After Sinking in, PARADIGM-HF Critiqued at HFSA Sessions

September 25, 2014

LAS VEGAS, NV — Its design was "outstanding," it was carried out "extraordinarily well," and its findings are "spectacular," Dr Marvin A Konstam (Tufts University, Boston, MA) said of the PARADIGM-HF trial, which had its public North American unveiling at the recent Heart Failure Society of America (HFSA) 2014 Scientific Meeting .

It was "a very well-designed trial," agreed Dr Lynne Warner Stevenson (Brigham and Women's Hospital, Boston, MA). "The effects on the primary end points are remarkable."

And Dr John GF Cleland (Harefield Hospital, Imperial College London, UK) called it "one of the landmark studies" in heart failure.

"But I think some of the criticisms are fair."

The three heart-failure thought leaders, none of whom were investigators with the trial, were themselves among those at the subspecialty meeting raising questions about PARADIGM-HF, which compared venerable enalapril to the recently developed angiotensin-receptor/neprilysin inhibitor (ARNI) LCZ696 (Novartis) in patients with reduced-LVEF heart failure, who were overwhelmingly in NYHA class 2–3.

Patients taking the new agent showed a 20% drop in the primary end point of CV death or HF hospitalization and in CV death on its own, along with a 16% fall in all-cause mortality (p<0.001 for all three end points), compared with those who received the older agent, as heartwire previously reported.

No one interviewed by heartwire soon after the trial's first public viewing at the recent European Society of Cardiology 2014 Congress had an unkind word about it. At the time, it would have been hard to give a substantive critique, and they generally didn't try, given that they had only just been presented with the results.

By the HFSA sessions a week later, the dust had settled a bit, and the potentially game-changing PARADIGM-HF results had been given some time to sink in. Among raised questions about the trial: Were there ways it was set up to favor the ARNI? Did randomized patients adequately reflect clinical practice? Was it greater renin-angiotensin-system (RAS) inhibition, not LCZ696 per se, that accounted for the benefits? Is the one trial enough to change guidelines recommendations?

ARNI Instead of ACE Inhibitors?

Among the top reservations about PARADIGM-HF is the extent to which it should replace ACE inhibitors in patients with systolic heart failure, for whom the agents are a mainstay. The trial randomized only patients who successfully completed a single-blind run-in phase consisting of enalapril for two weeks and then LCZ699 for twice that long—all after being stable for at least a month on either an ACE inhibitor or angiotensin-receptor blocker (ARB). Of the 10 521 entering the run-in phase, >2000 dropped out, mostly because they couldn't tolerate one of the two agents, often because of hypotension; ultimately 8442 were randomized.

"How do we know what will happen in the real world?" Konstam asked. The study doesn't show how many patients never before exposed to the drug would become hypotensive. In clinical practice, he said, "You don't get a run-in period."

As the discussant following the presentation of the trial by co–primary investigator Dr Milton Packer (University of Texas Southwestern, Dallas), Stevenson delved more deeply into that issue. "If we look at the whole sequence of intolerance" throughout the PARADIGM-HF run-in and randomized phases, 18% of patients who ever took LCZ696 dropped out.

That wasn't significantly different from the dropout rate for patients who didn't take the drug, she noted. "However, we have to be very careful about taking the remarkable positive end point that we see and generalizing it all the way back to the much larger population of patients [with low-EF heart failure], most of whom, of course, were not in this trial."

Also, Stevenson said, "It is not possible from this trial to compare the tolerability of enalapril with LCZ[696]. If the investigators had wanted to do that, they would have had to randomize the order during the run-in."

Were Patients Representative of Clinical Practice?

Only about 12% of PARADIGM-HF patients also had a biventricular pacemaker for cardiac resynchronization therapy (CRT), with (CRT-D) or without (CRT-P) implantable cardioverter defibrillator (ICD) capacity, Dr John JV McMurray (University of Glasgow, Scotland), a co–principal investigator for the trial, said in a presentation.

That's low, according to some critics who have had doubts whether the patients were on adequate guidelines-recommended therapy at the start.

But most of the trial's patients had been in NYHA class 2, and "at the time we started the trial, which was in 2009, there were only two large clinical trials with CRT-D, COMPANION and CARE-HF ," McMurray said. "And in fact, the guideline recommendations at that time were that CRT was indicated only in patients with a persistently low ejection fraction, class 3 or 4 symptoms, and of course a broad QRS complex."

But "I think there's no doubt that ICD use was lower than might be recommended by the guidelines." About 15% in the trial had ICDs, a third of those being CRT-D, "which was actually very similar to what happens in other trials, reflecting the practice throughout the world. And practice around the world seems to be at variance with the guidelines."

