A long-studied steroid fails its first proper trial in ataxia telangiectasia

Design

Phase 3, multicentre, randomised, double-blind, placebo-controlled

Population

105 children with ataxia telangiectasia across nine countries

Primary outcome

Change in ataxia rating scale at six months (ages 6-9)

Effect

Least-squares mean difference -1.30 (95% CI -2.77 to 0.18; p=0.085), not significant

NEAT, a phase 3 double-blind trial across nine countries, randomised 105 children with ataxia telangiectasia to intravenous erythrocyte-encapsulated dexamethasone or placebo every three to four weeks for six doses, with change in a standardised ataxia rating scale at six months as the primary endpoint in those aged 6-9 years.

There was no significant difference: the ataxia score changed by a least-squares mean of -1.30 versus placebo (95% CI -2.77 to 0.18; p=0.085). The drug was well tolerated, with no treatment-related serious adverse events and no adverse effects on growth, bone density or endocrine function.

This matters because encapsulated dexamethasone has been studied in this disease for two decades without proof from a randomised trial, and the result tempers expectations. The authors note six months may be too short to detect a sustained effect in a slowly progressive neurodegenerative disease, so it argues for longer, rigorous trials rather than continued off-trial use.

• Encapsulated dexamethasone did not significantly improve ataxia scores at six months.
• The treatment effect was -1.30 points versus placebo (95% CI -2.77 to 0.18; p=0.085).
• The drug was safe, with no treatment-related serious adverse events.
• A two-decade-old therapy now lacks randomised proof of benefit.
• Six months may be too short; longer rigorous trials are needed.

Why it matters

It holds a long-used but unproven therapy to a randomised standard, and it did not pass.

Don't overread it

A single six-month trial in a slowly progressive disease; absence of a significant effect here does not exclude a longer-term benefit.

The statistics, in plain English

A p value of 0.085 and a confidence interval crossing zero mean the result is not statistically significant; it does not prove the drug useless, but it provides no evidence of benefit at this timepoint.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.