The Optimal One-Month Dosing of Lemborexant for Moderate Obstructive Sleep Apnea (OSA) Patients With Low Arousal Threshold

The goal of this clinical trial is to investigate the 1-month treatment effects of different dose of lemborexant in moderate OSA adult patients (18-65 years of age) with low arousal threshold. The main questions it aims to answer are: Primary outcome measure: Apnea/hypopnea index (AHI) Secondary outcome measure: 1. Polysomnography parameters * Mean and nadir oxygen saturation * Sleep efficiency * Wake after sleep onset (WASO) * Sleep latency * Rapid eye movement (REM) latency * Percentage of time spent in Non-rapid eye movement (NREM) stage 1-3 and REM stage * Arousal index 2. Oxford Sleep Resistance Test (OSLER) test 3. Epworth Sleepiness Scale (ESS) 4. Functional Outcome of Sleep Questionnaire-short version (FOSQ-10T) 5. Pittsburgh Sleep Quality Index (PSQI) 6. Actigraphy parameters * Total sleep time * Sleep efficiency * Wake after sleep onset (WASO) * Sleep latency 7. Sleep diary parameters (Appendix 5) * Total sleep time * Sleep efficiency * Wake after sleep onset (WASO) * Sleep latency * Sleep quality Researchers will compare placebo to see if there is a difference in AHI. Participants will * participate a total of 3 phases of study (3-crossover trial) * each phrase the participants will receive either lemborexant 5 mg, lemborexant 10 mg, or placebo orally per day for 30 days with 2-week washout (depend on intervention arm whether the participants receive which intervention sequence) * complete three overnight in-laboratory polysomnography (2-week washout) at day 30 of each phrase * monitor actigraphy during day 1 to 29 of intervention of all periods * record sleep diary during day 1 to 29 of intervention of all periods * complete the OSLER test in the morning of the three overnight test * complete questionnaires including: Epworth Sleepiness Scale (ESS), Functional Outcome of Sleep Questionnaire-short version (FOSQ-10T), Pittsburgh Sleep Quality Index (PSQI) at baseline (prior to intervention) and the night before overnight in-laboratory polysomnography