Novel risk loci in LGI1-antibody encephalitis: genome-wide association study discovery and validation cohorts.

Encephalitis with antibodies to leucine-rich glioma-inactivated 1 (LGI1-Ab-E) is a common form of autoimmune encephalitis, presenting with seizures and neuropsychiatric changes, predominantly in older males. More than 90% of patients carry the human leucocyte antigen (HLA) class II allele, HLA-DRB1*07:01. However, this is also present in 25% of healthy controls. Therefore, we hypothesised the presence of additional genetic predispositions. In this genome-wide association study and meta-analysis, we studied a discovery cohort of 131 French LGI1-Ab-E and a validation cohort of 126 American, British and Irish LGI1-Ab-E patients, ancestry-matched to 2613 and 2538 European controls, respectively. Outside the known major HLA signal, we found two single nucleotide polymorphisms (SNPs) at genome-wide significance (p < 5 x 10-8), implicating PTPRD, a protein tyrosine phosphatase, and LINC00670, a non-protein coding RNA gene. Meta-analysis defined four additional non-HLA loci, including the protein coding COBL gene. Polygenic risk scores with and without HLA variants proposed a contribution of non-HLA loci. In silico network analyses suggested LGI1 and PTPRD mediated interactions via the established receptors of LGI1, ADAM22 and ADAM23. Our results identify new genetic loci in LGI1-Ab-E. These findings present opportunities for mechanistic studies and offer potential markers of susceptibility, prognostics and therapeutic responses.

Probabilistic Sleep Staging in MSLTs across Hypersomnia Disorders.

STUDY OBJECTIVES: This study aimed to identify novel markers of narcolepsy type 1 (NT1) using between-nap opportunity periods ('Lights On') and in-nap opportunity periods ('Lights Off') features of Multiple Sleep Latency Test (MSLT) recordings. We hypothesized that NT1 could be identified both from sleep-wake instability and patterns of sleepiness during wakefulness. Further, we explored if MSLTs from NT1 and narcolepsy type 2 (NT2) patients could be distinguished despite having the same diagnostic thresholds. METHODS: We analyzed 'Lights On' and 'Lights Off' periods of the MSLT, extracting 163 features describing sleepiness, microsleep, and sleep stage mixing using data from 177 patients with NT1, NT2, Idiopathic Hypersomnia (IH), and Subjective Hypersomnia (sH) from three sleep centers. These features were based on automated probabilistic sleep staging, also denoted as hypnodensities, using U-Sleep. Hypersomnias were differentiated using either or both features from 'Lights On' and 'Lights Off'. RESULTS: Patients with NT1 could be distinguished from NT2, IH, and sH using features solely from 'Lights On' periods with a sensitivity of 0.76 and specificity of 0.71. When using features from all periods of the MSLT, NT1 was distinguished from NT2 alone with a sensitivity of 0.77 and a specificity of 0.84. CONCLUSIONS: The findings of this study demonstrate microsleeps and sleep stage mixing as potential markers of the sleep attacks and unstable sleep-wake states common in NT1. Further, NT1 and NT2 could be frequently distinguished using 'Lights Off' features.

Lifestyle Profiling Using Wearables and Prediction of Glucose Metabolism in Individuals with Normoglycemia or Prediabetes.

This study examined the relationship between lifestyles (diet, sleep, and physical activity) and glucose responses at a personal level. 36 healthy adults in the Bay Area were monitored for their lifestyles and glucose levels using wearables and continuous glucose monitoring (NCT03919877). Gold-standard metabolic tests were conducted to phenotype metabolic characteristics. Through the lifestyle data (2,307 meals, 1,809 nights, and 2,447 days) and 231,206 CGM readings from metabolically-phenotyped individuals with normoglycemia or prediabetes, we found: 1) eating timing was associated with hyperglycemia, muscle insulin resistance (IR), and incretin dysfunction, whereas nutrient intakes were not; 2) timing of increased activity in muscle IS and IR participants was associated with differential benefits of glucose control; 3) Integrated ML models using lifestyle factors predicted distinct metabolic characteristics (muscle, adipose IR or incretin dysfunction). Our data indicate the differential impact of lifestyles on glucose regulation among individuals with different metabolic phenotypes, highlighting the value of personalized lifestyle modifications.

RESTORE: Once-nightly oxybate dosing preference and nocturnal experience with twice-nightly oxybates.

Objective/Background: Preference for extended-release, once-nightly sodium oxybate (ON-SXB, FT218) vs twice-nightly immediate-release (IR) oxybate was assessed in participants switching from IR oxybate to ON-SXB in an open-label/switch study, RESTORE (NCT04451668). Patients/Methods: Participants aged ≥16 years with narcolepsy who completed the phase 3 REST-ON trial, were oxybate-naive, or were on a stable IR oxybate dose (≥1 month) were eligible for RESTORE. For participants who switched from twice-nightly dosing to ON-SXB, initial doses were closest or equivalent to their previous nightly IR oxybate dose. These participants completed a questionnaire at baseline about nocturnal adverse events associated with the middle-of-the-night IR oxybate dose in the preceding 3 months, a preference questionnaire after 3 months of stable-dose ON-SXB, and an end-of-study (EOS) questionnaire. Results: There were 130 switch participants; 92/98 (93.9 %) who completed the preference questionnaire preferred ON-SXB. At baseline, 69.2 % reported missing their second IR oxybate dose at least once; in these cases, 80 % felt worse the next day. Approximately 39 % reported taking a second nightly IR oxybate dose >4 h after the first dose, 51 % of whom felt somewhat to extremely groggy/unsteady the next morning; 120 participants (92 %) reported getting out of bed after their second oxybate dose. Of those, 9 (8 %) reported falls and 5 (4 %) reported injuries. Of the switch participants who completed the EOS questionnaire, 91.2 % felt better able to follow the recommended ON-SXB dosing schedule. Conclusions: The second nightly IR oxybate dose presents significant treatment burdens and adherence concerns. Participants overwhelmingly preferred the once-nightly dosing regimen of ON-SXB.

Mortality Risk Assessment Using Deep Learning-Based Frequency Analysis of EEG and EOG in Sleep.

STUDY OBJECTIVES: To assess whether the frequency content of electroencephalography (EEG) and electrooculography (EOG) during nocturnal polysomnography (PSG) can predict all-cause mortality. METHODS: Power spectra from PSGs of 8,716 participants, included from the MrOS Sleep Study and the Sleep Heart Health Study (SHHS), were analyzed in deep learning-based survival models. The best-performing model was further examined using SHapley Additive Explanation (SHAP) for data-driven sleep-stage specific definitions of power bands, which were evaluated in predicting mortality using Cox Proportional Hazards models. RESULTS: Survival analyses, adjusted for known covariates, identified multiple EEG frequency bands across all sleep stages predicting all-cause mortality. For EEG, we found an all-cause mortality hazard ratio (HR) of 0.90 (CI95% 0.85-0.96) for 12-15 Hz in N2, 0.86 (CI95% 0.82-0.91) for 0.75-1.5 Hz in N3, and 0.87 (CI95% 0.83-0.92) for 14.75-33.5 Hz in REM sleep. For EOG, we found several low-frequency effects including an all-cause mortality HR of 1.19 (CI95% 1.11-1.28) for 0.25 Hz in N3, 1.11 (CI95% 1.03-1.21) for 0.75 Hz in N1, and 1.11 (CI95% 1.03-1.20) for 1.25-1.75 Hz in Wake. The gain in the concordance index (C-index) for all-cause mortality is minimal, with only a 0.24% increase: The best single mortality predictor was EEG N3 (0-0.5 Hz) with C-index of 77.78% compared to 77.54% for confounders alone. CONCLUSION: Spectral power features, possibly reflecting abnormal sleep microstructure, are associated with mortality risk. These findings add to a growing literature suggesting that sleep contains incipient predictors of health and mortality.

Associations between self-reported parasomnias and psychiatric illness in 370,000 patients with sleep disorders.

AIM: To assess self-reported parasomnias in patients with sleep disorders and explore relationships with psychiatric illness, comorbidities, subjective sleep assessments, and polysomnographic study results. METHODS: Results from intake questionnaires and polysomnographic assessments, collected from 240 sleep centers across 30 US states between 2004 and 2019, were analyzed retrospectively. Of 540,000 total patients, 371,889 who answered parasomnia-specific questions were included. Patients responding "often" or "always" to parasomnia-specific questions were considered "symptom-positive," whereas a "few times" or "never" were considered "symptom-negative" (controls). RESULTS: The study sample was 54.5% male with mean age 54 years (range, 2-107 years). Frequencies for the different parasomnias were 16.0% for any parasomnia, 8.8% for somniloquy, 6.0% for hypnagogic hallucinations, 4.8% for sleep-related eating disorder, 2.1% for sleep paralysis, and 1.7% for somnambulism. Frequent parasomnias were highly associated with diagnosed depression (odds ratio = 2.72). All parasomnias were associated with being younger and female and with symptoms of depression, anxiety, insomnia, restless legs, pain, medical conditions, fatigue, and sleepiness. Associations with objective sleep metrics showed characteristics of consolidated sleep and differentiated weakly between nonrapid eye movement sleep and rapid eye movement sleep parasomnias. Machine learning accurately classified patients with parasomnia versus controls (balanced accuracies between 71% and 79%). Benzodiazepines, antipsychotics, and opioids increased the odds of experiencing parasomnias, while antihistamines and melatonin reduced the odds. Z-drugs were found to increase the likelihood of a sleep-related eating disorder. CONCLUSION: Our findings suggest that parasomnias may be clinically relevant, yet understudied, symptoms of depression and anxiety. Further investigation is needed to quantify the nature of multimorbidity, including causality and implications for diagnosis and treatment.

Digital Measures Development: Lessons Learned from an Expert Workshop Addressing Cross-Therapeutic Area Measures of Sleep.

Introduction: The Digital Measures Development: Core Measures of Sleep project, led by the Digital Medicine Society (DiMe), emphasizes the importance of sleep as a cornerstone of health and the need for standardized measurements of sleep and its disturbances outside the laboratory. This initiative recognizes the complex relationship between sleep and overall health, addressing it as both a symptom of underlying conditions and a consequence of therapeutic interventions. It aims to fill a crucial gap in healthcare by promoting the development of accessible, nonintrusive, and cost-effective digital tools for sleep assessment, focusing on factors important to patients, caregivers, and clinicians. Methods: A central feature of this project was an expert workshop conducted on April 19th, 2023. The workshop convened stakeholders from diverse backgrounds, including regulatory, payer, industry, academic, and patient groups, to deliberate on the project's direction. This gathering focused on discussing the challenges and necessities of measuring sleep across various therapeutic areas, aiming to identify broad areas for initial focus while considering the feasibility of generalizing these measures where applicable. The methodological emphasis was on leveraging expert consensus to guide the project's approach to digital sleep measurement. Results: The workshop resulted in the identification of seven key themes that will direct the DiMe Core Digital Measures of Sleep project and the broader field of sleep research moving forward. These themes underscore the project's innovative approach to sleep health, highlighting the complexity of omni-therapeutic sleep measurement and identifying potential areas for targeted research and development. The discussions and outcomes of the workshop serve as a roadmap for enhancing digital sleep measurement tools, ensuring they are relevant, accurate, and capable of addressing the nuanced needs of diverse patient populations. Conclusion: The Digital Medicine Society's Core Measures of Sleep project represents a pivotal effort to advance sleep health through digital innovation. By focusing on the development of standardized, patient-centric, and clinically relevant digital sleep assessment tools, the project addresses a significant need in healthcare. The expert workshop's outcomes underscore the importance of collaborative, multi-stakeholder engagement in identifying and overcoming the challenges of sleep measurement. This initiative sets a new precedent for the integration of digital tools into sleep health research and practice, promising to improve outcomes for patients worldwide by enhancing our understanding and measurement of sleep.

Novel biomarkers derived from the Maintenance of Wakefulness Test as predictors of sleepiness and response to treatment.

The Maintenance of Wakefulness Test (MWT) is a widely accepted objective test used to evaluate daytime somnolence and is commonly used in clinical studies evaluating novel therapeutics for excessive daytime sleepiness. In the latter, sleep onset latency (SOL) is typically the sole MWT endpoint. Here, we explored microsleeps, sleep probability measures derived from automated sleep scoring, and quantitative electroencephalography (qEEG) features as additional MWT biomarkers of daytime sleepiness, using data from a phase 1B trial of the selective orexin receptor 2 agonist danavorexton (TAK-925) in people with narcolepsy type 1 (NT1) or type 2 (NT2). Danavorexton treatment reduced the rate and duration of microsleeps during the MWT in NT1 (days 1 and 7; p ≤ 0.005) and microsleep rate in NT2 (days 1 and 7; p < 0.0001). Use of an EEG-sleep-staging-derived measure to determine the probability of wakefulness for each minute revealed a novel metric to track changes in daytime sleepiness, which were consistent with the θ/α ratio, a known biomarker of drowsiness. The slopes of line-fits to both the log-transformed sleepiness score or log-transformed θ/α ratio correlated well to (inverse) MWT SOL for NT1 (R = 0.93 and R = 0.83, respectively) and NT2 (R = 0.97 and R = 0.84, respectively), suggesting that individuals with narcolepsy have increased sleepiness immediately after lights-off. These analyses demonstrate that novel EEG-based biomarkers can augment SOL as predictors of sleepiness and its response to treatment and provide a novel framework for the analysis of wake EEG in hypersomnia disorders.

HLA and KIR genetic association and NK cells in anti-NMDAR encephalitis.

