Author + information
- Received April 20, 2020
- Revision received May 28, 2020
- Accepted May 28, 2020
- Published online July 27, 2020.
- Jonathan W. Cunningham, MDa,
- Brian L. Claggett, PhDa,
- Eileen O’Meara, MDb,
- Margaret F. Prescott, PhDc,
- Marc A. Pfeffer, MD, PhDa,
- Sanjiv J. Shah, MDd,
- Margaret M. Redfield, MDe,
- Faiez Zannad, MD, PhDf,
- Lu-May Chiang, PhDc,
- Adel R. Rizkala, PharmDc,
- Victor C. Shi, MDc,
- Martin P. Lefkowitz, MDc,
- Jean Rouleau, MDb,
- John J.V. McMurray, MDg,
- Scott D. Solomon, MDa,∗ (, )@scottdsolomon and
- Michael R. Zile, MDh,i,∗∗ ()
- aCardiovascular Division, Brigham and Women’s Hospital, Boston, Massachusetts
- bMontreal Heart Institute, Université de Montréal, Montreal, Quebec, Canada
- cNovartis, East Hanover, New Jersey
- dNorthwestern University Feinberg School of Medicine, Chicago, Illinois
- eMayo Clinic, Rochester, Minnesota
- fCentre d'Investigations Cliniques-Plurithématique 1433, and Institut National de la Santé et de la Recherche Médicale U1116, Centre Hospitalier Regional Universitaire, French Clinical Research Infrastructure Network, Investigation Network Initiative Cardiovascular and Renal Clinical Trialists, Nancy, France
- gBritish Heart Foundation Cardiovascular Research Centre, University of Glasgow, Glasgow, United Kingdom
- hRalph H. Johnson Department of Veterans Affairs Medical Center, Charleston, South Carolina
- iDivision of Cardiology, Medical University of South Carolina, Charleston, South Carolina
- ↵∗Address for correspondence:
Dr. Scott D. Solomon, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115.
- ↵∗∗Dr. Michael R. Zile, Division of Cardiology, Medical University of South Carolina, Thurmond/Gazes Building, Room 323, 30 Courtenay Drive, Charleston, South Carolina 29425.
Background Myocardial fibrosis may contribute to the pathophysiology of heart failure with preserved ejection fraction. Given the biochemical targets of sacubitril/valsartan, this study hypothesized that circulating biomarkers reflecting the mechanisms that determine extracellular matrix homeostasis are altered by sacubitril/valsartan compared with valsartan alone.
Objectives This study investigated the effects of sacubitril/valsartan on biomarkers of extracellular matrix homeostasis and the association between biomarkers and the primary endpoint (total heart failure hospitalizations and cardiovascular death).
Methods N-terminal propeptide of collagen I and III, tissue inhibitor of matrix metalloproteinase 1, carboxyl-terminal telopeptide of collagen type I, and soluble ST2 were measured at baseline (n = 1,135) and 16 (n = 1,113) and 48 weeks (n = 1,016) after randomization. The effects of sacubitril/valsartan on these biomarkers were compared with those of valsartan alone. Baseline biomarker values and changes from baseline to 16 weeks were related to primary endpoint.
Results At baseline, all 5 biomarkers were higher than published referent control values. Sixteen weeks after randomization, sacubitril/valsartan decreased tissue inhibitor of matrix metalloproteinase 1 by 8% (95% confidence interval [CI]: 6% to 10%; p < 0.001), soluble ST2 by 4% (95% CI: 1% to 7%; p = 0.002), and N-terminal propeptide of collagen III by 3% (95% CI: 0% to 6%; p = 0.04) and increased carboxyl-terminal telopeptide of collagen type I by 4% (95% CI: 1% to 8%; p = 0.02) compared with valsartan alone, consistently in men and women and patients with left ventricular ejection fraction above or below the median of 57%. Higher levels of tissue inhibitor of matrix metalloproteinase 1 and soluble ST2 at baseline and increases in these markers at 16 weeks were associated with higher primary endpoint event rates.
