September 29th, 2014

The Implications of PARADIGM-HF for Clinical Practice


CardioExchange’s Harlan M. Krumholz and John Ryan interview Clyde Yancy and Mary Norine Walsh to get their perspectives on the much-discussed PARADIGM-HF trial, previously covered on CardioExchange.

Krumholz and Ryan: Do you see any red flags in this study?

Yancy: These results are among the top five developments in the treatment of heart failure in my 25-year career: ACE inhibitors, beta-blockers, MRAs, device therapy, and now angiotensin­–neprilysin inhibition. We have come far, and now with these data the foundation of our treatment regimen has changed yet again. When we change the foundation, it doesn’t mean we’ve raised the bar, it means we’ve elevated the floor. For our discipline and especially for our patients, this is a new day.

I don’t see any red flags in this study. The data are sound, and the design met the top tier of study constructs. However, no study can be comprehensive and no drug or device can be considered the panacea for heart failure, an extraordinarily complex and heterogeneous disease. There are several unanswered questions:

1. The number of trial participants in the U.S. was small, about 5%. Can we be comfortable extrapolating the overall results to the U.S.?

2. The standard of care in PARADIGM-HF involved a lesser degree of device penetration than seen in U.S. heart failure populations: ICD, 15% vs. 30%; CRT, 7% vs. 14%. Would the results remain as relevant in a U.S. treated cohort?

3. Older persons (>75 years) had a favorable point estimate but wide confidence interval. We shouldn’t be compelled by subgroup analyses, but this is a cohort in whom we consistently fail to have adequate data, and it is an ever-increasing patient population in heart failure. We need more data.

4.  Patients with NYHA class III and IV heart failure were not well represented in this study. Moreover, study eligibility required tolerability of full doses of either enalapril (10 mg twice daily) or the new regimen, which included valsartan (160 mg twice daily). This may have precluded entry of sicker patients and slanted the study toward mild to moderate heart failure.

5. Perhaps not a red flag but a yellow caution light was that African Americans — the cohort in whom angioedema was most prominent with the drug omapatrilat — were essentially absent from this study. There were about 450 participants of African descent, 219 on active drug. This is a paltry number to reassure us that the risk for angioedema in this group is not elevated. Yet the drug formulation is indeed different, and the overall incidence of angioedema in the study was <1% and not dissimilar between the ACE-inhibitor and angiotensin­–neprilysin inhibition groups. We ought be reassured, but given the burden of heart failure in African Americans, we should be certain.

Walsh: Not any red flags per se, but there are some interesting findings in the group studied.

This was a huge study: Approximately 10,000 patients were randomized at 1043 centers. That works out to an average enrollment of 10 patients per site. The relatively low enrollment per site suggests to me that investigators had trouble finding eligible patients, that the patient refusal rate was significant, or both.

The mean blood pressure was 122 mm Hg — pretty substantial when the systolic blood pressure could be as low as 100 mm Hg at enrollment and 95 mm Hg at randomization. This, and the fact that 70% of patients had NYHA class II symptoms at enrollment, suggests that the patients were pretty stable.

A minority of patients were enrolled at sites in the United States, which accounts for the low numbers of patients with ICD and CRT devices: 15% and 7%, respectively.

Each of these factors will come into play when considering this new drug for a patient who is “sicker” with lower blood pressure, poor functional status, and current ICD and/or CRT therapy.

In addition, the sponsor did the initial data analysis. Data transparency is important, and I hope that the sponsor will make the raw data available for analysis.

Krumholz and Ryan: If approved, given what you have seen so far, would you envision switching over many of your patients? And what sort of patients will you use this agent on — for example, patients with a new diagnosis of heart failure? Or will you be stopping ACE inhibition to start this agent?

