Rationale and design of the HINODE study: Heart failure indication and sudden cardiac death prevention trial Japan

Abstract Background Randomized trials in Western countries have provided evidence that prophylactic implantable cardioverter‐defibrillator (ICD) therapy reduces mortality in heart failure (HF) patients with reduced left ventricular ejection fraction. However, the risk of life‐threatening ventricular arrhythmias in Japanese HF patients sharing similar risk factors is still unknown. Methods The Heart Failure Indication and Sudden Cardiac Death Prevention Trial Japan trial (NCT03185832) is a prospective, multicenter registry designed to collect data on ventricular arrhythmia, HF events, and mortality in Japanese HF patients. Japanese patients with HF and 2‐5 predefined risk factors who were indicated for cardiac device implantation based on European Society of Cardiology guidelines were enrolled in four treatment arms: implantable cardioverter‐defibrillator (ICD), cardiac resynchronization therapy defibrillator (CRT‐D), HF pacing (PA; Pacemaker and cardiac resynchronization pacemaker), and nondevice (ND) cohorts and followed for a minimum of 12 months. Since it is anticipated that some baseline patient characteristics and risk factors will differ significantly from those reported in predominantly Western populations, event rates will be compared to a propensity‐matched population from the MADIT RIT trial. Primary endpoints are composite rates of first appropriately treated ventricular arrhythmias (VA) or/and life‐threatening VA symptoms for the ICD and CRT‐D cohorts. For nondevice and PA cohorts, the primary outcome is all‐cause mortality. Conclusions The Heart Failure Indication and Sudden Cardiac Death Prevention Trial Japan is a large prospective multicenter registry with defined device treatment cohorts and will provide data for risk stratification for cardiovascular events in Japanese HF patients.


| Study design
The HINODE trial (NCT03185832) is a prospective, multicenter registry designed to collect data on VA, HF events, and mortality in Japanese patients with HF. Forty centers were nominated by the steering committee, and site selection was based on an independent process, which took trial experience and subject availability into account. This study was conducted in accordance with the relevant parts of the International Conference on Harmonization Guidelines for GCP and the ethical principles of the Declaration of Helsinki.
All centers were asked to follow the Japanese Ethics Guideline for Clinical Research on Human Subjects issued by the Ministry of Health, Labour and Welfare. Study centers were opened for enrollment after obtaining approval from the ethics committee (EC). reporting, especially on endpoint-relevant events, data monitoring at study sites was conducted regularly.

| Characterization of study cohorts
Patients who underwent de novo implantation of ICD, CRT-D, pacemaker (PM), or CRT pacing (CRT-P) device within 45 days were screened for eligibility and enrollment. Screening was performed and informed consent for the participation of the study was obtained only after device implantation, and the decision of device implantation was solely determined by the investigator. Based on 2016 European Society of Cardiology (ESC) Guidelines for the diagnosis and treatment of acute and chronic heart failure, 6 patients were included in the ICD cohort as long as they met the above entry criteria and had any ICD (transvenous or subcutaneous) implanted, as well as reduced LVEF (≤35%), NYHA class II or III functional status, and were prescribed guideline directed medical therapy 13 for ≥3 months.
Likewise, patients were included into the CRT-D cohort if they had any CRT-D device implanted as long as they met the above criteria and had reduced LVEF (≤35%) and LBBB with QRS >130 ms or non-LBBB with QRS ≥150 ms despite optimal medical therapy (OMT) or alternatively were in AF with NYHA Class III, and a QRS duration ≥130 ms despite optimal medical therapy (OMT). ECGs collected up to 45 days prior to device implant and post implant were used to verify CRT-D enrollment criteria. Patients with a history of previous PM/ ICD/ CRT-P/ CRT-D were excluded from ICD/ CRT-D cohorts. In the PA cohort, the enrollment criteria were: expectation of >40% right ventricular pacing, reduced LVEF ≤50%, QRS >90 ms, and any previous HF admission to the hospital. Enrolled nondevice (ND) patients met ICD or CRT-D implantation criteria but were not implanted with a cardiac device. To control for the effects of risk factors on patient outcome, [14][15][16][17][18]   or a 60s delayed therapy at a heart rate ≥170 bpm (arm C). 21 The UNTOUCHED study confirmed the effect of high rate cut-off programming on resulting in a low rate of appropriate and inappropriate therapy for patients with S-ICD treatment, 25 similar to the outcome of MADIT RIT arm B and C. High rate cut-off and delayed therapy programming were also associated in MADIT RIT with a 50% mortality reduction and no increase of syncope. The programming according to protocol was required as best standard, deviations have been documented and monitored throughout the study. Device patients were tracked with the remote monitoring system LATITUDE.

| Follow-up
Enrollment was open for 24 months, from July 2017 until July 2019.
All participants remained in the study for a minimum of 12 months up to study close, ie, when the last participant completed the 12month visit.
After enrollment, patient follow-up occurred every 6 months until the close-out visit. Every other 6-month visit could be replaced by phone follow-up if the patient was connected to a home monitoring system and no standard of care in-clinic visit was performed.
During every visit, device assessments were performed according to standard of care. Any serious or nonserious adverse events (AEs) were recorded during follow-up. The onset of life-threatening arrhythmias, death, and HF events were documented for all device cohorts. Echocardiographic parameters and laboratory data were collected during the study period.

