User login
The Strange Untold Story of How Science Solved Narcolepsy
It was 1996, and Masashi Yanagisawa was on the brink of his next discovery.
The Japanese scientist had arrived at the University of Texas Southwestern in Dallas 5 years earlier, setting up his own lab at age 31. After earning his medical degree, he’d gained notoriety as a PhD student when he discovered endothelin, the body’s most potent vasoconstrictor.
Yanagisawa was about to prove this wasn’t a first-timer’s fluke.
His focus was G-protein–coupled receptors (GPCRs), cell surface receptors that respond to a range of molecules and a popular target for drug discovery. The Human Genome Project had just revealed a slew of newly discovered receptors, or “orphan” GPCRs, and identifying an activating molecule could yield a new drug. (That vasoconstrictor endothelin was one such success story, leading to four new drug approvals in the United States over the past quarter century.)
Yanagisawa and his team created 50 cell lines, each expressing one orphan receptor. They applied animal tissue to every line, along with a calcium-sensitive dye. If the cells glowed under the microscope, they had a hit.
“He was basically doing an elaborate fishing expedition,” said Jon Willie, MD, PhD, an associate professor of neurosurgery at Washington University School of Medicine in St. Louis, Missouri, who would later join Yanagisawa’s team.
It wasn’t long before the neon-green fluorescence signaled a match. After isolating the activating molecule, the scientists realized they were dealing with two neuropeptides.
No one had ever seen these proteins before. And no one knew their discovery would set off a decades-long journey that would finally solve a century-old medical mystery — and may even fix one of the biggest health crises of our time, as revealed by research published earlier in 2024. It’s a story of strange coincidences, serendipitous discoveries, and quirky details. Most of all, it’s a fascinating example of how basic science can revolutionize medicine — and how true breakthroughs happen over time and in real time.
But That’s Basic Science for You
Most basic science studies — the early, foundational research that provides the building blocks for science that follows — don’t lead to medical breakthroughs. But some do, often in surprising ways.
Also called curiosity-driven research, basic science aims to fill knowledge gaps to keep science moving, even if the trajectory isn’t always clear.
“The people working on the basic research that led to discoveries that transformed the modern world had no idea at the time,” said Isobel Ronai, PhD, a postdoctoral fellow in life sciences at Harvard University, Cambridge, Massachusetts. “Often, these stories can only be seen in hindsight,” sometimes decades later.
Case in point: For molecular biology techniques — things like DNA sequencing and gene targeting — the lag between basic science and breakthrough is, on average, 23 years. While many of the resulting techniques have received Nobel Prizes, few of the foundational discoveries have been awarded such accolades.
“The scientific glory is more often associated with the downstream applications,” said Ronai. “The importance of basic research can get lost. But it is the foundation for any future application, such as drug development.”
As funding is increasingly funneled toward applied research, basic science can require a certain persistence. What this under-appreciation can obscure is the pathway to discovery — which is often as compelling as the end result, full of unpredictable twists, turns, and even interpersonal intrigue.
And then there’s the fascinating — and definitely complicated — phenomenon of multiple independent discoveries.
As in: What happens when two independent teams discover the same thing at the same time?
Back to Yanagisawa’s Lab ...
... where he and his team learned a few things about those new neuropeptides. Rat brain studies pinpointed the lateral hypothalamus as the peptides’ area of activity — a region often called the brain’s feeding center.
“If you destroy that part of the brain, animals lose appetite,” said Yanagisawa. So these peptides must control feeding, the scientists thought.
Sure enough, injecting the proteins into rat brains led the rodents to start eating.
Satisfied, the team named them “orexin-A” and “orexin-B,” for the Greek word “orexis,” meaning appetite. The brain receptors became “orexin-1” and “orexin-2.” The team prepared to publish its findings in Cell.
But another group beat them to it.
Introducing the ‘Hypocretins’
In early January 1998, a team of Scripps Research Institute scientists, led by J. Gregor Sutcliffe, PhD, released a paper in the journal PNAS. They described a gene encoding for the precursor to two neuropeptides
As the peptides were in the hypothalamus and structurally like secretin (a gut hormone), they called them “hypocretins.” The hypocretin peptides excited neurons in the hypothalamus, and later that year, the scientists discovered that the neurons’ branches extended, tentacle-like, throughout the brain. “Many of the connected areas were involved in sleep-wake control,” said Thomas Kilduff, PhD, who joined the Sutcliffe lab just weeks before the hypocretin discovery. At the time, however, the significance of this finding was not yet clear.
Weeks later, in February 1998, Yanagisawa’s paper came out.
Somehow, two groups, over 1000 miles apart, had stumbled on the same neuropeptides at the same time.
“I first heard about [Yanagisawa’s] paper on NBC Nightly News,” recalls Kilduff. “I was skiing in the mountains, so I had to wait until Monday to get back to the lab to see what the paper was all about.”
He realized that Yanagisawa’s orexin was his lab’s hypocretin, although the study didn’t mention another team’s discovery.
“There may have been accusations. But as far as I know, it’s because [Yanagisawa] didn’t know [about the other paper],” said Willie. “This was not something he produced in 2 months. This was clearly years of work.”
‘Multiple Discovery’ Happens More Often Than You Think
In the mid-20th century, sociologist Robert Merton described the phenomenon of “multiple discovery,” where many scientific discoveries or inventions are made independently at roughly the same time.
“This happens much more frequently in scientific research than people suppose,” said David Pendlebury, head of research analysis at Clarivate’s Institute for Scientific Information, the analytics company’s research arm. (Last year, Pendlebury flagged the hypocretin/orexin discovery for Clarivate’s prestigious Citations Laureates award, an honor that aims to predict, often successfully, who will go on to win the Nobel Prize.)
“People have this idea of the lone researcher making a brilliant discovery,” Pendlebury said. “But more and more, teams find things at the same time.”
While this can — and does — lead to squabbling about who deserves credit, the desire to be first can also be highly motivating, said Mike Schneider, PhD, an assistant professor of philosophy at the University of Missouri, Columbia, who studies the social dynamics of science, potentially leading to faster scientific advancement.
The downside? If two groups produce the same or similar results, but one publishes first, scientific journals tend to reject the second, citing a lack of novelty.
Yet duplicating research is a key step in confirming the validity of a discovery.
That’s why, in 2018, the journal PLOS Biology created a provision for “scooped” scientists, allowing them to submit their paper within 6 months of the first as a complementary finding. Instead of viewing this as redundancy, the editors believe it adds robustness to the research.
‘What the Heck Is This Mouse Doing?’
Even though he’d been scooped, Yanagisawa forged on to the next challenge: Confirming whether orexin regulated feeding.
He began breeding mice missing the orexin gene. His team expected these “knockout” mice to eat less, resulting in a thinner body than other rodents. To the contrary, “they were on average fatter,” said Willie. “They were eating less but weighed more, indicating a slower metabolism.”
The researchers were befuddled. “We were really disappointed, almost desperate about what to do,” said Yanagisawa.
As nocturnal animals eat more at night, he decided they should study the mice after dark. One of his students, Richard Chemelli, MD, bought an infrared video camera from Radio Shack, filming the first 4 hours of the mice’s active period for several nights.
After watching the footage, “Rick called me and said, ‘Let’s get into the lab,’ ” said Willie. “It was four of us on a Saturday looking at these videos, saying, ‘What the heck is this mouse doing?’ ”
While exploring their habitat, the knockout mice would randomly fall over, pop back up after a minute or so, and resume normal activity. This happened over and over — and the scientists were unsure why.
They began monitoring the mice’s brains during these episodes — and made a startling discovery.
The mice weren’t having seizures. They were shifting directly into REM sleep, bypassing the non-REM stage, then quickly toggling back to wake mode.
“That’s when we knew these animals had something akin to narcolepsy,” said Willie.
The team recruited Thomas Scammell, MD, a Harvard neurologist, to investigate whether modafinil — an anti-narcoleptic drug without a clear mechanism — affected orexin neurons.
Two hours after injecting the mice with the medication, the scientists sacrificed them and stained their brains. Remarkably, the number of neurons showing orexin activity had increased ninefold. It seemed modafinil worked by activating the orexin system.
These findings had the potential to crack open the science of narcolepsy, one of the most mysterious sleep disorders.
Unless, of course, another team did it first.
The Mystery of Narcolepsy
Yet another multiple discovery, narcolepsy was first described by two scientists — one in Germany, the other in France — within a short span in the late 1800s.
It would be more than a hundred years before anyone understood the disorder’s cause, even though it affects about 1 in 2000 people.
“Patients were often labeled as lazy and malingerers,” said Kilduff, “since they were sleepy all the time and had this weird motor behavior called cataplexy” or the sudden loss of muscle tone.
In the early 1970s, William Dement, MD, PhD — “the father of sleep medicine” — was searching for a narcoleptic cat to study. He couldn’t find a feline, but several colleagues mentioned dogs with narcolepsy-like symptoms.
Dement, who died in 2020, had found his newest research subjects.
In 1973, he started a narcoleptic dog colony at Stanford University in Palo Alto, California. At first, he focused on poodles and beagles. After discovering their narcolepsy wasn’t genetic, he pivoted to dobermans and labradors. Their narcolepsy was inherited, so he could breed them to populate the colony.
Although human narcolepsy is rarely genetic, it’s otherwise a lot like the version in these dogs.
Both involve daytime sleepiness, “pathological” bouts of REM sleep, and the loss of muscle tone in response to emotions, often positive ones.
The researchers hoped the canines could unlock a treatment for human narcolepsy. They began laying out a path of dog kibble, then injecting the dogs with drugs such as selective serotonin reuptake inhibitors. They wanted to see what might help them stay awake as they excitedly chowed down.
Kilduff also started a molecular genetics program, trying to identify the genetic defect behind canine narcolepsy. But after a parvovirus outbreak, Kilduff resigned from the project, drained from the strain of seeing so many dogs die.
A decade after his departure from the dog colony, his work would dramatically intersect with that of his successor, Emmanuel Mignot, MD, PhD.
“I thought I had closed the narcolepsy chapter in my life forever,” said Kilduff. “Then in 1998, we described this novel neuropeptide, hypocretin, that turned out to be the key to understanding the disorder.”
Narcoleptic Dogs in California, Mutant Mice in Texas
It was modafinil — the same anti-narcoleptic drug Yanagisawa’s team studied — that brought Emmanuel Mignot to the United States. After training as a pharmacologist in France, his home country sent him to Stanford to study the drug, which was discovered by French scientists, as his required military service.
As Kilduff’s replacement at the dog colony, his goal was to figure out how modafinil worked, hoping to attract a US company to develop the drug.
The plan succeeded. Modafinil became Provigil, a billion-dollar narcolepsy drug, and Mignot became “completely fascinated” with the disorder.
“I realized quickly that there was no way we’d find the cause of narcolepsy by finding the mode of action of this drug,” Mignot said. “Most likely, the drug was acting downstream, not at the cause of the disorder.”
To discover the answer, he needed to become a geneticist. And so began his 11-year odyssey to find the cause of canine narcolepsy.
After mapping the dog genome, Mignot set out to find the smallest stretch of chromosome that the narcoleptic animals had in common. “For a very long time, we were stuck with a relatively large region [of DNA],” he recalls. “It was a no man’s land.”
Within that region was the gene for the hypocretin/orexin-2 receptor — the same receptor that Yanagisawa had identified in his first orexin paper. Mignot didn’t immediately pursue that gene as a possibility — even though his students suggested it. Why?
“The decision was simply: Should we lose time to test a possible candidate [gene] among many?” Mignot said.
As Mignot studied dog DNA in California, Yanagisawa was creating mutant mice in Texas. Unbeknownst to either scientist, their work was about to converge.
What Happened Next Is Somewhat Disputed
After diagnosing his mice with narcolepsy, Yanagisawa opted not to share this finding with Mignot, though he knew about Mignot’s interest in the condition. Instead, he asked a colleague to find out how far along Mignot was in his genetics research.
According to Yanagisawa, his colleague didn’t realize how quickly DNA sequencing could happen once a target gene was identified. At a sleep meeting, “he showed Emmanuel all of our raw data. Almost accidentally, he disclosed our findings,” he said. “It was a shock for me.”
Unsure whether he was part of the orexin group, Mignot decided not to reveal that he’d identified the hypocretin/orexin-2 receptor gene as the faulty one in his narcoleptic dogs.
Although he didn’t share this finding, Mignot said he did offer to speak with the lead researcher to see if their findings were the same. If they were, they could jointly submit their articles. But Mignot never heard back.
Meanwhile, back at his lab, Mignot buckled down. While he wasn’t convinced the mouse data proved anything, it did give him the motivation to move faster.
Within weeks, he submitted his findings to Cell, revealing a mutation in the hypocretin/orexin-2 receptor gene as the cause of canine narcolepsy. According to Yanagisawa, the journal’s editor invited him to peer-review the paper, tipping him off to its existence.
“I told him I had a conflict of interest,” said Yanagisawa. “And then we scrambled to finish our manuscript. We wrote up the paper within almost 5 days.”
For a moment, it seemed both papers would be published together in Cell. Instead, on August 6, 1999, Mignot’s study was splashed solo across the journal’s cover.
“At the time, our team was pissed off, but looking back, what else could Emmanuel have done?” said Willie, who was part of Yanagisawa’s team. “The grant he’d been working on for years was at risk. He had it within his power to do the final experiments. Of course he was going to finish.”
Two weeks later, Yanagisawa’s findings followed, also in Cell.
His paper proposed knockout mice as a model for human narcolepsy and orexin as a key regulator of the sleep/wake cycle. With orexin-activated neurons branching into other areas of the brain, the peptide seemed to promote wakefulness by synchronizing several arousal neurotransmitters, such as serotonin, norepinephrine, and histamine.
“If you don’t have orexin, each of those systems can still function, but they’re not as coordinated,” said Willie. “If you have narcolepsy, you’re capable of wakefulness, and you’re capable of sleep. What you can’t do is prevent inappropriately switching between states.”
Together, the two papers painted a clear picture: Narcolepsy was the result of a dysfunction in the hypocretin/orexin system.
After more than a century, the cause of narcolepsy was starting to come into focus.
“This was blockbuster,” said Willie.
By itself, either finding — one in dogs, one in mice — might have been met with skepticism. But in combination, they offered indisputable evidence about narcolepsy’s cause.
The Human Brains in Your Fridge Hold Secrets
Jerome Siegel had been searching for the cause of human narcolepsy for years. A PhD and professor at the University of California, Los Angeles, he had managed to acquire four human narcoleptic brains. As laughter is often the trigger for the sudden shift to REM sleep in humans, he focused on the amygdala, an area linked to emotion.
“I looked in the amygdala and didn’t see anything,” he said. “So the brains stayed in my refrigerator for probably 10 years.”
Then he was invited to review Yanagisawa’s study in Cell. The lightbulb clicked on: Maybe the hypothalamus — not the amygdala — was the area of abnormality. He and his team dug out the decade-old brains.
When they stained the brains, the massive loss of hypocretin-activated neurons was hard to miss: On average, the narcoleptic brains had only about 7000 of the cells versus 70,000 in the average human brain. The scientists also noticed scar tissue in the hypothalamus, indicating that the neurons had at some point died, rather than being absent from birth.
What Siegel didn’t know: Mignot had also acquired a handful of human narcoleptic brains.
Already, he had coauthored a study showing that hypocretin/orexin was undetectable in the cerebrospinal fluid of the majority of the people with narcolepsy his team tested. It seemed clear that the hypocretin/orexin system was flawed — or even broken — in people with the condition.
