Double-Blind, Randomized, Placebo-Controlled Trial Evaluating the Efficacy and Safety of Eplerenone in Japanese Patients With Chronic Heart Failure (J-EMPHASIS-HF)

Abstract

Background: The mineralocorticoid receptor antagonist eplerenone improved clinical outcomes among patients with heart failure with reduced ejection faction (HFrEF) in the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study. However, similar efficacy and safety have not been established in Japanese patients. We evaluated the efficacy and safety of eplerenone in patients with HFrEF in a multicenter, randomized, double-blind placebo-controlled outcome study (ClinicalTrials.gov Identifier: NCT01115855). The aim of the study was to evaluate efficacy predefined as consistency of the primary endpoint with that of EMPHASIS-HF at a point estimate of <1 for the hazard ratio.

Methods and Results: HFrEF patients with NYHA functional class II–IV and an EF ≤35% received eplerenone (n=111) or placebo (n=110) on top of standard therapy for at least 12 months. The primary endpoint was a composite of death from cardiovascular causes or hospitalization for HF. The primary endpoint occurred in 29.7% of patients in the eplerenone group vs. 32.7% in the placebo group [hazard ratio=0.85 (95% CI: 0.53–1.36)]. Hospitalization for any cause and changes in plasma BNP and LVEF were favorable with eplerenone. A total of 17 patients (15.3%) in the eplerenone group and 10 patients (9.1%) in the placebo group died. Adverse events, including hyperkalemia, were similar between the groups.

Conclusions: Eplerenone was well-tolerated in Japanese patients with HFrEF and showed results consistent with those reported in the EMPHASIS-HF study.

Mineralocorticoid receptor antagonists (MRAs), in addition to standard therapies such as angiotensin-converting enzyme (ACE) inhibitors, angiotensin-receptor blockers (ARBs), and β-blockers, have been shown to improve clinical outcomes among patients with heart failure with reduced ejection faction (HFrEF) in large-scale randomized clinical trials, including RALES¹ with spironolactone and the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study with eplerenone.² Additionally, a Japanese heart failure registry revealed that MRA use was associated with better prognosis among patients with HFrEF.³ By contrast, the addition of a renin inhibitor to evidence-based doses of ACE inhibitors did not result in reduced morbidity or mortality among patients with HFrEF,⁴ indicating that MRAs should be considered essential in combination with ACE inhibitors or ARBs for patients with HFrEF. Consequently, current guidelines recommend the use of a MRA with Class I, Level A evidence as an add-on treatment for patients with symptomatic HF (New York Heart Association [NYHA] functional class II–IV) receiving standard therapy such as an ACE inhibitor or ARB and β-blocker.^5,6 The MRA eplerenone selectively binds to mineralocorticoid receptors and competitively antagonizes the binding of aldosterone. In the EMPHASIS-HF study, a large-scale randomized clinical trial involving 2,737 patients with HFrEF and mild symptoms (NYHA functional class II), eplerenone decreased the relative risk of death from cardiovascular causes or hospitalization for HF by 37%, compared with placebo, in addition to standard of care. This beneficial effect was consistent across all prespecified subgroups, including different regions, age, renal function, primary diagnosis of HF, standard HF treatments, and medical history.^2,7–9

However, the efficacy and safety of eplerenone remain unestablished in Japanese patients. Therefore, we evaluated the effects of eplerenone in HFrEF patients in a multicenter, randomized, double-blind placebo-controlled study (NCT01115855) in Japan. To verify the statistically significant superiority of eplerenone to placebo using mortality and morbidity endpoints, a double-blind placebo-controlled study with thousands of patients is required, as was the case with the EMPHASIS-HF study. However, it was unrealistic to enroll this number of patients in a single-country trial in Japan. Therefore, the present study was designed to evaluate the consistency of results with those from the EMPHASIS-HF study. Consistency was predefined as a point estimate of the hazard ratio <1 in the primary endpoint, which was a composite of death from cardiovascular causes or hospitalization for HF.

Methods

Study Design

The study was a multicenter, randomized, double-blind placebo-controlled, parallel-group study (J-EMPHASIS-HF). Its primary objective was to evaluate whether eplerenone could demonstrate efficacy (the primary and secondary endpoints) in HFrEF patients with NYHA functional class II or higher, regardless of HF symptom severity. To demonstrate efficacy, the consistency of results with the EMPHASIS-HF study was predefined as a point estimate of the hazard ratio <1 in the primary endpoint of death from cardiovascular causes or hospitalization for HF, based on the concept of evaluating treatment effects across different regions in multiregional clinical trials.¹⁰ The planned target number of patients was 220 (180 NYHA functional class II and 40 NYHA functional class III/IV), and the actual number of patients was 221 (183 NYHA functional class II and 38 NYHA functional class III/IV).

The study was conducted according to the Declaration of Helsinki¹¹ and in compliance with the International Conference on Harmonization and Good Clinical Practice guidelines.¹² All local regulatory requirements and sponsoring company policies were followed. The study was monitored by an independent data safety monitoring committee.

Study Patients

All patients gave written informed consent prior to participation. Japanese patients ≥55 years of age who had chronic HF of either ischemic or non-ischemic etiology (duration ≥4 weeks); symptoms of NYHA functional class II or higher; left ventricular ejection fraction (LVEF) ≤30% (or ≤35% in addition to QRS duration >130 ms on ECG); and treatment with ACE inhibitor, ARB, β-blocker, or diuretic. Randomization was performed within 6 months after hospitalization for cardiovascular causes. Patients who had not been hospitalized for cardiovascular causes within 6 months before randomization could be enrolled if their plasma level of B-type natriuretic peptide (BNP) was ≥250 pg/mL or their plasma level of N-terminal proBNP (NT-proBNP) was ≥500 pg/mL for men and ≥750 pg/mL for women within 15 days of randomization.

