PARAGON-HF Misses Primary Endpoint, but Hints of Benefit Seen

PARIS, France—Sacubitril/valsartan (Entresto; Novartis) failed to significantly reduce the risk of hospitalization for heart failure or cardiovascular death in patients with heart failure with preserved ejection fraction (HFpEF) in the PARAGON-HF trial, dashing hopes that this drug would serve a major unmet need in this growing patient population. But the full results released here today provided some evidence that the treatment might improve outcomes in certain patient subsets.

Through a median follow-up of nearly 3 years, the rate of the composite primary endpoint per 100 patient-years was 12.8 with sacubitril/valsartan and 14.6 with valsartan alone (rate ratio 0.87; 95% CI 0.75-1.01), Scott Solomon, MD (Brigham and Women’s Hospital, Boston, MA), reported here at the European Society of Cardiology Congress 2019, saying the study “just missed statistical significance.”

That result was already known after drugmaker Novartis released the topline results about a month ago. But in the full details, published simultaneously online in the New England Journal of Medicine, exploratory secondary endpoints suggested that patients treated with sacubitril/valsartan were more likely to have an improvement in NYHA class and quality of life and less likely to have worsening renal function.

“Several key supportive analyses and secondary endpoints, including how patients feel and kidney function, are suggestive of benefit with sacubitril/valsartan compared with valsartan,” Solomon said in a press conference, noting that subgroup analyses pointed to significant benefits in terms of the primary endpoint in women and those in the lower end of the ejection fraction range included in the study (45% to 57%).

“These findings suggest that in this disease with no evidence-based therapy currently and a huge unmet need, some patients with heart failure with preserved ejection fraction, particularly those with ejection fractions below normal but not frankly reduced, . . . may benefit from sacubitril/valsartan,” he concluded.

But Mariell Jessup, MD, chief science/medical officer of the American Heart Association, provided a more sober assessment. Even though the trial was well conducted and included a clearly defined HFpEF population, “sadly, I don’t think there was a convincing signal that this would be the drug for HFpEF that we hoped,” she told TCTMD. “We hoped that this would finally be an answer for these patients.”

The subgroup analysis showing a potential benefit in women is intriguing, Jessup said, noting that preserved ejection fraction is more common than reduced ejection fraction in female heart failure patients.

But the take-away message for clinicians right now? “We still don’t have a proven therapy for HFpEF, which is disappointing,” Jessup said.

Shelley Zieroth, MD (University of Manitoba, Winnipeg, Canada), who was not involved in the trial, was more optimistic. “There was a lot of initial disappointment when we heard the preliminary topline results saying that the primary endpoint wasn’t met, but essentially today’s release of the paper really does give everybody sort of a glimmer of hope,” she told TCTMD.

Zieroth, who is president of the Canadian Heart Failure Society, said the trial “just barely missed statistical significance” on the primary endpoint, noting that it’s possible a greater difference wasn’t seen between the arms because of the use of an active comparator in valsartan.

She also highlighted the possibility of good results in women and patients at the lower end of the ejection fraction range.

“I don’t think that the door is shut for sacubitril/valsartan in HFpEF at all,” Zieroth said. “I think that there’s going to be more studies down the road and more exploration of the signals that we saw in PARAGON-HF.”

Searching for a Solution

Sacubitril/valsartan is currently approved to improve outcomes in patients with chronic heart failure with reduced ejection fraction (40% or lower) based on the results of the PARADIGM-HF trial, which showed that the angiotensin receptor-neprilysin inhibitor (ARNI) reduced a composite of cardiovascular death or hospitalizations for heart failure.

There are no therapies proven to improve outcomes in patients with preserved ejection fraction, however.

A phase II trial showed that sacubitril/valsartan lowered N-terminal pro-B-type natriuretic peptide levels and led to a larger reduction in left atrial size and a greater improvement in NYHA class compared with valsartan alone in HFpEF patients.

PARAGON-HF was designed to see whether sacubitril/valsartan could improve clinical outcomes as well. The trial, conducted at 848 centers in 43 countries, enrolled 4,822 patients 50 or older (mean age 73 years; 52% women) who had NYHA class II to IV heart failure, an ejection fraction of 45% or higher, elevated natriuretic peptides, and evidence of structural heart disease. They were randomized to sacubitril/valsartan (with target doses of 97 mg for sacubitril and 103 mg for valsartan) or valsartan alone (with a target dose of 160 mg) taken twice daily. All background medications were continued, with the exception of renin-angiotensin system inhibitors other than mineralocorticoid receptor antagonists.

