---
title: "AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: Clinical evidence"
description: "Formulary evidence on AD109 (aroxybutynin-atomoxetine, Oxnimbi) in Obstructive sleep apnea: Clinical evidence. Covers SynAIRgy, LunAIRo, APC-APN-306, MARIPOSA, APC-APN-201, APC-004, and 1 further heading. 380 evidence topics quoted from 46 sources."
canonical: "https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/clinical-evidence"
published: "2026-11-18T10:43:47-05:00"
modified: "2026-11-18T10:43:47-05:00"
---

# AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: Clinical evidence

<a id="clinical-evidence"></a>

## 3 Clinical Evidence

<a id="heading-3.1"></a>

### 3.1 Study Summaries

<a id="heading-3.1.1"></a>

#### 3.1.1 SynAIRgy

<a id="heading-3.1.1.1"></a>

##### 3.1.1.1 Objective

<a id="topic-3.1.1.1-1"></a>

###### Trial objective as stated in the publication

> “To evaluate the efficacy/safety of AD109 over 6 months in a diverse OSA population unable to use PAP.”
> “We conducted the phase 3 SynAIRgy clinical trial to evaluate the efficacy, safety, and tolerability of AD109 vs placebo over 6 months in adults with mild-to-severe OSA who were intolerant to or refused PAP therapy.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.1-2"></a>

###### Primary and secondary objectives in the design paper

> “The primary objective of the LunAIRo and SynAIRgy trials is to demonstrate the efficacy of AD109 versus placebo for upper airway obstruction, as measured by the AHI4, among individuals across the full range of OSA severity, from mild to very severe, over a 26-week treatment period.”
> “Secondary objectives are to demonstrate efficacy of AD109 vs placebo on OSA symptoms including fatigue, one of the most common and bothersome OSA symptoms [[13], [14], [15]], and on measures of oxygenation which have been shown to correlate more strongly than AHI with adverse cardiovascular outcomes [16].”
> “Exploratory objectives for both trials involve in-depth examination of the effects of AD109 on sleep parameters, and additional measures of airway obstruction.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="heading-3.1.1.2"></a>

##### 3.1.1.2 Location and Study Date

<a id="topic-3.1.1.2-1"></a>

###### Sites and countries

> “conducted at 69 sites in the United States and Canada”
> “Additionally, trial sites were limited to the United States and Canada and may not fully represent OSA in other geographies.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “The trial enrolled 646 adult participants from 69 centers in the United States and Canada.”

Source: [Apnimed SynAIRgy publication release (2026)](https://ir.apnimed.com/news-releases/news-release-details/apnimed-announces-publication-its-phase-3-synairgy-trial-ad109)

> “We conducted our SynAIRgy clinical trial with some study sites located in Canada”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.2-2"></a>

###### Number of centers, May to October 2025, superseded in May 2026

> “The trial enrolled 646 adult participants from 73 centers in the US and Canada.”

Source: [Apnimed SynAIRgy topline release (2025)](https://apnimed.com/article/ad109phase3toplineresults/)

> “A total of 64 sites have been identified in the U.S. for LunAIRo (Supplemental Table 1) and 73 sites have been identified in both the U.S. and Canada for SynAIRgy (Supplemental Table 2).”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.2-3"></a>

###### Enrollment period

> “Between December 20, 2023 and March 26, 2025, 646 participants with mild-to-severe OSA were enrolled and randomized to receive AD109 (n = 324) or placebo (n = 322) (Figure 1).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “We initiated the LunAIRo and SynAIRgy Phase 3 clinical trials for AD109 in September and November of 2023, respectively.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.2-4"></a>

###### Registry study dates

| Field | Quoted record |
| --- | --- |
| Study start (actual) | “2023-09-16” |
| Primary completion (actual) | “2025-03-10” |
| Study completion (actual) | “2025-03-14” |
| Enrollment (actual) | “646” |

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

<a id="topic-3.1.1.2-5"></a>

###### Registry results posting dates

| Field | Quoted record |
| --- | --- |
| Results first submitted | “2026-08-18” |
| Results first posted | “2026-09-11” |

Source: [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results)

<a id="heading-3.1.1.3"></a>

##### 3.1.1.3 Study Design

<a id="topic-3.1.1.3-1"></a>

###### Randomized, double-blind, placebo-controlled parallel-arm design

> “SynAIRgy (APC-APN-305; NCT05813275)35 was a phase 3, randomized, double-blind, placebo-controlled, 6-month parallel-arm study of AD109 (Figure S1), conducted at 69 sites in the United States and Canada in compliance with the Declaration of Helsinki and Good Clinical Practice guidelines outlined by the International Council for Harmonisation.”
> “Eligible participants were randomized 1:1 to receive AD109 or placebo and stratified according to baseline AHI (<15, 15 to <30, ≥30 events/hour) and PROMIS-Fatigue T-score (<58.5 vs ≥58.5).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.3-2"></a>

###### Registry design record

| Field | Quoted record |
| --- | --- |
| Official title | “Phase 3 Randomized Double-Blind Placebo-Controlled 6-month Parallel-Arm Study to Compare a Fixed Dose Combination of Aroxybutynin/Atomoxetine (AD109) to Placebo in Obstructive Sleep Apnea (SynAIRgy Study)” |
| Allocation | “Randomized” |
| Intervention model | “Parallel Assignment” |
| Masking | “Quadruple” |

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

<a id="topic-3.1.1.3-3"></a>

###### Visit schedule and polysomnography timing

> “Participants received AD109 or placebo during a 26-week period, including a 1-week run-in dose followed by 25 weeks at the target dose, then 2 weeks without dosing, followed by last visit at week 28 (Figure S1).”
> “Standard overnight polysomnography (PSG) was performed at baseline and weeks 4 and 26.”
> “Key patient-reported outcomes (PROs) were assessed using PROMIS-Fatigue and PROMIS-Sleep Impairment 8-item questionnaires at baseline and weeks 4, 13, and 26 (Supplementary Methods).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.3-4"></a>

###### Severity strata enrollment limits

> “To ensure representative distribution of baseline AHI4 severity, the size of AHI4 strata will be controlled through enrollment limits, with AHI4 <15 representing ∼20–40% of participants, AHI4 15–30 representing ∼20–40% of participants, and AHI4 >30 representing ∼20–40% of participants.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

> “Our trials were designed to potentially support a broader label indicated for patients across the range of OSA severities by ensuring no more than 40% of the overall study population in each trial could represent mild, moderate or severe patient subgroups.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.3-5"></a>

###### Blinding, central polysomnography scoring, and data monitoring

> “All study participants, investigators, site staff, staff at the PSG scoring center, staff at the CRO and sponsor staff are blinded to treatment assignment throughout the studies.”
> “The studies are conducted with the support of a PSG scoring center (Sleep Strategies, Ottawa, ON) responsible for centralized PSG scoring blinded to treatment assignment for both screening and outcome assessment and generation of standardized PSG variables.”
> “Each participant's baseline and on-treatment PSGs were evaluated by the same scorer to maintain consistency.”
> “An independent DMC, with expertise in sleep apnea, clinical trials, and biostatistics meets regularly to assess unblinded safety data and make recommendations for both trials.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.3-6"></a>

###### Separate sites from the LunAIRo trial

> “LunAIRo and SynAIRgy were performed in different centers, with the exception of a small number of sites that enrolled for SynAIRgy after the enrollment of LunAIRo was closed. No participant was enrolled in both studies.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.3-7"></a>

###### Exploratory cohort of GLP-1 receptor agonist users not analyzed

> “An additional exploratory cohort of up to 100 participants receiving a concomitant glucagon-like peptide-1 (GLP-1) receptor agonist specifically for weight loss (ie, not for treatment of type 2 diabetes) was initiated and planned to be analyzed separately to evaluate the potential impact of weight loss on the efficacy and safety of AD109 compared with placebo.”
> “Due to low enrollment in this cohort, the prespecified exploratory analyses are not included in this report.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.3-8"></a>

###### Optional open-label extension after completion

> “Participants completing either of these studies are eligible for an optional open label extension (OLE) study of AD109 in which safety is assessed for up to an additional year (Supplement Information).”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.3-9"></a>

###### Planned randomization and in-treatment polysomnography in the design abstract

> “Approximately 640 participants will be randomized in a 1:1 ratio to receive either placebo or AD109.”
> “Participants will have a PSG on treatment at week 4.”

Source: [Strollo et al. (2024)](https://academic.oup.com/sleep/article/47/Supplement_1/A233/7654816)

<a id="heading-3.1.1.4"></a>

##### 3.1.1.4 Eligibility Criteria

<a id="topic-3.1.1.4-1"></a>

###### Inclusion and exclusion criteria in the publication

> “Patients aged ≥18 years with mild-to-severe OSA who had failed or refused PAP therapy were screened for eligibility.”
> “Participants with AHI ≥5 events/hour were initially enrolled.”
> “An apnea was defined by an event with ≥90% reduction in airflow for ≥10 seconds and hypopnea by ≥30% reduction in airflow for ≥10 seconds associated with a ≥4% decrease in the oxygen saturation.”
> “The eligibility criterion of enrolling participants with AHI ≥5 events/hour was subsequently changed to ≥10 to ≤45 events/hour to avoid overrepresentation of participants with very severe or very mild OSA.”
> “Other inclusion criteria were an 8-item Patient-Reported Outcomes Measurement Information System (PROMIS)–Fatigue T-score >50, representing above-average fatigue and a body mass index (BMI) of 18.5 to 40 kg/m2 in men or 18.5 to 42 kg/m2 in women.”
> “Key exclusion criteria were clinically significant or medically uncontrolled cardiovascular disease, presence of narcolepsy, restless leg syndrome requiring medication, rapid eye movement (REM) sleep behavior disorder, bothersome symptoms of insomnia, and craniofacial malformation syndromes.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.4-2"></a>

###### Registry inclusion criteria, current version

> “PSG criteria: AHI of ≥10 and ≤45; ≤ 25% central or mixed apneas; and PLM arousal index ≤15”
> “PROMIS-Fatigue: raw score ≥17”
> “PAP intolerance or current PAP refusal.”
> “BMI between 18.5 and 40 kg/m2 for men, or 42 kg/m2 for women, inclusive.”

