PARAGON-HF: Is There a Role for the ARNI in HFpEF?

Ileana L. Piña, MD, MPH; Burkert Pieske, MD


September 16, 2019

This transcript has been edited for clarity.

Ileana L. Piña, MD, MPH: Hello. This is Ileana Piña from Wayne State University and Detroit Medical Center in Detroit, Michigan, in the United States. And this is my blog.

I borrowed Dr Pieske from Berlin, where he's the chair of cardiology at Charité Hospital, probably one of the most important institutions in Germany. Thank you for taking the time. I know it's been incredibly busy with the presentation of PARAGON-HF,[1] which we have been eagerly awaiting. Tell us a little bit about the study. What was its purpose?

Burkert Pieske, MD: Thank you, Ileana, for the opportunity to discuss PARAGON-HF, a state-of-the-art phase 3 outcome trial in heart failure with preserved ejection fraction (HFpEF). It is important to understand that HFpEF is a common disease; nowadays, almost half of our patients who are admitted for heart failure to our hospital have HFpEF.

Piña: In my place it's 40% of all admissions.

Pieske: It's a high proportion, and these patients are in a desperate situation because their signs and symptoms are as severe as in heart failure with reduced ejection fraction (HFrEF), but we have no good therapeutic strategies for them. We do not have any medication or therapy that would improve outcomes, so the recommendations are risk-factor control and diuretics.

Piña: Getting blood pressure control. That's all in the guidelines.

Pieske: Exactly. PARAGON-HF was a prospective trial in over 4400 patients that tested whether sacubitril/valsartan would be superior to valsartan alone in reducing cardiovascular death and repeat heart failure hospitalizations. Scott Solomon from Boston just presented these data at the European Society of Cardiology Congress (ESC) in Paris, and the key result was that there was a marginal 13% reduction in the primary outcome, which was a combination of cardiovascular disease and repeat heart failure hospitalization.

Piña: And was it time to first event?

Pieske: It is a little bit beyond time to first event. It is first heart failure hospitalization but then adding any subsequent hospitalizations.

Piña: So it's more the totality of hospitalizations as Pocock identified.[2]

Pieske: Exactly. And that 13% reduction just missed the level of significance of P < .05. We had a clear signal of efficacy that just missed the level of significance. I learned from the biostatistician that we needed just seven more events.

Piña: Oh my goodness. So why was the comparison done against valsartan?

Pieske: That's a very important question because, of course, valsartan itself may already have some benefit. Over 70% of the patients in PARAGON-HF were either on an ACE inhibitor or an ARB before randomization, probably for hypertension.

The investigators thought that it might hamper recruitment if we said, "You have to get rid of your antihypertensive therapy to be randomized."

Piña: So it was probably done for ease of enrollment because it was a hard trial to enroll. In the United States, most of my female patients were overweight and I couldn't put them in because there was a BMI cutoff. I had some of them on a diet.

I know it's an exploratory endpoint, but tell us about the mortality. What was the annual mortality?

Pieske: The annual event rate was in the range of 13 per 100 patients (13.6, if I recall correctly). That was the combination of deaths and heart failure hospitalizations, and the majority of the effect was derived from reductions in heart failure hospitalizations; there was very little effect on cardiovascular mortality or overall mortality.

Some interesting additional observations: First, the intervention was safe. It lowers blood pressure so hypotension was the most prominent adverse effect.

Piña: We know that about the drug.

Pieske: Yes, we know that from PARADIGM.[3] But as in PARADIGM, which was in HFrEF, it prevented progression of renal failure. And that was a significant effect.

There were other important observations from the PARAGON-HF subgroup analysis. We had a clear signal of heterogeneity of the effects of the drug in two subgroups, which have been discussed extensively over the past few days. Women benefited more than men and this interaction was highly significant. We don't understand why.

There was also an interaction in patients with an left ventricular ejection fraction (LVEF) below the median. So, 57% and lower.

Piña: Do you think that's the HF with midrange ejection fraction—the so-called HFmrEF population?

Pieske: It goes in the direction. The cut-off to include patients was an ejection fraction of 45% or higher and the median was 57%, so we're talking about those between 45% and 57%.

They had a clear benefit whereas in those with LVEF above 57%, the benefit faded. Obviously there is heterogeneity in patients with HFpEF.

Piña: We've been saying for years that HFpEF patients are not all the same.

If you go back and look at CHARM, for example, there were a lot of women in CHARM Preserved[4] and the populations are very similar. And if you go back into A-HeFT,[5] the women who were mostly hypertensive had a remarkable benefit from that combination early in the trial.

I wonder if it's the hypertensive women with HF who respond better quicker. Maybe it's all blood pressure–related. We may be looking at hypertensive heart disease in women.

Why do you think we have such a hard time getting a mortality signal with any of these trials?

Pieske: That's a tough question, Ileana. In PARAGON-HF you see that the curves are diverging all the time until the end of the trial. HFpEF is a disease that evolves over years and years, whereas HFrEF can sometimes occur immediately after myocardial infarction.

How can you reverse a disease that has evolved over years within 24 months? I think that is not an easy undertaking. I am personally convinced that longer-term follow-up of the more intermediate-risk patients—not all are 80 years or older—might detect signals of a mortality benefit.

Piña: It's going to take longer than the study follow-up period.

Pieske: Absolutely. We also have to adapt the way we do trials. We are not following these patients once they are censored or once the trial is over.

Piña: We keep using HFrEF information to set up our HFpEF trials.

I have one last question for you. The use of mineralocorticoid-receptor antagonists (MRAs) was kind of low in this trial: 20%-24%.

Why was that? TOPCAT[6] certainly had very strong signals for a benefit for spironolactone.

Pieske: I cannot fully explain it. There is some regional variability so there are countries with better uptake of MRAs. You do a good job in the US. You even now have a Class IIb recommendation in the guidelines.[7] We're not as far along in Europe.

Piña: You may want to do a subgroup analysis by region and take a look at the patients who were on an MRA, because I find clinically that it works very well with the ARB.

Pieske: That's very true. I use MRAs in my patients, especially when they are hypertensive; and many are on antihypertensive medication but still in the range of 140 mm Hg, and that blood pressure comes right down.

Piña: As long as the potassium is okay. And now we can even manage that.

I want to thank you for your time and I wish you luck. I look forward to other papers. You also have data on a Kansas City Cardiomyopathy Questionnaire that didn't change a lot. That's very intriguing. Every time we do one of these trials, we learn

Pieske: We have an ongoing trial on that. The PARALLAX trial will look at sacubitril/valsartan and quality of life.

Piña: We need to continue to learn from these trials. But I want to thank you. I wish you a lot of luck with the trial and I'm sure we're going to be hearing more. This is Ileana Piña, signing off. Please join us again for more information from the ESC. Have a great day.

For more Medscape Cardiology, follow us on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.