Key exclusion criteria were acute myocardial infarction or stroke within 30 days prior to randomization, serum potassium level >5.0 mEq/L, estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m² within 24 h prior to randomization, need for a potassium-sparing diuretic such as spironolactone, and any other clinically significant coexisting conditions.

Study Procedures

Patients were assigned to receive eplerenone or a matching placebo in a 1:1 ratio by centralized randomization with stratification by NYHA functional class (II and III/IV) at randomization and eGFR (30 to <50 mL/min/1.73 m² and ≥50 mL/min/1.73 m²) within 24 h before randomization.

Eplerenone was initiated at a dose of 25 mg once daily provided that the serum potassium level was <5.0 mEq/L when dosage was initiated, and increased after 4 weeks to 50 mg once daily (or initiated at 25 mg on alternate days and increased to 25 mg daily, if eGFR was 30 to <50 mL/min/1.73 m²). Thereafter, serum potassium level was measured at each visit except for months 2, 3, and 4. Investigators were instructed to decrease the dose of study drug if the serum potassium level was 5.5–5.9 mEq/L and to withhold the study drug if the serum potassium level was ≥6.0 mEq/L. Potassium was to be re-measured within 72 h after withholding the study drug, and the study drug was to be restarted only if the level was <5.0 mEq/L.

Patients were treated with the study drug for a maximum of 48 months. The study was completed when the last randomized patient had been followed for 1 year. Clinical events were captured until study completion.

Study Endpoints

The primary endpoint was a composite of death from cardiovascular causes or hospitalization for HF (first occurrence). The prespecified adjudicated secondary endpoints were death from any cause; death from cardiovascular causes; hospitalization for any cause; hospitalization for HF; hospitalization for cardiovascular cause; hospitalization for worsening renal function; hospitalization for hyperkalemia; addition or increase of HF medication due to worsening HF; stroke; myocardial infarction; a composite of death from cardiovascular causes, hospitalization for HF, or addition or increase of HF medication due to worsening HF; composite of death from any cause or hospitalization for any cause; and composite of death from HF or hospitalization for HF. Adjudication of the outcomes was carried out by an independent adjudication committee according to prespecified criteria. The analyses of LVEF and plasma BNP also contributed to efficacy assessment.

Statistical Analysis

The target sample size for this study was 220 patients: 180 in the NYHA functional class II group and 40 in the NYHA functional class III/IV group. The sample size for the NYHA functional class II cohort was calculated to have sufficient power to show consistency of the primary endpoint with the EMPHASIS-HF study. Consistency was defined as a point estimate of the hazard ratio <1 in the primary endpoint in this study to show directionally similar results to the EMPHASIS-HF study. With a sample size of 180 patients in the NYHA functional class II cohort, the probability of showing a point estimate of hazard ratio of <1 in the primary endpoint in this study was approximately 80%, given an expected 18% risk reduction in the hazard ratio of the primary endpoint. The target sample size for the NYHA functional class III/IV cohort was calculated at 40 patients to have the same distribution of NYHA functional classes as the Japanese HF patient population, referring to the approximately 20% proportion of NYHA functional class III/IV patients in the prospective cohort investigation of the CHART-1 study.^13,14 We planned to evaluate the consistency of the primary endpoint using the entire population of 220 patients. Ultimately, a total population of 221 patients was enrolled, and this entire population was used for the primary endpoint evaluation.

Because the study was designed to evaluate the consistency of primary endpoint results with the EMPHASIS-HF study, no confirmatory hypothesis testing was planned. P values were calculated for exploratory purposes after unblinding. The comparability of baseline characteristics between the 2 study groups was assessed by 2-sample t-test for continuous variables or Fisher’s exact test for categorical variables. The primary and secondary outcomes were analyzed using data from all patients who underwent randomization, according to the intention-to-treat principle, with the use of Kaplan-Meier estimates and Cox proportional-hazards models. Hazard ratios, 95% confidence intervals (CI), and P values were calculated with models adjusted for eGFR (<50 mL/min/1.73 m² and ≥50 mL/min/1.73 m²) and NYHA functional class (II and III/IV). For the interpretation of adjudicated secondary outcomes, a P-value of 0.05 was considered the threshold of significance for exploratory purposes. The consistency of the treatment effect was assessed among prespecified subgroups. The effect in each subgroup was analyzed with a Cox proportional-hazards model, without adjustment for covariates. The treatment-by-subgroup interaction was evaluated by using a Cox proportional-hazards model with terms for treatment, subgroup, and their interaction. Repeatedly measured plasma BNP, LVEF, serum potassium and creatinine levels, and systolic blood pressure were compared between the 2 study groups at each measurement time point by 2-sample t-test. P values for comparisons of the incidence of adverse events between treatments were determined using Fisher’s exact test.

Results

Study Patients

This study was conducted at 52 sites in Japan from July 30, 2010, through September 7, 2015 (Appendix 1,S1). A total of 221 patients were randomized: 111 to receive eplerenone and 110 to receive placebo. All patients were treated with a study drug and included in efficacy and safety analyses.

Demographic and other baseline characteristics were generally similar between the treatment groups (Table 1). The majority of patients were male (76.6% and 82.7% in the eplerenone and placebo groups, respectively), and the mean (±SD) age was 69.0±8.7 years in the eplerenone group and 68.4±7.7 years in the placebo group. There were 183 NYHA functional class II patients and 38 NYHA functional class III patients. In the majority of patients, the principal cause of HF was non-ischemic (72.1% in the eplerenone group and 60.9% in the placebo group). Patients’ demographic and other baseline characteristics were also generally similar to those of patients in the EMPHASIS-HF study, except for lower body mass index (22.6 vs. 27.5 kg/m²), eGFR (56.6 vs. 70.8 mL/min/1.73 m²), prevalence of ischemic HF (33.5% vs. 67.5%) and use of ACE inhibitor (49.8% vs. 77.6%), and higher prevalence of idiopathic dilated cardiomyopathy (45.2% vs. 22.3%) and use of ARB (36.7% vs. 19.3%) (Table 2).

