist and a DPP-4 inhibitor (i.e., alogliptin, linagliptin, saxagliptin, or sitagliptin). Although there are differences between the two classes of medications, both act on the incretin system. There is, however, preliminary evidence that the addition of exenatide BID to the combination of sitagliptin plus metformin produces additional (0.3%) A1C reduction beyond the combination of exenatide BID and metformin over 20 weeks.⁶³  Nonetheless, patients who are taking a DPP-4 inhibitor should discontinue it at the start of GLP-1 therapy.

Some additional patients also should not be considered for GLP-1 receptor agonist treatment. Such a therapy would not be appropriate in patients who have severe GI disease (e.g., gastroparesis).^43,44  Patients who are pregnant or nursing should be excluded from using a GLP-1 receptor agonist, which are classified as pregnancy category C agents, unless the benefits outweigh the risks to the fetus.^43–45  Because exenatide is cleared primarily by the kidneys, blood concentrations of exenatide (BID or QW) are not expected to be altered in patients with hepatic impairment.^43,44  Liraglutide should be used cautiously, although no dose adjustment is recommended in patients with hepatic impairment.⁴⁵ 

Managing patient expectations from the outset is key to maximizing benefits and minimizing limitations associated with GLP-1 receptor agonist. One problem noted by the author with some frequency is that patients expect immediate results with this type of therapy. With exenatide BID and liraglutide, it is not uncommon to see glucose-lowering within the first few days. However, exenatide QW requires a few weeks to begin showing effects; maximal benefit may not be seen for up to 10 weeks.

In addition to patients' glycemic control expectations, it is especially important to manage their expectations with regard to weight. It is not uncommon for patients to ask about “those new medications that cause weight loss.” GLP-1 receptor agonists should not be presented to patients as weight loss drugs. It must be made clear that, although weight loss occurs in ~ 80% of patients,⁶⁴  it is not possible to identify which patients will lose weight before initiating the therapy.

In the author's experience, patients fit into one of three groups with respect to the weight effects of the GLP-1 receptor agonists. The first is the small percentage of patients who do not lose weight but have the expected 0.5–1.5% reduction in A1C. For these patients, the lack of a weight effect should not be viewed as a clinical failure. The second group is those who experience a 2- to 5-lb weight loss, and the third group includes those who have a greater weight loss, perhaps as much as 20–40 lb. Some patients in the third group subsequently regain some of their lost weight.

Although patients with a BMI > 30 or 35 kg/m² are generally those who experience the greatest weight loss,⁶⁵  this has not always been the case in the author's experience, in which some patients with a BMI < 30 kg/m² have been observed to lose 15–20 lb. The loss in weight usually plateaus in about 8 weeks with exenatide BID and 8–12 weeks with liraglutide.

Consideration of a GLP-1 receptor agonist should involve discussing with patients their comfort with delivering a medication subcutaneously. This can be a significant issue for patients and providers, with concerns similar to those that arise when considering starting insulin.⁶⁶  Patient concerns include having to use a needle and give a “shot,” the social stigma associated with using an injected medication, the perceived complexity of the delivery system, and the belief that this in some way signifies that the disease process is getting worse. Providers also have concerns, including feeling that it takes more time to educate patients about using a subcutaneous medication, the availability and qualification of staff to teach patients how to use these medications, and concerns that patients will not accept their recommendation to use a GLP-1 receptor agonist. Such concerns on the part of patients and providers can create an environment of reluctance to advance diabetes care, thereby making it difficult for patients to attain glycemic control.⁶⁷ 

Numerous strategies can be implemented to address potential patient barriers. First, it may be helpful to ask patients if they have any concerns or issues they would like to discuss. Many patients with type 2 diabetes have searched for information online, talked with neighbors, or have otherwise gained some secondhand knowledge about these medications.

Once patients' concerns are identified, it is much easier to address them. It may not be possible to address more than two or three issues during a clinic visit, but patients should be assured that remaining issues will be addressed in subsequent visits or, alternatively, by other members of the diabetes health care team.

