In a paper in Science Translational Medicine (STM) this week, researchers offer a possible explanation. They show that the vaccine, called Pandemrix, triggers antibodies that can also bind to a receptor in brain cells that help regulate sleepiness. The work strongly suggests that Pandemrix, which was given to more than 30 million Europeans, triggered an autoimmune re action that led to narcolepsy in some people who are genetically at risk.
“They put together quite a convincing picture and provide a plausible explanation for what has happened,” says Pasi Penttinen, who heads the influenza program at the European Centre for Disease Prevention and Control in Stockholm. “It’s really the kind of work we’ve been waiting for for 5 years.” But the results still need to be confirmed in a larger study, the authors and other narcolepsy researchers say. A 2013 paper in STM by another group, documenting a different type of vaccine-triggered autoimmune re action, was retracted after the results proved irreproducible (Science, 1 August 2014, p. 498).
Narcolepsy, a mysterious malady that affects roughly one in 3000 people in Europe, most often appears in childhood or adolescence. Patients lose certain brain cells in the hypothalamus, leading to a deficiency of hypocretin, a molecule that helps regulate the sleep-wake cycle. Researchers suspect an autoimmune reaction is to blame because many people who develop narcolepsy—and just about everyone with the vaccine-associated form—have a specific variant in a gene in the HLA family, which helps the body distinguish its own proteins from those made by microbial invaders.
When they heard about the rise in narcolepsy in 2010, neuroscientist Lawrence Steinman of Stanford University in Palo Alto, California, and rheumatologist Sohail Ahmed, who at the time was global head of clinical sciences at Novartis’s vaccines and diagnostics division in Siena, Italy, began scouring databases for proteins expressed in the brain that might resemble those in the vaccine.
Their search turned up a suspect: a piece of a receptor for hypocretin resembles part of the H1N1 influenza nucleoprotein—which binds to the virus genome and plays a key role in its replication. “That was really an ‘Aha’ moment,” says Ahmed, who is now at GSK, which bought part of Novartis’s vaccine division. The flu vaccine is designed to trigger antibodies to influenza’s surface proteins, but if it elicits antibodies to the nucleoprotein as well, those might well latch on to the hypocretin receptor, and eventually lead to death of the cells, the researchers thought.
Some of the same researchers already had circumstantial evidence that the nucleoprotein might be an important player. In December, Outi Vaarala of the University of Helsinki and her colleagues had reported that Pandemrix contained a much higher level of nucleoprotein than a GSK vaccine called Arepanrix, which was associated with a much smaller risk for narcolepsy. The group also found that children with narcolepsy had an altered immune response to the nucleoprotein in Pandemrix.
In the new work, the researchers added serum from Finnish narcolepsy patients who had received Pandemrix to cells that were engineered to display human hypo cretin receptor 2 on their surface. Antibodies from the patients bound to these cells in 17 of 20 samples. Serum from Italians who had been vaccinated with a different pandemic vaccine from Novartis, called Focetria, did not have such antibodies. The researchers also showed that Focetria, which has not been linked to narcolepsy, had a much lower concentration of nucleoprotein than Pandemrix did.
If a vaccine can trigger narcolepsy, can the flu virus itself—either H1N1 or other influenza strains? Cases of narcolepsy increased after the 2009 pandemic in China, where Pandemrix wasn’t used, and researchers discovered that serum from some H1N1 influenza patients does bind to the hypo cretin receptor. But the jury is still out, says virologist Ilkka Julkunen of the University of Turku in Finland, a co-author on the paper.
In the meantime, the new insights might eventually lead to ways to help narcolepsy patients. Researchers could explore whether somehow unblocking the receptor might allow the hypocretin system to recover, says Noni MacDonald, a pediatric vaccine researcher at Dalhousie University in Halifax, Canada. The study also suggests ways to make influenza vaccines even safer, for instance by keeping the nucleoprotein levels low or removing the specific region of the protein that mimics the receptor. “We should do everything to understand the disease mechanism,” Julkunen says, “so that this kind of situation would never be repeated."