A series of 7 cases of patients with narcolepsy with hypocretin deficiency without the HLA DQB1*06:02 allele

Abstract

Study Objectives:

We report data collected from 2 reference European sleep centers on a series of patients with narcolepsy with hypocretin-1 deficiency and absence of the human leukocyte antigens (HLA) DQB1*06:02 allele.

Methods:

Clinical data, HLA DQ markers, and cerebrospinal fluid assessments were collected retrospectively from Caucasian patients with a diagnosis of narcolepsy type 1 with cerebrospinal fluid hypocretin-1 deficiency (< 110 pg/ml) and absence of the HLA DQB1*06:02 allele, with follow-up with at least 1 visit within the last 4 years, consecutively admitted to 2 European sleep centers (Lugano, Switzerland and Montpellier, France).

Results:

Seven patients (3 of 29 patients in Lugano and 4 of 328 in Montpellier) were diagnosed with narcolepsy with hypocretin-1 deficiency and absence of HLA DQB1*06:02 (ie, 2% of patients with narcolepsy type 1). Regarding the HLA-DQB1 genotyping, 4 cases were positive for HLA DQB1*03:01, 1 for DQB1*03:02, and 3 for DQB1*02:01. Three patients had atypical cataplexy and 1 had no cataplexy. Only 2 patients had both a mean sleep latency of less than 8 minutes and more than 2 sleep onset rapid eye movement periods on the Multiple Sleep Latency Test, indicative of a less severe condition.

Conclusions:

Although rare, this series of 7 cases confirms that hypocretin-deficient narcolepsy should not be excluded in the absence of HLA DQB1*06:02, especially if patients are carriers of other high-risk HLA-DQB1 alleles (DQB1*03:01, *03:02, *02:01). These data support the hypothesis that narcolepsy type 1 is a wider disease spectrum linked to the loss of hypocretin peptide.

Citation:

Miano S, Barateau L, De Pieri M, et al. A series of 7 cases of patients with narcolepsy with hypocretin deficiency without the HLA DQB1*06:02 allele. J Clin Sleep Med. 2023;19(12):2053–2057.

BRIEF SUMMARY

Current Knowledge/Study Rationale: The majority of sleep specialists reported not measuring cerebrospinal fluid hypocretin-1 in their routine narcolepsy assessment. In patients suspected of narcolepsy without the human leukocyte antigens (HLA) DQB1*06:02 allele cerebrospinal fluid hypocretin-1 is very rarely measured.

Study Impact: Although rare, our series of 7 cases showed that patients with hypocretin-deficient narcolepsy can be HLA DQB1*06:02-negative, especially if they carry other high-risk HLA-DQB1 alleles (03:01, 03:02, 02:01). These patients often presented with less severe sleepiness and fewer sleep onset rapid eye movement periods on the Multiple Sleep Latency Test and atypical or absent cataplexy. Altogether, depending on the clinical context, the measurement of cerebrospinal fluid hypocretin-1 levels may still be relevant in the absence of HLA DQB1*06:02. Our data support that narcolepsy type 1 could be a spectrum of diseases, mostly related to the low level of hypocretin.

INTRODUCTION

Narcolepsy type 1 (NT1) is an orphan sleep disorder clinically characterized by excessive daytime sleepiness (EDS) and cataplexy (ie, sudden muscle weakness) that results from the loss of a small hypothalamic neuronal population producing the wake-promoting peptide orexin/hypocretin.¹ NT1 is thought to be a sporadic acquired immune-mediated condition that develops in genetically predisposed people.^2–4 Genetic markers of NT1 are diverse but primarily include the human leukocyte antigens (HLA) class II DRB1*15:01–DQA1*01:02–DQB1*06:02 haplotype shared by 95% of patients with narcolepsy.^5–10 NT1 is one of the most strongly HLA-associated diseases with an odds ratio of ∼250 for DQB1*06:02 and a relative risk for NT1 2–4 times higher in DQB1*06:02-homozygous vs -heterozygous patients across all groups.^11,12 Symptom severity did not differ between DQB1*06:02-homozygous and -heterozygous patients, indicating that HLA-DQB1*06:02 homozygosity increases susceptibility to NT1 but not disease severity.¹¹ Other alleles such as HLA DQB1*03:01, but also DQB1*03:02, DQB1*03:04, and DQB1*06:09, increase susceptibility to NT1 when present in trans of DQB1*06:02.^10,13–15 In contrast, HLA DQB1*05:01, DQB1*06:01, and DQB1*06:03 alleles provide dominant protection against NT1. Unlike DQB1*06:02, HLA-DQB1∗03:01 is associated with a young age at onset (of 3 years difference).^10,16 By definition, all patients with NT1 have low or absent cerebrospinal fluid (CSF) hypocretin-1 levels (< 110 pg/ml).^17,18 A large multicenter study assessed the association between CSF hypocretin-1 levels and DQB1*06:02 presence in patients with narcolepsy (552 patients with cataplexy and 144 others without cataplexy).¹⁶ They found rare cases (all with cataplexy) without the allele DQB1*06:02 and with low CSF hypocretin-1, constituting 1.7% of all cases with cataplexy and 1.8% of cases with low CSF hypocretin-1 independently of cataplexy.¹⁶ Among them, 5 patients had severe cataplexy, often occurring without clear triggers.¹⁶

The third edition of the International Classification of Sleep Disorders (ICSD-3)¹⁸ does not include HLA genotyping in the diagnostic criteria for narcolepsy, but its presence is often sought. Conversely, although it is within the ICSD-3 criteria¹⁸ hypocretin-1 in the CSF is rarely routinely measured because of low accessibility, cost, and the invasive nature of the spinal tap. A recent survey showed that the majority of sleep specialists reported not measuring CSF hypocretin-1 in their assessment of narcolepsy, and only 1% routinely performed it.¹⁹ Typically, patients with suspected narcolepsy without the HLA DQB1*06:02 allele do not have CSF hypocretin-1 measured, because the levels are most often normal due to the good negative predictive value of this test.²⁰

In this context, we wanted to describe the clinical phenotype of patients with narcolepsy with hypocretin-1 deficiency and absence of HLA DQB1*06:02, collecting data from 2 reference European sleep centers.

METHODS

Clinical data, HLA DQ markers, and CSF assessments were collected retrospectively from patients with a diagnosis of NT1 with hypocretin-1 deficiency and absence of HLA DQB1*06:02 allele, with follow-up with at least 1 visit within the last 4 years, consecutively admitted to the Sleep Center at the Neurocenter of Southern Switzerland, Lugano, and the Reference National Center for Narcolepsy-Hypersomnia, Montpellier University Hospital, France.

The diagnosis followed standard current criteria according to ICSD-3^18,21: video-polysomnography (V-PSG) was followed by a Multiple Sleep Latency Test (MSLT), without any medication or substances that could have an effect on sleep (drug-naïve condition or withdrawal). The lumbar puncture was performed in both centers during the hospitalization as part of the diagnostic procedure. CSF hypocretin-1 levels were determined in duplicate using the I-125 radioimmunoassay kit from Phoenix Pharmaceuticals Inc (Belmont, CA). according to the manufacturer’s recommendations at the University of Montpellier (for samples collected in France) and at the Institute of Clinical Chemistry at the University Hospital Zurich (for samples collected in Switzerland). CSF hypocretin-1 values below 110 pg/ml defined hypocretin deficiency.

Blood was collected to genotype HLA DQB1 by Luminex reverse sequence-specific oli-gonucleotide (Lifecodes HLA SSO Rapid kits, Immucor, France) in the Department of Immunology at the University of Montpellier (for samples collected in France) and derived from next-generation sequencing with an Omixon Holotype kit and HLA Twin software (Beverly, MA) in the Department of Physiology at the University of Lausanne (for samples taken in Switzerland).

The local ethic committees approved the study protocol, and all patients (and parents of minors) gave their informed consent to collection of data.

RESULTS

Table 1 summarizes the demographic, polysomnography, and MSLT sleep parameters of the 7 cases.

Table 1.

Demographic, cataplexy, and PSG and MSLT parameters of the 7 cases of patients with narcolepsy with hypocretin deficiency without the HLA DQB1*06:02 allele.

