Effect of Vericiguat vs Placebo on Quality of Life in Patients With Heart Failure and Preserved Ejection Fraction: The VITALITY-HFpEF Randomized Clinical Trial

Key Points

Question  Does vericiguat improve the physical limitation score (PLS) of the Kansas City Cardiomyopathy Questionnaire (KCCQ) in patients with heart failure and preserved ejection fraction (HFpEF)?

Findings  In this randomized clinical trial, 789 patients with HFpEF after recent HF decompensation were randomized to receive vericiguat, 15 mg/d; vericiguat, 10 mg/d; or placebo. After 24 weeks, the mean changes in the KCCQ PLS (range, 0-100; higher values indicate better functioning) were 5.5 points in the 15-mg/d vericiguat group, 6.5 points in the 10-mg/d vericiguat group, and 6.9 points in the placebo group; differences between either vericiguat dosage and placebo were not statistically significant.

Meaning  Vericiguat did not improve the KCCQ PLS at 24 weeks in patients with HFpEF.

Importance  Patients with heart failure and preserved ejection fraction (HFpEF) are at high risk of mortality, hospitalizations, and reduced functional capacity and quality of life.

Objective  To assess the efficacy of the oral soluble guanylate cyclase stimulator vericiguat on the physical limitation score (PLS) of the Kansas City Cardiomyopathy Questionnaire (KCCQ).

Design, Setting, and Participants  Phase 2b randomized, double-blind, placebo-controlled, multicenter trial of 789 patients with chronic HFpEF and left ventricular ejection fraction 45% or higher with New York Heart Association class II-III symptoms, within 6 months of a recent decompensation (HF hospitalization or intravenous diuretics for HF without hospitalization), and with elevated natriuretic peptides, enrolled at 167 sites in 21 countries from June 15, 2018, through March 27, 2019; follow-up was completed on November 4, 2019.

Interventions  Patients were randomized to receive vericiguat, up-titrated to 15-mg (n = 264) or 10-mg (n = 263) daily oral dosages, compared with placebo (n = 262) and randomized 1:1:1.

Main Outcomes and Measures  The primary outcome was change in the KCCQ PLS (range, 0-100; higher values indicate better functioning) after 24 weeks of treatment. The secondary outcome was 6-minute walking distance from baseline to 24 weeks.

Results  Among 789 randomized patients, the mean age was 72.7 (SD, 9.4) years; 385 (49%) were female; mean EF was 56%; and median N-terminal pro–brain natriuretic peptide level was 1403 pg/mL; 761 (96.5%) completed the trial. The baseline and 24-week KCCQ PLS means for the 15-mg/d vericiguat, 10-mg/d vericiguat, and placebo groups were 60.0 and 68.3, 57.3 and 69.0, and 59.0 and 67.1, respectively, and the least-squares mean changes were 5.5, 6.4, and 6.9, respectively. The least-squares mean difference in scores between the 15-mg/d vericiguat and placebo groups was −1.5 (95% CI, −5.5 to 2.5; P = .47) and between the 10-mg/d vericiguat and placebo groups was −0.5 (95% CI, −4.6 to 3.5; P = .80). The baseline and 24-week 6-minute walking distance mean scores in the 15-mg/d vericiguat, 10-mg/d vericiguat, and placebo groups were 295.0 m and 311.8m , 292.1 m and 318.3 m, and 295.8 m and 311.4 m, and the least-squares mean changes were 5.0 m, 8.7 m, and 10.5 m, respectively. The least-squares mean difference between the 15-mg/d vericiguat and placebo groups was −5.5 m (95% CI, −19.7 m to 8.8 m; P = .45) and between the 10-mg/d vericiguat and placebo groups was −1.8 m (95% CI, −16.2 m to 12.6 m; P = .81), respectively. The proportions of patients who experienced symptomatic hypotension were 6.4% in the 15-mg/d vericiguat group, 4.2% in the 10-mg/d vericiguat group, and 3.4% in the placebo group; those with syncope were 1.5%, 0.8%, and 0.4%, respectively.

