From PARADIGM to PARAGON Further Evidence Supporting Continuous Heart Failure Spectrum

Neurohormonal inhibitors form the backbone of treatment in heart failure (HF) with low left ventricular ejection fraction (LVEF).1 In contrast, treatment of HF with near-normal/normal LVEF2 is based on expert consensus focused on congestion and comorbidity management.3 The usefulness of LVEF as a discriminator of HF phenotypes has been challenged and it has been suggested that HF is a syndrome across a continuous LVEF spectrum with varying degrees of neurohormonal activation and response to treatment.4, 5 The continuous HF spectrum concept is further supported by several large-scale randomized trials, which demonstrated that HF patients with an LVEF of 40–50% have similar clinical features as those with a LVEF < 40%.6 The recent pooled analysis of two trials,7 the Prospective Comparison of ARNI with ARB Global Outcomes in HF with Preserved Ejection Fraction (PARAGON-HF) trial8 and the Prospective Comparison of ARNI with ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure (PARADIGM-HF) trial,9 further strengthens one of the pivotal notions of the continuous HF spectrum concept, namely that neurohormonal overactivity and consequently the effectiveness of neurohormonal modulating agents, progressively decline as LVEF increases but both remain measurable at higher LVEF. Heart failure is a heterogeneous syndrome in which disease progression is associated with a dynamic evolution of functional and structural changes, leading to unique disease trajectories creating a spectrum of phenotypes with both overlapping and distinct characteristics.4 Accordingly, maladaptive neurohormonal overactivity, e.g. renin–angiotensin system (RAS), sympathetic nervous system, and other systems (arginine vasopressin, endothelin, adrenomedullin, etc.), prevails over the adaptive neurohormonal systems regardless of LVEF, but the magnitude of the difference declines from the lower to near-normal/normal LVEF. The major adaptive neurohormonal system involves the natriuretic peptides (NPs).10 Elevation of the NP levels are observed in HF regardless of LVEF, but the magnitude is higher in HF with low LVEF vs. HF with near-normal/normal LVEF. However, HF is associated with an increased ratio of unprocessed (inactive)/processed (active) NPs. In essence, HF involves a state of active NP deficiency where the myocardium continues to produce the NP prohormones that are not processed into active forms.11 Thus, throughout the continuous HF spectrum: (i) maladaptive neurohormonal overactivity is present but more so in the lower LVEF than in near-normal/normal LVEF, and (ii) adaptive neurohormonal overactivity represented by the NPs is defective. It is, therefore, anticipated that an agent like sacubitril/valsartan inhibiting maladaptive neurohormonal overactivity and augmenting the active NP forms will be effective throughout the continuous HF spectrum. However, effectiveness will be less in those with near-normal/normal LVEF (Figure 1). Sacubitril is a prodrug and is metabolized to sacubitrilat, which inhibits the endopeptidase neprilysin catalysing the degradation of several vasodilator peptides including the NPs (atrial NP more than B-type NP), as well as vasoconstrictor peptides.12, 13 Valsartan is an angiotensin receptor blocker (ARB). Sacubitril/valsartan provides natriuretic and diuretic properties and better blood pressure control as compared to RAS inhibitors alone.14 Initiation of treatment with sacubitril/valsartan causes an acute reduction in cardiac filling pressures.15 This is most likely due to sacubitril/valsartan induced rapid and strong reduction of soluble neprilysin activity leading to an increase in the levels of NPs.12 In addition, sacubitril/valsartan has the advantage of concomitantly blocking a pro-fibrotic/pro-hypertrophic mechanism while stimulating an anti-fibrotic/anti-hypertrophic mechanism.16 In a recent experimental study, sacubitril/valsartan proved more effective than valsartan at ameliorating cardiac fibrosis during pressure overload-induced hypertrophy, with distinct effects on cardiac fibroblast gene expression.17 PARAGON-HF8 and PARADIGM-HF9 were randomized, double-blind trials including patients with LVEF ≥45% and ≤40%, respectively. Both trials employed similar designs (eligibility criteria, run-in period), active comparators, and endpoint assessment. In PARAGON-HF sacubitril/valsartan was compared to valsartan, whereas in PARADIGM-HF to the angiotensin-converting enzyme inhibitor (ACEi) enalapril. In PARAGON-HF the rate ratio for the primary endpoint was 0.87 (95% confidence interval 0.75–1.01; P = 0.059). Of the pre-specified subgroups, suggestion of benefit was shown in patients with an LVEF 45–57% and in women. The results concerning the interaction between treatment effect and LVEF in PARAGON-HF are consistent with those obtained in the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) study18 and the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Programme,19 both showing a similar gradient of benefit with less benefit as LVEF goes up. Regarding the possible sex-related modification of the effect of treatment, there are several potential explanations, but a definite mechanistic basis for this finding could not be provided by PARAGON-HF.20 However, the sex difference may be partly due to a greater arterial stiffness in females compared with their male counterparts.21 In contrast to PARAGON-HF,8 PARADIGM-HF9 was stopped early after a median follow-up of 27 months, because the boundary for an overwhelming benefit had been crossed. In