Combination of GLP-1 receptor agonists and SGLT2 inhibitors in cardiometabolic disease: synergistic, additive, or non-additive effects? A narrative review

Abstract

Background

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and sodium-glucose cotransporter-2 inhibitors (SGLT2 inhibitors) are two novel classes of glucose-lowering agents with proven cardiometabolic benefits beyond glycaemic control. Both are now recommended in international guidelines for patients with type 2 diabetes mellitus (T2DM) and high cardiovascular risk. As they operate through complementary mechanisms, their combined use may confer additive or synergistic benefits on cardiometabolic outcomes.

Aim

This narrative review examines and summarizes the available evidence on the synergistic effects of GLP-1 RAs and SGLT2 inhibitors in cardiometabolic disease, encompassing their effects on glycaemic control, body weight, blood pressure, cardiovascular events, renal outcomes, and mortality.

Methods

A narrative review of randomised clinical trials, observational studies, post-hoc analyses, and meta-analyses evaluating the combination of GLP-1 RAs and SGLT2 inhibitors in patients with T2DM was conducted. Individual drug class evidence from landmark cardiovascular and renal outcome trials is also reviewed to contextualise the potential for synergy. Evidence from trials conducted in non-diabetic populations with obesity or heart failure with preserved ejection fraction (HFpEF) is also examined where relevant to the combination therapy rationale.

Results

The combination of GLP-1 RAs and SGLT2 inhibitors consistently produced greater reductions in HbA1c (− 1.0 to − 1.5%), body weight (− 3 to − 9 kg), and systolic blood pressure (− 4 to − 10 mmHg) compared with either agent alone across multiple randomised trials, including DURATION-8, AWARD-10, SUSTAIN-9, and LIRA-ADD2SGLT2. Observational studies suggest a significant reduction in major adverse cardiovascular events and heart failure hospitalisations with combination therapy versus individual agents taken separately, while both classes also confer cardiovascular and renal benefits in non-diabetic populations at standard doses. However, randomised trial data do not consistently confirm superiority in cardiovascular event rates. Effects on renal outcomes are inconsistent across studies. Combination therapy appears well tolerated, with no significant increase in serious adverse events beyond those expected from each agent individually.

Conclusions

The combination of GLP-1 RAs and SGLT2 inhibitors exerts additive benefits on glycaemic control, body weight, and blood pressure reduction. Evidence for synergistic cardiovascular and renal protection is promising but remains limited, largely derived from observational data and secondary analyses. Larger, dedicated randomised trials are needed to determine whether combination therapy confers superior cardiometabolic outcomes beyond either agent alone.

Introduction

Cardiovascular disease remains the leading cause of death worldwide. According to the World Health Organization, approximately 19.8 million persons died from cardiovascular causes in 2022, accounting for nearly one-third of all global deaths [1]. This burden is driven by a complex interplay of modifiable behavioural risk factors, including tobacco use, physical inactivity, unhealthy diet, and excessive alcohol consumption, as well as metabolic risk factors such as type 2 diabetes mellitus (T2DM), arterial hypertension, and dyslipidaemia. Among these, T2DM occupies a central role, conferring a two- to threefold increase in cardiovascular risk and representing one of the most prevalent and rapidly growing non-communicable diseases globally.

For decades, the management of T2DM focused primarily on achieving glycaemic targets, with the assumption that lowering blood glucose would translate into improved cardiovascular outcomes. However, large clinical trials such as ACCORD [2] and ADVANCE [3] demonstrated that intensive glucose lowering did not consistently reduce cardiovascular events and, in some cases, increased mortality. This challenged the glucocentric paradigm and highlighted the need for antidiabetic agents with proven cardiovascular safety and benefit. Nevertheless, glycaemic control itself remains an important determinant of cardiometabolic outcomes: meta-analyses of glucose-lowering trials have shown that improvements in HbA1c are associated with reductions in major cardiovascular events in T2DM [4], and glycaemic control is a major determinant of progression of diabetic kidney disease [5]. Modern therapeutic decisions therefore reflect both the glycaemic and the extra-glycaemic (cardiovascular and renal) effects of antidiabetic drugs.

Over the past decade, two novel classes of glucose-lowering agents have fundamentally transformed the cardiometabolic landscape: sodium-glucose cotransporter-2 inhibitors (SGLT2 inhibitors) and glucagon-like peptide-1 receptor agonists (GLP-1 RAs). Despite sharing the goal of glycaemic control, these agents operate through distinct mechanisms. SGLT2 inhibitors promote urinary glucose excretion through inhibition of proximal tubular glucose reabsorption. At the same time, GLP-1 RAs enhance glucose-dependent insulin secretion, suppress glucagon secretion, slow gastric emptying, and reduce appetite via activation of central and peripheral GLP-1 receptors. Both classes have demonstrated cardiometabolic benefits well beyond glycaemic control in landmark cardiovascular outcome trials. SGLT2 inhibitors have shown particular strength in reducing heart failure hospitalisations and slowing the progression of chronic kidney disease (CKD), as demonstrated in the EMPA-REG OUTCOME [6], DAPA-HF [7], EMPEROR-Reduced [8], CREDENCE [9], and DAPA-CKD [10] trials. GLP-1 RAs, in turn, have demonstrated significant reductions in major adverse cardiovascular events (MACE), particularly atherosclerotic endpoints, as shown in the LEADER [11], SUSTAIN-6 [12], REWIND [13], and FLOW [14] trials.

Given that these two drug classes target different pathophysiological pathways, it is biologically plausible that their combination may exert additive or synergistic effects on cardiometabolic outcomes. The American Heart Association formally recognised this integrated therapeutic framework in 2023, introducing the concept of cardiovascular–kidney–metabolic (CKM) syndrome to describe the pathophysiological and clinical interrelationships among cardiovascular disease, chronic kidney disease, obesity, and metabolic disorders [15]. Both SGLT2 inhibitors and GLP-1 RAs are central to the pharmacological management of CKM syndrome, and their combination represents a rational multi-pathway approach to addressing its full disease spectrum. Current guidelines from the European Society of Cardiology (ESC), the American Diabetes Association (ADA), and the European Association for the Study of Diabetes (EASD) recommend both drug classes in patients with T2DM and established cardiovascular disease or high cardiovascular risk. Their combined use is increasingly encountered in clinical practice. Nevertheless, the evidence base for their synergistic effects remains limited, with most data derived from secondary analyses of cardiovascular outcome trials, short-term randomised controlled trials, and observational studies.

The present narrative review aims to critically appraise and synthesise the available evidence on whether the combination of GLP-1 receptor agonists and SGLT2 inhibitors confers synergistic, additive, or non-additive effects across key cardiometabolic outcomes—including glycaemic control, body weight, blood pressure, major adverse cardiovascular events, heart failure, renal disease, and mortality. To this end, we draw on randomised clinical trials, prespecified subgroup analyses, observational cohort studies, and recent meta-analyses, and we discuss the mechanistic rationale, the strengths and limitations of the existing evidence, and the practical clinical implications for the management of patients with type 2 diabetes mellitus and complex cardiometabolic risk.

SGLT2 inhibitors

Effects on blood glucose and mechanisms of action

Sodium–glucose cotransporter-2 (SGLT2) inhibitors represent one of the most significant pharmacological advances in the management of cardiometabolic disease over the past decade. Initially developed as glucose-lowering agents for the treatment of T2DM, these medications have demonstrated substantial cardiovascular and renal protective effects that extend well beyond glycaemic control. Evidence from large randomised controlled trials has shown that SGLT2 inhibitors reduce the risk of heart failure hospitalisation, slow the progression of chronic kidney disease (CKD), and lower cardiovascular mortality in both diabetic and non-diabetic populations [16]. Consequently, these drugs have become a central element of modern cardiorenal-metabolic therapy and are now recommended in international clinical guidelines for patients with T2DM, heart failure, and CKD.

