Type 2 diabetes and cardiovascular disease: risk reduction and early intervention

ABSTRACT

People with type 2 diabetes (T2D) have a higher risk of cardiovascular (CV) disease (CVD) than those without. This increased risk begins with pre-diabetes, potentially 7–10 years before T2D is diagnosed. Selecting medication for patients with T2D should focus on reducing the risk of CVD and established CVD. Within the last decade, several antihyperglycemic agents with proven CV benefit have been approved for the treatment of hyperglycemia and for the prevention of primary and secondary CV events, including glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium-glucose cotransporter-2 inhibitors. T2D treatment guidelines recommend that an antihyperglycemic agent with proven CV benefit should be used after metformin in patients with high risk of or established CVD, regardless of glycated hemoglobin levels. Despite the availability of antihyperglycemic agents with proven CV benefit, and guidelines on when to use them, less than one in four patients with T2D and CVD receive this type of therapy. These findings suggest a potential gap between current recommendations and clinical practice. This article reviews the approved agents with CV indications, with a focus on injectable GLP-1RAs, and their place in the T2D treatment paradigm according to current guidelines. We aim to provide primary healthcare providers with in-depth information on subsets of patients who would benefit from this type of therapy and when it should be initiated, taking into consideration safety and tolerability and other disease factors. An individualized treatment approach is increasingly recommended in the management of T2D, employing a shared decision-making strategy between patients and healthcare professionals.

Video Abstract

© 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group

KEYWORDS:

• Cardiovascular disease
• incretin therapy
• antidiabetic drug
• diabetes complications
• type 2 diabetes

1 Introduction

Type 2 diabetes (T2D) is characterized by elevated blood glucose levels and is associated with an increased risk of developing micro- and macrovascular comorbidities, such as diabetic retinopathy, renal disease, and cardiovascular (CV) disease (CVD) [1]. Indeed, a recent study (CAPTURE) of 9,823 patients with T2D in 13 countries reported that 34.8% of patients attending a primary or specialist routine healthcare visit had established CVD, with most cases (85.8%) classified as atherosclerotic CVD (ASCVD) [2]. Timely initiation or intensification of appropriate therapy is important in patients with T2D, as extended periods of time with poor glycemic control has been associated with the increased risk of macrovascular events [3]. Moreover, guidelines advise that an antihyperglycemic agent with proven CV benefit should be initiated after metformin in patients with high risk of or established CVD, regardless of their glycated hemoglobin (HbA_1c) level [4–6].

There are several antihyperglycemic agents approved for use in patients with, or at risk of, CVD, including glucagon-like peptide-1 (GLP-1) receptor agonists (GLP-1RAs) and sodium-glucose cotransporter-2 inhibitors (SGLT2is) [4–6]. For the purpose of this review, we will focus on GLP-1RAs with a CV indication in the US for the management of patients with T2D who have CVD or CVD risk factors. Dulaglutide, liraglutide, and semaglutide are injectable GLP-1RAs that are approved to reduce the risk of major adverse CV events (MACE) compared with placebo in patients with T2D [7–9].

Despite the availability of antihyperglycemic agents with proven CV benefit, and guidelines on when they should be used, less than one in four adults with T2D in the CAPTURE study received such agents, irrespective of CVD status [10]. These findings indicate a gap between clinical practice and current T2D treatment guidelines. To help bridge this gap, the following review aims to identify subsets of patients who would benefit from a GLP-1RA with proven CV benefit and provide practical guidance on how and when to initiate it.

2 Literature search

We searched PubMed and Embase databases using the terms glucagon-like peptide-1 receptor agonists; type 2 diabetes; first or initial injectable therapy; insulin naïve; add on or combination; oral antidiabetic; inadequate control; intensification; switch; cardiovascular disease; atherosclerotic cardiovascular disease; early treatment. Records were limited to those in English language (N = 248 after duplicates were removed). Records were excluded during screening for the following reasons: not in a relevant drug/indication/population; preclinical study; reviews; reporting data on parameters considered out of the scope of this manuscript; not a randomized trial (N = 231). Records included in this review N = 17. Bibliographies of retrieved articles were used to source additional articles of relevance. US-based guidelines for the management of T2D were used for discussion on current standards of care. These included guidelines and standards of care published by the American Diabetes Association (ADA), American Association of Clinical Endocrinologists and American College of Endocrinology (AACE/ACE), and the American College of Cardiology (ACC). Searches were last updated 4 January 2020.

3 Cardiovascular benefits of injectable human glucagon-like peptide-1 receptor agonists

In three CV outcomes trials (CVOTs), dulaglutide, liraglutide, and semaglutide were found to reduce CV morbidity and mortality in patients with T2D with established CVD or risk factors for CVD [11–13]. In all three trials, the risk of MACE was reduced compared with placebo/standard of care (SoC), and these agents are currently indicated for use in patients with T2D and established CVD [7–9]. In addition, dulaglutide is indicated in patients with T2D and multiple CVD risk factors [8]. Due to their CV indications, these agents are not limited to second-line therapy and their use is not dependent on HbA_1c targets [7–9].

3.1 Dulaglutide

Dulaglutide is an injectable human GLP-1RA indicated as an adjunct to diet and exercise in adults with T2D. Doses are administered once weekly starting at 0.75 mg, increasing to 1.5 mg then 3.0 mg, to a maximum dose of 4.5 mg in 4-week increments, based on the level of glycemic control required [8] (Figure 1). The CV safety of dulaglutide was assessed in the REWIND trial, which had a median follow-up time of 5.4 years [11].