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Age, years	23	16	25	19	35	40	17
Sex	Male	Female	Male	Female	Female	Female	Female
Cataplexy	Typical	Atypical	Typical	Atypical	Atypical	Typical	Absence
Sleep latency at PSG (minutes)	2	16.2	6.7	20	22	27	18
Number of SOREMP at PSG	0	0	0	0	0	0	0
REM latency at PSG (minutes)	168.5	140.5	144.5	120	170	182	95
Sleep efficiency	94.5	95.8	80.8	87.7	76	66	43.7
Total sleep time (minutes)	430.5	396	331	429.5	372	343	470
Sleep stage N1 (%)	3.3	0.4	5	3	16.9	16.9	4
Sleep stage N2 (%)	51.2	64.4	59.1	51.7	63.5	44.7	51.8
Sleep stage N3 (%)	24.3	15.5	22.1	19.3	6.8	27.6	21.2
REM (%)	21.3	19.7	13.9	27.4	12.8	10.8	23
WASO (minutes)	23.2	1.3	72.1	28.5	95	119	15
AHI (events/h)	3.8	2.6	2.4	2.5	3.9	5.9	2.7
REM sleep without atonia	No	No	No	No	No	No	No
MSLT sleep latency (minutes)	3.4	10	6.2	13.4	11.4	5.8	9.4
No. of SOREMPs at MSLT	0	1	2	2	0	3	1

AHI = apnea-hypopnea index, HLA, human leukocyte antigens, MSLT = Multiple Sleep Latency Test, PSG = polysomnography, REM = rapid eyes movement, SOREMP = sleep onset REM period, WASO = wakefulness after sleep onset.

Cases from the Sleep Unit, Neurocenter of Southern Switzerland

Among the 29 patients diagnosed with hypocretin-1–deficient NT1, 3 cases were HLA DQB1*06:02-negative.

Case 1

Caucasian male, 23 years old (body mass index [BMI], 23.8 kg/m²). The V-PSG was normal, and the MSLT showed a pathological sleep latency (SL) of 3.4 minutes, with no sleep onset rapid eye movement periods (SOREMPs). He reported a positive family history for sleep paralysis in 2 sisters and EDS in 1 brother. He reported a mean sleep time of 7.5 hours, 1 hour awake during sleep, no symptoms of sleep inertia or sleep talking, rare episodes of sexsomnia, and rare hypnagogic hallucinations and sleep paralysis since he was 13 years old. He also reported EDS with sleep attacks since he was 18 years old. He described episodes of generalized cataplexy triggered by positive emotions. HLA-DQB1 genotyping was DQB1*02:01/02:01, and CSF hypocretin-1 was < 20 pg/ml. He received treatment with 100 mg/d modafinil plus 50 mg on demand, with satisfactory efficacy on EDS and ancillary symptoms.

Case 2

Caucasian female, 16 years old (BMI, 25.0 kg/m²). The V-PSG was normal, and the MSLT showed an SL of 10 minutes and 1 SOREMP. Family history was negative for sleep disorders. She reported a mean sleep time of about 8 hours at night and 1.5 hours during the afternoon, no sleep paralysis, visual hypnagogic hallucinations, and sleep inertia. She reported 2 recent abrupt falls, without any triggers, being atypical for cataplexy, and EDS for 1 year. She received a diagnosis of attention-deficit/hyperactivity disorder and learning disability and had been treated with methylphenidate for 5 years. She reported sleepwalking since was 13 years old. She was withdrawn from treatment 2 weeks before V-PSG and MSLT. HLA-DQB1 genotyping was DQB1*02:01/03:01. CSF hypocretin-1 level was undetectable. A treatment with modafinil was started with scarce efficacy; similarly, she did not respond to sodium oxybate. Now she is under treatment with 40 mg/d long-acting methylphenidate. She did not apparently report any other episodes of atypical cataplexy.

Case 3

Caucasian male, 25 years old (BMI, 24.5 kg/m²). The V-PSG showed a periodic leg movement index of 38.5 events/h, and the MSLT showed 2 SOREMPs and a mean SL of 6.2 minutes. Family history was positive for cardiovascular diseases and restless legs syndrome. He reported a night sleep time of 7.5 hours, no awakenings during sleep, and no sleep paralysis or hypnagogic hallucinations but reported EDS and brief sleep attacks for the last 2 years. He also reported mild symptoms of restless legs syndrome, with normal blood ferritin value. He also described monthly partial episodes of cataplexy. HLA-DQB1 genotyping was DQB1*03:01/06:04, and hypocretin-1 was undetectable. He started treatment with 100 mg/d modafinil and 50 mg when needed, with complete remission of symptoms.

Cases from the Reference National Center for Narcolepsy-Hypersomnia, Montpellier, France

Among the 328 patients diagnosed with hypocretin-1–deficient NT1, 4 cases were HLA DQB1*06:02-negative.