Conclusions Biomarkers reflecting extracellular matrix homeostasis are elevated in heart failure with preserved ejection fraction, favorably altered by sacubitril/valsartan, and have important prognostic value. (Prospective Comparison of ARNI With ARB Global Outcomes in HF With Preserved Ejection Fraction [PARAGON-HF]; NCT01920711)
Changes in the architecture, composition, and distribution of the myocardial extracellular matrix (ECM) play a central role in myocardial remodeling and left ventricular diastolic dysfunction (1–3). A change in ECM homeostatic control is postulated to occur during the development and progression of heart failure (HF) in patients with heart failure with reduced ejection fraction (HFrEF) and in those with heart failure with preserved ejection fraction (HFpEF). The changes in ECM homeostasis that lead to structural remodeling include changes in synthesis, processing, degradation, and turnover of proteins such as fibrillar collagen (4).
Although these changes occur within the myocardium itself, evidence of their occurrence can also be detected by examining circulating biomarkers (5–9). For example, maturation of newly synthesized collagen requires removal of the N-terminal and C-terminal propeptides; the concentration of these propeptides (such as N-terminal propeptide of collagen I [PINP] and N-terminal propeptide of collagen III [PIIINP]) reflect collagen synthesis rate. Activation of the principal collagen-producing myocardial cell, the fibroblast, is enhanced by soluble ST2 (sST2). Fibrillar collagen turnover and degradation result in liberation of collagen I telopeptides (CITPs). Collagen degradation by matrix metalloproteinases (MMPs) is inhibited by tissue inhibitor of matrix metalloproteinase (TIMP)-1.
Sacubitril/valsartan increases levels of vasoactive peptides including natriuretic peptides that may inhibit fibrosis (10). We therefore hypothesized that it would favorably alter these circulating biomarkers, as demonstrated in patients with HFrEF (11). The previous phase II PARAMOUNT (Prospective Comparison of ARNI With ARB on Management of Heart Failure With Preserved Ejection Fraction) study in HFpEF lacked sufficient sample size to examine the effects of sacubitril/valsartan on these biomarkers. The PARAGON-HF (Prospective Comparison of ARNI With ARB Global Outcomes in HF With Preserved Ejection Fraction) trial provided this unique opportunity (12).
The following hypotheses were investigated in this analysis of PARAGON-HF: 1) circulating biomarkers that reflect the determinants of ECM homeostasis are abnormal in patients with HFpEF; 2) baseline values of these biomarkers and change from baseline predict HF events independently of clinical covariates and other biomarkers with established prognostic value (natriuretic peptides and troponin); and 3) treatment with sacubitril/valsartan, compared with valsartan alone, favorably changes the levels of these biomarkers.
Study design and procedures
The design and primary results of the PARAGON-HF have been previously described (12,13). PARAGON-HF was a multicenter, double-blind clinical trial that randomized patients with chronic HFpEF in a 1:1 fashion to sacubitril/valsartan or valsartan alone. Eligibility criteria included signs and symptoms of HF, New York Heart Association functional classes II to IV, left ventricular ejection fraction (LVEF) ≥45%, age ≥50 years, diuretic therapy, evidence of structural heart disease, and elevated level of natriuretic peptides (with lower thresholds for patients with recent hospitalization for HF and higher thresholds in the presence of atrial fibrillation). Only patients who tolerated both study drugs during sequential run-in periods were eligible for randomization.
The pre-specified primary endpoint was a composite of total (first and recurrent) hospitalizations for HF and cardiovascular (CV) death. Key secondary endpoints included components of the primary endpoint. A blinded clinical events committee at Brigham and Women’s Hospital (Boston, Massachusetts) adjudicated these endpoints.
Patients recruited at selected sites were invited to participate in the collagen biomarker substudy. Most sites enrolling in this substudy were in North America or Europe; few centers in Asia and Latin America participated for logistical reasons. Five biomarkers were measured in 1,135 patients at the pre-specified baseline time point, after screening but prior to the start of the valsartan run-in period. Of these participants, 1,113 had a second measurement 16 weeks after randomization, and 1,016 had a third measurement 48 weeks after randomization. Analyses of patients at 16 and 48 weeks after randomization included only patients with available data at earlier time points as well.