Conclusions and Relevance  Among patients with HFpEF and recent decompensation, 24-week treatment with vericiguat at either 15-mg/d or 10-mg/d dosages compared with placebo did not improve the physical limitation score of the KCCQ.

Trial Registration  ClinicalTrials.gov Identifier: NCT03547583

Quiz Ref IDHeart failure with preserved ejection fraction (HFpEF) constitutes both a clinical challenge and a major public health issue. In 2017, it was estimated that 50% of all patients with HF were known to have HFpEF, and the relative proportion appears to be increasing. Moreover, HFpEF is associated with substantial morbidity and mortality.¹ Apart from relieving congestion with diuretics and modifying concomitant comorbidities, the absence of definitive therapy is especially problematic. The lack of therapies to address the impairment in functional capacity and quality of life imposed by HFpEF remains a major unmet need for patients and presents a significant clinical challenge.

Although the precise mechanisms underlying the causes and progression of this disorder remain unclear, it appears that a deficiency of cyclic guanosine monophosphate may be involved because it modulates the putative central and peripheral pathophysiologic pathways of HFpEF.²Quiz Ref ID Vericiguat, a novel oral soluble guanylate cyclase stimulator, directly generates cyclic guanosine monophosphate and restores the sensitivity of soluble guanylate cyclase to endogenous nitric oxide.³ In the phase 2 SOCRATES-PRESERVED randomized clinical trial, there was a suggestion of improvement in the physical limitation score (PLS) component of the Kansas City Cardiomyopathy Questionnaire (KCCQ) in patients with worsening HFpEF treated with vericiguat compared with placebo; however, the effect of vericiguat on an objective measure of exercise capacity (eg, 6-minute walking distance [6MWD]) was not evaluated.^4,5 The aims of the current study were to evaluate the efficacy of vericiguat on quality of life and exercise tolerance and to assess adverse events in patients with HFpEF.

The primary objective was to determine if vericiguat, up-titrated to either 10 mg/d or 15 m/d dosages, improved the KCCQ PLS when compared with placebo in patients with HFpEF after 24 weeks. The secondary outcome was change in 6MWD from baseline to 24 weeks.

VITALITY-HFpEF was a multicenter, randomized, double-blind, placebo-controlled, phase 2b trial that enrolled patients with HFpEF after a recent HF decompensation. The study design has been previously described (the trial protocol and statistical analysis plan are available in Supplement 1 and Supplement 2).⁶ Institutional review board or ethics committee approval was obtained at each study site. All patients provided written informed consent.

The steering committee designed the trial with national leaders from participating countries and regions and oversaw operations in collaboration with the Canadian VIGOUR Centre and the trial cosponsors, Bayer and Merck. An independent data and safety monitoring committee evaluated adverse events.

Quiz Ref IDThe trial included patients 45 years of age or older with HFpEF and New York Heart Association (NYHA) class II to III symptoms who were receiving standard-of-care therapy. The inclusion of race/ethnicity data was aligned with US Food and Drug Administration (FDA) guidance. Participants made the determination of race/ethnicity based on fixed categories. Patients had a history of chronic HF and had an HF decompensation within 6 months prior to randomization, defined as hospitalization for HF or intravenous diuretic treatment for HF without hospitalization. In addition, they were required to have elevated natriuretic peptide levels (N-terminal pro–brain natriuretic peptide [NT-proBNP] levels ≥300 pg/mL or BNP levels ≥100 pg/mL in sinus rhythm; or NT-proBNP levels ≥600 pg/mL or BNP levels ≥200 pg/mL in atrial fibrillation) within 30 days before randomization. Echocardiographic evidence of left ventricular EF of 45% or greater and either left ventricular hypertrophy or left atrial enlargement, acquired at participating sites within 12 months of randomization, was required.