“It looked like the cause of narcolepsy in humans was indeed this lack of orexin in the brain,” he said. “That was the hypothesis immediately. To me, this is when we established that narcolepsy in humans was due to a lack of orexin. The next thing was to check that the cells were missing.”
Now he could do exactly that.
As expected, Mignot’s team observed a dramatic loss of hypocretin/orexin cells in the narcoleptic brains. They also noticed that a different cell type in the hypothalamus was unaffected. This implied the damage was specific to the hypocretin-activated cells and supported a hunch they already had: That the deficit was the result not of a genetic defect but of an autoimmune attack. (It’s a hypothesis Mignot has spent the last 15 years proving.)
It wasn’t until a gathering in Hawaii, in late August 2000, that the two realized the overlap of their work.
To celebrate his team’s finding, Mignot had invited a group of researchers to Big Island. With his paper scheduled for publication on September 1, he felt comfortable presenting his findings to his guests, which included Siegel.
Until then, “I didn’t know what he had found, and he didn’t know what I had found, which basically was the same thing,” said Siegel.
In yet another strange twist, the two papers were published just weeks apart, simultaneously revealing that human narcoleptics have a depleted supply of the neurons that bind to hypocretin/orexin. The cause of the disorder was at last a certainty.
“Even if I was first, what does it matter? In the end, you need confirmation,” said Mignot. “You need multiple people to make sure that it’s true. It’s good science when things like this happen.”
How All of This Changed Medicine
Since these groundbreaking discoveries, the diagnosis of narcolepsy has become much simpler. Lab tests can now easily measure hypocretin in cerebrospinal fluid, providing a definitive diagnosis.
But the development of narcolepsy treatments has lagged — even though hypocretin/orexin replacement therapy is the obvious answer.
“Almost 25 years have elapsed, and there’s no such therapeutic on the market,” said Kilduff, who now works for SRI International, a non-profit research and development institute.
That’s partly because agonists — drugs that bind to receptors in the brain — are challenging to create, as this requires mimicking the activating molecule’s structure, like copying the grooves of an intricate key.
Antagonists, by comparison, are easier to develop. These act as a gate, blocking access to the receptors. As a result, drugs that promote sleep by thwarting hypocretin/orexin have emerged more quickly, providing a flurry of new options for people with insomnia. The first, suvorexant, was launched in 2014. Two others followed in recent years.
Researchers are hopeful a hypocretin/orexin agonist is on the horizon.
“This is a very hot area of drug development,” said Kilduff. “It’s just a matter of who’s going to get the drug to market first.”
One More Hypocretin/Orexin Surprise — and It Could Be The Biggest
Several years ago, Siegel’s lab received what was supposed to be a healthy human brain — one they could use as a comparison for narcoleptic brains. But researcher Thomas Thannickal, PhD, lead author of the UCLA study linking hypocretin loss to human narcolepsy, noticed something strange: This brain had significantly more hypocretin neurons than average.
Was this due to a seizure? A traumatic death? Siegel called the brain bank to request the donor’s records. He was told they were missing.
Years later, Siegel happened to be visiting the brain bank for another project and found himself in a room adjacent to the medical records. “Nobody was there,” he said, “so I just opened a drawer.”
Shuffling through the brain bank’s files, Siegel found the medical records he’d been told were lost. In the file was a note from the donor, explaining that he was a former heroin addict.
“I almost fell out of my chair,” said Siegel. “I realized this guy’s heroin addiction likely had something to do with his very unusual brain.”
Obviously, opioids affected the orexin system. But how?
“It’s when people are happy that this peptide is released,” said Siegel. “The hypocretin system is not just related to alertness. It’s related to pleasure.”
As Yanagisawa observed early on, hypocretin/orexin does indeed play a role in eating — just not the one he initially thought. The peptides prompted pleasure seeking. So the rodents ate.
In 2018, after acquiring five more brains, Siegel’s group published a study in Translational Medicine showing 54% more detectable hypocretin neurons in the brains of heroin addicts than in those of control individuals.
In 2022, another breakthrough: His team showed that morphine significantly altered the pathways of hypocretin neurons in mice, sending their axons into brain regions associated with addiction. Then, when they removed the mice’s hypocretin neurons and discontinued their daily morphine dose, the rodents showed no symptoms of opioid withdrawal.
This fits the connection with narcolepsy: Among the standard treatments for the condition are amphetamines and other stimulants, which all have addictive potential. Yet, “narcoleptics never abuse these drugs,” Siegel said. “They seem to be uniquely resistant to addiction.”
This could powerfully change the way opioids are administered.
“If you prevent the hypocretin response to opioids, you may be able to prevent opioid addiction,” said Siegel. In other words, blocking the hypocretin system with a drug like those used to treat insomnia may allow patients to experience the pain-relieving benefits of opioids — without the risk for addiction.
His team is currently investigating treatments targeting the hypocretin/orexin system for opioid addiction.
In a study published in July, they found that mice who received suvorexant — the drug for insomnia — didn’t anticipate their daily dose of opioids the way other rodents did. This suggests the medication prevented addiction, without diminishing the pain-relieving effect of opioids.
If it translates to humans, this discovery could potentially save millions of lives.
“I think it’s just us working on this,” said Siegel.
But with hypocretin/orexin, you never know.
A version of this article appeared on Medscape.com.
It was 1996, and Masashi Yanagisawa was on the brink of his next discovery.
The Japanese scientist had arrived at the University of Texas Southwestern in Dallas 5 years earlier, setting up his own lab at age 31. After earning his medical degree, he’d gained notoriety as a PhD student when he discovered endothelin, the body’s most potent vasoconstrictor.
Yanagisawa was about to prove this wasn’t a first-timer’s fluke.
His focus was G-protein–coupled receptors (GPCRs), cell surface receptors that respond to a range of molecules and a popular target for drug discovery. The Human Genome Project had just revealed a slew of newly discovered receptors, or “orphan” GPCRs, and identifying an activating molecule could yield a new drug. (That vasoconstrictor endothelin was one such success story, leading to four new drug approvals in the United States over the past quarter century.)
Yanagisawa and his team created 50 cell lines, each expressing one orphan receptor. They applied animal tissue to every line, along with a calcium-sensitive dye. If the cells glowed under the microscope, they had a hit.
“He was basically doing an elaborate fishing expedition,” said Jon Willie, MD, PhD, an associate professor of neurosurgery at Washington University School of Medicine in St. Louis, Missouri, who would later join Yanagisawa’s team.
It wasn’t long before the neon-green fluorescence signaled a match. After isolating the activating molecule, the scientists realized they were dealing with two neuropeptides.
No one had ever seen these proteins before. And no one knew their discovery would set off a decades-long journey that would finally solve a century-old medical mystery — and may even fix one of the biggest health crises of our time, as revealed by research published earlier in 2024. It’s a story of strange coincidences, serendipitous discoveries, and quirky details. Most of all, it’s a fascinating example of how basic science can revolutionize medicine — and how true breakthroughs happen over time and in real time.
But That’s Basic Science for You
Most basic science studies — the early, foundational research that provides the building blocks for science that follows — don’t lead to medical breakthroughs. But some do, often in surprising ways.
Also called curiosity-driven research, basic science aims to fill knowledge gaps to keep science moving, even if the trajectory isn’t always clear.
“The people working on the basic research that led to discoveries that transformed the modern world had no idea at the time,” said Isobel Ronai, PhD, a postdoctoral fellow in life sciences at Harvard University, Cambridge, Massachusetts. “Often, these stories can only be seen in hindsight,” sometimes decades later.
Case in point: For molecular biology techniques — things like DNA sequencing and gene targeting — the lag between basic science and breakthrough is, on average, 23 years. While many of the resulting techniques have received Nobel Prizes, few of the foundational discoveries have been awarded such accolades.
“The scientific glory is more often associated with the downstream applications,” said Ronai. “The importance of basic research can get lost. But it is the foundation for any future application, such as drug development.”
As funding is increasingly funneled toward applied research, basic science can require a certain persistence. What this under-appreciation can obscure is the pathway to discovery — which is often as compelling as the end result, full of unpredictable twists, turns, and even interpersonal intrigue.
And then there’s the fascinating — and definitely complicated — phenomenon of multiple independent discoveries.
As in: What happens when two independent teams discover the same thing at the same time?
Back to Yanagisawa’s Lab ...
... where he and his team learned a few things about those new neuropeptides. Rat brain studies pinpointed the lateral hypothalamus as the peptides’ area of activity — a region often called the brain’s feeding center.
“If you destroy that part of the brain, animals lose appetite,” said Yanagisawa. So these peptides must control feeding, the scientists thought.
Sure enough, injecting the proteins into rat brains led the rodents to start eating.
Satisfied, the team named them “orexin-A” and “orexin-B,” for the Greek word “orexis,” meaning appetite. The brain receptors became “orexin-1” and “orexin-2.” The team prepared to publish its findings in Cell.
But another group beat them to it.
Introducing the ‘Hypocretins’
In early January 1998, a team of Scripps Research Institute scientists, led by J. Gregor Sutcliffe, PhD, released a paper in the journal PNAS. They described a gene encoding for the precursor to two neuropeptides
As the peptides were in the hypothalamus and structurally like secretin (a gut hormone), they called them “hypocretins.” The hypocretin peptides excited neurons in the hypothalamus, and later that year, the scientists discovered that the neurons’ branches extended, tentacle-like, throughout the brain. “Many of the connected areas were involved in sleep-wake control,” said Thomas Kilduff, PhD, who joined the Sutcliffe lab just weeks before the hypocretin discovery. At the time, however, the significance of this finding was not yet clear.
Weeks later, in February 1998, Yanagisawa’s paper came out.
Somehow, two groups, over 1000 miles apart, had stumbled on the same neuropeptides at the same time.
“I first heard about [Yanagisawa’s] paper on NBC Nightly News,” recalls Kilduff. “I was skiing in the mountains, so I had to wait until Monday to get back to the lab to see what the paper was all about.”
He realized that Yanagisawa’s orexin was his lab’s hypocretin, although the study didn’t mention another team’s discovery.
“There may have been accusations. But as far as I know, it’s because [Yanagisawa] didn’t know [about the other paper],” said Willie. “This was not something he produced in 2 months. This was clearly years of work.”
‘Multiple Discovery’ Happens More Often Than You Think
In the mid-20th century, sociologist Robert Merton described the phenomenon of “multiple discovery,” where many scientific discoveries or inventions are made independently at roughly the same time.
“This happens much more frequently in scientific research than people suppose,” said David Pendlebury, head of research analysis at Clarivate’s Institute for Scientific Information, the analytics company’s research arm. (Last year, Pendlebury flagged the hypocretin/orexin discovery for Clarivate’s prestigious Citations Laureates award, an honor that aims to predict, often successfully, who will go on to win the Nobel Prize.)
“People have this idea of the lone researcher making a brilliant discovery,” Pendlebury said. “But more and more, teams find things at the same time.”
While this can — and does — lead to squabbling about who deserves credit, the desire to be first can also be highly motivating, said Mike Schneider, PhD, an assistant professor of philosophy at the University of Missouri, Columbia, who studies the social dynamics of science, potentially leading to faster scientific advancement.
The downside? If two groups produce the same or similar results, but one publishes first, scientific journals tend to reject the second, citing a lack of novelty.
Yet duplicating research is a key step in confirming the validity of a discovery.
That’s why, in 2018, the journal PLOS Biology created a provision for “scooped” scientists, allowing them to submit their paper within 6 months of the first as a complementary finding. Instead of viewing this as redundancy, the editors believe it adds robustness to the research.
‘What the Heck Is This Mouse Doing?’
Even though he’d been scooped, Yanagisawa forged on to the next challenge: Confirming whether orexin regulated feeding.
He began breeding mice missing the orexin gene. His team expected these “knockout” mice to eat less, resulting in a thinner body than other rodents. To the contrary, “they were on average fatter,” said Willie. “They were eating less but weighed more, indicating a slower metabolism.”
The researchers were befuddled. “We were really disappointed, almost desperate about what to do,” said Yanagisawa.
As nocturnal animals eat more at night, he decided they should study the mice after dark. One of his students, Richard Chemelli, MD, bought an infrared video camera from Radio Shack, filming the first 4 hours of the mice’s active period for several nights.
After watching the footage, “Rick called me and said, ‘Let’s get into the lab,’ ” said Willie. “It was four of us on a Saturday looking at these videos, saying, ‘What the heck is this mouse doing?’ ”
While exploring their habitat, the knockout mice would randomly fall over, pop back up after a minute or so, and resume normal activity. This happened over and over — and the scientists were unsure why.
They began monitoring the mice’s brains during these episodes — and made a startling discovery.
The mice weren’t having seizures. They were shifting directly into REM sleep, bypassing the non-REM stage, then quickly toggling back to wake mode.
“That’s when we knew these animals had something akin to narcolepsy,” said Willie.
The team recruited Thomas Scammell, MD, a Harvard neurologist, to investigate whether modafinil — an anti-narcoleptic drug without a clear mechanism — affected orexin neurons.
Two hours after injecting the mice with the medication, the scientists sacrificed them and stained their brains. Remarkably, the number of neurons showing orexin activity had increased ninefold. It seemed modafinil worked by activating the orexin system.
These findings had the potential to crack open the science of narcolepsy, one of the most mysterious sleep disorders.
Unless, of course, another team did it first.
The Mystery of Narcolepsy
Yet another multiple discovery, narcolepsy was first described by two scientists — one in Germany, the other in France — within a short span in the late 1800s.
It would be more than a hundred years before anyone understood the disorder’s cause, even though it affects about 1 in 2000 people.
“Patients were often labeled as lazy and malingerers,” said Kilduff, “since they were sleepy all the time and had this weird motor behavior called cataplexy” or the sudden loss of muscle tone.
In the early 1970s, William Dement, MD, PhD — “the father of sleep medicine” — was searching for a narcoleptic cat to study. He couldn’t find a feline, but several colleagues mentioned dogs with narcolepsy-like symptoms.
Dement, who died in 2020, had found his newest research subjects.
In 1973, he started a narcoleptic dog colony at Stanford University in Palo Alto, California. At first, he focused on poodles and beagles. After discovering their narcolepsy wasn’t genetic, he pivoted to dobermans and labradors. Their narcolepsy was inherited, so he could breed them to populate the colony.
Although human narcolepsy is rarely genetic, it’s otherwise a lot like the version in these dogs.
Both involve daytime sleepiness, “pathological” bouts of REM sleep, and the loss of muscle tone in response to emotions, often positive ones.
The researchers hoped the canines could unlock a treatment for human narcolepsy. They began laying out a path of dog kibble, then injecting the dogs with drugs such as selective serotonin reuptake inhibitors. They wanted to see what might help them stay awake as they excitedly chowed down.
Kilduff also started a molecular genetics program, trying to identify the genetic defect behind canine narcolepsy. But after a parvovirus outbreak, Kilduff resigned from the project, drained from the strain of seeing so many dogs die.
A decade after his departure from the dog colony, his work would dramatically intersect with that of his successor, Emmanuel Mignot, MD, PhD.
“I thought I had closed the narcolepsy chapter in my life forever,” said Kilduff. “Then in 1998, we described this novel neuropeptide, hypocretin, that turned out to be the key to understanding the disorder.”
Narcoleptic Dogs in California, Mutant Mice in Texas
It was modafinil — the same anti-narcoleptic drug Yanagisawa’s team studied — that brought Emmanuel Mignot to the United States. After training as a pharmacologist in France, his home country sent him to Stanford to study the drug, which was discovered by French scientists, as his required military service.
As Kilduff’s replacement at the dog colony, his goal was to figure out how modafinil worked, hoping to attract a US company to develop the drug.
