PARADIGM HF Teases With Visions of New Paradigm for Treating Chronic Heart Failure

May 20, 2014

ATHENS, GREECE — The phase 3 trial has a few quirks, one of its designers observed, as he related how its data safety monitoring board (DSMB) urged it be stopped because patients getting the test agent were showing a clear advantage. There had never been a corresponding phase 2 study. It wasn't placebo-controlled; it evaluated a novel potential treatment for chronic heart failure instead of—not on top of—an old-standby ACE inhibitor. PARADIGM-HF , the largest-ever drug trial in chronic HF, was prospectively designed to follow its patients for mortality for up to two decades.

And it had an audacious goal, which was "to provide compelling evidence to replace current inhibitors of the renin-angiotensin system [RAS] as the cornerstone of the treatment of chronic heart failure," Dr Milton S Packer (University of Texas Southwestern Medical Center, Dallas) told a cavernous and crowded auditorium this week at Heart Failure Congress 2014 of the European Society of Cardiology Heart Failure Association.

Actually, sponsor Novartis at the end of March had announced PARADIGM-HF's early termination. It meant, Packer stated, that "the study had demonstrated convincing superiority of LCZ696 over high doses of [ACE inhibitor] enalapril on cardiovascular mortality as well as on the primary end point," which was a composite of cardiovascular death or HF hospitalization.

LCZ696 remains the name of the test drug, an angiotensin-receptor/neprilysin inhibitor (ARNI) that can replicate effects of the angiotensin-receptor blocker (ARB) valsartan as well enhance levels of endogenous natriuretic-peptide vasodilators. "This would be expected to greatly potentiate the existing inhibitors of the renin-angiotensin system," Packer said.

Building on Earlier Promise

Dr Milton S Packer

The ARNI had already shown promise in hypertension; blood pressures fell off more sharply for patients given it rather than valsartan in a large 2010 trial, as reported by heartwire . LCZ696 was also seen to improve surrogate markers for heart-failure severity and cardiac structure in patients with preserved-ejection-fraction HF, for whom no solid drug therapies are currently available, in a small study also comparing it with valsartan.

Packer knew that PARADIGM HF had shown benefit for CV death by itself, in addition to CV death/HF hospitalization, because the trial had been designed to permit its early termination only if there appeared to be a "compelling effect" on both the single and composite end point, he said.

When we designed the stopping rules, we made them so conservative . . . that it was mathematically almost impossible to stop the trial early.

Indeed, in another anomalous feature of the trial, its target sample size wasn't determined by the primary composite end point. The trial was specifically powered to detect a 15% drop in relative risk for CV death alone[1].

"John and I know almost nothing about the results," Packer later told heartwire , referring to Dr John JV McMurray (University of Glasgow, Scotland), with whom he is co–principal investigator. "When we designed the stopping rules, we made them so conservative—we told ourselves on the executive committee—that it was mathematically almost impossible to stop the trial early. So no one was more surprised than we were."

Packer's teasing PARADIGM-HF presentation and also the vibe in the auditorium evoked the image of someone unexpectedly given a peek inside the biggest package in the pile of birthday presents, not knowing the contents exactly but delighted at the prospect of one of the finest gifts ever received.

The assigned discussant for Packer's presentation, Dr Clyde Yancy (Northwestern University, Chicago, IL), crystalized the feeling: "Most of us can only imagine the extent to which these results are positive, and all of us are intrigued regarding potential introduction of a therapy that may fundamentally change practice. There hasn't been this much anticipation in the field in nearly 20 years."

Later in his presentation he said hopefully, "We've needed something better in heart failure, and something better may have finally arrived."

Packer himself had concluded his own presentation by announcing that the PARADIGM HF data will soon be unlocked, and he hopes that what is known about the trial will be presented at the European College of Cardiology 2014 Congress in Barcelona, Spain, in late summer. "For those of you who will be there, we promise you a most interesting presentation."

PARADIGM HF launched in 2009 at ultimately 1042 sites throughout the world, randomizing 8458 patients in chronic NYHA class 2–4 heart failure, with an LV ejection fraction <35% and elevated natriuretic peptides, who completed at least four weeks of treatment with an ACE inhibitor or ARB equivalent to enalapril at 10 mg/day and maintained if possible on a stable dose of beta-blockers, preferably with aldosterone inhibitors. They were assigned to either 200-mg LCZ696 or 10-mg enalapril, both twice daily, on top of the other guidelines-based treatments.

Although the mean follow-up time when the trial was terminated has yet to be revealed, Packer told heartwire that if he had to make a guess, it would be perhaps three years, with some patients followed for as long as five years.

Milton's PARADIGM Lost?

Denied further details about the trial's outcomes, as was most everyone, and assigned the role of discussant knowing only some methodology and that the results were "positive," Yancy imposed perspective on the discussion by playing devil's advocate. He outlined possible scenarios that could potentially still keep LCZ696 from supplanting conventional RAS-inhibiting agents, at least just yet.

Dr Clyde Yancy

Top of the list: "Will the data be valid?" he asked. "Early trial termination often leads to overstated benefits and understated safety observations. We should take the time to vet all the results and these data carefully."