Patients were excluded if they were clinically unstable (as defined by need for any intravenous treatment within 24 hours prior to randomization), had a systolic blood pressure of 160 mm Hg or greater or less than 110 mm Hg, had a diastolic blood pressure of less than 40 mm Hg, had symptomatic hypotension, or had a resting heart rate of less than 50/min or 100/min or greater. Other exclusion criteria were use of intravenous inotropes any time between hospitalization and randomization, previous diagnosis of reduced left ventricular EF less than 40%, concurrent use of nitrates, phosphodiesterase type 5 inhibitors, or other soluble guanylate cyclase stimulators, acute coronary syndrome within 60 days prior to randomization, estimated glomerular filtration rate less than 30 mL/min/1.73 m², and specific HF etiologies including hypertrophic obstructive cardiomyopathy, pericardial disease, acute myocarditis, amyloidosis, sarcoidosis, or primary valvular heart disease requiring surgery or intervention.

Patients who met eligibility criteria were randomized in a 1:1:1 ratio to receive vericiguat, 15 mg/d; vericiguat, 10 mg/d; or placebo. Randomization was implemented using a computer-generated algorithm with stratification according to region and heart rhythm using a block size of 6. A maximal dosage of either 15 mg or 10 mg once daily was targeted in the 2 vericiguat treatment groups. Randomization occurred within 6 months after a HF decompensation event and within 4 weeks of screening (visit 1). In both active treatment groups, vericiguat, 2.5 mg once daily, was started at randomization. Vericiguat or placebo were up-titrated or sham-titrated at subsequent visits over the next 6 weeks. Dose modification was predicated on a mean sitting systolic blood pressure and the absence of symptoms indicative of hypotension with the intention to reach and maintain the target study dosage on up-titration. If the target dosage was not attained by the week 6 visit (42 ± 3 days after randomization) or dosing up-titration was temporarily interrupted or reduced, up-titration was resumed at subsequent visits at the discretion of investigators. The planned treatment duration was 24 weeks, followed by adverse events follow-up at 28 weeks after randomization.

The primary end point of the study was the change from baseline to week 24 in physical limitations as measured by the KCCQ PLS. The KCCQ is a heart failure–specific patient-reported measure of health status. The PLS along with the total symptom score are components of the clinical summary score, each of which qualify as clinical outcome assessments in HF according to the FDA. The KCCQ PLS is derived from patient self-report of limitations due to HF in performing 6 common, graded, everyday physical activities; it is scored on a 0- to 100-point scale, and higher scores indicate better physical functioning. The choice of the KCCQ PLS, which provides a measure of functional capacity in activities of daily life limited by HFpEF symptoms, was based on a prior phase 2 study of vericiguat.^4,7 The secondary end point was the change from baseline in 6MWD at 24 weeks. For the primary end point, data were collected using electronic patient-reported outcome capture. An independent data and safety monitoring board oversaw patient safety and adverse events during the trial.

Clinical event data were collected to explore the efficacy of study drug on clinical outcomes. These included all cardiovascular and noncardiovascular deaths, cardiovascular hospitalizations including HF and non-HF hospitalizations and non-HF cardiovascular hospitalizations (ie, myocardial infarction and stroke). All of these events were adjudicated by an independent clinical events committee whose members were unaware of treatment assignment.

The KCCQ PLS was calculated using an algorithm applying equal weighting to each of the items and transformation to a 0- to 100-point scale. A sample size of 735 patients (245 patients per study group) provided 80% power for rejecting either of the primary hypotheses (15-mg/d vericiguat vs placebo and 10-mg/d vericiguat vs placebo), assuming a mean difference from placebo in KCCQ PLS of 5 points for both dosage groups and a common standard deviation of 21 points, using a 2-sided type I error of .05.^8,9 The analysis of the primary end point was performed using a mixed-model repeated-measures approach to generate a least-squares mean, including all postbaseline assessments. Covariates adjusted for included baseline KCCQ PLS values, treatment group, region, heart rhythm, and study visit as fixed effects, and the interaction effect between study visit and treatment group. The analysis used 2 types of imputation. For participants who died during the study, their KCCQ PLS was imputed with the worst-case outcome (ie, score of 0) from the time of death onward. Missing data for all other reasons were imputed based on a pattern-mixture model with control-based pattern imputation.^10,11 A sensitivity analysis of the primary end point was performed without imputation following the missing-at-random assumption. The numbers of patients with either KCCQ PLS improvement or worsening from baseline by 5 or more points at 24 weeks were summarized for each treatment group and compared using a χ² test.