The plan succeeded. Modafinil became Provigil, a billion-dollar narcolepsy drug, and Mignot became “completely fascinated” with the disorder.
“I realized quickly that there was no way we’d find the cause of narcolepsy by finding the mode of action of this drug,” Mignot said. “Most likely, the drug was acting downstream, not at the cause of the disorder.”
To discover the answer, he needed to become a geneticist. And so began his 11-year odyssey to find the cause of canine narcolepsy.
After mapping the dog genome, Mignot set out to find the smallest stretch of chromosome that the narcoleptic animals had in common. “For a very long time, we were stuck with a relatively large region [of DNA],” he recalls. “It was a no man’s land.”
Within that region was the gene for the hypocretin/orexin-2 receptor — the same receptor that Yanagisawa had identified in his first orexin paper. Mignot didn’t immediately pursue that gene as a possibility — even though his students suggested it. Why?
“The decision was simply: Should we lose time to test a possible candidate [gene] among many?” Mignot said.
As Mignot studied dog DNA in California, Yanagisawa was creating mutant mice in Texas. Unbeknownst to either scientist, their work was about to converge.
What Happened Next Is Somewhat Disputed
After diagnosing his mice with narcolepsy, Yanagisawa opted not to share this finding with Mignot, though he knew about Mignot’s interest in the condition. Instead, he asked a colleague to find out how far along Mignot was in his genetics research.
According to Yanagisawa, his colleague didn’t realize how quickly DNA sequencing could happen once a target gene was identified. At a sleep meeting, “he showed Emmanuel all of our raw data. Almost accidentally, he disclosed our findings,” he said. “It was a shock for me.”
Unsure whether he was part of the orexin group, Mignot decided not to reveal that he’d identified the hypocretin/orexin-2 receptor gene as the faulty one in his narcoleptic dogs.
Although he didn’t share this finding, Mignot said he did offer to speak with the lead researcher to see if their findings were the same. If they were, they could jointly submit their articles. But Mignot never heard back.
Meanwhile, back at his lab, Mignot buckled down. While he wasn’t convinced the mouse data proved anything, it did give him the motivation to move faster.
Within weeks, he submitted his findings to Cell, revealing a mutation in the hypocretin/orexin-2 receptor gene as the cause of canine narcolepsy. According to Yanagisawa, the journal’s editor invited him to peer-review the paper, tipping him off to its existence.
“I told him I had a conflict of interest,” said Yanagisawa. “And then we scrambled to finish our manuscript. We wrote up the paper within almost 5 days.”
For a moment, it seemed both papers would be published together in Cell. Instead, on August 6, 1999, Mignot’s study was splashed solo across the journal’s cover.
“At the time, our team was pissed off, but looking back, what else could Emmanuel have done?” said Willie, who was part of Yanagisawa’s team. “The grant he’d been working on for years was at risk. He had it within his power to do the final experiments. Of course he was going to finish.”
Two weeks later, Yanagisawa’s findings followed, also in Cell.
His paper proposed knockout mice as a model for human narcolepsy and orexin as a key regulator of the sleep/wake cycle. With orexin-activated neurons branching into other areas of the brain, the peptide seemed to promote wakefulness by synchronizing several arousal neurotransmitters, such as serotonin, norepinephrine, and histamine.
“If you don’t have orexin, each of those systems can still function, but they’re not as coordinated,” said Willie. “If you have narcolepsy, you’re capable of wakefulness, and you’re capable of sleep. What you can’t do is prevent inappropriately switching between states.”
Together, the two papers painted a clear picture: Narcolepsy was the result of a dysfunction in the hypocretin/orexin system.
After more than a century, the cause of narcolepsy was starting to come into focus.
“This was blockbuster,” said Willie.
By itself, either finding — one in dogs, one in mice — might have been met with skepticism. But in combination, they offered indisputable evidence about narcolepsy’s cause.
The Human Brains in Your Fridge Hold Secrets
Jerome Siegel had been searching for the cause of human narcolepsy for years. A PhD and professor at the University of California, Los Angeles, he had managed to acquire four human narcoleptic brains. As laughter is often the trigger for the sudden shift to REM sleep in humans, he focused on the amygdala, an area linked to emotion.
“I looked in the amygdala and didn’t see anything,” he said. “So the brains stayed in my refrigerator for probably 10 years.”
Then he was invited to review Yanagisawa’s study in Cell. The lightbulb clicked on: Maybe the hypothalamus — not the amygdala — was the area of abnormality. He and his team dug out the decade-old brains.
When they stained the brains, the massive loss of hypocretin-activated neurons was hard to miss: On average, the narcoleptic brains had only about 7000 of the cells versus 70,000 in the average human brain. The scientists also noticed scar tissue in the hypothalamus, indicating that the neurons had at some point died, rather than being absent from birth.
What Siegel didn’t know: Mignot had also acquired a handful of human narcoleptic brains.
Already, he had coauthored a study showing that hypocretin/orexin was undetectable in the cerebrospinal fluid of the majority of the people with narcolepsy his team tested. It seemed clear that the hypocretin/orexin system was flawed — or even broken — in people with the condition.
“It looked like the cause of narcolepsy in humans was indeed this lack of orexin in the brain,” he said. “That was the hypothesis immediately. To me, this is when we established that narcolepsy in humans was due to a lack of orexin. The next thing was to check that the cells were missing.”
Now he could do exactly that.
As expected, Mignot’s team observed a dramatic loss of hypocretin/orexin cells in the narcoleptic brains. They also noticed that a different cell type in the hypothalamus was unaffected. This implied the damage was specific to the hypocretin-activated cells and supported a hunch they already had: That the deficit was the result not of a genetic defect but of an autoimmune attack. (It’s a hypothesis Mignot has spent the last 15 years proving.)
It wasn’t until a gathering in Hawaii, in late August 2000, that the two realized the overlap of their work.
To celebrate his team’s finding, Mignot had invited a group of researchers to Big Island. With his paper scheduled for publication on September 1, he felt comfortable presenting his findings to his guests, which included Siegel.
Until then, “I didn’t know what he had found, and he didn’t know what I had found, which basically was the same thing,” said Siegel.
In yet another strange twist, the two papers were published just weeks apart, simultaneously revealing that human narcoleptics have a depleted supply of the neurons that bind to hypocretin/orexin. The cause of the disorder was at last a certainty.
“Even if I was first, what does it matter? In the end, you need confirmation,” said Mignot. “You need multiple people to make sure that it’s true. It’s good science when things like this happen.”
How All of This Changed Medicine
Since these groundbreaking discoveries, the diagnosis of narcolepsy has become much simpler. Lab tests can now easily measure hypocretin in cerebrospinal fluid, providing a definitive diagnosis.
But the development of narcolepsy treatments has lagged — even though hypocretin/orexin replacement therapy is the obvious answer.
“Almost 25 years have elapsed, and there’s no such therapeutic on the market,” said Kilduff, who now works for SRI International, a non-profit research and development institute.
That’s partly because agonists — drugs that bind to receptors in the brain — are challenging to create, as this requires mimicking the activating molecule’s structure, like copying the grooves of an intricate key.
Antagonists, by comparison, are easier to develop. These act as a gate, blocking access to the receptors. As a result, drugs that promote sleep by thwarting hypocretin/orexin have emerged more quickly, providing a flurry of new options for people with insomnia. The first, suvorexant, was launched in 2014. Two others followed in recent years.
Researchers are hopeful a hypocretin/orexin agonist is on the horizon.
“This is a very hot area of drug development,” said Kilduff. “It’s just a matter of who’s going to get the drug to market first.”
One More Hypocretin/Orexin Surprise — and It Could Be The Biggest
Several years ago, Siegel’s lab received what was supposed to be a healthy human brain — one they could use as a comparison for narcoleptic brains. But researcher Thomas Thannickal, PhD, lead author of the UCLA study linking hypocretin loss to human narcolepsy, noticed something strange: This brain had significantly more hypocretin neurons than average.
Was this due to a seizure? A traumatic death? Siegel called the brain bank to request the donor’s records. He was told they were missing.
Years later, Siegel happened to be visiting the brain bank for another project and found himself in a room adjacent to the medical records. “Nobody was there,” he said, “so I just opened a drawer.”
Shuffling through the brain bank’s files, Siegel found the medical records he’d been told were lost. In the file was a note from the donor, explaining that he was a former heroin addict.
“I almost fell out of my chair,” said Siegel. “I realized this guy’s heroin addiction likely had something to do with his very unusual brain.”
Obviously, opioids affected the orexin system. But how?
“It’s when people are happy that this peptide is released,” said Siegel. “The hypocretin system is not just related to alertness. It’s related to pleasure.”
As Yanagisawa observed early on, hypocretin/orexin does indeed play a role in eating — just not the one he initially thought. The peptides prompted pleasure seeking. So the rodents ate.
In 2018, after acquiring five more brains, Siegel’s group published a study in Translational Medicine showing 54% more detectable hypocretin neurons in the brains of heroin addicts than in those of control individuals.
In 2022, another breakthrough: His team showed that morphine significantly altered the pathways of hypocretin neurons in mice, sending their axons into brain regions associated with addiction. Then, when they removed the mice’s hypocretin neurons and discontinued their daily morphine dose, the rodents showed no symptoms of opioid withdrawal.
This fits the connection with narcolepsy: Among the standard treatments for the condition are amphetamines and other stimulants, which all have addictive potential. Yet, “narcoleptics never abuse these drugs,” Siegel said. “They seem to be uniquely resistant to addiction.”
This could powerfully change the way opioids are administered.
“If you prevent the hypocretin response to opioids, you may be able to prevent opioid addiction,” said Siegel. In other words, blocking the hypocretin system with a drug like those used to treat insomnia may allow patients to experience the pain-relieving benefits of opioids — without the risk for addiction.
His team is currently investigating treatments targeting the hypocretin/orexin system for opioid addiction.
In a study published in July, they found that mice who received suvorexant — the drug for insomnia — didn’t anticipate their daily dose of opioids the way other rodents did. This suggests the medication prevented addiction, without diminishing the pain-relieving effect of opioids.
If it translates to humans, this discovery could potentially save millions of lives.
“I think it’s just us working on this,” said Siegel.
But with hypocretin/orexin, you never know.
A version of this article appeared on Medscape.com.
It was 1996, and Masashi Yanagisawa was on the brink of his next discovery.
The Japanese scientist had arrived at the University of Texas Southwestern in Dallas 5 years earlier, setting up his own lab at age 31. After earning his medical degree, he’d gained notoriety as a PhD student when he discovered endothelin, the body’s most potent vasoconstrictor.
Yanagisawa was about to prove this wasn’t a first-timer’s fluke.
His focus was G-protein–coupled receptors (GPCRs), cell surface receptors that respond to a range of molecules and a popular target for drug discovery. The Human Genome Project had just revealed a slew of newly discovered receptors, or “orphan” GPCRs, and identifying an activating molecule could yield a new drug. (That vasoconstrictor endothelin was one such success story, leading to four new drug approvals in the United States over the past quarter century.)
Yanagisawa and his team created 50 cell lines, each expressing one orphan receptor. They applied animal tissue to every line, along with a calcium-sensitive dye. If the cells glowed under the microscope, they had a hit.
“He was basically doing an elaborate fishing expedition,” said Jon Willie, MD, PhD, an associate professor of neurosurgery at Washington University School of Medicine in St. Louis, Missouri, who would later join Yanagisawa’s team.
It wasn’t long before the neon-green fluorescence signaled a match. After isolating the activating molecule, the scientists realized they were dealing with two neuropeptides.
No one had ever seen these proteins before. And no one knew their discovery would set off a decades-long journey that would finally solve a century-old medical mystery — and may even fix one of the biggest health crises of our time, as revealed by research published earlier in 2024. It’s a story of strange coincidences, serendipitous discoveries, and quirky details. Most of all, it’s a fascinating example of how basic science can revolutionize medicine — and how true breakthroughs happen over time and in real time.
But That’s Basic Science for You
Most basic science studies — the early, foundational research that provides the building blocks for science that follows — don’t lead to medical breakthroughs. But some do, often in surprising ways.
Also called curiosity-driven research, basic science aims to fill knowledge gaps to keep science moving, even if the trajectory isn’t always clear.
“The people working on the basic research that led to discoveries that transformed the modern world had no idea at the time,” said Isobel Ronai, PhD, a postdoctoral fellow in life sciences at Harvard University, Cambridge, Massachusetts. “Often, these stories can only be seen in hindsight,” sometimes decades later.
Case in point: For molecular biology techniques — things like DNA sequencing and gene targeting — the lag between basic science and breakthrough is, on average, 23 years. While many of the resulting techniques have received Nobel Prizes, few of the foundational discoveries have been awarded such accolades.
“The scientific glory is more often associated with the downstream applications,” said Ronai. “The importance of basic research can get lost. But it is the foundation for any future application, such as drug development.”
As funding is increasingly funneled toward applied research, basic science can require a certain persistence. What this under-appreciation can obscure is the pathway to discovery — which is often as compelling as the end result, full of unpredictable twists, turns, and even interpersonal intrigue.
And then there’s the fascinating — and definitely complicated — phenomenon of multiple independent discoveries.
As in: What happens when two independent teams discover the same thing at the same time?
Back to Yanagisawa’s Lab ...
... where he and his team learned a few things about those new neuropeptides. Rat brain studies pinpointed the lateral hypothalamus as the peptides’ area of activity — a region often called the brain’s feeding center.
“If you destroy that part of the brain, animals lose appetite,” said Yanagisawa. So these peptides must control feeding, the scientists thought.
Sure enough, injecting the proteins into rat brains led the rodents to start eating.
Satisfied, the team named them “orexin-A” and “orexin-B,” for the Greek word “orexis,” meaning appetite. The brain receptors became “orexin-1” and “orexin-2.” The team prepared to publish its findings in Cell.
But another group beat them to it.
Introducing the ‘Hypocretins’
In early January 1998, a team of Scripps Research Institute scientists, led by J. Gregor Sutcliffe, PhD, released a paper in the journal PNAS. They described a gene encoding for the precursor to two neuropeptides
As the peptides were in the hypothalamus and structurally like secretin (a gut hormone), they called them “hypocretins.” The hypocretin peptides excited neurons in the hypothalamus, and later that year, the scientists discovered that the neurons’ branches extended, tentacle-like, throughout the brain. “Many of the connected areas were involved in sleep-wake control,” said Thomas Kilduff, PhD, who joined the Sutcliffe lab just weeks before the hypocretin discovery. At the time, however, the significance of this finding was not yet clear.
Weeks later, in February 1998, Yanagisawa’s paper came out.
Somehow, two groups, over 1000 miles apart, had stumbled on the same neuropeptides at the same time.
“I first heard about [Yanagisawa’s] paper on NBC Nightly News,” recalls Kilduff. “I was skiing in the mountains, so I had to wait until Monday to get back to the lab to see what the paper was all about.”
He realized that Yanagisawa’s orexin was his lab’s hypocretin, although the study didn’t mention another team’s discovery.
“There may have been accusations. But as far as I know, it’s because [Yanagisawa] didn’t know [about the other paper],” said Willie. “This was not something he produced in 2 months. This was clearly years of work.”
‘Multiple Discovery’ Happens More Often Than You Think
In the mid-20th century, sociologist Robert Merton described the phenomenon of “multiple discovery,” where many scientific discoveries or inventions are made independently at roughly the same time.
“This happens much more frequently in scientific research than people suppose,” said David Pendlebury, head of research analysis at Clarivate’s Institute for Scientific Information, the analytics company’s research arm. (Last year, Pendlebury flagged the hypocretin/orexin discovery for Clarivate’s prestigious Citations Laureates award, an honor that aims to predict, often successfully, who will go on to win the Nobel Prize.)
“People have this idea of the lone researcher making a brilliant discovery,” Pendlebury said. “But more and more, teams find things at the same time.”
While this can — and does — lead to squabbling about who deserves credit, the desire to be first can also be highly motivating, said Mike Schneider, PhD, an assistant professor of philosophy at the University of Missouri, Columbia, who studies the social dynamics of science, potentially leading to faster scientific advancement.
The downside? If two groups produce the same or similar results, but one publishes first, scientific journals tend to reject the second, citing a lack of novelty.
Yet duplicating research is a key step in confirming the validity of a discovery.
That’s why, in 2018, the journal PLOS Biology created a provision for “scooped” scientists, allowing them to submit their paper within 6 months of the first as a complementary finding. Instead of viewing this as redundancy, the editors believe it adds robustness to the research.
‘What the Heck Is This Mouse Doing?’
Even though he’d been scooped, Yanagisawa forged on to the next challenge: Confirming whether orexin regulated feeding.
He began breeding mice missing the orexin gene. His team expected these “knockout” mice to eat less, resulting in a thinner body than other rodents. To the contrary, “they were on average fatter,” said Willie. “They were eating less but weighed more, indicating a slower metabolism.”
The researchers were befuddled. “We were really disappointed, almost desperate about what to do,” said Yanagisawa.
As nocturnal animals eat more at night, he decided they should study the mice after dark. One of his students, Richard Chemelli, MD, bought an infrared video camera from Radio Shack, filming the first 4 hours of the mice’s active period for several nights.
After watching the footage, “Rick called me and said, ‘Let’s get into the lab,’ ” said Willie. “It was four of us on a Saturday looking at these videos, saying, ‘What the heck is this mouse doing?’ ”
While exploring their habitat, the knockout mice would randomly fall over, pop back up after a minute or so, and resume normal activity. This happened over and over — and the scientists were unsure why.
They began monitoring the mice’s brains during these episodes — and made a startling discovery.
The mice weren’t having seizures. They were shifting directly into REM sleep, bypassing the non-REM stage, then quickly toggling back to wake mode.
“That’s when we knew these animals had something akin to narcolepsy,” said Willie.
The team recruited Thomas Scammell, MD, a Harvard neurologist, to investigate whether modafinil — an anti-narcoleptic drug without a clear mechanism — affected orexin neurons.
Two hours after injecting the mice with the medication, the scientists sacrificed them and stained their brains. Remarkably, the number of neurons showing orexin activity had increased ninefold. It seemed modafinil worked by activating the orexin system.
These findings had the potential to crack open the science of narcolepsy, one of the most mysterious sleep disorders.
Unless, of course, another team did it first.
The Mystery of Narcolepsy
Yet another multiple discovery, narcolepsy was first described by two scientists — one in Germany, the other in France — within a short span in the late 1800s.
It would be more than a hundred years before anyone understood the disorder’s cause, even though it affects about 1 in 2000 people.
“Patients were often labeled as lazy and malingerers,” said Kilduff, “since they were sleepy all the time and had this weird motor behavior called cataplexy” or the sudden loss of muscle tone.
In the early 1970s, William Dement, MD, PhD — “the father of sleep medicine” — was searching for a narcoleptic cat to study. He couldn’t find a feline, but several colleagues mentioned dogs with narcolepsy-like symptoms.
Dement, who died in 2020, had found his newest research subjects.
In 1973, he started a narcoleptic dog colony at Stanford University in Palo Alto, California. At first, he focused on poodles and beagles. After discovering their narcolepsy wasn’t genetic, he pivoted to dobermans and labradors. Their narcolepsy was inherited, so he could breed them to populate the colony.
Although human narcolepsy is rarely genetic, it’s otherwise a lot like the version in these dogs.
Both involve daytime sleepiness, “pathological” bouts of REM sleep, and the loss of muscle tone in response to emotions, often positive ones.
The researchers hoped the canines could unlock a treatment for human narcolepsy. They began laying out a path of dog kibble, then injecting the dogs with drugs such as selective serotonin reuptake inhibitors. They wanted to see what might help them stay awake as they excitedly chowed down.
Kilduff also started a molecular genetics program, trying to identify the genetic defect behind canine narcolepsy. But after a parvovirus outbreak, Kilduff resigned from the project, drained from the strain of seeing so many dogs die.
A decade after his departure from the dog colony, his work would dramatically intersect with that of his successor, Emmanuel Mignot, MD, PhD.
“I thought I had closed the narcolepsy chapter in my life forever,” said Kilduff. “Then in 1998, we described this novel neuropeptide, hypocretin, that turned out to be the key to understanding the disorder.”
Narcoleptic Dogs in California, Mutant Mice in Texas
It was modafinil — the same anti-narcoleptic drug Yanagisawa’s team studied — that brought Emmanuel Mignot to the United States. After training as a pharmacologist in France, his home country sent him to Stanford to study the drug, which was discovered by French scientists, as his required military service.
As Kilduff’s replacement at the dog colony, his goal was to figure out how modafinil worked, hoping to attract a US company to develop the drug.
The plan succeeded. Modafinil became Provigil, a billion-dollar narcolepsy drug, and Mignot became “completely fascinated” with the disorder.
“I realized quickly that there was no way we’d find the cause of narcolepsy by finding the mode of action of this drug,” Mignot said. “Most likely, the drug was acting downstream, not at the cause of the disorder.”
To discover the answer, he needed to become a geneticist. And so began his 11-year odyssey to find the cause of canine narcolepsy.
After mapping the dog genome, Mignot set out to find the smallest stretch of chromosome that the narcoleptic animals had in common. “For a very long time, we were stuck with a relatively large region [of DNA],” he recalls. “It was a no man’s land.”
Within that region was the gene for the hypocretin/orexin-2 receptor — the same receptor that Yanagisawa had identified in his first orexin paper. Mignot didn’t immediately pursue that gene as a possibility — even though his students suggested it. Why?
“The decision was simply: Should we lose time to test a possible candidate [gene] among many?” Mignot said.
As Mignot studied dog DNA in California, Yanagisawa was creating mutant mice in Texas. Unbeknownst to either scientist, their work was about to converge.
What Happened Next Is Somewhat Disputed
After diagnosing his mice with narcolepsy, Yanagisawa opted not to share this finding with Mignot, though he knew about Mignot’s interest in the condition. Instead, he asked a colleague to find out how far along Mignot was in his genetics research.
According to Yanagisawa, his colleague didn’t realize how quickly DNA sequencing could happen once a target gene was identified. At a sleep meeting, “he showed Emmanuel all of our raw data. Almost accidentally, he disclosed our findings,” he said. “It was a shock for me.”
Unsure whether he was part of the orexin group, Mignot decided not to reveal that he’d identified the hypocretin/orexin-2 receptor gene as the faulty one in his narcoleptic dogs.
Although he didn’t share this finding, Mignot said he did offer to speak with the lead researcher to see if their findings were the same. If they were, they could jointly submit their articles. But Mignot never heard back.
Meanwhile, back at his lab, Mignot buckled down. While he wasn’t convinced the mouse data proved anything, it did give him the motivation to move faster.
Within weeks, he submitted his findings to Cell, revealing a mutation in the hypocretin/orexin-2 receptor gene as the cause of canine narcolepsy. According to Yanagisawa, the journal’s editor invited him to peer-review the paper, tipping him off to its existence.
“I told him I had a conflict of interest,” said Yanagisawa. “And then we scrambled to finish our manuscript. We wrote up the paper within almost 5 days.”
For a moment, it seemed both papers would be published together in Cell. Instead, on August 6, 1999, Mignot’s study was splashed solo across the journal’s cover.
“At the time, our team was pissed off, but looking back, what else could Emmanuel have done?” said Willie, who was part of Yanagisawa’s team. “The grant he’d been working on for years was at risk. He had it within his power to do the final experiments. Of course he was going to finish.”
Two weeks later, Yanagisawa’s findings followed, also in Cell.
His paper proposed knockout mice as a model for human narcolepsy and orexin as a key regulator of the sleep/wake cycle. With orexin-activated neurons branching into other areas of the brain, the peptide seemed to promote wakefulness by synchronizing several arousal neurotransmitters, such as serotonin, norepinephrine, and histamine.
“If you don’t have orexin, each of those systems can still function, but they’re not as coordinated,” said Willie. “If you have narcolepsy, you’re capable of wakefulness, and you’re capable of sleep. What you can’t do is prevent inappropriately switching between states.”
Together, the two papers painted a clear picture: Narcolepsy was the result of a dysfunction in the hypocretin/orexin system.
After more than a century, the cause of narcolepsy was starting to come into focus.
“This was blockbuster,” said Willie.
By itself, either finding — one in dogs, one in mice — might have been met with skepticism. But in combination, they offered indisputable evidence about narcolepsy’s cause.
The Human Brains in Your Fridge Hold Secrets
Jerome Siegel had been searching for the cause of human narcolepsy for years. A PhD and professor at the University of California, Los Angeles, he had managed to acquire four human narcoleptic brains. As laughter is often the trigger for the sudden shift to REM sleep in humans, he focused on the amygdala, an area linked to emotion.
“I looked in the amygdala and didn’t see anything,” he said. “So the brains stayed in my refrigerator for probably 10 years.”
Then he was invited to review Yanagisawa’s study in Cell. The lightbulb clicked on: Maybe the hypothalamus — not the amygdala — was the area of abnormality. He and his team dug out the decade-old brains.
When they stained the brains, the massive loss of hypocretin-activated neurons was hard to miss: On average, the narcoleptic brains had only about 7000 of the cells versus 70,000 in the average human brain. The scientists also noticed scar tissue in the hypothalamus, indicating that the neurons had at some point died, rather than being absent from birth.
What Siegel didn’t know: Mignot had also acquired a handful of human narcoleptic brains.
Already, he had coauthored a study showing that hypocretin/orexin was undetectable in the cerebrospinal fluid of the majority of the people with narcolepsy his team tested. It seemed clear that the hypocretin/orexin system was flawed — or even broken — in people with the condition.
“It looked like the cause of narcolepsy in humans was indeed this lack of orexin in the brain,” he said. “That was the hypothesis immediately. To me, this is when we established that narcolepsy in humans was due to a lack of orexin. The next thing was to check that the cells were missing.”
Now he could do exactly that.
As expected, Mignot’s team observed a dramatic loss of hypocretin/orexin cells in the narcoleptic brains. They also noticed that a different cell type in the hypothalamus was unaffected. This implied the damage was specific to the hypocretin-activated cells and supported a hunch they already had: That the deficit was the result not of a genetic defect but of an autoimmune attack. (It’s a hypothesis Mignot has spent the last 15 years proving.)
It wasn’t until a gathering in Hawaii, in late August 2000, that the two realized the overlap of their work.
To celebrate his team’s finding, Mignot had invited a group of researchers to Big Island. With his paper scheduled for publication on September 1, he felt comfortable presenting his findings to his guests, which included Siegel.
Until then, “I didn’t know what he had found, and he didn’t know what I had found, which basically was the same thing,” said Siegel.
In yet another strange twist, the two papers were published just weeks apart, simultaneously revealing that human narcoleptics have a depleted supply of the neurons that bind to hypocretin/orexin. The cause of the disorder was at last a certainty.
“Even if I was first, what does it matter? In the end, you need confirmation,” said Mignot. “You need multiple people to make sure that it’s true. It’s good science when things like this happen.”
How All of This Changed Medicine
Since these groundbreaking discoveries, the diagnosis of narcolepsy has become much simpler. Lab tests can now easily measure hypocretin in cerebrospinal fluid, providing a definitive diagnosis.
But the development of narcolepsy treatments has lagged — even though hypocretin/orexin replacement therapy is the obvious answer.
“Almost 25 years have elapsed, and there’s no such therapeutic on the market,” said Kilduff, who now works for SRI International, a non-profit research and development institute.
That’s partly because agonists — drugs that bind to receptors in the brain — are challenging to create, as this requires mimicking the activating molecule’s structure, like copying the grooves of an intricate key.
Antagonists, by comparison, are easier to develop. These act as a gate, blocking access to the receptors. As a result, drugs that promote sleep by thwarting hypocretin/orexin have emerged more quickly, providing a flurry of new options for people with insomnia. The first, suvorexant, was launched in 2014. Two others followed in recent years.
Researchers are hopeful a hypocretin/orexin agonist is on the horizon.
“This is a very hot area of drug development,” said Kilduff. “It’s just a matter of who’s going to get the drug to market first.”
One More Hypocretin/Orexin Surprise — and It Could Be The Biggest
Several years ago, Siegel’s lab received what was supposed to be a healthy human brain — one they could use as a comparison for narcoleptic brains. But researcher Thomas Thannickal, PhD, lead author of the UCLA study linking hypocretin loss to human narcolepsy, noticed something strange: This brain had significantly more hypocretin neurons than average.
Was this due to a seizure? A traumatic death? Siegel called the brain bank to request the donor’s records. He was told they were missing.
Years later, Siegel happened to be visiting the brain bank for another project and found himself in a room adjacent to the medical records. “Nobody was there,” he said, “so I just opened a drawer.”
Shuffling through the brain bank’s files, Siegel found the medical records he’d been told were lost. In the file was a note from the donor, explaining that he was a former heroin addict.
“I almost fell out of my chair,” said Siegel. “I realized this guy’s heroin addiction likely had something to do with his very unusual brain.”
Obviously, opioids affected the orexin system. But how?
“It’s when people are happy that this peptide is released,” said Siegel. “The hypocretin system is not just related to alertness. It’s related to pleasure.”
As Yanagisawa observed early on, hypocretin/orexin does indeed play a role in eating — just not the one he initially thought. The peptides prompted pleasure seeking. So the rodents ate.
In 2018, after acquiring five more brains, Siegel’s group published a study in Translational Medicine showing 54% more detectable hypocretin neurons in the brains of heroin addicts than in those of control individuals.
In 2022, another breakthrough: His team showed that morphine significantly altered the pathways of hypocretin neurons in mice, sending their axons into brain regions associated with addiction. Then, when they removed the mice’s hypocretin neurons and discontinued their daily morphine dose, the rodents showed no symptoms of opioid withdrawal.
This fits the connection with narcolepsy: Among the standard treatments for the condition are amphetamines and other stimulants, which all have addictive potential. Yet, “narcoleptics never abuse these drugs,” Siegel said. “They seem to be uniquely resistant to addiction.”
This could powerfully change the way opioids are administered.
“If you prevent the hypocretin response to opioids, you may be able to prevent opioid addiction,” said Siegel. In other words, blocking the hypocretin system with a drug like those used to treat insomnia may allow patients to experience the pain-relieving benefits of opioids — without the risk for addiction.
His team is currently investigating treatments targeting the hypocretin/orexin system for opioid addiction.
In a study published in July, they found that mice who received suvorexant — the drug for insomnia — didn’t anticipate their daily dose of opioids the way other rodents did. This suggests the medication prevented addiction, without diminishing the pain-relieving effect of opioids.
If it translates to humans, this discovery could potentially save millions of lives.
“I think it’s just us working on this,” said Siegel.
But with hypocretin/orexin, you never know.
A version of this article appeared on Medscape.com.
Cutting-edge nasal tech could usher in a new era of medicine
Noses are like caverns – twisting, turning, no two exactly the same. But if you nose past anyone’s nostrils, you’ll discover a surprisingly sprawling space.
“The size of the nasal cavity is about the same as a large handkerchief,” said Hugh Smyth, PhD, a professor of molecular pharmaceutics and drug delivery at the University of Texas at Austin.
