This article is more than 5 years old. Information may no longer be current.
TOPCAT: Spironolactone failed to improve clinical outcomes in HFpEF
DALLAS — Compared with placebo, treatment with spironolactone did not yield significant reductions in CV mortality, aborted cardiac arrest or HF hospitalization in patients with HF and preserved ejection fraction.
Marc A. Pfeffer
“It’s unfortunate that the substantial progress we’ve had in the management of patients with reduced heart failure has not been translated to those with more preserved [heart failure],” Marc A. Pfeffer, MD, PhD, the Dzau professor of medicine at Harvard University Medical School, said during an AHA 2013 press conference.
Pfeffer and colleagues conducted the large, randomized, double blind, placebo-controlled TOPCAT study to determine whether mineralocorticoid receptor antagonists such as spironolactone produce the same clinical improvements in patients with HF with preserved ejection fraction (HFpEF) as those with reduced EF.
The trial included 3,445 patients with HFpEF (median age, 69 years) at 270 centers in the United States, Argentina, Brazil, Canada, Russia and the Republic of Georgia. Patients were randomly assigned spironolactone 15 mg titrated to 45 mg, with a target of 30 mg (n=1,722), or placebo (n=1,723). Mean spironolactone dose was 28 mg at 8 months, Pfeffer said.
Inclusion criteria included symptomatic HF, age ≥50 years and LVEF ≥45%. Patients also had to be hospitalized within the past year for HF or have elevated natriuretic peptides.
About one-third of patients were not taking medication toward the end of the trial, and the researchers were unable obtain the vital status of 67 patients in the spironolactone group and 65 in the placebo group.
The primary endpoint was CV death, HF hospitalization or aborted cardiac arrest. During a mean follow-up of 3.3 years, the primary endpoint occurred in 20.4% of the placebo group and 18.6% of the spironolactone group (HR=0.89; 95% CI, 0.77-1.04). Event rates were 6.6 and 5.9 per 100 patient-years, respectively.
The researchers observed a lower percentage of HF hospitalizations among patients receiving spironolactone vs. placebo (12% vs. 14.2%; HR=0.83; 95% CI, 0.69-0.99), but this difference was not statistically significant.
All-cause hospitalization was similar between the two groups (HR=0.94; 95% CI, 0.85-1.04), as was all-cause death (HR=0.91; 95% CI, 0.77-1.08). There were also no significant differences in the number or reports of serious adverse events between groups. However, compared with placebo, the spironolactone group exhibited more hyperkalemia (18.7% vs. 9.1%; P<.001), suggesting the need to carefully monitor potassium levels with use of spironolactone, Pfeffer said.
Patients on active therapy were also almost 50% more likely to experience doubling of creatinine above the upper limit of normal (P<.001). There was no difference between groups, however, in the number of patients with levels ≥3 mg/dL and the number of patients who required dialysis was not increased due to the monitoring system in place.
Of 22 prespecified subgroup analyses, results revealed one significant interaction with treatment among patients with elevated natriuretic peptides.
An exploratory, post-hoc analysis showed that placebo event rates varied dramatically by region: 31.8% (HR=0.82; 95% CI, 0.69-0.98) in the United States, Argentina, Brazil and Canada and 8.4% (HR=1.1; 95% CI, 0.79-1.51) in Russia and the Republic of Georgia. This is an area that may require further investigation, according to Pfeffer.
“In our trial, we do not show a significant difference in our primary endpoint, but we do show reductions in hospitalizations for HF. We stress that if physicians use [spironolactone] for that purpose, they have to carefully monitor potassium and creatinine,” Pfeffer said. – by Melissa Foster and Katie Kalvaitis
For more information:
Pfeffer MA. LBCT.03. Medical and surgical approaches to improving heart failure outcomes. Presented at: the American Heart Association Scientific Sessions; Nov. 16-20, 2013; Dallas.
