DAPA-HF

Published

McMurray JJV, et al. "Dapagliflozin in Patients With Heart Failure and Reduced Ejection Fraction". The New England Journal of Medicine. 2019. 381(21):1995-2008.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In individuals with heart failure with reduced ejection fraction (NYHA II-IV, LVEF ≤40%) with or without diabetes, does the addition of the SGLT-2 inhibitor dapagliflozin reduce rates of cardiovascular death or worsening heart failure?

Bottom Line

Among individuals with HFrEF (NYHA II-IV, LVEF ≤40%) with or without T2DM, the addition of the SGLT-2 inhibitor dapagliflozin decreased rates of CV death or worsening HF, as well as all-cause mortality.

Major Points

In 2008, the US Food and Drug Administration began requiring every new drug for diabetes to be tested in a cardiovascular outcomes trial. Because of this, the development of three new sodium-glucose cotransporter 2 (SGLT-2) inhibitors for diabetic patients was accompanied by three cardiovascular outcomes trials (CANVAS, EMPA-REG OUTCOME, and DECLARE-TIMI 58). EMPA-REG and CANVAS showed reductions in composite primary outcomes of CV mortality, nonfatal MI, or nonfatal stroke, as well as reductions in HF hospitalizations among patients with diabetes. Though the mechanisms of cardiovascular benefit remains unclear, it is likely to be driven by a reduction in HF death given that rates of MI were similar between treatment arms. In the more recent DECLARE-TIMI 58 study, dapagliflozin did not show a statistically significant reduction in the primary outcome of major adverse CV events but did show a reduction in the composite of CV death or hospitalization for HF. Prior studies were not powered to examine this class of medications in the setting of HF with reduced ejection fraction (HFrEF).

The 2019 DAPA-HF trial was conducted to study whether the addition of the SGLT-2 inhibitor dapagliflozin could benefit patients with HFrEF, either with or without T2DM. DAPA-HF randomly assigned nearly 5,000 patients receiving standard medical care to dapagliflozin or placebo. Results indicated that SGLT-2 inhibitor therapy resulted in a 4.9% absolute reduction in the primary outcome of CV death or worsening HF, defined as urgent evaluation or hospitalization for HF. There was also a 2.3% absolute reduction in all-cause mortality. Importantly, subgroup analyses demonstrated similar improvements in patients with and without T2DM. Adverse events were similar in both arms.

Overall, the results of DAPA-HF confirm the benefit of SGLT-2 inhibition in patients with HFrEF and T2DM, and provide initial evidence of their efficacy in patients with HFrEF without T2DM. These findings were later confirmed in the EMPEROR-Reduced (2020) trial. Of note, the related EMPEROR-Preserved (2021) trial and DELIVER (2022) confirmed benefit of SGLT2-inhibitors in HFmrEF and HFpEF.

Guidelines

ACC/AHA/HFSA heart failure (2022, adapted)

• If symptomatic chronic HFrEF, SGLT2i recommended to reduce HF hospitalizations and CVD mortality, regardless of diabetes status (COR 1, LOE A)
• If HFpEF or HFmrEF, SGLT2i can be useful for lowering HF mortality and HF hospitalizations (COR 2a, LOE B-R)

Design

• Multicenter, double-blind, parallel-group, randomized, controlled trial
• N=4,744 patients with HFrEF, EF ≤40%, and NYHA II-IV symptoms
  • Dapagliflozin (n=2,373)
  • Placebo (n=2,371)
• Setting: 410 centers in 20 countries (14% North America)
• Enrollment: 2017-2018
• Median follow-up: 18.2 months
• Analysis: Intention-to-treat
• Primary outcome: Composite of worsening HF or CV death

Population

Inclusion Criteria

Patients were included if they met all of the following criteria:

• Age of at least 18 years
• Ejection fraction of ≤40%
• NYHA Class II, III, or IV symptoms
• Plasma NT-proBNP level of:
  • ≥ 600pg/mL OR
  • ≥ 400pg/mL if they were hospitalized for HF within the past 12 months OR
  • ≥ 900pg/mL if patient had atrial fibrillation/flutter on baseline ECG.

Additionally, all patients were required to receive standard heart-failure device therapy (ICD, CRT or both) and standard drug therapy (ACEi/ARB/ARNI + beta-blocker unless contraindicated or resulting in unacceptable side effects). MRA (mineralocorticoid receptor antagonism) use was encouraged. All drug doses were individually tailored, and dosing of anti-hyperglycemics including insulin was titratable as seen fit by the prescribing physician.