Although these benefits are now well established, the mechanisms responsible for them are complex and not yet fully understood. Beyond promoting glucosuria and improving glycaemic control, SGLT2 inhibitors modulate renal haemodynamics, metabolic substrate utilisation, inflammatory signalling pathways, and neurohormonal systems.

Primary mechanism of glucose lowering

Glucose homeostasis in the kidney is regulated by sodium-glucose cotransporters located in the proximal tubule. Under physiological conditions, approximately 180 g of glucose is filtered daily through the glomerulus, nearly all of which is reabsorbed in the proximal tubule. Approximately 90% of filtered glucose is reabsorbed by SGLT2 transporters located in the S1 and S2 segments, while SGLT1 transporters reabsorb the remaining 10% in the S3 segment [17, 18]. SGLT2 inhibitors selectively block this transporter, reducing glucose and sodium reabsorption and promoting urinary glucose excretion via an insulin-independent mechanism, making these agents particularly useful in patients with insulin resistance or impaired β-cell function [19]. Pharmacological inhibition of SGLT2 leads to sustained glucosuria, with urinary glucose excretion of approximately 50–100 g per day, corresponding to a daily caloric deficit of approximately 200–300 kcal [19].

Metabolic effects

Beyond glucose-lowering, SGLT2 inhibitors produce clinically meaningful reductions in glycated haemoglobin (HbA1c) of approximately 0.6–1.0%, systolic blood pressure of approximately 3–5 mmHg, and body weight of approximately 2–3 kg [20]. The magnitude of HbA1c reduction, however, is strongly dependent on baseline glycaemic control: mean reductions reported in clinical trials should be interpreted in the context of baseline HbA1c values, with substantially larger reductions observed in patients with poorly controlled diabetes. A key metabolic effect is a shift towards ketone-body utilisation: reduced circulating insulin and a modest rise in glucagon stimulate hepatic ketogenesis, increasing β-hydroxybutyrate levels. These provide a more energy-efficient myocardial fuel source compared with glucose or fatty acids [21]. The glucose-lowering efficacy of SGLT2 inhibitors decreases as renal function declines; however, cardiovascular and renal protective effects persist even in individuals with markedly reduced kidney function [10, 22].

Effects on major cardiovascular events in diabetes

Cardiovascular outcome trials

The cardiovascular benefits of SGLT2 inhibitors were first established in large randomised cardiovascular outcome trials. The EMPA-REG OUTCOME trial demonstrated that empagliflozin significantly reduced the risk of MACE and lowered cardiovascular mortality by 38% in patients with T2DM and established cardiovascular disease [6]. The CANVAS Program verified the MACE benefit with canagliflozin, though it raised concerns about an increased risk of lower-limb amputations [23]. The DECLARE-TIMI 58 trial with dapagliflozin enrolled a broader population, including many without established cardiovascular disease; although MACE was not significantly reduced, heart failure hospitalisations and renal outcomes were significantly improved [24]. The VERTIS CV trial of ertugliflozin demonstrated cardiovascular safety, with a neutral MACE result, but a significant reduction in heart failure hospitalisation [25]. Importantly, although SGLT2 inhibitors appear to share broadly similar effects on HbA1c, body weight, heart failure, and renal outcomes, the results of cardiovascular outcome trials in T2DM are not entirely homogeneous with respect to MACE: a clear MACE benefit was demonstrated for empagliflozin and canagliflozin, whereas DECLARE-TIMI 58 and VERTIS CV showed neutral MACE results despite consistent benefits on heart failure hospitalisation. This suggests that the MACE benefit of SGLT2 inhibitors should not be extrapolated uncritically across all agents in the class, even if the heart failure and renal class effects appear robust.

Possible mechanisms

The mechanisms by which SGLT2 inhibitors reduce cardiovascular events and heart failure are multiple and distinct from their glucose-lowering effects. Haemodynamic effects are central to the HF benefit: natriuresis and osmotic diuresis reduce preload, while blood pressure reduction lowers afterload, collectively reducing myocardial wall stress [21]. SGLT2 inhibitors promote a shift in myocardial fuel utilisation towards ketone bodies (β-hydroxybutyrate), which generate more ATP per unit of oxygen than glucose or fatty acids, improving energetics in the metabolically stressed failing heart [21]. Inhibition of the myocardial sodium–hydrogen exchanger (NHE1) reduces intracellular sodium and calcium overload in cardiomyocytes, improving contractility and reducing arrhythmia susceptibility [26]. SGLT2 inhibitors stimulate erythropoiesis, raising haematocrit and improving oxygen delivery to ischaemic myocardium. They reduce epicardial and perivascular adipose tissue, attenuating pro-inflammatory adipokine secretion. Sympathetic nervous system modulation contributes to reductions in blood pressure and heart rate. RAAS attenuation provides additional cardiorenal protection. The predominance of haemodynamic and metabolic mechanisms explains why SGLT2 inhibitors benefit heart failure across the full ejection fraction spectrum—a key distinction from GLP-1 RAs, whose HF benefit is restricted to obese HFpEF.

Effects on heart failure

Subsequent trials established robust benefits in heart failure regardless of diabetic status. The DAPA-HF trial demonstrated that dapagliflozin significantly reduced the risk of worsening heart failure or cardiovascular death in patients with heart failure with reduced ejection fraction (HFrEF), including those without diabetes [7]. The EMPEROR-Reduced trial confirmed these findings with empagliflozin [8]. More recently, the EMPEROR-Preserved and DELIVER trials demonstrated benefits in heart failure with preserved ejection fraction (HFpEF), a condition historically lacking effective pharmacological treatments [27, 28]. Meta-analyses have demonstrated consistent reductions of approximately 25–30% in heart failure hospitalisation across diverse patient populations [16]. Tables 1 and 2 summarise the major cardiovascular outcome and heart failure trials, respectively.

Table 1.

Major cardiovascular outcome trials of SGLT2 inhibitors

Trial (Year)	Population	Drug	Primary outcome	Key findings
EMPA-REG OUTCOME (2015)	DM2 + established ASCVD	Empagliflozin	3-point MACE (CV death, MI, stroke)	38% reduction in CV death; significant reduction in HF hospitalisation; renal protection
CANVAS Program (2017)	DM2 with high CV risk	Canagliflozin	3-point MACE	Reduced MACE and HF hospitalisation; increased amputation risk observed
DECLARE-TIMI 58 (2019)	Broad DM2 population (many without ASCVD)	Dapagliflozin	CV death or HF hospitalisation	Significant reduction in HF hospitalisation; neutral MACE result
VERTIS CV (2020)	T2DM + established ASCVD	Ertugliflozin	3-point MACE	Cardiovascular safety established; neutral MACE result; significant reduction in HF hospitalisation (HR 0.70)

Table 2.

Major heart failure trials of SGLT2 inhibitors

Trial (Year)	Population	Drug	Primary outcome	Key findings
DAPA-HF (2019)	HFrEF ± diabetes	Dapagliflozin	CV death or worsening HF	26% reduction in composite outcome; benefit independent of diabetes status
EMPEROR-Reduced (2020)	HFrEF ± diabetes	Empagliflozin	CV death or HF hospitalisation	25% reduction in composite outcome; slower renal function decline
EMPEROR-Preserved (2021)	HFpEF ± diabetes	Empagliflozin	CV death or HF hospitalisation	First therapy to significantly improve HFpEF outcomes
DELIVER (2022)	HFmrEF/HFpEF ± diabetes	Dapagliflozin	CV death or worsening HF	Confirmed benefit across the full HF ejection fraction spectrum

Effects on kidney disease

Renal outcome trials

Dedicated renal outcome trials confirmed the nephroprotective class effect. The CREDENCE trial demonstrated that canagliflozin significantly reduced end-stage kidney disease, doubling of serum creatinine, and renal death in patients with T2DM and CKD, reducing the absolute risk of the renal composite endpoint by 5.0% points over 2.6 years [9]. The DAPA-CKD trial demonstrated that dapagliflozin reduced kidney disease progression and all-cause mortality in patients with CKD including those without diabetes, with an absolute risk reduction of approximately 5.3% [10]. The EMPA-KIDNEY trial confirmed these benefits across a wide range of kidney function levels [22]. These relative risk reductions of 30–40% provide clinicians with meaningful context for prescribing decisions. Table 3 summarises the major renal outcome trials.