a pooled analysis,7 the treatment effect of sacubitril/valsartan compared with RAS inhibition for the composite of total HF hospitalization and cardiovascular death declined with increasing LVEF but remained superior to RAS inhibition up until a LVEF of approximately 60%. Interaction analyses suggested that LVEF was an effect modifier for the treatment effect (interaction P = 0.02). Guidelines recommend the ACEi enalapril used as the comparator in PARADIGM-HF for the treatment of HF with low LVEF (class IA recommendation), but do not endorse the ARB valsartan used as the comparator in PARAGON-HF for the treatment of HF with near-normal/normal LVEF.1 Thus, interpretation of the PARAGON-HF findings requires clarification of whether valsartan acted as an inert or an effective agent, as in the case of the latter it could dampen any potential beneficial effects of sacubitril/valsartan. In the Irbesartan in Heart Failure with Preserved Ejection Fraction Study (I-PRESERVE) trial there was a high proportion of patients receiving an ACEi, which increased from 25% in both groups at baseline to 39% in the irbesartan and 40% in the placebo group.22 The combination of RAS inhibitors is potentially harmful and is not recommended and may have obscured the beneficial effects of irbesartan. The CHARM-Preserved study examined the effects of the ARB candesartan in patients with LVEF > 40% and showed that HF hospitalization was lower in the candesartan group.23 A post hoc analysis of the CHARM Programme evaluating the effects of candesartan across the entire LVEF spectrum, demonstrated that candesartan reduced the primary endpoint (cardiovascular death or HF hospitalization) up until a LVEF of approximately 50% and recurrent HF hospitalizations up until a LVEF of approximately 60%.19 Thus, based on the CHARM Programme data, valsartan cannot be considered inert in HF up to a LVEF of approximately 60% and this may have contributed to an attenuation of the effect of sacubitril/valsartan in PARAGON-HF. The effect of background therapy in the two studies should also be considered as it may differentially attenuate the treatment response if it is not inert. As neurohormonal overactivity is less in HF with near-normal/normal LVEF than in HF with low LVEF, if patients are treated with a neurohormonal inhibitor (e.g. beta-blocker) the relative attenuation of the effect of an add-on neurohormonal inhibitor (e.g. sacubitril/valsartan) will be greater in HF with near-normal/normal LVEF than in HF with low LVEF. About 80% of PARAGON-HF8 and 90% of PARADIGM-HF9 patients were treated with beta-blockers. Thus, sacubitril/valsartan was given on top of beta-blockers. Although beta-blockers are not recommended for HF with near-normal/normal LVEF, they cannot be considered ineffective in this setting. A meta-analysis of randomized trials demonstrated that beta-blockers improve LVEF and prognosis for patients with HF in sinus rhythm and LVEF < 40%, but a similar benefit is observed in the subgroup of patients with LVEF 40–49%.24 Approximately 25% of patients in PARAGON-HF and 55% in PARADIGM-HF were treated with mineralocorticoid receptor antagonists (MRAs), which in those with near-normal/normal LVEF of the TOPCAT trial showed no significant effect.25 However, a post hoc analysis of the TOPCAT trial revealed an approximately four-fold difference in the primary outcome between patients randomized in Russia and Georgia vs. Americas.26 The lack of identification of canrenone in 30% of participants from Russia compared with 3% from the Americas (P < 0.001) established regional discrepancies in the reported and actual use of spironolactone.27 Thus, a greater relative attenuation of the sacubitril/valsartan effect (predominantly by beta-blockers) in PARAGON-HF by background treatment cannot be excluded. The PARAGON-HF and PARADIGM-HF trials demonstrated that the efficacy of sacubitril/valsartan vs. RAS inhibition, on top of treatment with beta-blockers and in a proportion with MRAs, varies with LVEF. Greater treatment benefit was observed in patients with low LVEF compared with higher LVEF, but effectiveness, particularly for HF hospitalization, remained measurable up until a LVEF of approximately 60%. These findings are in accordance with the continuous spectrum theory of HF purporting that maladaptive neurohormonal overactivity and consequently the effectiveness of neurohormonal modulating agents progressively decline as LVEF increases but both remain measurable at higher LVEF. These and data from previous trials with neurohormonal modulation suggest that careful consideration with these agents should be given in managing HF patients not only with low (< 40%) but higher LVEF as well. The Australian clinical guidelines for the management of HF28 and a recent expert consensus meeting report of the Heart Failure Association of the European Society of Cardiology29 endorse consideration of neurohumoral inhibitors in HF patients with LVEF between 40% and 49%. The possibility that with the use of sacubitril/valsartan HF patients with LVEF up to approximately 60% may also benefit should be considered. Conflict of interest: F.T. has received research support and honoraria from Amgen, Bayer, Boehringer Ingelheim, Elpen, Genesis Pharma, Lilly, Menarini, Merck, Novartis, Sanofi, Servier, Vianex and WinMedica. A.X. has received honoraria from Novartis. J.B. serves as a consultant for Amgen, Array, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CVRx, G3 Pharmaceutical, Innolife, Janssen, LivaNova, Luitpold, Merck, Novartis, NovoNordisk, Relypsa, Roche, Sanofi, V-Wave Limited, and Vifor.