A recent report in Science Traslational Medicine (1) has deserved a lot of attention by mainstream media. Headlines referred to it as the confirmation that narcolepsy is an autoimmune disease. Narcolepsy is an interesting disease both clinically (sleep attacks, cataplexy, sleep paralysis, visual hallucinations during early sleep and awakening…) and pathophysiologically. Current knowledge points at a selective death or damage in the neurons of the anterior part of the hypothalamus responsible of orexin production. Orexin (or hypocretin) is a secreted proteic neurotransmitter that regulates awakeness and apettite and whose levels in CSF are significantly lower in patients with narcolepsy than in controls. As usual, the exact cause of narcolepsy is unknown, although recent studies suggest that an autoimmune response, probably triggered by an environmental factor (let’s say, a virus), is the key process in its development (2).The paper by De la Herrán-Arita and colleagues reports an interesting, well-performed study that, contrary to what mainstream media say, does not demonstrate that narcolepsy is an autoimmune disease. But we’ll come to that later.

The autoimmune hypothesis of narcolepsy is not new at all. Among the genes that have been associated to narcolepsy, most play important roles in the immune system (3). The strongest association was found with the HLA-DQB1*0602 allele, which more than 95% of the narcolepsy patients carry. But other genes, related to the immune system have also been implicated (4). However, as in any other complex disease, genes don’t explain everything. What the HLA system does is to present antigens (the targets of an immune response, regardless of it is against a pathogen or an autoimmune one) to T cells (lymphocytes) and T cells are the ones enabled to kill a cell carrying that antigen or to call other cells so they are the ones killing the antigenic cell.  One of the main functions of HLA is to present viral particles to these lymphocytes. However, in many autoimmune diseases, the viral particles can have some similarities in structure with endogenous proteins and, thus, the immune system can divert the immune response to a self protein. Once the autoimmune response is set is usually very difficult to get rid of those autoreactive lymphocytes.

So, following this paradigm, it has been found an increase in the incidence of narcolepsy in two situations: the H1N1 influenza outburst and the H1N1 influenza vaccination campaing (5), meaning that one of the environmental triggers of narcolepsy are influenza-virus. In other reports it was described that Streptococci were the environmental trigger (6). What De la Herrán-Arita and co-workers demonstrate is, first, that some orexin pepetides bind to DQ0602 and that CD4 (helper) T cells secrete inflammatory cytokines (interferon) when incubated with orexin peptides bound to DQ0602. Second they demonstrate that patients with an increase in CD4 T cells reactivity against orexin also have a (slightly) increased reactivity against H1N1 influenza virus peptides. Finally, when those patients were vaccinated with Pandemrix (H1N1 vaccine), the number of orexin-reacting cells increased significantly. This summarizes a tremendous amount of work and a beautiful paper. In my opinion, however, a couple of clarifications have to be made with regard to the “this paper demonstrates that narcolepsy is an autoimmune disease” statement. To be honest, those extraordinary claims are not mentioned by the authors. Mainstream media are te ones concluding so strongly about the study findings. In fact, most of the clarifications have already been mentioned by the authors in the discussion, but i think some of them are worth to be highlighted because are useful for other “autoimmune diseases”.

– First, the finding of a specific immune response does not mean that the cause of a disease is autoimmune. Immune responses against self antigens are present in degenerative, infectious, toxic and genetic disorders as a way to clear debris from a damaged tissue, as a stochastic phenomenon or as a secondary immune reaction not causing damage by itself. So describing immune mechanisms does not imply autoimmune disease. A true autoimmune disease has to fulfill the Koch postulates that, so far, narcolpesy does not.

Linking with the previous point is interesting to note that narcolepsy patients react more strongly than controls against two orexin peptides, but there are several other orexin peptides to which patients and controls react exactly the same and to the same extent patients do against the two narcolepsy-specific peptides. Moreover siblings and monzigotic twins of narcolepsy patients that do not have the disease (but that share the HLA DQ0602) do not show increased reactivity against the orexin peptides. That can be due to key differences in the T cell receptor repertoire between patients and controls but then the important fact is neither the HLA nor the virus, but the T cell receptor (that is the one recognizing the peptide bound to the HLA). In that sense a recent paper by the same group has refined the genetic association studies and found that TCR beta chain is also a genetic risk factor for narcolepsy (4).