Introduction: Genetic predisposition to autoimmune encephalitis with antibodies against N-methyl-D-aspartate receptor (NMDAR) is poorly understood. Given the diversity of associated environmental factors (tumors, infections), we hypothesized that human leukocyte antigen (HLA) and killer-cell immunoglobulin-like receptors (KIR), two extremely polymorphic gene complexes key to the immune system, might be relevant for the genetic predisposition to anti-NMDAR encephalitis. Notably, KIR are chiefly expressed by Natural Killer (NK) cells, recognize distinct HLA class I allotypes and play a major role in anti-tumor and anti-infection responses. Methods: We conducted a Genome Wide Association Study (GWAS) with subsequent control-matching using Principal Component Analysis (PCA) and HLA imputation, in a multi-ethnic cohort of anti-NMDAR encephalitis (n=479); KIR and HLA were further sequenced in a large subsample (n=323). PCA-controlled logistic regression was then conducted for carrier frequencies (HLA and KIR) and copy number variation (KIR). HLA-KIR interaction associations were also modeled. Additionally, single cell sequencing was conducted in peripheral blood mononuclear cells from 16 cases and 16 controls, NK cells were sorted and phenotyped. Results: Anti-NMDAR encephalitis showed a weak HLA association with DRB1*01:01~DQA1*01:01~DQB1*05:01 (OR=1.57, 1.51, 1.45; respectively), and DRB1*11:01 (OR=1.60); these effects were stronger in European descendants and in patients without an underlying ovarian teratoma. More interestingly, we found increased copy number variation of KIR2DL5B (OR=1.72), principally due to an overrepresentation of KIR2DL5B*00201. Further, we identified two allele associations in framework genes, KIR2DL4*00103 (25.4% vs. 12.5% in controls, OR=1.98) and KIR3DL3*00302 (5.3% vs. 1.3%, OR=4.44). Notably, the ligands of these KIR2DL4 and KIR3DL3, respectively, HLA-G and HHLA2, are known to act as immune checkpoint with immunosuppressive functions. However, we did not find differences in specific KIR-HLA ligand interactions or HLA-G polymorphisms between cases and controls. Similarly, gene expression of CD56dim or CD56bright NK cells did not differ between cases and controls. Discussion: Our observations for the first time suggest that the HLA-KIR axis might be involved in anti-NMDAR encephalitis. While the genetic risk conferred by the identified polymorphisms appears small, a role of this axis in the pathophysiology of this disease appears highly plausible and should be analyzed in future studies.

HLA-DR3 ~ DQ2 associates with sensory neuropathy in paraneoplastic neurological syndromes with Hu antibodies.

OBJECTIVES: To investigate the association between human leukocyte antigen (HLA) and paraneoplastic neurological syndromes (PNS) with Hu antibodies, and potential specificities according to clinical presentation and cancer status. METHODS: HLA genotypes at four-digit resolution were imputed from available genome-wide association data. Allele carrier frequencies were compared between patients (whole cohort, n = 100, and according to clinical presentation and cancer status) and matched healthy controls (n = 508) using logistic regression controlled by the three main principal components. RESULTS: The clinical presentation of 100 anti-Hu patients involved the central nervous system (28, 28%), the peripheral nervous system (36, 36%) or both combined (36, 36%). Cancer diagnosis was certain in 75 (75%). HLA association analyses revealed that anti-Hu PNS patients were more frequently carriers of DQA1*05:01 (39% vs. 19%, OR = 2.8 [1.74-4.49]), DQB1*02:01 (39% vs. 18%, OR = 2.88 [1.79-4.64]) and DRB1*03:01 (41% vs. 19%, OR = 2.92 [1.80-4.73]) than healthy controls. Remarkably, such DR3 ~ DQ2 association was absent in patients with pure central involvement, but more specific to those manifesting with peripheral involvement: DQA1*05:01 (OR = 3.12 [1.48-6.60]), DQB1*02:01 (OR = 3.35 [1.57-7.15]) and DRB1*03:01 (OR = 3.62 [1.64-7.97]); being even stronger in cases with sensory neuropathy, DQA1*05:01 (OR = 4.41 [1.89-10.33]), DQB1*02:01 (OR = 4.85 [2.04-11.53]) and DRB1*03:01 (OR = 5.79 [2.28-14.74]). Similarly, DR3 ~ DQ2 association was only observed in patients with cancer. DISCUSSION: Patients with anti-Hu PNS show different HLA profiles according to clinical presentation and, probably, cancer status, suggesting pathophysiological differences.

From sleep patterns to heart rhythm: Predicting atrial fibrillation from overnight polysomnograms.