Figure 1. Administration and dose-escalation information for GLP-1RAs with proven CVD benefit [7–9]. CVD, cardiovascular disease; GI, gastrointestinal; GLP-1RA, glucagon-like peptide-1 receptor agonist.

The REWIND trial enrolled a broad range of male and female patients with T2D (N = 9,901). Patients were either aged ≥50 years and had established CVD, or aged ≥60 years and did not have established CVD but had multiple CV risk factors including tobacco use, dyslipidemia, hypertension, or abdominal obesity [11]. Established CVD was defined as history of at least one of the following: myocardial ischemia; ischemic stroke; coronary, carotid, or peripheral artery revascularization; unstable angina; or hospitalization for unstable angina [11]. The addition of dulaglutide 1.5 mg to existing antihyperglycemic treatment (up to two oral antihyperglycemic drugs [OADs] with or without basal insulin) and CV SoC led to significantly fewer patients experiencing the composite outcome of nonfatal myocardial infarction (MI), nonfatal stroke, or death from CV causes, compared with those who received placebo/SoC alone (hazard ratio [HR], 0.88; 95% confidence interval [CI]: 0.79, 0.99; P = 0.026). These results were mostly driven by the reduction in the individual MACE component of nonfatal stroke. HRs for the individual MACE components were 0.91 (95% CI: 0.78, 1.06; P = 0.21) for CV death, 0.96 (95% CI: 0.79, 1.16; P = 0.65) for nonfatal MI and 0.76 (95% CI: 0.61, 0.95; P = 0.017) for nonfatal stroke [11].

Notably, only 31.5% of patients in the REWIND trial had previous CVD, as defined above, suggesting that patients older than 60 with multiple risk factors for CVD could benefit from early treatment with dulaglutide to prevent primary CV events [11]. The fact that mean baseline HbA_1c was 7.2% in the REWIND trial suggests that the CV benefits observed could be attributed to the drug rather than a dramatic improvement in glycemia [11].

Several cardiometabolic parameters were improved in patients receiving dulaglutide compared with placebo in the REWIND trial, including HbA_1c, body weight, body mass index (BMI), systolic blood pressure, arterial blood pressure, pulse pressure, total cholesterol, low-density lipoprotein (LDL) cholesterol, and waist-to-hip circumference [11] (Table 1). Some of these cardiometabolic markers are thought to be predictors of CVD [14], and the reduction in MACE in the REWIND trial could, at least in part, be attributed to the improvements seen in these parameters.

Table 1. Summary of CV risk factor outcomes in GLP-1RA CVOTs.

	Dulaglutide Gerstein et al. [11]			Liraglutide Marso et al. [13]		Semaglutide Marso et al. [12]
Outcome (change from baseline)	Dulaglutide 1.5 mg	Placebo	Liraglutide 1.8 mg		Placebo		Semaglutide 0.5 mg	Placebo0.5 mg	Semaglutide 1.0 mg	Placebo 1.0 mg
BMI, kg/m^2	–1.08 (0.03)	–0.55 (0.03)	NR		NR		NR	NR	NR	NR
Difference, P value	–0.53 (–0.61, – 0.46), P < 0.0001		NR				NR		NR
Systolic BP, mm Hg	–3.15 (0.13)	–1.44 (0.13)	-		-		–3.44	–2.17	–5.37	–2.78
Difference, P value	–1.70 (–2.07, – 1.33), P < 0.0001		–1.2 (–1.9, – 0.5)				–1.27 (–2.77, 0.23), P = NS		–2.59 (–4.09, – 1.08), P < 0.001
Diastolic BP, mm Hg	–1.93 (0.08)	–2.05 (0.08)	-		-		–1.37	–1.42	–1.57	–1.71
Difference, P value	0.12 (–0.10, 0.34), P = 0.28		0.6 (0.2, 1.0)				0.04 (–0.83, 0.92), P = NS		0.14 (–0.74, 1.03), P = NS
Total cholesterol, mM	–0.20 (0.01)	–0.13 (0.01)	NR		NR		0.97	1.00	0.97	0.99
Difference, P value	–0.07 (–0.10, –0.03), P = 0.0002		NR				0.97 (0.95, 1.00), P < 0.05^a		0.99 (0.97, 1.01), P = NS^a
LDL cholesterol, mM	–0.15 (0.01)	–0.09 (0.01)	NR		NR		0.97	1.01	0.98	0.99
Difference, P value	–0.05 (–0.08, – 0.02), P = 0.001		NR				0.96 (0.93, 0.99), P < 0.05^a		0.99 (0.96, 1.03), P = NS^a
HDL cholesterol	-	-	NR		NR		0.99	0.99	1.01	0.97
Difference, P value	-		NR				1.00 (0.99, 1.02), P = NS^a		1.04 (1.02, 1.06), P < 0.0001^a
Waist:hip (women)	–0.005 (0.001)	0.0 (0.001)	NR		NR		NR	NR	NR	NR
Difference, P value	–0.005 (–0.008, – 0.0012), P = 0.009		NR				NR		NR
Waist:hip (men)	–0.007 (0.001)	–0.004 (0.001)	NR		NR		NR	NR	NR	NR
Difference, P value	–0.003 (–0.006, – 0.0001), P = 0.04		NR				NR		NR

Data are not comparable due to differences between the trials in patient populations and analysis methods.

