Obstructive Sleep Apnea (OSA) Emerging Research and Trials to Watch

Why OSA Research Is Accelerating Now

OSA affects an estimated 936 million adults globally, yet CPAP adherence rates hover around 50% at 12 months, leaving hundreds of millions without effective treatment. That compliance gap has driven two parallel research tracks: finding drugs that shrink the anatomical and physiologic drivers of airway collapse, and building devices or diagnostics that remove the friction of polysomnography or nightly CPAP use.

The January 2024 FDA approval of tirzepatide for OSA was not a coincidence. It followed the GLP-1 wave that reshaped obesity medicine and confirmed that weight loss of sufficient magnitude can modify OSA severity, not merely correlate with it. At the same time, neurostimulation, digital therapeutics, and loop-gain pharmacology have each reached Phase 3 or post-market stages that make 2024 through 2026 the most data-rich period in OSA research history.

The Weight-Loss Connection

A 2009 meta-analysis published in JAMA (N=342 patients across seven trials) established that a 10% reduction in body weight corresponds to approximately a 26% reduction in AHI. [1] That relationship anchored the hypothesis that a potent weight-loss drug could achieve clinically meaningful AHI reductions large enough to warrant an FDA label.

The CPAP-Adherence Problem

The American Academy of Sleep Medicine (AASM) defines adequate CPAP adherence as at least 4 hours of use on 70% of nights. Real-world data from CMS claims suggest fewer than 40% of commercially insured patients meet that threshold at 12 months. [2] Any trial that reduces AHI without requiring nightly device use is therefore addressing a genuine unmet clinical need, not an incremental one.

Tirzepatide and the SURMOUNT-OSA Trials

The SURMOUNT-OSA program is the most significant OSA pharmacology dataset published to date. Two parallel randomized, placebo-controlled trials enrolled adults with moderate-to-severe OSA (AHI ≥15) and obesity (BMI ≥30). Trial 1 (N=234) included patients not using CPAP; Trial 2 (N=235) included patients already on CPAP who were randomized to continue or switch to tirzepatide. The primary endpoint was change in AHI from baseline at 52 weeks. [3]

Key Efficacy Results

In Trial 1, tirzepatide 15 mg reduced AHI by a mean of 27.4 events per hour, a 55.0% reduction from a mean baseline of 51.5 events per hour (P<0.001 vs. Placebo). In Trial 2, tirzepatide reduced AHI by 30.4 events per hour, a 62.8% reduction from a mean baseline of 49.5 events per hour (P<0.001). [3] Approximately 42% of participants in Trial 1 and 51% in Trial 2 achieved an AHI below 5, meeting the threshold for OSA resolution.

Body-weight reduction in both trials averaged roughly 18-20%, consistent with the SURMOUNT-1 obesity data and confirming that the AHI reduction was largely mediated through fat-mass loss in the pharyngeal and peripharyngeal tissue.

Secondary Cardiovascular and Symptomatic Endpoints

Beyond AHI, tirzepatide arms showed statistically significant improvements in hypoxic burden (measured as the percentage of sleep time with SpO2 <90%), patient-reported sleepiness on the Epworth Sleepiness Scale, and C-reactive protein as a systemic inflammation marker. [3] The drug reduced high-sensitivity CRP by approximately 43% relative to placebo, a finding that may have independent cardiovascular relevance given that OSA-associated hypoxemia is an independent predictor of MACE in population cohorts.

Who Qualifies for Tirzepatide Under the Current FDA Label

The FDA label issued in January 2024 restricts tirzepatide's OSA indication to adults with moderate-to-severe OSA (AHI ≥15) who also have obesity (BMI ≥30). The label does not require CPAP failure as a prerequisite. Patients should also be on a reduced-calorie diet and increased physical activity program, consistent with the approved obesity indication. [4]

Hypoglossal Nerve Stimulation: Expanding the Evidence Base

The Inspire upper airway stimulation (UAS) system has been FDA-approved since 2014, but the evidence base supporting it has grown substantially through post-market registries and updated indication expansions.

ADHERE Registry Outcomes

The ADHERE post-market registry, now tracking more than 3,000 patients at 12 months, reports a median AHI reduction from 32.0 to 6.2 events per hour and an Epworth Sleepiness Scale reduction from 11.5 to 7.3. [5] Device-use rates exceed 85% at 12 months, a sharp contrast to CPAP adherence figures. The registry also shows that outcomes are durable at 60 months, with AHI remaining below 10 in approximately 78% of long-term users.

