The first approved drug for idiopathic hypersomnia, and how to read its trial

Until August 2021 no medication had regulatory approval anywhere for idiopathic hypersomnia, a central disorder of overwhelming daytime sleepiness, and clinicians borrowed drugs from narcolepsy off-label [s1]. That month lower-sodium oxybate became the first drug the US Food and Drug Administration approved specifically for the condition, on the strength of a single phase 3 trial [s1] — a trial whose design decides how much its striking headline number is actually worth.

What idiopathic hypersomnia is

Idiopathic hypersomnia is characterised mainly by excessive daytime sleepiness, often with prolonged night-time sleep and pronounced sleep inertia — the heavy, disoriented grogginess on waking that patients describe as the hardest part to explain to anyone who has not lived it [s1]. "Idiopathic" is doing real work in the name: unlike narcolepsy type 1, it has no established biomarker, and it is diagnosed by the pattern of symptoms and sleep testing rather than by a specific lesion or deficiency. That absence of a mechanism is also why its drug cupboard was, for decades, bare.

The guideline that mapped a thin evidence base

How thin was visible in the American Academy of Sleep Medicine's 2021 clinical practice guideline for central disorders of hypersomnolence, which graded each option "strong" (a "we recommend") or "conditional" (a "we suggest") using the GRADE method, with every intervention compared to no treatment [s2]. For idiopathic hypersomnia in adults, only one drug earned a strong recommendation — modafinil — while clarithromycin, methylphenidate, pitolisant, and sodium oxybate were all conditional suggestions [s2]. A single strong recommendation, and it for a wake-promoting agent first approved for narcolepsy, is a fair summary of where treatment stood when this trial reported.

How the trial was built — and why the design matters

The study enrolled adults aged 18–75 with idiopathic hypersomnia at 50 specialist sleep centres across six European countries and the United States [s1]. It used a randomised-withdrawal design, and understanding that design is the whole of reading the result honestly.

First came an open-label run-in: every participant took lower-sodium oxybate — an oral solution taken once or twice nightly, individually titrated over 10–14 weeks, dose range 2.5–9.0 g/night — followed by a two-week stable-dose stretch [s1]. Only then were participants who had reached a stable dose randomised 1:1 to either continue the drug or switch to a matched placebo for a two-week, double-blind withdrawal period, with the primary outcome being the change in Epworth Sleepiness Scale (ESS) score across that blinded window [s1].

The open-label numbers look dramatic: across the 154 people in the safety population, mean ESS fell from 15.7 (SD 3.8) at baseline to 6.1 (4.0) by the end of the stable-dose period [s1]. That is a large drop, but it is uncontrolled — no placebo group, and expectation, regression to the mean, and the attention of a trial all push in the same direction. It cannot be read as the drug's effect.

The controlled comparison is the withdrawal. Of the 115 participants randomised — 59 to placebo, 56 to continue the drug — those switched to placebo saw their sleepiness return, while those who stayed on lower-sodium oxybate held steady, a least-squares mean difference of −6.5 points on the ESS (95% CI −8.0 to −5.0; p < 0.0001) [s1]. That gap is what the approval rests on.

What a randomised-withdrawal design does and does not show

This design answers a narrow, honest question: among people who tolerated the drug and stabilised on it, does stopping make them worse? It cannot tell you how the average newly diagnosed patient will fare, because everyone who did not tolerate or respond during the open-label months never reached randomisation — an enrichment that inflates the apparent effect relative to a conventional parallel-group trial. It measures loss on withdrawal, not gain on starting. And the blinded window was two weeks; durability beyond that is not something these data establish.

The safety ledger

Lower-sodium oxybate is a salt-reduced formulation of sodium oxybate, itself a controlled substance with a restricted distribution programme in narcolepsy, so tolerability is not incidental. In the safety population the treatment-emergent adverse events were nausea (34 of 154, 22%), headache (27, 18%), dizziness (19, 12%), anxiety (17, 11%), and vomiting (17, 11%); no deaths were reported during the study [s1]. The trial was funded by the drug's manufacturer, Jazz Pharmaceuticals, a fact that belongs next to any single-trial approval [s1].

Where this leaves patients

The result is genuine and, for a condition that had nothing formally approved, consequential — but it is one industry-funded trial, in an enriched population, over a short blinded window, for a controlled drug that a fifth of users found nauseating [s1]. That idiopathic hypersomnia sits close to narcolepsy is worth keeping in view; the two are distinguished by testing, and the mortality signals seen in narcolepsy cohorts, covered in our analysis of a US veterans study, do not transfer automatically. What to watch is longer-term effectiveness data and any head-to-head comparison against the modafinil that the guideline still recommends first [s2]. This article describes trial and guideline evidence and is not medical advice.

Sources

• Safety and efficacy of lower-sodium oxybate in adults with idiopathic hypersomnia: a phase 3, placebo-controlled, double-blind, randomised withdrawal study — The Lancet Neurology, 2021-12-20
• Treatment of central disorders of hypersomnolence: an American Academy of Sleep Medicine clinical practice guideline — Journal of Clinical Sleep Medicine, 2021-09-01