^aLipid parameters are ratio to baseline (mg/dL) and were analyzed on log-scale with means at baseline presented as geometric means.

BMI, body mass index; BP, blood pressure; CV, cardiovascular; CVOTs, CV outcomes trials; GLP-1RA, glucagon-like peptide-1 receptor agonist; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NR, not reported; NS, non-significant.

3.2 Liraglutide

Liraglutide is an injectable human GLP-1RA indicated as an adjunct to diet and exercise in patients aged ≥10 years with T2D. For the purpose of this review, the pediatric indication will not be discussed further. For adults with T2D, once-daily doses of liraglutide start at 0.6 mg, increasing to 1.2 mg, then 1.8 mg in weekly increments, if additional glycemic control is required [9] (Figure 1). The LEADER trial assessed the CV safety of liraglutide and had a median follow-up time of 3.8 years [13].

In the LEADER trial, 9,340 male and female patients with T2D aged ≥50 years with at least one preexisting CV condition were enrolled. In addition, patients aged ≥60 years were included if they had at least one CV risk factor. Definitions of both the preexisting CVD and the risk factors differed somewhat from the REWIND trial. For example, the prior CVD definition included the presence of chronic kidney disease or stenosis ≥50%. Risk factors included microalbuminuria or proteinuria, hypertension and left ventricular hypertrophy, left ventricular systolic or diastolic dysfunction, or an ankle-brachial index of <0.9 [13].

In treatment-naïve patients, or those taking one or more OAD or insulin, the incidence of first occurrence of the composite outcome of death from CV causes, nonfatal MI, or nonfatal stroke was significantly lower with liraglutide than with placebo/CV SoC (HR, 0.87; 95% CI: 0.78, 0.97; P < 0.001 for noninferiority; P = 0.01 for superiority). The incidence was also lower for all-cause mortality (HR, 0.85; 95% CI: 0.74, 0.97; P = 0.02) and death from CV causes (HR, 0.78; 95% CI: 0.66, 0.93; P = 0.007). Nonfatal MI and nonfatal stroke occurred less frequently with liraglutide than with placebo/CV SoC but without statistical significance (HR, 0.88; 95% CI: 0.75, 1.03; P = 0.11 for nonfatal MI and HR, 0.89; 95% CI: 0.72, 1.11; P = 0.30 for nonfatal stroke) [13].

HbA_1c was reduced by 0.40 percentage points, and body weight and systolic blood pressure were improved with liraglutide versus placebo/CV SoC [13] (Table 1). The trial did not report other cardiometabolic parameters assessed in the REWIND trial such as BMI, arterial blood pressure, pulse pressure, total cholesterol, LDL cholesterol, and waist-to-hip circumference. Thus, it is not possible to compare these parameters between the REWIND and LEADER trials.

A post hoc analysis of LEADER was conducted to determine if the reduced risk of first MACE, as measured by the primary hypothesis [13], would remain with first and subsequent events included (total MACE) [15]. There was a 15.7% relative risk reduction for total MACE with liraglutide compared with placebo (HR, 0.84; 95% CI: 0.76, 0.93), suggesting that patients with a high risk of CV events could benefit from treatment with this GLP-1RA [15].

3.3 Semaglutide

Semaglutide is a once-weekly injectable human GLP-1RA indicated as an adjunct to diet and exercise in adults with T2D. The starting dose is 0.25 mg, increasing to 0.5 mg after 4 weeks, then to 1.0 mg [7]. If additional glycemic control is required, a dose of 2.0 mg can be initiated after 4 weeks on 1.0 mg [7] (Figure 1). To evaluate the CV safety of semaglutide, a pre-approval SUSTAIN 6 CVOT was conducted over 2.1 years [12].

SUSTAIN 6 enrolled 3,297 patients ≥50 years of age who were treatment-naïve, or who had been treated with up to two OADs with or without basal or premixed insulin. Patients aged ≥50 years were included if they had established CVD, while patients aged ≥60 years were included if they had at least one CV risk factor. Prior CVD and risk factors differed from the REWIND trial [11] but were similar to those in the LEADER trial [12].

Since this was a pre-approval trial, it was much shorter and enrolled fewer patients than the REWIND and LEADER CVOTs, and it was not prespecified to show superiority. However, the accrual of more events and a much larger observed effect between treatment arms than estimated led to a statistically significant result showing lower risk of the primary outcome in patients treated with semaglutide than with placebo/CV SoC. In the overall trial population, the composite outcome of CV death, nonfatal MI, or nonfatal stroke was significantly lower among patients receiving semaglutide than among those receiving placebo/CV SoC (HR, 0.74; 95% CI: 0.58, 0.95; P < 0.001 for noninferiority; P = 0.02 for superiority) [12]. Similar to the findings in the REWIND trial [11], the reduction in the composite outcome for semaglutide was mainly driven by a significant reduction in nonfatal stroke. The individual MACE component HRs were 0.74 (95% CI: 0.51, 1.08; P = 0.12) for nonfatal MI, 0.61 (95% CI: 0.38, 0.99; P = 0.04) for nonfatal stroke, and 0.98 (95% CI: 0.65, 1.48; P = 0.92) for CV death [12].