Case 4

Caucasian female, 19 years old (BMI, 23.2 kg/m²). The V-PSG was normal except for some episodes of neck myoclonus during rapid eye movement sleep. MSLT showed a mean SL of 13.4 minutes and 2 SOREMPs. Rapid eye movement sleep behavior disorder, sleep talking, and bruxism were reported in the family. She received a diagnosis of attention-deficit disorder during childhood and her mother had dyslexia. She described a mean night sleep time of 7 hours during the week, some short episodes of wakefulness during sleep, sleep inertia in the morning, sleep talking, night terrors, and frequent episodes of sleep paralysis and hypnagogic hallucinations. She also reported partial unilateral cataplexy (left knee), usually triggered by positive and negative emotions, but sometimes spontaneous. She had had severe EDS since she was 15 years old, a few months after contracting viral meningitis. EDS was characterized by at least 3 naps per day, lasting sometimes more than 1 hour. HLA-DQB1 genotyping was DQB1*02:02/03:01, and CSF hypocretin-1 level was undetectable. She started treatment with 200 mg/d modafinil, which was poorly tolerated (headache, palpitations). Next therapies (first pitolisant, then methylphenidate) were also poorly tolerated.

Case 5

Caucasian female, 35 years old (BMI, 22.8 kg/m²). The V-PSG was normal and MSLT showed a mean SL of 11.4 minutes without SOREMPs. Sleepwalking and bruxism were found in the family. She reported EDS evolving since she was a child, worsening in adolescence, and brief sleep attacks all day long. She was diagnosed with attention-deficit/hyperactivity disorder in the past year with high intellectual potential and was treated with methylphenidate. She reported a night sleep time of 12 hours, 1 awakening of 1.5 hours during sleep, and no sleep talking, sleep paralysis, or hypnagogic hallucinations. During holidays, she could sleep 16 out of 24 hours. She described atypical unilateral cataplexy, evolving from the age of 20 years, without triggering factors. HLA-DQB1 genotyping was DQB1*02:01/03:01. CSF hypocretin-1 level was 50 pg/ml. She started treatment with poor response to methylphenidate (54 mg/d); then, sodium oxybate (9 g/night) was introduced but was not tolerated and was ineffective. Finally, she was given a combination of pitolisant (18 mg/d) and solriamfetol (150 mg/d) with good efficacy on EDS and cataplectic events but not on prolonged sleep time.

Case 6

Caucasian female, 40 years old (BMI, 29.1 kg/m²; gained 6 kg in the last year). The V-PSG showed many periodic leg movements (index 55 per hour). MSLT showed 3 SOREMPs and a mean SL of 5.8 minutes. Obstructive sleep apnea was reported in the family. Her mother had rheumatoid arthritis. She reported a sleep time of about 8–9 hours at night and 1.5 hours in the afternoon. She reported hypnagogic hallucinations, without sleep paralysis. She had EDS for 4 months, without sleep inertia. She reported cataplexy for 3 months, triggered by emotions (both positive and negative) and orgasm. HLA-DQB1 genotyping was DQB1*03:03/05:01. CSF hypocretin-1 level was undetectable. She started treatment with 300 mg/d modafinil, which was effective.

Case 7

Caucasian female, 17 years old (BMI, 22.5 kg/m²; gained 9 kg in the last year). The V-PSG was normal. MSLT showed a mean SL of 9.4 minutes and 1 SOREMP at the first session only. A prolonged 32-hour PSG assessment on the Bedrest protocol showed a total sleep time of 22.5 hours of sleep, supporting first the diagnosis of idiopathic hypersomnia.²² Insomnia and bruxism were found in the family and attention-deficit/hyperactivity disorder in a brother. She had a past history of major depressive disorder treated with escitalopram. This treatment was stopped 6 months before sleep assessment because of complete remission. She presented sleepwalking and night terrors when she was a child. She reported a sleep time of about 8 hours at night and 1.5 hours in the afternoon. During holidays, she could sleep 18 out of 24 hours. She did not report sleep paralysis, hypnagogic hallucinations, or cataplexy. She reported nightmares and sleep inertia in the morning. She had severe EDS for 18 months. HLA DQB1* genotyping was *03:02/04:02. CSF hypocretin-1 level was undetectable. She started treatment with modafinil and then pitolisant, both of which were ineffective.