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

<a id="topic-3.1.1.4-3"></a>

###### Apnea-hypopnea index eligibility range in protocol amendment 5.0, February 2024

> “Version 5.0 February 29, 2024”
> “AHI4 (Hypopneas defined by 4% oxygen desaturation) of ≥10 and ≤45”

Source: [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf)

<a id="topic-3.1.1.4-4"></a>

###### Registry exclusion criteria

> “Insomnia disorder characterized by difficulty initiating or maintaining sleep, or use within the past month of sedative-hypnotics or other medication for the purpose of treating or avoiding insomnia symptoms.”
> “Clinically significant cardiac disease, e.g., ventricular arrhythmia, untreated or unstable coronary artery disease, cardiac failure.”

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

<a id="topic-3.1.1.4-5"></a>

###### Definitions of PAP failure and PAP refusal

> “PAP failure is defined as no PAP use for ≥3 months prior to randomization, or return/removal of PAP device from the home.”
> “PAP refusal is defined as refusal of PAP after prior positive sleep study, or prior refusal of provider-recommended sleep study due to unwillingness to consider PAP.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.4-6"></a>

###### PROMIS-Fatigue threshold expressed as raw score and T-score

> “A PROMIS-Fatigue raw score of 17 is equivalent to a T-score of 50.4.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.4-7"></a>

###### Additional exclusion criteria in the design paper

> “Clinically significant or medically uncontrolled cardiovascular disease, or resting heart rate >100), untreated or unstable coronary artery disease, heart failure, cerebrovascular event or revascularization within 3 months.”
> “Schizophrenia, schizoaffective disorder, generalized anxiety disorder or bipolar disorder according to DSM-5 criteria”
> “Attempted suicide within 1 year prior to screening, or current suicidal ideation.”
> “Current bothersome symptoms of bladder outlet obstruction including difficulty initiating or maintaining urinary flow, straining, or sensation of incomplete bladder emptying”
> “Blood pressure >145/90 mmHg”
> “Narrow angle glaucoma”
> “Women who are pregnant or nursing”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.4-8"></a>

###### Disallowed concomitant medications

> “Atomoxetine is metabolized by CYP2D6 and aroxybutynin by CYP3A4, and consequently, strong pharmacologic inhibitors of these metabolic enzymes are not allowed as concomitant therapy.”
> “Strong CYP2D6 inhibitors, and strong CYP3A4 inhibitors”
> “Modafinil, armodafinil, solriamfetol, and pitolisant”
> “Oral anticholinergics, and cholinesterase inhibitors”
> “Drugs used exclusively for weight loss”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.4-9"></a>

###### Apnea-hypopnea index eligibility in protocol versions 2.0 and 4.0, July 2023, superseded in February 2024

> “Version 2.0 July 26, 2023”
> “the V2 PSG must be >5 and ≤45, with ≤ 25% central or mixed apneas”

Source: [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf)

<a id="topic-3.1.1.4-10"></a>

###### Apnea-hypopnea index eligibility in congress abstracts, April 2024 and May 2025, superseded in May 2026

> “Participants were enrolled over the full range of OSA severity as measured by apnea-hypopnea index with 4% desaturation (AHI4) >5 with no upper limit.”

Source: [Strollo et al. (2025b)](https://academic.oup.com/sleep/article/48/Supplement_1/A339/8134797)

> “Eligible participants include adults with AHI4 (4% desaturation definition for hypopneas) ≥5, and a BMI < 42kg/m2 for women and < 40 kg/m2 for men who either decline or are unable to tolerate continuous positive airway pressure treatment.”

Source: [Strollo et al. (2024)](https://academic.oup.com/sleep/article/47/Supplement_1/A233/7654816)

<a id="heading-3.1.1.5"></a>

##### 3.1.1.5 Treatment

<a id="topic-3.1.1.5-1"></a>

###### AD109 regimen with one-week run-in and no dose reduction

> “AD109 dose was aroxybutynin 2.5 mg/atomoxetine 37.5 mg for week 1 and aroxybutynin 2.5 mg/atomoxetine 75 mg for the remaining weeks (dose reductions not permitted).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.5-2"></a>

###### Registry intervention descriptions

| Arm | Quoted record |
| --- | --- |
| AD109 | “Aroxybutynin 2.5 mg/atomoxetine 75 mg tablet, administered orally once daily at bedtime (Weeks 2-26). Week 1 run-in: aroxybutynin 2.5 mg/atomoxetine 37.5 mg tablet.” |
| Placebo | “Matching placebo tablet administered orally once daily at bedtime for 26 weeks, with run-in and target dose matched to active arm packaging” |

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

<a id="topic-3.1.1.5-3"></a>

###### Dispensing and dosing rules

> “The IMP is dispensed as a blister package of 7 run-in dose tablets and a bottle containing 33 target dose tablets.”
> “Dose reduction is not permitted, and dosing is discontinued if the participant experiences intolerance to the assigned study drug.”
> “Participants are not prohibited from initiating PAP during the trial if deemed by the participant's primary care provider to be in the participant's best medical interest.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.5-4"></a>

###### Run-in dose in base notation and reason dose reduction was not permitted

> “A lower dose of AD109 (2.3/37.5mg) was administered during the one-week run-in period; however, dose reduction was not permitted in the Phase 3 trials, to maximize evaluation of efficacy.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.5-5"></a>

###### Adherence calculation

> “The medication adherence (percent compliant) was calculated as medication adherence = 100 × (total number of doses received/total number of doses expected).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="heading-3.1.1.6"></a>

##### 3.1.1.6 Study Outcomes

<a id="heading-3.1.1.6.1"></a>

###### 3.1.1.6.1 Rationale for Using Apnea-Hypopnea Index as the Primary Endpoint

<a id="topic-3.1.1.6.1-1"></a>

###### Primary endpoint: change in AHI with 4% desaturation hypopnea rule at week 26

> “The primary efficacy endpoint was the change in AHI from baseline to week 26.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “AHI, Apnea-Hypopnea Index, is the number of apneas or hypopneas recorded via polysomnography during the study per hour of sleep. Apnea is defined as a cessation of airflow lasting at least 10 seconds, hypopnea as a decrease in airflow by at least 30% from baseline for at least 10 seconds occurring with a drop in oxygen saturation (SpO2) by at least 4%.”

Source: [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275)

> “Change from baseline to week 26 in AHI4”
> “AHI4 = apnea-hypopnea index based on 4% hypopnea desaturation”

Source: [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf)

<a id="topic-3.1.1.6.1-2"></a>

###### Key secondary and exploratory endpoints

> “The key secondary efficacy endpoints (at week 26) were change from baseline in oxygen desaturation index based on 3% desaturation (ODI), PROMIS-Fatigue T-score, hypoxic burden based on 4% desaturation (HB), PROMIS-Sleep Impairment T-score, and proportion of participants with ≥50% reduction in AHI.”
> “Prespecified exploratory endpoints (at week 26) included change from baseline in OSA severity category (none [AHI <5], mild [AHI 5 to <15], moderate [AHI 15 to <30], and severe [AHI ≥30]), proportion of participants with AHI reduction of ≥30% to ≥90% by 10% increments, and proportion of breaths with snoring.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.6.1-3"></a>

###### Other secondary endpoints in the design paper

> “Other secondary endpoints in both trials include the change from baseline in the PGI-S score for fatigue at Week 26, the patient global impression of change (PGI-C) score for fatigue at Week 26, and the change from baseline in ESS at Week 26.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.6.1-4"></a>

###### Rationale for AHI, oxygen desaturation index, and hypoxic burden

> “The primary efficacy endpoint in most OSA trials is the AHI which historically has been the principal biomarker for the disease [[22], [23], [24], [25]].”
> “Compared to AHI4, that only considers the frequency of obstructive events (events/hour of sleep), the HB4 also incorporates the length and the depth of the apnea/hypopnea-related chronic intermittent hypoxia during sleep [16].”
> “This metric has demonstrated a stronger correlation with cardiovascular mortality than AHI in recent epidemiological cohort analyses [16].”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.6.1-5"></a>

###### Rationale for PROMIS-Fatigue as a key secondary endpoint

> “Fatigue is one of the most common and most bothersome symptoms of OSA and is a distinct symptom from sleepiness [13].”
> “PROMIS-Fatigue may provide increased sensitivity to symptom improvement in OSA [31], particularly in the substantial proportion of OSA patients without EDS [32].”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.6.1-6"></a>

###### Content validity and meaningful change of the patient-reported outcome measures, SynAIRgy exit interviews

> “Qualitative exit interviews were conducted with 32 participants within 10 days of completing the trial.”
> “Mapping demonstrated all PROMs have strong concept coverage.”
> “Most participants reported a one-point change on response options for individual items from the PROMIS-Fatigue (61%-96%), PROMIS-Sleep Impairment (52%-96%), and ESS (62-96%) was clinically meaningful, regardless of baseline response.”
> “A one-point change on PROMs represents a clinically meaningful improvement from the patient's perspective.”