Because sacubitril/valsartan did not significantly reduce the rate of the composite primary endpoint, all other analyses were considered exploratory. There was no difference in CV death (8.5% with sacubitril/valsartan vs 8.9% with valsartan alone; HR 0.95; 95% CI 0.79-1.16), with a nonsignificantly lower rate of hospitalization for heart failure with sacubitril/valsartan (rate ratio 0.85; 95% CI 0.72-1.00).

Several secondary outcomes indicated a potential advantage for sacubitril/valsartan. Patients taking the ARNI were more likely to have an improvement in NYHA class (15.0% vs 12.6%) and at least a 5-point improvement on the Kansas City Cardiomyopathy Questionnaire clinical summary score (33.0% vs 29.6%); they were less likely to have worsening renal function (1.4% vs 2.7%).

Safety and tolerability were similar to what was seen in patients with reduced ejection fraction in the PARADIGM-HF trial, with a higher rate of hypotension and lower rates of elevated potassium and creatinine, Solomon reported. Angioedema was more frequent in the sacubitril/valsartan arm (0.6% vs 0.2%), although none of the cases was associated with airway compromise, Solomon said.

Should Doctors Use This in Practice?

Asked whether he would consider using sacubitril/valsartan in patients with borderline low ejection fraction, Solomon said he thinks those with so-called midrange ejection fraction would benefit from treatment. “I’m not speaking for regulators or for the guidelines, but I suspect in this group there’s at least some rationale for treating patients with this therapy,” he said during the press conference.

Solomon’s fellow steering committee co-chair agreed. “Normally we’re very cautious about subgroups,” he said, but in this case there’s supporting evidence from other heart failure trials that have suggested a benefit in patients with preserved ejection fraction, such as the CHARM-Preserved trial of candesartan and the TOPCAT trial of spironolactone. “That’s to me what lends credibility to this, this sort of external support from other trials to suggest that these patients with mild systolic dysfunction above the rates that we would normally consider reduced ejection fraction have something to gain from probably all of these treatments,” John McMurray, MD (University of Glasgow, Scotland), said.

I’m not speaking for regulators or for the guidelines, but I suspect in this group there’s at least some rationale for treating patients with this therapy. Scott Solomon

Jessup acknowledged that it’s possible that there are subtypes of HFpEF that would respond well to treatment with sacubitril/valsartan, predicting: “I think HFpEF is so difficult to treat that I won’t be surprised if clinicians—if they’re frustrated with how their patients are feeling—might try the drug. I won’t be surprised if I hear that’s happening.”

Whether that’s justified based on these results is unclear, she said. But, she added, digging deeper into what’s going on in the HFpEF subsets who showed a potential benefit may give clinicians a better idea about who might get something out of taking sacubitril/valsartan.

Whether the therapy should be used in HFpEF patients in practice based on the PARAGON-HF results will be “an issue of great debate in the upcoming weeks and months,” according to Zieroth, who indicated that a case can be made for using sacubitril/valsartan in at least one subset of patients.

“I think there should be great enthusiasm for the signal we saw in women. Women are usually underrecognized and underenrolled in clinical trials; 51% of this study population were women and they showed a significant benefit in a prespecified analysis,” she said. “I think that that’s really important, and women should advocate for consideration of this drug in the management of HFpEF.”

More broadly, there was a nonsignificant trend toward a reduction in hospitalizations for heart failure as well as advantages for sacubitril/valsartan on some of the secondary outcomes, Zieroth pointed out. “The totality of the evidence should be looked at in terms of using sacubitril/valsartan in select groups of HFpEF patients.”

Two panelists at the press conference who were not involved in PARAGON-HF supported assessing trial results based on more than just whether the primary endpoint did or did not meet statistical significance.

P. Gabriel Steg, MD (Hôpital Bichat, Paris, France), said such a black-and-white view of trials “is misleading, and this is why many in the field are moving to different types of analysis such as Bayesian analysis of trials, which give us a little more leeway in interpreting the data.”

In the case of HFpEF, Steg added, “I think the unmet clinical need in this population is absolutely huge. We have nothing that really has strong evidence of benefit. And I think that everything in this trial points to substantial potential benefit. . . . I think this is a very important advance in the field.”

Deepak Bhatt, MD (Brigham and Women’s Hospital), agreed with Steg’s comments. “I look at the totality of the evidence, as I think most good practicing physicians do,” he observed. And looking at the results presented by Solomon, Bhatt said, “To me, PARAGON-HF was positive in those patients that had an EF of, say, 50%, which isn’t normal.”