And It’s Back to the Drawing Board for PDE9 Inhibition in Heart Failure

PHOENIX — The development of a PDE9 inhibitor for heart failure hit a major roadblock, as two phase II placebo-controlled trials were suspected of having tested the wrong dose.

In both heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF), tovinontrine failed to meet expectations for effectiveness regarding improvement in NT-proBNP at week 12, reported James Udelson, MD, of Tufts Medical Center in Boston, at the Heart Failure Society of America (HFSA) annual meeting.

CYCLE-1 trial showed that in HFrEF, tovinontrine brought NT-proBNP down by a nonsignificant placebo-adjusted average 4.4% with the high dose (50 mg BID, P=0.45) — the main dose chosen for the study — whereas there was a more promising drop of 10.8% with the medium dose (25 mg BID, P=0.05).

The high dose again failed in CYCLE-2’s HFpEF cohort: NT-proBNP at week 12 went the wrong direction with tovinontrine 50 mg BID (-1.7% vs -10.1% with placebo, placebo-adjusted mean change +9.4%).

“Did we have the wrong dose? Yes, that’s pretty clear,” said Udelson.

But it was not over for tovinontrine, since there was evidence of an inverse dose response in CYCLE-1. Median change in NT-proBNP from baseline was most favorable with low-dose tovinontrine (-16% with dose of 2.5 mg BID) and worse with increasing doses (-6% with 25 mg BID, -1% with 50 mg BID).

Tovinontrine thus has another shot in the CYCLE-3 trial set to start in 2027. That phase IIb study will recruit in HFrEF with the hypothesis that a lower tovinontrine dose, causing less intense PDE9 inhibition and less downstream cGMP elevation, may result in greater reduction in NT-proBNP, Udelson told the audience at HFSA.

If that goes well, the investigators will proceed with testing in HFpEF.

The interest in PDE9 inhibition stems from observations that PDE9 expression is increased in heart failure, particularly in HFpEF. Recently, the CARDINAL-HF trial provided proof of mechanism in 60 HFrEF patients that a PDE9 inhibitor does raise cGMP elevation, with or without sacubitril/valsartan (Entresto) use, and with no major safety concerns.

Based on the present data, Udelson said there is a suspicion that there is a ceiling for how much myocardial cGMP can be elevated before other mechanisms kick in to prevent further NT-proBNP reduction. Indeed, in a post hoc analysis, tovinontrine recipients with small cGMP increases had more consistent NT-proBNP reductions in all dose levels. “This analysis supports that modest increases in plasma cGMP with tovinontrine may lead to greater reductions in NT-proBNP versus placebo,” according to Udelson.

CYCLE-1 was an international trial in people with ejection fraction 40% or below who had NT-proBNP at least 600 pg/mL at baseline, showed an estimated glomerular filtration rate (eGFR) of 30 mL/min/1.73 m² or better, were on stable guideline-directed medical therapy (GDMT), and had New York Heart Association class II or III symptoms.

Investigators had 529 eligible people randomized to tovinontrine at various doses (2.5 mg to 50 mg BID) versus placebo. This cohort averaged age 69 and was 78% men and 94% white. Most were on triple GDMT (88%) or quadruple GDMT (60%), with 62% on sacubitril/valsartan.

Across tovinontrine doses, plasma cGMP increased from baseline to week 4 and week 12; the placebo group had a drop in cGMP, according to Udelson.

CYCLE-2 had similar enrollment criteria as CYCLE-1 but sought HFpEF patients with ejection fractions >40% and <60%.

This came out to 277 patients randomized to tovinontrine 50 mg BID or placebo for this study. Mean age here was 72, 64% were men, and 92% white. GDMT at baseline was dual-agent for most people (76%) while a minority had stable triple therapy (42%); 31% were on sacubitril/valsartan at baseline.

As for safety in CYCLE-1 and CYCLE-2, there was some excess in treatment-emergent adverse events in the tovinontrine groups, mostly mild events, with minimal changes in blood pressure at 12 weeks. Renal changes were transient and reversible upon treatment discontinuation without additional treatment, according to Udelson.

There is no PDE9 inhibitor available in clinical practice to date.