Five biomarkers of ECM homeostasis—sST2, PINP, PIIINP, TIMP-1, and CITP—were prospectively selected for measurement by the trial’s executive committee and sponsor. Concentrations of these biomarkers were measured centrally at Clinical Reference Laboratory (Lenexa, Kansas) from frozen samples collected at individual sites. Samples were stored at −80°C. TIMP-1 was assayed in plasma and sST2, PINP, PIIINP, and CITP in serum. Commercial enzyme-linked immunosorbent assays for TIMP-1 (Quantikine, R&D Systems, Minneapolis, Minnesota) and sST2 (Presage, Critical Diagnostics, San Diego, California), and radioimmunoassays for PINP, PIIINP, and CITP (Orion Diagnostica, Espoo, Finland) were used. The coefficients of variance, lower limit of detection or lower limit of quantitation, and measuring range for each biomarker were previously published (11).
Baseline biomarker levels were described using median (interquartile range) and compared qualitatively with referent control values (5,7,14–17) as previously described (11). Baseline characteristics of the included patients were described using proportions for categorical variables, mean ± SD for normally distributed continuous variables, and median and interquartile range for skewed continuous variables and were compared using Pearson chi-square test, Student’s t-test, and Wilcoxon rank sum test, respectively. Changes in biomarker levels 16 and 48 weeks after randomization, compared to baseline, were described using geometric mean with 95% confidence intervals (CIs) for each treatment group; only patients with available data at all 3 time points were included in assessments of 48-week change. Biomarker changes from baseline were compared between treatment groups using linear regression with log-transformed biomarker values as the outcome and adjusting for log-transformed baseline biomarker level.
The relationships among log-transformed and baseline levels and incident rates of subsequent clinical outcomes were modeled using restricted cubic splines with 3 knots; log-transformed effects were reported per SD. These biomarkers were evaluated individually and together in recurrent events regression models using the method of Lin et al. (18) adjusted for 23 clinical covariates: age; sex; region; history of diabetes; hypertension; history of stroke; history of myocardial infarction; ischemic cause of HF; New York Heart Association functional class; prior HF hospitalization; angiotensin-converting enzyme inhibitor and/or angiotensin receptor blocker; mineralocorticoid antagonist; diuretic; beta-blocker; atrial fibrillation at screening; body mass index; LVEF; systolic blood pressure; diastolic blood pressure; serum potassium; estimated glomerular filtration rate; high-sensitivity troponin T (hsTnT); and N-terminal pro–B-type natriuretic peptide (NT-proBNP). The association between log-transformed and biomarker changes from baseline to 16 weeks and subsequent clinical outcomes were assessed using landmark analysis, both with and without adjustment for age, sex, New York Heart Association functional class, body mass index, estimated glomerular filtration rate, NT-proBNP, and hsTNT. Effect modification by randomized therapy was assessed via interaction terms. Similar time-to–first event Cox regression models were used for the endpoint of CV death.
All participants in PARAGON-HF provided written informed consent. The study protocols were approved by local ethics committees and institutional review boards at each participating site. Statistical analysis was performed in STATA version 14.1 (StataCorp, College Station, Texas). A 2-sided p value of <0.05 was considered significant.
The baseline characteristics of the study population are presented in Table 1 and Supplemental Table 1. Patients with available biomarker data, compared with patients without available data, were less likely to be Asian, on average 1 year older, with higher body mass index, lower estimated glomerular filtration rate, and lower rate of prevalent HF hospitalization in the past 9 months (all p < 0.001). Included patients also had a higher frequency of angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker, diuretic, and beta-blocker use at baseline. Within the biomarker substudy, baseline characteristics of patients randomized to sacubitril/valsartan and valsartan alone were balanced (Table 1). Two parameters had clinically minor but statistically significant differences: patients randomized to sacubitril/valsartan had greater prevalence of prior stroke (14% vs. 8%; p = 0.01) and hypertension (97% vs. 95%; p = 0.02).
Baseline biomarker values versus referent control subjects
Table 2 compares the baseline values of biomarkers in patients enrolled in PARAGON-HF with published values of these biomarkers from referent control subjects. Compared with referent control subjects, patients with HFpEF had greater median sST2, TIMP-1, PINP, PIIINP, and CITP. TIMP-1 and CITP exceeded referent control median values in >90% of patients.
Biomarker levels differed by sex: sST2, PIIINP, TIMP-1, and CITP were higher in men, whereas PINP was higher in women (Supplemental Table 2A). There were no statistically significant differences in baseline biomarker levels between patients with LVEF ≤57% or >57% (the median value) (Supplemental Table 2B).
Effects of treatment with sacubitril/valsartan versus valsartan on biomarkers
Figure 1 shows the effect of treatment with sacubitril/valsartan or valsartan on geometric mean biomarker levels 16 and 48 weeks after randomization. Sacubitril/valsartan decreased TIMP-1 by 8% (95% CI: 6% to 10%; p < 0.001), sST2 by 4% (95% CI: 1% to 7%; p = 0.002), and PIIINP by 3% (95% CI: 0% to 6%; p = 0.04) at 16 weeks, compared with valsartan after adjustment for baseline value (Figures 1A and 1C). CITP was increased by 4% (95% CI: 1% to 8%; p = 0.02) with sacubitril/valsartan compared with valsartan. There were no statistically significant differences in PINP between treatment groups.
At 48 weeks after randomization, reductions in TIMP-1 and sST2 with sacubitril/valsartan compared with valsartan were maintained (Figure 1B). PINP and PIIINP were numerically lower and CITP numerically higher with sacubitril/valsartan, but these differences were not statistically significant.
The effect of sacubitril/valsartan compared with valsartan on biomarker levels 16 weeks after randomization was similar in men and women and patients with lower versus higher LVEF (Figure 2), despite differences in primary endpoint treatment effect between these groups.
Baseline biomarker values versus outcomes
The risk of the primary outcome (composite of CV death and total heart failure hospitalizations) increased with greater baseline values of TIMP-1, sST2, PIIINP, and CITP, but not PINP, in unadjusted analysis (Table 3, Figures 3A and 3B). After adjustment for 23 covariates including LVEF, NT-proBNP, hsTNT, and randomized treatment group, sST2 and TIMP-1 retained statically significant associations with the primary endpoint. When all biomarkers and clinical covariates were included in 1 model, only higher TIMP-1 concentration remained independently associated with risk of the primary endpoint. Thus, in these patients with HFpEF, baseline values of TIMP-1 predicted outcomes independently of the other 4 profibrotic biomarkers, NT-proBNP, and hsTNT. Associations of biomarker levels with the components of the primary endpoint, CV death and HF hospitalization, were consistent in direction and approximate magnitude (Supplemental Table 3), though only sST2 independently predicted CV death after multivariable adjustment due to lower statistical power.
Baseline biomarker levels did not modify the effect of sacubitril/valsartan compared with valsartan on the primary endpoint (p for interaction >0.05 for all).
Relationship between change in biomarker from baseline and outcomes
Increases in biomarker levels from baseline to week 16 were associated with increased risk of subsequent primary endpoint events for all 5 biomarkers after adjustment for baseline value. After further adjustment for 7 covariates, changes in TIMP-1 (rate ratio [RR]: 1.23 per SD increase; 95% CI: 1.03 to 1.47; p = 0.02), sST2 (RR: 1.15 per SD increase; 95% CI: 1.02 to 1.31; p = 0.03), and CITP (RR: 1.26 per SD increase; 95% CI 1.03 to 1.55; p = 0.03) remained associated with the primary endpoint (Supplemental Table 4). The relationships between change in TIMP-1 and sST2 from baseline to 16 weeks after randomization and risk of subsequent primary endpoints are presented in Figures 3C and 3D.
Circulating biomarkers reflecting mechanisms of ECM homeostasis (sST2), collagen synthesis (PINP, PIIINP), and collagen degradation and turnover (TIMP-1, CITP) were abnormal in patients with HFpEF enrolled in the PARAGON-HF trial, suggesting that profibrotic signaling may play a role in the pathophysiology of HFpEF. Higher baseline levels of sST2 and TIMP-1 and increases in these markers over time were associated with greater risk of HF hospitalization and CV death. Sacubitril/valsartan altered these biomarkers in a pattern that suggests a decrease in profibrotic signaling. All of these findings were independent of sex and EF. In aggregate, these data suggest that biomarkers reflecting ECM homeostasis are elevated in HFpEF, altered by sacubitril/valsartan, and have important prognostic value (Central Illustration).