Source: [Ojile et al. (2026)](https://pubmed.ncbi.nlm.nih.gov/42247851/)

<a id="topic-3.1.1.6.1-7"></a>

###### Investigators' rationale for the clinical meaningfulness of the AHI change, treatment policy and on-treatment estimands

> “At the group level, treatment was associated with a shift in disease severity from the moderate to mild range (from a median of 19.8 at baseline to 13.3 at week 26 in the ITT, and from a median of 19.3 at baseline to 10.6 at week 26 in the on-treatment estimand).”
> “This magnitude of improvement, we believe, is clinically meaningful, as mild OSA has been associated with lower cardiovascular, metabolic, and neurocognitive risk compared to moderate OSA.”
> “At 26 weeks, >25% of participants in the AD109 arm achieved ≥50% reduction in AHI, a clinically meaningful improvement based on American Academy of Sleep Medicine guidelines,45,46 although this effect was not statistically significant compared with the placebo arm.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.6.1-8"></a>

###### Company position on relative rather than absolute AHI change in mild and moderate disease

> “In populations with mild and moderate OSA, meaningful clinical benefits are better reflected by relative reductions in AHI and HB, responder analyses, and shifts to lower disease severity categories, rather than by absolute reductions alone.”
> “Therefore, we believe that assessment of the clinical meaningfulness of response to treatment requires consideration of multiple objective and subjective measures.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.6.1-9"></a>

###### FDA pre-NDA feedback on the clinical meaningfulness of the primary endpoint results

> “In connection with the FDA’s review of our top-line Phase 3 data, at the pre-NDA meeting, the FDA provided feedback about the clinical meaningfulness of certain endpoint results, including the primary endpoint and the limitations of the patient-reported outcomes (PROs).”
> “The FDA indicated that additional analyses and justification may be needed to support that the data demonstrates a clinically meaningful benefit for the treatment of OSA.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.6.1-10"></a>

###### Editorial comment on the clinical significance of the absolute AHI change, treatment policy estimand

> “Among 646 participants (median apnea-hypopnea index [AHI], 19.6), treatment with AD109 over 26 weeks resulted in a placebo-adjusted reduction in AHI of approximately 4 events per hour.”
> “The clinical significance of this magnitude of reduction is uncertain when considered as a continuous measure.”
> “Although such threshold-based outcomes have recognized limitations, these findings suggest that a subset of patients derives a physiologically meaningful benefit.”

Source: [Mehra and Mokhlesi (2026)](https://academic.oup.com/ajrccm/article/212/7/1427/8679096)

<a id="topic-3.1.1.6.1-11"></a>

###### Protocol amendment 6.0 reordering the primary and key secondary endpoints, April 2025

> “Protocol Version: 6.0”
> “17 April 2025”
> “The order of endpoints, including the primary endpoint and key secondary endpoints, was changed to align with recent regulatory approvals for the treatment of OSA related to obesity.”

Source: [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf)

<a id="topic-3.1.1.6.1-12"></a>

###### Primary endpoint: responder proportion in protocol version 4.0, December 2023, superseded in April 2025

> “Version 4.0 December 15, 2023”
> “The sample size for the primary cohort has been selected based on the number of participants required to provide at least 90% power for the primary endpoint (proportion of participants with a ≥ 50% reduction from baseline to week 26 in AHI) and the first key secondary endpoint (change from baseline to week 26 in PROMIS Fatigue T-score).”
> “The primary clinical question of interest for the primary endpoint is: what is the difference”
> “between AD109 and placebo in the AHI response rate at Week 26”

Source: [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf)

<a id="topic-3.1.1.6.1-13"></a>

###### Primary endpoint: responder proportion in the registry record, March 2025, superseded in November 2025

> “Viewing V22 (2025-03-18)”
> “23 2025-11-18 Study Status Outcome Measures”
> “Efficacy of AD109 on airway obstruction in mild to sever OSA”
> “Proportion of participants with reduction in AHI (AHI, average number of events for every hour of sleep), measured by polysomnography”

Source: [ClinicalTrials.gov NCT05813275 version 22](https://clinicaltrials.gov/study/NCT05813275?tab=history&a=22)

<a id="topic-3.1.1.6.1-14"></a>

###### Primary endpoint: responder proportion in congress abstracts, April 2024 and May 2025, superseded in May 2025

> “The primary endpoint is the response rate in the AD109 arm vs. placebo arm, where response is defined as >50% reduction in AHI4 from baseline to 26-week polysomnogram (PSG).”

Source: [Strollo et al. (2024)](https://academic.oup.com/sleep/article/47/Supplement_1/A233/7654816)

> “The primary efficacy endpoint is the proportion of participants with ≥50% reduction in AHI4 at six months.”

Source: [Strollo et al. (2025b)](https://academic.oup.com/sleep/article/48/Supplement_1/A339/8134797)

<a id="heading-3.1.1.7"></a>

##### 3.1.1.7 Statistical Analysis Description

<a id="heading-3.1.1.7.1"></a>

###### 3.1.1.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.1.7.1-1"></a>

###### Planned sample size and power

> “The sample size (640 participants) was calculated based on the number of participants required to provide >90% power for the primary endpoint and one of the key secondary efficacy endpoints (change from baseline to week 26 in PROMIS-Fatigue T-score) at a 2-sided significance level of 0.05 (Supplementary Methods).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.1-2"></a>

###### Sample size assumptions from the MARIPOSA trial

> “In a previous Phase 2 study (MARIPOSA), the standard deviation of the placebo group change from baseline in AHI4 was approximately 9.0.”
> “Based on the observed dropout rates in MARIPOSA, it is assumed that the dropout rate in the placebo group will be 5% and the dropout rate in the AD109 group will be 25%.”
> “A total of 640 participants (320/group) will provide >98% power to detect a treatment difference of ≥ 4 points between AD109 and placebo for the change in AHI4.”
> “With these assumptions, simulation demonstrated that 640 participants (320/group) would provide approximately 90% power to detect a treatment difference of 2.5 points and approximately 95% power to detect a treatment difference of 2.75 points.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.1.7.1-3"></a>

###### Participants randomized and analyzed

| Population | AD109 | Placebo | Total |
| --- | --- | --- | --- |
| Randomized, all cohorts | “AD109 (n = 324)” | “placebo (n = 322)” | “646 participants” |
| Randomized, main cohort | “AD109 (n = 319)” | “placebo (n = 320)” | “639 participants” |
| Intent-to-treat set (treatment policy estimand) | “AD109 (n = 302)” | “Placebo (n = 313)” | “ITT (N = 615)” |
| On-treatment estimand | “AD109 (n = 260)” | “Placebo (n = 296)” | “On-treatment estimand (N = 556)” |
| Safety analysis set | “AD109a (n = 325)” | “Placebo (n = 319)” | “Safety analysis set (N = 644)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.1-4"></a>

###### Participants excluded from the efficacy analyses

> “The 7 participants (5 in the AD109 arm and 2 in the placebo arm) randomized and treated in the exploratory cohort are included in the safety analysis but not in the efficacy analyses.”
> “Sixteen participants enrolled at site 1416 in the AD106 arm were excluded from the efficacy analyses because the site was terminated due to Good Clinical Practice noncompliance.”
> “Six participants enrolled at site 1416 in the placebo arm were excluded from the efficacy analyses because the site was terminated due to Good Clinical Practice noncompliance.”
> “These numbers reflect participants by assigned treatment, but 2 participants who were assigned to placebo received AD109 instead.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="heading-3.1.1.7.2"></a>

###### 3.1.1.7.2 Description of Analysis Sets

<a id="topic-3.1.1.7.2-1"></a>

###### Treatment policy estimand (intent-to-treat) and on-treatment estimand

> “Primary analyses of the efficacy endpoints were performed using an intent-to-treat approach (ITT); supportive analyses were performed using the on-treatment estimand.”
> “The ITT analysis included all randomized participants who received ≥1 dose of AD109/placebo and evaluated endpoints regardless of treatment adherence or discontinuation.”
> “For the ITT estimand, missing data were assumed to be missing-not-at-random, and endpoints were analyzed with a pattern-mixture model using a jump-to-reference approach by means of sequential modeling with multiple imputation.”
> “The on-treatment estimand included all randomized participants who received ≥1 dose of AD109/placebo and who had ≥1 post-baseline PSG assessment on treatment.”
> “The on-treatment estimand focused on endpoints observed during active treatment exposure and excluded data collected after treatment discontinuation, with missing values imputed using a last observation carried forward approach.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.2-2"></a>

###### Registry definitions of the Primary Full Analysis Set and While-on-Treatment population

> “The Primary Full Analysis Set includes all participants who were not from the terminated site 1416, who were randomized into the main cohort and received at least 1 dose of IMP.”

