ARNI Again Shows Modest Benefits in HF With Mid-Range EF

Starting sacubitril/valsartan (Entresto) for recently decompensated heart failure with ejection fraction (EF) over 40% improved a key prognostic marker and, for those in the 40-60% range, clinical outcomes as well, the PARAGLIDE-HF trial showed.

The angiotensin receptor-neprilysin inhibition (ARNI) reduced time-averaged N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels compared with valsartan alone (ratio of change 0.85, 95% CI 0.73-0.99, P=0.049), reported Robert J. Mentz, MD, of the Duke Clinical Research Institute in Durham, North Carolina, and colleagues in the Journal of the American College of Cardiology.

While the primary endpoint just squeaked by on statistical significance, the secondary hierarchical outcome of cardiovascular death, heart failure (HF) hospitalizations, urgent HF visits, and change in NT-proBNP favored sacubitril/valsartan without reaching significance (unmatched win ratio 1.19, P=0.16).

"These data add to the evidence supporting a potential treatment benefit of Sac/Val in those with EF >40% (particularly in those with EF below normal), and may influence future guidance for use of Sac/Val in this population, regardless of treatment setting (hospital vs clinic) or HF chronicity (acute on chronic vs de novo HF)," the researchers concluded.

However, an accompanying editorial by Hector Ventura, MD, of the John Ochsner Heart and Vascular Institute in New Orleans, and colleagues drew a less favorable conclusion, noting that "the margin of benefit was less than anticipated on the basis of the prospective power calculation. The finding that clinically meaningful outcomes on cardiac events did not show a difference suggest that the magnitude of reduction in the natriuretic peptide levels may not have been sufficient to reach threshold for clinical benefit."

The findings follow the PARAGON-HF trial that was the basis for FDA expansion of the indication for sacubitril/valsartan beyond HF with reduced EF, although noting that benefits are most clearly evident with an EF below normal.

That trial narrowly missed its primary endpoint in reducing HF hospitalization and cardiovascular death compared with valsartan alone in people with an EF of 45% or higher, but the benefit was significant in the subgroup with an EF at or below the median of 57%.

In PARAGLIDE-HF, too, the subgroup with an EF below normal (≤60%) had a larger NT-proBNP change (OR 0.78, 95% CI 0.61-0.98) and a significant benefit over valsartan alone in terms of the hierarchical outcome (win ratio 1.46, 95% CI 1.09-1.95).

"Overall, these data ... suggest that the evidence for treatment benefit of ARNI in unselected patients with HFpEF [HF with preserved EF] appears limited," Ventura's group argued.

Given that a post-hoc analysis of PARAGON-HF signaled a greater benefit with sacubitril/valsartan in patients recently hospitalized for HF, PARAGLIDE-HF enrolled patients within 30 days of a worsening heart failure event requiring IV diuretic treatment, whether an HF hospitalization, emergency department visit, or out-of-hospital urgent HF visit.

Enrollment also required that patients be medically stabilized, with a systolic blood pressure over 100 mm Hg for the 6 hours prior to randomization, no increase in IV diuretic agents or use of IV vasodilators within the last 6 hours, and no IV inotropes for 24 hours before randomization.

PARAGLIDE-HF randomized 466 patients with EF over 40% to sacubitril/valsartan titrated to a target dose of 97/103 mg twice daily or to valsartan titrated to 160 mg twice daily. Just over half of the participants were women, 22% were Black patients, and the mean age was 70 years. While all the participants had an EF over 40%, the median was 55%. One-third of the patients had de novo HF, and 69.5% were enrolled in the hospital.

Sacubitril/valsartan reduced worsening renal function compared with valsartan alone (OR 0.61, 95% CI 0.40-0.93) but increased symptomatic hypotension (24% vs 16%, OR 1.73, 95% CI 1.09-2.76).

The increased hypotension in patients without enrollment limited by run-in phase results suggested "it may be prudent to delay initiation of this therapy in those patients exhibiting ongoing requirements for adjustments of antihypertensive therapy or diuretic agents or in those patients who have not demonstrated stability to previous use of vasodilators," Ventura and colleagues noted.

The editorialists pointed to a potential reason the clinical outcome endpoint missed significance in the overall study: "Notably, data to analyze the primary endpoint were not available for a substantial number of patients because of higher numbers of missing biomarker values and patient withdrawals in the sacubitril/valsartan group."

The 45-60% EF subgroup gave a "glimmer" of hope but still wasn't that impressive, according to Ventura and co-authors.

The win ratio methodology doesn't easily translate into magnitude of effect, durability of benefit, or eventual cost-effectiveness, they pointed out. "This analytic methodology also ignores ties (neither win nor loss), and given that ties occurred in more than one-third of the comparisons in PARAGLIDE-HF, the win ratio calculation, which is restricted only within the category of winners and losers, likely overestimates treatment effect."

"Furthermore, hierarchical endpoints in the win ratio calculation included the biomechanistic surrogate NT-proBNP as a final tie breaker, and this category was responsible for the majority of 'wins' for sacubitril/valsartan. In the absence of this subcategory, clinical endpoints alone would not have reached significance, even in the subgroup with EF ≤0.60," the editorialists wrote.

However, the trial was underpowered for and had too few clinical events over a duration of observation too short to really demonstrate benefit in clinically meaningful outcomes, they added. "Thus, in aggregate, although the study suggests some evidence of a beneficial trend of sacubitril/valsartan in HFpEF and a recent episode of worsening HF, the data are far from conclusive."

Comment