Significant improvements in cardiometabolic risk factors were observed with semaglutide versus placebo. HbA_1c and body weight were reduced with semaglutide 0.5 mg and 1.0 mg versus placebo. Other improvements with semaglutide 0.5 mg versus placebo included total cholesterol and LDL cholesterol. There were also reductions in systolic blood pressure, high-density lipoprotein cholesterol, triglycerides, and free fatty acids with semaglutide 1.0 mg versus placebo [12] (Table 1).

A post hoc analysis of SUSTAIN 6 was conducted to assess the effects of sex, age (50–65 years and >65 years), and baseline CV risk (prior MI or stroke vs no prior MI or stroke, and established CVD vs CV risk factors alone, including chronic kidney disease) on CV outcomes. Across all subgroups, semaglutide consistently reduced the risk of the first occurrence of MACE and individual MACE components compared with placebo/CV SoC. In addition, revascularization, HbA_1c, and body weight were all reduced with semaglutide versus placebo/CV SoC. These results support the use of semaglutide for the reduction of MACE in patients of different ages, sexes, and CV status [16].

Two recent meta-analyses analyzed the effects of all GLP-1RAs on CV outcomes compared with placebo/CV SoC [17,18]. In these analyses, which each included 56,004 patients, the combined effect of GLP-1RAs (including medications with and without US Food and Drug Administration indication for the reduction of MACE: lixisenatide, liraglutide, subcutaneous semaglutide, exenatide once weekly, albiglutide, dulaglutide, and oral semaglutide) was found to be beneficial on kidney outcomes, hospitalization for heart failure, and CV and all-cause mortality. The risk of MACE was reduced by 12–13% and there was a 9–17% risk reduction for all-cause mortality, kidney outcomes, and hospitalization for heart failure (Table 2). The incidence of macroalbuminuria was also reduced without affecting the progression of diabetic renal disease [17,18]. Furthermore, Kristensen et al. reported that there was no increase in risk of severe hypoglycemia, pancreatitis, or pancreatic cancer versus placebo [17].

Table 2. Key data reported in meta-analyses of GLP-1RA CVOTs.

	Meta-analyses of GLP-1RAs vs placebo
Outcome, HR [95% CI]; P value	Kristensen et al. [17]	Giugliano et al. [18]
MACE	0.88 [0.82, 0.94]; P < 0.001	0.87 [0.80, 0.96]; P < 0.011
All-cause mortality	0.88 [0.83, 0.95]; P = 0.001	0.89 [0.79, 0.99]; P < 0.034
Hospitalization for heart failure	0.91 [0.83, 0.99]; P = 0.028	0.91 [0.86, 0.97]; P = 0.013
Renal^a	0.83 [0.78, 0.89]; P < 0.001	0.83 [0.69, 1.00]; P < 0.047

GLP-1RAs included in the meta-analyses were lixisenatide, liraglutide, s.c. semaglutide, exenatide, albiglutide, dulaglutide, and oral semaglutide (N = 56,004).

^aComposite kidney outcome including macroalbuminuria consisted of development of macroalbuminuria, doubling of serum creatinine or 40% or greater decline in estimated glomerular filtration rate, development of end-stage kidney disease, or death due to kidney disease.

CI, confidence interval; CVOT, cardiovascular outcomes trial; GLP-1RA, glucagon-like peptide-1 receptor agonist; HR, hazard ratio; MACE, major adverse cardiovascular events; s.c., subcutaneous.

3.4 Physiological cardiovascular benefits with glucagon-like peptide-1 receptor agonists

There is a growing body of evidence to elucidate the potential mechanisms of action for the observed CV benefits of GLP-1RAs. In studies involving rat models and human subjects, improvements in endothelial function, cardiac function, and various CVD risk factors including systolic blood pressure and postprandial dyslipidemia have been observed [19–25].

Preclinical trials suggest that native GLP-1 attenuates the progression of atherosclerosis, vascular inflammation, and vasoconstriction, demonstrating the cardioprotective role of GLP-1 and thus the potential CV benefits of GLP-1RAs [23–25]. In a clinical trial using infused exogenous GLP-1, endothelial function was improved in patients with T2D and coronary artery disease [19]. In addition, liraglutide improved cardiac function in patients with T2D [26] and with T2D and heart failure [20]. Liraglutide has also been shown to reduce some risk factors for CVD in patients with T2D, including ambulatory systolic blood pressure in patients with hypertension [21], and also improve postprandial dyslipidemia and reduce postprandial atherogenic remnant particles and triglyceride-rich lipoproteins [22].

Collectively, this body of evidence supports the use of injectable GLP-1RAs in patients with CVD, and in those with CVD risk factors, to prevent primary and secondary CV events [11–13,15]. Since T2D is a multifactorial disease with varying characteristics that require different treatment approaches, other aspects such as body weight and hypoglycemia should be considered in addition to CV status when selecting an injectable therapy for individual patients [4–6].

