💤 The molecule that won this year's Albert Lasker Basic Medical Research Award was found by scientists who thought they were studying appetite. It turned out to be the brain's switch for staying awake, and it explained a disease in which people cannot. Then the pharmaceutical industry spent the next 28 years building the opposite drug first.

On September 9, 2026, the Lasker Foundation gave its basic research award to Masashi Yanagisawa, director and professor at the International Institute for Integrative Sleep Medicine (WPI-IIIS) at the University of Tsukuba, and Emmanuel Mignot, professor at Stanford University School of Medicine. The citation reads: "For identification of orexin, a brain peptide that maintains wakefulness, and the discovery that orexin deficiency causes uncontrolled sleep in narcolepsy—revealing a key insight into the nature of sleep."

What nobody put in a press release is the timing. In the seven weeks before that announcement, three national drug regulators approved the first medicine that does what the discovery implied one should. China cleared it on July 22, the United States on August 5, Japan on August 24.

A hunt for an appetite molecule turned up the brain's wake switch

In the years that led up to that 1998 paper, Yanagisawa's lab at the University of Texas Southwestern Medical Center in Dallas, where he was also an investigator of the Howard Hughes Medical Institute, had a problem on its hands. The genome had handed biologists a pile of receptors with no known partner. Orphan receptors. Something in the body binds them, and nobody knew what.

His team ground up rat brain tissue and looked for whatever switched two of these orphans on. They found two peptides. Injected into rat brains, the peptides made the animals eat. So they took the Greek for appetite, orexis, and called them orexins. The paper landed in Cell on February 20, 1998, with Takeshi Sakurai as first author, under the title "Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior."

The word sleep is not in that title.

That same year, Luis de Lecea and Greg Sutcliffe at Scripps published the same peptides from a different direction and called them hypocretins. That is why the literature still writes both names, usually with one in brackets. Two labs had the same molecule, and neither knew what it was really for.

In 1999, two labs reached the same answer two weeks apart

To test the feeding hypothesis, Yanagisawa's group built mice with the orexin gene deleted. The expectation was mice that ate less.

What they got were mice that collapsed. Animals would be moving normally, then their muscles would go slack and their brain waves would show REM sleep breaking into waking life. That is cataplexy, the signature of narcolepsy. The 1999 Cell paper was titled, with no hedging at all, "Narcolepsy in orexin knockout mice."

Meanwhile at Stanford, Mignot had been grinding through roughly a decade of genetic mapping on an inherited form of narcolepsy in Dobermans and Labradors. The gene he finally cornered was orexin receptor 2. His paper ran in Cell two weeks from Yanagisawa's.

One lab had removed the molecule and produced the disease. The other had started from the disease and arrived at the receptor for that molecule. Humans followed in 2000, when Seiji Nishino, working with Mignot, reported in The Lancet that seven of nine narcolepsy patients had hypocretin-1 in their spinal fluid below the assay's detection limit of 40 picograms per milliliter. Postmortem brains later showed that 85 to 95 percent of the orexin-producing neurons were simply gone.

The cells that make the wakefulness peptide die, so the patient cannot stay awake. For a neurological disease, that is an unusually clean mechanism.

So why did the sleeping pills come first?

Putting the missing molecule back is the obvious next move: a drug that switches the receptor on. An agonist.

Drugmakers did close to the opposite, because blocking a receptor is much easier than mimicking a peptide at one. If orexin keeps you awake, a molecule that blocks its receptor should make you sleepy. That is not a narcolepsy treatment but a sleeping pill, aimed at a complaint that is the mirror image of the disease the discovery explained.

Suvorexant, sold by Merck as Belsomra, was approved by the FDA on August 13, 2014, and in Japan on September 26 the same year. Lemborexant, sold as Dayvigo by the Japanese firm Eisai, followed in 2019. Daridorexant, sold as Quviviq by Idorsia, arrived in January 2022. Three companies shipped three drugs on one mechanism, and every one of them was aimed at insomnia.

Patients with narcolepsy, in the meantime, got one drug for the sleepiness and another for the cataplexy, and nothing at all for the missing peptide.