It is also important to avoid suggesting that advancing to this (or any other) glucose-lowering medication signifies that the disease is getting worse. Rather it is important to make clear that diabetes is a progressive illness that requires medication adjustment to maintain management goals.

Clinicians may find it especially helpful to discuss with patients the benefits and limitations of GLP-1 receptor agonists, noting that they are much less likely than insulin therapy to cause hypoglycemia or weight gain. The potential of losing weight with a GLP-1 receptor agonist can be a strong motivator for patients who have struggled with weight gain while using other glucose-lowering agents. Similarly, learning that the risk of hypoglycemia is much lower than with insulin or a secretagogue may help relieve patients' anxiety. It can also be valuable to have patients check their FBG and PPG levels a few times for the week or two after starting GLP-1 receptor agonist therapy. Seeing these levels decrease can be another motivating factor that can blunt initial concerns regarding issues such as self-injection and GI adverse effects.

Two of the GLP-1 receptor agonists, liraglutide and exenatide QW, do not require dose timing with meals. Remembering to take a medication in advance of a meal, as is necessary for exenatide BID, can be challenging for some patients.

All three of the GLP-1 receptor agonists have administration devices/systems that make both the dosing and teaching of the medications relatively simple. Liraglutide and exenatide BID are delivered with pen devices with 31- or 32-gauge needles, whereas exenatide QW has a delivery kit that simplifies the mixing of diluent and powdered medication but requires a syringe with an 8-mm, 23-gauge needle. Despite the larger-gauge administration needle required with exenatide QW, some patients may decide that the once-weekly administration is a key benefit because they can select the day of the week that they want to administer their dose and stick with it. Other patients may decide they like the once-daily administration of liraglutide because they do not need to worry about timing their dose with a meal.

Because patients usually have concerns about giving themselves an injection, avoiding use of the word “shot” is suggested. Patients associate shots with painful antibiotic and vaccine injections they have had in the past. It is also helpful to acknowledge that not wanting to “stick holes” in themselves is a very rational decision. Differentiating subcutaneous administration from intramuscular injection is important. GLP-1 receptor agonists use much smaller, shorter needles and are delivered into the subcutaneous fat. They are not intramuscular injections.

It is often helpful to patients to make a connection between doing something that they perceived as unpleasant and attaining the goals that they want to attain. Although it is important to establish and follow achievement of the numeric goals of diabetes control, it is also important to emphasize the greater sense of well-being patients may feel when they achieve improved blood glucose control.

Addressing these issues and educating people with type 2 diabetes is a team process. The author typically discusses treatment options with patients, explaining the benefits and limitations of each choice. Once a decision is made, much of the detailed patient education is provided by clinic staff. For this reason, it is important to work with staff to develop a coordinated process that can be followed when patients begin therapy with a GLP-1 receptor agonist or other glucose-lowering therapy. It is important that the teaching is consistent from both providers and support staff when providing patient education. In addition, it is important for those involved in providing patient education to be knowledgeable about the disease and the treatments, including how each delivery device works. For the three GLP-1 receptor agonists, understanding—and demonstrating—the differences among the devices is crucial before presenting to patients.

Local diabetes education programs and certified diabetes educators can be quite helpful with this process. More information about how to find a local diabetes educator may be found at the American Association of Diabetes Educators website (http://www.diabeteseducator.org/DiabetesEducation/Find.html). In addition, a wide variety of patient education resources are available from the ADA (http://www.diabetes.org/living-with-diabetes/?loc=GlobalNavLWD) and AACE (http://resources.aace.com).

The GLP-1 receptor agonist class of medications offers health care providers an important novel treatment option with distinct benefits and limitations compared to other classes of glucose-lowering medications. Working in collaboration with patients, providers and their staff can individualize treatment and implement strategies to improve patient acceptance and self-management with a GLP-1 receptor agonist (Table 3). Doing so will undoubtedly help patients with type 2 diabetes attain their metabolic and treatment goals.

Funding for the development of this article was provided by Novo Nordisk to the Primary Care Education Consortium, which provided editorial assistance to the author. The author received no financial compensation for this article. The author independently made the decision to submit his article and is solely responsible for all content.