“It’s very accessible tissue, and it has a lot of blood flow,” said Dr. Smyth. “The speed of onset can often be as fast as injections, sometimes even faster.”
It’s nothing new to get medicines via your nose. For decades, we’ve squirted various sprays into our nostrils to treat local maladies like allergies or infections. Even the ancients saw wisdom in the nasal route.
But recently, the nose has gained scientific attention as a gateway to the rest of the body – even the brain, a notoriously difficult target.
The upshot: Someday, inhaling therapies could be as routine as swallowing pills.
The nasal route is quick, needle free, and user friendly, and it often requires a smaller dose than other methods, since the drug doesn’t have to pass through the digestive tract, losing potency during digestion.
But there are challenges.
How hard can it be?
Old-school nasal sprayers, mostly unchanged since the 1800s, aren’t cut out for deep-nose delivery. “The technology is relatively limited because you’ve just got a single spray nozzle,” said Michael Hindle, PhD, a professor of pharmaceutics at Virginia Commonwealth University, Richmond.
These traditional devices (similar to perfume sprayers) don’t consistently push meds past the lower to middle sections inside the nose, called the nasal valve – if they do so at all: In a 2020 Rhinology study (doi: 10.4193/Rhin18.304) conventional nasal sprays only reached this first segment of the nose, a less-than-ideal spot to land.
Inside the nasal valve, the surface is skin-like and doesn’t absorb very well. Its narrow design slows airflow, preventing particles from moving to deeper regions, where tissue is vascular and porous like the lungs. And even if this structural roadblock is surpassed, other hurdles remain.
The nose is designed to keep stuff out. Nose hair, cilia, mucus, sneezing, coughing – all make “distributing drugs evenly across the nasal cavity difficult,” said Dr. Smyth. “The spray gets filtered out before it reaches those deeper zones,” potentially dripping out of the nostrils instead of being absorbed.
Complicating matters is how every person’s nose is different. In a 2018 study, Dr. Smyth and a research team created three dimensional–printed models of people’s nasal cavities. They varied widely. “Nasal cavities are very different in size, length, and internal geometry,” he said. “This makes it challenging to target specific areas.”
Although carefully positioning the spray nozzle can help, even something as minor as sniffing too hard (constricting the nostrils) can keep sprays from reaching the absorptive deeper regions.
Still, the benefits are enough to compel researchers to find a way in.
“This really is a drug delivery challenge we’ve been wrestling with,” said Dr. Hindle. “It’s not new formulations we hear about. It’s new devices and delivery methods trying to target the different nasal regions.”
Delivering the goods
In the late aughts, John Hoekman was a graduate student in the University of Washington’s pharmaceutics program, studying nasal drug delivery. In his experiments, he noticed that drugs distributed differently, depending on the region targeted – aiming for the upper nasal cavity led to a spike in absorption.
The results convinced Mr. Hoekman to stake his future on nasal drug delivery.
In 2008, while still in graduate school, he started his own company, now known as Impel Pharmaceuticals. In 2021, Impel released its first product: Trudhesa, a nasal spray for migraines. Although the drug itself – dihydroergotamine mesylate – was hardly novel, used for migraine relief since 1946 (Headache. 2020 Jan;60[1]:40-57), it was usually delivered through an intravenous line, often in the ED.
But with Mr. Hoekman’s POD device – short for precision olfactory delivery – the drug can be given by the patient, via the nose. This generally means faster, more reliable relief, with fewer side effects. “We were able to lower the dose and improve the overall absorption,” said Mr. Hoekman.
The POD’s nozzle is engineered to spray a soft, narrow plume. It’s gas propelled, so patients don’t have to breathe in any special way to ensure delivery. The drug can zip right through the nasal valve into the upper nasal cavity.
Another company – OptiNose – has a “bidirectional” delivery method that propels drugs, either liquid or dry powder, deep into the nose.
“You insert the nozzle into your nose, and as you blow through the mouthpiece, your soft palate closes,” said Dr. Hindle. With the throat sealed off, “the only place for the drug to go is into one nostril and out the other, coating both sides of the nasal passageways.”
The device is only available for Onzetra Xsail, a powder for migraines. But another application is on its way.
In May, OptiNose announced that the FDA is reviewing Xhance, which uses the system to direct a steroid to the sinuses. In a clinical trial, patients with chronic sinusitis who tried the drug-device combo saw a decline in congestion, facial pain, and inflammation.
Targeting the brain
Both of those migraine drugs – Trudhesa and Onzetra Xsail – are thought to penetrate the upper nasal cavity. That’s where you’ll find the olfactory zone, a sheet of neurons that connects to the olfactory bulb. Located behind the eyes, these two nerve bundles detect odors.
“The olfactory region is almost like a back door to the brain,” said Mr. Hoekman.
By bypassing the blood-brain barrier, it offers a direct pathway – the only direct pathway, actually – between an exposed area of the body and the brain. Meaning it can ferry drugs straight from the nasal cavity to the central nervous system.
Nose-to-brain treatments could be game-changing for central nervous system disorders, such as Parkinson’s disease, Alzheimer’s, or anxiety.
But reaching the olfactory zone is notoriously hard. “The vasculature in your nose is like a big freeway, and the olfactory tract is like a side alley,” explained Mr. Hoekman. “It’s very limiting in what it will allow through.” The region is also small, occupying only 3%-10% of the nasal cavity’s surface area.
Again, POD means “precision olfactory delivery.” But the device isn’t quite as laser focused on the region as its name implies. “We’re not at the stage where we’re able to exclusively deliver to one target site in the nose,” said Dr. Hindle.
While wending its way toward the olfactory zone, some of the drug will be absorbed by other regions, then circulate throughout the body.
“About 59% of the drug that we put into the upper nasal space gets absorbed into the bloodstream,” said Mr. Hoekman.
Janssen Pharmaceuticals’ Spravato – a nasal spray for drug-resistant depression – is thought to work similarly: Some goes straight to the brain via the olfactory nerves, while the rest takes a more roundabout route, passing through the blood vessels to circulate in your system.
A needle-free option
Sometimes, the bloodstream is the main target. Because the nose’s middle and upper stretches are so vascular, drugs can be rapidly absorbed.
This is especially valuable for time-sensitive conditions. “If you give something nasally, you can have peak uptake in 15-30 minutes,” said Mr. Hoekman.
Take Narcan nasal spray, which delivers a burst of naloxone to quickly reverse the effects of opioid an overdose. Or Noctiva nasal spray. Taken just half an hour before bed, it can prevent frequent nighttime urination.
There’s also a group of seizure-stopping sprays, known as “rescue treatments.” One works by temporarily loosening the space between nasal cells, allowing the seizure drug to be quickly absorbed through the vessels.
This systemic access also has potential for drugs that would otherwise have to be injected, such as biologics.
The same goes for vaccines. Mucosal tissue inside the nasal cavity offers direct access to the infection-fighting lymphatic system, making the nose a prime target for inoculation against certain viruses.
Inhaling protection against viruses
Despite the recent surge of interest, nasal vaccines faced a rocky start. After the first nasal flu vaccine hit the market in 2001, it was pulled due to potential toxicity and reports of Bell’s palsy, a type of facial paralysis.
FluMist came in 2003 and has been plagued by problems ever since. Because it contains a weakened live virus, flu-like side effects can occur. And it doesn’t always work. During the 2016-2017 flu season, FluMist protected only 3% of kids, prompting the Centers for Disease Control and Prevention to advise against the nasal route that year.
Why FluMist can be so hit-or-miss is poorly understood. But generally, the nose can pose an effectiveness challenge. “The nose is highly cycling,” said Dr. Hindle. “Anything we deposit usually gets transported out within 15-20 minutes.”
For kids – big fans of not using needles – chronically runny noses can be an issue. “You squirt it in the nose, and it will probably just come back out in their snot,” said Jay Kolls, MD, a professor of medicine and pediatrics at Tulane University, New Orleans, who is developing an intranasal pneumonia vaccine.
Even so, nasal vaccines became a hot topic among researchers after the world was shut down by a virus that invades through the nose.
“We realized that intramuscular vaccines were effective at preventing severe disease, but they weren’t that effective at preventing transmission,” said Michael Diamond, MD, PhD, an immunologist at Washington University in St. Louis.
Nasal vaccines could solve that problem by putting an immune barrier at the point of entry, denying access to the rest of the body. “You squash the infection early enough that it not only prevents disease,” said Dr. Kolls, “but potentially prevents transmission.”
And yes, a nasal COVID vaccine is on the way
In March 2020, Dr. Diamond’s team began exploring a nasal COVID vaccine. Promising results in animals prompted a vaccine development company to license the technology. The resulting nasal vaccine – the first for COVID – has been approved in India, both as a primary vaccine and a booster.
It works by stimulating an influx of IgA, a type of antibody found in the nasal passages, and production of resident memory T cells, immune cells on standby just beneath the surface tissue in the nose.
By contrast, injected vaccines generate mostly IgG antibodies, which struggle to enter the respiratory tract. Only a tiny fraction – an estimated 1% – typically reach the nose.
Nasal vaccines could also be used along with shots. The latter could prime the whole body to fight back, while a nasal spritz could pull that immune protection to the mucosal surfaces.
Nasal technology could yield more effective vaccines for infections like tuberculosis or malaria, or even safeguard against new – sometimes surprising – conditions.
In a 2021 Nature study, an intranasal vaccine derived from fentanyl was better at preventing overdose than an injected vaccine. “Through some clever chemistry, the drug [in the vaccine] isn’t fentanyl anymore,” said study author Elizabeth Norton, PhD, an assistant professor of microbiology and immunology at Tulane University. “But the immune system still has an antibody response to it.”
Novel applications like this represent the future of nasal drug delivery.
“We’re not going to innovate in asthma or COPD. We’re not going to innovate in local delivery to the nose,” said Dr. Hindle. “Innovation will only come if we look to treat new conditions.”
A version of this article originally appeared on WebMD.com.
Noses are like caverns – twisting, turning, no two exactly the same. But if you nose past anyone’s nostrils, you’ll discover a surprisingly sprawling space.
“The size of the nasal cavity is about the same as a large handkerchief,” said Hugh Smyth, PhD, a professor of molecular pharmaceutics and drug delivery at the University of Texas at Austin.
“It’s very accessible tissue, and it has a lot of blood flow,” said Dr. Smyth. “The speed of onset can often be as fast as injections, sometimes even faster.”
It’s nothing new to get medicines via your nose. For decades, we’ve squirted various sprays into our nostrils to treat local maladies like allergies or infections. Even the ancients saw wisdom in the nasal route.
But recently, the nose has gained scientific attention as a gateway to the rest of the body – even the brain, a notoriously difficult target.
The upshot: Someday, inhaling therapies could be as routine as swallowing pills.
The nasal route is quick, needle free, and user friendly, and it often requires a smaller dose than other methods, since the drug doesn’t have to pass through the digestive tract, losing potency during digestion.
But there are challenges.
How hard can it be?
Old-school nasal sprayers, mostly unchanged since the 1800s, aren’t cut out for deep-nose delivery. “The technology is relatively limited because you’ve just got a single spray nozzle,” said Michael Hindle, PhD, a professor of pharmaceutics at Virginia Commonwealth University, Richmond.
These traditional devices (similar to perfume sprayers) don’t consistently push meds past the lower to middle sections inside the nose, called the nasal valve – if they do so at all: In a 2020 Rhinology study (doi: 10.4193/Rhin18.304) conventional nasal sprays only reached this first segment of the nose, a less-than-ideal spot to land.
Inside the nasal valve, the surface is skin-like and doesn’t absorb very well. Its narrow design slows airflow, preventing particles from moving to deeper regions, where tissue is vascular and porous like the lungs. And even if this structural roadblock is surpassed, other hurdles remain.
The nose is designed to keep stuff out. Nose hair, cilia, mucus, sneezing, coughing – all make “distributing drugs evenly across the nasal cavity difficult,” said Dr. Smyth. “The spray gets filtered out before it reaches those deeper zones,” potentially dripping out of the nostrils instead of being absorbed.
Complicating matters is how every person’s nose is different. In a 2018 study, Dr. Smyth and a research team created three dimensional–printed models of people’s nasal cavities. They varied widely. “Nasal cavities are very different in size, length, and internal geometry,” he said. “This makes it challenging to target specific areas.”
Although carefully positioning the spray nozzle can help, even something as minor as sniffing too hard (constricting the nostrils) can keep sprays from reaching the absorptive deeper regions.
Still, the benefits are enough to compel researchers to find a way in.
“This really is a drug delivery challenge we’ve been wrestling with,” said Dr. Hindle. “It’s not new formulations we hear about. It’s new devices and delivery methods trying to target the different nasal regions.”
Delivering the goods
In the late aughts, John Hoekman was a graduate student in the University of Washington’s pharmaceutics program, studying nasal drug delivery. In his experiments, he noticed that drugs distributed differently, depending on the region targeted – aiming for the upper nasal cavity led to a spike in absorption.
The results convinced Mr. Hoekman to stake his future on nasal drug delivery.
In 2008, while still in graduate school, he started his own company, now known as Impel Pharmaceuticals. In 2021, Impel released its first product: Trudhesa, a nasal spray for migraines. Although the drug itself – dihydroergotamine mesylate – was hardly novel, used for migraine relief since 1946 (Headache. 2020 Jan;60[1]:40-57), it was usually delivered through an intravenous line, often in the ED.
But with Mr. Hoekman’s POD device – short for precision olfactory delivery – the drug can be given by the patient, via the nose. This generally means faster, more reliable relief, with fewer side effects. “We were able to lower the dose and improve the overall absorption,” said Mr. Hoekman.
The POD’s nozzle is engineered to spray a soft, narrow plume. It’s gas propelled, so patients don’t have to breathe in any special way to ensure delivery. The drug can zip right through the nasal valve into the upper nasal cavity.
Another company – OptiNose – has a “bidirectional” delivery method that propels drugs, either liquid or dry powder, deep into the nose.
“You insert the nozzle into your nose, and as you blow through the mouthpiece, your soft palate closes,” said Dr. Hindle. With the throat sealed off, “the only place for the drug to go is into one nostril and out the other, coating both sides of the nasal passageways.”
The device is only available for Onzetra Xsail, a powder for migraines. But another application is on its way.
In May, OptiNose announced that the FDA is reviewing Xhance, which uses the system to direct a steroid to the sinuses. In a clinical trial, patients with chronic sinusitis who tried the drug-device combo saw a decline in congestion, facial pain, and inflammation.
Targeting the brain
Both of those migraine drugs – Trudhesa and Onzetra Xsail – are thought to penetrate the upper nasal cavity. That’s where you’ll find the olfactory zone, a sheet of neurons that connects to the olfactory bulb. Located behind the eyes, these two nerve bundles detect odors.
“The olfactory region is almost like a back door to the brain,” said Mr. Hoekman.
By bypassing the blood-brain barrier, it offers a direct pathway – the only direct pathway, actually – between an exposed area of the body and the brain. Meaning it can ferry drugs straight from the nasal cavity to the central nervous system.
Nose-to-brain treatments could be game-changing for central nervous system disorders, such as Parkinson’s disease, Alzheimer’s, or anxiety.
But reaching the olfactory zone is notoriously hard. “The vasculature in your nose is like a big freeway, and the olfactory tract is like a side alley,” explained Mr. Hoekman. “It’s very limiting in what it will allow through.” The region is also small, occupying only 3%-10% of the nasal cavity’s surface area.
Again, POD means “precision olfactory delivery.” But the device isn’t quite as laser focused on the region as its name implies. “We’re not at the stage where we’re able to exclusively deliver to one target site in the nose,” said Dr. Hindle.
While wending its way toward the olfactory zone, some of the drug will be absorbed by other regions, then circulate throughout the body.
“About 59% of the drug that we put into the upper nasal space gets absorbed into the bloodstream,” said Mr. Hoekman.
Janssen Pharmaceuticals’ Spravato – a nasal spray for drug-resistant depression – is thought to work similarly: Some goes straight to the brain via the olfactory nerves, while the rest takes a more roundabout route, passing through the blood vessels to circulate in your system.
A needle-free option
Sometimes, the bloodstream is the main target. Because the nose’s middle and upper stretches are so vascular, drugs can be rapidly absorbed.
This is especially valuable for time-sensitive conditions. “If you give something nasally, you can have peak uptake in 15-30 minutes,” said Mr. Hoekman.
Take Narcan nasal spray, which delivers a burst of naloxone to quickly reverse the effects of opioid an overdose. Or Noctiva nasal spray. Taken just half an hour before bed, it can prevent frequent nighttime urination.
There’s also a group of seizure-stopping sprays, known as “rescue treatments.” One works by temporarily loosening the space between nasal cells, allowing the seizure drug to be quickly absorbed through the vessels.
This systemic access also has potential for drugs that would otherwise have to be injected, such as biologics.
The same goes for vaccines. Mucosal tissue inside the nasal cavity offers direct access to the infection-fighting lymphatic system, making the nose a prime target for inoculation against certain viruses.
Inhaling protection against viruses
Despite the recent surge of interest, nasal vaccines faced a rocky start. After the first nasal flu vaccine hit the market in 2001, it was pulled due to potential toxicity and reports of Bell’s palsy, a type of facial paralysis.
FluMist came in 2003 and has been plagued by problems ever since. Because it contains a weakened live virus, flu-like side effects can occur. And it doesn’t always work. During the 2016-2017 flu season, FluMist protected only 3% of kids, prompting the Centers for Disease Control and Prevention to advise against the nasal route that year.
Why FluMist can be so hit-or-miss is poorly understood. But generally, the nose can pose an effectiveness challenge. “The nose is highly cycling,” said Dr. Hindle. “Anything we deposit usually gets transported out within 15-20 minutes.”
For kids – big fans of not using needles – chronically runny noses can be an issue. “You squirt it in the nose, and it will probably just come back out in their snot,” said Jay Kolls, MD, a professor of medicine and pediatrics at Tulane University, New Orleans, who is developing an intranasal pneumonia vaccine.
Even so, nasal vaccines became a hot topic among researchers after the world was shut down by a virus that invades through the nose.
“We realized that intramuscular vaccines were effective at preventing severe disease, but they weren’t that effective at preventing transmission,” said Michael Diamond, MD, PhD, an immunologist at Washington University in St. Louis.
Nasal vaccines could solve that problem by putting an immune barrier at the point of entry, denying access to the rest of the body. “You squash the infection early enough that it not only prevents disease,” said Dr. Kolls, “but potentially prevents transmission.”
And yes, a nasal COVID vaccine is on the way
In March 2020, Dr. Diamond’s team began exploring a nasal COVID vaccine. Promising results in animals prompted a vaccine development company to license the technology. The resulting nasal vaccine – the first for COVID – has been approved in India, both as a primary vaccine and a booster.
It works by stimulating an influx of IgA, a type of antibody found in the nasal passages, and production of resident memory T cells, immune cells on standby just beneath the surface tissue in the nose.
By contrast, injected vaccines generate mostly IgG antibodies, which struggle to enter the respiratory tract. Only a tiny fraction – an estimated 1% – typically reach the nose.
Nasal vaccines could also be used along with shots. The latter could prime the whole body to fight back, while a nasal spritz could pull that immune protection to the mucosal surfaces.
Nasal technology could yield more effective vaccines for infections like tuberculosis or malaria, or even safeguard against new – sometimes surprising – conditions.
In a 2021 Nature study, an intranasal vaccine derived from fentanyl was better at preventing overdose than an injected vaccine. “Through some clever chemistry, the drug [in the vaccine] isn’t fentanyl anymore,” said study author Elizabeth Norton, PhD, an assistant professor of microbiology and immunology at Tulane University. “But the immune system still has an antibody response to it.”
Novel applications like this represent the future of nasal drug delivery.
“We’re not going to innovate in asthma or COPD. We’re not going to innovate in local delivery to the nose,” said Dr. Hindle. “Innovation will only come if we look to treat new conditions.”
A version of this article originally appeared on WebMD.com.
Noses are like caverns – twisting, turning, no two exactly the same. But if you nose past anyone’s nostrils, you’ll discover a surprisingly sprawling space.
“The size of the nasal cavity is about the same as a large handkerchief,” said Hugh Smyth, PhD, a professor of molecular pharmaceutics and drug delivery at the University of Texas at Austin.
“It’s very accessible tissue, and it has a lot of blood flow,” said Dr. Smyth. “The speed of onset can often be as fast as injections, sometimes even faster.”
It’s nothing new to get medicines via your nose. For decades, we’ve squirted various sprays into our nostrils to treat local maladies like allergies or infections. Even the ancients saw wisdom in the nasal route.
But recently, the nose has gained scientific attention as a gateway to the rest of the body – even the brain, a notoriously difficult target.
The upshot: Someday, inhaling therapies could be as routine as swallowing pills.
The nasal route is quick, needle free, and user friendly, and it often requires a smaller dose than other methods, since the drug doesn’t have to pass through the digestive tract, losing potency during digestion.
But there are challenges.
How hard can it be?
Old-school nasal sprayers, mostly unchanged since the 1800s, aren’t cut out for deep-nose delivery. “The technology is relatively limited because you’ve just got a single spray nozzle,” said Michael Hindle, PhD, a professor of pharmaceutics at Virginia Commonwealth University, Richmond.
These traditional devices (similar to perfume sprayers) don’t consistently push meds past the lower to middle sections inside the nose, called the nasal valve – if they do so at all: In a 2020 Rhinology study (doi: 10.4193/Rhin18.304) conventional nasal sprays only reached this first segment of the nose, a less-than-ideal spot to land.
Inside the nasal valve, the surface is skin-like and doesn’t absorb very well. Its narrow design slows airflow, preventing particles from moving to deeper regions, where tissue is vascular and porous like the lungs. And even if this structural roadblock is surpassed, other hurdles remain.
The nose is designed to keep stuff out. Nose hair, cilia, mucus, sneezing, coughing – all make “distributing drugs evenly across the nasal cavity difficult,” said Dr. Smyth. “The spray gets filtered out before it reaches those deeper zones,” potentially dripping out of the nostrils instead of being absorbed.
Complicating matters is how every person’s nose is different. In a 2018 study, Dr. Smyth and a research team created three dimensional–printed models of people’s nasal cavities. They varied widely. “Nasal cavities are very different in size, length, and internal geometry,” he said. “This makes it challenging to target specific areas.”
Although carefully positioning the spray nozzle can help, even something as minor as sniffing too hard (constricting the nostrils) can keep sprays from reaching the absorptive deeper regions.
Still, the benefits are enough to compel researchers to find a way in.
“This really is a drug delivery challenge we’ve been wrestling with,” said Dr. Hindle. “It’s not new formulations we hear about. It’s new devices and delivery methods trying to target the different nasal regions.”
Delivering the goods
In the late aughts, John Hoekman was a graduate student in the University of Washington’s pharmaceutics program, studying nasal drug delivery. In his experiments, he noticed that drugs distributed differently, depending on the region targeted – aiming for the upper nasal cavity led to a spike in absorption.
The results convinced Mr. Hoekman to stake his future on nasal drug delivery.
In 2008, while still in graduate school, he started his own company, now known as Impel Pharmaceuticals. In 2021, Impel released its first product: Trudhesa, a nasal spray for migraines. Although the drug itself – dihydroergotamine mesylate – was hardly novel, used for migraine relief since 1946 (Headache. 2020 Jan;60[1]:40-57), it was usually delivered through an intravenous line, often in the ED.
But with Mr. Hoekman’s POD device – short for precision olfactory delivery – the drug can be given by the patient, via the nose. This generally means faster, more reliable relief, with fewer side effects. “We were able to lower the dose and improve the overall absorption,” said Mr. Hoekman.
The POD’s nozzle is engineered to spray a soft, narrow plume. It’s gas propelled, so patients don’t have to breathe in any special way to ensure delivery. The drug can zip right through the nasal valve into the upper nasal cavity.
Another company – OptiNose – has a “bidirectional” delivery method that propels drugs, either liquid or dry powder, deep into the nose.
“You insert the nozzle into your nose, and as you blow through the mouthpiece, your soft palate closes,” said Dr. Hindle. With the throat sealed off, “the only place for the drug to go is into one nostril and out the other, coating both sides of the nasal passageways.”
The device is only available for Onzetra Xsail, a powder for migraines. But another application is on its way.
In May, OptiNose announced that the FDA is reviewing Xhance, which uses the system to direct a steroid to the sinuses. In a clinical trial, patients with chronic sinusitis who tried the drug-device combo saw a decline in congestion, facial pain, and inflammation.
Targeting the brain
Both of those migraine drugs – Trudhesa and Onzetra Xsail – are thought to penetrate the upper nasal cavity. That’s where you’ll find the olfactory zone, a sheet of neurons that connects to the olfactory bulb. Located behind the eyes, these two nerve bundles detect odors.
“The olfactory region is almost like a back door to the brain,” said Mr. Hoekman.
By bypassing the blood-brain barrier, it offers a direct pathway – the only direct pathway, actually – between an exposed area of the body and the brain. Meaning it can ferry drugs straight from the nasal cavity to the central nervous system.
Nose-to-brain treatments could be game-changing for central nervous system disorders, such as Parkinson’s disease, Alzheimer’s, or anxiety.
But reaching the olfactory zone is notoriously hard. “The vasculature in your nose is like a big freeway, and the olfactory tract is like a side alley,” explained Mr. Hoekman. “It’s very limiting in what it will allow through.” The region is also small, occupying only 3%-10% of the nasal cavity’s surface area.
Again, POD means “precision olfactory delivery.” But the device isn’t quite as laser focused on the region as its name implies. “We’re not at the stage where we’re able to exclusively deliver to one target site in the nose,” said Dr. Hindle.
While wending its way toward the olfactory zone, some of the drug will be absorbed by other regions, then circulate throughout the body.
“About 59% of the drug that we put into the upper nasal space gets absorbed into the bloodstream,” said Mr. Hoekman.
Janssen Pharmaceuticals’ Spravato – a nasal spray for drug-resistant depression – is thought to work similarly: Some goes straight to the brain via the olfactory nerves, while the rest takes a more roundabout route, passing through the blood vessels to circulate in your system.
A needle-free option
Sometimes, the bloodstream is the main target. Because the nose’s middle and upper stretches are so vascular, drugs can be rapidly absorbed.
This is especially valuable for time-sensitive conditions. “If you give something nasally, you can have peak uptake in 15-30 minutes,” said Mr. Hoekman.
Take Narcan nasal spray, which delivers a burst of naloxone to quickly reverse the effects of opioid an overdose. Or Noctiva nasal spray. Taken just half an hour before bed, it can prevent frequent nighttime urination.
There’s also a group of seizure-stopping sprays, known as “rescue treatments.” One works by temporarily loosening the space between nasal cells, allowing the seizure drug to be quickly absorbed through the vessels.
This systemic access also has potential for drugs that would otherwise have to be injected, such as biologics.
The same goes for vaccines. Mucosal tissue inside the nasal cavity offers direct access to the infection-fighting lymphatic system, making the nose a prime target for inoculation against certain viruses.
Inhaling protection against viruses
Despite the recent surge of interest, nasal vaccines faced a rocky start. After the first nasal flu vaccine hit the market in 2001, it was pulled due to potential toxicity and reports of Bell’s palsy, a type of facial paralysis.
FluMist came in 2003 and has been plagued by problems ever since. Because it contains a weakened live virus, flu-like side effects can occur. And it doesn’t always work. During the 2016-2017 flu season, FluMist protected only 3% of kids, prompting the Centers for Disease Control and Prevention to advise against the nasal route that year.
Why FluMist can be so hit-or-miss is poorly understood. But generally, the nose can pose an effectiveness challenge. “The nose is highly cycling,” said Dr. Hindle. “Anything we deposit usually gets transported out within 15-20 minutes.”
For kids – big fans of not using needles – chronically runny noses can be an issue. “You squirt it in the nose, and it will probably just come back out in their snot,” said Jay Kolls, MD, a professor of medicine and pediatrics at Tulane University, New Orleans, who is developing an intranasal pneumonia vaccine.
Even so, nasal vaccines became a hot topic among researchers after the world was shut down by a virus that invades through the nose.
“We realized that intramuscular vaccines were effective at preventing severe disease, but they weren’t that effective at preventing transmission,” said Michael Diamond, MD, PhD, an immunologist at Washington University in St. Louis.
Nasal vaccines could solve that problem by putting an immune barrier at the point of entry, denying access to the rest of the body. “You squash the infection early enough that it not only prevents disease,” said Dr. Kolls, “but potentially prevents transmission.”
And yes, a nasal COVID vaccine is on the way
In March 2020, Dr. Diamond’s team began exploring a nasal COVID vaccine. Promising results in animals prompted a vaccine development company to license the technology. The resulting nasal vaccine – the first for COVID – has been approved in India, both as a primary vaccine and a booster.
It works by stimulating an influx of IgA, a type of antibody found in the nasal passages, and production of resident memory T cells, immune cells on standby just beneath the surface tissue in the nose.
By contrast, injected vaccines generate mostly IgG antibodies, which struggle to enter the respiratory tract. Only a tiny fraction – an estimated 1% – typically reach the nose.
Nasal vaccines could also be used along with shots. The latter could prime the whole body to fight back, while a nasal spritz could pull that immune protection to the mucosal surfaces.
Nasal technology could yield more effective vaccines for infections like tuberculosis or malaria, or even safeguard against new – sometimes surprising – conditions.
In a 2021 Nature study, an intranasal vaccine derived from fentanyl was better at preventing overdose than an injected vaccine. “Through some clever chemistry, the drug [in the vaccine] isn’t fentanyl anymore,” said study author Elizabeth Norton, PhD, an assistant professor of microbiology and immunology at Tulane University. “But the immune system still has an antibody response to it.”
Novel applications like this represent the future of nasal drug delivery.
“We’re not going to innovate in asthma or COPD. We’re not going to innovate in local delivery to the nose,” said Dr. Hindle. “Innovation will only come if we look to treat new conditions.”
A version of this article originally appeared on WebMD.com.
How your voice could reveal hidden disease
: First during puberty, as the vocal cords thicken and the voice box migrates down the throat. Then a second time as aging causes structural changes that may weaken the voice.
But for some of us, there’s another voice shift, when a disease begins or when our mental health declines.
This is why more doctors are looking into voice as a biomarker – something that tells you that a disease is present.
Vital signs like blood pressure or heart rate “can give a general idea of how sick we are. But they’re not specific to certain diseases,” says Yael Bensoussan, MD, director of the University of South Florida, Tampa’s Health Voice Center and the coprincipal investigator for the National Institutes of Health’s Voice as a Biomarker of Health project.
“We’re learning that there are patterns” in voice changes that can indicate a range of conditions, including diseases of the nervous system and mental illnesses, she says.
Speaking is complicated, involving everything from the lungs and voice box to the mouth and brain. “A breakdown in any of those parts can affect the voice,” says Maria Powell, PhD, an assistant professor of otolaryngology (the study of diseases of the ear and throat) at Vanderbilt University, Nashville, Tenn., who is working on the NIH project.
You or those around you may not notice the changes. But researchers say voice analysis as a standard part of patient care – akin to blood pressure checks or cholesterol tests – could help identify those who need medical attention earlier.