The primary analysis was performed for all randomized patients who received at least 1 dose of study treatment and had at least 1 observed KCCQ PLS assessment at both baseline and postbaseline (excluding adverse event follow-up). The data set used to analyze adverse events included all patients who received at least 1 dose of vericiguat or placebo.

The study used multiplicity control across the primary and secondary hypotheses using an overall 2-sided type I error level of .05. The 2 primary hypotheses were tested with the Bonferroni procedure, each at the 2-sided .025 level. Only the primary and secondary analyses were prespecified in the testing procedure and thus were α adjusted. All other prespecified analyses were considered exploratory. Details of the multiplicity control are included in the statistical analysis plan (Supplement 2). All analyses were performed using SAS version 9.4 (SAS Institute Inc).

The original statistical analysis plan was formulated on December 3, 2018, and amended on November 21, 2019, prior to closure of the clinical database. The last patient was enrolled on April 23, 2019, and the last patient follow-up safety visit was November 4, 2019; unblinding of the data followed on December 4, 2019.

Between June 15, 2018, and April 23, 2019, 979 patients were screened for inclusion in 21 countries, and 789 patients were randomized at 168 sites (Figure 1). Of these, 264 were randomly assigned to receive vericiguat, 15 mg/d; 263 to receive vericiguat, 10 mg/d; and 262 to receive placebo.

The baseline characteristics were well balanced across the 3 study groups and were consistent with contemporary epidemiological studies of HFpEF (Table 1). The mean age was 72.7 years, and 48.8% were female. There were frequent coexisting comorbidities known to be common in HFpEF, including hypertension (92%), coronary artery disease (46%), diabetes (45%), and atrial fibrillation (35%). All patients had symptomatic HF; 59% had NYHA class II symptoms and the rest had NYHA class III symptoms. The index HF event was discharge from a HF hospitalization in 65.8% of patients, and treatment with intravenous diuretics without hospitalization accounted for 30.4%. The median elapsed time from HF event to randomization was 28 days. Blood pressure and heart rate were well controlled at baseline (mean systolic blood pressure, 129 mm Hg; mean heart rate, 70/min). Patients had preserved left ventricular EF (mean baseline EF, 56%), 53.9% had evidence of stage 3 chronic kidney disease (mean estimated glomerular filtration rate, 59.5 mL/min/1.73 m²), and the median NT-proBNP level was high (1403 pg/mL). At baseline, the site-reported echocardiographic findings showed that 84% of patients had left ventricular hypertrophy, 94% had left atrial enlargement, and 78% had both abnormalities. These characteristics, typical for HFpEF, were balanced across the 3 treatment groups.

Baseline medical therapy included loop diuretics (86%), β-blockers (86%), angiotensin-converting enzyme inhibitors (41%), angiotensin receptor blockers (39%), mineralocorticoid antagonists (45%), and calcium channel blockers (39%). Their use was balanced across the treatment groups, except for lower use of calcium channel blockers in the 10-mg/d vericiguat group. Sodium-glucose cotransporter 2 inhibitors were rarely used. Five patients had implantable defibrillators, 4 in the 15-mg/d vericiguat group and 1 in the placebo group.

The mean duration of study treatment was 149.5 (SD, 45.2) days. Eighty-four percent of patients (n = 663) completed at least 18 weeks (−3 days) of treatment, and 572 patients (72.6%) completed at least 24 weeks (−3 days) of treatment. The proportion of patients who reached the target dosage by week 18 was 81.1% in the 15-mg/d vericiguat group and 88.9% in the 10-mg/d vericiguat group, and the proportion of patients who reached the target dosage by week 24 was 83.0% in the 15-mg/d vericiguat group and 89.3% in the 10-mg/d vericiguat group. Mean treatment adherence (on a continuous scale, with 100 indicating that study drug was taken as expected, lower values indicating less drug taken, and higher values indicating more drug taken) was 99.3 (SD, 8.5) and 97.7 (SD, 12.5) for the 15- and 10-mg/d vericiguat groups, respectively, and 98.2 (SD, 9.7) for the placebo group.

