PROVE-HF Cracks the 'Holy Grail' of Cardiac Remodeling in HFrEF

Interviewer: Ileana L. Piña, MD, MPH; Interviewee: James L. Januzzi, MD


October 07, 2019

This transcript has been edited for clarity.

Ileana L. Piña, MD, MPH: Hello. I'm Ileana Piña from Wayne State University and the Detroit Medical Center in Detroit, Michigan, and this is my blog. I'm coming to you from the 2019 European Society of Cardiology Congress in the beautiful city of Paris.

I am especially happy today to be with Jim Januzzi, who is a named professor at Mass General and Harvard, plus a very good friend and colleague. We've been doing clinical trials together for several years. I am an executive committee member of the trial we're going to be talking about, PROVE-HF,[1] which is a trial of the angiotensin receptor neprilysin inhibitor (ARNI) sacubitril/valsartan. Welcome, Jim, and thank you so much for your time because I know this is a crazy meeting.

James L. Januzzi, MD: Ileana, it's a delight to be here with you.

PROVE-HF: Background

Piña: Thank you. Why did you think of this work and what made you build it in this fashion?

Januzzi: First I'll give a little bit of background about the PROVE-HF study. We know that in patients with heart failure and reduced ejection fraction (HFrEF), sacubitril/valsartan reduced cardiovascular risk. It's a Class 1 recommendation in the guidelines[2] on the basis of the PARADIGM-HF trial.[3] The problem is that the understanding about the mechanism of benefit remains entirely elusive. No one really had a good sense for how [it works].

Piña: Isn't that amazing that we did a trial and we didn't know? We knew it in the angiotensin-converting enzyme (ACE) inhibitors, we knew it in the angiotensin-receptor blockers (ARBs).

Januzzi: Yes, and the same story is developing around the sodium-glucose cotransporter-2 (SGLT2) inhibitors. We have no idea how they work, and yet here they are. So, with sacubitril/valsartan there was a clear, unmistakable benefit in terms of cardiovascular death and heart failure hospitalization, and we sought to understand why that would be. One of the observations from studies of sacubitril/valsartan is that the drug causes a robust reduction in N-terminal pro–B-type natriuretic peptide (NT-proBNP). We've seen that in PARADIGM-HF, PIONEER-HF,[4] and numerous trials. In turn, those reductions in NT-proBNP are strongly associated with the clinical benefit of sacubitril/valsartan. The question is, why?

One of the things that we and others have shown in the past is that NT-proBNP is much more than a biomarker of congestion. It is strongly associated with cardiac structure and function and left atrial pressure. The PROVE-HF study examined how changes in NT-proBNP after initiation of sacubitril/valsartan associated with improvements in structural measures of EF, left ventricular and left atrial volumes, and diastolic function.

Piña: When we did the early ACE inhibitor trials, we knew that ACE inhibitors prevented the ventricle from continued remodeling. We never proved that it totally stopped the remodeling or that it reversed it like the beta-blockers did.

Januzzi: Exactly, and that is a really important analogy. The degree of reverse cardiac remodeling seen with beta-blocker therapy is superimposable with the benefits with respect to mortality. Things that reverse cardiac remodeling, like cardiac resynchronization therapy (CRT), tend to reduce cardiac risk dramatically.

Piña: Correct. We did GUIDE-IT,[5] which in some cases was frustrating because we were not able to prove the point; and you did PROTECT,[6] which was a very positive study where, even in your own site with a smaller number of patients, you were able to show a lot of statistical significance. To me that was the background for thinking with this one.

Januzzi: Absolutely. We've learned a few things from GUIDE-IT and we've also taken some things away from PROTECT. In both of these trials, which were studies aimed at lowering NT-proBNP with the goal to improve outcomes, we found that if you lower NT-proBNP, patients do much better. The reason why GUIDE-IT was neutral was because both arms had substantial lowering, so they were superimposable. We just published a paper[7] in the Journal of the American College of Cardiology showing just how dominant it is if you get the NT-proBNP down. Patients do so much better. They feel better, their Kansas City Cardiomyopathy Questionnaire (KCCQ) quality-of-life scores improve, they are less likely to be readmitted, and, importantly, their left ventricular function improves. That set the stage for the PROVE-HF study. Isn't it amazing how things come together?

Patient Population

Piña: It's so cool. How many patients were enrolled?

Januzzi: We recruited 794 patients in 78 sites across America in under a year.

Piña: Where everybody is complaining about recruitment being bad.

Januzzi: This was a study of patients eligible for sacubitril/valsartan treatment, so we didn't have the usual long list of inclusion and exclusion criteria. This was a typical HFrEF population, with a mean age of 65 years and predominantly male (about 30% women).

Piña: I won't yell at you for that; 30% is not too bad.

Januzzi: It's not bad for a HFrEF trial. It is wonderful to see the demographic balance. Importantly, since this was an American trial, nearly 30% of our study participants were black patients, which is a relatively poorly studied group of patients with sacubitril/valsartan. There were the usual mix of comorbidities with a fair amount of obesity and diabetes. The mean EF at entry was 28%.

Piña: Which is almost exactly like HF-ACTION.[8]

Januzzi: Unlike HF-ACTION, the population was a bit more symptomatic. Their NT-proBNPs were 816 pg/mL. This was not a healthy population by any stretch.

Piña: And hospitalizations?

Januzzi: Many of the patients had a hospitalization within the 12 months before we brought them in.