Comparison to published studies
Baseline values of the biomarkers examined in PARAGON-HF were comparable to HF cohorts derived from both epidemiologic studies and randomized clinical trials. TIMP-1 levels exceeded the referent control median in >90% of patients and were similar to patients with HFrEF in PARADIGM-HF (Prospective Comparison of ARNI with ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure) (7,11,15). PIIINP and CITP levels exceeded referent controls and were similar to those of patients with HFpEF in the Cardiovascular Health Study (17). Elevation in PIIINP was similar to that of patients with HFpEF in I-PRESERVE (Irbesartan in Heart Failure With Preserved Systolic Function) (19) and is the only marker studied with confirmed association to histologically proven myocardial fibrosis (20,21). sST2 levels exceeded subjects of similar age and sex in the Framingham Heart Study (14). PINP levels exceeded referent controls in the SHIP (Study of Health in Pomerania) study and were slightly lower than for patients with HFpEF in I-PRESERVE (19,22).
Relationship to outcome
In the current study, both baseline and change from baseline values of sST2 and TIMP-1 had a significant relationship with the primary study outcome of CV death and HF hospitalization. These results in this carefully defined HFpEF population are similar to cohorts of patients with HFrEF derived from epidemiologic studies and randomized clinical trials (11,23). However, to our knowledge, this is the first study to demonstrate this relationship in patients with HFpEF. The relationships between higher levels of sST2 and TIMP-1 (and increases in these markers) and greater risk of CV death and HF hospitalization remained statistically significant even after robust multivariable adjustment for all baseline covariates including NT-proBNP and hsTNT.
When all 5 measured biomarkers of ECM homeostasis were added to the multivariate model, TIMP-1 retained a significant relationship with the primary endpoint. Thus, TIMP-1 had prognostic significance independent of the other biomarkers of ECM homeostasis and natriuretic peptides and troponin. Similar associations for TIMP-1 were also observed in patients with HFrEF in the PARADIGM-HF trial (11). TIMP-1 was the strongest predictor of all-cause mortality after age in a cohort of >5,000 Icelandic patients (24), and in the Framingham Heart Study, TIMP-1 was associated with LV hypertrophy and systolic dysfunction by echocardiogram (25). TIMP-1 appears to identify patients with chronic HF at the highest risk for events, regardless of EF.
Effects of sacubitril/valsartan
The changes in biomarker levels observed with sacubitril/valsartan compared with valsartan alone would be predicted to reflect significant changes in ECM homeostasis. TIMP-1 inhibits a number of MMPs, thus inhibiting the catabolism of insoluble fibrillar collagen and promoting its accumulation. In PARAGON-HF and PARADIGM-HF, treatment with sacubitril/valsartan decreased TIMP-1; therefore sacubitril/valsartan may alter ECM homeostasis in a manner that may reduce profibrotic signaling. sST2 serves as a decoy for interleukin 33, facilitates profibrotic signaling through the ST2 receptor and may lead to increased collagen synthesis. In the PARAGON-HF, PARADIGM-HF, and PIONEER-HF (Comparison of Sacubitril/Valsartan Versus Enalapril on Effect on NT-proBNP in Patients Stabilized From an Acute Heart Failure Episode) trials, treatment with sacubitril/valsartan decreased sST-2 and therefore may alter ECM homeostasis in a manner that may reduce profibrotic and inflammatory signaling (11,26). The magnitude of TIMP-1 and sST2 reduction with sacubitril/valsartan was similar regardless of EF in this study and compared with reduction in patients with reduced EF in PARADIGM-HF.
The increase in CITP with sacubitril/valsartan may appear inconsistent with reductions in TIMP-1 and sST2, as all 3 markers are postulated to associate with greater profibrotic signaling. One explanation is that chronic elevations in CITP reflect greater collagen turnover (which may adversely affect prognosis), whereas short-term changes are driven by collagen degradation (which may facilitate more normal ECM homeostasis). In chronic pathological states, collagen turnover increases to a higher steady state level. Existing normal collagen is degraded and replaced with newly synthesized collagen that is processed and cross-linked in an abnormal fashion. This increased collagen turnover would be characterized by increased propeptides and telopeptides, the pattern observed in this study and others, in which patients with HFpEF had higher values of PINP, PIIINP, and CITP than did referent control subjects or patients with diabetes and hypertension but not heart failure (7,17,27). These clinical data are also supported by data from a recent porcine model of pressure-overload that results in HFpEF and abnormal ECM architecture (28). On the other hand, the short-term increase in CITP with sacubitril/valsartan may reflect greater collagen degradation, particularly when viewed in combination with a coordinate decrease in TIMP-1. In this substudy, sacubitril/valsartan decreased TIMP-1, which would be expected to allow increased MMP degradation of fibrillar collagen and the release of increased CITP. Moreover, although the increase in CITP cannot be dismissed, baseline CITP was not independently associated with the primary endpoint and the increase in CITP with sacubitril/valsartan was less statistically robust than decreases in TIMP-1 and sST2 and was observed only at 16 and not 48 weeks.