The distribution of KCCQ PLSs at baseline and 24 weeks according to the 3 treatment groups is presented in eFigure 1 in Supplement 3. The pattern of KCCQ PLS missing data is presented in eTable 1 in Supplement 3. Figure 2 shows the primary results for the KCCQ PLS outcome. The baseline and 24-week KCCQ PLS means for the 15-mg/d vericiguat, 10-mg/d vericiguat, and placebo groups were 60.0 and 68.3, 57.3 and 69.0, and 59.0 and 67.1, respectively, and the least-squares mean changes were 5.5, 6.4, and 6.9, respectively. Quiz Ref IDThere were no significant differences in KCCQ PLS changes from baseline to 24 weeks between the placebo group and either vericiguat treatment group. This was also the case for the different domains of the KCCQ (eTable 2 in Supplement 3). The least-squares estimate for the change from baseline to 24 weeks in the 15-mg/d vericiguat group vs the placebo group was −1.5 points (95% CI, −5.5 to 2.5 points; P = .47) and in the 10-mg/d vericiguat group vs the placebo group was −0.5 points (95% CI, −4.6 to 3.5 points; P = .80)(Figure 2). Sensitivity analyses that did not impute missing data yielded similar results (eTable 3 in Supplement 3). Similarly, a responder analysis of 5-point improvement or worsening on the KCCQ PLS showed no statistically significant differences between the placebo group and the vericiguat treatment groups.

In eFigure 2A in Supplement 3, the primary outcome for the 3 treatment groups is shown according to relevant subgroups. There was no heterogeneity in the KCCQ PLS according to these groupings, including baseline NYHA class or KCCQ score. In eFigures 2A and 2B in Supplement 3, the absolute differences in mean KCCQ PLS changes from baseline to week 24 between treatment groups are shown.

The distribution of 6MWD measurements at baseline and 24 weeks is shown in eFigure 3 in Supplement 3. The 6MWD outcomes (Figure 3) were comparable between the 3 treatment groups. The baseline and 24-week 6MWD means for the 15-mg/d vericiguat, 10-mg/d vericiguat, and placebo groups were 295.0 m and 311.8 m, 292.1 m and 318.3 m, and 295.8 m and 311.4 m, respectively, and the least-squares mean changes were 5.0 m, 8.7 m, and 10.5 m, respectively. Quiz Ref IDThe least-squares mean difference between the 15-mg/d vericiguat and placebo groups was −5.5 m (95% CI, −19.7 m to 8.8 m; P = .45) and between the 10-mg/d vericiguat and placebo groups was −1.8 m (95% CI, −16.2 m to 12.6 m; P = .81). Sensitivity analyses without imputing missing data yielded similar results (eTable 4 in Supplement 3).

Table 2 shows the summary of patients with adverse events and clinical outcomes. The percentage of patients with adverse events was 65.2% in the 15-mg/d vericiguat group, 62.2% in the 10-mg/d vericiguat group, and 65.6% in the placebo group. Syncope occurred in 4 patients in the 15-mg/d vericiguat group, 2 patients in the 10-mg/d vericiguat group, and 1 patient in the placebo group. Symptomatic hypotension occurred in 6.4% of patients in the 15-mg/d vericiguat group, 4.2% in the 10-mg/d vericiguat group, and 3.4% in the placebo group. As shown in eTable 5A in Supplement 3, mean changes in systolic blood pressure at 24 weeks were −3.1 (SD, 14.94) mm Hg in the 15-mg/d vericiguat group, −3.8 (SD, 4.01) mm Hg in the 10-mg/d vericiguat group, and −1.2 (SD, 15.42) mm Hg in the placebo group. There were no significant changes between the groups in diastolic blood pressure (eTable 5B) or heart rate (eTable 5C) at 24 weeks. Treatment-emergent serious adverse events of anemia were reported in 0.8% of patients in the 15-mg/d vericiguat group, 1.1% in the 10-mg/d vericiguat group, and 0.4% in the placebo group. Mean baseline hemoglobin levels were 12.9 (SD, 1.7) g/dL in the 15-mg/d vericiguat group, 13.1 (SD, 1.8) g/dL in the 10-mg/d vericiguat group, and 13.0 (SD, 1.9) g/dL in the placebo group; mean changes at 24 weeks were −0.40 (SD, 1.14) g/dL, −0.27 (SD, 1.13) g/dL, and −0.02 (SD, 1.13) g/dL, respectively.

