Status of KLS Research at Stanford University

November 2007

Toward a better definition of the clinical syndrome of KLS

Dr. Arnulf (Hospital Pitié-Salpétrière in Paris, France) conducted a metanalysis of the KLS literature during a sabbatical year at Stanford University. This landmark publication is now available in the medical journal called Brain (Arnulf et al., 2005), and can be provided to physicians who know little about the disorder.  A total of 195 published reports about KLS from 1962-2004 were collected and those in foreign language translated.  Cases not consistent with true KLS were excluded.  Clinical data for a remaining 186 cases were compiled to produce the first comprehensive summary statistics for the disorder.  Primary KLS cases were found mostly in men (68%) and occurred sporadically worldwide. The median age of onset was 15 years (81% during the second decade of life) and the syndrome lasted 8 years, with seven episodes of 10 days, recurring every 3.5 months (median values). The disease lasted longer in women and in patients experiencing less frequent episodes during the first year. An important finding of this analysis was

the unappreciated observation that KLS is not simply a sleep disorder characterized by hypersomnia, hyperphagia and hypersexuality.  Indeed, beside increased sleep amounts during episodes, cognitive and perceptual abnormalities were always reported, most typically described by the patient as being in a fog, underwater, almost in a dream, or somehow separated from the surrounding reality.  This under-appreciated symptom was found to be more consistent than hyperphagia or hypersexuality, symptoms often overemphasized because of their dramatic value.  Hyperphagia and hypersexuality are in fact present only in some episodes, mild or not present at all.  The diversity of symptoms reported by KLS patients suggests that the pathology affects a large portion of the brain.  Indeed, this picture is consistent with a recent brain imaging study that showed a hypoperfusion of the thalamus (a sleep and sensory gating area) during episodes, and inconsistent hypoperfusion in the hypothalamus (controlling sleep, appetite and instinctual behaviors) and in multiple cortical regions (responsible for perception and cognition).

A second finding was the lack of efficacy for most reported attempted therapies. Indeed, it was found that the only agent that may have more effect than doing nothing was Lithium, but the response rate in preventing recurrence was only 41% and this was based only on 29 reports (our second, as yet unpublished study, actually suggests a lower response rate of approximately 20%). Similarly, stimulants were found to be active on sleepiness, but as they did not act on abnormal cognition, and were often not helpful if not deleterious.

A registry of active KLS cases

More recently, thanks to the KLS Foundation, we have been able to collect new clinical data and genetic material on a large number of KLS cases.  A KLS questionnaire was created, with the aim of gathering further clinical information and to explore other potential predisposing and causal factors.  Age, sex and ethnically matched controls were studied using a similar questionnaire. Finally, we collected blood samples from patients and from healthy controls to create a registry of samples for research studies. 

Thanks to the dedication of patients, amazing help from the KLS Foundation and hard work from our staff, particularly Isabelle Arnulf and Jennifer File (patient coordinator), we were able to collect over 100 new KLS cases.  In many instances we also collected blood samples from parents, which will help future genetic analysis.  For us, this surprising success meant that a lot of patients were starving for help, and that the disease is probably more frequent than believed.  The prevalence of less than 1 case of KLS per million typically quoted is in my opinion a gross underestimation.  A detailed report is now under review for publication in a prestigious scientific journal, Annals of Neurology.

New Insights into KLS:
The systematic collection of clinical data in a new series of KLS patients in comparison with healthy controls led to a few additional insights. 
First, we found that disease duration was longer than in the metanalysis, a median of 12 years instead of 8 years. 
Second, as in the prior study, we found that treatments were generally ineffective and in this case lithium or other mood stabilizers such as anticonvulsivants had similarly poor response rates (20%).
Third, we found that patients in between episodes were quite normal; indeed, we found that patients were only slightly more anxious and overweight versus age, sex and ethnically matched controls. Most notably, patients were not more depressed, and did not have increased family history for depression or other psychiatric and neurological disorders.  This last finding is important because some psychiatrists still think KLS is a variant of bipolar disorder due to the recurrence pattern, the hypersomnia and the possible beneficial effect of Lithium or other mood stabilizers.

Familial predisposition and a possible increased risk in Ashkenazi Jews
One of the surprising findings of this study was the observation that in 5 cases out of 108, more than one family member was affected with KLS.  This is surprising considering how rare KLS is believed to be. More than 20 such families with multiple cases of KLS are known to the KLS Foundation, suggesting either a genetic predisposition (genes passed from the parents to several children in the family) or shared environmental factors (for example being exposed to the same virus within the same family).

Further, we found that 14% of the KLS cases recruited in the United States were Jewish Ashkenazi. This figure is much higher than the 2% proportion of such individuals in the general US population and was also much higher than Jewish patients in our database of subjects with other sleep disorders.  This was only partially a surprise, as one of the largest previously published case series of KLS cases was from Israel (Gadoth et al., 2001).  To us, this result suggest that Ashkenazi Jews, who are known to carry special genetic changes because of their origin from a relatively small number of individuals 4,000 years ago, may have a specific genetic factor predisposing to KLS.  Importantly however, many more cases were non-Jewish and it is always possible that the inflated value is the result of a recruitment bias, for example if the American Jewish population is more aware of the disorder than other populations.  To illustrate this point, very few African Americans and Latinos were found in our recently collected KLS sample, possibly because they did not come to our attention due to socioeconomic differences.