Table 3.

Major renal outcome trials of SGLT2 inhibitors

Trial (Year)	Population	Drug	Primary outcome	Key findings
CREDENCE (2019)	DM2 + CKD	Canagliflozin	ESRD, doubling of creatinine, renal death	30% reduction in renal composite outcome; trial stopped early for benefit
DAPA-CKD (2020)	CKD ± diabetes	Dapagliflozin	Sustained eGFR decline, ESRD, renal/CV death	39% reduction in renal composite; benefit demonstrated in non-diabetics
EMPA-KIDNEY (2022)	CKD ± diabetes	Empagliflozin	Kidney disease progression or CV death	28% reduction in renal composite outcomes; broad CKD population included

Mechanisms of renal protection

One of the most important physiological effects of SGLT2 inhibitors is the restoration of tubuloglomerular feedback (TGF), a key mechanism regulating glomerular filtration. By inhibiting sodium reabsorption in the proximal tubule, SGLT2 inhibitors increase sodium delivery to the macula densa, activate the TGF mechanism, and cause afferent arteriolar vasoconstriction, thereby reducing intraglomerular pressure [29]. Clinically, this manifests as a modest initial decline in estimated glomerular filtration rate (eGFR) of approximately 3–5 mL/min/1.73 m², followed by long-term stabilisation of renal function [10]. Additionally, by reducing proximal tubular workload, SGLT2 inhibitors decrease renal oxygen consumption and improve medullary oxygenation, reducing hypoxia-induced renal injury and fibrosis.

Safety profile

SGLT2 inhibitors are generally well tolerated. The most common adverse events are genital mycotic infections due to increased urinary glucose excretion, occurring in approximately 10% of women and 2–3% of men [30]. Rare but clinically significant adverse events include euglycaemic diabetic ketoacidosis, particularly during physiological stress; the risk is substantially higher in patients with type 1 diabetes mellitus or those following ketogenic diets, in whom these agents should be used with particular caution or avoided. Although early concerns arose regarding lower-extremity amputations, particularly with canagliflozin (hazard ratio 1.97 in the CANVAS trial [23], subsequent observational analyses suggested this signal is context-dependent and not a class-wide effect [31]. Rare cases of Fournier’s gangrene prompted an FDA safety warning in 2018. SGLT2 inhibitors promote osmotic diuresis, which may cause volume depletion in susceptible patients, particularly those receiving concomitant diuretics. Monitoring of volume status, blood pressure, and renal function is recommended at the initiation of treatment. The association with urinary tract infections appears less pronounced than initially feared, with large clinical trials generally showing incidence rates comparable to placebo [30].

Guideline recommendations

Cumulative evidence has fundamentally reshaped international treatment guidelines. Organisations such as the ESC, ADA, and Kidney Disease: Improving Global Outcomes (KDIGO) now recommend SGLT2 inhibitors as first-line or foundational therapy for patients with T2DM at high cardiovascular risk, heart failure across the spectrum of ejection fraction, and CKD with albuminuria. Recent guideline updates recommend initiating SGLT2 inhibitors in patients with CKD and an eGFR of 20–45 mL/min/1.73 m², particularly when albuminuria is present (urine albumin-to-creatinine ratio ≥ 30 mg/g). This paradigm shift underscores their role in comprehensive cardiorenal risk reduction and provides a logical transition to the examination of complementary agents, such as GLP-1 receptor agonists.

GLP-1 receptor agonists

Effects on blood glucose

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) represent an important therapeutic class in the management of cardiometabolic disease. Initially developed as glucose-lowering agents for T2DM, these drugs have demonstrated substantial metabolic and cardiovascular benefits that extend beyond glycaemic control. GLP-1 RAs mimic the endogenous incretin hormone glucagon-like peptide-1, which is secreted by intestinal L-cells in response to nutrient ingestion. Several GLP-1 RAs have been approved for T2DM treatment, including exenatide, liraglutide, lixisenatide, dulaglutide, and semaglutide; an oral formulation of semaglutide has also been introduced in recent years.

Pancreatic and incretin mechanisms

GLP-1 enhances glucose-dependent insulin secretion from pancreatic β-cells while simultaneously suppressing glucagon secretion from pancreatic α-cells, improving post-prandial glucose control and reducing hepatic glucose production. GLP-1 RAs have a significantly longer half-life than endogenous GLP-1 due to resistance to degradation by dipeptidyl peptidase-4 (DPP-4), leading to sustained insulin secretion in response to hyperglycaemia while minimising the risk of hypoglycaemia. Experimental studies suggest that GLP-1 RAs may also promote β-cell proliferation and inhibit apoptosis, potentially contributing to long-term preservation of insulin-secreting capacity [32] however, these findings derive primarily from experimental rodent models and have not been reliably confirmed in human studies.

Effects on major cardiovascular events

Cardiovascular outcome trials

The LEADER trial demonstrated that liraglutide significantly reduced the incidence of MACE compared with placebo in patients with T2DM and high cardiovascular risk [11]. The SUSTAIN-6 trial showed that semaglutide significantly lowered the risk of MACE, with a particular reduction in stroke [12]. The REWIND trial evaluated dulaglutide and demonstrated a significant reduction in cardiovascular events in a broad population, including many without established cardiovascular disease [13]. In contrast, the ELIXA trial (lixisenatide) demonstrated cardiovascular safety without superiority over placebo [33], and the EXSCEL trial (once-weekly exenatide) showed cardiovascular safety with a non-significant trend toward MACE reduction [34]. Meta-analyses including more than 60,000 participants have confirmed that GLP-1 RAs reduce MACE by approximately 12%, cardiovascular mortality by 12%, and all-cause mortality by 12% [35].

The magnitude of cardiovascular benefit appears particularly pronounced for atherosclerotic outcomes such as myocardial infarction and stroke, whereas the effects on heart failure outcomes are generally more modest compared with those observed with SGLT2 inhibitors. This suggests that the two drug classes exert complementary cardioprotective mechanisms, providing a strong rationale for combination therapy. It should be emphasised, however, that the GLP-1 RA class is substantially heterogeneous—even more so than the SGLT2 inhibitor class. Heterogeneity has been observed not only in cardiovascular outcomes (with clear MACE benefit demonstrated for liraglutide, semaglutide, dulaglutide, albiglutide, and efpeglenatide, but neutral results for lixisenatide and exenatide) but also in the magnitude of effects on HbA1c and body weight, where short-acting agents (e.g., lixisenatide), albiglutide, and exenatide produce smaller reductions than the long-acting GLP-1 analogues semaglutide and dulaglutide. Differences in molecular structure, pharmacokinetics, half-life, and receptor affinity likely underlie these intra-class differences, and findings from individual trials should not be uncritically extrapolated to the class as a whole.

Cardiovascular outcomes in obesity without diabetes (SELECT)

The SELECT (Semaglutide Effects on Cardiovascular Outcomes in People With Overweight or Obesity) trial extended the cardiovascular benefits of GLP-1 RAs to a non-diabetic population. This multicentre, double-blind, randomised, placebo-controlled, event-driven trial enrolled 17,604 patients aged 45 years or older with pre-existing atherosclerotic cardiovascular disease and a BMI of 27 kg/m² or greater, but without diabetes. Participants were randomised 1:1 to subcutaneous semaglutide 2.4 mg once weekly or placebo, with escalation over 16 weeks to the target maintenance dose, with and a mean follow-up of 39.8 months [36]. The primary composite outcome of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke occurred in 6.5% of the semaglutide group versus 8.0% of the placebo group (hazard ratio 0.80; 95% CI 0.72–0.90; p < 0.001), representing a 20% relative risk reduction. Reductions in all-cause mortality (HR 0.81; 95% CI 0.71–0.93) and heart failure hospitalisation (HR 0.82; 95% CI 0.71–0.96) were also observed. Cardiovascular benefits were independent of baseline HbA1c, degree of weight loss, or baseline adiposity, suggesting pleiotropic mechanisms beyond glycaemic and weight effects. Notably, a prespecified subgroup analysis demonstrated that semaglutide reduced MACE in patients with both HFrEF (HR 0.65; 95% CI 0.49–0.87) and HFpEF (HR 0.69; 95% CI 0.51–0.91) at baseline [37]. SELECT is the first large randomised trial to demonstrate cardiovascular event reduction with a GLP-1 RA in patients without diabetes, establishing a new indication for this drug class in secondary cardiovascular prevention.

Possible cardiovascular mechanisms

The cardiovascular benefits of GLP-1 RAs are predominantly mediated through anti-atherosclerotic rather than haemodynamic mechanisms, explaining why their benefit is most consistent for MACE (particularly stroke and MI) and less pronounced for heart failure, where SGLT2 inhibitors have a clear advantage. GLP-1 receptors are expressed in endothelial cells of the aorta and coronary arteries; activation improves endothelial function through increased nitric oxide bioavailability, reduces oxidative stress, attenuates vascular inflammation, and exerts anti-thrombotic effects by reducing platelet aggregation and enhancing fibrinolysis [38, 39]. These combined effects stabilise atherosclerotic plaques and reduce the risk of acute coronary events. Crucially, albiglutide—a GLP-1 RA with negligible effect on body weight—significantly reduced MACE by 22% in HARMONY-OUTCOMES, providing strong evidence that cardiovascular protection is not simply a consequence of weight loss [40]. GLP-1 receptors are also expressed in cardiac tissue, particularly in sinoatrial node cells and atrial cardiomyocytes; activation modestly increases heart rate and may improve myocardial glucose utilisation during ischaemia-reperfusion. GLP-1 RA therapy is also consistently associated with modest systolic blood pressure reductions of approximately 2–5 mmHg [41], which represent a further potential mechanism of cardiovascular risk reduction. Sympathetic nervous system modulation, reduced CRP levels, improved endothelial progenitor cell function, and reduced hepatic steatosis are additional contributing mechanisms. The complementary, non-overlapping nature of GLP-1 RA and SGLT2 inhibitor cardiovascular mechanisms provides a strong mechanistic rationale for additive benefit when combined.

Several additional cardiovascular outcome trials complement the evidence above. The HARMONY-OUTCOMES trial demonstrated that albiglutide significantly reduced MACE by 22% versus placebo in patients with T2DM and established cardiovascular disease [40]. Notably, albiglutide has a negligible effect on body weight, which supports the notion that GLP-1 RA cardiovascular protection is at least partly independent of weight reduction. The AMPLITUDE-O trial demonstrated that efpeglenatide significantly reduced MACE by 27% and composite renal outcomes versus placebo in patients with T2DM at high cardiovascular risk [42]. The PIONEER-6 trial established cardiovascular safety for oral semaglutide 14 mg, with a non-significant 21% reduction in MACE [43]. Most recently, the SOUL trial randomised 9650 patients with T2DM and atherosclerotic cardiovascular disease and/or CKD to oral semaglutide or placebo, demonstrating a statistically significant 14% reduction in the primary cardiovascular composite outcome [44]. For SGLT2 inhibitors, the VERTIS CV trial of ertugliflozin demonstrated cardiovascular safety, with a neutral MACE result, but with significant reductions in heart failure hospitalisations [25].

Effects on heart failure

The STEP-HFpEF trial (Research Study to Investigate How Well Semaglutide Works in People Living With Heart Failure and Obesity) specifically addressed the unmet therapeutic need in the obesity phenotype of HFpEF. This randomised, double-blind, placebo-controlled trial enrolled 529 patients with symptomatic HFpEF (left ventricular ejection fraction ≥ 45%), a BMI of 30 kg/m² or greater, and without known diabetes or HbA1c ≥ 6.5%. Patients were randomised 1:1 to once-weekly semaglutide 2.4 mg or placebo for 52 weeks [45]. The trial had dual primary endpoints: the change from baseline in the Kansas City Cardiomyopathy Questionnaire clinical summary score (KCCQ-CSS) and body weight. Semaglutide produced a mean improvement in KCCQ-CSS of 16.6 points versus 8.7 points with placebo (estimated treatment difference 7.8 points; 95% CI 4.8–10.9), representing the largest KCCQ benefit recorded with any pharmacological agent in HFpEF. Body weight was reduced by 13.3% with semaglutide versus 2.6% with placebo (estimated treatment difference − 10.7%; 95% CI − 11.9 to − 9.4). Confirmatory secondary endpoints were also met: semaglutide significantly improved six-minute walk distance, the hierarchical composite outcome, and reduced C-reactive protein levels, reflecting meaningful anti-inflammatory effects. A companion trial, STEP-HFpEF DM, confirmed these findings in a similar population with concomitant type 2 diabetes [46]. A prespecified pooled analysis of STEP-HFpEF and STEP-HFpEF DM demonstrated consistent benefits across patient subgroups defined by age, sex, BMI, systolic blood pressure, and baseline CRP, with serious adverse events occurring substantially less frequently with semaglutide than with placebo [46]. Important caveats regarding the BMI dependence of these findings, the restriction to HFpEF, and the obesity context are discussed in detail in Sect. "Possible mechanisms and important caveats". These trials represent important early evidence for pharmacological treatment of obesity-related HFpEF. They are particularly relevant in the context of combination therapy with SGLT2 inhibitors, given the limited baseline use of SGLT2 inhibitors in these trials and the complementary mechanisms of action of the two drug classes.

Possible mechanisms and important caveats

The mechanisms by which GLP-1 RAs improve outcomes in heart failure require careful contextualisation. Three fundamental limitations of the current evidence must be explicitly acknowledged before drawing clinical conclusions.

First, all positive randomised evidence pertains exclusively to HFpEF. No randomised trial has demonstrated benefit in HFrEF. Mechanistic concerns that modestly increased heart rate and sympathetic activation with GLP-1 RAs might be detrimental in HFrEF remain unresolved, and GLP-1 RAs are not recommended in HFrEF guidelines. Any clinical application must be strictly confined to the HFpEF phenotype.

Second, a prespecified subgroup analysis of pooled data from STEP-HFpEF and STEP-HFpEF DM showed that the symptomatic benefit of semaglutide was most pronounced in patients with baseline BMI ≥ 35 kg/m² and effectively disappeared in patients with lower degrees of obesity [46]. This strongly implies that the benefit is primarily mediated through substantial weight reduction rather than direct cardiac GLP-1 receptor effects. Patients with HFpEF and BMI below 35 kg/m² should not be expected to derive the same symptomatic benefit.

Third, all randomised trial data on GLP-1 RAs in HF were obtained from patients who were overweight or obese (minimum BMI of 30 kg/m² in STEP-HFpEF). Results cannot be extrapolated to non-obese HFpEF populations.

Plausible mechanisms in obesity-related HFpEF include: (i) substantial weight reduction (− 13.3% in STEP-HFpEF), decreasing myocardial wall stress and filling pressures; (ii) anti-inflammatory effects, with CRP reductions partly independent of weight loss; (iii) reduction in epicardial adipose tissue with its pro-inflammatory and pro-fibrotic paracrine effects on adjacent myocardium; and (iv) improvements in skeletal muscle function and exercise capacity. These mechanisms are fundamentally distinct from those of SGLT2 inhibitors, which act by unloading volume and pressure across the full ejection fraction spectrum. This distinction defines the scientific basis for combination therapy in obese HFpEF: the two drug classes address different, complementary pathophysiological targets.