Disclosure: Pfeffer reports financial ties to Aastrom, Amgen, Bristol-Myers Squibb, Celladon, Cerenis, Concert, Genzyme, Hamilton Health Sciences, Keryx, Medtronic, Merck, Novartis, Roche, Sanofi-Aventis, Servier, Teva, University of Oxford and Xoma.
Perspective
Back to Top
HFpEF is overtaking HF with reduced ejection fraction (HFrEF). As the population gets older, we are seeing more and more cases. Obesity and hypertension also contribute to the rising incidence of HFpEF. We haven’t made many gains in the treatment of HFpEF in the last 20 years or so. In HFrEF, we have several things that work, including beta-blockers, ACE inhibitors and aldosterone blockers. Now, TOPCAT provides a potential clue for the treatment of HFpEF. The data aren’t compelling because we didn’t reach the primary endpoint. Also, when you look at the geographic distribution, we don’t know what happened in Russia and the Republic of Georgia.
We have to look at those data further. I think some people are going to look at these data and continue to use spironolactone. I’m sure others will look at the data and do the opposite, since the trial didn’t reach its primary endpoint. It is going to be up to the individual clinician to decide.
Perspective
Back to Top
The rationale for this trial was very strong — not just because this drug has been proven effective in HF with reduced ejection fraction (HFrEF) or hypertension, but also because, specifically in HFpEF, it has been shown that aldosterone levels correlate with mortality, suggesting that aldosterone activation of the receptor as occurs in the kidney and vascular smooth muscle cells is important. Perhaps even more importantly, it has been shown that obesity, diabetes, hypertension, coronary disease, atherosclerosis and the high-sodium diet that these HFpEF patients have can lead to myocardial oxidative stress, which can activate this receptor independent of aldosterone levels. That’s very important, and this was a very sound rationale for this trial.
The study population overall was appropriate; however, I agree with the investigators that this very marked regional variation in event rates at enrolling sites could have affected the results of this trial. One thing that is important to know, as opined in the design paper for this trial, is that enrollment was slow at first … and that necessitated a readjustment of the study design such that the number of patients was reduced and the duration of follow-up was lengthened. That could have been very important for the outcome of this trial. Two previous trials, RALES and EMPHASIS, were stopped early because of the very dramatic effect that was seen early on. In those trials, mean follow-up was 21 or 24 months as opposed to TOPCAT where the mean duration of follow-up was much longer at 40 months. Perhaps related to that, the percent of patients still on the active study drug was much less in TOPCAT than in RALES or EMPHASIS … If they would have had brisker enrollment so that the trial duration could have been shorter and if event rates were more consistent across study sites, it’s very likely that the TOPCAT investigators would have seen a significant reduction in the primary endpoint at their originally planned follow-up of 3 years.
Nonetheless, at least in this HFpEF population, when you think about the global public health impact of what this drug would be, that’s not so compelling because, overall, all-cause hospitalizations and all-cause mortality did not seem to be meaningfully impacted by this drug. We also have to acknowledge that, without the careful monitoring that occurs in a clinical trial, the prevalence of worsening renal function and hyperkalemia would likely be more common in clinical practice.
However, I do think this trial, although not statistically significant, showed a signal of benefit, and I found the subgroup analysis to be very compelling. This suggests that if you use biomarkers such as BNP or ideally a biomarker that identified patients with the highest levels of oxidative stress that this kind of therapy in those types of patients may very well be effective, and this deserves further study.
In the setting of yet another negative trial in HFpEF, should we give up on clinical trials in HFpEF? I'd emphatically say no. It's becoming the most predominant form of HF so we can't give up on this disease. However, the role of the NIH and the Heart Failure Clinical Research Network may need to expand … and I think this trial again emphasizes that we need to start designing trials that are very cognizant of the pathophysiology and focuses the intervention just on those patients with that particular pathophysiology that would be ameliorated by the drug. Given the poor quality of life in this population, we need to focus more on endpoints which reflect patient centered outcomes.