The overall mortality rate was 4.1% (n = 32) and included 10 patients (3.8%) in the 15-mg/d vericiguat group, 15 (5.7%) in the 10-mg/d vericiguat group, and 7 (2.7%) in the placebo group. There were 8 cardiovascular deaths (3.0%) in the 15-mg/d vericiguat group and 12 (4.6%) in the 10-mg/d vericiguat group compared with 4 (1.5%) in the placebo group, evenly distributed throughout the study follow-up period.

In this phase 2b study of vericiguat in patients with HFpEF who experienced a recent worsening HF event, vericiguat (15 mg/d and 10 mg/d) did not significantly improve the primary outcome of KCCQ PLS or the secondary end point of 6MWD. The rationale for this study was based on post hoc analysis of the KCCQ data from SOCRATES-PRESERVED, in which an exploratory analysis showed benefit in KCCQ PLS in 68 patients receiving 10 mg/d of vericiguat. Several design modifications were introduced in the trial to increase the likelihood of detecting a true difference. These included a larger sample size to provide adequate power to detect a clinically meaningful change in KCCQ PLS, a longer duration of treatment (24 weeks vs 12 weeks), and inclusion of a higher target dosage.

There are several possibilities for why this trial failed to confirm the hypothesis-generating observations of the prior study. The SOCRATES-PRESERVED findings may have resulted from the play of chance. That study population had, on average, lower baseline KCCQ scores compared with the current population (mean KCCQ PLS, 54.6 vs 58.8), consistent with the lower proportion of NYHA class II patients and higher proportion of NYHA class III patients.⁵ The placebo group in the prior investigation showed a more modest score improvement of 4.8 points over the 12-week observation period, whereas the improvement in KCCQ PLS of 7.3 points in the placebo group in the current study exceeded the 5-point change considered to be a minimally important difference.⁶ It appears that the natural history of KCCQ improvement following recent HFpEF hospitalization remains poorly understood and warrants future study.

Other possibilities include inclusion of a less symptomatic HFpEF population with a higher KCCQ PLS in the current study, or the greater blood pressure–lowering effects of the 15-mg dosage of vericiguat tested. Enrolling a less ill population with stable quality-of-life measures established over months might have yielded a different result. However, these findings, in contrast to those observed in the VICTORIA trial of patients with HF with reduced EF,¹² may also reflect that nitric oxide is not a key operative factor in the progression of HFpEF; indeed, lack of benefit with oral nitrates, inhaled nitrites, and phosphodiesterase inhibitors suggests that direct soluble guanylate cyclase stimulation with vericiguat was ineffective.^13-15

This study has several limitations. First, it is possible that not all enrolled patients in this trial had HFpEF. Given that the eligibility criteria required a clinical diagnosis, a history of HF hospitalization or need for intravenous diuretics in the outpatient setting, elevated natriuretic peptide levels, and echocardiographic confirmation of structural cardiac atrial or ventricular remodeling, it is likely that the population enrolled had the intended clinical syndrome. However, phenotypical heterogeneity and its potential effect on outcomes cannot be excluded.

Second, although there were nominally more cardiovascular deaths in the vericiguat groups compared with placebo, the numbers of events are too small to make any definitive conclusions. Because of the imputation of data for these patients, any improvement evident in KCCQ PLS would have been substantially mitigated. Third, the optimal method to account for missing data in analyses of patient-reported outcomes is not clear. In the current study, an a priori approach was planned to impute the worst-case scenario (ie, score of 0) in cases of death. Such a strict approach may overinflate the effect of any chance imbalance in missing data.

Among patients with HFpEF and recent decompensation, 24-week treatment with vericiguat at either 10-mg/d or 15-mg/d dosages compared with placebo did not improve the physical limitation score of the KCCQ.

Corresponding Author: Paul W. Armstrong, MD, 4-120 Katz Group Centre for Pharmacy and Health Research, University of Alberta, Edmonton, AB T6G 2E1, Canada (paul.armstrong@ualberta.ca).

Accepted for Publication: August 6, 2020.

Correction: This article was corrected for errors in figure labeling on February 2, 2021.

Author Contributions: Dr Armstrong had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Armstrong, Lam, Anstrom, Ezekowitz, Hernandez, O’Connor, Pieske, Ponikowski, Shah, Solomon, Voors, Vlajnic, Carvalho, Bamber, Blaustein, Roessig, Butler.

Acquisition, analysis, or interpretation of data: Armstrong, Lam, Anstrom, Ezekowitz, Hernandez, O’Connor, Pieske, Ponikowski, Voors, She, Vlajnic, Carvalho, Bamber, Blaustein, Roessig, Butler.

Drafting of the manuscript: Armstrong, Lam, Anstrom, Ezekowitz, She, Vlajnic, Roessig, Butler.

Critical revision of the manuscript for important intellectual content: Armstrong, Lam, Anstrom, Ezekowitz, Hernandez, O’Connor, Pieske, Ponikowski, Shah, Solomon, Voors, Vlajnic, Carvalho, Bamber, Blaustein, Roessig, Butler.

Statistical analysis: Armstrong, Anstrom, She, Vlajnic, Bamber.

Obtained funding: Armstrong, Blaustein.

Administrative, technical, or material support: Carvalho, Bamber, Butler.

Supervision: Armstrong, Anstrom, O’Connor, Pieske, Ponikowski, Bamber, Blaustein, Roessig, Butler.