| Primary endpoints
Primary endpoints of ICD and CRT-D cohorts were the composite rate of the first appropriately treated VA (by ATP or shock) or life-threatening symptoms associated with VA (defined as hemodynamic instability requiring treatment), whichever came first under MADIT Arm B (Intervention at high cut-off rate: VT ≥200 bpm) or C (Intervention delay ≥60 second before therapy ≥170 bpm) programming conditions. 19,20 In the ND and PA cohorts, the primary endpoint was defined as all-cause mortality.

| Secondary endpoints
In the ICD and CRT-D cohorts, the secondary endpoint was all-cause mortality. In addition, HF events were regarded as the secondary endpoint in all the device cohorts. In the nondevice cohort, the rate of SCD was defined as the secondary endpoint.

| Event adjudication
In order to ensure that data would be directly comparable with

| Statistical analysis
This study was designed to evaluate event rates in the Japanese patient population based on per treatment analysis and to compare TA B L E 2 Overview of programming options for defibrillator treatment them with historical data from landmark trials. The primary endpoint for each cohort was analyzed by a specific sample size calculation to reach 90% power. The sample size for each primary endpoint was estimated using exact binomial methods for comparison of a single proportion to a performance goal using a one-sided significance level of 2.5%. (Table 3). Expected performance and performance goals for primary endpoints were sourced from previous studies.
Performance measures were estimated based the following event rates, 5% of the ICD treated patients and 3% of the CRT-D treated patients will present an appropriately treated first VA-associated symptom (by ATP or shock) within 1 year (MADIT-RIT) 20 ; in the nondevice cohort, all-cause mortality within 1 year was estimated to be 10% (MADIT-II) 3 ; and in the PA cohort, all-cause mortality within 1 year was estimated to be 10% (CARE-HF). 5 Analysis of the primary and secondary endpoints on freedom from all-cause mortality and first VA will be performed for each cohort using the Kaplan-Meier curve for estimation of event free rates at 12/24-months.
Cohort event rates on VA, HF, and death will be descriptively compared with historical data. For best possible comparison between MADIT RIT and HINODE patient estimates will be further compared by subgroups within each cohort and to MADIT RIT outcomes using log-rank tests. Separate propensity score (PS) match analyses will be performed to compare event rates for HV cohorts to MADIT RIT. For the HV cohort, ICD patients with CRT-D indication (LVEF >35% and LBBB with QRS>130 ms or QRS>150 ms) will be removed, and for HV and ND cohorts' patients with chronic AF will be removed prior to PS match to parallel MADIT RIT exclu- Baseline characteristics included those with high relevance toward the endpoint and availability in both studies. QRS width for MADIT RIT -ICD indicated that patients are not available, and BMI will be excluded from the HV match because of burden on model fit.
Event rates will be re-estimated in matched HINODE and MADIT RIT cohorts and compared using a log-rank test stratified by the quintiles of the PS.
Additionally, the rate of SCD for the ICD or CRT-D indicated nondevice cohort will be compared with the rate of appropriately treated VAs or SCD in the ICD and CRT-D cohorts. Exploratory analyses for the pooled ICD and CRT-D cohorts versus nondevice cohorts will be performed using the propensity score matching method to compare the all-cause mortality rate and the composite rate of HF events between the cohorts.
Characteristics for ICD, CRT-D, and ND cohorts will be compared using Tukey's test for continuous variables and chi-squared test with Bonferroni correction for categorical variables. A significance level of 0.05 will be used.

| D ISCUSS I ON
The present study aims to evaluate the risk of SCD and HF events in Japanese patients using the risk stratification in the ESC guideline. 6 Although indication of primary defibrillator therapy was clearly described in the Japanese guideline published in 2011, 22 the proportion of patients receiving a defibrillator device as primary prevention remains lower than in western countries. 2,9 The CHART-2 study reported that only 30.4% of enrolled Japanese HF patients that satisfied the Class I indication and 6.6% of the patients that satisfied Class IIa indication received ICD therapy. 10 Previous registries reported lower rates of SCD in the Japanese population, 7,8 which may have impacted decision making for implant of primary defibrillator device. Nevertheless, rate of SCD satisfying Class I and IIa indications in the CHART-2 study was higher than previous registries. 7,8 Additionally, the Seattle Proportional Risk Model (a risk model calculated from western countries) also predicts future events in Japanese patients. 23 Thus, SCD event rates in Japan may be higher than assumed, and Japanese HF patients share similar risk factors with patients in western countries. To improve patient management, a large registry is warranted to clarify the incidence and therapeutic impact of implantable devices in Japanese HF patients.
The MADIT-RIT trial enrolled 68 Japanese primary prevention patients with ICD and CRT-D therapy. Analysis of this MADIT RIT subcohort presented similar rates of supraventricular, ventricular arrhythmias, appropriate ICD therapy, and mortality, but they had