Piña: Those are the NT-proBNP numbers that I see when people leave the hospital.

Januzzi: Also, a substantial percentage had relatively new-onset heart failure. One of the neat things about this study that we designed purposefully was that it included patients who were not represented in the PARADIGM-HF trial. This includes patients with new-onset heart failure, patients not taking an ACE inhibitor or an ARB before being started on sacubitril/valsartan, patients with low natriuretic peptide values, and patients who could not be titrated to target dose. Remember: In the PARADIGM-HF study run-in, if they could not get to target, they were out.

It's a real-world population. This study really provides useful information for the clinician.

Piña: And hopefully we're going to be able to see some of those subgroups.

Januzzi: We're planning on it. We already have some preliminary data that reflect the interesting effect of sacubitril/valsartan.


Januzzi: In the group as a whole, we found that from a baseline of 816 pg/mL, within 2 weeks the NT-proBNP had fallen by 30% and was maximally reduced essentially within the first month or two, suggesting that it's fast and that the lowest dose gets you most of the NT-proBNP lowering.

Piña: But we targeted the higher dose?

Januzzi: Sixty-five percent of the patients got to the highest dose. We never encouraged clinicians not to titrate, but we did want to study whether achieving the target dose was associated with less cardiac remodeling. We found that the degree of NT-proBNP reduction correlated quite significantly with the reverse cardiac remodeling in these patients. For the group as a whole, by 6 months we saw an average 5.2% increase in the EF, and by 1 year a 9.5% rise in the EF.

Piña: Very significant. Isn't that more than some of the beta-blockers?

Januzzi: It's more than any other drug therapy. The only therapy that is associated with that kind of lift in EF is CRT.

Piña: Right, which may not be there when you turn off the pacer.

Januzzi: That is an interesting point. Twenty-five percent of our patients had a 13% or more rise in their EF. One in four patients had a double-digit improvement. Really remarkable. Along with the EF changes, we saw improvements in left ventricular volumes, improvements in left atrial volume, reduction in E/e′—so, improved diastolic function.

Piña: Yes. We saw the filling of the ventricle getting better, which is something we have never seen with the other RAS inhibitors.

Januzzi: Absolutely not. We also found that the left ventricular mass index also fell by 16 g/m2.

Piña: Mass is the best measure of remodeling. In the background therapy, how many of them were on mineralocorticoid receptor antagonists (MRAs)?

Januzzi: About 35% came into the study on MRAs.

Piña: That is even higher than in PARAGON-HF.[9]

Januzzi: It's higher than in the CHAMP-HF registry. It's not where we want to see them but it's better than it was in other studies. Ninety-eight percent were on a beta-blocker and 75% were on an ACE or an ARB. Remember, many of the patients had new-onset heart failure, so now we get to the populations that were not studied in PARADIGM-HF. We can say that patients with new-onset heart failure not taking an ACE or an ARB, patients with low natriuretic peptides, and patients not getting to a target dose have comparable reductions in NT-proBNP and substantial reverse cardiac remodeling. Really remarkable.

Evaluating Biomarkers and Beyond

Piña: We also got a whole series of other biomarkers which hopefully we're going to get a chance to study. Did we get ST2? Did we get galectin-3?

Januzzi: We have ST2 but we do not have galectin-3. We have high-sensitivity troponin and multiple natriuretic peptides. Nasrien Ibrahim from my group published a paper[10] looking at the effects of sacubitril/valsartan in raising atrial natriuretic peptide (ANP). We're planning on measuring ANP in this population. We actually have preliminary data from other studies to suggest that ANP may be the thing that sacubitril/valsartan is pushing up to exert benefits, so we'll be looking at that. It's a very early biomarker with a very short half-life. We're going to examine just how much ANP rise occurs and how it correlates with improvements in left ventricular function.

Piña: So we may look at it parallel to the proBNP.

Januzzi: ProBNP, fibrosis markers, injury. One really important thing that I'm very happy about with this study is that we go beyond the mechanism and examine how patients feel. We try to understand the journey that these patients are going through because it's not all about their EF.

Piña: Patients say, "Can I get up in the morning and do what I want to do?" That is really what it's about.

Januzzi: It's really exciting to start looking at things like how much dyspnea they have and their KCCQ, something that I know is near and dear to your heart.

Piña: Near and dear to my heart because I use it all the time.

Januzzi: Absolutely. We'll be unpacking all of those data. The PROVE-HF paper was published in the Journal of the American Medical Association (JAMA) and it's already available. It is remarkable how this story has gone from the early NT-proBNP experience to the GUIDE-IT experience. The GUIDE-IT NT-proBNP response paper was published just 2 weeks ago and now PROVE-HF lands.

Piña: Are all of the other substudies lined up?

Januzzi: We have probably about a half a dozen substudies within the next few months to complete. We have a large number of different populations.

Piña: We are completing a very important picture that was not complete. We knew little pieces of it. Remodeling is the holy grail.

Januzzi: In HFrEF, no question about it.

Piña: It's the holy grail. I want to thank you for coming today, and thank you for all the work and your leadership.

Januzzi: Ileana, it's been wonderful. Thanks so much.

Piña: I want to remind our audience that this paper is the PROVE-HF trial, published in JAMA. Dr Januzzi is the first author. We have some beautiful tables in there, and hopefully you will take a look at your patients to see where they fit in. We're very excited about this. Thank you for joining me today, and come back and see us soon. This is Ileana Piña, signing off.

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.