Source: [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results)

<a id="topic-3.1.1.7.2-3"></a>

###### Statistical model and responder analysis

> “The primary analysis for the primary and secondary endpoints defined on a continuous scale used a restricted maximum likelihood–based mixed model for repeated measurements in combination with the Newton–Raphson algorithm.”
> “Analysis of the proportion of participants with ≥50% reduction in AHI at week 26 (key secondary endpoint not on a continuous scale) was conducted using the stratified Cochran–Mantel–Haenszel procedure, and estimates of risk difference between arms, along with the 2-sided 95% CI, were calculated based on the Cochran–Mantel–Haenszel stratum weights38 and the Sato variance estimator.39”
> “For the proportion of participants with ≥50% reduction in AHI, participants who discontinued treatment were considered nonresponders.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.2-4"></a>

###### Multiplicity control and hierarchical testing order

> “For the primary endpoint, no adjustment for multiplicity was needed and a 2-sided 5% significance level was used.”
> “Adjustment for multiplicity of the key secondary endpoints was conducted through a closed testing approach, where each key secondary endpoint was tested in a prespecified order: (1) change from baseline to week 26 in ODI, (2) change from baseline to week 26 in PROMIS-Fatigue T-score, (3) change from baseline to week 26 in HB, (4) change from baseline to week 26 in PROMIS-Sleep Impairment T-score, and (5) proportion of participants with ≥50% reduction in AHI at week 26.”
> “Key secondary endpoints that are nominally statistically significant but occur after a nonsignificant result in the prespecified hierarchical testing sequence should be interpreted with caution, as formal statistical significance cannot be claimed.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.2-5"></a>

###### Relative change in AHI on the log scale added as an exploratory analysis in statistical analysis plan version 2.0, April 2025

> “SAP Text Version Number: 2.0”
> “SAP Text Date: 22-Apr-2025”
> “Protocol Version and Date: V6.0, 17-Apr-2025”
> “Added an exploratory analysis of AHI4 to assess the relative change on the log scale in section 8.1”

Source: [Apnimed SynAIRgy statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/SAP_001.pdf)

<a id="topic-3.1.1.7.2-6"></a>

###### Safety analysis set

> “Safety was evaluated in the safety analysis set, including all randomized participants who received ≥1 dose of AD109/placebo (Supplementary Methods).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.7.2-7"></a>

###### Per-protocol population as described by the company

> “In addition, supportive per-protocol analyses were performed in a prespecified per-protocol population, defined prior to database lock and unblinding, consisting of participants who were adherent to treatment (≥70% compliance) and had no major protocol deviations, providing additional context on clinical response to treatment under conditions of consistent use.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.1.8"></a>

##### 3.1.1.8 Results

<a id="heading-3.1.1.8.1"></a>

###### 3.1.1.8.1 Participant Disposition

<a id="topic-3.1.1.8.1-1"></a>

###### Trial completion and completion of six months of treatment

> “Among participants who received treatment, 84.6% (539/637) completed the trial, with 83.3% (265/318) in the AD109 group and 85.9% (274/319) in the placebo group.”
> “Overall, 74.7% (476/637) completed 6 months of treatment (65.7% [209/318] with AD109 and 83.7% [267/319] with placebo; Figure 1).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.1-2"></a>

###### Treatment adherence

> “In the AD109 arm, treatment adherence (percent compliant) was 95.7% (SD, 9.2%) for participants completing the study on treatment, and 90.2% (SD, 35.2%) for those discontinuing treatment.”
> “In the placebo arm, treatment adherence was 95.5% (SD, 9.5%) for participants completing the study on treatment, and 77.5% (SD, 42.4%) for those discontinuing treatment.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.1-3"></a>

###### Overall treatment discontinuation before week 26

> “Limitations of this trial include that the substantial drug discontinuation rate before 26 weeks (34.3% for AD109 vs 16.3% for placebo) likely diminished the magnitude of benefits observed.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “In SynAIRgy, among participants randomized in the main cohort, AEs led to treatment discontinuation prior to month 6 in 21.2% of participants receiving AD109 compared with 3.1% on placebo, contributing to overall discontinuation rates of 34.6% for AD109 and 16.5% for placebo at month 6.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.1-4"></a>

###### Timing of treatment discontinuation

> “AE-related discontinuations occurred predominantly early after treatment initiation, with markedly fewer events later in the treatment period (Figure S10).”
> “In contrast, non-AE-related discontinuations were similar between the AD109 and placebo arms and were distributed more uniformly over time and generally at lower counts.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.1-5"></a>

###### Registry participant flow and reasons for study withdrawal

| Milestone or reason | AD109 | Placebo |
| --- | --- | --- |
| Started | “324” | “322” |
| Completed | “269” | “276” |
| Not completed | “55” | “46” |
| Adverse event | “4” | “1” |
| Lost to follow-up | “15” | “15” |
| Physician decision | “2” | “2” |
| Site terminated by sponsor | “8” | “2” |
| Withdrawal by subject | “21” | “22” |
| Other or not reported | “5” | “4” |

Source: [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results)

<a id="topic-3.1.1.8.1-6"></a>

###### Main cohort completion in the congress abstract

> “Eligible participants (N=639) were randomized (AD109: n=319; PBO: n=320) with 539 (84.6%) completing the trial.”

Source: [Strollo et al. (2025c)](https://journal.chestnet.org/article/S0012-3692%2825%2905441-8/fulltext)

<a id="heading-3.1.1.8.2"></a>

###### 3.1.1.8.2 Baseline Characteristics

<a id="topic-3.1.1.8.2-1"></a>

###### Baseline characteristics, intent-to-treat set, AHI with 4% desaturation hypopnea rule

| Characteristic | AD109 | Placebo | Total |
| --- | --- | --- | --- |
| Participants | “AD109 (n = 302)” | “Placebo (n = 313)” | “Total (N = 615)” |
| Age, years, median (minimum-maximum) | “59 (24-83)” | “58 (25-87)” | “58 (24-87)” |
| Female, n (%) | “150 (49.7)” | “153 (48.9)” | “303 (49.3)” |
| Black or African American, n (%) | “55 (18.2)” | “62 (19.8)” | “117 (19.0)” |
| White, n (%) | “212 (70.2)” | “216 (69.0)” | “428 (69.6)” |
| Asian, n (%) | “23 (7.6)” | “22 (7.0)” | “45 (7.3)” |
| Hispanic or Latino, n (%) | “63 (20.9)” | “69 (22.0)” | “132 (21.5)” |
| Body weight, kg, mean (SD) | “93.3 (17.0)” | “93.1 (17.1)” | “93.2 (17.0)” |
| BMI, kg/m2, median (minimum-maximum) | “32.5 (18.5-41.9)” | “32.4 (20.1-42.0)” | “32.4 (18.5-42.0)” |
| BMI 18.5-29.9 kg/m2, n (%) | “107 (35.4)” | “102 (32.6)” | “209 (34.0)” |
| BMI 30-34.9 kg/m2, n (%) | “100 (33.1)” | “111 (35.5)” | “211 (34.3)” |
| BMI 35 kg/m2 or higher, n (%) | “95 (31.5)” | “100 (31.9)” | “195 (31.7)” |
| AHI, events/h, median (minimum-maximum) | “19.8 (5.2-101.8)” | “19.1 (5.1-70.0)” | “19.6 (5.1-101.8)” |
| Mild OSA, AHI 5 to <15, n (%) | “105 (34.8)” | “109 (34.8)” | “214 (34.8)” |
| Moderate OSA, AHI 15 to <30, n (%) | “125 (41.4)” | “134 (42.8)” | “259 (42.1)” |
| Severe OSA, AHI 30 or higher, n (%) | “72 (23.8)” | “70 (22.4)” | “142 (23.1)” |
| ODI, events/h, median (minimum-maximum) | “27.3 (6.8-96.3)” | “26.9 (4.5-79.3)” | “27.0 (4.5-96.3)” |
| HB, %min/h, median (minimum-maximum) | “33.7 (0.8-379.8)” | “28.8 (2.4-272.8)” | “31.1 (0.8-379.8)” |
| PROMIS-Fatigue T-score, mean (SD) | “59.2 (7.0)” | “58.9 (6.7)” | “59.0 (6.8)” |
| PROMIS-Sleep Impairment T-score, mean (SD) [n] | “58.9 (7.2) [299]” | “58.6 (7.4) [309]” | “58.7 (7.3) [608]” |
| ESS score, mean (SD) [n] | “10.0 (4.6) [299]” | “10.1 (4.7) [311]” | “10.0 (4.6) [610]” |
| PAP intolerance, n (%) | “208 (68.9)” | “203 (64.9)” | “411 (66.8)” |
| PAP refusal, n (%) | “94 (31.1)” | “111 (35.5)” | “205 (33.3)” |
| Hypertension, n (%) | “152 (50.3)” | “153 (48.9)” | “305 (49.6)” |
| Diabetes, n (%) | “52 (17.2)” | “51 (16.3)” | “103 (16.7)” |
| Mood disturbances, n (%) | “60 (19.9)” | “45 (14.4)” | “105 (17.1)” |
| Insomnia, n (%) | “17 (5.6)” | “15 (4.8)” | “32 (5.2)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.2-2"></a>

###### Distribution of baseline severity and body mass index in the intent-to-treat set

> “The distribution of baseline OSA severity was broad and right-skewed with 34.8% (214/615) mild, 42.1% (259/615) moderate, and 23.1% (142/615) severe (Table 1, Figure S2).”
> “Baseline insomnia was assessed based on participant-reported medical history obtained at screening and includes initial insomnia (difficulty falling asleep at the beginning of the night), insomnia, and middle insomnia (difficulty maintaining sleep, characterized by frequent or prolonged nighttime awakenings).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.2-3"></a>

###### Registry baseline age and mean AHI with 4% desaturation hypopnea rule, Primary Full Analysis Set

| Characteristic | AD109 | Placebo | Total |
| --- | --- | --- | --- |
| Age, years, mean (SD) | “57.5(10.67)” | “57.1(11.11)” | “57.3(10.89)” |
| Baseline AHI4, events/hr, mean (SD) | “22.13(11.819)” | “21.47(10.607)” | “21.79(11.214)” |

Source: [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results)

<a id="topic-3.1.1.8.2-4"></a>

###### Baseline characteristics, all 646 randomized participants, AHI with 4% desaturation hypopnea rule

> “At baseline, the mean (SD) age was 57.1 (11.0) years, and mean (SD) BMI was 32.3 (5.0) kg/m2 with a balanced proportion of female (n=317, 49.1%) and male (329, 50.9 %) participants.”
> “The mean (SD) AHI4 at baseline was 22 (11) with 34.4%, 42.4%, and 23.2% of participants having mild (AHI4 5–< 15) moderate (AHI4 15–< 30), or severe (AHI4 ≥30) OSA, respectively.”
> “Mean (SD) PROMIS-Fatigue T-score was 59.1 (7.0), mean (SD) PROMIS-Sleep Impairment T-score score was 58.5 (7.5), and mean (SD) ESS was 10.1 (4.7).”