BACKGROUND: Atrial fibrillation (AF) is often asymptomatic and thus under-observed. Given the high risks of stroke and heart failure among patients with AF, early prediction and effective management are crucial. Given the prevalence of obstructive sleep apnea among AF patients, electrocardiogram (ECG) analysis from polysomnography (PSG) offers a unique opportunity for early AF prediction. Our aim is to identify individuals at high risk of AF development from single­lead ECGs during standard PSG. METHODS: We analyzed 18,782 single­lead ECG recordings from 13,609 subjects undergoing PSG at the Massachusetts General Hospital sleep laboratory. AF presence was identified using ICD-9/10 codes. The dataset included 15,913 recordings without AF history and 2054 recordings from patients diagnosed with AF between one month to fifteen years post-PSG. Data were partitioned into training, validation, and test cohorts ensuring that individual patients remained exclusive to each cohort. The test set was held out during the training process. We employed two different methods for feature extraction to build a final model for AF prediction: Extraction of hand-crafted ECG features and a deep learning method. For extraction of ECG-hand-crafted features, recordings were split into 30-s windows, and those with a signal quality index (SQI) below 0.95 were discarded. From each remaining window, 150 features were extracted from the time, frequency, time-frequency domains, and phase-space reconstructions of the ECG. A compilation of 12 statistical features summarized these window-specific features per recording, resulting in 1800 features (12 × 150). A pre-trained deep neural network from the PhysioNet Challenge 2021 was updated using transfer learning to discriminate recordings with and without AF. The model processed PSG ECGs in 16-s windows to generate AF probabilities, from which 13 statistical features were extracted. Combining 1800 features from feature extraction with 13 from the deep learning model, we performed a feature selection and subsequently trained a shallow neural network to predict future AF and evaluated its performance on the test cohort. RESULTS: On the test set, our model exhibited sensitivity, specificity, and precision of 0.67, 0.81, and 0.3, respectively, for AF prediction. Survival analysis revealed a hazard ratio of 8.36 (p-value: 1.93 × 10-52) for AF outcomes using the log-rank test. CONCLUSIONS: Our proposed ECG analysis method, utilizing overnight PSG data, shows promise in AF prediction despite modest precision, suggesting false positives. This approach could enable low-cost screening and proactive treatment for high-risk patients. Refinements, including additional physiological parameters, may reduce false positives, enhancing clinical utility and accuracy.

From Sleep Patterns to Heart Rhythms: Predicting Atrial Fibrillation from Overnight Polysomnograms.

Background: Atrial fibrillation (AF) is often asymptomatic and thus under-observed. Given the high risks of stroke and heart failure among patients with AF, early prediction and effective management are crucial. Importantly, obstructive sleep apnea is highly prevalent among AF patients (60-90%); therefore, electrocardiogram (ECG) analysis from polysomnography (PSG), a standard diagnostic tool for subjects with suspected sleep apnea, presents a unique opportunity for the early prediction of AF. Our goal is to identify individuals at a high risk of developing AF in the future from a single-lead ECG recorded during standard PSGs. Methods: We analyzed 18,782 single-lead ECG recordings from 13,609 subjects at Massachusetts General Hospital, identifying AF presence using ICD-9/10 codes in medical records. Our dataset comprises 15,913 recordings without a medical record for AF and 2,056 recordings from patients who were first diagnosed with AF between 1 day to 15 years after the PSG recording. The PSG data were partitioned into training, validation, and test cohorts. In the first phase, a signal quality index (SQI) was calculated in 30-second windows and those with SQI < 0.95 were removed. From each remaining window, 150 hand-crafted features were extracted from time, frequency, time-frequency domains, and phase-space reconstructions of the ECG. A compilation of 12 statistical features summarized these window-specific features per recording, resulting in 1,800 features. We then updated a pre-trained deep neural network and data from the PhysioNet Challenge 2021 using transfer-learning to discriminate between recordings with and without AF using the same Challenge data. The model was applied to the PSG ECGs in 16-second windows to generate the probability of AF for each window. From the resultant probability sequence, 13 statistical features were extracted. Subsequently, we trained a shallow neural network to predict future AF using the extracted ECG and probability features. Results: On the test set, our model demonstrated a sensitivity of 0.67, specificity of 0.81, and precision of 0.3 for predicting AF. Further, survival analysis for AF outcomes, using the log-rank test, revealed a hazard ratio of 8.36 (p-value of 1.93 × 10 -52 ). Conclusions: Our proposed ECG analysis method, utilizing overnight PSG data, shows promise in AF prediction despite a modest precision indicating the presence of false positive cases. This approach could potentially enable low-cost screening and proactive treatment for high-risk patients. Ongoing refinement, such as integrating additional physiological parameters could significantly reduce false positives, enhancing its clinical utility and accuracy.

Objectively measured daytime sleepiness predicts weight change among adults: Findings from the Wisconsin Sleep Cohort Study.

