Glucagon-Like Peptide-1 Receptor Agonists Across the Spectrum of Heart Failure

Abstract

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have been used to reduce body weight in overweight or people with obesity and to improve glycemic control and cardiovascular outcomes among people with type 2 diabetes (T2D) and a high cardiovascular risk. However, the effects of GLP-1 RAs may be modified by the presence of heart failure (HF). In this review, we summarize the evidence for the use of GLP-1 RA across a patient's risk with a particular focus on HF. After a careful review of the literature, we challenge the current views about the use of GLP-1 RAs and suggest performing active HF screening (with directed clinical history, physical examination, an echocardiogram, and natriuretic peptides) before initiating a GLP-1 RA. After HF screening, we suggest GLP-1 RA treatment decisions as follows: (1) in people with T2D without HF, GLP-1 RAs should be used for reducing the risk of myocardial infarction and stroke, with a possible effect to reduce the risk of HF hospitalizations; (2) in patients with HF and preserved ejection fraction, GLP-1 RAs do not reduce HF hospitalizations but may reduce atherosclerotic events, and their use may be considered in an individualized manner; and (3) in patients with HF and reduced ejection fraction, the use of GLP-1 RAs warrants caution due to potential risk of worsening HF events and arrhythmias, pending risk–benefit data from further studies.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have been used to reduce body weight in overweight or people with obesity and to improve glycemic control and cardiovascular outcomes among people with type 2 diabetes (T2D) and a high cardiovascular risk (1-3).

Despite the well-documented benefits of GLP-1 RA, these may not be generalized to all patient populations. For instance, in patients with heart failure (HF), particularly HF with reduced ejection fraction (HFrEF), some studies suggested that GLP-1 RA may produce harmful effects, increasing the risk of HF hospitalizations and ventricular arrhythmias (4-6). Despite these data, current recommendations no not provide guidance on GLP-1 RA prescription to patients with HF nor do they differentiate between the prevention of HF in those at risk vs the treatment of patients with overt HF. Furthermore, these recommendations do not differentiate among those with different HF phenotypes (7).

Based on the available data, we postulate that the decision to use a GLP-1 RA should be weighted vs the risk of each individual patient, particularly according to HF status, symptoms, and EF. In this review, we provide a practical guidance for the use of GLP-1 RA according to a patient's risk and HF status, backed by data from randomized controlled trials.

Overweight or Obesity Without Diabetes or Heart Failure

Obesity is a major public health issue affecting approximately 40% of the adult population worldwide, and is associated with an increased risk of cardiovascular events (1). GLP-1 RAs reduce body weight as adjuncts to lifestyle intervention.

The potential cardiovascular benefits of GLP-1 RA in overweight or people with obesity without T2D are not well-established, mainly because GLP-1 RA trials in overweight people or with obesity were relatively small or with a short follow-up time. Moreover, overweight or people with obesity without T2D or additional cardiovascular risk factors have a low rate of cardiovascular events.

A meta-analysis of randomized controlled trials including 11 430 overweight people or people with obesity without T2D randomized to either a GLP-1 RA or placebo, suggested that GLP-1 RA might reduce the risk of any adverse cardiovascular event (3). The occurrence of HF events or cardiac arrythmias was rare, in concordance with the low cardiovascular risk of this population.

The ongoing Semaglutide Effects on Heart Disease and Stroke in Patients With Overweight or Obesity trial (SELECT; NCT03574597) and Study of Tirzepatide on the Reduction on Morbidity and Mortality in Adults With Obesity (SURMOUNT-MMO; NCT05556512) will provide more robust evidence on the effect of the GLP-1 RA on cardiovascular outcomes in overweight people or people with obesity without T2D. However, a recent meta-analysis suggested that GLP-1 RAs potentially reduce cardiovascular events in overweight people or people with obesity without HF (3).

Type 2 Diabetes Without Heart Failure

Patients with T2D with high cardiovascular risk but without HF, benefit from GLP-1 RAs for the reduction of atherosclerotic cardiovascular events (ie, myocardial infarction, ischemic stroke, or cardiovascular death: 3-point major adverse cardiac event [3P-MACE]).

Several trials comparing GLP-1 RA vs placebo performed analysis stratified on HF status. However, it is important to highlight that in these trials 80% or more of the enrolled patients did not have HF diagnosis, although an echocardiogram or natriuretic peptides were rarely available in these trials (8-15).

Overall, the findings suggest that GLP-1 RAs may be useful to prevent HF hospitalizations among patients without HF (9-14, 16-21). Specifically, the effect of GLP-1 RA vs placebo on the composite of HF hospitalizations or cardiovascular death in patients without HF history, gave a meta-analyzed pooled hazard ratio (HR) of 0.85, with a 95% CI ranging from 0.76 to 0.94. More specifically, the effect of GLP-1 RAs vs placebo on HF hospitalizations alone yielded a borderline result (HR of 0.81, 95% CI 0.63-1.03). The effect of GLP-1 RAs vs placebo on cardiovascular mortality gave a meta-analyzed HR of 0.85 (95% CI 0.76-0.94). GLP-1 RAs were also superior to placebo in reducing 3P-MACE, with a meta-analyzed HR of 0.88 (95% CI 0.83-0.93) (22).

The GRADE (A Comparative Effectiveness Study of Major Glycemia-lowering Medications for Treatment of Type 2 Diabetes; NCT01794143) trial compared the effectiveness of 4 commonly used glucose-lowering medications (insulin glargine, glimepiride, liraglutide, and sitagliptin) on top of metformin (23). Compared with all other treatments, assignment to liraglutide resulted in lower rate of HF hospitalizations (HR 0.49, 95% CI 0.28-0.86) but the number of HF events was small and there was a lower rate of death from cardiovascular causes (HR 0.47, 95% CI 0.23-0.93). Liraglutide also tended to reduce 3P-MACE (HR 0.75, 95% CI 0.54-1.03). Results according to HF subgroups were not presented in this analysis; however, few patients were expected to have symptomatic HF. The GRADE population had a low cardiovascular risk, with overall HF hospitalization rates inferior to 0.4 events per 100 patient-years and cardiovascular death rates inferior to 0.3 events per 100 patient-years due to the exclusion of patients with a history of a major cardiovascular event in the year before randomization, HF with a New York Heart Association (NYHA) functional classification of III or higher, and an estimated glomerular filtration rate of less than 30 mL/min/1.73 m².

Together, these results support the use of GLP-1 RA for reducing atherosclerotic events in patients with T2D without HF, with a possible effect to reduce HF hospitalizations.

Type 2 Diabetes With Heart Failure and Preserved Ejection Fraction

In contrast to patients with T2D without HF, those with HF (representing 15-20% of patients in trials) did not seem to benefit from GLP-1 RA for reducing HF hospitalizations or cardiovascular death. GLP-1 RAs may reduce atherosclerotic events in patients with T2D and HF (8-15).

The effect of GLP-1 RAs vs placebo on the composite of HF hospitalizations or cardiovascular death in patients with HF gave a meta-analyzed HR of 0.96 (95% CI 0.84-1.08). The meta-analyzed effect on HF hospitalizations alone gave an HR of 0.95 (95% CI 0.81-1.11). The meta-analyzed effect on cardiovascular death gave an HR of 0.96 (95% CI 0.82-1.12) (22).

Despite the neutral effect on HF hospitalizations and cardiovascular death, GLP-1 RAs may be superior to placebo for reducing major atherosclerotic events in patients with T2D and HF, with a pooled 3P-MACE HR of 0.85 (95% CI 0.75-0.97) (22).

The characterization of HF in T2D trials with GLP-1 RA was poor, mostly lacking detailed description on signs and symptoms, as well as natriuretic peptides and echocardiogram. Still, is likely that most HF patients enrolled in T2D trials were stable outpatients with HF with preserved ejection fraction (HFpEF). This statement is supported with data from the EXSCEL (Exenatide Study of Cardiovascular Event Lowering Trial; NCT01144338) trial where EF data were available for 33.2% (n = 4892) of the patients with 9.6% (n = 469) having an EF lower than 40% (ie, >90% of the patients with echocardiographic data available had an EF of 40% or greater) (17).

Dedicated HFpEF trials with GLP-1 RAs are ongoing (eg, NCT04847557 and NCT04788511) and will provide more evidence about the effect of these agents in patients with HFpEF.

Based on available data, GLP-1 RAs did not reduce HF-related events in patients with T2D and HF, most of whom presumably had HFpEF. However, GLP-1 RAs may be considered for reducing ischemic events in patients with HFpEF with a high atherosclerotic risk. However, adequately powered trials should be conducted to test the effect of GLP-1 RAs on cardiovascular (including HF) outcomes.

Heart Failure With Reduced Ejection Fraction

The effect of the GLP-1 RA liraglutide was studied in 3 small trials enrolling patients with HFrEF.

Patients with severely symptomatic HFrEF who had been recently hospitalized for worsening HF were included in the FIGHT trial (the Heart Failure Network Functional Impact of GLP-1 for Heart Failure Treatment; NCT01800968) and randomly assigned to either liraglutide (n = 154) or placebo (n = 146) and followed for 180 days (5). Despite no differences between liraglutide and placebo on the global rank score including time to death, time to HF rehospitalization, and time-averaged change in N-terminal probrain natriuretic peptide (NT-pro-BNP) (primary outcome), and no differences in echocardiographic parameters or health status, liraglutide treatment showed a tendency for increasing the risk of rehospitalization for cardiovascular reasons (HR 1.33, 95% CI 0.95-1.85), HF rehospitalizations (HR 1.33, 95% CI 0.83-2.12), and the risk of an emergency department visit, HF hospitalization, or all-cause death (HR 1.36, 95% CI 0.99-1.85) (5). In a post hoc analysis of the FIGHT trial, reassessing the trial outcomes using total events (first and recurrent), a tendency toward increased risk of total HF hospitalizations or all-cause deaths (incidence rate ratio 1.41, 95% CI 0.98-2.04) and total arrhythmias (incidence rate ratio 1.76, 95% CI 0.92-3.37) was found with liraglutide. Of note, the risk of HF hospitalizations or all-cause deaths was higher among patients in NYHA class III-IV (interaction P = .008), and the risk of arrhythmic events was higher among those without an implanted cardiac device (interaction P = .047) (4).

Another relatively small (n = 241) trial with a short follow-up (180 days) studying the effect of liraglutide vs placebo in stable HFrEF was the LIVE trial (Effect of Liraglutide on Left Ventricular Function in Stable Chronic Heart Failure Patients with and without Diabetes). Although patients in LIVE were more stable than in FIGHT, the results from LIVE also suggested an increased risk of adverse cardiac events with liraglutide, particularly ventricular tachycardias and atrial fibrillation (6).

Another small trial (A Multi-center, Placebo-controlled Study to Evaluate the Safety of GSK716155 and Its Effects on Myocardial Metabolism, Myocardial Function, and Exercise Capacity in Patients With NYHA Class II/III Congestive Heart Failure; NCT01357850) compared weekly placebo (n = 30) with albiglutide 30 mg (n = 27) during 12 weeks. No detectable effect of albiglutide on cardiac function or myocardial glucose use was found, but a modest increase in peak oxygen consumption was observed with albiglutide, which the authors attributed to noncardiac effects of albiglutide (24).

A post hoc analysis of published EXSCEL data (17) suggested that patients with an EF <40% had an increased risk of HF hospitalizations with exenatide vs placebo. In EXSCEL, there were 249 first HF hospitalizations during a median follow-up of 3.2 years among the 4892 participants with baseline left ventricular EF (LVEF) available (33.2% of EXSCEL population). A significant interaction of LVEF with the effect of exenatide on HF hospitalizations was found with an increased risk of HF hospitalizations observed in patients with a LVEF <40%: LVEF >55% (OR 0.73, 95% CI 0.48-1.13); LVEF 40-55% (OR 0.77, 95% CI 0.50-1.20); LVEF <40% (OR 1.70, 95% CI 1.02-2.83; interaction P = .027). A meta-analysis of EXSCEL and FIGHT trials suggested an increased risk of HF hospitalizations with GLP-1 RAs in patients with LVEF <40% with little heterogeneity across studies: meta-analyzed OR 1.49 (95% CI 1.05-2.10) (overall treatment effect P = 0.02 and heterogeneity I² = 0%) (25).

Together, these results suggest that GLP-1 RAs may be associated with an increased risk of HF hospitalizations in patients with HFrEF and, until further randomized studies are available, GLP-1 RAs should be avoided in patients with HFrEF.

Integrated View of the Role of GLP-1 RA Across the Spectrum of Heart Function

The evidence summarized above supports that the effects of GLP-1 RAs are modified by HF status (Fig. 1). Among people with T2D or obesity without HF, treatment with GLP-1 RAs may be used to improve metabolic status and reduce atherosclerotic events, with a possible effect to reduce HF hospitalizations. In patients with stable HFpEF (15-20% of people with T2D), GLP-1 RAs have not reduced HF hospitalizations or cardiovascular mortality but have reduced atherosclerotic events and may be used in selected high-risk patients. In patients with HFrEF, GLP-1 RAs may increase the risk of adverse outcomes, particularly worsening HF and ventricular arrhythmias, and the use of these agents should not be recommended (Fig. 1).

Current guidelines support the use of GLP-1 RAs or SGLT2 inhibitors (with or without metformin) for patients with T2D and cardiovascular disease, including HF (1). The evidence here summarized challenges this “conventional” view by providing an “updated” view whereby ruling out the presence of HF, particularly HFrEF, is relevant for treatment decisions.

Practical Guidance for the Use of GLP-1 RA

The decision to use GLP-1 RAs must be individualized. Before starting GLP-1 RAs, all patients should be evaluated for symptoms and signs of HF and have natriuretic peptides (either BNP or NT-pro-BNP) measured. If natriuretic peptides are elevated (>35 pg/mL for BNP and >125 pg/mL for NT-pro BNP) an echocardiogram should be performed, as recommended by international guidelines (26, 27). This active HF screening allows the categorization of the patients into 3 categories:

1. No HF: GLP-1 RAs are recommended for reducing atherosclerotic cardiovascular events in high-risk patients, with a possible effect to reduce HF hospitalizations.

2. HFpEF: After an SGLT2 inhibitor (first-line agent), GLP-1 RAs may be used in selected patients to reduce atherosclerotic cardiovascular events if the atherosclerotic risk is high.

3. HFrEF: GLP-1 RAs should be avoided in patients with HFrEF, until further evidence is produced. SGLT2 inhibitor (first line) and metformin (second line) should be preferred. Given that other antidiabetic drugs may also not be appropriate in this setting (eg, DDP4 inhibitors and thiazolidinediones may increase the risk of HF hospitalization, and sulfonylureas and insulin increase the risk of hypoglycemic events) (28), a higher HbA1c level may be acceptable in patients with HFrEF already on an SGLT2 inhibitor and metformin. The risk of adverse events with GLP-1 RAs in HFrEF may be particularly increased in those in NYHA class III-IV and in patients without an implanted cardiac device (4).

These bullet points are illustrated in Fig. 2, which may serve as a tool to support routine clinical decisions.

Mechanisms Supporting the Use of GLP-1 RAs According to HF Status

The mechanisms underlying the beneficial effects of GLP-1 RA on HF prevention (among patients without HF), the neutral (HFpEF) or potentially harmful (HFrEF) effects in patients with HF are probably multifactorial. GLP-1 RAs can reduce epicardial fat (28, 29), an effect (together with the reduction of atherosclerotic cardiovascular disease) that may explain why GLP-1 RAs may possibly reduce HF hospitalizations (Fig. 1). On the other hand, GLP-1 RAs are known to increase heart rate, which may be deleterious to patients with HF (30). Furthermore, the GLP-1 receptor, expressed in cardiomyocytes and sinoatrial node cells, signals through a cyclic adenosine monophosphate–dependent pathway, which may induce intracellular calcium overload and increase the risk of ventricular ectopy in high-risk patients, such as those with severely depressed LVEF (7, 17, 31-33).