DISCUSSION

Our case series describe 7 Caucasian patients (3 among 29 patients in Lugano, Switzerland, and 4 among 328 in Montpellier, France) with narcolepsy with hypocretin-1 deficiency and absence of HLA DQB1*06:02 (ie, 2% of patients with NT1). According to the ICSD-3 criteria, all patients were diagnosed with NT1, and EDS and CSF hypocretin-1 levels were undetectable or very low in all patients. Three patients had atypical cataplexy (ie, attacks purely unilateral, no clear precipitants for episodes or if only negative emotions act as triggers) and 1 had no cataplexy.²³ Interestingly, only 2 cases had both a mean SL of less than 8 minutes and more than 2 SOREMPs on the MSLT. Two SOREMPs or more on the MSLT were documented in 3 cases and a mean SL on the MSLT below 8 minutes in 3 of the 7 cases. Of note, 2 patients had no SOREMP despite being hypocretin-deficient. Finally, none of our patients had rapid eye movement sleep without atonia.

Regarding the HLA-DQB1 genotyping, 4 cases (2 from Switzerland and 2 from France) were positive for HLA DQB1*03:01, and 1 case was positive for DQB1*03:02 (from France), 2 alleles already known as risk factors for NT1 in association with HLA DQB1*06:02.^{10,14–16,24} We also found the presence of HLA DQB1*02:01 in 3 cases (2 from Lugano and 1 from France), an association never reported with NT1. In a recent multicenter study, only 9 among 552 patients with cataplexy and low CSF hypocretin-1 did not carry the DQB1*06:02 allele (eg, often with the presence of DQB1*03:01, *03:02, *05:01, and *06:01 alleles) (ie, 1.6% of patients with NT1).²⁵ This value is very close to that of our study, although the data show higher figures in Switzerland, possibly due to a smaller population size. This limited amount of data demonstrates that patients with hypocretin-deficient narcolepsy may be HLA DQB1 *06:02-negative, carriers of other at-risk HLA-DQB1 alleles, often with less severe sleepiness and less SOREMPs on the MSLT, frequently without positive MSLT criteria for narcolepsy, sometimes even without any SOREMP, and, finally, with atypical or even absent cataplexy. We found no obese patients, no young children, and mainly women (78%) among our patients. Altogether, depending on the clinical context, the measurement of CSF hypocretin-1 levels may be relevant in the absence of HLA DQB1*06:02. Our data support that NT1 could be a spectrum of diseases, mostly related to the low level of hypocretin.²⁶

This case series study has several limitations. First, the sample is very small, preventing us from performing statistical analyses to compare these patients with others carrying the HLA DQB1*06:02 allele. Second, the studied population was selected among patients referred to specialized centers for central disorders of hypersomnolence, which could be a selection bias. Our description, however, can show that these patients seem very heterogeneous in terms of severity, clinical phenotype, MSLT results, and drug response. The rarity and heterogeneity of narcolepsy with low CSF hypocretin-1 levels in the absence of HLA DQB1*06:02 would benefit from a larger cohort, in multiple reference centers, such as among the European Narcolepsy Network to compare them to typical NT1 patients with positive HLA DQB1*06:02.²⁷

In conclusion, although rare, our series of 7 cases showed that patients with hypocretin-deficient narcolepsy can be HLA DQB1*06:02-negative, especially if they carry other high-risk HLA-DQB1 alleles (*03:01, *03:02, *02:01). Depending on the clinical context, the measurement of CSF hypocretin-1 levels may still be relevant in the absence of HLA DQB1*06:02. These data support the hypothesis that NT1 is a wider disease spectrum linked to the loss of hypocretin peptide.

DISCLOSURE STATEMENT

All authors have seen and approved the manuscript. Work for this study was performed at the Sleep Center at the Neurocenter of Southern Switzerland, Lugano, and the Reference National Center for Narcolepsy-Hypersomnia, Montpellier University Hospital, France. L. Barateau received funds for traveling to conferences from Idorsia and Bioprojet and board engagements from Jazz, Takeda, Idorsia, and Bioprojet. Y. Dauvilliers received funds for seminars, board engagements and travel to conferences from Jazz, Idorsia, Takeda, Avadel, and Bioprojet. The other authors report no conflicts of interest.

ABBREVIATIONS

BMI

body mass index

CSF

cerebrospinal fluid

EDS

excessive daytime sleepiness

HLA

human leukocyte antigens

ICSD

International Classification of Sleep Disorders

MSLT

Multiple Sleep Latency Test

NT1

narcolepsy type 1

SL

sleep latency

SOREMP

sleep onset rapid eye movement period

V-PSG

video-polysomnography