Givinostat slowed, but did not stop, motor decline in Duchenne muscular dystrophy

The Food and Drug Administration approved givinostat, an oral drug that acts on muscle biology rather than on the dystrophin gene itself, for Duchenne muscular dystrophy. The approval was recorded in the agency's Drugs@FDA database on 21 March 2024 under new drug application 217865, held by Italfarmaco [s2]. The product, brand name Duvyzat, is an oral suspension of givinostat hydrochloride [s2].

Duchenne muscular dystrophy is the most common childhood muscular dystrophy and is caused by a deficiency of dystrophin [s1]. Givinostat is a histone deacetylase inhibitor; preclinical and phase 2 data had suggested it might help counteract the downstream effects of that deficiency [s1]. Unlike the gene therapies and exon-skipping drugs that target the genetic defect, givinostat works further downstream, on the muscle tissue the disease damages.

What the pivotal trial tested

The approval rests on EPIDYS, a phase 3, double-blind, placebo-controlled trial run at 41 tertiary care sites in 11 countries and published in The Lancet Neurology [s1]. Eligible participants were ambulant, male and aged at least 6 years, with a genetically confirmed diagnosis and at least six months of systemic corticosteroids behind them [s1]. Boys were randomly assigned 2:1 to oral givinostat or matching placebo twice a day for 72 weeks [s1].

The trial split participants by how much fat had already replaced muscle in the thigh — the baseline vastus lateralis fat fraction, measured by magnetic resonance spectroscopy. Group A comprised boys with a fraction above 5% but no more than 30%, while group B comprised those with a fraction of 5% or less, or more than 30% [s1]. The primary endpoint was measured in group A: the change in the four-stair climb assessment between baseline and 72 weeks, analysed by intention to treat [s1]. The dose itself was flexible, set by weight and reduced if a boy could not tolerate it, and treatment assignment was masked to the boys, investigators, and site and sponsor staff [s1].

Two features of the design deserve note. An interim futility assessment was run once the first 50 group A boys reached 12 months, after which the sample size was adapted using masked data; and the starting dose of givinostat was reduced following a protocol amendment [s1].

What it found

Between 6 June 2017 and 22 February 2022, 359 boys were assessed and 179 enrolled, with a median age of 9.8 years [s1]. All were randomised — 118 to givinostat, 61 to placebo — and 170 (95%) completed the study [s1]. Of the 179, 120 (67%) fell into group A, of whom 114 (95%) completed [s1].

On the primary endpoint, the four-stair climb time worsened in both arms over 72 weeks — but by less in the treated group. The geometric least-squares mean ratio was 1.27 (95% CI 1.17–1.37) with givinostat and 1.48 (1.32–1.66) with placebo, giving a between-group ratio of 0.86 (95% CI 0.745–0.989; p=0.035) [s1]. A ratio above 1 means the task took longer than at baseline, so both groups declined; the treated boys simply declined more slowly.

On safety, the most common adverse events with givinostat were diarrhoea, in 43 (36%) of 118 boys versus 11 (18%) of 61 on placebo, and vomiting, in 34 (29%) versus 8 (13%) [s1]. No treatment-related deaths occurred [s1].

How to read it

The honest framing is that givinostat slowed, rather than halted, the loss of motor function over the trial year [s1]. The effect was statistically significant but modest, and the result turns on a single primary endpoint in one subgroup of the trial, measured over 72 weeks. Whether a smaller decline on a timed stair-climb translates into differences families notice in daily life — and whether the benefit persists over the many years the disease is managed — is not something a 72-week trial can answer [s1].

The trial was funded by the drug's manufacturer, Italfarmaco [s1]. The dose was reduced after the interim safety analysis, and the investigators reported no new safety signals thereafter [s1]. An extension study is evaluating long-term safety and efficacy [s1].

What to watch

The near-term questions are practical: which patients the prescribing information covers, how givinostat sits alongside corticosteroids and the dystrophin-directed therapies, and whether the gastrointestinal effects seen in the trial shape real-world tolerability. Because the drug acts on muscle biology rather than the genetic defect, it is in principle agnostic to which dystrophin mutation a boy carries — a contrast with the exon-skipping drugs, which are tied to specific mutations [s1]. The extension data will decide whether the modest slowing holds, grows or fades with time [s1].

This article describes research and regulatory news and is not medical advice. Decisions about Duchenne treatment are for patients, families and their treating clinicians.