MINNEAPOLIS — The investigational monoclonal antibody ponsegromab, which has proved promising in the oncology space, appeared to fall flat for the treatment of people with heart failure in a phase 2 trial.

Not only did the drug fail to meet its primary outcomes of improved health status or physical activity in this patient population, as compared with placebo but it was also associated with an unexpected negative consequence.

“The surprising finding was that neutralizing GDF-15 [growth differentiation factor-15] with ponsegromab in this population actually increased the risk of worsening heart failure events by more than twofold,” study investigator David Berg, MD, MPH, cardiologist at Brigham and Women’s Hospital and assistant professor of medicine at Harvard Medical School in Boston, told Medscape Medical News.

Berg and colleagues conducted an international, multicenter, proof-of-concept study, dubbed GARDEN-TIMI 74, to see if ponsegromab could lower levels of GDF-15 in a heart failure population based on its success in treating people with cancer. Higher circulating levels of this well-established biomarker are associated with higher heart failure symptom burden, greater physical impairment, and increased risk for adverse heart failure outcomes. 

The cumulative incidence of a composite endpoint of cardiovascular death and worsening heart failure was 19.8% in the ponsegromab cohort vs 11.6% in the placebo group by 22 weeks (hazard ratio [HR], 2.24; 90% CI, 1.43-3.50). This finding was driven primarily by worsening heart failure events, reported in 19.3% of the treatment group vs 9.1% of the placebo participants (HR, 2.84; 90% CI, 1.74-4.64).

Berg presented the findings of this phase 2 trial during a late-breaking research session at the Heart Failure Society of America (HFSA) 2025 Annual Scientific Meeting.

Some Caveats

“This is a very important trial — a really well-conducted trial of an important question,” said Christopher O’Connor, MD, president of the Inova Schar Heart and Vascular in Fairfax, Virginia, who was invited to discuss the results. However, he raised some concerns about the design, including the choice of dose groups, suggesting that a wider range would have been preferable.

Researchers selected the primary dose of 300 mg, administered subcutaneously every 4 weeks, based on pharmacokinetic and pharmacodynamic modeling with the aim of completely neutralizing GDF-15, which was achieved with this dose, Berg said in a follow-up interview. The investigators included 100 mg and 200 mg doses to assess pharmacokinetics and pharmacodynamics.

“Had our proof-of-concept trial been positive, the next step would have been to do a full dose-ranging study,” Berg added.

O’Connor proposed that a sicker cohort might have benefitted more from the therapy. Berg noted, however, that the study participants did have fairly advanced disease.

“For an ambulatory heart failure population, the NT-proBNP [N-terminal pro-B-type natriuretic peptide] levels of enrolled patients were actually quite high, reflecting a more advanced heart failure population than is typically enrolled in clinical trials,” Berg said.

He also highlighted participants’ median qualifying Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS) score of 48, on a scale of 0 to 100, where higher numbers correspond to better health.

“This number is much lower than we typically see in heart failure trials, reflecting the highly symptomatic population of advanced heart failure patients in the study,” Berg said.

Bold Ideas and Borrowed Information

Previous studies of ponsegromab showed that patients with cancer cachexia and elevated GDF-15 gained weight and experienced increased physical activity. Even so, O’Connor said, “This was a pretty risky thing to do — to borrow information from another disease space.”

“Our trial acutely highlights that we must keep striving to help these patients, who are very much in need of new options,” Berg said. “I agree with what Dr O’Connor said in his commentary, which is that we should continue to bring creative and bold ideas to the table — sometimes borrowed from other disease states — like we did in this trial.”

Researchers terminated GARDEN-TIMI 74 early during a planned interim analysis in November 2024. At the time, Pfizer, the study sponsor, concluded the trial was unlikely to meet its objectives.

Key Findings

For the trial, Berg and colleagues randomly assigned 433 people with heart failure to one of the three dosage groups. Twenty-one people received 100 mg, 17 people received 200 mg, and 197 received 300 mg ponsegromab via injection once every 4 weeks. Another 198 received placebo injections.

Participants were enrolled between September 2022 and August 2024 at 115 sites in 11 countries. All had evidence of fatigue, cachexia, or functional impairment. Median age was 75 years, median BMI was 25, and 29% met criteria for cachexia. All participants also had a left ventricular ejection fraction below 50%, with a median value of 35%.

Seventy-eight participants were taking an SGLT2 inhibitor, and 47% were on quadruple therapy at baseline. The median NT-proBNP level was 2156 pg/mL, and the median GDF-15 level was 3958 pg/mL.

Ponsegromab successfully suppressed more than 96% of GDF-15 levels across all doses and time points. In addition, expected weight gain occurred, with a median increase of 1.4 kg from baseline to week 22.

Compared with the placebo, researchers found no significant differences in change in KCCQ-CSS measures across any treatment groups from baseline to week 22. For example, scores only increased 0.31 points in the 300-mg cohort during that period.

At the same time, they observed no statistically significant differences in improvements in 6-minute walk distance between the 300-mg treatment group and the placebo group.

In terms of biomarkers, NT-proBNP levels were almost identical at baseline (2110 pg/mL in the ponsegromab group and 2119 pg/mL in the placebo group). By 12 weeks, levels increased in the treatment group and decreased in the placebo group for a ponsegromab-to-placebo ratio of 1.13. Levels in both groups decreased by 22 weeks (ponsegromab-to-placebo ratio, 1.11).

The inflammation marker high-sensitivity C-reactive protein decreased in the placebo cohort and increased in the ponsegromab cohort. At week 22, the treatment-to-placebo ratio was 1.53, suggesting an increasing inflammation burden over this period with the monoclonal antibody treatment.

‘Fundamental Questions’

The trial “raises fundamental questions about the role of GDF-15 in heart failure,” Berg said. He noted the possibility of a direct protective role for GDF-15 in people with heart failure, in which case neutralizing GDF-15 with ponsegromab may have disrupted an important compensatory response. It is also feasible that the increased heart failure risk in GARDEN-TIMI 74 was related to metabolic dysregulation or the proinflammatory effects of GDF-15 neutralization.

“This trial underscores the importance of conducting rigorous phase 2 trials to fully characterize the impact of novel therapies, including potential safety signals, before diving into much larger phase 3 trials,” Berg said.

The GARDEN-TIMI 74 trial was supported by a grant from Pfizer to Brigham and Women’s Hospital. Berg reported receiving research grants and consulting fees/honoraria from Pfizer. O’Connor reported having no conflicts of interest.