Baseline biomarker values were compared with those of referent control subjects from previous publications, not with those of patients enrolled contemporaneously as part of PARAGON-HF. However, studies with similar assays and patient demographics were selected, except for the presence of HF. Plasma and serum biomarker concentrations may be influenced by processes outside the myocardium, such as renal or hepatic dysfunction, bone or skin disease, systemic inflammation, malignancy, or pregnancy. However, patients with these conditions were excluded from the PARAGON-HF trial. Comparison to the active comparator valsartan, which may itself have antifibrotic effects, may have resulted in smaller treatment effects than would have been observed with a placebo control. The assays for TIMP-1, PIIINP, and CITP are indicated for research use only; the assays for sST2 and PINP are approved for clinical use by the U.S. Food and Drug Administration.
Patients with HFpEF who were enrolled in the PARAGON-HF trial had elevated levels of circulating biomarkers reflecting abnormal ECM homeostasis. Baseline biomarker levels and short-term changes were associated with subsequent risk of HF events independently of clinical risk factors. Sacubitril/valsartan significantly altered these biomarkers. These data suggest that interstitial myocardial fibrosis contributes to HFpEF pathogenesis and prognosis and that antifibrotic properties may contribute to the mechanism of potential benefit of sacubitril/valsartan in patients with HFpEF.
COMPETENCY IN MEDICAL KNOWLEDGE: Reductions in biomarkers of perturbed ECM homeostasis during treatment with sacubitril/valsartan are associated with improved clinical outcomes in patients with HFpEF.
TRANSLATION OUTLOOK: Further studies are warranted to understand the effects of sacubitril/valsartan on fibrotic signaling on the pathobiology of ECM homeostasis in patients with HFpEF.
PARAGON-HF was sponsored by Novartis. Dr. Cunningham has received a National Heart, Lung, and Blood Institute T32 postdoctoral training grant (T32HL094301). Dr. Claggett has received consulting fees from AOBiome, Biogen, Boehringer Ingelheim, Corvia, Gilead, and MyoKardia. Dr. O’Meara has received HSFC grant for the Bio-AIMI-HF study; has participated in clinical trials and steering committees for Amgen and American Regent; and has received honoraria for clinical trials, speaking, and consulting for her or her institution from AstraZeneca, Bayer, Boehringer Ingelheim, Merck, Novartis, and Pfizer. Drs. Prescott, Chiang, Rizkala, Shi, and Lefkowitz are employees of Novartis. Dr. Pfeffer has received consulting fees from AstraZeneca, DalCor, GlaxoSmithKline, Novo Nordisk, Sanofi, Jazz, MyoKardia, Servier, Takeda, and Corvidia. Dr. Shah has received grants from the National Institutes of Health (R01 HL140731, R01 HL120728, R01 HL107577, and R01 HL149423), American Heart Association, Actelion, AstraZeneca, Corvia, and Novartis; and has received consulting fees from Actelion, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Cardiora, Eisai, Ionis, Ironwood, Merck, Novartis, Pfizer, Sanofi, and United Therapeutics. Dr. Redfield has served as an unpaid consultant for Novartis. Dr. Zannad has received personal fees from Janssen, Novartis, Bayer, Boston Scientific, Amgen, CVRx, Boehringer Ingelheim, AstraZeneca, Vifor Fresenius, Cardior Pharmaceuticals, Cerno Pharmaceuticals, Applied Therapeutics, and Merck; and has received other payments from CVCT and Cardiorenal. Dr. Rouleau has received personal fees from Novartis and AstraZeneca. Dr. McMurray has consulted for Bayer, Cardiorentis, Amgen, Theracos, AbbVie, DalCor, Pfizer, Merck, AstraZeneca, GlaxoSmithKline, Bristol-Myers Squibb, Vifor Fresenius, KRUK, and Novartis; all payments were made through a consultancy with Glasgow University and were not personal payments. Dr. Solomon has received grants and personal fees from Novartis, Alnylam, Amgen, AstraZeneca, Bristol-Myers Squibb, Gilead, GlaxoSmithKline, MyoKardia, Theracos, and Cytokinetics; has received grants from Celladon, Bellerophon, Ionis, Lone Star Heart, Mesoblast, Eidos, MyoKardia, National Institutes of Health/National Heart, Lung, and Blood Institute, and Sanofi Pasteur; and has received personal fees from Akros, Corvia, Ironwood, Merck, Roche, Takeda, Quantum Genomics, AOBiome, Janssen, Cardiac Dimensions, Tenaya, and Daiichi-Sankyo. Dr. Zile has received a grant from Novartis; and has received consulting fees from Novartis, Abbott, Boston Scientific, CVRx, EBR, Endotronics, Ironwood, Merck, Medtronic, and MyoKardia V Wave.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC author instructions page.
