ARTICLE IN BRIEF
Investigators provide new evidence that human narcolepsy is an autoimmune disease that results when people with certain gene variants are exposed to a virus or a vaccine that induces the body to launch an attack on hypocretin-producing cells.
For decades, circumstantial evidence has suggested that some people have a genetic susceptibility to narcolepsy, which produces excessive daytime sleepiness, cataplexy, and sleep abnormalities. However, unlike narcoleptic dogs, which lack a receptor for hypocretin (a hormone that promotes wakefulness), humans predisposed to narcolepsy seem to require an environmental trigger that somehow damages or destroys the 70,000 or so brain cells that produce hypocretin.
Now researchers at the Stanford University School of Medicine have found compelling evidence that human narcolepsy is an autoimmune disease that results when people with certain gene variants are exposed to a virus or a vaccine that induces the body to launch an attack on hypocretin-producing cells.
Also, signs of this autoimmune attack appear very early in the course of the disease, which may enable a prompt diagnosis with a simple blood test that would allow treatment to begin sooner, according to veteran sleep researcher Emmanuel Mignot, MD, PhD, director of the Stanford Center for Sleep Sciences and Medicine, the lead author of the study published in the Dec. 18 issue of Science Translational Medicine.
“With a blood test we might be able to diagnose milder cases of narcolepsy and do immune suppression before it's too late,” said Dr. Mignot, who is also the Craig Reynolds professor of sleep medicine at the Stanford School of Medicine.
The research also provides the first evidence in vivo that autoimmune disease results from molecular mimicry, in which a virus or a bacterium displays molecules so similar to those found in the body's own tissues that the immune system attacks both.
“This is a huge finding both for narcolepsy and for immunology in general,” Dr. Mignot said. “People have suspected for a long time that molecular mimicry was involved in autoimmune disease, but now we have demonstrated that the immune system sometimes gets confused and fails to distinguish self from virus. I think narcolepsy will become a model for other autoimmune disorders.”
UNRAVELLING THE NARCOLEPSY MECHANISM
Dr. Mignot has been trying to unravel the mechanism behind narcolepsy for nearly 25 years. In a 2000 paper in The Lancet, Dr. Mignot and colleagues reported that hypocretin could not be detected in the cerebrospinal fluid of people with narcolepsy, which implied that the hormone was essential for maintaining normal wakefulness. Later that year, in Nature Medicine, they reported that unlike narcoleptic dogs, which lack the gene needed to produce receptors for hypocretin, virtually all human cases of narcolepsy involve the destruction of hypocretin-producing cells in the brain.
In 2009, in Nature Genetics, they reported that a genome-wide association study of more than 4,000 people identified a genetic variant of the human leukocyte antigen (HLA) in virtually all cases of narcolepsy, while the variant appears in only 18-25 percent of the population. The researchers also found that people with narcolepsy are far more likely to have a variant of the T-cell receptor alpha (TCRA) gene, which encodes a receptor protein on the surface of T cells that enables the T cells to recognize and attack bacteria and viruses. Together these findings provided powerful evidence that narcolepsy is an autoimmune disease involving T cells.
Subsequent research by Dr. Mignot and others found increases in the incidence of narcolepsy among people exposed to the H1N1 flu virus and a flu vaccine used in conjunction with an adjuvant, providing strong evidence of environmental factors that trigger the disease in genetically susceptible people.
In their Science Translational Medicine paper, Dr. Mignot and colleagues showed that people with narcolepsy possess a genetic variant of HLA — DQ0602 — which stimulates CD4+ T cells to attack two 13-amino acid peptides derived from hypocretin. These epitopes evoked an immune response to hypocretin in 23 patients with narcolepsy, but not in healthy DQ0602-positive control subjects. Also, in pairs of identical twins, the one with narcolepsy carried T cells activated by the hypocretin peptides, while the healthy twin did not.
This provides compelling evidence that narcolepsy requires genetic susceptibility plus an environmental trigger, such as exposure to the H1N1 flu virus, or Streptococcus pyogenes, said Dr. Mignot.
“We suspected narcolepsy was an autoimmune problem from all the genetic data that pointed to that conclusion, and from the fact that all these hypocretin cells were killed,” he said. “But we couldn't find the epitope — the cross-reactive molecule, or autoantigen. In this study we have found the exact piece of hypocretin that triggers the autoimmune response, and that is very important.”