Often, all it takes is a smartphone – “something that’s cheap, off-the-shelf, and that everyone can use,” says Ariana Anderson, PhD, director of the University of California, Los Angeles, Laboratory of Computational Neuropsychology.
“You can provide voice data in your pajamas, on your couch,” says Frank Rudzicz, PhD, a computer scientist for the NIH project. “It doesn’t require very complicated or expensive equipment, and it doesn’t require a lot of expertise to obtain.” Plus, multiple samples can be collected over time, giving a more accurate picture of health than a single snapshot from, say, a cognitive test.
Over the next 4 years, the Voice as a Biomarker team will receive nearly $18 million to gather a massive amount of voice data. The goal is 20,000-30,000 samples, along with health data about each person being studied. The result will be a sprawling database scientists can use to develop algorithms linking health conditions to the way we speak.
For the first 2 years, new data will be collected exclusively via universities and high-volume clinics to control quality and accuracy. Eventually, people will be invited to submit their own voice recordings, creating a crowdsourced dataset. “Google, Alexa, Amazon – they have access to tons of voice data,” says Dr. Bensoussan. “But it’s not usable in a clinical way, because they don’t have the health information.”
Dr. Bensoussan and her colleagues hope to fill that void with advance voice screening apps, which could prove especially valuable in remote communities that lack access to specialists or as a tool for telemedicine. Down the line, wearable devices with voice analysis could alert people with chronic conditions when they need to see a doctor.
“The watch says, ‘I’ve analyzed your breathing and coughing, and today, you’re really not doing well. You should go to the hospital,’ ” says Dr. Bensoussan, envisioning a wearable for patients with COPD. “It could tell people early that things are declining.”
Artificial intelligence may be better than a brain at pinpointing the right disease. For example, slurred speech could indicate Parkinson’s, a stroke, or ALS, among other things.
“We can hold approximately seven pieces of information in our head at one time,” says Dr. Rudzicz. “It’s really hard for us to get a holistic picture using dozens or hundreds of variables at once.” But a computer can consider a whole range of vocal markers at the same time, piecing them together for a more accurate assessment.
“The goal is not to outperform a ... clinician,” says Dr. Bensoussan. Yet the potential is unmistakably there: In a recent study of patients with cancer of the larynx, an automated voice analysis tool more accurately flagged the disease than laryngologists did.
“Algorithms have a larger training base,” says Dr. Anderson, who developed an app called ChatterBaby that analyzes infant cries. “We have a million samples at our disposal to train our algorithms. I don’t know if I’ve heard a million different babies crying in my life.”
So which health conditions show the most promise for voice analysis? The Voice as a Biomarker project will focus on five categories.
Voice disorders (cancers of the larynx, vocal fold paralysis, benign lesions on the larynx)
Obviously, vocal changes are a hallmark of these conditions, which cause things like breathiness or “roughness,” a type of vocal irregularity. Hoarseness that lasts at least 2 weeks is often one of the earliest signs of laryngeal cancer. Yet it can take months – one study found 16 weeks was the average – for patients to see a doctor after noticing the changes. Even then, laryngologists still misdiagnosed some cases of cancer when relying on vocal cues alone.
Now imagine a different scenario: The patient speaks into a smartphone app. An algorithm compares the vocal sample with the voices of laryngeal cancer patients. The app spits out the estimated odds of laryngeal cancer, helping providers decide whether to offer the patient specialist care.
Or consider spasmodic dysphonia, a neurological voice disorder that triggers spasms in the muscles of the voice box, causing a strained or breathy voice. Doctors who lack experience with vocal disorders may miss the condition. This is why diagnosis takes an average of nearly 4.5 years, according to a study in the Journal of Voice, and may include everything from allergy testing to psychiatric evaluation, says Dr. Powell. Artificial intelligence technology trained to recognize the disorder could help eliminate such unnecessary testing.
Neurological and neurodegenerative disorders (Alzheimer’s, Parkinson’s, stroke, ALS)
For Alzheimer’s and Parkinson’s, “one of the first changes that’s notable is voice,” usually appearing before a formal diagnosis, says Anais Rameau, MD, an assistant professor of laryngology at Weill Cornell Medicine, New York, and another member of the NIH project. Parkinson’s may soften the voice or make it sound monotone, while Alzheimer’s disease may change the content of speech, leading to an uptick in “umms” and a preference for pronouns over nouns.
With Parkinson’s, vocal changes can occur decades before movement is affected. If doctors could detect the disease at this stage, before tremor emerged, they might be able to flag patients for early intervention, says Max Little, PhD, project director for the Parkinson’s Voice Initiative. “That is the ‘holy grail’ for finding an eventual cure.”
Again, the smartphone shows potential. In a 2022 Australian study, an AI-powered app was able to identify people with Parkinson’s based on brief voice recordings, although the sample size was small. On a larger scale, the Parkinson’s Voice Initiative collected some 17,000 samples from people across the world. “The aim was to remotely detect those with the condition using a telephone call,” says Dr. Little. It did so with about 65% accuracy. “While this is not accurate enough for clinical use, it shows the potential of the idea,” he says.
Dr. Rudzicz worked on the team behind Winterlight, an iPad app that analyzes 550 features of speech to detect dementia and Alzheimer’s (as well as mental illness). “We deployed it in long-term care facilities,” he says, identifying patients who need further review of their mental skills. Stroke is another area of interest, because slurred speech is a highly subjective measure, says Dr. Anderson. AI technology could provide a more objective evaluation.
Mood and psychiatric disorders (depression, schizophrenia, bipolar disorders)
No established biomarkers exist for diagnosing depression. Yet if you’re feeling down, there’s a good chance your friends can tell – even over the phone.
“We carry a lot of our mood in our voice,” says Dr. Powell. Bipolar disorder can also alter voice, making it louder and faster during manic periods, then slower and quieter during depressive bouts. The catatonic stage of schizophrenia often comes with “a very monotone, robotic voice,” says Dr. Anderson. “These are all something an algorithm can measure.”
Apps are already being used – often in research settings – to monitor voices during phone calls, analyzing rate, rhythm, volume, and pitch, to predict mood changes. For example, the PRIORI project at the University of Michigan is working on a smartphone app to identify mood changes in people with bipolar disorder, especially shifts that could increase suicide risk.
The content of speech may also offer clues. In a University of California, Los Angeles, study published in the journal PLoS One, people with mental illnesses answered computer-programmed questions (like “How have you been over the past few days?”) over the phone. An app analyzed their word choices, paying attention to how they changed over time. The researchers found that AI analysis of mood aligned well with doctors’ assessments and that some people in the study actually felt more comfortable talking to a computer.
Respiratory disorders (pneumonia, COPD)
Beyond talking, respiratory sounds like gasping or coughing may point to specific conditions. “Emphysema cough is different, COPD cough is different,” says Dr. Bensoussan. Researchers are trying to find out if COVID-19 has a distinct cough.
Breathing sounds can also serve as signposts. “There are different sounds when we can’t breathe,” says Dr. Bensoussan. One is called stridor, a high-pitched wheezing often resulting from a blocked airway. “I see tons of people [with stridor] misdiagnosed for years – they’ve been told they have asthma, but they don’t,” says Dr. Bensoussan. AI analysis of these sounds could help doctors more quickly identify respiratory disorders.
Pediatric voice and speech disorders (speech and language delays, autism)
Babies who later have autism cry differently as early as 6 months of age, which means an app like ChatterBaby could help flag children for early intervention, says Dr. Anderson. Autism is linked to several other diagnoses, such as epilepsy and sleep disorders. So analyzing an infant’s cry could prompt pediatricians to screen for a range of conditions.
ChatterBaby has been “incredibly accurate” in identifying when babies are in pain, says Dr. Anderson, because pain increases muscle tension, resulting in a louder, more energetic cry. The next goal: “We’re collecting voices from babies around the world,” she says, and then tracking those children for 7 years, looking to see if early vocal signs could predict developmental disorders. Vocal samples from young children could serve a similar purpose.
And that’s only the beginning
Eventually, AI technology may pick up disease-related voice changes that we can’t even hear. In a new Mayo Clinic study, certain vocal features detectable by AI – but not by the human ear – were linked to a three-fold increase in the likelihood of having plaque buildup in the arteries.
“Voice is a huge spectrum of vibrations,” explains study author Amir Lerman, MD. “We hear a very narrow range.”
The researchers aren’t sure why heart disease alters voice, but the autonomic nervous system may play a role, because it regulates the voice box as well as blood pressure and heart rate. Dr. Lerman says other conditions, like diseases of the nerves and gut, may similarly alter the voice. Beyond patient screening, this discovery could help doctors adjust medication doses remotely, in line with these inaudible vocal signals.
“Hopefully, in the next few years, this is going to come to practice,” says Dr. Lerman.
Still, in the face of that hope, privacy concerns remain. Voice is an identifier that’s protected by the federal Health Insurance Portability and Accountability Act, which requires privacy of personal health information. That is a major reason why no large voice databases exist yet, says Dr. Bensoussan. (This makes collecting samples from children especially challenging.) Perhaps more concerning is the potential for diagnosing disease based on voice alone. “You could use that tool on anyone, including officials like the president,” says Dr. Rameau.
But the primary hurdle is the ethical sourcing of data to ensure a diversity of vocal samples. For the Voice as a Biomarker project, the researchers will establish voice quotas for different races and ethnicities, ensuring algorithms can accurately analyze a range of accents. Data from people with speech impediments will also be gathered.
Despite these challenges, researchers are optimistic. “Vocal analysis is going to be a great equalizer and improve health outcomes,” predicts Dr. Anderson. “I’m really happy that we are beginning to understand the strength of the voice.”
A version of this article first appeared on WebMD.com.
: First during puberty, as the vocal cords thicken and the voice box migrates down the throat. Then a second time as aging causes structural changes that may weaken the voice.
But for some of us, there’s another voice shift, when a disease begins or when our mental health declines.
This is why more doctors are looking into voice as a biomarker – something that tells you that a disease is present.
Vital signs like blood pressure or heart rate “can give a general idea of how sick we are. But they’re not specific to certain diseases,” says Yael Bensoussan, MD, director of the University of South Florida, Tampa’s Health Voice Center and the coprincipal investigator for the National Institutes of Health’s Voice as a Biomarker of Health project.
“We’re learning that there are patterns” in voice changes that can indicate a range of conditions, including diseases of the nervous system and mental illnesses, she says.
Speaking is complicated, involving everything from the lungs and voice box to the mouth and brain. “A breakdown in any of those parts can affect the voice,” says Maria Powell, PhD, an assistant professor of otolaryngology (the study of diseases of the ear and throat) at Vanderbilt University, Nashville, Tenn., who is working on the NIH project.
You or those around you may not notice the changes. But researchers say voice analysis as a standard part of patient care – akin to blood pressure checks or cholesterol tests – could help identify those who need medical attention earlier.
Often, all it takes is a smartphone – “something that’s cheap, off-the-shelf, and that everyone can use,” says Ariana Anderson, PhD, director of the University of California, Los Angeles, Laboratory of Computational Neuropsychology.
“You can provide voice data in your pajamas, on your couch,” says Frank Rudzicz, PhD, a computer scientist for the NIH project. “It doesn’t require very complicated or expensive equipment, and it doesn’t require a lot of expertise to obtain.” Plus, multiple samples can be collected over time, giving a more accurate picture of health than a single snapshot from, say, a cognitive test.
Over the next 4 years, the Voice as a Biomarker team will receive nearly $18 million to gather a massive amount of voice data. The goal is 20,000-30,000 samples, along with health data about each person being studied. The result will be a sprawling database scientists can use to develop algorithms linking health conditions to the way we speak.
For the first 2 years, new data will be collected exclusively via universities and high-volume clinics to control quality and accuracy. Eventually, people will be invited to submit their own voice recordings, creating a crowdsourced dataset. “Google, Alexa, Amazon – they have access to tons of voice data,” says Dr. Bensoussan. “But it’s not usable in a clinical way, because they don’t have the health information.”
Dr. Bensoussan and her colleagues hope to fill that void with advance voice screening apps, which could prove especially valuable in remote communities that lack access to specialists or as a tool for telemedicine. Down the line, wearable devices with voice analysis could alert people with chronic conditions when they need to see a doctor.
“The watch says, ‘I’ve analyzed your breathing and coughing, and today, you’re really not doing well. You should go to the hospital,’ ” says Dr. Bensoussan, envisioning a wearable for patients with COPD. “It could tell people early that things are declining.”
Artificial intelligence may be better than a brain at pinpointing the right disease. For example, slurred speech could indicate Parkinson’s, a stroke, or ALS, among other things.
“We can hold approximately seven pieces of information in our head at one time,” says Dr. Rudzicz. “It’s really hard for us to get a holistic picture using dozens or hundreds of variables at once.” But a computer can consider a whole range of vocal markers at the same time, piecing them together for a more accurate assessment.
“The goal is not to outperform a ... clinician,” says Dr. Bensoussan. Yet the potential is unmistakably there: In a recent study of patients with cancer of the larynx, an automated voice analysis tool more accurately flagged the disease than laryngologists did.
“Algorithms have a larger training base,” says Dr. Anderson, who developed an app called ChatterBaby that analyzes infant cries. “We have a million samples at our disposal to train our algorithms. I don’t know if I’ve heard a million different babies crying in my life.”
So which health conditions show the most promise for voice analysis? The Voice as a Biomarker project will focus on five categories.
Voice disorders (cancers of the larynx, vocal fold paralysis, benign lesions on the larynx)
Obviously, vocal changes are a hallmark of these conditions, which cause things like breathiness or “roughness,” a type of vocal irregularity. Hoarseness that lasts at least 2 weeks is often one of the earliest signs of laryngeal cancer. Yet it can take months – one study found 16 weeks was the average – for patients to see a doctor after noticing the changes. Even then, laryngologists still misdiagnosed some cases of cancer when relying on vocal cues alone.
Now imagine a different scenario: The patient speaks into a smartphone app. An algorithm compares the vocal sample with the voices of laryngeal cancer patients. The app spits out the estimated odds of laryngeal cancer, helping providers decide whether to offer the patient specialist care.
Or consider spasmodic dysphonia, a neurological voice disorder that triggers spasms in the muscles of the voice box, causing a strained or breathy voice. Doctors who lack experience with vocal disorders may miss the condition. This is why diagnosis takes an average of nearly 4.5 years, according to a study in the Journal of Voice, and may include everything from allergy testing to psychiatric evaluation, says Dr. Powell. Artificial intelligence technology trained to recognize the disorder could help eliminate such unnecessary testing.
Neurological and neurodegenerative disorders (Alzheimer’s, Parkinson’s, stroke, ALS)
For Alzheimer’s and Parkinson’s, “one of the first changes that’s notable is voice,” usually appearing before a formal diagnosis, says Anais Rameau, MD, an assistant professor of laryngology at Weill Cornell Medicine, New York, and another member of the NIH project. Parkinson’s may soften the voice or make it sound monotone, while Alzheimer’s disease may change the content of speech, leading to an uptick in “umms” and a preference for pronouns over nouns.
With Parkinson’s, vocal changes can occur decades before movement is affected. If doctors could detect the disease at this stage, before tremor emerged, they might be able to flag patients for early intervention, says Max Little, PhD, project director for the Parkinson’s Voice Initiative. “That is the ‘holy grail’ for finding an eventual cure.”
Again, the smartphone shows potential. In a 2022 Australian study, an AI-powered app was able to identify people with Parkinson’s based on brief voice recordings, although the sample size was small. On a larger scale, the Parkinson’s Voice Initiative collected some 17,000 samples from people across the world. “The aim was to remotely detect those with the condition using a telephone call,” says Dr. Little. It did so with about 65% accuracy. “While this is not accurate enough for clinical use, it shows the potential of the idea,” he says.
Dr. Rudzicz worked on the team behind Winterlight, an iPad app that analyzes 550 features of speech to detect dementia and Alzheimer’s (as well as mental illness). “We deployed it in long-term care facilities,” he says, identifying patients who need further review of their mental skills. Stroke is another area of interest, because slurred speech is a highly subjective measure, says Dr. Anderson. AI technology could provide a more objective evaluation.
Mood and psychiatric disorders (depression, schizophrenia, bipolar disorders)
No established biomarkers exist for diagnosing depression. Yet if you’re feeling down, there’s a good chance your friends can tell – even over the phone.
“We carry a lot of our mood in our voice,” says Dr. Powell. Bipolar disorder can also alter voice, making it louder and faster during manic periods, then slower and quieter during depressive bouts. The catatonic stage of schizophrenia often comes with “a very monotone, robotic voice,” says Dr. Anderson. “These are all something an algorithm can measure.”
Apps are already being used – often in research settings – to monitor voices during phone calls, analyzing rate, rhythm, volume, and pitch, to predict mood changes. For example, the PRIORI project at the University of Michigan is working on a smartphone app to identify mood changes in people with bipolar disorder, especially shifts that could increase suicide risk.
The content of speech may also offer clues. In a University of California, Los Angeles, study published in the journal PLoS One, people with mental illnesses answered computer-programmed questions (like “How have you been over the past few days?”) over the phone. An app analyzed their word choices, paying attention to how they changed over time. The researchers found that AI analysis of mood aligned well with doctors’ assessments and that some people in the study actually felt more comfortable talking to a computer.
Respiratory disorders (pneumonia, COPD)
Beyond talking, respiratory sounds like gasping or coughing may point to specific conditions. “Emphysema cough is different, COPD cough is different,” says Dr. Bensoussan. Researchers are trying to find out if COVID-19 has a distinct cough.
Breathing sounds can also serve as signposts. “There are different sounds when we can’t breathe,” says Dr. Bensoussan. One is called stridor, a high-pitched wheezing often resulting from a blocked airway. “I see tons of people [with stridor] misdiagnosed for years – they’ve been told they have asthma, but they don’t,” says Dr. Bensoussan. AI analysis of these sounds could help doctors more quickly identify respiratory disorders.
Pediatric voice and speech disorders (speech and language delays, autism)
Babies who later have autism cry differently as early as 6 months of age, which means an app like ChatterBaby could help flag children for early intervention, says Dr. Anderson. Autism is linked to several other diagnoses, such as epilepsy and sleep disorders. So analyzing an infant’s cry could prompt pediatricians to screen for a range of conditions.
ChatterBaby has been “incredibly accurate” in identifying when babies are in pain, says Dr. Anderson, because pain increases muscle tension, resulting in a louder, more energetic cry. The next goal: “We’re collecting voices from babies around the world,” she says, and then tracking those children for 7 years, looking to see if early vocal signs could predict developmental disorders. Vocal samples from young children could serve a similar purpose.
And that’s only the beginning
Eventually, AI technology may pick up disease-related voice changes that we can’t even hear. In a new Mayo Clinic study, certain vocal features detectable by AI – but not by the human ear – were linked to a three-fold increase in the likelihood of having plaque buildup in the arteries.
“Voice is a huge spectrum of vibrations,” explains study author Amir Lerman, MD. “We hear a very narrow range.”
The researchers aren’t sure why heart disease alters voice, but the autonomic nervous system may play a role, because it regulates the voice box as well as blood pressure and heart rate. Dr. Lerman says other conditions, like diseases of the nerves and gut, may similarly alter the voice. Beyond patient screening, this discovery could help doctors adjust medication doses remotely, in line with these inaudible vocal signals.
“Hopefully, in the next few years, this is going to come to practice,” says Dr. Lerman.
Still, in the face of that hope, privacy concerns remain. Voice is an identifier that’s protected by the federal Health Insurance Portability and Accountability Act, which requires privacy of personal health information. That is a major reason why no large voice databases exist yet, says Dr. Bensoussan. (This makes collecting samples from children especially challenging.) Perhaps more concerning is the potential for diagnosing disease based on voice alone. “You could use that tool on anyone, including officials like the president,” says Dr. Rameau.
But the primary hurdle is the ethical sourcing of data to ensure a diversity of vocal samples. For the Voice as a Biomarker project, the researchers will establish voice quotas for different races and ethnicities, ensuring algorithms can accurately analyze a range of accents. Data from people with speech impediments will also be gathered.
Despite these challenges, researchers are optimistic. “Vocal analysis is going to be a great equalizer and improve health outcomes,” predicts Dr. Anderson. “I’m really happy that we are beginning to understand the strength of the voice.”
A version of this article first appeared on WebMD.com.
: First during puberty, as the vocal cords thicken and the voice box migrates down the throat. Then a second time as aging causes structural changes that may weaken the voice.
But for some of us, there’s another voice shift, when a disease begins or when our mental health declines.
This is why more doctors are looking into voice as a biomarker – something that tells you that a disease is present.
Vital signs like blood pressure or heart rate “can give a general idea of how sick we are. But they’re not specific to certain diseases,” says Yael Bensoussan, MD, director of the University of South Florida, Tampa’s Health Voice Center and the coprincipal investigator for the National Institutes of Health’s Voice as a Biomarker of Health project.
“We’re learning that there are patterns” in voice changes that can indicate a range of conditions, including diseases of the nervous system and mental illnesses, she says.
Speaking is complicated, involving everything from the lungs and voice box to the mouth and brain. “A breakdown in any of those parts can affect the voice,” says Maria Powell, PhD, an assistant professor of otolaryngology (the study of diseases of the ear and throat) at Vanderbilt University, Nashville, Tenn., who is working on the NIH project.
You or those around you may not notice the changes. But researchers say voice analysis as a standard part of patient care – akin to blood pressure checks or cholesterol tests – could help identify those who need medical attention earlier.
Often, all it takes is a smartphone – “something that’s cheap, off-the-shelf, and that everyone can use,” says Ariana Anderson, PhD, director of the University of California, Los Angeles, Laboratory of Computational Neuropsychology.
“You can provide voice data in your pajamas, on your couch,” says Frank Rudzicz, PhD, a computer scientist for the NIH project. “It doesn’t require very complicated or expensive equipment, and it doesn’t require a lot of expertise to obtain.” Plus, multiple samples can be collected over time, giving a more accurate picture of health than a single snapshot from, say, a cognitive test.
Over the next 4 years, the Voice as a Biomarker team will receive nearly $18 million to gather a massive amount of voice data. The goal is 20,000-30,000 samples, along with health data about each person being studied. The result will be a sprawling database scientists can use to develop algorithms linking health conditions to the way we speak.
For the first 2 years, new data will be collected exclusively via universities and high-volume clinics to control quality and accuracy. Eventually, people will be invited to submit their own voice recordings, creating a crowdsourced dataset. “Google, Alexa, Amazon – they have access to tons of voice data,” says Dr. Bensoussan. “But it’s not usable in a clinical way, because they don’t have the health information.”
Dr. Bensoussan and her colleagues hope to fill that void with advance voice screening apps, which could prove especially valuable in remote communities that lack access to specialists or as a tool for telemedicine. Down the line, wearable devices with voice analysis could alert people with chronic conditions when they need to see a doctor.
“The watch says, ‘I’ve analyzed your breathing and coughing, and today, you’re really not doing well. You should go to the hospital,’ ” says Dr. Bensoussan, envisioning a wearable for patients with COPD. “It could tell people early that things are declining.”
Artificial intelligence may be better than a brain at pinpointing the right disease. For example, slurred speech could indicate Parkinson’s, a stroke, or ALS, among other things.
“We can hold approximately seven pieces of information in our head at one time,” says Dr. Rudzicz. “It’s really hard for us to get a holistic picture using dozens or hundreds of variables at once.” But a computer can consider a whole range of vocal markers at the same time, piecing them together for a more accurate assessment.
“The goal is not to outperform a ... clinician,” says Dr. Bensoussan. Yet the potential is unmistakably there: In a recent study of patients with cancer of the larynx, an automated voice analysis tool more accurately flagged the disease than laryngologists did.
“Algorithms have a larger training base,” says Dr. Anderson, who developed an app called ChatterBaby that analyzes infant cries. “We have a million samples at our disposal to train our algorithms. I don’t know if I’ve heard a million different babies crying in my life.”
So which health conditions show the most promise for voice analysis? The Voice as a Biomarker project will focus on five categories.
Voice disorders (cancers of the larynx, vocal fold paralysis, benign lesions on the larynx)
Obviously, vocal changes are a hallmark of these conditions, which cause things like breathiness or “roughness,” a type of vocal irregularity. Hoarseness that lasts at least 2 weeks is often one of the earliest signs of laryngeal cancer. Yet it can take months – one study found 16 weeks was the average – for patients to see a doctor after noticing the changes. Even then, laryngologists still misdiagnosed some cases of cancer when relying on vocal cues alone.
Now imagine a different scenario: The patient speaks into a smartphone app. An algorithm compares the vocal sample with the voices of laryngeal cancer patients. The app spits out the estimated odds of laryngeal cancer, helping providers decide whether to offer the patient specialist care.
Or consider spasmodic dysphonia, a neurological voice disorder that triggers spasms in the muscles of the voice box, causing a strained or breathy voice. Doctors who lack experience with vocal disorders may miss the condition. This is why diagnosis takes an average of nearly 4.5 years, according to a study in the Journal of Voice, and may include everything from allergy testing to psychiatric evaluation, says Dr. Powell. Artificial intelligence technology trained to recognize the disorder could help eliminate such unnecessary testing.
Neurological and neurodegenerative disorders (Alzheimer’s, Parkinson’s, stroke, ALS)
For Alzheimer’s and Parkinson’s, “one of the first changes that’s notable is voice,” usually appearing before a formal diagnosis, says Anais Rameau, MD, an assistant professor of laryngology at Weill Cornell Medicine, New York, and another member of the NIH project. Parkinson’s may soften the voice or make it sound monotone, while Alzheimer’s disease may change the content of speech, leading to an uptick in “umms” and a preference for pronouns over nouns.
With Parkinson’s, vocal changes can occur decades before movement is affected. If doctors could detect the disease at this stage, before tremor emerged, they might be able to flag patients for early intervention, says Max Little, PhD, project director for the Parkinson’s Voice Initiative. “That is the ‘holy grail’ for finding an eventual cure.”
Again, the smartphone shows potential. In a 2022 Australian study, an AI-powered app was able to identify people with Parkinson’s based on brief voice recordings, although the sample size was small. On a larger scale, the Parkinson’s Voice Initiative collected some 17,000 samples from people across the world. “The aim was to remotely detect those with the condition using a telephone call,” says Dr. Little. It did so with about 65% accuracy. “While this is not accurate enough for clinical use, it shows the potential of the idea,” he says.
Dr. Rudzicz worked on the team behind Winterlight, an iPad app that analyzes 550 features of speech to detect dementia and Alzheimer’s (as well as mental illness). “We deployed it in long-term care facilities,” he says, identifying patients who need further review of their mental skills. Stroke is another area of interest, because slurred speech is a highly subjective measure, says Dr. Anderson. AI technology could provide a more objective evaluation.
Mood and psychiatric disorders (depression, schizophrenia, bipolar disorders)
No established biomarkers exist for diagnosing depression. Yet if you’re feeling down, there’s a good chance your friends can tell – even over the phone.
“We carry a lot of our mood in our voice,” says Dr. Powell. Bipolar disorder can also alter voice, making it louder and faster during manic periods, then slower and quieter during depressive bouts. The catatonic stage of schizophrenia often comes with “a very monotone, robotic voice,” says Dr. Anderson. “These are all something an algorithm can measure.”
Apps are already being used – often in research settings – to monitor voices during phone calls, analyzing rate, rhythm, volume, and pitch, to predict mood changes. For example, the PRIORI project at the University of Michigan is working on a smartphone app to identify mood changes in people with bipolar disorder, especially shifts that could increase suicide risk.
The content of speech may also offer clues. In a University of California, Los Angeles, study published in the journal PLoS One, people with mental illnesses answered computer-programmed questions (like “How have you been over the past few days?”) over the phone. An app analyzed their word choices, paying attention to how they changed over time. The researchers found that AI analysis of mood aligned well with doctors’ assessments and that some people in the study actually felt more comfortable talking to a computer.
Respiratory disorders (pneumonia, COPD)
Beyond talking, respiratory sounds like gasping or coughing may point to specific conditions. “Emphysema cough is different, COPD cough is different,” says Dr. Bensoussan. Researchers are trying to find out if COVID-19 has a distinct cough.
Breathing sounds can also serve as signposts. “There are different sounds when we can’t breathe,” says Dr. Bensoussan. One is called stridor, a high-pitched wheezing often resulting from a blocked airway. “I see tons of people [with stridor] misdiagnosed for years – they’ve been told they have asthma, but they don’t,” says Dr. Bensoussan. AI analysis of these sounds could help doctors more quickly identify respiratory disorders.
Pediatric voice and speech disorders (speech and language delays, autism)
Babies who later have autism cry differently as early as 6 months of age, which means an app like ChatterBaby could help flag children for early intervention, says Dr. Anderson. Autism is linked to several other diagnoses, such as epilepsy and sleep disorders. So analyzing an infant’s cry could prompt pediatricians to screen for a range of conditions.
ChatterBaby has been “incredibly accurate” in identifying when babies are in pain, says Dr. Anderson, because pain increases muscle tension, resulting in a louder, more energetic cry. The next goal: “We’re collecting voices from babies around the world,” she says, and then tracking those children for 7 years, looking to see if early vocal signs could predict developmental disorders. Vocal samples from young children could serve a similar purpose.
And that’s only the beginning
Eventually, AI technology may pick up disease-related voice changes that we can’t even hear. In a new Mayo Clinic study, certain vocal features detectable by AI – but not by the human ear – were linked to a three-fold increase in the likelihood of having plaque buildup in the arteries.
“Voice is a huge spectrum of vibrations,” explains study author Amir Lerman, MD. “We hear a very narrow range.”
The researchers aren’t sure why heart disease alters voice, but the autonomic nervous system may play a role, because it regulates the voice box as well as blood pressure and heart rate. Dr. Lerman says other conditions, like diseases of the nerves and gut, may similarly alter the voice. Beyond patient screening, this discovery could help doctors adjust medication doses remotely, in line with these inaudible vocal signals.
“Hopefully, in the next few years, this is going to come to practice,” says Dr. Lerman.
Still, in the face of that hope, privacy concerns remain. Voice is an identifier that’s protected by the federal Health Insurance Portability and Accountability Act, which requires privacy of personal health information. That is a major reason why no large voice databases exist yet, says Dr. Bensoussan. (This makes collecting samples from children especially challenging.) Perhaps more concerning is the potential for diagnosing disease based on voice alone. “You could use that tool on anyone, including officials like the president,” says Dr. Rameau.
But the primary hurdle is the ethical sourcing of data to ensure a diversity of vocal samples. For the Voice as a Biomarker project, the researchers will establish voice quotas for different races and ethnicities, ensuring algorithms can accurately analyze a range of accents. Data from people with speech impediments will also be gathered.
Despite these challenges, researchers are optimistic. “Vocal analysis is going to be a great equalizer and improve health outcomes,” predicts Dr. Anderson. “I’m really happy that we are beginning to understand the strength of the voice.”
A version of this article first appeared on WebMD.com.