3.5 Cardiovascular benefits of sodium-glucose cotransporter-2 inhibitors

The effects of SGLT2is (canagliflozin, dapagliflozin, empagliflozin, and ertugliflozin) on CV outcomes have been analyzed in a meta-analysis of six placebo-controlled CVOTs in patients with T2D, with or without established CVD (N = 46,969) [27]. Overall, the risk of MACE was significantly reduced with SGLT2i treatment compared with placebo (HR, 0.90; 95% CI: 0.85, 0.95; Q statistic P = 0.27). The presence or absence of CVD did not modify the treatment outcome on MACE (HR, 0.89; 95% CI: 0.84, 0.95; and HR, 0.94; 95% CI: 0.83, 1.07; P = 0.63 for interaction) [27]. For the individual components of MACE, SGLT2is significantly reduced the risk of CV death (HR, 0.85; 95% CI: 0.78, 0.93; Q statistic P = 0.02) and all-cause mortality (HR, 0.87; 95% CI: 0.81, 0.93; Q statistic P = 0.06) but not stroke (HR, 0.96; 95% CI: 0.87, 1.07; Q statistic P = 0.29).

4 Other considerations for injectable glucagon-like peptide-1 receptor agonists

4.1 Body weight

The need to reduce body weight might be an important consideration for some patients with T2D. In all three CVOTs of the injectable GLP-1RAs with a proven CV benefit, there were larger reductions in body weight compared with placebo/CV SoC for dulaglutide (least square means 1.46 kg), liraglutide (2.3 kg), and semaglutide (2.9 kg [0.5 mg]; 4.3 kg [1.0 mg]) [11–13]. These findings suggest that these three agents with a CV indication could be appropriate in patients with a compelling need to lose weight.

Reductions in body weight were greater in patients who received semaglutide, liraglutide, or dulaglutide versus basal insulin glargine U100 [28–30]. Patients receiving semaglutide 0.5 mg and 1.0 mg had body weight reductions of 3.47 kg and 5.17 kg, respectively, while patients receiving insulin glargine had a weight gain of 1.15 kg [28]. Patients receiving liraglutide had mean weight loss of 1.8 kg while those receiving insulin glargine had a mean weight gain of 1.6 kg [29]. Reductions in body weight for dulaglutide (1.33 kg [0.75 mg]; 1.87 kg [1.5 mg]) were similar to those observed with liraglutide and patients receiving insulin glargine also gained weight in this trial (1.44 kg) [30]. The reductions in body weight observed with GLP-1RAs compared with insulin glargine indicate that when initiating injectable therapy, a GLP-1RA would be more beneficial than basal insulin when body weight is a concern.

4.2 Hypoglycemia

Hypoglycemia is an important consideration when selecting a therapy for the management of T2D. Since the HbA_1c-lowering effects of GLP-1RAs are glucose-dependent [31], hypoglycemia is not common with this class of therapy [32–35]. However, the risk of hypoglycemia increases when GLP-1RAs are used in combination with insulin or insulin secretagogues [7–9]. If there is a need to reduce the risk of hypoglycemia when selecting therapy for T2D, GLP-1RAs are generally favorable to insulin [6].

In a placebo-controlled trial for dulaglutide 1.5 mg, 3.0 mg, and 4.5 mg added on to metformin, incidences of hypoglycemia (glucose level <54 mg/dL) through 36 weeks of treatment were 1.1%, 0.3%, and 1.1%, respectively. Incidences of severe hypoglycemia (defined as requiring the assistance of another person to administer carbohydrate, glucagon, or other resuscitative actions) were 0.2%, 0.0%, and 0.2% [8]. For liraglutide 0.6 mg, 1.2 mg, and 1.8 mg as add-on to metformin, the rate of hypoglycemia that did not require third-party assistance for treatment was 0.05 events/year, and the rate of events that did require third-party assistance for treatment was 0.001 events/year [9]. In a monotherapy trial of semaglutide 0.5 mg and 1.0 mg there were no severe or blood glucose-confirmed symptomatic events (glucose level ≤56 mg/dL) reported [7].

5 Safety and tolerability of glucagon-like peptide-1 receptor agonists

5.1 Common adverse events

Gastrointestinal (GI) symptoms (including nausea, vomiting, diarrhea, and constipation) are the most common side effects of GLP-1RA therapy (predominantly nausea) (Table 3). In a pool of placebo-controlled trials for dulaglutide 0.75 mg and 1.5 mg, nausea was reported in 12.4% and 21.1% of patients compared with 5.3% of patients receiving placebo. In general, the incidence of nausea does not appear to occur in a dose-dependent manner. Comparable rates of nausea were reported in patients receiving higher doses of dulaglutide (15.6% of patients receiving dulaglutide 3.0 mg and 16.4% of patients receiving dulaglutide 4.5 mg) [8]. The proportions of patients reporting nausea were similar for liraglutide 1.2 mg and 1.8 mg (18.0% and 20.0%) [9], and semaglutide 0.5 mg and 1.0 mg (15.8% and 20.3%) [7] (Table 3). Furthermore, proportions of patients reporting nausea were similar with semaglutide 1.0 mg (14.6%) and 2.0 mg (14.4%) in SUSTAIN FORTE [36].

Table 3. Management of common GI AEs associated with GLP-1RAs with proven CV benefit.

