02/24/2014 12:50 pm ET | Updated Apr 26, 2014

New Stanford Study Paves the Way to a Diagnostic Blood Test for Narcolepsy

By Dr. Emmanuel Mignot

In a recently published paper, Alberto De la Herrán-Arita and colleagues closed the loop on the autoimmune basis of narcolepsy, showing that in patients with cataplexy reactive CD4+ T cells are present in most patients but not in controls. This discovery has two consequences. First, narcolepsy joins the club of diseases with an undisputable autoimmune basis. Because of the relative simplicity of the HLA association, the small number of epitopes presented, and the relationship with a specific triggering antigen in the 2009 H1N1 flu, it is likely to be a major model to understand autoimmunity in the future. We predict that narcolepsy will become the prototype of several other diseases that affect neurons. We believe these have been difficult to detect, as they likely primarily involve T cell rather than B cell immunity, a speculation we hope to discuss further in a future blog. Second is the possibility of diagnosing narcolepsy with a blood test, a development that is rapidly implementable, and the object of this discussion.

The principles of the test are simple. It has three components that must be combined and incubated together for 24 hours for the reaction to occur. The first is called T2.0602. For a long time, narcolepsy has been known to be selectively associated with two Human Leukocyte Antigen (HLA) gene variants DQB1*06:02 and DQA1*0102 that are almost a prerequisite to developing narcolepsy but are present in 25 percent of healthy subjects. We thus engineered a specialized cell line that only express DQB1*06:02 and DQA1*0102, which together create a protein composed of the alpha-01:02 chain and beta-06:02 chain, which we denote as T2.0602 and use it as antigen presenting cells.

The second ingredient is the autoantigen, natural pieces of hypocretins (HCRT56-68 and HCRT87-99) that are only abnormally recognized by narcoleptic immune cells. These 13 amino acid peptides will bind DQB0602 on the surface of T2.0602.

The third component is the patient sample, which consist of isolated CD4+ T cells. Typically, peripheral blood mononuclear cells (PBMCs) have been isolated days or weeks before from fresh blood and frozen until the day of the experiment. The CD4+ T cells which constitute about 50 percent of the PBMC are negatively isolated by magnetic beads technology from Invitrogen.

We combine the three components together in a plastic well then detect the IFN-gamma secretion resulted from the reaction as showed in below.

The following figure explains the principle of the test:


If abnormal, autoimmune T cells are present in the sample, they will recognize the DQB602-HCRT56-68 and DQB602-HCRT87-99 complexes, get abnormally activated and release a chemical called interferon gamma (INF-gamma), which we can detect using brown or blue color. As a result, any sample that contains T cells that are auto-reactive to hypocretin will create small spots of color, one for each T cell that locally releases INF-gamma. For each sample, a third well is used as a negative control: it only has the T cells and the T2.0602, but not the small auto-antigen that creates the sandwich activation and thus never reacts. A fourth well, a positive control, uses a chemical that activates all cells so that a continuous blue color is observed.

The following figure shows you the most recent example of what we obtained testing a series of patients and controls:


(Courtesy of Ling Lin.)

As can be seen, the pattern shown above is extremely clear (B, C, E = narcolepsy versus A, D, F = controls). In our first study, comparing approximately 50 patients (all with cataplexy) and 50 DQB0602 controls, we found the test to be very specific, meaning that the observation of this pattern of spots was only observed in narcolepsy. For a few narcolepsy patients who had the disease for more than 20 years, the activation was very weak, probably because once all the hypocretin cells are destroyed, the auto-reactive T cells may become quiescent. We are now testing a number of older narcoleptic patients and increasing the sensitivity and specificity of the assay to see if this pattern is general and to determine if we can still detect some auto-reactive cells in these patients. Overall, the impression is that for patients who have narcolepsy-cataplexy for less than 20 years, the sensitivity and specificity is at least as high as the Multiple Sleep Latency Test (MSLT), the gold standard for diagnosis.

Our next question is to test patients and a population sample with unexplained sleepiness: narcolepsy without cataplexy, idiopathic hypersomnia, or sleep apnea patients who remain sleepy. It is our suspicion that some cases without cataplexy will have reactivity, the question is: how many? It will also be important to test subjects both in sleep clinics and in the general population, as our recent study in the Wisconsin Sleep Cohort suggest that 2 cases without cataplexy may exist for each case with cataplexy. These cases do not go to the sleep clinic and frequently have obesity and sleep apnea, confounding the picture. Stanford has also filed a patent on the diagnostic method. Stay tuned for future developments as the science advances.


De la Herrán-Arita AK, Kornum BR, Mahlios J, Jiang W, Lin L, Hou T, Macaubas C, Einen M, Plazzi G, Crowe C, Newell EW, David MM, Mellins ED, Mignot E. "CD4+ T Cell Autoimmunity to Hypocretin/Orexin and Cross-Reactivity to a 2009 H1N1 Influenza A Epitope in Narcolepsy." Sci Transl Med. 18 December 2013, Vol. 5, Issue 216, p. 216ra176 Sci. Transl. Med. DOI: 10.1126/scitranslmed.3007762.
See for download.

Goldbart A, Peppard P, Finn L, Ruoff CM, Barnet J, Young T, and Mignot E. "Narcolepsy and predictors of positive MSLTs in the Wisconsin Sleep Cohort." Sleep. In press. See: for download.

Dr. Mignot is the director of the Stanford Center for Sleep Sciences and Medicine. This Center is the birthplace of sleep medicine and includes research, clinical, and educational programs that have advanced the field and improved patient care for decades. To learn more, visit us at: