An anti-inflammatory drug failed to ease symptoms in a common heart failure

A drug designed to dampen inflammation inside the failing heart did not improve symptoms or exercise capacity in a phase 2b trial published in Nature Medicine [s1]. Mitiperstat, an inhibitor of the enzyme myeloperoxidase, missed both of its co-primary goals in patients with heart failure and a preserved or only mildly reduced ejection fraction [s1].

The result matters because this form of heart failure — where the heart pumps with a normal or near-normal ejection fraction but stiffens and fills poorly — is common, disabling and short on treatments that make patients feel better. A drug attacking a genuinely new target was a plausible candidate, and it did not work.

The idea being tested

Myeloperoxidase (MPO) is an enzyme released by certain white blood cells. Its oxidant products reduce the availability of nitric oxide and are implicated in coronary microvascular dysfunction, stiffening of heart-muscle cells and interstitial scarring — mechanisms thought to drive heart failure with preserved and mildly reduced ejection fraction [s1]. Mitiperstat, an oral MPO inhibitor also known in development as AZD4831, was intended to interrupt that cascade [s3].

The trial was industry-sponsored: its manufacturer, AstraZeneca, is the registered lead sponsor of the study [s3]. That is worth stating plainly, because a negative result reported by the company that stood to benefit from a positive one is, if anything, a mark in favour of the finding's reliability.

What the trial did

This was a multicentre, randomised, double-blind, placebo-controlled, three-arm, parallel-group phase 2b trial in patients with heart failure and an ejection fraction above 40% [s1]. In total, 711 patients, 45% of them women, were randomised in equal proportions to mitiperstat 2.5 mg, mitiperstat 5 mg or placebo, and treated for 48 weeks [s1].

The two co-primary endpoints, assessed at 16 weeks, were the Kansas City Cardiomyopathy Questionnaire Total Symptom Score (KCCQ-TSS), a validated measure of heart-failure symptoms, and the six-minute walk distance (6MWD), a physical measure of exercise capacity [s1]. Secondary endpoints, followed for up to 48 weeks, included natriuretic peptides, inflammatory markers and echocardiographic measures of heart structure [s1].

What it found

With the two mitiperstat doses pooled, neither co-primary endpoint improved versus placebo. The placebo-corrected difference in KCCQ-TSS was −1.4 points (95% confidence interval −3.9 to 1.2; P=0.29), and the difference in six-minute walk distance was 3.8 metres (95% CI −3.1 to 10.8; P=0.28) [s1]. Both confidence intervals straddle zero, and neither reaches anything a patient would notice. No secondary endpoint improved either [s1].

On safety, adverse and serious adverse events, including infections, were similar across groups, with one exception: maculopapular rash occurred in 3.6% of mitiperstat-treated patients versus 0.4% on placebo [s1]. The authors concluded that mitiperstat was safe and well tolerated but did not improve symptoms or exercise function in this population [s1].

How to read it

This is a clean negative trial, and its design gives the reader confidence in that verdict. The endpoints were patient-centred — how people feel and how far they can walk — rather than surrogate laboratory markers, and the trial was double-blind and placebo-controlled at a reasonable size for phase 2b [s1]. When such a trial shows a flat result on both co-primary measures, the most parsimonious conclusion is that the drug, at these doses, does not help.

The finding also fits a wider pattern for this compound. In a separate phase 2a trial in metabolic dysfunction-associated steatohepatitis (MASH), mitiperstat 5 mg produced no significant change versus placebo in the liver-enzyme and fibrosis markers it targeted at 12 weeks (ALT +7.5%, P=0.893; the collagen marker Pro-C3 −0.9%, P=0.396), and the authors concluded the study did not provide proof of concept [s2]. Two negative readouts in two different diseases weaken the broader case that inhibiting myeloperoxidase translates into clinical benefit, at least so far.

Caution cuts the other way too. This is a phase 2b trial, not the final word: it tested particular doses over a defined window, and a null result does not prove the biological target is irrelevant, only that this drug at these doses did not move these outcomes [s1]. Heart failure with preserved ejection fraction has a long history of promising mechanisms that faltered in trials, which is part of why each negative result is worth recording rather than quietly shelving.

Why it matters

For patients with this common and stubborn form of heart failure, the treatment cupboard remains thin, and a candidate with a novel mechanism has come up empty on the outcomes that matter to them. Publishing the negative result — rather than burying it — is how the field avoids repeating the same bet.

What to watch

The near-term question is whether AstraZeneca continues mitiperstat's development in its other programmes, including chronic obstructive pulmonary disease, after two negative phase 2 readouts, and whether any biomarker signal justifies a different dose or population [s1][s2].

This article describes trial results and is not medical advice. Treatment decisions are for patients and their clinicians.

Sources

• Myeloperoxidase inhibition with mitiperstat in heart failure with preserved or mildly reduced ejection fraction: a randomized phase 2b trial — Nature Medicine, 9 September 2026 (primary results)
• A Phase 2a Trial of Myeloperoxidase Inhibitor Mitiperstat in Non-Cirrhotic MASH — Liver International, 18 October 2025 (prior negative readout)
• Study to Evaluate the Efficacy and Safety of AZD4831 (NCT04986202) — ClinicalTrials.gov (trial registration and sponsor)