An old blood-pressure drug slowed a fatal childhood brain disease

Vanishing white matter is one of the more devastating inherited brain diseases, usually striking children between the ages of one and six, stripping away the ability to walk, and causing early death with no treatment that changes its course [s1]. A phase 1/2 trial in the Lancet Neurology reports that guanabenz, a decades-old blood-pressure drug, may slow that decline — a genuinely hopeful signal, tempered by a study design that cannot prove it on its own [s1].

The rationale

The disease is caused by a genetic defect in eukaryotic initiation factor 2B, a central regulator of what cells do under stress — the integrated stress response [s1]. Guanabenz, an α2-adrenergic antihypertensive, happens to inhibit that stress response, and it had shown benefit in a mouse model of the disease [s1]. That is the chain of reasoning behind repurposing it: a cheap, long-available drug aimed at a newly understood mechanism in a condition with nothing else to offer.

What they did

The trial was single-arm — every participant received guanabenz, with no concurrent placebo group [s1]. It was run from Amsterdam University Medical Center with international recruitment [s1]. Eligible children had vanishing white matter confirmed by MRI and genetic testing, were aged six years or younger at onset, had a disease duration of eight years or less, and could still walk at least ten steps with, at most, light support from one hand [s1]. Oral guanabenz was started at 0.15 mg/kg per day and titrated over roughly six weeks toward a target of 2 mg/kg per day [s1]. The primary efficacy outcome was the time to loss of walking with support — a concrete, meaningful milestone in a disease defined by progressive motor decline [s1]. Because there was no randomised comparison group, each treated child was matched 1:2 to untreated historical controls from the Vanishing White Matter Registry, on similar age of onset and similar disability at the same disease duration [s1] [s2].

What it showed

Between May 2021 and May 2024, 33 patients were screened, found eligible, and enrolled, of whom 31 completed the trial [s1]. The 33 treated children had a significantly lower risk of losing the ability to walk with support than the 66 historical controls, with a hazard ratio of 0.33 (95% confidence interval, 0.16 to 0.69; log-rank p=0.0061) — roughly a two-thirds lower rate of that milestone over the period studied [s1]. No life-threatening events or deaths occurred during the trial [s1].

The safety cost

Guanabenz was far from side-effect-free, especially at the start. There were 63 serious adverse events in 25 (76%) of the 33 patients, and 30 (48%) of those events were judged likely or very likely related to the drug [s1]. Most striking were hallucinations: 24 drug-related reactions were hallucinations, occurring in 18 (55%) of the 33 children [s1]. These came intermittently, mostly within the first four months of treatment, and mostly resolved within four months of first appearing [s1]. After that initial phase, the trial reports, guanabenz was well tolerated and no patient stopped because of side effects [s1].

How much to trust it

The central caveat is built into the design, and the investigators say so plainly: this was a non-randomised comparison against historical controls, not a controlled trial [s1]. Matching on age of onset and disability cannot rule out that treated children differed in other, unmeasured ways from registry patients recorded in earlier years, and expectations on the part of families and clinicians can shape an open, unblinded outcome like walking. The number of children is also small — 33 — as is inevitable in so rare a disease. A hazard ratio of 0.33 is a large effect, but the confidence interval reaches to 0.69, and the whole estimate rests on a comparison the authors caution should be confirmed in a long-term extension study [s1].

Set against those limits is the fact that this is an academic, publicly and charitably funded effort — backed by the Dutch research organisation ZonMw, the Dutch Brain Foundation, the European Leukodystrophy Association and the VWM Families Foundation [s1] — pursuing a repurposed generic drug rather than a commercial product.

What to watch

The disease-modifying effect now needs confirmation in the planned long-term extension, and ideally in a design that does not depend on historical controls [s1]. The early hallucination signal will also need careful management if the drug moves toward wider use. For families facing a diagnosis that has offered no options at all, the result is meaningful — but it is a first, uncontrolled signal, not a settled treatment.

Sources

• [s1] Safety and efficacy of guanabenz in early-childhood onset vanishing white matter: primary analysis of a single-arm, phase 1/2 trial. The Lancet Neurology. 12 August 2026.
• [s2] EU Clinical Trials Register. Guanabenz in vanishing white matter (EudraCT 2017-001438-25).