The primary end point, a composite of cardiovascular death or hospitalization for HF, favored the ARNI drug. Most important, overall death was lower in the ARNI vs enalapril group at 17% vs 19.8%, respectively.
There is no other way to describe PARADIGM other than to say it was a positive trial. Most heart-failure experts are enthusiastic. Mainstream media, too, are sold. And only hours after the FDA approved the drug, I received an upbeat email from a Novartis sales representative.
The ocean breezes and warm sun of Barcelona also had its effect on me. I was initially
positive on the trial. Among other strengths, I noted the large numbers of patients in both groups; the good baseline CHF therapy; biologic plausibility of neprilysin inhibition; and the highly significant
P values.
Time for thought and reflection is a potent elixir in the practice of medicine. I feel differently now. Closer inspection of the trial and discussion with experts in methodology have led me to these 10 reasons for caution.
- The active arm of PARADIGM featured a high dose of valsartan against a low to moderate dose of enalapril. Was angiotensin blockade equivalent? Was enalapril a straw-man comparator?
- PARADIGM had an active run-in phase in which 12% of patients dropped out. There are no active run-in phases out here in the real world.
- PARADIGM enrolled a select group of young patients. The mean age was 63 years; 80% were male, and only 6% were black. How does that cohort generalize to the millions of elderly, female, and nonwhite CHF patients with comorbid diseases?
- The absolute difference in death rate was 2.8% over 2.5 years. The translation: 97.2% of patients experienced the same result—they did not die.
- PARADIGM was terminated early. That fact introduces bias.
- More than 1000 centers randomized patients in PARADIGM-HF. That requires a lot of quality control over the raw data.
- More patients in the ARNI group experienced low blood pressure. Any clinician knows low blood pressure is a major problem in CHF treatment, especially in the elderly and those with comorbidities.
- A group of European authorsraised concern that neprilysin inhibition of β-amyloid peptide could increase amyloid deposition in the brain and predispose to dementia. These authors concede no cognition issues were seen in PARADIGM, but they also noted the 2.5-year trial was not designed to assess a problem that may accrue over the long term. (Drugs don't affect just the heart.)
- Cost is another reason for pause. Some estimates have valsartan/sacubitril costing over $4500 per year.comorbid diseases.[4] The conversations, I am afraid, will focus on the new chemical rather than the powerful effects of old-fashioned therapies, such as exercise, diets, and attention to sleep.
Summary
The new ARNI drug might prove to be as successful in the real world as it did in the clinical trial. Maybe. But the real world is a place where drugs aren't free and where patients aren't always young, male, and screened with run-in periods.
In elementary school, I learned that replication of astounding results is a tenet of good science. We don't seem to need that anymore when it comes to clinical science. That's too bad, because the history of medicine is replete with examples where we let hope, bias, and marketing get the best of us.
Do not dismiss the problem of cost and distraction. When we spend money on one therapy, we have less for other therapies. When we spend precious bandwidth in the clinic speaking about expensive new pills, we have less time to promote the basics of health.
I am not against progress. History may prove that combining valsartan and sacubitril was a great idea. The
early signs are positive. But years of practicing electrophysiology have taught me the value of slow medicine. So slowly is how I will proceed with this new class of chemicals.