Also, the enrolled patients have yet to be characterized in full, at least publicly, so it remains possible that PARADIGM HF could actually and less usefully be a trial in only mild heart failure. Yancy noted that 70% of its patients entered in NYHA functional class 2.

How broadly will its results apply? Less than 10% of the population was from North America, only about 5% from the US; fewer than 500 patients may not be enough to lock in a positive decision from the FDA. About 10% were from Russia, "and we've previously seen the angst created by the group of patients who came from the Russian cohort, particularly, in the TOPCAT study," Yancy noted. "Thus, this is a trial where outcomes in the different subgroups may matter greatly."

We're either on the precipice of peril or the horizon of a true paradigm change.

Yancy also referred to another major clinical-trial experience with a functionally similar agent, omapatrilat, which had been derailed as a potential hypertension or heart failure drug a dozen years ago, in part because it posed a worrisome risk of angioedema, particularly in African Americans. The angioedema had been attributed to the drug's inhibitory effects on aminopeptidase. "A multinational black population may not be a sufficient surrogate for what was [previously] seen in the African American group," he proposed.

"What is especially different about this LCZ preparation, regardless of nomenclature, vs the previous compound, omapatrilat? Is it just the absence of the effect on aminopeptidase? Is it the dose effect as described? Or is it the patient population studied? This will require much more thought and deliberation."

Given those possible pitfalls, Yancy said of the excitement over the trial's early termination, "We're either on the precipice of peril or the horizon of a true paradigm change, as the name of the study would infer."

There are "important reassurances" that it's the latter. There's consensus among expert observers of PARADIGM HF that—given what's known about it—the trial appears to have been well designed. There was excellent background therapy, including wide-scale use of aldosterone antagonists, he observed. The trial went for "the highest possible bar," cardiovascular mortality. And "there may be strength in the multiple different cohorts that were recruited. Indeed, these results likely are scalable."

Betting on LCZ696

PARADIGM HF's unusual design and intentions could be seen as needlessly risky, given the enormous capital and reputational investment typical of a major drug clinical trial. But they may partly represent the trialists' attempt to subvert a curse attributed to philosopher George Santayana—that is, they wanted to remember the past so as to avoid repeating it. The omapatrilat experience weighed heavily in PARADIGM HF's origins, in particular the OVERTURE trial, which marked the end of the drug's bid to become the next best thing for heart failure.

That trial failed to show 40-mg omapatrilat once daily to be superior to 10-mg enalapril twice daily for the primary end point of death from any cause or HF hospitalization, specifically one that required IV drug therapy. Randomized patients had been in NYHA 2–4 heart failure with an LVEF <30% and an HF hospitalization within the previous year.

But the new drug was at least equivalent to the old one for the primary end point. "If it had been compared with placebo it would have shown a clear benefit," according to Packer himself at the time.

We saw things in the OVERTURE trial that we believed in, and we thought that if we ever had a chance to do it again, that it would be positive.

"If there's one study that I had to do over again in my whole career, it would have been that trial," Packer told heartwire . "The reason is that omapatrilat is a once-a-day drug in hypertension, but it was a twice-a-day drug in heart failure. We gave it once a day in OVERTURE and didn't realize until we analyzed the results of the trial that it didn't work once a day in heart failure. We were acting on a mechanism without acting on it for a 24-hour period. To know that at the end of a trial was heartbreaking."

Also, the HF-hospitalization component of the primary end point required the use of IV drug therapy; but it was typical in some OVERTURE countries to go with intensified oral medications under circumstances the trial designers—they learned too late—would have expected IV drugs to be initiated. That caused many otherwise-valid HF hospitalizations not to be counted toward the primary end point in some countries, according to Packer.

So, he said, "We saw things in the OVERTURE trial that we believed in, and we thought that if we ever had a chance to do it again, that it would be positive."

But there would have to be a few changes, including a different drug. Omapatrilat inhibits three enzymes key to its effects in the trial; it works as an ACE inhibitor, not an ARB, and it suppresses aminopeptidase P along with neprilysin. ACE inhibitors carry a greater risk of angioedema than ARBs. So the agent that became LCZ696 was designed based on neprilysin inhibition with angiotensin-receptor (not ACE) blocking and without significant aminopeptidase inhibition.

All that gave them the confidence to be creative about the next phase 3 test of the strategy in heart failure, PARADIGM HF. For example, usually a new drug passes a phase 2 trial on the road to regulatory approval. But there's no FDA mandate to conduct such a trial.

In phase 2 studies, Packer observed, clinicians gain experience with a drug and, often, find what's thought to be an optimal dosage. "But the major reason that you do phase 2 is to build the political capital inside the [sponsoring] company to set aside an enormous amount of money to do phase 3. So we made a persuasive case that we had everything that we needed to know without doing phase 2," he said.

"Many companies put too much faith in the importance and interpretability of phase 2. The correlation between what happens in phase 2 and phase 3 is not very good." So their argument was, 'If it's not very good, and you have external data that suggest you should go forward, skip it.' "

PARADIGM HF is funded by Novartis Pharmaceuticals, from which Packer, McMurray, or their institutions have received consulting fees and travel and research support related to the conduct of the trial. Yancy had no disclosures.


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.