KLS, HLA and autoimmunity: a disappointing hypothesis

Many possible etiologies have been suggested for KLS.  The presence of flu-like symptoms at the onset of the disease in many cases suggests a possible infectious cause or at least an infectious trigger.  In 2002, Dauvilliers et al. made an exciting finding, reporting on the observation that 47% of his KLS cases in Europe, 30 in total, were Human Leukocyte Antigen (HLA)-DQB1*02 positive in contrast to approximately 25% of controls.  HLA are proteins that are very important for regulating the immune system, the system that fights infections in our body.  Most diseases with an HLA association are autoimmune diseases, conditions where the immune system mistakenly attacks one’s own self, for example in multiple sclerosis where the immune system attacks insulating components in our nerves.  The hypothesis was very appealing, as many autoimmune diseases have a characteristic remitting-relapsing course like KLS, corresponding to periods of sudden immune activation, and may be initiated by environmental or infectious triggers.  To further test this hypothesis, we examined the HLA-DR and DQ associations in our large sample of recently collected KLS patients, but were unable to find a correlation, suggesting that the first report was a chance finding.  Dr. Arnulf, in Paris, France, also tried immune related therapies, most notably intravenous immunoglobulin (a therapy effective in some autoimmune diseases) but did not observe consistent effects.

The environment/genetic interaction hypothesis and the search for a possible pathogen causing KLS

As mentioned above, our working hypothesis is that of a genetic predisposition to an infectious process that may affect the brain and would result in a waxing and waning brain infection/inflammation thereby producing KLS episodes.  After several years, the immune system could successfully keep the process at bay and the disease would terminate.  In some aspect, KLS could be similar to herpes infections, which often reoccur in life.
To address this issue, we are now collecting nasal swab samples from patients at the beginning of a KLS episode. We hope these samples will identify a possible viral agent that could be involved in KLS. We have now collected approximately 10 samples, and are hoping to obtain 20 such samples in order to initiate a molecular analysis by Dr. DeRisi at UCSF, who specializes in identifying unknown viruses.  In parallel, we are sending serum samples to screen for the presence of a number of viruses known to affect the brain, such as the recently identified HHV-6.

Collaboration with Israel

The observation that KLS preferentially affects Ashkenazi Jews offers a unique research opportunity. Indeed, if this is the case, it is likely that whatever genetic factor is involved in predisposing to KLS, it is enriched in this population. Further, as this ethnic group descends from a relatively smaller number of individuals, it is more likely that only one or few genetic subtypes are involved in this population, which greatly increases the probability of finding a genetic culprit. 

Based on the above, our first goal has been to collect as many KLS cases in this ancestry as possible in the US and elsewhere. In this endeavor, we have been extremely fortunate to recruit Dr. Yakov Sivan from Israel. Thanks to him, we now have more than 40 cases and controls from this ethnic subgroup.  A genome wide association is currently ongoing with these samples.

A need for education

Another area where we feel much remains to be done is in education.  Indeed, very little exists in term of diagnostic and therapeutic guidelines.  We believe that to fill this gap we need to publish intensively in high impact journals.  The publications mentioned above, published in Brain (mostly European readership) and under revision at Annals of Neurology, will be extremely effective.  We are also trying to publish commentaries on this intriguing disease as frequently as possible.

KLS International research team
Principal investigators and staff:
Emmanuel Mignot MD, PhD, Director, Center for Narcolepsy and Hypersomnia research, Stanford University USA
Isabelle Arnulf, MD, PhD, Fédération des Pathologies du Sommeil, Hôpital Pitié-Salpêtrière, Paris, France
Juliette Faraco, PhD, Center for Narcolepsy and Hypersomnia research, Stanford University
Joachim Hallmayer, MD, PhD, Stanford University, USA
Neil Risch, PhD Genetic epidemiology, University of California, San Francisco, USA
Yakov Sivan, MD, Dana-Children’s Hospital, Tel-Aviv, Israel

Clinical patient coordinators:
USA: Jacob Markovitz, Stanford University, USA
Israel: Esti Cohen Gabizon, Dana-Children’s Hospital. Tel-Aviv, Israel
Europe: Isabelle Arnulf, Hôpital Pitié-Salpêtrière, Paris, France

References:

Arnulf I, Zeitzer JM, File J, Farber N, Mignot E. Kleine-Levin syndrome: a systematic review of 186 cases in the literature. Brain. 2005 Dec;128(Pt 12):2763-76.

Dauvilliers Y, Mayer G, Lecendreux M, Neidhart E, Peraita-Adrados R, Sonka K, et al. Kleine–Levin syndrome: an autoimmune hypothesis based on clinical and genetic analyses. Neurology 2002; 59: 1739–45.

Gadoth N, Kesler A, Vainstein G, Peled R, Lavie P. Clinical and polysomnographic characteristics of 34 patients with Kleine–Levin syndrome. J Sleep Res 2001; 10: 337–41.

Huang YS, Guilleminault C, Kao PF, Liu FY. SPECT findings in the Kleine-Levin syndrome. Sleep. 2005 Aug 1;28(8):955-60.

Mignot E., A step forward for restless legs syndrome. Nat Genet. 2007 Aug;39(8):938-9.