Table 1. Baseline Characteristics of the Patients in the J-EMPHASIS-HF Study

Characteristic	Eplerenone (n=111)	Placebo (n=110)	P value
Age (years)	69.0±8.7	68.4±7.7	0.58
Male sex, n (%)	85 (76.6)	91 (82.7)	0.32
BMI (kg/m2)	22.5±3.6	22.7±3.4	0.73
Heart rate (beats/min)	74±11	75±14	0.77
Systolic blood pressure (mmHg)	118±19	117±19	0.72
Diastolic blood pressure (mmHg)	70±12	69±11	0.47
LVEF (%)	25.6±5.0	26.6±4.0	0.10
NYHA functional class			0.86
II	91 (82.0)	92 (83.6)
III/IV	20 (18.0)	18 (16.4)
Principal cause of HF, n (%)			0.29
Ischemic heart disease	31 (27.9)	43 (39.1)
Idiopathic dilated cardiomyopathy	52 (46.8)	48 (43.6)
Hypertension	9 (8.1)	6 (5.5)
Other causes	19 (17.1)	13 (11.8)
Medical history, n (%)
Hospitalization for HF	90 (81.1)	79 (71.8)	0.11
Hypertension	63 (56.8)	66 (60.0)	0.68
Diabetes mellitus	37 (33.3)	51 (46.4)	0.05
Angina pectoris	11 (9.9)	25 (22.7)	0.01
MI	28 (25.2)	39 (35.5)	0.11
Percutaneous coronary intervention	27 (24.3)	36 (32.7)	0.18
Coronary artery bypass grafting	7 (6.3)	19 (17.3)	0.01
Atrial fibrillation or flutter	37 (33.3)	39 (35.5)	0.78
Stroke	16 (14.4)	15 (13.6)	1.00
ICD	8 (7.2)	7 (6.4)	1.00
CRT/CRT-D	16 (14.4)	10 (9.1)	0.30
BNP (pg/mL)	469.3±375.4	435.6±391.5	0.51
N-terminal pro-BNP (pg/mL)	2,635.8±2,143.6	2,354.3±1,874.3	0.30
Hemoglobin (g/dL)	13.4±1.6	13.5±1.5	0.47
Serum creatinine (mg/dL)	0.99±0.26	1.02±0.28	0.35
eGFR (mL/min/1.73 m2)	57.1±15.8	56.1±14.6	0.60
Serum potassium (mEq/L)	4.2±0.4	4.2±0.4	0.69
Medication at randomization visit, n (%)
Diuretic	95 (85.6)	93 (84.5)	0.85
ACE inhibitor	56 (50.5)	54 (49.1)	0.89
ARB	41 (36.9)	40 (36.4)	1.00
ACE inhibitor, ARB, or both	92 (82.9)	90 (81.8)	0.86
β-blocker	98 (88.3)	92 (83.6)	0.34
Digitalis	12 (10.8)	19 (17.3)	0.18
Antiarrhythmic drug	23 (20.7)	26 (23.6)	0.63
Antiplatelet	48 (43.2)	57 (51.8)	0.23
Oral anticoagulant	50 (45.0)	66 (60.0)	0.03
Statin	50 (45.0)	62 (56.4)	0.11

Numerical data are expressed as mean±SD or n (%). ACE, angiotensin-converting enzyme; ARB, angiotensin-receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; CRT, cardiac-resynchronization therapy; CRT-D, implantable cardioverter-defibrillator with cardiac resynchronization; eGFR, estimated glomerular filtration rate; HF, heart failure; ICD, implantable cardioverter-defibrillator; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association.

Table 2. Baseline Characteristics of the Patients in the J-EMPHASIS-HF and EMPHASIS-HF Studies

Characteristic	J-EMPHASIS-HF (n=221)	EMPHASIS-HF (n=2,737)
Age (years)	68.7 (8.2)	68.6 (7.6)
Male sex, n (%)	176 (79.6)	2,127 (77.7)
BMI (kg/m2)	22.6 (3.48)	27.5 (4.9)
Heart rate (beats/min)	74.3 (12.6)	71.7 (12.5)
Systolic blood pressure (mmHg)	117.6 (18.6)	124.1 (16.9)
Diastolic blood pressure (mmHg)	69.5 (11.7)	74.6 (10.2)
LVEF (%)	26.1 (4.5)	26.1 (4.7)
Principal cause of HF, n (%)*
Ischemic heart disease	74 (33.5)	1,840 (67.5)
Idiopathic dilated cardiomyopathy	100 (45.2)	609 (22.3)
Hypertension	15 (6.8)	162 (5.9)
Other causes	32 (14.5)	114 (4.2)
Medical history, n (%)
Hospitalization for HF	169 (76.5)	1,440 (52.6)
Hypertension	129 (58.4)	1,819 (66.5)
Diabetes mellitus	88 (39.8)	859 (31.4)
Angina pectoris	36 (16.3)	1,189 (43.4)
MI	67 (30.3)	1,381 (50.5)
Percutaneous coronary intervention	63 (28.5)	596 (21.8)
Coronary artery bypass grafting	26 (11.8)	516 (18.9)
Atrial fibrillation or flutter	76 (34.4)	844 (30.8)
Stroke	31 (14.0)	262 (9.6)
ICD	15 (6.8)	362 (13.2)
CRT/CRT-D	26 (11.8)	230 (8.4)
Hemoglobin (g/Dl)	13.5 (1.5)	13.8 (1.6)
Serum creatinine (mg/dL)	1.01 (0.27)	1.15 (0.30)
eGFR at baseline (mL/min/1.73 m2)	56.6 (15.2)	70.8 (21.8)
Serum potassium (mEq/L)	4.22 (0.39)	4.32 (0.43)
Medication at randomization visit, n (%)
Diuretic	188 (85.1)	2,326 (85.0)
ACE inhibitor	110 (49.8)	2,123 (77.6)
ARB	81 (36.7)	527 (19.3)
ACE inhibitor, ARB, or both	182 (82.4)	2,557 (93.4)
β-blocker	190 (86.0)	2,374 (86.7)
Digitalis	31 (14.0)	740 (27.0)
Antiarrhythmic drug	49 (22.2)	388 (14.2)
Antiplatelet	105 (47.5)	1,807 (66.0)
Oral anticoagulant	116 (52.5)	863 (31.5)
Statin	112 (50.7)	1,670 (61.0)

Numerical data are expressed as mean±SD or n (%). *Patients without data on principal cause of HF were excluded from the analysis. Abbreviations as in Table 1.

Study Drug Administration and Follow-up

Among patients with baseline eGFR ≥50 mL/min/1.73 m², 68 of 74 patients (91.9%) in the eplerenone group and 63 of 71 patients (88.7%) in the placebo group had a maximum dose of 50 mg/day. The mean (±SD) dose was 45.5±9.5 mg/day in the eplerenone group and 45.2±10.9 mg/day in the placebo group at the end of treatment. In patients with baseline eGFR <50 mL/min/1.73 m², 35 of 37 patients (94.6%) in the eplerenone group and 39 of 39 patients (100.0%) in the placebo group had a maximum dose of 25 mg/day. The mean (±SD) dose was 21.1±6.0 mg/day in the eplerenone group and 21.8±5.7 mg/day in the placebo group at the end of treatment.

A total of 36 patients (32.4%) in the eplerenone group and 36 patients (32.7%) in the placebo group discontinued study drug treatment (Table S1). The median duration of study treatment was 749 days in the eplerenone group and 742 days in the placebo group. The median duration of follow-up was 862 days in the eplerenone group and 895 days in the placebo group.

Efficacy Outcomes

The primary endpoint, a composite of death from cardiovascular causes or hospitalization for HF (first occurrence), occurred in 33 patients (29.7%) in the eplerenone group and in 36 patients (32.7%) in the placebo group. The hazard ratio of time to the first occurrence of the primary endpoint was 0.85 with a 95% CI of 0.53–1.36 (Table 3, Figure 1A). The effect of eplerenone on the primary outcome was generally consistent across the prespecified subgroups (Figure 2).

Table 3. Primary and Secondary Outcomes in the J-EMPHASIS-HF Study

Outcome*	Eplerenone (n=111)	Placebo (n=110)	Hazard ratio	95% CI	P value
Primary outcome
Death from CV causes or hospitalization for HF	33 (29.7)	36 (32.7)	0.85	(0.53, 1.36)	0.50
Secondary outcomes
Death from CV causes, hospitalization for HF, or addition or increase of HF medication due to worsening HF	42 (37.8)	45 (40.9)	0.86	(0.56, 1.31)	0.48
Death from any cause	17 (15.3)	10 (9.1)	1.77	(0.81, 3.87)	0.15
Death from CV causes	14 (12.6)	6 (5.5)	2.40	(0.92, 6.24)	0.07
Hospitalization for any cause	45 (40.5)	58 (52.7)	0.65	(0.44, 0.97)	0.03
Hospitalization for HF	27 (24.3)	33 (30.0)	0.75	(0.45, 1.25)	0.27
Hospitalization for CV causes	35 (31.5)	44 (40.0)	0.70	(0.45, 1.10)	0.12
Death from any cause or hospitalization for any cause	48 (43.2)	61 (55.5)	0.66	(0.45, 0.97)	0.03
Death from HF or hospitalization for HF	29 (26.1)	33 (30.0)	0.81	(0.49, 1.33)	0.40
Addition or increase of HF medication due to worsening HF	38 (34.2)	43 (39.1)	0.83	(0.53, 1.28)	0.40
Fatal or non-fatal stroke	3 (2.7)	4 (3.6)	0.79	(0.18, 3.53)	0.76
Fatal or non-fatal MI	1 (0.9)	1 (0.9)	1.11	(0.07, 17.85)	0.94
New-onset atrial fibrillation or flutter	4 (3.6)	2 (1.8)	2.11	(0.39, 11.56)	0.39
New-onset diabetes mellitus	1 (0.9)	2 (1.8)	0.51	(0.05, 5.66)	0.59
Hospitalization for worsening renal function	2 (1.8)	2 (1.8)	1.13	(0.16, 8.04)	0.90
Hospitalization for hyperkalemia	0	0	–	–	–

*Data are shown as number of patients (%). CI, confidence interval; CV, cardiovascular; HF, heart failure.

Figure 1.

Kaplan-Meier plots for (A) death from cardiovascular causes or hospitalization for heart failure (the primary outcome), (B) hospitalization for any cause, (C) hospitalization for heart failure, and (D) death from any cause. CI, confidence interval.

Figure 2.

Hazard ratios for death from cardiovascular causes or hospitalization for heart failure (the primary outcome) according to subgroups. Data are missing for some patients in some subgroups. ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BNP, B-type natriuretic peptide; CI, confidence interval; CRT, cardiac-resynchronization therapy; CV, cardiovascular; GFR, glomerular filtration rate; ICD, implantable cardioverter-defibrillator; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; UACR, urine albumin-creatinine ratio.

The rate of hospitalization for any cause was siginificantly lower in the eplerenone group (hazard ratio, 0.65; 95% CI, 0.44–0.97, P=0.03) (Table 3, Figure 1B). Death from any cause or hospitalization for any cause was also significantly lower in the eplerenone group (P=0.03) (Table 3). Hospitalization for HF occurred in 27 patients (24.3%) in the eplerenone group and in 33 patients (30.0%) in the placebo group (hazard ratio, 0.75; 95% CI, 0.45–1.25) (Table 3, Figure 1C).

The rate of death from cardiovascular causes tended to be higher in the eplerenone group than in the placebo group (12.6% vs. 5.5%); however, this difference did not reach statistical significance (95% CI, 0.92–6.24; P=0.07). Death from any cause occurred in 17 patients (15.3%) in the eplerenone group and 10 patients (9.1%) in the placebo group (hazard ratio, 1.77; 95% CI, 0.81–3.87; P=0.15) (Table 3, Figure 1D).

The most common cause of death was worsening HF (7 patients in the eplerenone group and 3 in the placebo group), and 2 patients in each group died during study drug treatment (Table S2). None of the 3 deaths in the placebo group was complicated by events other than HF at the time of death, but 4 of the 7 deaths in the eplerenone group were complicated by comorbidities, including multi-organ failure, disseminated intravascular coagulation, rhabdomyolysis, sepsis, emphysema, and cardiac sarcoidosis.

Death during the treatment period (i.e., discontinuation of treatment due to death) occurred in 6 patients in the eplerenone group and in 5 patients in the placebo group, reflecting comparable rates (Table S2). By contrast, death after discontinuation of treatment occurred in 11 patients in the eplerenone group and in 5 patients in the placebo group. The reasons for discontinuation in patients who died after discontinuation in the eplerenone group did not include hyperkalemia or worsening renal function (Table S3).

After unblinding, additional analysis was performed according to patient groups who completed vs. discontinued the study drug treatment. Hazard ratios in the completed group were generally lower than those in the discontinued group. Hazard ratios for death from any cause and death from cardiovascular causes in the completed group were both 1.06 (Table S4). On-treatment analysis indicated that the hazard ratio of death from any cause was lower than that of the primary analysis (1.36 vs. 1.77). The hazard ratio of death from cardiovascular causes was also lower (1.74 vs. 2.40) (Tables 3,4).

Table 4. On-Treatment Analysis of Primary and Secondary Outcomes in the J-EMPHASIS-HF Study

Outcome*	Eplerenone (n=111)	Placebo (n=110)	Hazard ratio	95% CI	P value
Primary outcome
Death from CV causes or hospitalization for HF	29 (26.1)	35 (31.8)	0.76	(0.46, 1.25)	0.28
Secondary outcomes
Death from CV causes, hospitalization for HF, or addition or increase of HF medication due to worsening HF	39 (35.1)	44 (40.0)	0.83	(0.54, 1.27)	0.39
Death from any cause	7 (6.3)	5 (4.5)	1.36	(0.43, 4.31)	0.60
Death from CV causes	7 (6.3)	4 (3.6)	1.74	(0.51, 5.97)	0.38
Hospitalization for any cause	42 (37.8)	55 (50.0)	0.66	(0.44, 0.99)	0.04
Hospitalization for HF	25 (22.5)	33 (30.0)	0.69	(0.41, 1.16)	0.16
Hospitalization for CV causes	33 (29.7)	44 (40.0)	0.66	(0.42, 1.04)	0.07
Death from any cause or hospitalization for any cause	45 (40.5)	57 (51.8)	0.68	(0.46, 1.01)	0.06
Death from HF or hospitalization for HF	25 (22.5)	33 (30.0)	0.69	(0.41, 1.16)	0.16
Addition or increase of HF medication due to worsening HF	35 (31.5)	42 (38.2)	0.79	(0.51, 1.24)	0.31
Fatal or non-fatal stroke	3 (2.7)	2 (1.8)	1.38	(0.23, 8.35)	0.73
Fatal or non-fatal MI	1 (0.9)	1 (0.9)	1.10	(0.07, 17.58)	0.95
New-onset atrial fibrillation or flutter	4 (3.6)	1 (0.9)	3.70	(0.41, 33.60)	0.24
New-onset diabetes mellitus	1 (0.9)	2 (1.8)	0.50	(0.05, 5.55)	0.57
Hospitalization for worsening renal function	2 (1.8)	2 (1.8)	1.01	(0.14, 7.30)	0.99
Hospitalization for hyperkalemia	0	0	–	–	–

On-treatment analysis after unblinding, censoring at 30 days after the last dose. *Data are shown as number of patients (%). Abbreviations as in Table 3.

Time-dependent changes in plasma BNP levels and LVEF are shown in Figure 3A,B. Plasma BNP was decreased and LVEF was increased in the eplerenone group compared with the placebo group.

Figure 3.

Time-dependent changes of (A) plasma B-type natriuretic peptide levels and (B) left ventricular ejection fraction. Data are shown as mean±SE. *P<0.05.

Safety Results

A total of 2 patients (1.8%) in each group were hospitalized for worsening renal function; none were hospitalized for hyperkalemia (Table 3).

As detailed above, 36 patients each in the eplerenone group (32.4%) and placebo group (32.7%) discontinued study drug treatment (Table S1).The most common reason for treatment discontinuation was adverse events (16 patients [14.4%] in the eplerenone group, 18 patients [16.4%] in the placebo group).The second most common reason was death (6 patients [5.4%] in the eplerenone group, 5 patients [4.5%] in the placebo group).

Relevant adverse events are summarized in Table 5. The incidence of any adverse event, including hyperkalemia, was generally similar between treatment groups. The incidence of hypokalemia was lower in the eplerenone group compared with the placebo group (1.8% vs. 10.0%, P=0.01).

