Screening for obstructive sleep apnea in patients with cancer - a machine learning approach.

Background: Obstructive sleep apnea (OSA) is a highly prevalent sleep disorder associated with daytime sleepiness, fatigue, and increased all-cause mortality risk in patients with cancer. Existing screening tools for OSA do not account for the interaction of cancer-related features that may increase OSA risk. Study Design and Methods: This is a retrospective study of patients with cancer at a single tertiary cancer institution who underwent a home sleep apnea test (HSAT) to evaluate for OSA. Unsupervised machine learning (ML) was used to reduce the dimensions and extract significant features associated with OSA. ML classifiers were applied to principal components and model hyperparameters were optimized using k-fold cross-validation. Training models for OSA were subsequently tested and compared with the STOP-Bang questionnaire on a prospective unseen test set of patients who underwent an HSAT. Results: From a training dataset of 249 patients, kernel principal component analysis (PCA) extracted eight components through dimension reduction to explain the maximum variance with OSA at 98%. Predictors of OSA were smoking, asthma, chronic kidney disease, STOP-Bang score, race, diabetes, radiation to head/neck/thorax (RT-HNT), type of cancer, and cancer metastases. Of the ML models, PCA + RF had the highest sensitivity (96.8%), specificity (92.3%), negative predictive value (92%), F1 score (0.93), and ROC-AUC score (0.88). The PCA + RF screening algorithm also performed better than the STOP-Bang questionnaire alone when tested on a prospective unseen test set. Conclusions: The PCA + RF ML model had the highest accuracy in screening for OSA in patients with cancer. History of RT-HNT, cancer metastases, and type of cancer were identified as cancer-related risk factors for OSA.

The need to promote sleep health in public health agendas across the globe.

Healthy sleep is essential for physical and mental health, and social wellbeing; however, across the globe, and particularly in developing countries, national public health agendas rarely consider sleep health. Sleep should be promoted as an essential pillar of health, equivalent to nutrition and physical activity. To improve sleep health across the globe, a focus on education and awareness, research, and targeted public health policies are needed. We recommend developing sleep health educational programmes and awareness campaigns; increasing, standardising, and centralising data on sleep quantity and quality in every country across the globe; and developing and implementing sleep health policies across sectors of society. Efforts are needed to ensure equity and inclusivity for all people, particularly those who are most socially and economically vulnerable, and historically excluded.

Dual inhibition of IDO1/TDO2 enhances anti-tumor immunity in platinum-resistant non-small cell lung cancer.

BACKGROUND: The impact of non-small cell lung cancer (NSCLC) metabolism on the immune microenvironment is not well understood within platinum resistance. We have identified crucial metabolic differences between cisplatin-resistant (CR) and cisplatin-sensitive (CS) NSCLC cells with elevated indoleamine 2,3-dioxygenase-1 (IDO1) activity in CR, recognized by increased kynurenine (KYN) production. METHODS: Co-culture, syngeneic, and humanize mice models were utilized. C57BL/6 mice were inoculated with either Lewis lung carcinoma mouse cells (LLC) or their platinum-resistant counterpart (LLC-CR) cells. Humanized mice were inoculated with either A (human CS cells) or ALC (human CR cells). Mice were treated with either IDO1 inhibitor or TDO2 (tryptophan 2,3-dioxygenase-2) inhibitor at 200 mg/kg P.O. once a day for 15 days; or with a new-in-class, IDO1/TDO2 dual inhibitor AT-0174 at 170 mg/kg P.O. once a day for 15 days with and without anti-PD1 antibody (10 mg/kg, every 3 days). Immune profiles and KYN and tryptophan (TRP) production were evaluated. RESULTS: CR tumors exhibited a more highly immunosuppressive environment that debilitated robust anti-tumor immune responses. IDO1-mediated KYN production from CR cells suppressed NKG2D on immune effector natural killer (NK) and CD8+ T cells and enhanced immunosuppressive populations of regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs). Importantly, while selective IDO1 inhibition attenuated CR tumor growth, it concomitantly upregulated the TDO2 enzyme. To overcome the compensatory induction of TDO2 activity, we employed the IDO1/TDO2 dual inhibitor, AT-0174. Dual inhibition of IDO1/TDO2 in CR mice suppressed tumor growth to a greater degree than IDO1 inhibition alone. Significant enhancement in NKG2D frequency on NK and CD8+ T cells and a reduction in Tregs and MDSCs were observed following AT-1074 treatment. PD-L1 (programmed death-ligand-1) expression was increased in CR cells; therefore, we assessed dual inhibition + PD1 (programmed cell death protein-1) blocking and report profound anti-tumor growth and improved immunity in CR tumors which in turn extended overall survival in mice. CONCLUSION: Our study reports the presence of platinum-resistant lung tumors that utilize both IDO1/TDO2 enzymes for survival, and to escape immune surveillance as a consequence of KYN metabolites. We also report early in vivo data in support of the potential therapeutic efficacy of the dual IDO1/TDO2 inhibitor AT-0174 as a part of immuno-therapeutic treatment that disrupts tumor metabolism and enhances anti-tumor immunity.