Exclusion Criteria

• Unacceptable side effects associated with SGLT2i
• Type 1 Diabetes
• Hypotension/ sBP <95 mm HG
• Estimated glomerular filtration rate (eGFR) ≤30 ml/m/1.73m2

• Current decompensated HF or HF hospitalization <4 weeks prior
• MI, unstable angina, stroke, or TIA within 3 months to enrolment

Baseline Characteristics

Baseline characteristics were well balanced. Below is reported baseline characteristics from the intervention group (dapagliflozin).

• Type 2 diabetes: 42%
• Mean age: 66 years
• Mean BMI: 28 kg/m2
• Female: 24%
• Mean Ejection Fraction: 31%
• Mean eGFR: 66 ml/min/1.73 m²
• NYHA Classification
  • NYHA II: 68%
  • NYHA III: 32%
  • NYHA IV: 1%
• HF medication
  • ACEi/ARB/ARNI: 93%
  • Beta-blocker: 96%
  • MRA: 71%
  • Diuretics: 93%
  • ICD: 26%

Interventions

After a 14-day screening period, patients were randomly assigned to receive either:


• Intervention: dapagliflozin 10mg once daily
• Control: placebo

Outcomes

Comparisons are dapagliflozin vs. placebo.

Primary Outcomes

Worsening heart failure (hospitalization or urgent visit resulting in IV therapy for HF) or CV mortality

16.3% vs. 21.2% (HR 0.74; 95% CI 0.65-0.85; P<0.001)

Components of the primary composite outcome are included in this section for simplicity

Hospitalization/urgent visit for heart failure: 10.0% vs. 13.7% (HR 0.70; 95% CI 0.59-0.83; P not given)

CV mortality: 9.6% vs. 11.5% (HR 0.82; 95% CI 0.69-0.98; P not given)

Secondary Outcomes

Cardiovascular death or heart-failure hospitalization

16.1% vs 20.9% (HR 0.75; 95% CI 0.65-0.85; P<0.001)

Changes in KCCQ (Kansas City Cardiomyopathy Questionnaire) total symptom score at 8 months

6.1±18.6 vs 3.3±19.2 (HR 1.18; 95% CI 1.11-1.26; P<0.001)

Worsening renal function

1.2% vs 1.6% (HR 0.71; 95% CI 0.44-1.16; P not given)

All cause mortality

11.6% vs 13.9% (HR 0.83; 95% CI 0.71-0.97; P not given)

Subgroup Analysis

For the primary outcome

NYHA class II: HR 0.63 (95% CI 0.52-0.75)

NYHA class III or IV: HR 0.90 (95% CI 0.74-1.09)

No P-value for interaction is presented, though the authors note that a greater effect for SGLT2 inhibition may be present among those with NYHA class II HF symptoms.

T2DM at baseline: HR 0.75 (95% CI 0.63-0.90)

No T2DM at baseline: HR 0.73 (95% CI 0.60-0.88)

There was no other meaningful difference between groups, when analyzed by age, sex, race, geographic region, LVEF level, level of NT-proBNP, HF hospitalization history, use of mineralocorticoids at baseline, AF or flutter on ECG, ischemic vs. other cardiomyopathy, BMI, or eGFR.

Adverse Events

Discontinuation due to adverse event

4.7% vs 4.9% (P=0.79)

Renal adverse event

6.5% vs 7.2% (P=0.36)

Fracture

2.1% vs 2.1% (P=0.36)

Amputation

0.5% vs 0.5% (P=1.00)

Major hypoglycemia (requiring intervention of another person)

0.2% vs 0.2% (P=n/a; only occurred in patients with pre-existing diabetes)

DKA

0.1% vs 0% (P=n/a; only occurred in patients with pre-existing diabetes

Fournier’s Gangrene

0 vs <0.1 (P=N/A; only occurred in 1 patient in the placebo group)

Criticisms

• Patients were largely those with moderate heart failure, thus further study in patients with severe HF will be needed. Of note, there appears to be a differential benefit by NYHA symptom class, with the primary benefit being noted among patients with more mild NYHA class II symptoms.
• Unclear effects of ARNI in this population, though a post-hoc analysis found attenuation of the benefit among those using sacubitril-valsartan (HR 0.75; 95% CI 0.50, 1.13) when compared to those not using this ARNI (HR 0.74; 95% CI 0.65, 0.86).
• Rates of more common adverse effects and predominant reasons for discontinuation SGLT2i in earlier studies such as genitourinary infections were not reported
• SGLT2i medications have diuretic properties and diuretics are staples of HF management. It is unclear how diuresis may have contributed to the different outcomes between the treatment and placebo arms.

Funding

AstraZeneca (producers of dapagliflozin)