Effects on kidney disease

The FLOW trial provided dedicated evidence for the renal effects of GLP-1 RAs. This double-blind, placebo-controlled trial randomised 3533 patients with T2DM and CKD to receive semaglutide 1 mg subcutaneously once weekly or placebo, with a median follow-up of 3.4 years. Semaglutide significantly reduced the risk of the primary composite renal outcome—including sustained eGFR decline of at least 50%, onset of end-stage kidney disease, or death from renal or cardiovascular causes—by 24% compared with placebo [14]. These findings establish semaglutide as the first GLP-1 RA with demonstrated benefit in a dedicated renal outcomes trial, complementing the nephroprotective effects observed with SGLT2 inhibitors. The contribution of glycaemic improvement to this renal benefit deserves emphasis: because glycaemic control is itself a major determinant of progression of diabetic kidney disease [5], the magnitude of HbA1c reduction achieved with GLP-1 RAs likely accounts for a substantial fraction of the renal effect observed in patients with T2DM. Whether GLP-1 RAs exert clinically meaningful renoprotection independent of glycaemic improvement remains less well established. Data from non-diabetic populations are limited but informative. A prespecified analysis of renal endpoints in the SELECT trial (semaglutide 2.4 mg in 17,604 patients with obesity and established CVD without diabetes) demonstrated a significant slowing of eGFR decline and a reduction in the composite renal outcome (eGFR decline ≥ 50%, ESKD, or renal death) compared with placebo over a mean follow-up of 39.8 months [47]. Reductions in albuminuria were also observed in participants with elevated baseline UACR, with the magnitude of UACR reduction broadly consistent with that reported in T2DM trials despite the absence of glycaemic improvement as a contributing mechanism. Post-hoc analyses of the STEP-1 and STEP-HFpEF trials similarly suggest reductions in albuminuria and modest preservation of eGFR in non-diabetic participants with obesity. Although these data derive from secondary or post-hoc analyses rather than dedicated renal outcome trials in non-diabetic CKD, they collectively support the existence of weight-mediated, anti-inflammatory, and direct vascular renoprotective mechanisms of GLP-1 RAs that are partly independent of glycaemic improvement. Dedicated randomised renal outcome trials of GLP-1 RAs in non-diabetic CKD populations are nonetheless lacking and represent an important evidence gap. Table 4 summarises the major cardiovascular and renal outcome trials of GLP-1 RAs.

Table 4.

Major cardiovascular and renal outcome trials of GLP-1 receptor agonists

Trial (Year)	Population	Drug	Primary outcome	Key findings
ELIXA (2015)	DM2 + recent acute coronary syndrome	Lixisenatide	3-point MACE	Demonstrated cardiovascular safety but no significant reduction in MACE
LEADER (2016)	DM2 + high CV risk	Liraglutide	3-point MACE	13% reduction in MACE; 22% reduction in CV death
SUSTAIN-6 (2016)	DM2 + high CV risk	Semaglutide (s.c.)	3-point MACE	26% reduction in MACE; significant reduction in stroke
EXSCEL (2017)	DM2 with or without CVD	Exenatide (weekly)	3-point MACE	Demonstrated CV safety; non-significant trend toward reduced MACE
REWIND (2019)	DM2 with or without CVD	Dulaglutide	3-point MACE	Significant reduction in MACE; benefit extended to primary prevention population
FLOW (2024)	DM2 + CKD	Semaglutide (s.c.) 1 mg	Renal composite (eGFR decline ≥ 50%, ESKD, renal/CV death)	24% reduction in primary renal composite; first dedicated renal outcomes trial for GLP-1 RAs
SELECT (2023)	No diabetes; established CVD; BMI ≥ 27 kg/m2	Semaglutide (s.c.) 2.4 mg	3-point MACE (CV death, MI, stroke)	20% reduction in MACE (HR 0.80); first GLP-1 RA CVOT in non-diabetic patients; benefits independent of weight loss magnitude
STEP-HFpEF (2023)	HFpEF (LVEF ≥ 45%); BMI ≥ 30 kg/m2; no diabetes	Semaglutide (s.c.) 2.4 mg	KCCQ-CSS change and body weight change (dual primary)	KCCQ-CSS improved by 7.8 points vs. placebo; body weight reduced by 10.7%; first pharmacological therapy to significantly improve symptoms in obesity-related HFpEF
HARMONY-OUTCOMES (2018)	T2DM + established CVD	Albiglutide	3-point MACE	22% reduction in MACE (HR 0.78); negligible weight effect supports weight-independent CV protection
AMPLITUDE-O (2021)	T2DM + high CV risk or CKD	Efpeglenatide	3-point MACE + renal composite	27% reduction in MACE (HR 0.73); significant reduction in composite renal outcome
PIONEER-6 (2019)	T2DM + high CV risk	Oral semaglutide 14 mg	3-point MACE	CV safety established; non-significant 21% MACE reduction (HR 0.79)
SOUL (2025)	T2DM + ASCVD and/or CKD (n = 9,650; ~31% on SGLT2i)	Oral semaglutide 14 mg	3-point MACE	Significant 14% reduction in CV composite; benefit consistent regardless of baseline SGLT2i use

Effects on body weight

GLP-1 RAs slow gastric emptying, delaying nutrient absorption and reducing postprandial glucose excursions. They also act on hypothalamic appetite centers, promoting satiety and reducing caloric intake. This central effect contributes to the significant weight reduction observed during treatment, with most trials reporting losses of 3–6 kg, and larger reductions observed with higher doses used for obesity treatment [48].

Effects on metabolic dysfunction-associated steatotic liver disease (MASLD)

Emerging evidence suggests that GLP-1 RAs exert beneficial effects on metabolic dysfunction-associated steatotic liver disease (MASLD, formerly termed NAFLD/NASH). GLP-1 RAs reduce hepatic steatosis and inflammation through multiple mechanisms, including reduction in body weight, decreased hepatic lipogenesis, improved insulin sensitivity, and direct hepatocyte effects via GLP-1 receptor activation. The LEAN trial demonstrated that liraglutide significantly improved liver histology in patients with non-alcoholic steatohepatitis (NASH) compared with placebo [49]. More recently, semaglutide demonstrated significant reductions in liver fat content and hepatic inflammation in phase 2 trials. The phase 3 ESSENCE trial is currently evaluating semaglutide in biopsy-confirmed MASH (metabolic dysfunction-associated steatohepatitis). In combination therapy, SGLT2 inhibitors also reduce hepatic steatosis through ketogenesis and caloric deficit, suggesting complementary hepatoprotective mechanisms when the two drug classes are combined. Whether combination therapy offers an additive benefit in MASLD/MASH remains to be determined in dedicated trials.

Safety profile

GLP-1 receptor agonists are generally well tolerated, although gastrointestinal symptoms represent the most commonly reported adverse effects. Nausea, vomiting, and diarrhoea frequently occur during the initial phase of treatment but are typically mild and transient. Other reported adverse events include injection-site reactions and headaches. Because insulin secretion induced by GLP-1 RAs is glucose-dependent, the risk of hypoglycaemia is low when these agents are used as monotherapy or in combination with non-hypoglycaemic medications.

Several additional safety considerations merit attention for GLP-1 RAs. Although early pharmacoepidemiological data raised concerns about pancreatitis, large randomised cardiovascular outcome trials have not confirmed a clinically meaningful increase in risk. Cholelithiasis (gallstone disease) is a recognised class effect, likely mediated by altered gallbladder motility and rapid weight loss. Pharmacovigilance reports have described cases of non-arteritic anterior ischaemic optic neuropathy (NAION); causality has not been established, but clinicians should be aware. Rodent studies identified C-cell hyperplasia, raising concerns about papillary thyroid carcinoma; long-term human cardiovascular outcome trial data have not confirmed an elevated risk, but GLP-1 RAs remain contraindicated in patients with a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia type 2. With respect to combination therapy, elderly and frail patients warrant particular attention as they are underrepresented in randomised trials. They are at disproportionate risk of sarcopenia from rapid weight loss, and of volume depletion from the combined diuretic effects of both drug classes. Renal function, hydration, and body composition should be monitored carefully in this population.