Facial and Intraoral Photographic Traits Related to Sleep Apnea in a Clinical Sample with Genetic Ancestry Analysis.

Rationale: Craniofacial and pharyngeal morphology influences risk for obstructive sleep apnea (OSA). Quantitative photography provides phenotypic information about these anatomical factors and is feasible in large samples. However, whether associations between morphology and OSA severity differ among populations is unknown. Objectives: The aim of this study was to examine this question in a large sample encompassing people from different ancestral backgrounds. Methods: Participants in SAGIC (Sleep Apnea Global Interdisciplinary Consortium) with genotyping data were included (N = 2,393). Associations between photography-based measures and OSA severity were assessed using linear regression, controlling for age, sex, body mass index, and genetic ancestry. Subgroups (on the basis of 1000 Genomes reference populations) were identified: European (EUR), East Asian, American, South Asian, and African (AFR). Interaction tests were used to assess if genetically determined ancestry group modified these relationships. Results: Cluster analysis of genetic ancestry proportions identified four ancestrally defined groups: East Asia (48.3%), EUR (33.6%), admixed (11.7%; 46% EUR, 27% Americas, and 22% AFR), and AFR (6.4%). Multiple anatomical traits were associated with more severe OSA independent of ancestry, including larger cervicomental angle (standardized ß [95% confidence interval (CI)] = 0.11 [0.06-0.16]; P < 0.001), mandibular width (standardized ß [95% CI] = 0.15 [0.10-0.20]; P < 0.001), and tongue thickness (standardized ß [95% CI] = 0.06 [0.02-0.10]; P = 0.001) and smaller airway width (standardized ß [95% CI] = -0.08 [-0.15 to -0.002]; P = 0.043). Other traits, including maxillary and mandibular depth angles and lower face height, demonstrated different associations with OSA severity on the basis of ancestrally defined subgroups. Conclusions: We confirm that multiple facial and intraoral photographic measurements are associated with OSA severity independent of ancestral background, whereas others differ in their associations among the ancestrally defined subgroups.

Novel neuroelectrophysiological age index associated with imaging features of brain aging and sleep disorders.

Sleep architecture and microstructures alter with aging and sleep disorder-led accelerated aging. We proposed a sleep EEG based brain age prediction model using convolutional neural networks. We then associated the estimated brain age index with brain structural aging features, sleep disorders and various sleep parameters. Our model also showed a higher BAI (predicted brain age minus chronological age) is associated with cortical thinning in various functional areas. We found a higher BAI for sleep disorder groups compared to healthy sleepers, as well as significant differences in the spectral pattern of EEG among different sleep disorders (lower power in slow and ϑ waves for sleep apnea vs. higher power in ß and σ for insomnia), suggesting sleep disorder-dependent pathomechanisms of aging. Our results demonstrate that the new EEG-BAI can be a biomarker reflecting brain health in normal and various sleep disorder subjects, and may be used to assess treatment efficacy.

Synaptic dysfunction connects autism spectrum disorder and sleep disturbances: A perspective from studies in model organisms.

Sleep disturbances (SD) accompany many neurodevelopmental disorders, suggesting SD is a transdiagnostic process that can account for behavioral deficits and influence underlying neuropathogenesis. Autism Spectrum Disorder (ASD) comprises a complex set of neurodevelopmental conditions characterized by challenges in social interaction, communication, and restricted, repetitive behaviors. Diagnosis of ASD is based primarily on behavioral criteria, and there are no drugs that target core symptoms. Among the co-occurring conditions associated with ASD, SD are one of the most prevalent. SD often arises before the onset of other ASD symptoms. Sleep interventions improve not only sleep but also daytime behaviors in children with ASD. Here, we examine sleep phenotypes in multiple model systems relevant to ASD, e.g., mice, zebrafish, fruit flies and worms. Given the functions of sleep in promoting brain connectivity, neural plasticity, emotional regulation and social behavior, all of which are of critical importance in ASD pathogenesis, we propose that synaptic dysfunction is a major mechanism that connects ASD and SD. Common molecular targets in this interplay that are involved in synaptic function might be a novel avenue for therapy of individuals with ASD experiencing SD. Such therapy would be expected to improve not only sleep but also other ASD symptoms.