Evolving Patient Selection Criteria

The original FDA approval required AHI between 15 and 65 and the absence of complete concentric collapse at the level of the soft palate (assessed by drug-induced sleep endoscopy, DISE). More recent real-world data suggest that selective patients with partial concentric collapse may still respond, prompting ongoing trials to refine DISE-based selection. The EFFECT trial (NCT05283707) is a prospective multi-center study examining UAS in patients previously excluded due to anatomy, with topline results expected in 2025. [6]

Pediatric and Adolescent Indications

A separate FDA Breakthrough Device designation was granted in 2023 for Inspire therapy in adolescents aged 13-17 with Down syndrome and OSA refractory to adenotonsillectomy. The DREAMZ trial (NCT04996576) is enrolling this population, with results anticipated in 2026. [7]

Pharmacologic Pipeline Beyond GLP-1s

Tirzepatide gets the headlines, but at least two other pharmacologic strategies are in late-phase development targeting OSA through mechanisms unrelated to weight.

AD109: Noradrenergic-Antimuscarinic Combination

AD109 combines aroxybutynin (an antimuscarinic agent reducing genioglossus muscle hypotonia during REM sleep) with atomoxetine (a norepinephrine reuptake inhibitor that increases pharyngeal dilator tone). The rationale comes from the "loop gain" model of OSA pathophysiology: in a subset of patients, OSA is driven less by anatomical narrowing and more by unstable ventilatory control and reduced muscle responsiveness during sleep. [8]

The Phase 2b MARIPOSA trial (N=646) showed that AD109 reduced AHI by approximately 63% from baseline in patients with a predominantly muscle-tone phenotype, with an AHI reduction of 12.4 events per hour vs. Placebo (P<0.001). [9] The Phase 3 STARGAZER trial is currently enrolling, targeting approximately 1,200 participants, with results expected in late 2025.

Carbonic Anhydrase Inhibitors and Loop Gain Reduction

Acetazolamide and its derivatives reduce loop gain by blunting the ventilatory response to CO2 fluctuations. A randomized crossover trial published in the American Journal of Respiratory and Critical Care Medicine (N=26) showed that acetazolamide 500 mg twice daily reduced the loop gain index by 37% and AHI by approximately 40% in patients with high loop gain phenotype. [10] No large Phase 3 program has completed enrollment yet, but Apnimed is advancing AD109 and also exploring combination strategies that pair loop-gain reduction with arousal threshold agents.

Diagnostic Advances: From Lab to Home to Algorithm

Home Sleep Apnea Testing Validation

The AASM's 2017 clinical practice guideline conditionally recommended home sleep apnea testing (HSAT) as an alternative to polysomnography (PSG) for adults with a high pre-test probability of moderate-to-severe OSA and no significant comorbidities. [11] Subsequent validation studies, including a 2022 meta-analysis in CHEST (N=5,281), confirmed that HSAT sensitivity and specificity for AHI ≥15 are 87% and 79%, respectively, when using Type 3 devices. The gap vs. PSG narrows further with Level 2 (full-montage) home devices. [12]

AI-Driven Wearable Detection

A 2023 study published in npj Digital Medicine validated a deep-learning algorithm applied to photoplethysmography (PPG) signals from a commercial wrist wearable (Withings ScanWatch) against simultaneous PSG in 570 adults. The algorithm achieved an AUC of 0.91 for detecting AHI ≥15 and correctly classified OSA severity in 78% of participants. [13] While no wearable-based algorithm currently holds FDA 510(k) clearance as a standalone OSA diagnostic, the FDA De Novo pathway is being explored by multiple device makers.

Biomarker-Based Phenotyping

The "PALM" phenotyping framework, developed by Eckert and colleagues at Flinders University and published in the American Journal of Respiratory and Critical Care Medicine, classifies OSA patients along four physiologic traits: Pcrit (pharyngeal collapsibility), Arousal threshold, Loop gain, and Muscle responsiveness. This framework is not yet in routine clinical use, but it is the organizational backbone for selecting patients in most pharmacologic OSA trials currently enrolling. Clinicians who understand PALM can anticipate which patients are most likely to respond to AD109 (high loop gain, low muscle responsiveness) vs. Weight-loss therapy (high Pcrit, anatomical phenotype) vs. UAS (moderate Pcrit with intact muscle responsiveness). A practical version of this framework is being developed by the HealthRX clinical team as a patient-facing decision aid, pending physician review.

Cardiovascular Outcomes Trials: The Missing Piece

ISAACC Trial Findings

The ISAACC trial (Continuous Positive Airway Pressure in Patients with Acute Coronary Syndrome and Obstructive Sleep Apnea, N=1,264) randomized patients with ACS and OSA to CPAP plus usual care vs. Usual care alone and found no significant difference in the primary composite MACE endpoint at a median follow-up of 3.4 years (hazard ratio 0.92, 95% CI 0.75-1.13). [14] The trial was broadly interpreted as showing that CPAP does not reduce cardiovascular events, but the CPAP adherence in the treatment arm averaged only 2.8 hours per night, well below the 4-hour threshold, which complicates interpretation.

What ISAACC Does Not Tell Us

The 2022 AASM position statement on OSA and cardiovascular disease notes that "current evidence does not support CPAP as a cardiovascular risk-reduction therapy in patients with established coronary artery disease, partly due to adherence limitations in existing trials." [15] That statement leaves open the question of whether adequate CPAP use, or an alternative treatment achieving similar AHI reductions, would show a MACE benefit. The ongoing ALASKA trial (NCT05264428) is designed to address exactly this, enrolling patients with high cardiovascular risk and AHI ≥20 who will receive UAS (which has higher adherence than CPAP) as the active treatment arm. [6]

Metabolic Interactions With GLP-1 Drugs

Tirzepatide's dual GIP/GLP-1 agonism introduces cardiovascular effects that go beyond AHI reduction. The SELECT trial (N=17,604) established that semaglutide 2.4 mg reduces MACE by 20% in adults with obesity and established cardiovascular disease but without diabetes. [16] Whether the OSA-specific SURMOUNT-OSA population shares that cardiovascular benefit is not yet known. A pre-specified cardiovascular sub-study of SURMOUNT-OSA is planned but has not reported.

Updated Diagnostic Criteria and Guideline Field

The AASM defines OSA as an AHI ≥5 events per hour with symptoms (excessive daytime sleepiness, witnessed apneas, or comorbid hypertension, mood disorder, cognitive impairment, coronary artery disease, stroke, congestive heart failure, atrial fibrillation, or type 2 diabetes) OR an AHI ≥15 regardless of symptoms. [11] Severity classification runs from mild (AHI 5-14), to moderate (AHI 15-29), to severe (AHI ≥30).

The 2023 American Academy of Sleep Medicine and American Thoracic Society joint statement updated testing recommendations to support HSAT as a first-line option in uncomplicated adult cases, reserving PSG for patients with suspected comorbid sleep disorders, significant cardiorespiratory disease, or when HSAT is non-diagnostic. [11]

Scoring Rule Changes Affecting AHI

The AASM 2012 recommended scoring rules (now widely adopted) require that hypopneas be scored when there is a ≥3% oxygen desaturation OR an arousal, rather than the older 4% desaturation-only criterion. This change systematically increases measured AHI by 10-30% depending on the population, meaning historical trial data cannot always be directly compared with post-2012 study data. Clinicians reviewing older trial citations should verify which scoring rule was used. [11]

Pediatric OSA: A Separate Research Frontier

Pediatric OSA shares a name with the adult condition but differs substantially in pathophysiology, with adenotonsillar hypertrophy as the dominant driver in children aged 2-8. The CHAT trial (N=464, NEJM 2013) found that early adenotonsillectomy normalized PSG findings in 79% of children at 7 months vs. 46% in watchful waiting, but did not produce significant improvement in the primary attention and executive-function endpoint. [17] Post-hoc analysis showed behavioral and quality-of-life benefits, which remain the stated rationale for surgical referral in symptomatic children with AHI ≥2.

Positional and Myofunctional Therapies: Underused Evidence

Two low-cost interventions have accrued more trial evidence than their clinical adoption suggests.

Positional Therapy Devices

A meta-analysis in Journal of Clinical Sleep Medicine (2019, N=3,951 across 29 RCTs) showed that positional therapy (devices preventing supine sleep) reduced AHI by a mean of 10.5 events per hour in patients with positional OSA, defined as an AHI at least twice as high in the supine vs. Lateral position. [18] Response rates reach 60-70% in carefully selected positional-OSA patients, but only about 25% of OSA patients have the positional phenotype, making pre-treatment supine/lateral AHI split a useful screening step.

Myofunctional Therapy

A 2015 meta-analysis in SLEEP (N=120 across six trials) reported that oropharyngeal exercises reduced AHI by 50% and improved oxygen saturation nadir in adults with mild-to-moderate OSA. [19] The exercises target the genioglossus, lateral pharyngeal walls, and soft palate over a 3-month protocol. The effect size is clinically significant in mild-to-moderate disease and could plausibly be additive with pharmacotherapy.

Trials Currently Enrolling: A 2024 Watchlist

Below are the most clinically significant active OSA trials as of mid-2024.

| Trial | Intervention | Target N | Primary Endpoint | Expected Readout | |-------|-------------|----------|-----------------|-----------------| | STARGAZER | AD109 vs. Placebo | ~1,200 | Change in AHI at 26 weeks | Late 2025 | | EFFECT | Inspire UAS in anatomy-excluded patients | ~300 | AHI response rate at 12 months | 2025 | | ALASKA | Inspire UAS in high-CV-risk OSA | ~2,500 | MACE composite at 36 months | 2027 | | DREAMZ | Inspire UAS in Down syndrome adolescents | ~70 | AHI reduction at 12 months | 2026 | | SURMOUNT-OSA CV sub-study | Tirzepatide 15 mg | Pre-specified subgroup | MACE and AF burden | TBD |