Device usage was greater in North America, where 54% of patients had ICDs, alone or as CRT-D; that compared with only 12% in the rest of the world.

Stevenson said she was unconcerned about the level of ICD penetration in the study, "as this would not have had an impact on the hospitalization benefit, and [the risk of] sudden deaths are relatively low in the class-3 population."

Why 10 mg Twice Daily and Not 20 mg Twice Daily?

"The reason we used [enalapril] 10 mg twice a day is quite simply because 20 mg twice a day is rarely reached," according to McMurray. CONSENSUS , a 1987 trial with only 127 patients actually receiving enalapril, and the 1991 treatment portion of the SOLVD trial with 1284 recipients, are the only two placebo-controlled trials that have showed improved outcomes with any ACE inhibitor, he said. Moreover, CONSENSUS is the only one to have had a target dose of 20 mg twice daily, and only 22% of patients taking enalapril were successfully titrated to that target; the mean total daily dosage was only 18.4 mg.

The target in SOLVD-Treatment was 10 mg twice daily, and the achieved total daily dosage was 16.7 mg, McMurray noted; similar ranges short of a 10-mg twice-daily target had been achieved in other enalapril trials such as the prevention part of SOLVD, V-HeFT 2 , and OVERTURE (2002).

In PARADIGM-HF, with >4000 patients who received the ACE inhibitor, "We achieved a daily dose of 18.9 [mg] or thereabout," he said.

Thus, "In well-treated patients, LCZ[696] maintained clinical stability longer and more effectively than the highest doses of enalapril ever achieved in a clinical-outcomes trial," Packer said when presenting PARADIGM-HF at the HFSA sessions.

Others have proposed that the trial's apparent gains from LCZ696 actually had more to do with the suppression of the renin-angiotensin system than a contribution from the ARNI's neprilysin-inhibitor moiety. Packer pointed out, after all, that the PARADIGM-HF dosage of 200 mg twice daily delivers about 320-mg/day valsartan.

"We think it's clearly no, because we have not only studies comparing high- to low-dose ACE inhibitors with ARBs [ATLAS and HEAAL , together including about 7000 patients], but we of course have trials where an ARB was added to an ACE inhibitor [ Val-HeFT and CHARM-Added , together including >7500 patients]. And in those trials of intensification of renin-angiotensin blockade, we simply did not see the sort of mortality benefits that were received in the PARADIGM-HF trial."

Nor did the significant mean systolic blood-pressure reduction from the ARNI vs enalapril (p<0.001) account for the primary-end-point benefit in a covariate analysis, according to McMurray.

LCZ696: Ready for Prime-Time?

"I don't want to second-guess the guidelines committees, but surely this must [deserve] a class I [level] a indication," Cleland said. "What intrigues me is what the indication will be for an ACE inhibitor in new guidelines. Will they also include Ia for the same group of patients?"

Stevenson said, "I don't believe it is time [for ARNIs] to replace ACE inhibitors or ARBs," given all the patients who didn't tolerate the drug or didn't meet the inclusion criteria.

"The advent of LCZ[696] enriches our respect for the complexity of our patients' individual responses. I'm not really sure which actions of ACE inhibitors are the most important for which patients. I'm not [even] sure which actions of beta-blockers are most important for which patients. And I'm certainly not sure of all the hormones affected by LCZ[696] or which are going to be the most important for each patient. These results should inspire us to do new research on these kinds of individual responses and the challenge of optimizing our therapies in patient-centered ways," she said.

"I don't think that LCZ[696] is ready for a level I indication; that is a higher bar," added Stevenson. "I think we could see a level IIa indication based on the strong results that we've seen."

Eight thousand patients is a lot, but not enough to answer every question about the new agent, Cleland observed. "We are going to need more clinical trials."

Should they be morbidity and mortality outcomes trials, he wondered, or "more of the nature of working out what we should do in clinical practice?" Also, he added, much will be learned about the therapy once it's introduced into clinical practice.

Cleland discloses receiving research support from Novartis, Amgen, and Phillips; speaking fees and/or receiving honoraria from Novartis, Amgen, Servier, and Sorin; and consulting for Novartis, Amgen, Medtronic, Phillips, Merck, Siemens, Servier, and Sorin. Konstam discloses receiving research support from Merck and Otsuka and consulting for Merck, Otsuka, Johnson & Johnson, Amgen, Novartis, Ono, and Pfizer. McMurray discloses that his institution is paid by Novartis for his involvement in the PARADIGM-HF and ATMOSPHERE trials and that he has received reimbursement from the company for travel related to the trials. Packer discloses consulting for Novartis. Stevenson discloses consulting for St Jude Medical.


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.