– Second, CD4 T cells are key cells in an immune response but do not cause damage to the tissue because they are not meant to cause damage, just help others do it. The ones that usually cause damage are B cells or CD8 T cells. CD8 T cells recognize HLA class I, which has not been found to be a risk factor in narcolepsy. B cells secrete antibodies but autoantibodies have not been found in all patients with narcolepsy. If that were the case, as in other autoimmune diseases such as myasthenia gravis or rheumatoid arthritis, they could be a great diagnostic tool and, so far, they are not. However, some years ago, another group not only described that TRIB2 antibodies (7) are present in a significant proportion of narcolepsy patients with cataplexy but that the innoculation of those antibodies to rodents led to hypersomnia, cognitive disturbances and cataplexy and, importantly, to orexin-secreting neurons loss  in those same animals (8). The demonstration that passive transfer of IgG determines narcolepsy is a much more solid proof that it is in fact an autoimmune disease that the paper we are currently reviewing.

Linking with that it arises, to me, a big question. Considering that CD4 T cells have been found to react against orexin peptides, the logical move is to try to demonstrate anti-orexin antibodies. I bet the authors tested them and did not find them (but this is just what i would have done and maybe they did not). Which is then the mechanism through which that CD4 T cell leads to selective orexin-neuron damage? It might be that after an acute immune response, when all orexin-secreting neurons are dead, the antibodies dissappear but, still, big knowledge gap there…

To sum-up… Scientific reports in mainstream media, and this is not the authors fault, usually not only overclaim when reporting new findings but also, very importantly, tend to forget the scientific context of the finding and condemn previous findings to oblivion. The study by De la Herrán-Arita is an important study and deserves credit for it but,

It’s not the first study to suggest that narcolepsy is autoimmune. Several others have suggested that before from different perspectives: epidemiological, clinical and pathophysiological.

It does not demonstrate that it is an autoimmune disease (that still needs some work to be demonstrated) and other studies demonstrate that with more compelling evidence. I believe narcolepsy is, indeed, autoimmune, but claiming that it with this study is overclaiming.

It does not link the disease with the Pandemrix vaccine or with the influenza virus. That was demonstrated before with epidemiological studies. What the study does is to provide evidence of a possible molecular mimicry link between orexin and H1N1 influenza.

Finally, for the record, the aim of this post is just to clarify, just as authors do in the discussion, the mainstream media overclaims, not to criticize an otherwise interesting study.

So, regardless of this particular study, evidence in favor of narcolepsy being an autoimmune disease is compelling and, thus, efforts to try to treat it as such (and not only the symptoms), need to be started. Narcolepsy might be another neuroimmune disease with potential disease-modifying treatment and neuroimmunologists should  be implicated.


1: De la Herrán-Arita AK et al, CD4+ T Cell Autoimmunity to Hypocretin/Orexin and Cross-Reactivity to a 2009 H1N1 Influenza A Epitope in Narcolepsy. Sci Transl Med. 2013 Dec 18;5(216):216ra176.
2: Singh AK, Mahlios J, Mignot E. Genetic association, seasonal infections and autoimmune basis of narcolepsy. J Autoimmun. 2013 Jun;43:26-31.
3: Tafti M, et al, DQB1 Locus Alone Explains Most of the Risk and Protection in Narcolepsy with Cataplexy in Europe. Sleep.
2014 Jan 1;37(1):19-25.
4: Han F et al Genome wide analysis of narcolepsy in China implicates novel immune loci and reveals changes in association prior to versus after the 2009 H1N1 influenza pandemic. PLoS Genet. 2013 Oct;9(10):e1003880
5: Dauvilliers Y, et al Increased risk of narcolepsy in children and adults after pandemic H1N1 vaccination in France. Brain. 2013 Aug;136(Pt 8):2486-96.
6: Aran A, et al Elevated anti-streptococcal antibodies in patients with recent narcolepsy
Sleep. 2009 Aug;32(8):979-83.
7: Cvetkovic-Lopes V, et al Elevated Tribbles homolog 2-specific antibody levels
in narcolepsy patients.
J Clin Invest. 2010 Mar;120(3):713-9.
8: Katzav A, et al Passive transfer of narcolepsy: anti-TRIB2 autoantibody positive patient IgG causes hypothalamic orexin neuron loss and sleep attacks in mice. J Autoimmun. 2013 Sep;45:24-30.