OBJECTIVE: Body mass index (BMI) trajectories are associated with night-time sleep, but it is not clear how they relate to daytime sleepiness in population data. This study aimed to examine longitudinal associations between levels and changes in daytime sleepiness and BMI trajectories among men and women. METHODS: We estimated growth curve models among 827 participants in the Wisconsin Sleep Cohort Study (mean [sd] age = 55.2 [8.0] years at baseline). The outcome variable was BMI (kg/m2) and the key predictor was daytime sleepiness measured by Multiple Sleep Latency Test (MSLT) scores. Covariates included demographics, health behaviors, retirement status, stimulant use, and depressive symptoms. In sensitivity analyses, we evaluated the potential effects of cardiovascular disease, shift work status, and sleep apnea on the robustness of sleepiness and BMI associations. RESULTS: At the between-person level, men who were sleepier had higher BMI levels. At the within-person level, age moderated the positive association between sleepiness and BMI among women. Specifically, young women who became sleepier over time gained more BMI than older women with comparable increases in sleepiness. Furthermore, while BMI tended to increase with age among women, BMI trajectories were steeper among sleepy women than among well-rested women, who experienced less increase in BMI over time. CONCLUSION: The study suggested that levels and changes in daytime sleepiness as objectively measured by MSLT scores are associated with body mass among adults.

Comparative Study of Paraneoplastic and Nonparaneoplastic Autoimmune Encephalitis With GABABR Antibodies.

BACKGROUND AND OBJECTIVES: While patients with paraneoplastic autoimmune encephalitis (AE) with gamma-aminobutyric-acid B receptor antibodies (GABABR-AE) have poor functional outcomes and high mortality, the prognosis of nonparaneoplastic cases has not been well studied. METHODS: Patients with GABABR-AE from the French and the Dutch Paraneoplastic Neurologic Syndromes Reference Centers databases were retrospectively included and their data collected; the neurologic outcomes of paraneoplastic and nonparaneoplastic cases were compared. Immunoglobulin G (IgG) isotyping and human leukocyte antigen (HLA) genotyping were performed in patients with available samples. RESULTS: A total of 111 patients (44/111 [40%] women) were enrolled, including 84 of 111 (76%) paraneoplastic and 18 of 111 (16%) nonparaneoplastic cases (cancer status was undetermined for 9 patients). Patients presented with seizures (88/111 [79%]), cognitive impairment (54/111 [49%]), and/or behavioral disorders (34/111 [31%]), and 54 of 111 (50%) were admitted in intensive care unit (ICU). Nonparaneoplastic patients were significantly younger (median age 54 years [range 19-88] vs 67 years [range 50-85] for paraneoplastic cases, p < 0.001) and showed a different demographic distribution. Nonparaneoplastic patients more often had CSF pleocytosis (17/17 [100%] vs 58/78 [74%], p = 0.02), were almost never associated with KTCD16-abs (1/16 [6%] vs 61/70 [87%], p < 0.001), and were more frequently treated with second-line immunotherapy (11/18 [61%] vs 18/82 [22%], p = 0.003). However, no difference of IgG subclass or HLA association was observed, although sample size was small (10 and 26 patients, respectively). After treatment, neurologic outcome was favorable (mRS ≤2) for 13 of 16 (81%) nonparaneoplastic and 37 of 84 (48%) paraneoplastic cases (p = 0.03), while 3 of 18 (17%) and 42 of 83 (51%) patients had died at last follow-up (p = 0.008), respectively. Neurologic outcome no longer differed after adjustment for confounding factors but seemed to be negatively associated with increased age and ICU admission. A better survival was associated with nonparaneoplastic cases, a younger age, and the use of immunosuppressive drugs. DISCUSSION: Nonparaneoplastic GABABR-AE involved younger patients without associated KCTD16-abs and carried better neurologic and vital prognoses than paraneoplastic GABABR-AE, which might be due to a more intensive treatment strategy. A better understanding of immunologic mechanisms underlying both forms is needed.

A Deep Transfer Learning Approach for Sleep Stage Classification and Sleep Apnea Detection Using Wrist-Worn Consumer Sleep Technologies.

Obstructive sleep apnea (OSA) is a common, underdiagnosed sleep-related breathing disorder with serious health implications Objective - We propose a deep transfer learning approach for sleep stage classification and sleep apnea (SA) detection using wrist-worn consumer sleep technologies (CST). Methods - Our model is based on a deep convolutional neural network (DNN) utilizing accelerometers and photo-plethysmography signals from nocturnal recordings. The DNN was trained and tested on internal datasets that include raw data from clinical and wrist-worn devices; external validation was performed on a hold-out test dataset containing raw data from a wrist-worn CST. Results - Training on clinical data improves performance significantly, and feature enrichment through a sleep stage stream gives only minor improvements. Raw data input outperforms feature-based input in CST datasets. The system generalizes well but performs slightly worse on wearable device data compared to clinical data. However, it excels in detecting events during REM sleep and is associated with arousal and oxygen desaturation. We found; cases that were significantly underestimated were characterized by fewer of such event associations. Conclusion - This study showcases the potential of using CSTs as alternate screening solution for undiagnosed cases of OSA. Significance - This work is significant for its development of a deep transfer learning approach using wrist-worn consumer sleep technologies, offering comprehensive validation for data utilization, and learning techniques, ultimately improving sleep apnea detection across diverse devices.