	Percentage of patients with AE in placebo-controlled clinical trials									Management strategies
	Placebo [8]	Dulaglutide [8]		Placebo [9]	Liraglutide [9]		Placebo [7]	Semaglutide [7]		Implement slower dose escalation if required [7–9] Avoid eating food that is high in fat [37] Smaller meal sizes [46] Avoid eating beyond the feeling of being full [7–9]
		0.75 mg	1.5 mg		1.2 mg	1.8 mg		0.5 mg	1.0 mg
N	568	836	834	661	645	1,024	262	260	261
Any GI AEs, %	21.3	31.6	41.0	-	-	-	15.3	32.7	36.4
Nausea	5.3	12.4	21.1	5.0	18.0	20.0	6.1	15.8	20.3
Vomiting	2.3	6.0	12.7	2.0	6.0	9.0	2.3	5.0	9.2
Diarrhea	6.7	8.9	12.6	4.0	10.0	12.0	1.9	8.5	8.8

Data pooled across placebo-controlled randomized trials for each agent. Direct comparisons between agents may not be possible.

AE, adverse event; CV, cardiovascular; GI, gastrointestinal; GLP-1RA, glucagon-like peptide-1 receptor agonist.

Overall, in clinical trials for dulaglutide, liraglutide, and semaglutide, GI adverse events (AEs) were mostly mild to moderate in severity and tended to be transient, occurring mostly during dose escalation and not usually leading to treatment discontinuation [33–35].

5.2 Infrequent and rare adverse events

Diabetic retinopathy complications (DRCs), defined as vitreous hemorrhage, blindness, or the need for treatment with an intravitreal agent or photocoagulation, were assessed in the CVOTs for liraglutide and semaglutide [12,13]. Proportions of patients reporting DRCs were similar between liraglutide (2.3%) and placebo (2.0%) [13]. A greater difference in the rates of DRCs among patients receiving semaglutide compared with placebo in the SUSTAIN 6 trial was noted (1.49 vs 0.86 incidence rate per 100 patient years, respectively) [12]. This may have been due to rapid lowering of HbA_1c in the initial 16 weeks of the trial [12], since intensive glucose lowering has been associated with a temporary worsening of diabetic retinopathy, particularly in patients with long-standing poor glycemic control [38]. Retinopathy was assessed according to different criteria in REWIND compared with LEADER and SUSTAIN 6 and occurred in 1.9% of patients receiving dulaglutide and 1.5% of those receiving placebo [11].

For both dulaglutide and semaglutide, the proportions of patients with DRCs were larger among patients with a history of diabetic retinopathy at baseline (dulaglutide 8.5%, placebo 6.2%; semaglutide 8.2%, placebo 5.2%) compared with patients without a known history of diabetic retinopathy (dulaglutide 1.0%, placebo 1.0%; semaglutide 0.7%, placebo 0.4%) [7,8]. Despite the overall low risk of diabetic retinopathy, patients with a history of diabetic retinopathy should be monitored for disease worsening when initiating treatment with dulaglutide or semaglutide [7,8].

There was a very low incidence of acute pancreatitis in clinical studies of GLP-1RAs [33–35], and two meta-analyses of GLP-1RA CVOTs have found that the risk of acute pancreatitis is not increased with GLP-1RA treatment [39,40]. Patients with a history of chronic or acute pancreatitis were excluded from GLP-1RA trials [11–13], therefore it is not established if patients with a history of these disorders are at an increased risk of developing pancreatitis. GLP-1RA treatment should be discontinued if pancreatitis is suspected, and not restarted if it is confirmed [7–9].

Although only based on nonclinical observations, GLP-1RAs are contraindicated in patients with a personal or family history of medullary thyroid cancer (MTC) or with multiple endocrine neoplasia syndrome type 2 [7–9]. These conditions have very rarely been observed in humans receiving GLP-1RAs [7–9]. Nevertheless, patients should be counseled regarding the potential risk for MTC with the use of GLP-1RA therapy and informed of symptoms of thyroid tumors (for example, a mass in the neck, dysphagia, dyspnea, persistent hoarseness) [7–9].

6 Practical guidance on initiating a glucagon-like peptide-1 receptor agonist with proven cardiovascular benefit, and management strategies for gastrointestinal adverse events

The current ADA guidelines recommend that blood pressure should be measured at every routine clinical visit with a patient with T2D since hypertension is a major risk factor for ASCVD [41]. The ACC recommend that CV clinicians screen patients for T2D, since many patients cared for in this setting have known T2D, undiagnosed T2D, or prediabetes [42]. In addition to routine blood pressure measurements, it is recommended to use the ACC/American Heart Association ASCVD risk calculator to estimate 10-year risk of a primary ASCVD event [43]. Therapeutic options for treating CVD risk factors such as hypertension and dyslipidemia are based on the individual profiles and are beyond the scope of this review. For information on treating CVD risk factors, we guide readers to the ADA guidelines on CVD risk management [41]. Based on clinical assessment, a GLP-1RA or SGLT2i with proven CV benefit is recommended in patients with T2D and established ASCVD or established kidney disease as part of their CV risk reduction and/or antihyperglycemic regimen [41]. For patients with multiple risk factors for ASCVD, a GLP-1RA or SGLT2i with proven CV benefit is recommended to reduce the risk of MACE [4,5,41]. Initiating a GLP-1RA or SGLT2i with proven CV benefit in these subsets of patients should be based on their CVD status rather than HbA_1c measurements [4–6] (Figure 2).

Figure 2. Treatment decision-making infographic [4–6].

^aIf SGLT2i is not tolerated or contraindicated. ASCVD, atherosclerotic cardiovascular disease; CKD, chronic kidney disease; GLP-1RA, glucagon-like peptide-1 receptor agonist; HbA_1c, glycated hemoglobin; HF, heart failure; OAD, oral antihyperglycemic drug; SGLT2i, sodium-glucose cotransporter-2 inhibitor; T2D, type 2 diabetes.

