SGLT2 inhibitors: How a diabetes drug class evolves into the HF realm

Despite current established therapy, heart failure (HF) remains a leading cause of morbidity and mortality worldwide. Patients with HF had a substantially reduced median survival rate, regardless of age groups, compared with the general population.

Being an insidious disease which starts even before the first symptoms show, the gradual downward progressing nature of the disease course across the spectrum of stages highlights just how important both prevention and treatment are for HF.

Prevention of symptomatic HF has been highlighted in recent international clinical practice guidelines for HF, [Eur Heart J 2016;37:2129–2200; J Card Fail 2017;23:628–651] given the progressive nature of HF. Identifying at-risk individuals with preclinical cardiac structural/functional precursors of HF is therefore critical, as early initiation of treatment at the preclinical stage can help prevent HF from progressing and improve clinical outcomes. [N Engl J Med 1992;327:685–691; J Am Coll Cardiol 2013;62:1365–1372] Furthermore, the AHA/ACC guidelines have recommended using biomarkers to screen for patients at risk of developing HF.

Once patients develop HF, aggressive treatment is imperative to prevent or delay the adverse trajectory of HF, considering the high mortality risk associated with the disease.  

One major comorbidity often seen in HF patients is diabetes mellitus. HF and diabetes are intricately related: HF patients are at a greater risk of diabetes, while patients with diabetes also have an increased risk of HF. Diabetes puts patients at risk of HF partly due to insulin resistance; while diabetes intensifies the risk for HF through cardiometabolic excesses such as oxidative stress, inflammation, and decoy cytokines. In addition, diabetes also increased the risk of HF-related complications, including death. [Am J Cardiol 1974;34:29–34] As such, having both HF and diabetes substantially worsens the outcomes than having either condition alone.

While currently available HF therapies, for instance diuretics and the RAAS* inhibitors, have demonstrated clinical benefits, in particular for patients with HF with reduced ejection fraction (HFrEF), these treatments are also associated with various adverse effects — including chronic neurohormonal activation and volume depletion. [Curr Hypertens Rep 2016;18:13; J Am Coll Cardiol 2008;52:1527–1539] Hence, this calls for an urgent need for novel therapies with better safety profile. 

Insights from CVOTs

New insights into HF therapeutics have emerged unexpectedly from trials on antidiabetic agents for the treatment of type 2 diabetes (T2D). In 2008, the US FDA mandated that all new antidiabetic medications be tested for cardiovascular (CV) safety. Among these are the SGLT-2** inhibitors, which have been approved for T2D.

The first cardiovascular outcome trial (CVOT) results for an SGLT-2 inhibitor came from the EMPA-REG OUTCOME trial. As with many CVOT trials, the primary endpoint was the 3-point major adverse cardiovascular event (MACE), consisting of a composite of CV death, nonfatal stroke or myocardial infarction (MI). The study population was focused on T2D patients with atherosclerotic cardiovascular disease (ASCVD), while HF, per se, was neither a criterion for inclusion nor exclusion in the study. The resulting population was ~10 percent of the 7,000 patients who had HF and the remaining 90 percent without HF at baseline.

In addition to a significant reduction in the primary composite CV outcome (10.5 percent vs 12.1 percent; hazard ratio [HR], 0.86; p=0.04), empagliflozin also led to significantly lower rates of CV deaths (3.7 percent vs 5.9 percent; HR, 0.62; p<0.001) and all-cause mortality (5.7 percent and 8.3 percent; HR, 0.68; p<0.001) compared with placebo. [N Engl J Med 2015;373:2117-2128]

Somewhat unexpectedly, empagliflozin also significantly reduced the risk for heart failure hospitalization (HHF) by 35 percent compared with placebo (2.7 percent vs 4.1 percent; HR, 0.65; p=0.002). Also, the risk reduction in HHF was seen early in the trial and the benefits continued throughout the study.

The benefit of HHF risk reduction was subsequently replicated in CVOTs for the other SGLT-2 inhibitors including dapagliflozin (in DECLARE-TIMI 58) and canagliflozin (in CANVAS). Even though there are inconsistencies in MACE or CV deaths across the CVOTs, one thing that stood out consistently was the significant reduction in HHF risk of around 30 percent on average across the three trials.

In DECLARE-TIMI 58, HHF risk was 27 percent lower with dapagliflozin (HR, 0.73, 95 percent confidence interval [CI], 0.61– 0.88) and 33 percent lower with canagliflozin in the CANVAS trial (HR 0.67, 95 percent CI 0.52–0.87). [N Engl J Med 2019;380:347-357; N Engl J Med 2017;377:644-657]

While the consistent reductions in HHF risks across the CVOTs validated the HHF benefit as a class effect, these trials also evoke other questions, one of which being: is the benefit similar for patients with HF and those without HF? In other words, are the effects seen about prevention or treatment of HF? 

As the majority of the study population are patients without HF, one lesson that can be drawn from these CVOTs is that these SGLT-2 inhibitors can prevent HF in a population of high-risk diabetes patients with established CVD.

Dedicated trials on HF patients

Given that only minority of the population for the above CVOTs were patients with HF at baseline, the question of treatment can only be answered by dedicated trials enrolling HF patients as a study entry criterion.

SGLT-2 inhibitors were initially studied as an antidiabetic agent in view of their glycosuric effect, which increases urinary excretion of glucose and sodium. With glycosuria, fluid is also excreted through osmotic diuresis, which can lead to decongestion — believed to be part of the mechanism mediating their benefits in HF. Loss of glucose — equivalent to loss of calorie — can translate into reduced adiposity and weight loss.

Meanwhile, adipose tissue represents a site for the production of decoy cytokines, which have direct adverse effect on the heart, or indirect effects via activation of neurohormonal systems, oxidative stress, and inflammation. Neurohormones and decoy cytokines, in turn, exert adverse effects on the heart by causing LV structural and functional changes and aortic stiffness. Taken together, it follows that whatever benefits SGLT-2 inhibitors have on diabetes should benefit the heart as well.

To test the hypothesis that SGLT-2 inhibitor is in fact a CV drug in its own right, regardless of diabetes status, the dedicated HF outcome trials enrolled HF patients with or without diabetes. One such trial is DAPA-HF, which randomized 4,744 patients with HFrEF (defined as an ejection fraction of ≤40 percent) to receive once-daily dapagliflozin 10 mg or placebo, on top of recommended therapy. The composite primary outcome comprised worsening HF events (referred to hospitalization or an urgent visit leading to intravenous therapy for HF) or CV deaths. Of note, about 55 percent of the patients were free of diabetes and 45 percent had diabetes at baseline. [N Engl J Med 2019;381:1995-2008]

Dapagliflozin led to a significant 26 percent reduction in the composite primary outcome compared with placebo (HR, 0.74; p<0.001). Impressively, the number needed to treat was as low as 21. Again, the treatment effects occurred early and continued throughout the study — highlighting that treatment should be started early on as the benefits begin to accrue early.

Similarly, the individual component of HHF (HR, 0.70, 95 percent CI, 0.59–0.83) and CV death (HR, 0.82, 95 percent CI, 0.69–0.98) were both reduced in the dapagliflozin vs the placebo arm.

When the analysis was stratified by diabetes status, dapagliflozin led to significantly lower risk of the primary outcomes regardless of whether the HF patients had diabetes (HR, 0.75, 95 percent CI, 0.63–0.90) or did not (HR, 0.60, 95 percent CI, 0.60–0.88). Also, the reduction in primary outcome persisted across prespecified subgroups, irrespective of age, sex, race, geographic regions, BMI, prior HHF, or aetiology of HF.

Another point to note is that the patients were already on very good medical therapy for HF at baseline, so any benefits seen with dapagliflozin were incremental on top of existing therapy.

In addition, dapagliflozin also appeared to have renal protective effects, as indicated by fewer event in the secondary endpoint of worsening renal function (HR, 0.71; p=0.17) compared with placebo, although the difference between groups were not statistically significant.

Another HF outcome trial with similar design, EMPEROR-Reduced, enrolled similar population of patients (n=3,730) with HFrEF, who were randomized to receive once-daily empagliflozin 10 mg or placebo, on top of recommended therapy. Compared to DAPA-HF, patients in EMPEROR-Reduced were sicker — with higher NT-proBNP levels (by 400-500 pg/mL), lower EF (27 percent vs 31 percent), and lower eGFR*** (61 vs 66 mL/minute/1.73 m²). [N Engl J Med 2020;383:1413-1424]

Consistent with DAPA-HF, the empagliflozin SGLT-2 inhibitor significantly reduced the primary composite outcome of CV death or hospitalization for worsening HF by 25 percent (HR, 0.75; p<0.001).

The key secondary endpoint of adjudicated HHF (consisting of first and recurrent events) was also significantly lower with empagliflozin than placebo (HR, 0.70; p=0.0003).

In contrast to the renal endpoint in DAPA-HF, empagliflozin showed a pronounced renal protective benefit by significantly delaying the decline in eGFR compared with placebo (annual eGFR decline rate, -0.55 vs -2.28 mL/minute/1.73 m²; p<0.001).

In terms of safety profile, both dapagliflozin and empagliflozin were well tolerated. There were no differences between groups in the rates of hypoglycaemia, volume depletion, and renal dysfunction in DAPA-HF. For EMPEROR-Reduced, genital mycotic infection was more frequent in the empagliflozin group. Across the three CVOTs mentioned above, increased rate of genital mycotic infection with SGLT-2 inhibitors has also been observed, and will be a side effect clinicians have to watch out for. 

Overall, the results from both DAPA-HF and EMPEROR-Reduced are complimentary and strengthen the role of SGLT2 inhibitors in the treatment of HF. Data from the EMPEROR-Reduced trial also extend the use of empagliflozin to a wider group of sicker patients.

 
*RAAS: Renin-Angiotensin-Aldosterone System
**SGLT-2: Sodium-glucose co-transporter-2
***eGFR: Estimated glomerular filtration rate