A liver problem, and then three approvals in 33 days

Takeda did go after the agonist. Its first serious candidate, TAK-994, entered phase 2 with 73 patients, and it worked. At week 8, the highest dose improved the maintenance of wakefulness test by 32.6 minutes. Placebo moved 2.5 minutes in the wrong direction.

Then the liver enzymes came back. Eight patients crossed predefined ALT or AST thresholds, three met the Hy's law criteria for serious drug-induced liver injury, and the program was stopped. The results eventually reached the New England Journal of Medicine in 2023, as the record of a drug that no longer existed.

A second molecule came out of the same chemistry: TAK-861, oveporexton. Its two pivotal trials, FirstLight with 168 patients and RadiantLight with 105, cleared every primary and secondary endpoint at week 12 with p-values below 0.001 at all doses.

Approvals then arrived in a rush. China's National Medical Products Administration (NMPA) cleared it on July 22, 2026 for adults and adolescents from age 16, making China the first country anywhere to approve an orexin receptor agonist. The FDA followed on August 5 for adults, under the name Orzeyful. Japan's Ministry of Health, Labour and Welfare approved it on August 24. Thirty-three days, three markets.

Takeda calls it the only approved medicine that goes after the disease as a whole instead of one symptom at a time. Approved is not the same as available. In the United States the drug is still waiting on the Drug Enforcement Administration's scheduling decision, and its product site said as of September 12 that it expects availability by November 2026. In Japan the company said on August 24 that it was preparing the launch.

So the prize for explaining narcolepsy landed sixteen days after Japan approved the drug built on that explanation. Award juries do not usually admit to reading the news, but the sequence is hard to miss.

Found in Texas, continued in Japan

Every experiment described above happened in Dallas. The orphan receptor screen, the naming, the knockout mice. Yanagisawa spent roughly two decades in Texas, on American institutional money, running an American lab.

He came back in 2012, to run the WPI-IIIS at Tsukuba, an institute created that year under Japan's World Premier International Research Center Initiative. WPI started in 2007 and has produced 18 centers. Each gets roughly 700 million yen a year, about 4.5 million dollars, for ten years, stretched to fifteen years for the centers chosen in fiscal 2012 or earlier, and each is required to recruit at least 30 percent of its researchers from overseas. The program was designed to build places that scientists would move to rather than move away from. Yanagisawa is the clearest case of it working in reverse, bringing one home.

Japan sits on the other side of the award too. Nishino, first author of the Lancet paper that carried the finding into human patients, was on Mignot's side of the story.

There is a familiar claim that Japan has the world's highest rate of narcolepsy. It rests on older Japanese surveys built from questionnaires and clinical interviews, which put prevalence somewhere between 160 and 590 cases per 100,000 people. When researchers instead worked through the insurance claims of 6,110,751 people in the Japan Medical Data Center, they arrived at 37.5 per 100,000 as of the end of 2019, and put part of the older range down to overreporting. Set that beside the same kind of data elsewhere: two American claims analyses found 38.9 to 44.3 and 53.3 per 100,000, while European estimates run from 4.4 to 21.0. On like-for-like data Japan sits next to the United States rather than above it. Whether the country is an outlier turns out to depend almost entirely on who is counting and how.

The drug is Takeda's and the discovery is Yanagisawa's. Japan was still the third country to approve it, 33 days behind China and 19 days behind the United States.

This year Japanese researchers appear in both of the Lasker's research categories. Alongside Yanagisawa, three Chugai Pharmaceutical researchers won the clinical award for Hemlibra (emicizumab), a bispecific antibody for hemophilia A that more than 30,000 patients worldwide had used by the end of June 2026. Four Japanese laureates in one year, and the first since Kazutoshi Mori of Kyoto University, whose award was announced on September 9, 2014. Same calendar date, twelve years apart.

The Lasker is often called America's Nobel, and it often precedes one. Not always. Mori has been waiting twelve years.

If you have taken a prescription sleeping pill in the last decade, there is a fair chance it was a byproduct of a 1998 search for an appetite molecule in a Texas laboratory. What do doctors reach for first where you live, and has the narcolepsy drug reached your country yet?

References