An overview of the PARAGON-HF trial with special consideration to the rationale behind the design, the cardiovascular end points, and the safety profile.
John McMurray, MBChB: The most recent attempt to identify treatments for patients with heart failure and preserved ejection fraction [HFpEF] was with sacubitril/valsartan and the PARAGON-HF trial, which was the sister trial to PARADIGM-HF, which was the heart failure in reduced ejection fractions [HFrEF], sacubitril/valsartan trial. The PARAGON-HF trial overall showed a borderline effect with sacubitril/valsartan on the primary composite outcome of cardiovascular death and heart failure hospitalizations. The P value was just above .05, so the trial was not technically statistically significant, although if we added a few worsening heart failure events that needed emergency treatment but didn’t lead to hospital admission, that made the overall treatment effect significant, so it was borderline.
It was driven, as you would expect, by around a 15% reduction in heart failure hospitalization, but the interesting thing was that that benefit was concentrated in patients with an ejection fraction below the median value in the trial, which was 57%. In other words, what we were seeing was that there was another drug that seemed to be effective in patients who have what we would today call heart failure with midrange ejection fraction: patients with an ejection fraction that’s certainly not normal; it’s below normal, but it’s not in the HFrEF range. In other words, it’s not below 40%. But those people with heart failure and ejection fraction below normal seem to be getting some benefit from sacubitril/valsartan, from another drug called candesartan, and from mineralocorticoid receptor antagonists, like spironolactone and eplerenone. In December, there will be an advisory committee convened by the US Food and Drug Administration to review these data to ask this question: is it appropriate to use sacubitril/valsartan and maybe other drugs in patients who have heart failure with preserved ejection fraction, but who have an ejection fraction in the lower end of the range of patients included in the trials, patients who we might describe as midrange or those with mildly reduced ejection fraction? It will be interesting to see what the advisory committee recommends to the FDA because if they were to recommend using these treatments in those patients, then that would be the first time that we have had an approved therapy for any patient with heart failure with preserved ejection fraction.
Scott D. Solomon, MD: PARAGON-HF was designed to determine whether sacubitril/valsartan could benefit patients with heart failure with preserved ejection fraction, in the way we had previously been able to show that it would benefit patients with heart failure with reduced ejection fraction with the PARADIGM-HF trial. PARAGON-HF was based on a pilot study called PARAMOUNT that demonstrated in 300 patients a reduction in NT-proBNP [N-terminal pro-b-type natriuretic peptide], improvement in left atrial size, and improvement in New York Heart Association class. We did a fairly large outcomes trial in 4700 patients to demonstrate benefit with sacubitril/valsartan in heart failure with preserved ejection fraction. As you may know, the overall results of the trial were just short of statistical significance. We saw a 13% reduction in a composite end point of cardiovascular death and total heart failure hospitalizations with a P value of .059. This was driven almost entirely by a reduction in heart failure hospitalization, with not much of an effect on cardiovascular deaths. We don’t expect to see much of an effect on cardiovascular deaths in HFpEF because these patients have a lower risk for cardiovascular death than patients with heart failure with reduced ejection fraction.
Importantly and interestingly, with PARAGON-HF, we saw something that was similar to what we'd seen previously in the CHARM trial and in the TOPCAT trial with spironolactone, which is that the patients who appeared to benefit the most are the patients with ejection fractions that were below normal. So not frankly reduced ejection fraction like in HFrEF patients, but ejection fraction that was below what we normally consider within the normal range. That raises the possibility that one size is not going to fit all in patients with HFpEF. In fact, patients who have heart failure with ejection fraction that is not frankly reduced, but is not normal, might be the kind of patient who would benefit from therapies that we already know to be beneficial in heart failure with reduced ejection fraction.