Conflict of Interest Disclosures: Dr Armstrong reported receiving personal fees from Merck and Bayer during the conduct of the study as well as grants from Sanofi-Aventis Recherche & Developpement, Boehringer Ingelheim, and CSL Limited and personal fees from AstraZeneca and Novartis. Dr Lam reported receiving personal fees from Bayer during the conduct of the study as well as grants from Boston Scientific, Bayer, Roche Diagnostics, AstraZeneca, Medtronic, and Vifor Pharma and personal fees from Abbott Diagnostics, Amgen, Applied Therapeutics, AstraZeneca, Bayer, Biofourmis, Boehringer Ingelheim, Boston Scientific, Corvia Medical, Cytokinetics, Darma, Eko.ai, JanaCare, Janssen Research & Development, Medtronic, Menarini Group, Merck, MyoKardia, Novartis, Novo Nordisk, Radcliffe Group, Roche Diagnostics, Stealth BioTherapeutics, The Corpus, Vifor Pharma, and WebMD. In addition, Dr Lam has a patent pending on a method for diagnosis and prognosis of chronic heart failure (PCT/SG2016/050217) and a patent issued for a clinical workflow that recognizes and analyses 2D and Doppler echocardiogram images for automated cardiac measurements and diagnosis, prediction, and prognosis of heart disease (16/216,929). Dr Lam is also cofounder and nonexecutive director of EKo.ai Pte Ltd. Dr Ezekowitz reported receiving personal fees from Bayer and Merck during the conduct of the study as well as grants and personal fees from American Regent, Novartis, AstraZeneca, Boehringer Ingelheim, Amgen, and Cytokinetics (additional disclosures available online at https://thecvc.ca). Dr Hernandez reported receiving personal fees from Bayer and Merck during the conduct of the study as well as grants and personal fees from AstraZeneca and Novartis and personal fees from Amgen and Cytokinetics. Dr O’Connor reported receiving consulting for Bayer, Bristol Myers Squibb Foundation, and Dey. Dr Pieske reported receiving personal fees from Merck and Bayer during the conduct of the study as well as personal fees from Novartis, Servier, Medscape, and Bristol Myers Squibb. Dr Ponikowski reported receiving personal fees from Merck during the conduct of the study as well as receiving personal fees from and participating in clinical trials for Amgen, Boehringer Ingelheim, Servier, AstraZeneca, RenalGuardSolution, and Cibiem; receiving grants and personal fees from and participating in clinical trials for Vifor Pharma; and receiving personal fees from Pfizer and Respicardia. Dr Shah reported receiving personal fees from Bayer during the conduct of the study as well as grants and personal fees from Actelion, AstraZeneca, Novartis, and Pfizer; grants from Corvia; and personal fees from Cyclerion, Amgen, Boehringer Ingelheim, Cardiora, Cytokinetics, MyoKardia, Merck, Shifamed, Eisai, Ionis, Novo Nordisk, Sanofi, and Tenax. Dr Solomon reported receiving personal fees from Bayer during the conduct of the study as well as grants from Alnylam, Amgen, AstraZeneca, Bellerophon, Bayer, Bristol Myers Squibb, Celladon, Cytokinetics, Eidos, Gilead, GlaxoSmithKline, Ionis, Lone Star Heart, Mesoblast, MyoKardia, Neurotronik, the National Institutes of Health/National Heart, Lung, and Blood Institute, Novartis, Respicardia, Sanofi Pasteur, and Theracos and personal fees from Akros, Alnylam, Amgen, Arena, AstraZeneca, Bayer, Bristol Myers Squibb, Cardior, Cardurion, Corvia, Cytokinetics, Daiichi-Sankyo, Gilead, GlaxoSmithKline, Ironwood, Merck, MyoKardia, Novartis, Roche, Takeda, Theracos, Quantum Genetics, Cardurion, AoBiome, Janssen, Cardiac Dimensions, Sanofi Pasteur, Tenaya, Dinaqor, Tremeau, CellProThera, and Moderna. Dr Voors reported receiving personal fees from Merck and Bayer during the conduct of the study as well as personal fees from AstraZeneca, Cytokinetics, Novartis, and MyoKardia and grants and personal fees from Boehringer Ingelheim, Novo Nordisk, and Roche Diagnostics. Ms Vlajnic is an employee of Bayer. Dr Carvalho is an employee of Bayer. Mr Bamber is an employee of Bayer. Dr Blaustein reported receiving personal fees from Merck during the conduct of the study and is an employee of Merck. Dr Roessig is an employee of Bayer. Dr Butler reported receiving personal fees from Bayer and Merck during the conduct of the study as well as personal fees from Abbott, Adrenomed, Amgen, Applied Therapeutics, Array, AstraZeneca, BerlinCures, Boehringer Ingelheim, Cardior, CVRx, Foundry, G3 Pharma, Imbria, Impulse Dynamics, Innolife, Janssen, LivaNova, Luitpold, Medtronic, Novartis, Novo Nordisk, Relypsa, Roche, Sanofi, Sequana Medical, V-Wave Limited, and Vifor. No other disclosures were reported.

Funding/Support: Funding for this research was provided by Bayer and Merck Sharp & Dohme Corp, a subsidiary of Merck & Co Inc.

Role of the Funder/Sponsor: The sponsors, in collaboration with the steering committee, participated in the design and conduct of the study and the collection, management, analysis, and interpretation of the data. Preparation, review, and approval of the manuscript and the decision to submit the manuscript for publication were undertaken by the steering committee. The sponsors did not have the right to veto publication or to control the decision regarding the choice of journal for submission.

Group Information: The members of the VITALITY-HFpEF Study Group are listed in eAppendix 1 in Supplement 3.

Data Sharing Statement: See Supplement 4.

Additional Contributions: Elizabeth E. S. Cook, BA, an employee of the Duke Clinical Research Institute, and Cecilia Freitas, MD, MSc, PhD, an employee of Bayer, provided editorial assistance. No compensation outside of their regular salaries was received.