Yancy: I see this question differently than simply addressing the logistics of “ in whom and when?” I agree that early implementation of this new therapy should be pursued. However, the emphasis is on implementation, which necessarily comes with a new dynamic in clinical medicine — implementation science, or “the study of methods to promote the integration of research findings into practice.” Thus, we should change the paradigm of new drug introduction and proceed diligently with real-time data acquisition, ongoing surveillance for side effects, toxicity, and careful assessment of outcomes, especially hospitalizations for heart failure. Our focus should be on the elderly, those with devices, and African Americans. We ought not just to “make this drug available” but, rather, to create an infrastructure to thoughtfully dispense this therapy in a model that helps us learn as we go.

With that in mind, and with great hope that a sophisticated registry will be place, I would advise following the clinical trial data: Target patients already tolerating full doses of an ACE inhibitor, and proceed with transitioning to the angiotensin­–neprilysin inhibitor — or initiating this therapy in newly diagnosed patients while making certain that full doses can be tolerated. Remember, we have data in hand that any dose of an ACE inhibitor improves survival (see the ATLAS trial), but we don’t have the same data for the angiotensin­–neprilysin inhibitor.

The rush will be to offer this new drug to the patient failing current therapies. Such use would be empiric, as that is not the scenario in which these data were acquired. We should not offer false hope.

Walsh: The decision to use the new drug, either at the time of a new HF with reduced ejection fraction diagnosis or as a replacement for an ACE inhibitor in a patient with established HFrEF, is a preference-sensitive decision. I see this as an opportunity to share the decision with my patients, after reviewing the results of PARADIGM-HF and any information/restrictions provided by the FDA. My explanation of the trial results and benefits of the new medication will be influenced by my patient’s clinic status and characteristics and how he/she compares with patients randomized in the trial. Ultimately, the decision will rest with the patient.

Krumholz and Ryan: Will cost affect your decision about who should receive this medication?

Yancy: At a population level, no. The only metric that will matter is cost-effectiveness, which certainly should be met given the reduction in hospitalizations. Provided that the cost is not excessive, this therapy should qualify as sound healthcare economics.

But, on a patient level, yes. We see patients already on extensive multidrug regimens and typically with a still-significant out-of-pocket expense for copays. As well, what will happen in community hospitals, especially safety-net hospitals and VA hospitals, where formulary costs are under intense pressure? Unlike for a new iterative therapy, where market forces can predominate in determining cost, for a breakthrough compound that changes the foundation and whose use will be broad, is there a different economic model that might work?

The sign of good research is that more questions are raised and that new opportunities for better care are realized; such is the case here. My hope is that the introduction of this therapy will lessen the burden of heart failure and usher into the clinical sphere a new era in caring for HF patients.

Walsh: Cost will clearly play a role. During the past few years, medical treatment of HFrEF has benefited from the availability of relatively inexpensive, evidence-based generic medications. If approved, how the insurers “tier” this new medication with regard to copayment rules will have a big effect on the out-of-pocket cost to patients. A 90-day supply of 40 mg enalapril tablets is currently available for as little as $10 per month in the U.S. That is a hard price to beat. Each patient will need to consider cost, as well as outcome benefit, and make an informed decision.

2 Responses to “The Implications of PARADIGM-HF for Clinical Practice”

  1. H Robert Silverstein, MD says:

    PARADIGM-HF should have compared “2 to 2”: enalapril (a cough-inducing and now rarely clinically used ACE I vs the non-cough inducing ARB valsartan which in & of itself raises issues) plus spironolactone or epleronone. The enthusiasm for LCZ696 seems hysterical & inappropriate to me. HRS, MD, FACC

    • Judith Andersen, AB, MD says:

      I agree with Dr Silverstein — except that when one plans a clinical trial,data that trickle in,from problematical either parallel or primary studies, may make the primary comparison irrelevant.That seems to be true in this case . Cannot see that this study clarifies a forward thinking treatment path. The side-effect uses are important, and although cost is always an issue,do not think this will be a life-changing result. Could be wrong.

      No competing interests for either.

      Thank you,

      Judith C Andersen,MD