Predicting brain age based on sleep EEG and DenseNet.

We proposed a sleep EEG-based brain age prediction model which showed higher accuracy than previous models. Six-channel EEG data were acquired for 6 hours sleep. We then converted the EEG data into 2D scalograms, which were subsequently inputted to DenseNet used to predict brain age. We then evaluated the association between brain aging acceleration and sleep disorders such as insomnia and OSA.The correlation between chronological age and expected brain age through the proposed brain age prediction model was 80% and the mean absolute error was 5.4 years. The proposed model revealed brain age increases in relation to the severity of sleep disorders.In this study, we demonstrate that the brain age estimated using the proposed model can be a biomarker that reflects changes in sleep and brain health due to various sleep disorders.Clinical Relevance-Proposed brain age index can be a single index that reflects the association of various sleep disorders and serve as a tool to diagnose individuals with sleep disorders.

Defining Extreme Phenotypes of OSA Across International Sleep Centers.

BACKGROUND: Extreme phenotypes of OSA have not been systematically defined. RESEARCH QUESTION: This study developed objective definitions of extreme phenotypes of OSA by using a multivariate approach. The utility of these definitions for identifying characteristics that confer predisposition toward or protection against OSA is shown in a new prospective sample. STUDY DESIGN AND METHODS: In a large international sample, race-specific liability scores were calculated from a weighted logistic regression that included age, sex, and BMI. Extreme cases were defined as individuals with an apnea-hypopnea index (AHI) ≥ 30 events/hour but low likelihood of OSA based on age, sex, and BMI (liability scores > 90th percentile). Similarly, extreme controls were individuals with an AHI < 5 events/hour but high likelihood of OSA (liability scores < 10th percentile). Definitions were applied to a prospective sample from the Sleep Apnea Global Interdisciplinary Consortium, and differences in photography-based craniofacial and intraoral phenotypes were evaluated. RESULTS: This study included retrospective data from 81,338 individuals. A total of 4,168 extreme cases and 1,432 extreme controls were identified by using liability scores. Extreme cases were younger (43.1 ± 14.7 years), overweight (28.6 ± 6.8 kg/m2), and predominantly female (71.1%). Extreme controls were older (53.8 ± 14.1 years), obese (34.0 ± 8.1 kg/m2), and predominantly male (65.8%). These objective definitions identified 29 extreme cases and 87 extreme controls among 1,424 Sleep Apnea Global Interdisciplinary Consortium participants with photography-based phenotyping. Comparisons suggest that a greater cervicomental angle increases risk for OSA in the absence of clinical risk factors, and smaller facial widths are protective in the presence of clinical risk factors. INTERPRETATION: This objective definition can be applied in sleep centers throughout the world to consistently define OSA extreme phenotypes for future studies on genetic, anatomic, and physiologic pathways to OSA.

Reinventing polysomnography in the age of precision medicine.

For almost 50 years, sleep laboratories around the world have been collecting massive amounts of polysomnographic (PSG) physiological data to diagnose sleep disorders, the majority of which are not utilized in the clinical setting. Only a small fraction of the information available within these signals is utilized to generate indices. For example, the apnea-hypopnea index (AHI) remains the primary tool for diagnostic and therapeutic decision-making for obstructive sleep apnea (OSA) despite repeated studies showing it to be inadequate in predicting clinical consequences. Today, there are many novel approaches to PSG signals, making it possible to extract more complex metrics and analyses that are potentially more clinically relevant for individual patients. However, the pathway to implement novel PSG metrics/analyses into routine clinical practice is unclear. Our goal with this review is to highlight some of the novel PSG metrics/analyses that are becoming available. We suggest that stronger academic-industry relationships would facilitate the development of state-of-the-art clinical research to establish the value of novel PSG metrics/analyses in clinical sleep medicine. Collectively, as a sleep community, it is time to reinvent how we utilize the polysomnography to move us towards Precision Sleep Medicine.

High-throughput sleep phenotyping produces robust and heritable traits in Diversity Outbred mice and their founder strains.