Although our focus is on pharmacotherapy for reducing the risk of MACE in people with T2D, it is worth noting the importance of non-pharmacological interventions as part of an early approach to preventing CV events. Patients with metabolic syndrome are likely to have elevated levels of tumor necrosis factor (TNF)-alpha (a marker of inflammation), which are further increased after a high-fat meal in association with endothelial dysfunction [47]. Early prevention of increasing levels of TNF-alpha and worsening of endothelial function should also be part of CV risk prevention. Lifestyle management continues to be an important strategy for reducing risks for T2D, hypertension and dyslipidemia, weight loss, increased physical activity, and medical nutrition therapy allows some patients to reduce ASCVD risk factors [44,45]. Use of internet and mobile-based digital platforms may help to reinforce healthy behaviors [44].

Management strategies can be employed to help mitigate potential GI AEs such as nausea, vomiting, and diarrhea [46]. These strategies include slow dose escalation, especially when initiating a GLP-1RA [46] (Table 3). For semaglutide and dulaglutide, which are administered once weekly, dose escalation is recommended in at least 4-week increments to help alleviate potential GI AEs [7,8] (Figure 1). Dose escalation in 1-week increments is recommended for liraglutide, a once-daily GLP-1RA [9] (Figure 1). But again, slower dose escalation may be indicated. The administration guidance for liraglutide also recommends starting at the lowest dose in the case of a missed dose for more than 3 days to mitigate possible GI effects as a result of re-initiation [9] (Figure 1). In the authors’ clinical experience, people who have had GI issues with metformin may benefit from a slower dose escalation of GLP-1RA and/or reduction of metformin at the time of GLP-1RA initiation. Signs of constipation should be monitored due to decreased gastric emptying and there may be a need to increase fiber and fluids.

Patients may experience loss of appetite after initiating treatment with a GLP-1RA, which could lead to nausea if the patient continues to eat past the feeling of being full. Ways in which nausea can be minimized include reducing meal size and not eating beyond the feeling of being full, and avoiding fatty foods for a few weeks (Table 3). Incorporating such strategies should be part of a shared decision-making process between healthcare providers and their patients [4].

Patient engagement is important in managing AEs and healthcare providers should help patients to understand what AEs are possible and, if they occur, their likely reduction over time and how they may be managed. A shared decision-making approach between the patient with T2D and the healthcare professional may improve patient engagement with treatment and managing AEs. The language used by healthcare providers when discussing diabetes with their patients is important in this context. A joint document between the American Association of Diabetes Educators (since renamed the Association of Diabetes Care and Education Specialists) and the ADA providing recommendations for language to enhance communication is available [48] and is a useful tool when educating patients on appropriate management of AEs. This includes a focus on using non-judgmental and fact-based language to describe a patient’s condition (they are a person with diabetes rather than a diabetic, for example), and to avoid making patients feel that they are failing if they are having issues with treatment adherence or reaching goals [48].

7 The place of glucagon-like peptide-1 receptor agonists in the treatment of type 2 diabetes

The ADA, ACC, and the AACE/ACE all recommend that an antihyperglycemic agent with proven CV benefit should be used in patients with, or at risk of, CVD. These agents include GLP-1RAs (injectable semaglutide, dulaglutide, or liraglutide) and SGLT2is (oral canagliflozin, dapagliflozin, or empagliflozin) [4–6]. There are differences in terms of the first-line therapy among the guidelines. The ACC guidelines discuss the possibility of using a GLP-1RA or SGLT2i as first-line therapy (i.e. before metformin) in patients with newly diagnosed, treatment-naïve T2D and with established CVD or risk factors for CVD [4]. This approach is supported by the European Society of Cardiology guidelines [49]; however, the ADA and AACE/ACE guidelines currently state that GLP-1RAs or SGLT2is should be first-line therapy only in patients where metformin cannot be used. At the same time, the ADA and AACE/ACE guidelines recommend initiating a GLP-1RA or SGLT2i in patients with, or at risk of, CVD regardless of HbA_1c target, suggesting these agents should be given to such patients even if further add-on therapy is needed for glycemic control [4–6] (Figure 2).

In addition to reducing the risk of MACE, there are some additional factors to consider when selecting which agent to initiate first, either a GLP-1RA or an SGLT2i. If there is a need to reduce the risk of heart failure and/or diabetic kidney disease, an SGLT2i is recommended initially. A combination of both a GLP-1RA and an SGLT2i can also be beneficial when additional glycemic control is required. GLP-1RAs are recommended for treatment intensification prior to insulin therapy in patients who have not achieved their glycemic target and have a compelling need to lose weight and/or avoid hypoglycemia [4,6].

7.1 Timing of glucagon-like peptide-1 receptor agonists treatment: the earlier the better

Given that patients with T2D are at increased risk of CVD, and many already have indicators of CVD at the time of T2D diagnosis [50], it is important that appropriate treatment is not delayed. This is highlighted by Paul et al., who reported a significant increase in the risk of MI, heart failure, stroke, and a composite of these three events when treatment intensification was delayed by 1 year in conjunction with HbA_1c levels above 7.0% [3]. Another study found that time to treatment intensification with injectable therapy (insulin or GLP-1RA) decreased with increasing HbA_1c targets [51]. This finding suggests that initiating GLP-1RAs in primary or specialist care might be considered HbA_1c-dependent. However, T2D treatment guidelines state that a therapeutic agent with proven CV benefit should be used after metformin, regardless of HbA_1c, in patients with, or at risk of, CVD [4–6].

Patients should be assessed for their CV risk (moderate, high, very high) to inform choices between primary and secondary CV prevention. In the presence of organ damage or multiple CV risk factors, a patient in primary CV prevention can be at very high CV risk, and therefore requires the same intensified therapy as a patient in secondary CV prevention [49]. While treatment for T2D that reduces the risk of CV events should be initiated as early as possible, there is evidence emerging that intensive multifactorial therapy can reduce MACE and mortality early after initiation with durable effects [52].

T2D with ASCVD comorbidity is associated with increased healthcare utilization costs compared with T2D without ASCVD, arising from increases in medical and pharmacy costs [53]. Simplified treatment regimens could reduce associated healthcare utilization costs. In addition, early intervention with GLP-1RAs has the potential to reduce healthcare costs due to improved longer-term health outcomes [54]. When considering the economic impact of treatment intensification, the initial cost of the drug should be offset against total healthcare costs associated with complications that may occur when glycemic targets are not met, or when CV events occur [55,56].

7.2 Concomitant therapy

Patients with T2D may benefit from simplified treatment regimens since the presence of multimorbidity can result in many different treatments being prescribed. In the SUSTAIN 6 and LEADER CVOTs, greater proportions of patients in the placebo groups than in the semaglutide and liraglutide groups received additional T2D medications during the trials [12,13]. Notably, significant differences were observed for the addition of insulin (basal and basal-bolus), sulfonylureas, and thiazolidinediones [12,13]. Insulin and sulfonylureas can increase the risk of hypoglycemia and weight gain while thiazolidinediones are associated with an increased risk of weight gain [5]. Patients receiving such medications could benefit from switching to a GLP-1RA to reduce the risk of hypoglycemia, potentially reduce body weight, and simplify their treatment regimen.

8 Limitations and future scope

The main limitation of this article is that it is a narrative review and not systematic. In addition, the scope was limited to the CV benefits of injectable GLP-1RAs; further exploration of intensive multifactorial treatment for the prevention of CV events in high-risk patient populations would be useful for primary care providers when managing these patients.

9 Conclusion

Patients with T2D are at high risk of CVD morbidity and mortality, starting well before the diagnosis of diabetes is made. There is a need for early treatment with GLP-1RAs, in line with guidelines and regardless of HbA_1c targets, to help improve long-term outcomes in patients with T2D with, or at risk of, CVD. Even if GLP-1RA therapy has not been initiated early, the intensification of adding a GLP-1RA to insulin still offers benefits for reducing CV risk later in the course of diabetes. While the most common AEs with GLP-1RAs are GI disorders, they are generally manageable and reduce over a few weeks following initiation. A shared decision-making approach between healthcare professionals and patients may improve patient engagement with treatment and managing AEs.

List of abbreviations

Table

AACE/ACE	American Association of Clinical Endocrinologists and AmericanCollege of Endocrinology
ACC	American College of Cardiology
ADA	American Diabetes Association
AE	adverse event
ASCVD	atherosclerotic cardiovascular disease
BMI	body mass index
BP	blood pressure
CI	confidence interval
CKD	chronic kidney disease
CV	cardiovascular
CVD	cardiovascular disease
CVOT	cardiovascular outcomes trial
DRC	diabetic retinopathy complications
GI	gastrointestinal
GLP-1RA	glucagon-like peptide-1 receptor agonist
HbA1c	glycated hemoglobin
HDL	high-density lipoprotein
HF	heart failure
HR	hazard ratio
LDL	low-density lipoprotein
MACE	major adverse cardiovascular events
MI	myocardial infarction
MTC	medullary thyroid cancer
NR	not reported
NS	non-significant
OAD	oral antihyperglycemic drug
s.c.	subcutaneous
SGLT2i	sodium-glucose cotransporter-2 inhibitor
SoC	standard of care
TNF	tumor necrosis factor
T2D	type 2 diabetes

Authorship

The authors were involved with conceptualization, drafting, and/or critically reviewing all drafts during the development of the review article, and all authors provided their final approval for submission.

Disclosure of any financial/other conflicts of interest

Debbie Hinnen has attended advisory boards and served on speakers’ bureaus for Boehringer Ingelheim, Eli Lilly and Co., Janssen Pharmaceuticals, Novo Nordisk, and Sanofi. Davida Kruger has attended advisory boards for Eli Lilly and Co., Novo Nordisk, and Sanofi, and served on speakers’ bureaus for AstraZeneca, Eli Lilly and Co., Janssen Pharmaceuticals, and Novo Nordisk. Melissa Magwire has attended advisory boards and acted as a consultant for Boehringer Ingelheim, Eli Lilly and Co., and Novo Nordisk.

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Declaration of funding

This article was supported by Novo Nordisk Inc.; the company was provided with the opportunity to perform a medical accuracy review.

Acknowledgments

Under the direction of the authors, medical writing and editorial support were provided by Debbie Day of Axis, a division of Spirit Medical Communications Group Ltd. (funded by Novo Nordisk Inc.).

Data availability statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00325481.2022.2126235