Source: [Strollo et al. (2025b)](https://academic.oup.com/sleep/article/48/Supplement_1/A339/8134797)

<a id="topic-3.1.1.8.2-5"></a>

###### Baseline oxygenation measures, all 646 randomized participants

| Characteristic | Quoted value |
| --- | --- |
| ODI3, mean (SD) | “29 (13)” |
| HB4, mean (SD) | “43 (40)” |

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="heading-3.1.1.8.3"></a>

###### 3.1.1.8.3 Efficacy Results

<a id="topic-3.1.1.8.3-1"></a>

###### Primary and key secondary endpoints at week 26, treatment policy estimand, AHI with 4% desaturation hypopnea rule

| Endpoint | AD109 | Placebo | Difference | 95% CI | P value |
| --- | --- | --- | --- | --- | --- |
| Change in AHI, LS mean (95% CI), events/h (primary) | “−3.3 (−5.1 to −1.5)” | “0.7 (−1.0 to 2.4)” | “−4.0” | “(−6.4 to −1.6)” | “P = .001” |
| Change in ODI, LS mean (95% CI), events/h | “−3.7 (−5.6 to −1.7)” | “0.9 (−1.1 to 2.8)” | “−4.5” | “(−7.3 to −1.8)” | “P = .001” |
| Change in PROMIS-Fatigue T-score, LS mean (95% CI) | “−7.4 (−8.3 to −6.5)” | “−6.2 (−7.1 to −5.3)” | “−1.2” | “(−2.4 to 0.0)” | “P = .059” |
| Percent reduction in geometric mean HB, LS mean (95% CI) | “44.7 (34.5-53.4)” | “8.5 (‒8.1 to 22.5)” | “39.6” | “(23.6-52.3)” | “P <.0001” |
| Change in PROMIS-Sleep Impairment T-score, LS mean (95% CI) | “−7.6 (−8.5 to −6.6)” | “−6.9 (−7.8 to −5.9)” | “−0.7” | “(−2.0 to 0.7)” | “P = .316” |
| Participants with ≥50% reduction in AHI, n (%) | “78 (25.8)” | “62 (19.8)” | “5.7” | “(−1.0 to 12.4)” | “P = .182” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-2"></a>

###### Relative reduction in AHI with 4% desaturation hypopnea rule at week 26, treatment policy estimand

> “These changes represent a model-estimated 44.1% geometric mean reduction in AHI from baseline to week 26 in the AD109 arm vs 17.6% reduction in the placebo arm (P <.0001; Figure 2A).”
> “The median (95% CI) AHI was reduced from 19.8 (18.3-21.2) in the moderate severity range at baseline to 13.3 (10.7-15.9) in the mild severity range at week 26 with AD109, and from 19.1 (17.0-20.2) at baseline to 17.7 (15.3-20.8) at week 26 with placebo (Figure 2B; Table S1).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-3"></a>

###### Week 4 placebo-adjusted AHI change and responder difference, treatment policy estimand

> “The placebo-adjusted reduction at week 4 was −5.7 (−7.5 to −3.8); P <.0001.”
> “The treatment difference at week 4 was 21.7 (14.8-28.5); P <.0001.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-4"></a>

###### Registry primary endpoint result for change in AHI with 4% desaturation hypopnea rule at 6 months, treatment policy and while-on-treatment estimands

| Analysis | AD109, LS mean | Placebo, LS mean | LS mean difference | 95% CI lower limit | 95% CI upper limit | P value |
| --- | --- | --- | --- | --- | --- | --- |
| “Primary Full Analysis Set (Treatment Policy Estimand)” | “-3.28” | “0.71” | “-3.99” | “-6.40” | “-1.59” | “0.001” |
| “While-on-Treatment population” | “-6.11” | “0.44” | “-6.54” | “-8.80” | “-4.28” | “<0.0001” |

Source: [ClinicalTrials.gov NCT05813275 API record](https://clinicaltrials.gov/api/v2/studies/NCT05813275)

> “The While-on-Treatment population is the MITT set, which includes participants in the PFAS who had at least one post-baseline on-treatment polysomnography (PSG) assessment.”
> “Analyses were conducted under the While-on-Treatment Estimand”

Source: [ClinicalTrials.gov NCT05813275 API record](https://clinicaltrials.gov/api/v2/studies/NCT05813275)

<a id="topic-3.1.1.8.3-5"></a>

###### Primary and key secondary endpoints at week 26, on-treatment estimand, AHI with 4% desaturation hypopnea rule

| Endpoint | AD109 | Placebo | Difference | 95% CI | P value |
| --- | --- | --- | --- | --- | --- |
| Change in AHI, LS mean (95% CI), events/h (primary) | “−6.1 (−7.8 to −4.4)” | “0.4 (−1.2 to 2.0)” | “−6.5” | “(−8.8 to −4.3)” | “P <.0001” |
| Change in ODI, LS mean (95% CI), events/h | “−6.5 (−8.4 to −4.6)” | “0.7 (−1.1 to 2.5)” | “−7.2” | “(−9.8 to −4.7)” | “P <.0001” |
| Change in PROMIS-Fatigue T-score, LS mean (95% CI) | “−7.5 (−8.4 to −6.6)” | “−6.2 (−7.0 to −5.3)” | “−1.4” | “(−2.6 to −0.1)” | “P = .030” |
| Percent reduction in geometric mean HB, LS mean (95% CI) | “60.5 (52.8-66.9)” | “14.7 (‒0.7 to 27.8)” | “53.6” | “(41.0-63.6)” | “P <.0001” |
| Change in PROMIS-Sleep Impairment T-score, LS mean (95% CI) | “−7.6 (−8.6 to −6.7)” | “−6.8 (−7.7 to −5.9)” | “−0.8” | “(−2.2 to 0.5)” | “P = .231” |
| Participants with ≥50% reduction in AHI, n (%) | “103 (39.6)” | “66 (22.3)” | “17.3” | “(9.6-25.0)” | “P <.0001” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-6"></a>

###### Relative and median AHI with 4% desaturation hypopnea rule at week 26, on-treatment estimand

> “In the mITT (AD109: n=260; PBO: n=296), the AHI4 (events/hr) was reduced from a median (IQR) of 19.3 (12.7–28.7) at baseline to 10.6 (5.3–20.0) at Week 26 with AD109 and from 19.2 (12.9–29.1) at baseline to 18.0 (9.2–30.5) at Week 26 with PBO.”
> “In the mITT, AD109 significantly reduced AHI4 (LS mean [95% CI]) at Week 26 by −6.1 (−7.8, −4.4) vs 0.4 (−1.2, 2.0) with PBO (p<0.0001), representing a 55.6% improvement (vs 19.1% with PBO; p<0.0001).”

Source: [Strollo et al. (2025c)](https://journal.chestnet.org/article/S0012-3692%2825%2905441-8/fulltext)

<a id="topic-3.1.1.8.3-7"></a>

###### Week 4 placebo-adjusted AHI change and responder difference, on-treatment estimand

> “The placebo-adjusted reduction at week 4 was −6.7 (−8.7 to −4.8); P <.0001.”
> “The treatment difference at week 4 was 27.1 (19.7-34.4); P <.0001.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-8"></a>

###### Relative AHI reduction by estimand as reported by the company

> “SynAIRgy: Treatment Policy Estimand: AD109 significantly reduced AHI by 44.1% vs placebo (17.6%; p<0.0001); On Treatment Estimand: AD109 significantly reduced AHI by 55.6% vs placebo (19.1%; p<0.0001)”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.3-9"></a>

###### Topline primary endpoint result and relative AHI reduction, as reported in the company release

> “The SynAIRgy trial met its primary endpoint: mean change from baseline in the apnea-hypopnea index (AHI) at 26 weeks compared to placebo (p=0.001).”
> “Study participants treated with AD109 achieved a mean reduction in AHI of 55.6% from baseline and achieved significantly improved oxygenation and reduced disease severity”

Source: [Apnimed SynAIRgy topline release (2025)](https://apnimed.com/article/ad109phase3toplineresults/)

<a id="topic-3.1.1.8.3-10"></a>

###### Responders with at least 50% AHI reduction at weeks 4 and 26, treatment policy and on-treatment estimands

> “SynAIRgy: Treatment Policy Estimand: AD109 reduced AHI≥50% in 25.8% of participants vs placebo (19.8%; p=0.182); On Treatment Estimand: AD109 significantly reduced AHI ≥50% in 39.6% of participants vs placebo (22.3%; p<0.0001).”
> “For participants achieving an AHI reduction of > 50%, a supportive treatment policy analysis that treated only missing data as non-response demonstrated statistically significant differences favoring AD109 (week 26 p=0.006), supporting the responder findings despite unequal discontinuation rates and a non-trivial placebo response.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