Guideline recommendations

Current guidelines from the ESC, ADA, and EASD recommend GLP-1 RAs as the preferred option in patients with T2DM who have established atherosclerotic cardiovascular disease or are at high cardiovascular risk, irrespective of baseline HbA1c or current glucose-lowering therapy. The 2023 ESC Guidelines and ADA/EASD consensus reports recommend GLP-1 RAs with proven cardiovascular benefit as first-line add-on therapy regardless of whether glycaemic targets have been achieved on metformin. In contrast to SGLT2 inhibitors, GLP-1 RAs are not currently recommended as primary treatment for heart failure or CKD in the absence of T2DM. However, the FLOW trial results may prompt future revisions to guidelines.

Combination therapy: GLP-1 RAs and SGLT2 inhibitors

The rationale for combining GLP-1 RAs and SGLT2 inhibitors rests on their complementary and largely non-overlapping mechanisms of action. GLP-1 RAs act primarily through anti-atherosclerotic, anti-inflammatory, and weight-reducing pathways. In contrast, SGLT2 inhibitors act predominantly through haemodynamic mechanisms—natriuresis, osmotic diuresis, reduction of intraglomerular pressure, and improved myocardial substrate utilisation. Together, these drug classes address multiple pathophysiological defects simultaneously, providing a mechanistic basis for potentially additive cardiometabolic and renal benefit [50, 51]. The following sections review the evidence for additive effects across key clinical outcomes.

Glycaemic control

Several randomised clinical trials demonstrate that GLP-1 RA and SGLT2 inhibitor combination therapy has an additive effect on HbA1c reduction, with decreases ranging from approximately 1.0–1.5% depending on the agent and dose used. Observational studies and meta-analyses corroborate these findings.

The DURATION-8 trial, a multinational double-blind randomised phase 3 study with 28 week follow-up, showed that once-weekly exenatide combined with once-daily dapagliflozin in patients with T2DM inadequately controlled with metformin produced an HbA1c reduction of − 2.0% (95% CI − 2.2 to − 1.8), significantly greater than exenatide alone (− 1.6%) or dapagliflozin alone (− 0.6%) [52]. The AWARD-10 trial (24-week follow-up) randomised 424 patients with T2DM inadequately controlled on an SGLT2 inhibitor (with or without metformin), with mean baseline HbA1c of approximately 8.0%, mean BMI ~ 32 kg/m², and mean age 57 years; dulaglutide 1.5 mg added to SGLT2 inhibitor therapy reduced HbA1c by − 1.34%, significantly greater than placebo (− 0.54%) [53]. The SUSTAIN-9 trial randomised 302 patients with T2DM on stable SGLT2 inhibitor therapy (with or without metformin or sulfonylurea), mean baseline HbA1c 8.0%, mean BMI ~ 32 kg/m², mean age ~ 57 years, and mean diabetes duration ~ 9 years; once-weekly semaglutide 1 mg added to SGLT2 inhibitor therapy reduced HbA1c by − 1.42% (estimated treatment difference versus placebo) over 30 weeks [54]. The LIRA-ADD2SGLT2i trial showed that liraglutide 1.8 mg added to SGLT2 inhibitors reduced HbA1c by − 0.98% versus −  0.3% for placebo, with 51.8% of liraglutide-treated patients achieving HbA1c < 7.0% [55]. A post-hoc analysis of the CANVAS study additionally showed that canagliflozin added to GLP-1 RA therapy led to significant HbA1c reductions of − 1.0% and − 1.06% with 100 mg and 300 mg, respectively [64]. A meta-analysis confirmed that GLP-1 RA/SGLT2 inhibitor combination therapy produced greater HbA1c reductions than either GLP-1 RA (weighted mean difference − 0.61%) or SGLT2 inhibitor alone (− 0.85%) [56]. Table 5 summarises the key randomised trials of GLP-1 RA and SGLT2 inhibitor combination therapy.

Table 5.

Key randomised trials of GLP-1 RA and SGLT2 inhibitor combination therapy

Trial (Year)	Population	Combination	Primary outcome	Key findings
DURATION-8 (2016)	T2DM on metformin (n = 695)	Exenatide QW + dapagliflozin	HbA1c reduction (28 wk)	Combo—2.0% vs. exenatide—1.6% vs. dapa—0.6%; weight —3.4 kg vs. dapa alone; SBP—4.0 mmHg
AWARD-10 (2018)	T2DM on SGLT2i ± metformin (n = 424)	Dulaglutide 1.5 mg + SGLT2i	HbA1c reduction (24 wk)	Combo—1.34% vs. placebo—0.54%; weight —3.1 vs. —2.1 kg; no increase in serious AEs
SUSTAIN-9 (2019)	T2DM on SGLT2i ± OAD (n = 302)	Semaglutide 1 mg QW + SGLT2i	HbA1c ETD (30 wk)	ETD—1.42%; weight —3.81 kg; SBP—6.32 mmHg; well tolerated
LIRA-ADD2SGLT2i (2020)	T2DM on SGLT2i ± metformin (n = 303)	Liraglutide 1.8 mg + SGLT2i	HbA1c reduction (26 wk)	Combo—0.98% vs.− 0.3%; 51.8% achieved HbA1c < 7.0%; no additional weight loss vs. SGLT2i alone
Vernstrom et al. (2024)	T2DM/obesity at CV risk (n = 106)	Empagliflozin + semaglutide	Vascular function (32 wk)	Weight —8.8 kg (9.9%); SBP—10 mmHg (24-hr); UACR—36% vs. placebo

AE, Adverse event; ETD, Estimated treatment difference; OAD, Oral antidiabetic drug; QW, Once weekly; SBP, Systolic blood pressure; UACR, Urinary albumin-to-creatinine ratio; wk, Weeks

Body weight

Moderate body weight reductions are consistently observed with GLP-1 RA/SGLT2 inhibitor combination therapy, though the magnitude appears to depend on the specific GLP-1 RA used and its dose. In the SUSTAIN-9 trial, semaglutide added to SGLT2 inhibitor therapy led to a greater weight reduction (− 3.81 kg) than placebo [54]. The combination of empagliflozin plus semaglutide for 16 weeks resulted in a weight reduction of − 8.8 kg (9.9%), greater than with either agent alone [57]. In the DURATION-8 study, the combination of exenatide and dapagliflozin led to greater weight loss and a higher proportion of patients achieving > 5% weight loss compared with either agent alone [52]. The AWARD-10 study showed an absolute weight loss of − 3.1 kg with dulaglutide 1.5 mg added to SGLT2 inhibitors, significantly greater than placebo (− 2.1 kg, p = 0.028) [53]. Notably, liraglutide added to SGLT2 inhibitors in the LIRA-ADD2SGLT2i trial was not associated with a significant additional reduction in body weight [55], suggesting drug-specific rather than class-wide weight effects. A meta-analysis showed that the combination produced greater weight reductions than GLP-1 RA alone (− 2.59 kg) but comparable reductions to SGLT2 inhibitor monotherapy [56].

Major adverse cardiovascular events (MACE)

Results on cardiovascular event rates are mixed across study designs. Several randomised trials have not demonstrated a lower rate of cardiovascular events with combination therapy. A post-hoc analysis of the CANVAS trial found no significant difference in the effect of canagliflozin on MACE between patients treated with or without a GLP-1 RA at baseline [58]; of the 10,142 CANVAS participants, approximately 9.9% were receiving a GLP-1 RA at baseline, substantially limiting the statistical power of this subgroup analysis. The SOUL trial, which randomised 9650 patients with T2DM and atherosclerotic cardiovascular disease and/or CKD to oral semaglutide or placebo, showed that the 14% reduction in the primary cardiovascular composite outcome was consistent regardless of baseline SGLT2 inhibitor use [44] approximately 31% of SOUL participants were receiving an SGLT2 inhibitor at baseline, representing the largest combination subgroup reported in any GLP-1 RA cardiovascular outcome trial to date. The FLOW trial similarly reported that cardiovascular event rates were similar irrespective of SGLT2 inhibitor use [14]; approximately 16% of FLOW participants were receiving an SGLT2 inhibitor at baseline.