- Abbreviations and Acronyms
- confidence interval
- collagen I telopeptide
- extracellular matrix
- heart failure
- heart failure with a preserved ejection fraction
- heart failure with reduced ejection fraction
- high-sensitivity troponin T
- left ventricular ejection fraction
- matrix metalloproteinase
- N-terminal pro–B-type natriuretic peptide
- N-terminal propeptide of collagen III
- N-terminal propeptide of collagen I
- rate ratio
- soluble ST2
- tissue inhibitor of matrix metalloproteinase
- Received April 20, 2020.
- Revision received May 28, 2020.
- Accepted May 28, 2020.
- 2020 The Authors
- Kasner M.,
- Westermann D.,
- Lopez B.,
- et al.
- Zile M.R.,
- Baicu C.F.,
- Ikonomidis J.S.,
- et al.
- López B.,
- Ravassa S.,
- González A.,
- et al.
- González A.,
- Schelbert E.B.,
- Díez J.,
- Butler J.
- Duprez D.A.,
- Gross M.D.,
- Kizer J.R.,
- Ix J.H.,
- Hundley W.G.,
- Jacobs D.R.
- Zile M.R.,
- Jhund P.S.,
- Baicu C.F.,
- et al.,
- for the PARAMOUNT Investigators
- Zile M.R.,
- Desantis S.M.,
- Baicu C.F.,
- et al.
- Zannad F.,
- Alla F.,
- Dousset B.,
- Perez A.,
- Pitt B.,
- Rales Investigators
- Iraqi W.,
- Rossignol P.,
- Angioi M.,
- et al.
- Tamura N.,
- Ogawa Y.,
- Chusho H.,
- et al.
- Zile M.R.,
- O'Meara E.,
- Claggett B.,
- et al.
- Solomon S.D.,
- McMurray J.J.V.,
- Anand I.S.,
- et al.,
- for the PARAGON-HF Investigators and Committees
- Solomon S.D.,
- Rizkala A.R.,
- Gong J.,
- et al.
- Coglianese E.E.,
- Larson M.G.,
- Vasan R.S.,
- et al.
- Barasch E.,
- Gottdiener J.S.,
- Aurigemma G.,
- et al.
- Lin D.Y.,
- Wei L.J.,
- Yang I.,
- Ying Z.
- Krum H.,
- Elsik M.,
- Schneider H.G.,
- et al.
- López B.,
- González A.,
- Ravassa S.,
- et al.
- Michelsen J.,
- Wallaschofski H.,
- Friedrich N.,
- et al.
- Aimo A.,
- Januzzi J.L. Jr..,
- Bayes-Genis A.,
- et al.
- LaRocca G.,
- Aspelund T.,
- Greve A.M.,
- et al.
- Morrow D.A.,
- Velazquez E.J.,
- DeVore A.D.,
- et al.
- Zhang Z.Y.,
- Ravassa S.,
- Yang W.Y.,
- et al.
- Torres W.M.,
- Barlow S.C.,
- Moore A.,
- et al.