Narcolepsy type 1 is one of the cleaner mysteries in neurology, because for once we know exactly what has gone wrong. Somewhere in the hypothalamus, a small population of neurons that make a wakefulness chemical called orexin (also written hypocretin) has been destroyed, most likely by the patient’s own immune system. Lose those cells and the brain can no longer hold the boundary between sleep and waking: people collapse into sleep during the day, and strong emotion can drop their muscle tone without warning, the symptom clinicians call cataplexy. The deficit is specific, measurable in spinal fluid, and has been understood for a quarter century. What has been missing all that time is a drug that walks up to the same receptor those dead neurons used to signal and switches it on. As of this week there is one, and the road it took to market ran straight through a liver ward.

On August 5, the FDA approved Takeda’s oveporexton, brand name Orzeyful, for narcolepsy type 1 in adults. It is the first orexin receptor 2 agonist to reach the market, and that mechanism is the whole event. Every other narcolepsy medicine on the shelf works downstream of the actual lesion: the amphetamine-class stimulants and modafinil push wakefulness by other chemistry, pitolisant and solriamfetol prod adjacent systems, sodium oxybate consolidates night sleep. Each patches a symptom the missing orexin was supposed to prevent. Oveporexton aims at the receptor the disease switched off. That is a genuinely different place to stand.

To see why the approval is an event and not just a press release, go back to 2021, because this is the second time Takeda has walked an oral orexin agonist into late-stage trials, and the first time ended in a hospital. The predecessor molecule, TAK-994, worked. It also poisoned livers. In October 2021 Takeda halted both Phase 2 studies early after 8 patients blew past the trial’s liver-enzyme safety thresholds, and 3 of them met what hepatologists call Hy’s law, the pattern that predicts a real risk of fatal drug-induced liver failure. The efficacy data, later published in the New England Journal of Medicine, were striking and beside the point, because a drug that makes you awake and wrecks your liver is not a drug. Investigators traced the damage to reactive metabolites of that specific molecule, a chemistry problem local to TAK-994 rather than an unavoidable tax on hitting the orexin receptor. That distinction is the only reason the program survived. Takeda went back to the bench, changed the molecule, and came back with oveporexton.

THE 2021 SIGNAL
3 of 8 met Hy's law
Of the 8 patients who blew past liver-enzyme safety thresholds in the halted TAK-994 program, 3 met Hy's law, the pattern that predicts a real risk of fatal drug-induced liver failure. Source: Takeda TAK-994 program update, 2021

This time the safety readout held. Across the two pivotal Phase 3 trials, FirstLight and RadiantLight, 273 adults across 19 countries, Takeda reported no serious treatment-related adverse events, and the most common complaints were the almost reassuringly minor kind: insomnia, more frequent urination, extra saliva. The efficacy is what made sleep physicians sit up. Every primary and secondary endpoint was met at a p-value below 0.001. On the Maintenance of Wakefulness Test, patients on the 2 mg twice-daily dose pushed their ability to stay awake toward the range of people who don’t have the disease, and roughly 85 percent drove their Epworth Sleepiness Scale scores to 10 or below, which is to say into normal territory. Weekly cataplexy attacks fell. The American Academy of Sleep Medicine’s summary notes gains not just in daytime sleepiness and cataplexy but in the sleep paralysis, hallucinations, and broken nighttime sleep that travel with the disorder. For a condition managed until now with a stack of drugs each covering one symptom, a single agent moving all of them at once is a real shift in the standard of care.

REACHED NORMAL SLEEPINESS
85%
of patients on the 2 mg dose
Share of oveporexton patients who drove Epworth Sleepiness Scale scores to 10 or below, into the range of people without the disease. Source: CHEST, pivotal Phase 3 trials, 2025

Now the part the launch materials would rather you glide past. Takeda is selling this as the first medicine “to treat the underlying cause” of narcolepsy type 1, and that phrase does a lot of quiet work. Oveporexton does not regrow the dead neurons, and it does nothing to stop the autoimmune process that killed them. It is a receptor agonist you swallow twice a day, presumably for the rest of your life, standing in for a signal your brain can no longer send. Better than a stimulant, yes. The same thing as fixing the cause, no. The gap matters because it sets both patient expectations and, eventually, price. This is an orphan-market drug heading to specialty pharmacies once the DEA finishes assigning it a controlled-substance schedule, a step expected within about 90 days, and Takeda has already told investors it expects the franchise to reach $2 billion to $3 billion in peak annual sales. A company watching generics chew into its older blockbusters has every commercial reason to dress a chronic, lifelong prescription as a cure-adjacent breakthrough. The science is solid. The adjective is marketing.

And there is the ghost in the room the approval paperwork does not dwell on: this is a drug class with a liver-injury body already in the ground. Oveporexton’s pivotal trials ran twelve weeks and turned up no serious treatment-related adverse events, but twelve weeks is a short window in which to judge a medicine people will take for decades, and the failure that killed TAK-994 was a slow-building hepatic signal, not an overnight one. More than 95 percent of trial completers rolled into a long-term extension study, and that data, running past a year and beyond, is what will actually show whether the re-engineered chemistry stays clean under years of exposure. Two numbers will decide whether this is a durable breakthrough or a clean 12-week trial with a long tail nobody watched: the liver-enzyme readouts coming out of that extension, and whatever monitoring language the FDA wrote into the full label. The receptor was never the problem. Proving the new molecule isn’t either is the work of the next two years, and it starts with the first extension liver panel Takeda reports.

Sources

  1. FDA – Approves First Drug to Treat the Full Range of Narcolepsy Type 1 Symptoms (Aug 5, 2026)
  2. Takeda – Positive Results from Two Pivotal Phase 3 Studies of Oveporexton (FirstLight, RadiantLight)
  3. CHEST – Oveporexton (TAK-861): Efficacy and Safety from Two Pivotal Phase 3 Trials
  4. American Academy of Sleep Medicine – FDA Approves Orzeyful for Narcolepsy Type 1 in Adults
  5. Takeda – Update on the TAK-994 Clinical Program (hepatotoxicity halt, 2021)
  6. NEJM – Oral Orexin Receptor 2 Agonist TAK-994 in Narcolepsy Type 1 (2023)
  7. PMC – TAK-994 Mechanistic Investigation into Drug-Induced Liver Injury
  8. SEC – Takeda Pharmaceutical Co. Ltd., Form 6-K (Aug 5, 2026)