In contrast, several large observational studies suggest a clinically meaningful cardiovascular benefit from combination therapy on atherosclerotic outcomes. A nested-cohort study from the UK (n = 336,334 patients with T2DM without CVD) showed that the adjusted odds ratio for MACE associated with GLP-1 RA/SGLT2 inhibitor combination was 0.70 [59]. A large UK observational study found that the combination of GLP-1 RA and an SGLT2 inhibitor was associated with a ~ 30% lower risk of MACE compared with either agent alone [60].

Two recent meta-analyses deserve specific mention. Neuen et al. [61] conducted a systematic review and meta-analysis of randomised trial subgroup data, finding that the effects of GLP-1 RAs on cardiovascular outcomes were consistent regardless of background SGLT2 inhibitor use, and vice versa. The complementary SMART-C collaborative meta-analysis [62] addressed the mirror question: whether SGLT2 inhibitor benefits are consistent regardless of background GLP-1 RA use. Across 12 randomised trials including 73,238 participants with diabetes—of whom only 3065 (4.2%) were receiving a GLP-1 RA at baseline—SGLT2 inhibitors reduced MACE consistently in patients both receiving and not receiving a GLP-1 RA (HR 0.81 vs. 0.90; p-heterogeneity = 0.31). Effects on HF hospitalisation (p-heterogeneity = 0.90) and CKD progression (p-heterogeneity = 0.81) were also consistent. The appropriate interpretation of these subgroup analyses warrants careful emphasis: non-significant interaction terms do not mean that the two drug classes “do not work in combination”. On the contrary, they indicate that the added drug worked as well in patients on background therapy as in those without background therapy—a finding fully consistent with an additive effect. The principal limitation is statistical power: subgroups receiving combination therapy in individual trials are typically too small to demonstrate a statistically significant within-subgroup effect or to establish additivity formally. The available randomised evidence is therefore most consistent with additive rather than synergistic effects on MACE, with no signal of subadditivity or interference; observational data point in the same direction. A definitive answer awaits a dedicated, adequately powered combination trial. With respect to lipid profiles, a meta-analysis demonstrated that GLP-1 RA/SGLT2 inhibitor combination therapy was associated with modest reductions in LDL-cholesterol and triglycerides compared with either agent alone [63], providing an additional potential mechanism for cardiovascular risk reduction beyond glycaemic, weight, and blood pressure effects.

Combination therapy with GLP-1 RAs and SGLT2 inhibitors also consistently reduces systolic blood pressure more than either agent alone, with decreases of approximately 4–10 mmHg across trials [52, 54, 56, 57, 64]. These additive blood pressure effects represent a further potential mechanism of cardiovascular risk reduction, complementing the direct cardiometabolic benefits. However, current guidelines do not recommend the combination specifically as an anti-hypertensive strategy.

Heart failure

Heart failure outcomes with combination therapy warrant separate consideration from MACE because the underlying pathophysiological targets, the patient populations in whom benefit has been demonstrated, and the strength of evidence differ substantially between the two drug classes. As discussed in Sect. "Effects on heart failure", SGLT2 inhibitors reduce heart failure hospitalisations across the full spectrum of ejection fraction (HFrEF, HFmrEF, and HFpEF) in both diabetic and non-diabetic patients, supported by DAPA-HF, EMPEROR-Reduced, EMPEROR-Preserved, and DELIVER. In contrast, the heart failure benefit of GLP-1 RAs is restricted to obese HFpEF (STEP-HFpEF, STEP-HFpEF DM), is concentrated in patients with BMI ≥ 35 kg/m², and has not been demonstrated in HFrEF.

Direct randomised evidence for the combination in heart failure is limited. The STEP-HFpEF trials enrolled few patients on background SGLT2 inhibitor therapy, precluding a meaningful subgroup analysis of the combination, and the heart failure trials of SGLT2 inhibitors enrolled few patients on GLP-1 RAs. The principal observational evidence comes from a large retrospective US electronic health record study in patients with T2DM and HFpEF, which showed that adding a GLP-1 RA to SGLT2 inhibitor therapy was associated with a lower risk of heart failure exacerbations (RRR 32%; HR 0.62), lower all-cause emergency hospitalisations (RRR 15%; HR 0.74), reduced new-onset atrial fibrillation (RRR 25%; HR 0.81), and lower pulmonary hypertension risk (HR 0.81) compared with SGLT2 inhibitor alone [65]. The UK nested-cohort study by Wright et al. [59] similarly reported a markedly lower risk of heart failure with combination therapy (adjusted OR 0.43) in patients with T2DM without CVD. These observational signals are encouraging but must be interpreted cautiously, given the inherent risks of confounding by indication and channelling bias.

Mechanistically, the case for additive benefit in obesity-related HFpEF is strong: SGLT2 inhibitors act through volume regulation, reduction of intraglomerular and ventricular filling pressures, and improved myocardial energetics, while GLP-1 RAs at obesity doses act through substantial weight reduction, anti-inflammatory effects, and reduction of epicardial adipose tissue. These mechanisms are largely non-overlapping. Whether this mechanistic complementarity translates into clinically meaningful additive benefit on heart failure outcomes—and specifically whether the combination is superior to either drug alone in obese HFpEF—remains to be established in dedicated randomised trials. In HFrEF and in non-obese HFpEF, the addition of a GLP-1 RA to SGLT2 inhibitor therapy cannot currently be recommended, given the absence of supporting trial evidence and unresolved theoretical concerns regarding GLP-1 RA-induced increases in heart rate and sympathetic tone in HFrEF.

Renal outcomes

Data on the renal effects of combination therapy are heterogeneous, and the most informative findings emerge when results are read together rather than in isolation. The CANVAS trial post-hoc analysis found no significant difference in composite renal outcomes between patients treated with or without GLP-1 RA therapy [58]—a finding that, as noted for MACE, is consistent with an additive effect rather than an interaction. The combination of empagliflozin and semaglutide for 32 weeks reduced the urinary albumin-to-creatinine ratio (UACR) by 36% compared with placebo [57]. A large US observational study reported lower risks of acute kidney injury (RRR 29%; HR 0.70) and new renal replacement therapy (RRR 50%; HR 0.47) with GLP-1 RA plus SGLT2 inhibitor versus SGLT2 inhibitor alone [65]. The CONFIDENCE trial, which evaluated the combination of empagliflozin and finerenone, showed a − 36% change in UACR from baseline; however, similar reductions were observed regardless of baseline GLP-1 RA use [66], suggesting that the renal benefit of the SGLT2 inhibitor component may not be substantially augmented by concomitant GLP-1 RA therapy in all patient populations.

A particularly important nuance comes from the prespecified subgroup analysis of the FLOW trial [14]. The overall renal benefit of semaglutide 1 mg in T2DM and CKD (24% reduction in the primary composite renal outcome) was numerically attenuated in the subgroup of patients receiving an SGLT2 inhibitor at baseline (approximately 16% of the trial population), in whom the point estimate for benefit was substantially smaller and the confidence interval crossed unity. While the formal interaction test was not statistically significant—meaning the data do not exclude a true additive effect—the direction and magnitude of the subgroup estimates raise the possibility that a substantial proportion of the renal benefit of GLP-1 RAs may overlap, mechanistically, with that of SGLT2 inhibitors and may therefore not be fully additive when the two are combined. This pattern contrasts with MACE and HF outcomes, where subgroup estimates are generally consistent across baseline therapy strata.