Table 5. Selected Investigator-Reported Adverse Events, and Those Leading to Permanent Withdrawal of the Study Drug, According to Study Group in the J-EMPHASIS-HF Study*

Event**	Adverse event			Adverse event leading to study drug discontinuation
	Eplerenone (n=111)	Placebo (n=110)	P value	Eplerenone (n=111)	Placebo (n=110)	P value
All events	102 (91.9)	103 (93.6)	0.80	19 (17.1)	20 (18.2)	0.86
Hyperkalemia	8 (7.2)	6 (5.5)	0.78	2 (1.8)	1 (0.9)	1.00
Hypokalemia	2 (1.8)	11 (10.0)	0.01	0	0	–
Renal impairment	5 (4.5)	10 (9.1)	0.19	0	1 (0.9)	0.50
Hypotension	4 (3.6)	7 (6.4)	0.37	0	0	–
Gynecomastia	0	0	–	0	0	–

*Patients who received at least one dose of the study drug were included in this safety analysis. **Data are shown as number of patients (%).

Time-dependent changes in serum potassium and creatinine, and systolic blood pressure are shown in Figure S1. Serum potassium increased in the eplerenone group compared with the placebo group, while serum creatinine and systolic blood pressure did not differ between groups.

Discussion

We evaluated the effects of eplerenone in Japanese HFrEF patients with NYHA functional class II or higher and LVEF ≤35% by administering eplerenone (up to 50 mg daily) or placebo in addition to standard medical therapy. The background characteristics of the patients in the J-EMPHASIS-HF study were comparable between the eplerenone and placebo groups (Table 1) as well as being similar to the EMPHASIS-HF study (Table 2). The subjects enrolled in this study were well managed with standard therapies for HF: 82% (182/221) received ACE inhibitors or ARBs, 86% (190/221) had β-blockers, and 85% (188/221) were on diuretics. Eplerenone was evaluated in addition to these guideline-recommended drugs.

Efficacy in this study was predefined as consistency of the primary endpoint with that of the EMPHASIS-HF trial at a point estimate of <1 for the hazard ratio. The primary endpoint was a composite of death from cardiovascular causes or hospitalization for HF. The primary endpoint of the Japanese trial had a point estimate of 0.85 for the hazard ratio, suggesting that efficacy was in line with that of the EMPHASIS-HF trial (Table 3). The effect of eplerenone on the primary outcome was generally consistent across the prespecified subgroups, except that the hazard ratio appeared to be higher in the subgroup of patients without ACE inhibitor or ARB treatment (Figure 2). We cannot draw a conclusion from this finding based on the J-EMPHASIS-HF study alone, because of the small sample size of the subgroup without ACE inhibitor or ARB treatment (16 patients in the eplerenone group and 18 patients in the placebo group).

Furthermore, the J-EMPHASIS-HF study investigated trends in BNP and LVEF, biomarkers relevant to mortality and morbidity, throughout the study period, which were not investigated in the EMPHASIS-HF study. Compared with the placebo group, BNP tended to decrease and LVEF tended to increase with eplerenone treatment, and these effects persisted throughout the treatment period (Figure 3). Therefore, even though eplerenone did not reduce the primary composite outcome of death from cardiovascular causes or hospitalization for HF with a P-value <0.05, it showed favorable efficacy in reducing hospitalization for any cause and in the changes in plasma BNP and LVEF.

Eplerenone had a favorable safety profile and was well-tolerated in these Japanese HFrEF patients. The incidence of any adverse event was comparable between the eplerenone and placebo groups (91.9% and 93.6%, respectively) (Table 5), and both were higher than in the EMPHASIS-HF study (72.0% and 73.6%, respectively).² This difference might be related to the longer duration of follow-up in the J-EMPHASIS-HF study. The median duration of follow-up was 862 days (29 months) in the eplerenone group and 895 days (30 months) in the placebo group in the J-EMPHASIS-HF study, whereas the median duration of follow-up for all patients was 21 months in the EMPHASIS-HF study.² The incidence of hyperkalemia was similar between the eplerenone and placebo groups, but hypokalemia occurred less frequently in the eplerenone group (Table 5). Serum potassium tended to increase slightly in the eplerenone group compared with the placebo group, suggesting the need to monitor serum potassium and adjust the eplerenone dose accordingly (Figure S1). Gynecomastia was not reported in this study, suggesting an advantage of eplerenone over the non-selective MRA spironolactone, for which sex-hormone-related side effects have been reported.¹

In the J-EMPHASIS-HF study, the majority of deaths in the eplerenone group (10/17 cases) occurred >30 days after discontinuation of treatment. Although the exact cause of the imbalance in deaths was not clear, considering the temporal relationship between death and the treatment, eplerenone itself might not be the direct cause of these deaths. The death rate in the placebo group was low. Kaplan-Meier plots for death from any cause demonstrated that the rate of death in the eplerenone group was 6.21 person-years in the J-EMPHASIS-HF study and 6.93 in the EMPHASIS-HF study, whereas in the placebo group it was 3.51 in the J-EMPHASIS-HF study and 8.89 in the EMPHASIS-HF study (Figure S2). Moreover, the cumulative event rate of death during 1 year in the placebo group in the J-EMPHASIS-HF study (0.92%) was much lower than in the EMPHASIS-HF study (8.08%), CHART-1 trial (7.3%),¹³ and JCARE-CARD trial (8.9%).¹⁵ We compared the patients’ background characteristics and follow-up periods with the EMPHASIS-HF study, as well as with the real-world data reported in the CHART-1 and JCARE-CARD trials,^13,15 but could not find a clear reason to explain the low absolute death rate in the placebo group of the J-EMPHASIS-HF study. Therefore, this low death rate was considered to have occurred by chance because of the small sample size.