Future Perspectives

Notwithstanding the current evidence about the relevance of HF status for GLP-1 RA cardiovascular effects, further research in this field is required. More research is needed on the combined use of GLP-1 RAs and SGLT2 inhibitors and on the efficacy and safety of GLP-1 RAs (and dual glucose-dependent insulinotropic polypeptide [GIP] and GLP-1 RA agonists) among patients with HFpEF. In this regard, the effect of semaglutide in HFpEF will be studied in the STEP-HFpEF (Research Study to Investigate How Well Semaglutide Works in People Living With Heart Failure and Obesity; NCT04788511) trial, enrolling 516 patients with HFpEF and obesity, and in the STEP-HFpEF-DM (Research Study to Look at How Well Semaglutide Works in People Living With Heart Failure, Obesity and Type 2 Diabetes) trial, enrolling 610 patients with HFpEF and T2D. The ongoing SUMMIT (A Study of Tirzepatide in Participants With Heart Failure With Preserved Ejection Fraction and Obesity; NCT04847557) trial will assess the effect of tirzepatide in 700 patients with HFpEF and obesity. Despite the clinical importance of these trials, they will likely be underpowered to assess the impact of GLP-1 RAs (or dual GIP/GLP-1 RA) on HF hospitalizations or mortality.

Ideally, further research on the use of GLP-1 RAs in patients with HF (including HFrEF) should be incentivized by regulatory agencies, with further and larger outcome randomized controlled trials testing the efficacy and safety of GLP-1 RAs in HF patients across a wide range of EFs. Nonetheless, based on the available data, enrolling patients with EF <40% possibly raises ethical issues. Still, based on the available evidence, these agents should be avoided in patients with HFrEF, until further evidence is produced.

Conclusion

This article provides evidence for an “updated” view of the use of GLP-1 RAs in clinical practice based on HF status. After active HF screening, if the patient has T2D and atherosclerotic risk without HF, then GLP-1 RAs should be considered for improving cardiovascular outcomes. If the patient has T2D and HFpEF, a GLP-1 RA may be considered after an SGLT2 inhibitor (first line agent) if the atherosclerotic risk is high. However, if the patient has HFrEF, GLP-1 RAs should be avoided until further evidence is produced.

Funding

This work was financed by national funds through FCT (Fundação para a Ciência e Tecnologia), I.P., within the scope of the Cardiovascular R&D Center (UIDB/00051/2020 and UIDP/00051/2020) and RISE (LA/P/0053/2020).

Disclosures

J.P.F. is a consultant for Boehringer Ingelheim and AstraZeneca. He has received research support from Novartis, Boehringer Ingelheim, AstraZeneca, and Bayer through his institution. J.S.N. has received consulting or speaker fees from AstraZeneca, BIAL, Boehringer Ingelheim, Lilly, Merck, and Novo Nordisk. All other authors report not having conflicts of interest regarding the content of this work.

Data Availability

All data used in this manuscript is fully available in the respective publications cited throughout the manuscript.

References

Abbreviations

 
• 3P-MACE

  3-point major adverse cardiac event

 
• EF

  ejection fraction

 
• GIP

  glucose-dependent insulinotropic polypeptide

 
• GLP-1 RA

  glucagon-like peptide-1 receptor agonist

 
• HF

  heart failure

 
• HFpEF

  heart failure with preserved ejection fraction

 
• HFrEF

  heart failure with reduced ejection fraction

 
• LV

  left ventricular

 
• NT-pro-BNP

  N-terminal probrain natriuretic peptide

 
• NYHA

  New York Heart Association

 
• T2D

  type 2 diabetes