> “Clinically meaningful response was observed as early as week 4 based on the ≥50% reduction in AHI responder rates, suggesting that the lower treatment difference at week 26 could be attributed to the higher early treatment discontinuation rate with AD109 relative to placebo, since participants who discontinued treatment were classified as nonresponders in this analysis.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-11"></a>

###### Severity category improvement and disease control at week 26, treatment policy estimand

> “AD109 improved OSA disease severity category in 41.8% of participants (vs 33.6% for placebo), with 17.6% achieving complete OSA disease control (AHI <5) at week 26 vs 9.3% for placebo (P = .012; Figure 2C).”
> “Reductions ≥70% in AHI at week 26 were reported in 17.2% (52/302) of the AD109 arm vs 8.9% (28/313) in the placebo arm (P = .003; Figure 3A).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-12"></a>

###### Severity category improvement and disease control at week 26, on-treatment estimand

> “Results from the SynAIRgy trial as published in AJRCCM (on treatment estimand):”
> “22.3% of participants achieved disease control (AHI <5 events/hour)”

Source: [Apnimed SynAIRgy publication release (2026)](https://ir.apnimed.com/news-releases/news-release-details/apnimed-announces-publication-its-phase-3-synairgy-trial-ad109)

> “In the mITT population, AD109 achieved a significant reduction in mean AHI from baseline (55.6%, p<0.0001) compared to placebo (19.1%, p<0.0001) at 26 weeks.”
> “51.2% of participants treated with AD109 showed a reduction in OSA disease severity category”
> “22.3% of participants treated with AD109 achieved complete OSA disease control (defined as AHI <5 events/hour)”

Source: [Apnimed CHEST 2025 release (2025)](https://apnimed.com/article/apnimed-to-present-additional-phase-3-data-for-ad109-an-investigational-oral-pill-for-obstructive-sleep-apnea-at-chest-2025-annual-meeting/)

<a id="topic-3.1.1.8.3-13"></a>

###### Oxygen desaturation index with 3% desaturation criterion, relative change by estimand

> “SynAIRgy: Treatment Policy Estimand: AD109 significantly reduced ODI3 of 34.3% (3.7 events/hr) vs placebo (10.8%; +0.9 events/hr; p=0.001) at week 26; On Treatment Estimand: AD109 significantly reduced ODI3 of 44.0% (6.5 events/hr) vs placebo (11.2%; +0.7 events/hr; p<0.0001) at week 26.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.3-14"></a>

###### Oxygen desaturation index and hypoxic burden, on-treatment estimand, as reported in the CHEST 2025 company release

> “Significant improvements in oxygenation as assessed by mean reductions in HB4 (60.5%, p<0.0001) and ODI3 (44.8%, p<0.0001)”

Source: [Apnimed CHEST 2025 release (2025)](https://apnimed.com/article/apnimed-to-present-additional-phase-3-data-for-ad109-an-investigational-oral-pill-for-obstructive-sleep-apnea-at-chest-2025-annual-meeting/)

<a id="topic-3.1.1.8.3-15"></a>

###### Hypoxic burden with 4% desaturation criterion outside formal hierarchical testing, treatment policy estimand

> “Based on the prespecified hierarchical testing approach for secondary endpoints, formal endpoint testing ended before analysis of HB.”
> “The geometric mean percent reduction in HB at week 26 was 44.7% for participants in the AD109 arm vs 8.5% for participants in the placebo arm (nominal P <.0001; Table 2).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-16"></a>

###### Non-REM, REM, and supine AHI with 4% desaturation hypopnea rule at week 26, on-treatment estimand

| Measure | AD109 change, LS mean (95% CI) | Placebo change, LS mean (95% CI) | Difference | 95% CI | P value |
| --- | --- | --- | --- | --- | --- |
| Non-REM AHI, events/h | “−3.7 (−5.6 to −1.8)” | “0.8 (−1.0 to 2.6)” | “−4.5” | “(−7.1 to −2.0)” | “P <.001” |
| REM AHI, events/h | “−3.6 (−6.7 to −0.5)” | “−0.2 (−2.7 to 2.3)” | “−3.4” | “(−7.4 to 0.6)” | “P = .092” |
| Supine AHI, events/h | “−7.4 (−10.0 to −4.8)” | “−2.5 (−4.9 to −0.1)” | “−4.9” | “(−8.4 to −1.4)” | “P = .006” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-17"></a>

###### Apnea index and hypopnea index at week 26, on-treatment estimand

> “At week 26 in SynAIRgy, AD109 was associated with a 77.2% mean reduction from baseline in the apnea index among patients who remained on treatment (vs 20.2% on placebo), compared with a smaller mean relative reduction of approximately 45.3% in the hypopnea index (vs 12.4% on placebo).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.3-18"></a>

###### Arousal indexes at week 26, on-treatment estimand

| Measure | AD109 change, median (IQR) | Placebo change, median (IQR) |
| --- | --- | --- |
| Respiratory arousal index, events/h | “0.1 (−5.0 to 5.8)” | “1.3 (−3.5 to 7.5)” |
| Total arousal index, events/h | “2.9 (−2.5 to 10.5)” | “1.6 (−3.9 to 8.7)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-19"></a>

###### Subgroups and adjustment for REM sleep and body mass index, treatment policy and on-treatment estimands

> “Similar improvements in AHI at week 26 were observed among participants in the AD109 arm across various subgroups including sex, baseline OSA severity categories (mild, moderate, and severe), and BMI classes (Figure S3).”
> “AD109 treatment effects on AHI were consistent across estimands, with larger effect sizes observed in the supportive on-treatment estimand in participants who tolerated treatment (Table 3, Figure 2D and E, Figure S3).”
> “The evaluation of the treatment effect on AHI and HB at week 26 performed with an adjusted model including the change from baseline in proportion of REM sleep and BMI is provided in Table S3 and shows that the reduction in AHI is independent from reduction in REM and BMI in the treatment arm.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.3-20"></a>

###### Snoring, exploratory endpoint

> “AD109 reduced snoring vs placebo (Table S2).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="heading-3.1.1.8.4"></a>

###### 3.1.1.8.4 Health-related Quality of Life and Patient-reported Outcomes

<a id="topic-3.1.1.8.4-1"></a>

###### PROMIS-Fatigue at week 26, treatment policy and on-treatment estimands

> “AD109 reduced PROMIS-Fatigue T-scores at week 26 by a mean of 7.4 points compared with 6.2 points in the control arm, such that the difference between arms did not meet statistical significance (P = .059; Table 2).”
> “In the supportive on-treatment estimand analyses, PROMIS-Fatigue T-scores fell at week 26 by 7.5 points with AD109 vs 6.2 points with placebo (P = .030; Table 3).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.4-2"></a>

###### PROMIS-Fatigue by baseline fatigue and daytime sleepiness, treatment policy estimand

> “The impact of AD109 on PROMIS-Fatigue T-score at week 26 was more pronounced in participants who were more fatigued (P = .049) or had more daytime sleepiness (P = .007) at baseline (Figure S6).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.4-3"></a>

###### PROMIS-Sleep Impairment at week 26, treatment policy estimand

> “The PROMIS-Sleep Impairment T-score improved by 7.6 points in the AD109 arm but the improvement in the placebo arm was similar, such that no treatment effect was observed between arms (P = .316; Table 2).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.4-4"></a>

###### PROMIS-Fatigue and PROMIS-Sleep Impairment at week 26, on-treatment estimand, congress abstract

> “In the mITT, the change from baseline PROMIS-Fatigue and PROMIS-Sleep Impairment T-scores (LS mean [95% CI]) at Week 26 were −7.5 (−8.4, −6.6) and −7.6 (−8.6, −6.7) with AD109 compared to −6.2 (−7.0, −5.3) and −6.8 (−7.7, −5.9) with PBO, respectively (p=0.03; p=0.23).”

Source: [Strollo et al. (2025c)](https://journal.chestnet.org/article/S0012-3692%2825%2905441-8/fulltext)

<a id="topic-3.1.1.8.4-5"></a>

###### Placebo response in patient-reported outcomes

> “The substantial improvement in symptoms observed in both arms suggests that the placebo effect may have impaired the ability to detect true clinical improvements.”
> “In the context of PRO assessments, participants were never specifically asked if they were aware of receiving placebo, so we cannot rule out the possibility of expectation bias influencing responses.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “In the SynAIRgy clinical trial, although we did not observe a greater improvement in fatigue overall with AD109 which may have been due to a high placebo effect, in participants with more EDS symptoms at baseline, AD109 led to improvements in symptoms of fatigue relative to placebo.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.4-6"></a>

###### Editorial comment on patient-reported outcomes

> “Notably, no overall significant improvements were observed in patient-reported outcomes, such as fatigue, in the background of overall minimal baseline symptoms.”
> “However, among participants with greater baseline symptom burden of fatigue and sleepiness, improvements in Patient-Reported Outcomes Measurement Information System (PROMIS) Fatigue scores were observed, suggesting that symptomatic benefit may be confined to selected subgroups.”

Source: [Mehra and Mokhlesi (2026)](https://academic.oup.com/ajrccm/article/212/7/1427/8679096)

<a id="topic-3.1.1.8.4-7"></a>

###### Epworth Sleepiness Scale and patient global impression endpoints

No evidence found.