STUDY OBJECTIVES: This study describes high-throughput phenotyping strategies for sleep and circadian behavior in mice, including examinations of robustness, reliability, and heritability among Diversity Outbred (DO) mice and their eight founder strains. METHODS: We performed high-throughput sleep and circadian phenotyping in male mice from the DO population (n = 338) and their eight founder strains: A/J (n = 6), C57BL/6J (n = 14), 129S1/SvlmJ (n = 6), NOD/LtJ (n = 6), NZO/H1LtJ (n = 6), CAST/EiJ (n = 8), PWK/PhJ (n = 8), and WSB/EiJ (n = 6). Using infrared beam break systems, we defined sleep as at least 40 s of continuous inactivity and quantified sleep-wake amounts and bout characteristics. We developed assays to measure sleep latency in a new environment and during a modified Murine Multiple Sleep Latency Test, and estimated circadian period from wheel-running experiments. For each trait, broad-sense heritability (proportion of variability explained by all genetic factors) was derived in founder strains, while narrow-sense heritability (proportion of variability explained by additive genetic effects) was calculated in DO mice. RESULTS: Phenotypes were robust to different inactivity durations to define sleep. Differences across founder strains and moderate/high broad-sense heritability were observed for most traits. There was large phenotypic variability among DO mice, and phenotypes were reliable, although estimates of heritability were lower than in founder mice. This likely reflects important nonadditive genetic effects. CONCLUSIONS: A high-throughput phenotyping strategy in mice, based primarily on monitoring of activity patterns, provides reliable and heritable estimates of sleep and circadian traits. This approach is suitable for discovery analyses in DO mice, where genetic factors explain some proportion of phenotypic variation.

Blood pressure monitoring in sleep: time to wake up.

Hypertension is a highly common condition with well-established adverse consequences. Ambulatory blood pressure monitoring has repeatedly been shown to better predict cardiovascular outcomes and mortality, compared to single office visit blood pressure. Non-dipping of sleep-time blood pressure is an independent marker for increased cardiovascular risk. We review blood pressure variability and the challenges of blood pressure monitoring during sleep. Although pathological sleep such as obstructive sleep apnea has been associated with non-dipping of sleep-time blood pressure, blood pressure is not routinely measured during sleep due to lack of unobtrusive blood pressure monitoring technology. Second, we review existing noninvasive continuous blood pressure monitoring technologies. Lastly, we propose including sleep-time blood pressure monitoring during sleep studies and including sleep studies in patients undergoing ambulatory blood pressure monitoring.

Age attenuates the transcriptional changes that occur with sleep in the medial prefrontal cortex.

Sleep abnormalities are common with aging. Studies show that sleep plays important roles in brain functions, and loss of sleep is associated with increased risks for neurological diseases. Here, we used RNA sequencing to explore effects of age on transcriptome changes between sleep and sleep deprivation (SD) in medial prefrontal cortex and found that transcriptional changes with sleep are attenuated in old. In particular, old mice showed a 30% reduction in the number of genes significantly altered between sleep/wake and, in general, had smaller magnitudes of changes in differentially expressed genes compared to young mice. Gene ontology analysis revealed differential age effects on certain pathways. Compared to young mice, many of the wake-active functions were similarly induced by SD in old mice, whereas many of the sleep-active pathways were attenuated in old mice. We found similar magnitude of changes in synaptic homeostasis genes (Fos, Arc, and Bdnf) induced by SD, suggesting intact synaptic upscaling on the transcript level during extended wakefulness with aging. However, sleep-activated processes, such as DNA repair, synaptogenesis, and axon guidance, were sensitive to the effect of aging. Old mice expressed elevated levels of immune response genes when compared to young mice, and enrichment analysis using cell-type-specific markers indicated upregulation of microglia and oligodendrocyte genes in old mice. Moreover, gene sets of the two cell types showed age-specific sleep/wake regulation. Ultimately, this study enhances understanding of the transcriptional changes with sleep and aging, providing potential molecular targets for future studies of age-related sleep abnormalities and neurological disorders.

Symptom Subtypes of Obstructive Sleep Apnea Predict Incidence of Cardiovascular Outcomes.

Rationale: Symptom subtypes have been described in clinical and population samples of patients with obstructive sleep apnea (OSA). It is unclear whether these subtypes have different cardiovascular consequences.Objectives: To characterize OSA symptom subtypes and assess their association with prevalent and incident cardiovascular disease in the Sleep Heart Health Study.Methods: Data from 1,207 patients with OSA (apnea-hypopnea index ≥ 15 events/h) were used to evaluate the existence of symptom subtypes using latent class analysis. Associations between subtypes and prevalence of overall cardiovascular disease and its components (coronary heart disease, heart failure, and stroke) were assessed using logistic regression. Kaplan-Meier survival analysis and Cox proportional hazards models were used to evaluate whether subtypes were associated with incident events, including cardiovascular mortality.Measurements and Main Results: Four symptom subtypes were identified (disturbed sleep [12.2%], minimally symptomatic [32.6%], excessively sleepy [16.7%], and moderately sleepy [38.5%]), similar to prior studies. In adjusted models, although no significant associations with prevalent cardiovascular disease were found, the excessively sleepy subtype was associated with more than threefold increased risk of prevalent heart failure compared with each of the other subtypes. Symptom subtype was also associated with incident cardiovascular disease (P < 0.001), coronary heart disease (P = 0.015), and heart failure (P = 0.018), with the excessively sleepy again demonstrating increased risk (hazard ratios, 1.7-2.4) compared with other subtypes. When compared with individuals without OSA (apnea-hypopnea index < 5), significantly increased risk for prevalent and incident cardiovascular events was observed mostly for patients in the excessively sleepy subtype.Conclusions: OSA symptom subtypes are reproducible and associated with cardiovascular risk, providing important evidence of their clinical relevance.

Effect of cyclical intermittent hypoxia on Ad5CMVCre induced solitary lung cancer progression and spontaneous metastases in the KrasG12D+; p53fl/fl; myristolated p110fl/fl ROSA-gfp mouse.

BACKGROUND: Epidemiological data suggests that obstructive sleep apnea (OSA) is associated with increased cancer incidence and mortality. We investigate the effects of cyclical intermittent hypoxia (CIH), akin to the underlying pathophysiology of OSA, on lung cancer progression and metastatic profile in a mouse model. METHODS: Intrathoracic injection of Ad5CMVCre virus into a genetically engineered mouse (GEM) KrasG12D+/-; p53fl/fl; myristolated-p110αfl/fl-ROSA-gfp was utilized to induce a solitary lung cancer. Male mice were then exposed to either CIH or Sham for 40-41 days until harvest. To monitor malignant progression, serial micro CT scans with respiratory gating (no contrast) was performed. To detect spontaneous metastases in distant organs, H&E and immunohistochemistry were performed. RESULTS: Eighty-eight percent of injected Ad5CMVCre virus was recovered from left lung tissue, indicating reliable and accurate injections. Serial micro CT demonstrated that CIH increases primary lung tumor volume progression compared to Sham on days 33 (p = 0.004) and 40 (p<0.001) post-injection. In addition, CIH increases variability in tumor volume on day 19 (p<0.0001), day 26 (p<0.0001), day 33 (p = 0.025) and day 40 (p = 0.004). Finally, metastases are frequently detected in heart, mediastinal lymph nodes, and right lung using H&E and immunohistochemistry. CONCLUSIONS: Using a GEM mouse model of metastatic lung cancer, we report that male mice with solitary lung cancer have accelerated malignant progression and increased variability in tumor growth when exposed to cyclical intermittent hypoxia. Our results indicate that cyclical intermittent hypoxia is a pathogenic factor in non-small cell lung cancer that promotes the more rapid growth of developing tumors.

Home sleep apnea testing: comparison of manual and automated scoring across international sleep centers.

PURPOSE: To determine the agreement between the manual scoring of home sleep apnea tests (HSATs) by international sleep technologists and automated scoring systems. METHODS: Fifteen HSATs, previously recorded using a type 3 monitor, were saved in European Data Format. The studies were scored by nine experienced technologists from the sleep centers of the Sleep Apnea Global Interdisciplinary Consortium (SAGIC) using the locally available software. Each study was scored separately by human scorers using the nasal pressure (NP), flow derived from the NP signal (transformed NP), or respiratory inductive plethysmography (RIP) flow. The same procedure was followed using two automated scoring systems: Remlogic (RLG) and Noxturnal (NOX). RESULTS: The intra-class correlation coefficients (ICCs) of the apnea-hypopnea index (AHI) scoring using the NP, transformed NP, and RIP flow were 0.96 [95% CI 0.93-0.99], 0.98 [0.96-0.99], and 0.97 [0.95-0.99], respectively. Using the NP signal, the mean differences in AHI between the average of the manual scoring and the automated systems were - 0.9 ± 3.1/h (AHIRLG vs AHIMANUAL) and - 1.3 ± 2.6/h (AHINOX vs AHIMANUAL). Using the transformed NP, the mean differences in AHI were - 1.9 ± 3.3/h (AHIRLG vs AHIMANUAL) and 1.6 ± 3.0/h (AHINOX vs AHIMANUAL). Using the RIP flow, the mean differences in AHI were - 2.7 ± 4.5/h (AHIRLG vs AHIMANUAL) and 2.3 ± 3.4/h (AHINOX vs AHIMANUAL). CONCLUSIONS: There is very strong agreement in the scoring of the AHI for HSATs between the automated systems and experienced international technologists. Automated scoring of HSATs using commercially available software may be useful to standardize scoring in future endeavors involving international sleep centers.

Opportunities for utilizing polysomnography signals to characterize obstructive sleep apnea subtypes and severity.

BACKGROUND: Obstructive sleep apnea (OSA) is a heterogeneous sleep disorder with many pathophysiological pathways to disease. Currently, the diagnosis and classification of OSA is based on the apnea-hypopnea index, which poorly correlates to underlying pathology and clinical consequences. A large number of in-laboratory sleep studies are performed around the world every year, already collecting an enormous amount of physiological data within an individual. Clinically, we have not yet fully taken advantage of this data, but combined with existing analytical approaches, we have the potential to transform the way OSA is managed within an individual patient. Currently, respiratory signals are used to count apneas and hypopneas, but patterns such as inspiratory flow signals can be used to predict optimal OSA treatment. Electrocardiographic data can reveal arrhythmias, but patterns such as heart rate variability can also be used to detect and classify OSA. Electroencephalography is used to score sleep stages and arousals, but specific patterns such as the odds-ratio product can be used to classify how OSA patients responds differently to arousals. OBJECTIVE: In this review, we examine these and many other existing computer-aided polysomnography signal processing algorithms and how they can reflect an individual's manifestation of OSA. SIGNIFICANCE: Together with current technological advance, it is only a matter of time before advanced automatic signal processing and analysis is widely applied to precision medicine of OSA in the clinical setting.

Symptom-Based Subgroups of Koreans With Obstructive Sleep Apnea.

STUDY OBJECTIVES: Recently, the Icelandic Sleep Apnea Cohort (ISAC) identified three subgroups in patients with obstructive sleep apnea (OSA) recruited from the sleep clinic based on clinical symptoms and comorbidities: excessively sleepy, minimally symptomatic, and disturbed sleep. This study sought to determine whether the three OSA subgroups are applicable to a population-based cohort in Korea. METHODS: Study subjects are participants of an ongoing cohort study in Korea. Of the 2,918 participants, 422 new moderate to severe OSA cases (apneahypopnea index [AHI] ≥ 15 events/h) were diagnosed by home sleep studies. All participants completed a detailed questionnaire on sleep-related symptoms, comorbidities, medication, and other information. A latent class analysis was performed. RESULTS: When examining solutions for up to 10 clusters, the a priori three-cluster solution was the optimal clustering solution. The three-cluster solution demonstrated a subgroup presentation similar to the clusters identified in the ISAC. The minimally symptomatic subgroup was most prevalent (55.7%) in the Korean cohort. Among the three subgroups, there were no differences in mean AHI and body mass index; however, the disturbed sleep subgroup was older and had more females. CONCLUSIONS: Combined with the ISAC data, we suggest that the three-symptom cluster solution for patients with OSA may be more widely applicable, irrespective of ethnicity and study population.

P4 medicine approach to obstructive sleep apnoea.

P4 medicine is an evolving approach to personalized medicine. The four Ps offer a means to: Predict who will develop disease and co-morbidities; Prevent rather than react to disease; Personalize diagnosis and treatment; have patients Participate in their own care. P4 medicine is very applicable to obstructive sleep apnoea (OSA) because each OSA patient has a different pathway to disease and its consequences. OSA has both structural and physiological mechanisms with different clinical subgroups, different molecular profiles and different consequences. This may explain why there are different responses to alternative therapies, such as intraoral devices and hypoglossal nerve stimulation therapy. Currently, technology facilitates patients to participate in their own care from screening for OSA (snoring and apnoea apps) to monitoring response to therapy (sleep monitoring, blood pressure, oxygen saturation and heart rate) as well as monitoring their own continuous positive airway pressure (CPAP) compliance. We present a conceptual framework that provides the basis for a new, P4 medicine approach to OSA and should be considered more in depth: predict and prevent those at high risk for OSA and consequences, personalize the diagnosis and treatment of OSA and build in patient participation to manage OSA.