Precision OSA Therapy Based on Phenotypes and Endotypes
ConditionObstructive Sleep Apnea
Trial statusNot yet recruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
Age18-80
SponsorNanjing Medical University
About this trial
Analyzing the phenotypic and endotypic characteristics of Sleep Apnea, along with DISE obstruction situations, is crucial for precise diagnosis and treatment. In this study, we aim to construct and apply a multidimensional predictive model based on four aspects: basic physiological characteristics of OSA, clinical phenotypes, mechanistic endotypes, and DISE obstruction levels. The study will begin by categorizing the clinical phenotypes; subsequently, it will quantify endotypic indicators based on PSG signal information and construct the PALM scale for Chinese individuals. Following this, a comprehensive clinical profile and a treatment efficacy prediction model for OSA patients will be built based on the results from the aforementioned multidimensional data.
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
Patients aged between 18 and 80 years.
Diagnosed with Obstructive Sleep Apnea (OSA)(apnea-hypopnea index≥5/h).
First-time diagnosis, with no previous surgical interventions or CPAP treatment for OSA.
Ability and willingness to provide informed consent for participation in the study.
Disqualifiers
History of severe stroke or cerebral hemorrhage, or presence of neurological or psychiatric conditions that could affect study results.
Presence of active malignancies or other severe underlying diseases, such as severe liver or kidney dysfunction. Diagnosed with diabetes or other significant vascular diseases.
Presence of severe chronic obstructive pulmonary disease (COPD), severe asthma, severe pulmonary hypertension, or heart failure caused by any condition.
Pregnancy or having other conditions that make participation in this study unsuitable.
Trial population
This study will involve participants diagnosed with OSA who are not currently undergoing any active treatment for the condition. The study aims to explore the detailed phenotypic and endotypic characteristics of these patients to better understand OSA dynamics and develop predictive models for individualized treatment approaches.
Trial design
Design model
Cohort
Time perspective
Prospective
Treatments tested in this trial
Clinical and Endotypic Assessmen
Other interventionThis observational study involves a detailed clinical and endotypic assessment of patients diagnosed with Obstructive Sleep Apnea (OSA). Assessments include polysomnography (PSG) to measure sleep patterns and disturbances, drug-induced sleep endoscopy (DISE) to evaluate upper airway obstruction, and various biomarker analyses to characterize endotypic traits. The study aims to collect comprehensive phenotypic and endotypic data to develop predictive models for OSA patient characterization and management.
Treatment groups
200 Participants
are divided into 1 treatment groupGroup A: OSA Comprehensive Assessment Group1 intervention
Trial outcomes
Primary outcomes
1
Phenotype and Endotype Classification
This outcome measure will evaluate the new phenotypic and endotypic features in OSA patients by integrating various clinical symptoms, traditional and novel PSG metrics, upper airway imaging indicators, diaphragm morphology and function parameters, and DISE results. A comprehensive classification system will be developed by combining these data points to classify OSA patients into distinct clinical phenotypes and endotypes. This classification aims to provide insights into the underlying pathophysiology of OSA and to better understand patient-specific characteristics for personalized treatment plans.
Time frame
12 months post-enrollment.2
Multidimensional Predictive Model
This outcome measure will assess the effectiveness of a multidimensional predictive model constructed using clinical phenotype, endotype, novel biomarkers, and DISE results. We used six commonly employed supervised machine learning algorithms: Random Forest, XGBoost, Support Vector Classifier (SVC), Logistic Regression, Multi-layer Perceptron (MLP), and Stacking Regression to classify OSA patients based on their survival status. The Stacking Regression model was designed by combining the outputs of Random Forest, XGBoost, and Support Vector Regression. The best-performing model will be selected to compute the final prediction, providing a powerful tool to predict the treatment response and clinical outcomes for OSA patients.
Time frame
End of the study, expected 24 months after enrollment.Sponsors and contacts
Click on the lead sponsor to view all of their trials.
Source ClinicalTrials.gov
This trial is not recruiting at the moment. You can still explore other options: