FINEARTS-HF: New Analyses Ask, Who Benefits Most From Finerenone?

Two new analyses of the FINEARTS-HF trial may help physicians decide how, and for whom, finerenone can best be used to treat heart failure with mildly reduced or preserved ejection fraction (HFmrEF/HFpEF) .

Both studies, scheduled as late-breaking trials at the Heart Failure Society of America (HFSA) 2024 scientific session, were presented online after the meeting’s last-minute cancellation due to Hurricane Helene.

In the first analysis, which was presented by senior author John J. McMurray, MD, PhD (University of Glasgow, Scotland), investigators showed the nonsteroidal mineralocorticoid receptor antagonist (MRA) reduced the risk of total HF events or cardiovascular mortality—the study’s primary endpoint—regardless of baseline LVEF.

For the second analysis, which was led by Akshay Desai, MD (Brigham and Women’s Hospital, Boston, MA), investigators provided evidence that the benefit of finerenone (Kerendia; Bayer AG) may be strongest in those with a more recent worsening HF event. While investigators saw a trend toward greater absolute risk reductions in the study’s primary endpoint in the group with a recent worsening HF event, the relative reduction in risk did not hold up in formal interaction testing.

“I think it is important in interpreting these data to highlight that while provocative and hypothesis-generating, they do not conclusively suggest that the benefit of finerenone is confined to those with a recent—within 3 months—worsening HF event,” Desai told TCTMD. “As with all subgroup analyses, the primary trial result should rule unless there is compelling evidence of a treatment effect by subgroup interaction, and there was no such interaction identified here.”

For now, Desai advised that clinicians should not use the FINEARTS-HF subgroup analysis to exclude eligible patients from finerenone, but rather “to provide reassurance that those high-risk patients with recent worsening heart failure benefit at least as much, and perhaps more, than those remote from a worsening HF event or without a prior worsening HF event without a concerning signal of differential safety.”

The findings, he added, support current guidelines that encourage physicians to “leverage” episodes of worsening heart failure, and heart failure hospitalizations in particular, as a good time to optimize treatment of HFmrEF/HFpEF, just as is done for HFrEF.

As with all subgroup analyses, the primary trial result should rule. Akshay Desai

John Teerlink, MD (University of California, San Francisco), the discussant following the HFSA presentations, also stressed the need to be cautious when interpreting interaction effects in a subgroup analysis.

The finding of greater benefit in those with a more recent worsening HF event is “intriguing,” he said, especially since there was no treatment effect in those who had a worsening HF event more than 3 months out or in those with no prior worsening HF event at all. 

“Do these results suggest this subgroup of patients should not receive finerenone?” he asked. “Until additional data emerges, I believe that in the absence of any mortality benefit overall, and the apparent absence or minimal benefit in this subgroup of patients, there is not a compelling need to initiate finerenone in patients who have no recent worsening heart failure event.”

Like Desai, though, Teerlink believes the more important message from the FINEARTS-HF analysis is to start finerenone in hospital or within a week of a worsening HF event in HFmrEF/HFpEF patients, because doing so “significantly improves the care and outcomes of our patients.”   

Proximal to Worsening HF

In FINEARTS-HF, whose main results were recently reported by TCTMD, use of finerenone reduced the relative risk of total worsening HF events or death from cardiovascular causes by 16% when compared with placebo in 6,001 patients with symptomatic HF and LVEF 40% or greater (plus elevated levels of natriuretic peptides and evidence of structural heart disease). The benefit was driven by a reduction in worsening HF events, with no evidence of a reduction in cardiovascular mortality.

In the trial, explained Desai, enrollment included hospitalized patients and those with a recent worsening HF event. Of those randomized, 20.3% were enrolled during or within 7 days of a recent worsening HF event, 33.8% were enrolled with an event occurring between 7 days and 3 months, and 15.6% with an event more than 3 months ago. Overall, 30.3% had no history of a worsening HF event.

“It was a very intentional enrollment of a hospitalized, or recently hospitalized, population,” said Desai.

Patients randomized closer to a worsening HF event tended to be sicker, with higher baseline rates of atrial fibrillation and NT-proBNP levels, lower LVEF and estimated glomerular filtration rate (eGFR), and worse NYHA functional class. In the analysis, which was published simultaneously in the Journal of the American College of Cardiology, clinical event rates were highest in those randomized to treatment within 7 days of a worsening HF event—roughly twofold higher when compared with those randomized 3 months out from a worsening HF event.

Finerenone compared with placebo reduced the primary endpoint by 26% (RR 0.74; 95% CI 0.57-0.95) in those with a recent worsening HF event (< 7 days) and by 21% (RR 0.79; 95% CI 0.64-0.97) in those with a worsening event between 7 days and 3 months. In contrast, there was no benefit to finerenone versus placebo in those with a worsening HF event more than 3 months out from randomization or in those enrolled without worsening event (RR 0.99; 95% CI 0.81-1.21).

Formal interaction testing, however, looking at the treatment effect of finerenone based on time from the worsening HF event was not statistically significant (P = 0.07 for interaction).

Despite the “borderline” interaction, “it is important to recognize that this P value is not corrected for multiple testing in the context of the many possible interactions examined,” said Desai. “As well, the pattern of waning efficacy with greater distance from the most recent worsening heart failure event was not as apparent in examination of the time-to-first occurrence of cardiovascular death or worsening heart failure event, which is the traditional trial composite endpoint.”

Additionally, there was a suggestion of higher rates of discontinuation of finerenone compared with placebo in those more than 3 months out from a worsening HF event, which might have undermined the treatment benefit, said Desai.

Absolute reductions in the primary endpoint were greatest for those randomized proximal to a worsening HF event: 7.8% lower versus placebo in those with a worsening HF event within 7 days, 4.6% lower in those with worsening event between 7 days and 3 months, and 0.1% lower in those with worsening event greater than 3 months out or no worsening HF event (P = 0.011 for trend).

There was no difference in the risk of adverse events across the groups of patients stratified by time from a worsening HF event to randomization.

Benefit Seen Across LVEF Spectrum

The second FINEARTS-HF analysis, which was published in Circulation, looked at the treatment effect with finerenone across the spectrum of LVEF.

“There has been a suggestion in previous trials, particularly with neurohormonal modulating therapies, that efficacy might vary according to ejection fraction, with possible attenuation of the benefit of treatments in patients with higher ejection fractions,” said McMurray. “That was most clearly and recently seen with sacubitril/valsartan.”

In contrast, the benefits of SGLT2 inhibitors are consistent across the entire spectrum of LVEF, investigators note.

The analysis stratified the FINEARTS-HF population into three groups based on LVEF: < 50% (n = 2,172), ≥ 50% to < 60% (n = 2,674), and ≥ 60% (n = 1,147). Those with a lower LVEF tended to be older, were more commonly female, had worse quality-of-life scores, and had higher mean body mass index, systolic blood pressure, and a greater prevalence of kidney dysfunction. They were also more likely to have a history of MI or prior HF hospitalization and had higher NT-proBNP concentrations.   

McMurray said they observed no difference in the treatment effects by LVEF category for the primary endpoint, as well as none for the endpoints assessed individually.

“In all of the endpoints in each of the subgroups, there was no significant heterogeneity,” he said. “Generally, the point estimates line up and the interaction P values were clearly nonsignificant.” 

Safety endpoints did not differ significantly across the LVEF groups.

One caveat of the analysis is that there were few patients and clinical events in patients with LVEF > 70%, but further studies are planned, including REDEFINE-HF and CONFIRMATION-HF, which should provide more clarity in HF patients with LVEF at this upper range, said McMurray.

“Although it represents a small group of patients, it will be interesting to see the clinical course and treatment of finerenone in people with improved or recovered ejection fractions,” said Teerlink. He noted that a soon to-be-released consensus statement from the HFSA, Heart Failure Association of the European Society of Cardiology, and Japanese Heart Failure Society will advocate for starting some HF therapies, such as SGLT2 inhibitors, without knowledge of LVEF in patients with a worsening HF event.

“We don’t need to wait for the echocardiogram results, necessarily,” said Teerlink. “FINEARTS reinforces this potential role of MRAs across the ejection fraction spectrum, with the continuing need to monitor for hyperkalemia, worsening kidney function, and hypotension.”