<a id="heading-3.1.1.8.5"></a>

###### 3.1.1.8.5 Safety Results

<a id="topic-3.1.1.8.5-1"></a>

###### Overall safety summary, safety analysis set

| Event | AD109 (n = 325), n (%) | Placebo (n = 319), n (%) |
| --- | --- | --- |
| Any TEAE | “230 (70.8)” | “149 (46.7)” |
| Any treatment-related TEAE | “190 (58.5)” | “75 (23.5)” |
| Any severe TEAE | “15 (4.6)” | “10 (3.1)” |
| Any serious TEAE | “5 (1.5)” | “8 (2.5)” |
| Any TEAE leading to treatment discontinuation | “69 (21.2)” | “10 (3.1)” |
| Dry mouth | “115 (35.4)” | “29 (9.1)” |
| Insomnia (initial, middle, or unspecified) | “67 (20.6)” | “15 (4.7)” |
| Nausea | “37 (11.4)” | “2 (0.6)” |
| Urinary hesitation | “29 (8.9)” | “1 (0.3)” |
| Constipation | “17 (5.2)” | “4 (1.3)” |
| Somnolence | “17 (5.2)” | “10 (3.1)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-2"></a>

###### Treatment discontinuation due to adverse events

> “Overall, 21.2% (69/325) of participants receiving AD109 and 3.1% (10/319) of participants receiving placebo discontinued treatment due to an AE (Table 6).”
> “The primary (occurring in ≥7 participants) AEs leading to AD109 discontinuation were insomnia, urinary hesitation, and nausea (Table 6).”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-3"></a>

###### Adverse events leading to treatment discontinuation in two or more participants

| Event | AD109 (n = 325), n (%) |
| --- | --- |
| Insomnia | “16 (4.9)” |
| Urinary hesitation | “8 (2.5)” |
| Nausea | “7 (2.2)” |
| Erectile dysfunction or sexual dysfunction | “5 (1.5)” |
| Dry mouth | “4 (1.2)” |
| Hypertension | “3 (0.9)” |
| Palpitations | “3 (0.9)” |
| Upper abdominal pain | “2 (0.6)” |
| Feeling jittery | “2 (0.6)” |
| Urinary retention | “2 (0.6)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-4"></a>

###### Severity of adverse events and serious adverse events

> “The most common AEs were generally mild in severity (Table S4).”
> “Serious AEs were reported by 1.5% (5/325) of participants who received AD109 and 2.5% (8/319) of participants who received placebo; all were deemed by investigators to be unrelated to study medication (Table 6).”
> “In the AD109 arm, 5 participants experienced the following serious adverse events (all unrelated to AD109 treatment): 1 participant had cholelithiasis and cholestatic hepatitis, 1 participant had gastritis and bacterial pneumonia, 1 participant had asthma and chronic obstructive pulmonary disease, 1 participant had cholecystitis, and 1 participant had malignant melanoma.”
> “No deaths were reported in the trial.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-5"></a>

###### Registry adverse events at the 5% reporting threshold, safety population

| Event | AD109, affected/at risk (%) | Placebo, affected/at risk (%) |
| --- | --- | --- |
| Serious adverse events, total | “5/325 (1.54%)” | “8/319 (2.51%)” |
| Other adverse events, total | “176/325 (54.15%)” | “50/319 (15.67%)” |
| Dry mouth | “115/325 (35.38%)” | “29/319 (9.09%)” |
| Nausea | “37/325 (11.38%)” | “2/319 (0.63%)” |
| Constipation | “17/325 (5.23%)” | “4/319 (1.25%)” |
| Somnolence | “17/325 (5.23%)” | “10/319 (3.13%)” |
| Middle insomnia | “38/325 (11.69%)” | “10/319 (3.13%)” |
| Initial insomnia | “30/325 (9.23%)” | “4/319 (1.25%)” |
| Urinary hesitation | “29/325 (8.92%)” | “1/319 (0.31%)” |

Source: [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results)

<a id="topic-3.1.1.8.5-6"></a>

###### Any adverse event counts in the congress abstract

> “Adverse events (AEs) were predominantly mild with ≥1 AE reported by 232 (71.4%) participants with AD109 and 151 (47.3%) with PBO.”
> “No treatment-related serious AEs were identified.”

Source: [Strollo et al. (2025c)](https://journal.chestnet.org/article/S0012-3692%2825%2905441-8/fulltext)

<a id="topic-3.1.1.8.5-7"></a>

###### Heart rate and blood pressure, safety analysis set

| Measure | AD109, mean change (SD) [n] | Placebo, mean change (SD) [n] |
| --- | --- | --- |
| Heart rate, beats/min, baseline day 1 to week 26 day 1 | “1.5 (9.8) [268]” | “−1.9 (10.1) [277]” |
| Heart rate, beats/min, baseline day 2 to week 26 day 2 | “3.6 (9.4) [267]” | “0.0 (9.2) [276]” |
| Systolic blood pressure, mm Hg, baseline day 1 to week 26 day 1 | “1.4 (11.8) [268]” | “1.7 (11.9) [277]” |
| Diastolic blood pressure, mm Hg, baseline day 1 to week 26 day 1 | “0.9 (8.0) [268]” | “−0.5 (7.7) [277]” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-8"></a>

###### Heart rate of 100 beats per minute or higher at any time on treatment

> “Changes in blood pressure and heart rate were small, although a small subgroup of participants showed tachycardia while on AD109.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

| Measure | AD109 (n = 325), n (%) | Placebo (n = 319), n (%) | P value |
| --- | --- | --- | --- |
| Heart rate ≥100 beats/min, on study | “20 (6.3)” | “6 (1.9)” | “.006” |
| Systolic blood pressure ≥140 mm Hg, on study | “83 (26.3)” | “82 (26.5)” | “.939” |
| Diastolic blood pressure ≥90 mm Hg, on study | “65 (20.6)” | “64 (20.7)” | “.956” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-9"></a>

###### Sleep architecture at week 26, post hoc analysis in the safety analysis set

| Measure | AD109 change, median (IQR) | Placebo change, median (IQR) |
| --- | --- | --- |
| Total sleep time, min | “−8.8 (−52.1 to 35.0)” | “7.5 (−27.2 to 40.0)” |
| Sleep efficiency, % | “−1.9 (−10.3 to 7.8)” | “0.6 (−5.9 to 8.1)” |
| Stage N1, % | “3.0 (−1.7 to 7.8)” | “−0.3 (−3.8 to 3.8)” |
| Stage N2, % | “7.3 (−0.8 to 13.8)” | “2.2 (−4.4 to 10.0)” |
| Stage N3, % | “−1.6 (−6.1 to 0.6)” | “−1.3 (−6.1 to 0.8)” |
| Stage REM, % | “−7.5 (−13.6 to −0.9)” | “−0.2 (−5.0 to 5.2)” |

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “Small changes in sleep architecture were observed consistent with the higher prevalence of insomnia.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-10"></a>

###### Body mass index change at week 26

> “Changes in BMI at week 26 were small across treatment groups; median change was −0.1 kg/m2 (IQR, −1.1 to 0.6) with AD109 and 0.1 kg/m2 (IQR, −0.4 to 0.9) with placebo.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

<a id="topic-3.1.1.8.5-11"></a>

###### Insomnia as a reason for discontinuation and baseline insomnia assessment

> “The population studied must have been intolerant to or refused PAP therapy to enroll; thus, the trial may have been enriched with participants who may have exhibited some degree of subclinical insomnia.41 However, insomnia was not assessed in a standardized manner in this trial.”

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/)

> “Insomnia was the main reason for discontinuation in 4.9% of participants in SynAIRgy and 3% of participants in LunAIRo.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.1.8.5-12"></a>

###### Editorial comment on tolerability and long-term safety

> “Adverse events, primarily anticholinergic and stimulant-related, including dry mouth, insomnia, and urinary symptoms, were common, and treatment discontinuation occurred in 21.2% of participants.”
> “The long-term cognitive safety of chronic antimuscarinic therapy remains uncertain, with observational studies linking cumulative anticholinergic exposure to dementia risk, although causality has not been established.”