The relevance of glycaemic control as a determinant of renal outcomes deserves emphasis in this context. Improvement in HbA1c is a major determinant of progression of diabetic kidney disease [5], and the renal effects observed with GLP-1 RAs in T2DM trials likely reflect substantial contributions from glycaemic improvement, in addition to direct, weight-related, and anti-inflammatory effects. In patients already well controlled on an SGLT2 inhibitor, the incremental glycaemic improvement attainable by adding a GLP-1 RA may be smaller, which partly explains the attenuation observed in the FLOW subgroup analysis. Conversely, in patients with poor glycaemic control on SGLT2 inhibitor monotherapy, the additional renal benefit of a GLP-1 RA may be more substantial. Taken together, the renal evidence is currently the least homogeneous of the cardiometabolic outcomes considered in this review, and a dedicated randomised trial of the combination versus each agent alone, with hard renal endpoints, is needed before firm conclusions can be drawn.

Mortality

Data on mortality outcomes with combination therapy are largely derived from retrospective observational studies and prespecified subgroup analyses of randomised trials; they should be interpreted with caution given the inherent risk of confounding. A large retrospective observational study showed that patients with atherosclerotic cardiovascular disease and heart failure who received a GLP-1 RA/SGLT2 inhibitor combination had a lower incidence of mortality or hospitalisation than those receiving SGLT2 inhibitor alone (HR 0.78, 95% CI 0.74–0.83) [67]. The SOUL trial showed that oral semaglutide decreased mortality independently of baseline SGLT2 inhibitor use, with consistent hazard ratios of approximately 0.91 in both groups [44]. A large US observational study also reported lower all-cause mortality with combination therapy compared with the SGLT2 inhibitor alone (RRR 35%; HR 0.64) [65].

Safety of combination therapy

Combination therapy with GLP-1 RAs and SGLT2 inhibitors appears to be generally safe and well-tolerated. Adverse events are predominantly mild and transient and are not associated with significantly higher rates of treatment discontinuation. Gastrointestinal symptoms, particularly nausea, are the most commonly reported adverse effects and are primarily attributable to the GLP-1 RA component. The post-hoc CANVAS analysis showed higher incidences of genital mycotic infections and osmotic diuresis-related events with canagliflozin added to GLP-1 RA, consistent with the known class effects of SGLT2 inhibitors [64]. No episodes of major or severe hypoglycaemia were reported in the DURATION-8 trial [52]. Serious adverse event rates were low and comparable across treatment groups in AWARD-10 [53] and SUSTAIN-9 [54]. In the LIRA-ADD2SGLT2i trial, a higher proportion of liraglutide-treated patients reported adverse events (66.3% vs. 47.0%), primarily gastrointestinal, though serious adverse events remained low (2.5% vs. 1.0%) [55]. A retrospective observational study reported a higher association of diabetic retinopathy in the GLP-1 RA/SGLT2 inhibitor group compared with the SGLT2 inhibitor alone (HR 1.36) [65], a finding that warrants further investigation. Particular consideration should be given to elderly and frail patients, who are systematically underrepresented in randomised trials and who are at disproportionate risk of two complementary hazards of combination therapy: sarcopenia and muscle wasting from rapid weight loss driven by the GLP-1 RA, and volume depletion from the natriuretic and osmotic-diuretic effects of the SGLT2 inhibitor, particularly in those concurrently receiving loop diuretics. In this population, treatment initiation should be cautious, dose escalation gradual, and clinical monitoring of hydration status, renal function, blood pressure, and body composition should be more frequent than in younger or non-frail patients (Table 5).

Discussion

The reviewed data are most consistent with additive effects of GLP-1 RA and SGLT2 inhibitor combination therapy on glycaemic control, body weight, and blood pressure; probable but unproven additivity on MACE; additivity on heart failure that is largely confined to obese HFpEF; and uncertain, possibly less-than-additive, effects on renal outcomes. The contribution of glycaemic improvement itself, particularly to renal outcomes, is often under-recognised. None of the reviewed trials has shown supra-additive (synergistic) effects of GLP-1 RAs and SGLT2 inhibitors on any outcome.

The randomised evidence on the combination is limited. Subgroup analyses of CVOTs (CANVAS, SOUL, FLOW) and recent meta-analyses [61, 62] consistently show that the benefits of one class are not modified by background use of the other. As discussed in Sect. "Major adverse cardiovascular events (MACE)", this pattern of non-significant interaction terms reflects additivity, not absence of effect, and should not be read as evidence that the two classes “do not work together”. The main limitation is statistical power: combination subgroups in individual trials are small (approximately 31% of SOUL participants, 16% of FLOW, 4.2% of the SMART-C SGLT2 inhibitor pool, 33% of STEP-HFpEF DM participants), and no trial was designed to test the combination as a primary question. SELECT and STEP-HFpEF, although often cited here, enrolled no patients on background SGLT2 inhibitors; they extend GLP-1 RA benefit to non-diabetic populations but do not directly address the combination.

Observational studies provide complementary but lower-quality evidence. Large electronic health record cohorts and registries [59, 60, 65, 67] point consistently in the same direction as the randomised data—lower MACE, lower heart failure events, and lower mortality with combination therapy—but are subject to confounding by indication, channelling bias, healthy-adherer effects, and residual confounding. They strengthen the rationale for, but cannot replace, dedicated randomised trials.

Given the complementary mechanisms—GLP-1 RAs reducing atherosclerotic events and weight, SGLT2 inhibitors protecting against heart failure and CKD progression—the combination is a rational strategy for patients with T2DM and complex cardiometabolic risk. Larger dedicated randomised trials with cardiovascular and renal endpoints are needed to define its place in routine care. The cardiovascular and HFpEF benefits of high-dose semaglutide in non-diabetic obesity (SELECT, STEP-HFpEF) strengthen the rationale for studying the combination in this broader population. Future research should prioritise trials testing whether obesity-dose GLP-1 RAs combined with SGLT2 inhibitors confer additive benefit in patients with HFpEF, obesity, and established cardiovascular disease, irrespective of diabetes status.

The field is moving quickly. The ongoing RENAL LIFECYCLE trial may inform combination trial design in advanced CKD, and tirzepatide—a dual GIP/GLP-1 receptor agonist with greater weight loss than semaglutide in the SURMOUNT programme—is increasingly used alongside SGLT2 inhibitors, with cardiovascular outcomes awaited from SURMOUNT-MMO. A detailed review of tirzepatide is beyond our scope, but the same questions raised here — mechanistic plausibility versus demonstrated additivity, intra-class heterogeneity, the role of glycaemic control, and the absence of dedicated trials—will apply with equal force to the next generation of incretin-based combinations.

Conclusions

The combination of GLP-1 receptor agonists and SGLT2 inhibitors provides additive benefits on glycaemic control, body weight, and blood pressure. Evidence for additive cardiovascular and renal protection is biologically plausible and supported by observational data, but a dedicated randomised trial with hard outcomes is still lacking. The combination is safe and well tolerated, with largely non-overlapping adverse effects.

Five practical questions can be answered from the current evidence. First, combination therapy is clearly indicated when glycaemic targets are not met on either agent alone: HbA1c reductions of 1.0–1.5% are well established. Second, in patients with very high cardiovascular risk already on one class, adding the other is reasonable and is increasingly supported by guidelines and observational data, although a confirmatory randomised trial is still missing. Third, in T2DM with symptomatic HFpEF on an SGLT2 inhibitor, adding semaglutide at obesity doses (where indicated) is supported by STEP-HFpEF, but additive benefit beyond SGLT2 inhibitor monotherapy awaits prospective confirmation. Fourth, in T2DM with CKD on an SGLT2 inhibitor, adding a GLP-1 RA for further renal protection is plausible from FLOW, but the FLOW subgroup result urges caution and a dedicated trial is needed. Fifth, adding an SGLT2 inhibitor to a GLP-1 RA in non-diabetic obesity, purely for further weight loss, is not currently supported and would be off-label.

Author contributions

E.S. and C.T. conceptualised the review. E.S., C.T., and M.S. conducted the literature search and wrote the original draft. K.K. and V.K. supervised and revised the manuscript. All authors approved the final version.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable

Competing interests

The authors declare no competing interests.