HLA-DQB1*05 subtypes and not DRB1*10:01 mediates risk in anti-IgLON5 disease.

Anti-IgLON5 disease is a rare and likely underdiagnosed subtype of autoimmune encephalitis. The disease displays a heterogeneous phenotype that includes sleep, movement and bulbar-associated dysfunction. The presence of IgLON5-antibodies in CSF/serum, together with a strong association with HLA-DRB1*10:01∼DQB1*05:01, supports an autoimmune basis. In this study, a multicentric human leukocyte antigen (HLA) study of 87 anti-IgLON5 patients revealed a stronger association with HLA-DQ than HLA-DR. Specifically, we identified a predisposing rank-wise association with HLA-DQA1*01:05∼DQB1*05:01, HLA-DQA1*01:01∼DQB1*05:01 and HLA-DQA1*01:04∼DQB1*05:03 in 85% of patients. HLA sequences and binding cores for these three DQ heterodimers were similar, unlike those of linked DRB1 alleles, supporting a causal link to HLA-DQ. This association was further reflected in an increasingly later age of onset across each genotype group, with a delay of up to 11 years, while HLA-DQ-dosage dependent effects were also suggested by reduced risk in the presence of non-predisposing DQ1 alleles. The functional relevance of the observed HLA-DQ molecules was studied with competition binding assays. These proof-of-concept experiments revealed preferential binding of IgLON5 in a post-translationally modified, but not native, state to all three risk-associated HLA-DQ receptors. Further, a deamidated peptide from the Ig2-domain of IgLON5 activated T cells in two patients, compared with one control carrying HLA-DQA1*01:05∼DQB1*05:01. Taken together, these data support a HLA-DQ-mediated T-cell response to IgLON5 as a potentially key step in the initiation of autoimmunity in this disease.

Strengths, weaknesses, opportunities, and threats of using AI-enabled technology in sleep medicine: a commentary.

Over the past few years, artificial intelligence (AI) has emerged as a powerful tool used to efficiently automate several tasks across multiple domains. Sleep medicine is perfectly positioned to leverage this tool due to the wealth of physiological signals obtained through sleep studies or sleep tracking devices and abundance of accessible clinical data through electronic medical records. However, caution must be applied when utilizing AI, due to intrinsic challenges associated with novel technology. The Artificial Intelligence in Sleep Medicine Committee of the American Academy of Sleep Medicine reviews advancements in AI within the sleep medicine field. In this article, the Artificial Intelligence in Sleep Medicine committee members provide a commentary on the scope of AI technology in sleep medicine. The commentary identifies 3 pivotal areas in sleep medicine that can benefit from AI technologies: clinical care, lifestyle management, and population health management. This article provides a detailed analysis of the strengths, weaknesses, opportunities, and threats associated with using AI-enabled technologies in each pivotal area. Finally, the article broadly reviews barriers and challenges associated with using AI-enabled technologies and offers possible solutions. CITATION: Bandyopadhyay A, Oks M, Sun H, et al. Strengths, weaknesses, opportunities, and threats of using AI-enabled technology in sleep medicine: a commentary. J Clin Sleep Med. 2024;20(7):1183-1191.

Fatigued but not sleepy? An empirical investigation of the differentiation between fatigue and sleepiness in sleep disorder patients in a cross-sectional study.

OBJECTIVE: Sleepiness and fatigue are common complaints among individuals with sleep disorders. The two concepts are often used interchangeably, causing difficulty with differential diagnosis and treatment decisions. The current study investigated sleep disorder patients to determine which factors best differentiated sleepiness from fatigue. METHODS: The study used a subset of participants from a multi-site study (n = 606), using a cross-sectional study design. We selected 60 variables associated with either sleepiness or fatigue, including demographic, mental health, and lifestyle factors, medical history, sleep questionnaires, rest-activity rhythms (actigraphy), polysomnographic (PSG) variables, and sleep diaries. Fatigue was measured with the Fatigue Severity Scale and sleepiness was measured with the Epworth Sleepiness Scale. A Random Forest machine learning approach was utilized for analysis. RESULTS: Participants' average age was 47.5 years (SD 14.0), 54.6% female, and the most common sleep disorder diagnosis was obstructive sleep apnea (67.4%). Sleepiness and fatigue were moderately correlated (r = 0.334). The model for fatigue (explained variance 49.5%) indicated depression was the strongest predictor (relative explained variance 42.7%), followed by insomnia severity (12.3%). The model for sleepiness (explained variance 17.9%), indicated insomnia symptoms was the strongest predictor (relative explained variance 17.6%). A post hoc receiver operating characteristic analysis indicated depression could be used to discriminate fatigue (AUC = 0.856) but not sleepiness (AUC = 0.643). CONCLUSIONS: The moderate correlation between fatigue and sleepiness supports previous literature that the two concepts are overlapping yet distinct. Importantly, depression played a more prominent role in characterizing fatigue than sleepiness, suggesting depression could be used to differentiate the two concepts.

Stability and Volatility of Human Rest-Activity Rhythms: Insights from Very Long Actograms (VLAs).

Importance: Wrist-worn activity monitors provide biomarkers of health by non-obtrusively measuring the timing and amount of rest and physical activity (rest-activity rhythms, RARs). The morphology and robustness of RARs vary by age, gender, and sociodemographic factors, and are perturbed in various chronic illnesses. However, these are cross-sectionally derived associations from recordings lasting 4-10 days, providing little insights into how RARs vary with time. Objective: To describe how RAR parameters can vary or evolve with time (~months). Design Setting and Participants: 48 very long actograms ("VLAs", ≥90 days in duration) were identified from subjects enrolled in the STAGES (Stanford Technology, Analytics and Genomics in Sleep) study, a prospective cross-sectional, multi-site assessment of individuals > 13 years of age that required diagnostic polysomnography to address a sleep complaint. A single 3-year long VLA (author GD) is also described. Exposures/Intervention: None planned. Main Outcomes and Measures: For each VLA, we assessed the following parameters in 14-day windows: circadian/ultradian spectrum, pseudo-F statistic ("F"), cosinor amplitude, intradaily variability, interdaily stability, acrophase and estimates of "sleep" and non-wearing. Results: Included STAGES subjects (n = 48, 30 female) had a median age of 51, BMI of 29.4kg/m2, Epworth Sleepiness Scale score (ESS) of 10/24 and a median recording duration of 120 days. We observed marked within-subject undulations in all six RAR parameters, with many subjects displaying ultradian rhythms of activity that waxed and waned in intensity. When appraised at the group level (nomothetic), averaged RAR parameters remained remarkably stable over a ~4 month recording period. Cohort-level deficits in average RAR robustness associated with unemployment or high BMI (>29.4) also remained stable over time. Conclusions and Relevance: Through an exemplary set of months-long wrist actigraphy recordings, this study quantitatively depicts the longitudinal stability and dynamic range of human rest-activity rhythms. We propose that continuous and long-term actigraphy may have broad potential as a holistic, transdiagnostic and ecologically valid monitoring biomarker of changes in chronobiological health. Prospective recordings from willing subjects will be necessary to precisely define contexts of use.

Role of Daytime Continuous Polysomnography in the Diagnosis of Pediatric Narcolepsy Type 1.

BACKGROUND AND OBJECTIVES: Narcolepsy type 1 (NT1) is still largely underdiagnosed or diagnosed too late in children. Difficulties in obtaining rapid and reliable diagnostic evaluations of the condition in clinical practice partially explain this problem. Predictors of NT1 include cataplexy and sleep-onset REM periods (SOREMPs), documented during nocturnal polysomnography (N-PSG) or through the multiple sleep latency test (MSLT), although low CSF hypocretin-1 (CSF hcrt-1) is the definitive biological disease marker. Obtaining reliable MSLT results is not always feasible in children; therefore, this study aimed to validate daytime continuous polysomnography (D-PSG) as an alternative diagnostic tool. METHODS: Two hundred consecutive patients aged younger than 18 years (112 with NT1; 25 with other hypersomnias, including narcolepsy type 2 and idiopathic hypersomnia; and 63 with subjective excessive daytime sleepiness) were randomly split into 2 groups: group 1 (n = 133) for the identification of diagnostic markers and group 2 (n = 67) for the validation of the detected markers. The D-PSG data collected included the number of spontaneous naps, total sleep time, and the number of daytime SOREMPs (d-SOREMP). D-PSG data were tested against CSF hcrt-1 deficiency (NT1 diagnosis) as the gold standard using receiver operating characteristic (ROC) curve analysis in group 1. ROC diagnostic performances of single and combined D-PSG parameters were tested in group 1 and validated in group 2. RESULTS: In group 1, the areas under the ROC curve (AUCs) were 0.91 (95% CI 0.86-0.96) for d-SOREMPs, 0.81 (95% CI 0.74-0.89) for the number of spontaneous naps, and 0.70 (95% CI 0.60-0.79) for total sleep time. A d-SOREMP count ≥1 (sensitivity of 95% and specificity of 72%), coupled with a diurnal total sleep time above 60 minutes (sensitivity of 89% and specificity of 91%), identified NT1 in group 1 with high reliability (area under the ROC curve of 0.93, 95% CI 0.88-0.97). These results were confirmed in the validation group with an AUC of 0.88 (95% CI 0.79-0.97). DISCUSSION: D-PSG recording is an easily performed, cost-effective, and reliable tool for identifying NT1 in children. Further studies should confirm its validity with home D-PSG monitoring. These alternative procedures could be used to confirm NT1 diagnosis and curtail diagnostic delay.