In the J-EMPHASIS-HF study, 7 patients in the eplerenone group and 19 in the placebo group underwent revascularization by coronary artery bypass grafting (P=0.01), and 27 patients in the eplerenone group and 36 in the placebo group underwent percutaneous coronary intervention (Table 1). Revascularization has short-term and long-term mortality benefits,¹⁶ which might have contributed to better survival in the placebo group. Also, 5 patients with LVEF ≤15% were enrolled in this study, all of whom were randomized to the eplerenone group and 2 of whom died. More subjects in the eplerenone group were ≥75 years old or had a history of myocardial infarction and stroke (Table S5). Although the background characteristics of the entire population were generally balanced between the eplerenone and placebo groups (Table 1), these differences in background characteristics might have affected mortality outcomes because of the small sample size.

Study Limitations

There are several potential limitations to highlight. First, the sample size in the J-EMPHASIS-HF study was as small (221 patients) compared with the EMPHASIS-HF study (2,737 patients). This sample size was suitable to evaluate the consistency of efficacy results with the EMPHASIS-HF study, but not to assess statistical significance. Second, 11 of 17 deaths in the eplerenone group occurred after study drug discontinuation, which might suggest the importance of continuous treatment with MRAs. However, this observation should be verified by further studies. Third, some parameters that might be useful for evaluating the effect of eplerenone, such as plasma aldosterone levels, were not investigated in this study. Fourth, the increase in LVEF in the eplerenone group might suggest reverse remodeling, as previously described for spironolactone.^17–19 However, echocardiographic data other than LVEF, such as LV dimensions, were not collected in this study.

In conclusion, eplerenone was well-tolerated in Japanese patients with HFrEF and showed results consistent with those reported in the EMPHASIS-HF study.

Acknowledgments

We thank all the patients, investigators, and staff who participated in this study.

Disclosures

This trial was designed and performed by the sponsor (Pfizer). The data were collected and analyzed and the first draft of the manuscript was written by the sponsor. It was fully reviewed and revised by the authors.

Conflict of Interest Statement

H.T. has received speakers’ bureau/honoraria from MSD, Otsuka Pharmaceutical, Ono Pharmaceutical, Daiichi-Sankyo, Takeda Pharmaceutical, Mitsubishi Tanabe Pharma, Teijin Pharma, Nippon Boehringer Ingelheim, Bayer Yakuhin, Pfizer Japan, and Bristol-Myers Squibb; consultation fees from Novartis Pharma K.K., Ono Pharmaceutical, Pfizer Japan, and Bayer Yakuhin; scholarship funds from Novartis Pharma K.K., Daiichi-Sankyo, Astellas Pharma, Takeda Pharmaceutical, Bayer Yakuhin, Nippon Boehringer Ingelheim, Mitsubishi Tanabe Pharma, and Sanofi K.K.; and honoraria for writing promotional material for Igaku-shoin and Medical Review. H.I. has received scholarship funds from Pfizer Japan. M.K. has received speakers’ bureau/honoraria from AstraZeneca K.K. and Mitsubishi Tanabe Pharma; research grants from Takeda Pharmaceutical and Mitsubishi Tanabe Pharma; and scholarship funds from Takeda Pharmaceutical. I.K. has received speakers’ bureau/honoraria from Takeda Pharmaceutical, Mitsubishi Tanabe Pharma, Nippon Boehringer Ingelheim, Astellas Pharma, Shionogi, and Toa Eiyo; scholarship funds from Takeda Pharmaceutical, Nippon Boehringer Ingelheim, Mitsubishi Tanabe Pharma, Edwards Lifesciences Corporation, Sumitomo Dainippon Pharma, Teijin Pharma, and Toa Eiyo; and honoraria for writing promotional material for Ishiyaku Publishers, Igaku-shoin, Bunkodo, and Medical Review. T.M. has received speakers’ bureau/honoraria and scholarship funds from Pfizer Japan. M.Y. has received speakers’ bureau/honoraria from Ono Pharmaceutical and scholarship funds from Boston Scientific Japan, Takeda Pharmaceutical, Otsuka Pharmaceutical, Teijin Pharma, Nippon Boehringer Ingelheim, Daiichi-Sankyo, Shionogi, and MSD. K.Y. has received speakers’ bureau/honoraria from Otsuka Pharmaceutical, Ono Pharmaceutical, Mitsubishi Tanabe Pharma, Toa Eiyo, Takeda Pharmaceutical, Medtronic, Bristol-Myers Squibb, and Pfizer Japan; research funds from Astellas Pharma; and scholarship funds from St. Jude Medical Japan, Otsuka Pharmaceutical, Daiichi-Sankyo, Ono Pharmaceutical, Biotronik Japan, Japan Lifeline, Sanwa Kagaku Kenkyusho, Nippon Boehringer Ingelheim, Abbott Vascular Japan, Bayer Yakuhin, Teijin Pharma, Mitsubishi Tanabe Pharma, Novartis Pharma K.K., Fukuda Denshi, Taisho Toyama Pharmaceutical, and Fukuda Life Tec. T.I., K.S., Y.Y., T.Y., T.T., and M.M. have no conflict of interest. J.Z., A.O., Y.I., and K.K. are Pfizer employees. This study was sponsored by Pfizer.

Appendix 1. Members of the Study Group (1)

Steering Committee Members

Masunori Matsuzaki (St Hill Hospital and Yamaguchi University), Hiroyuki Tsutsui (Kyushu University), Hiroshi Ito (Okayama University), Masafumi Kitakaze (National Cerebral and Cardiovascular Center), Issei Komuro (the University of Tokyo), Toyoaki Murohara (Nagoya University), Tohru Izumi (Kitasato University and Niigata Minami Hospital), Kenji Sunagawa (Kyushu University).

Adjudication Committee Members

Yoshio Yasumura (Osaka Police Hospital), Masafumi Yano (Yamaguchi University), Kazuhiro Yamamoto (Tottori University).