The appearance of the H1N1 virus preceded an outbreak of narcolepsy in several countries, including England, Sweden, and Ireland, suggesting that the immune system of those who developed the disease failed to distinguish hypocretin cells from the hemagglutinin molecule found in the virus. GlaxoSmithKline, the manufacturer of the Pandemrix flu vaccine used in the 2009-2010 H1N1 outbreak, has identified at least 900 narcolepsy cases in people who received the vaccine, which contained portions of the virus. The Pandemrix vaccine, used with an adjuvant that induced a stronger immune response, was never used in the United States, and has been withdrawn from the market.
Dr. Mignot suspects that narcolepsy exists on a spectrum, with cases at the milder end now potentially identifiable with a blood test.
“This is speculation on my part, but I think onset in some cases is probably milder and harder to detect, and progressive,” he said. “People learn to live with it.”
A blood test could allow T cells from a subject to be exposed in vitro to cells containing the HLA variant associated with narcolepsy. “Then you add the piece of hypocretin that binds to [the haplotype] DQB1∗06:02,” said Dr. Mignot. “T cells from normal people won't proliferate, but if you put T cells from patients (with narcolepsy), they will get activated.”
The Science Translational Medicine paper is very impressive in the way it finds differences between narcoleptics and healthy people, Jerome Siegel, PhD, professor of psychiatry and biobehavioral sciences and chief of neurobiology research at the University of California, Los Angeles Center for Sleep Research, told Neurology Today.
“That hasn't been seen before,” said Dr. Siegel, also a professor in residence at the Semel Institute for Neuroscience and Human Behavior. “They found that narcoleptics differ from normals in ways that can be detected in the blood. Particularly impressive is their finding that identical twins discordant for narcolepsy show a T-cell difference.”
Using this information to develop a blood test for narcolepsy would be extremely helpful, Dr. Siegel added, especially for those with subtle disease that does not include the most conspicuous symptom, cataplexy. (Only about two-thirds of those with narcolepsy experience the sudden, transient muscle weakness of cataplexy.)
“People with narcolepsy without cataplexy have lower hypocretin than normal, but not as low as those who have narcolepsy with cataplexy,” Dr. Siegel said. “They experience sleepiness, but other disorders such as sleep apnea and hypersomnia produce that. Narcolepsy often is not diagnosed for 10 years or more after symptom onset.”
Michael J. Thorpy, MD, director of the Sleep-Wake Disorders Center at the Montefiore Medical Center in the Bronx, also expressed hope that Dr. Mignot's research will lead to earlier diagnosis of narcolepsy. “Because of the consequences of narcolepsy, the earlier we can get it diagnosed the better,” he said.
At the moment, treatments exist only for the symptoms of narcolepsy. Excessive sleepiness can be treated with modafinil (Provigil) and armodafinil (Nuvigil), and antidepressants are sometimes prescribed off-label for cataplexy. Sodium oxybate (Xyrem, Alcover) is approved by the US Food and Drug Administration and can help both sleepiness and cataplexy in some patients.
Since narcolepsy results from the death of cells that do not regenerate, prevention offers the most attractive strategy. However, that would require early diagnosis, before all hypocretin-producing cells have died. Immunosuppressive and anti-inflammatory drugs have been shown to delay onset of canine genetic narcolepsy and reduce the severity of symptoms, and some narcolepsy patients have shown improvement after being treated early in their disease with intravenous immunoglobulins.
“But that's still experimental, so it's not being done clinically,” Dr. Thorpy said. “People have used steroids and plasmapheresis, but the effectiveness may depend on when you give these therapies. There's a possibility that if we could get early diagnosis with a blood test and then administer some immune therapy, we may see further benefit. This paper suggests that a test could be developed that could lead to a confirmed diagnosis of narcolepsy. That would be great.”
•. De la Herrán-Arita AK, Kornum BR, Mahlios J, et al. CD4+ T cell autoimmunity to hypocretin/orexin and cross-reactivity to a 2009 H1N1 influenza A epitope in narcolepsy. Science Transl Med. 2013; 5:(216ra176).
•. Dauvilliers Y, Montplaisir J, Cochen V, et al. Post-H1N1 narcolepsy-cataplexy. Sleep. 2010; 33: 1428–1430.
•. Partinen M, Saarenpää-Heikkilä O, Ilveskoski I, et al. Increased incidence and clinical picture of childhood narcolepsy following the 2009 H1N1 pandemic vaccination campaign in Finland. PLOS One. 2012; 7:(3): e33723.
•. Nishino S, Ripley B, Overeem S, et al. Hypocretin (orexin) deficiency in human narcolepsy. Lancet. 2000; 355:(9197):39–40.