Data Monitoring Committee Members

Mitsuyoshi Urashima (Jikei University), Akihiro Hata (Toshima Hospital), Yasunori Sato (Chiba University).

Medical Advisors

Tsutomu Yoshikawa (Sakakibara Heart Institute), Takayoshi Tsutamoto (Toyosato Hospital).

Supplementary Files

Supplementary File 1

Appendix S1. Members of the Study Group (2)

Figure S1. Time-dependent changes in serum potassium levels (A), serum creatinine levels (B), and systolic blood pressure (C).

Figure S2. Kaplan-Meier plots for death from any cause in the EMPHASIS-HF and J-EMPHASIS-HF studies.

Table S1. Prevalence and reasons for discontinuation of treatment

Table S2. Causes of death during treatment and after discontinuation

Table S3. Reasons for discontinuation among patients who died after discontinuation of treatment

Table S4. Primary and secondary outcomes in completed and discontinued patients

Table S5. Baseline characteristics of the patients

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-17-0323

References

• 1.   Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al; Randomized Aldactone Evaluation Study Investigators. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 1999; 341: 709–717.
• 2.   Zannad F, McMurray JJ, Krum H, van Veldhuisen DJ, Swedberg K, Shi H, et al; EMPHASIS-HF Study Group. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med 2011; 364: 11–21.
• 3.   Hamaguchi S, Kinugawa S, Tsuchihashi-Makaya M, Goto K, Goto D, Yokota T, et al. Spironolactone use at discharge was associated with improved survival in hospitalized patients with systolic heart failure. Am Heart J 2010; 160: 1156–1162.
• 4.   McMurray JJ, Krum H, Abraham WT, Dickstein K, Køber LV, Desai AS, et al; ATMOSPHERE Committees Investigators. Aliskiren, enalapril, or aliskiren and enalapril in heart failure. N Engl J Med 2016; 374: 1521–1532.
• 5.   Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al; Authors/Task Force Members. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2016; 37: 2129–2200.
• 6.   Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Drazner MH, et al. 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation 2013; 128: e240–e327.
• 7.   Swedberg K, Zannad F, McMurray JJ, Krum H, van Veldhuisen DJ, Shi H, et al; EMPHASIS-HF Study Investigators. Eplerenone and atrial fibrillation in mild systolic heart failure: Results from the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study. J Am Coll Cardiol 2012; 59: 1598–1603.
• 8.   Preiss D, van Veldhuisen DJ, Sattar N, Krum H, Swedberg K, Shi H, et al. Eplerenone and new-onset diabetes in patients with mild heart failure: Results from the Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF). Eur J Heart Fail 2012; 14: 909–915.
• 9.   Eschalier R, McMurray JJ, Swedberg K, van Veldhuisen DJ, Krum H, Pocock SJ, et al; EMPHASIS-HF Investigators. Safety and efficacy of eplerenone in patients at high risk for hyperkalemia and/or worsening renal function: Analyses of the EMPHASIS-HF study subgroups (Eplerenone in Mild Patients Hospitalization And Survival Study in Heart Failure). J Am Coll Cardiol 2013; 62: 1585–1593.
• 10.   ICH Steering Committee. Ethnic factors in the acceptability of foreign clinical data (ICH-E5(R1). 1998. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E5_R1/Step4/E5_R1__Guideline.pdf (accessed December 14, 2016).
• 11.   World Medical Association. World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects. Bull World Health Organ 2001; 79: 373–374.
• 12.   ICH Steering Committee. Guideline for good clinical practice E6(R1). 1996. https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E6/E6_R1_Guideline.pdf (accessed February 15, 2017).
• 13.   Shiba N, Watanabe J, Shinozaki T, Koseki Y, Sakuma M, Kagaya Y, et al; CHART Investigators. Analysis of chronic heart failure registry in the Tohoku district: Third year follow-up. Circ J 2004; 68: 427–434.
• 14.   Shiba N, Shimokawa H. Chronic heart failure in Japan: Implications of the CHART studies. Vasc Health Risk Manag 2008; 4: 103–113.
• 15.   Tsuchihashi-Makaya M, Hamaguchi S, Kinugawa S, Yokota T, Goto D, Yokoshiki H, et al; JCARE-CARD Investigators. Characteristics and outcomes of hospitalized patients with heart failure and reduced vs preserved ejection fraction: Report from the Japanese Cardiac Registry of Heart Failure in Cardiology (JCARE-CARD). Circ J 2009; 73: 1893–1900.
• 16.   Petrie MC, Jhund PS, She L, Adlbrecht C, Doenst T, Panza JA, et al; STICH Trial Investigators. Ten-year outcomes after coronary artery bypass grafting according to age in patients with heart failure and left ventricular systolic dysfunction: An analysis of the extended follow-up of the STICH trial (Surgical Treatment for Ischemic Heart Failure). Circulation 2016; 134: 1314–1324.
• 17.   Tsutamoto T, Wada A, Maeda K, Mabuchi N, Hayashi M, Tsutsui T, et al. Effect of spironolactone on plasma brain natriuretic peptide and left ventricular remodeling in patients with congestive heart failure. J Am Coll Cardiol 2001; 37: 1228–1233.
• 18.   Cicoira M, Zanolla L, Rossi A, Golia G, Franceschini L, Brighetti G, et al. Long-term, dose-dependent effects of spironolactone on left ventricular function and exercise tolerance in patients with chronic heart failure. J Am Coll Cardiol 2002; 40: 304–310.
• 19.   Chan AK, Sanderson JE, Wang T, Lam W, Yip G, Wang M, et al. Aldosterone receptor antagonism induces reverse remodeling when added to angiotensin receptor blockade in chronic heart failure. J Am Coll Cardiol 2007; 50: 591–596.