Source: [Mehra and Mokhlesi (2026)](https://academic.oup.com/ajrccm/article/212/7/1427/8679096)

<a id="heading-3.1.1.9"></a>

##### 3.1.1.9 Study Limitations

<a id="topic-3.1.1.9-1"></a>

###### Summary

- Design: 26 weeks of treatment against placebo, with polysomnography at baseline and weeks 4 and 26; no active comparator. Dose reduction was not permitted, and participants could start PAP during the trial.
- Population: adults intolerant of or refusing PAP, with a PROMIS-Fatigue T-score above 50 and a BMI cap of 40 (men) or 42 (women) kg/m2. Exclusions included clinically significant cardiovascular disease, blood pressure above 145/90 mmHg, bothersome insomnia, specified psychiatric disorders, and narrow angle glaucoma. Sites were limited to the United States and Canada, which the authors state may not represent OSA in other geographies.
- Eligibility AHI changed during enrollment: protocol version 2.0 (July 2023) required more than 5 to 45 events/h, version 5.0 (February 2024) required 10 to 45, and the publication states participants with AHI of 5 or more were enrolled first. The 2024 congress abstract described AHI4 of 5 or more, and the 2025 abstract AHI4 above 5 with no upper limit. Enrollment ran from December 20, 2023 to March 26, 2025.
- Severity mix: enrollment limits capped each AHI4 stratum at about 20% to 40%, and the authors state the population skewed toward mild-to-moderate disease, which can limit absolute effects; the editorial gives a median baseline AHI of 19.6.
- Endpoint change: protocol version 4.0 (December 2023) named the proportion with at least 50% AHI reduction as the primary endpoint. Amendment 6.0 (17 April 2025) reordered the primary and key secondary endpoints “to align with recent regulatory approvals for the treatment of OSA related to obesity”. The registry version of 2025-03-18 named the proportion of participants with reduction in AHI as the primary outcome, and the next version, submitted 2025-11-18, changed the outcome measures. The amendment postdates the registry's actual primary completion date (2025-03-10). Statistical analysis plan version 2.0 (22 April 2025) added the relative change in AHI on the log scale as an exploratory analysis.
- Estimands and missing data: the primary treatment policy analysis assumed data missing not at random (jump-to-reference multiple imputation). The supportive on-treatment estimand excluded data collected after treatment discontinuation and imputed missing values by last observation carried forward; it included 260 participants on AD109 and 296 on placebo, against 302 and 313 in the treatment policy set. The topline release reported the on-treatment relative reduction (55.6%) beside the treatment policy primary result (p=0.001).
- Analysis-set exclusions: 22 participants (16 AD109, 6 placebo) at site 1416, terminated for Good Clinical Practice noncompliance, and 7 participants in the exploratory GLP-1 receptor agonist cohort were excluded from efficacy analyses; that cohort's analyses were not reported because of low enrollment.
- Discontinuation: treatment discontinuation before 26 weeks was 34.3% with AD109 and 16.3% with placebo in the publication, which the authors state likely diminished the observed benefit; the prospectus reports 34.6% and 16.5% at month 6 in the main cohort. Treatment discontinuation due to adverse events was 21.2% and 3.1%; study withdrawal due to adverse events in the registry was 4 and 1 participants.
- Hierarchical testing: under the treatment policy estimand, PROMIS-Fatigue (P = .059) did not reach significance, so formal testing ended before hypoxic burden, and the responder endpoint was not significant (P = .182).
- Patient-reported outcomes: the authors cite a placebo response in both arms and possible expectation bias, and state insomnia was not assessed in a standardized manner although the PAP-intolerant population may have been enriched for subclinical insomnia.
- Clinical meaningfulness: at the pre-NDA meeting the FDA gave feedback on the clinical meaningfulness of the primary endpoint results and the limitations of the PROs, and indicated additional analyses and justification may be needed. The editorial states the clinical significance of the placebo-adjusted reduction of about 4 events per hour is uncertain as a continuous measure, and that durability, long-term safety, and effects on cardiovascular events require further study.
- Sponsor: the prospectus describes SynAIRgy as a trial the company conducted.

Source: [Strollo et al. (2026c)](https://pmc.ncbi.nlm.nih.gov/articles/PMC13318230/); [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/); [Apnimed SynAIRgy protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/Prot_000.pdf); [Apnimed SynAIRgy statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/75/NCT05813275/SAP_001.pdf); [ClinicalTrials.gov NCT05813275](https://clinicaltrials.gov/study/NCT05813275); [ClinicalTrials.gov NCT05813275 version 22](https://clinicaltrials.gov/study/NCT05813275?tab=history&a=22); [ClinicalTrials.gov NCT05813275 results](https://clinicaltrials.gov/study/NCT05813275?tab=results); [Strollo et al. (2024)](https://academic.oup.com/sleep/article/47/Supplement_1/A233/7654816); [Strollo et al. (2025b)](https://academic.oup.com/sleep/article/48/Supplement_1/A339/8134797); [Apnimed SynAIRgy topline release (2025)](https://apnimed.com/article/ad109phase3toplineresults/); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm); [Mehra and Mokhlesi (2026)](https://academic.oup.com/ajrccm/article/212/7/1427/8679096)

<a id="heading-3.1.2"></a>

#### 3.1.2 LunAIRo

<a id="heading-3.1.2.1"></a>

##### 3.1.2.1 Objective

<a id="topic-3.1.2.1-1"></a>

###### Primary and secondary objectives in the design paper

> “The primary objective of the LunAIRo and SynAIRgy trials is to demonstrate the efficacy of AD109 versus placebo for upper airway obstruction, as measured by the AHI4, among individuals across the full range of OSA severity, from mild to very severe, over a 26-week treatment period.”
> “Additional objectives for the LunAIRo trial include exploring both long-term efficacy and safety endpoints over a 51-week placebo-controlled period, and exploring time points as early as 4 weeks to evaluate rapid onset of efficacy of AD109.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.1-2"></a>

###### Study rationale and purpose in the protocol

> “The present study, LunAIRo (APC-APN-304), is a Phase 3 study designed to examine the efficacy and safety of AD109 compared to placebo at 6-months and 1 year.”
> “The primary purpose of conducting this study is to show that AD109 is safe and superior to placebo for airway obstruction (AHI4).”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.1-3"></a>

###### Brief summary in the ClinicalTrials.gov record

> “This is a Phase 3 Randomized Double-Blind Placebo-Controlled 1-year Parallel-Arm Study to Compare a Fixed Dose Combination of AD109 to Placebo in Obstructive Sleep Apnea.”

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="heading-3.1.2.2"></a>

##### 3.1.2.2 Location and Study Date

<a id="topic-3.1.2.2-1"></a>

###### Study dates in the ClinicalTrials.gov record

> “Study Start (Actual) 2023-08-28”
> “Primary Completion (Actual) 2025-05-07”
> “Study Completion (Actual) 2025-05-07”
> “Results First Posted 2026-11-03”

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="topic-3.1.2.2-2"></a>

###### Number of locations in the ClinicalTrials.gov record

> “This study has 64 locations”

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="topic-3.1.2.2-3"></a>

###### Sites in the United States

> “A total of 64 sites have been identified in the U.S. for LunAIRo (Supplemental Table 1) and 73 sites have been identified in both the U.S. and Canada for SynAIRgy (Supplemental Table 2).”
> “LunAIRo and SynAIRgy were performed in different centers, with the exception of a small number of sites that enrolled for SynAIRgy after the enrollment of LunAIRo was closed. No participant was enrolled in both studies.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

> “The trial enrolled 660 adult participants from 64 centers in the US.”

Source: [Apnimed LunAIRo topline release (2025)](https://apnimed.com/article/ad109toplinephase3results/)

<a id="topic-3.1.2.2-4"></a>

###### Trial initiation and completion in the prospectus

> “We initiated the LunAIRo and SynAIRgy Phase 3 clinical trials for AD109 in September and November of 2023, respectively.”
> “LunAIRo was conducted in the United States and SynAIRgy was conducted in the United States and Canada.”
> “Both LunAIRo and SynAIRgy trials were completed in 2025 and the OLE trial was completed in 2026.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.3"></a>

##### 3.1.2.3 Study Design

<a id="topic-3.1.2.3-1"></a>

###### Design, randomization, and duration in the protocol

> “The LunAIRo Study (APC-APN-304) is a randomized, double blind, placebo-controlled, 1-year parallel-arm study of AD109 vs placebo in participants with mild to severe OSA who are intolerant of, or currently refuse, PAP.”
> “The first week of dosing will be with a lower dose (run-in) of AD109, or placebo, followed by the target dose taken 50-weeks, then a two-week period without dosing prior to the End of Study visit at 53 weeks.”
> “No interim analysis is planned.”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.3-2"></a>

###### Design details in the ClinicalTrials.gov record

| Field | Quoted record |
| --- | --- |
| Official title | “Phase 3 Randomized Double-Blind Placebo-Controlled 1-year Parallel-Arm Study to Compare a Fixed Dose Combination of Aroxybutynin/Atomoxetine (AD109) to Placebo in Obstructive Sleep Apnea (LunAIRo Study)” |
| Other study ID | “APC-APN-304” |
| Allocation | “Randomized” |
| Interventional model | “Parallel Assignment” |
| Masking | “Quadruple (ParticipantCare ProviderInvestigatorOutcomes Assessor)” |
| Enrollment | “Enrollment (Actual) 660” |

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="topic-3.1.2.3-3"></a>

###### Stratified randomization and severity strata

> “Randomization is stratified according to baseline AHI4 (<15, 15 to 30, >30) and 8-item PROMIS-Fatigue raw score (<25 vs ≥ 25; a raw score of 25 is equivalent to a T-score of 58.5).”
> “To ensure representative distribution of baseline AHI4 severity, the size of AHI4 strata will be controlled through enrollment limits, with AHI4 <15 representing ∼20–40% of participants, AHI4 15–30 representing ∼20–40% of participants, and AHI4 >30 representing ∼20–40% of participants.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

> “Our trials were designed to potentially support a broader label indicated for patients across the range of OSA severities by ensuring no more than 40% of the overall study population in each trial could represent mild, moderate or severe patient subgroups.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.3-4"></a>

###### Polysomnography schedule and central scoring

> “Objective efficacy data are gathered with in-laboratory polysomnograms (PSG) performed at 13, 26, and 51 weeks for the LunAIRo study, and at 4 and 26 weeks for the SynAIRgy study.”
> “The studies are conducted with the support of a PSG scoring center (Sleep Strategies, Ottawa, ON) responsible for centralized PSG scoring blinded to treatment assignment for both screening and outcome assessment and generation of standardized PSG variables.”
> “Each participant's baseline and on-treatment PSGs were evaluated by the same scorer to maintain consistency.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.3-5"></a>

###### Ambulatory blood pressure monitoring and blinding

> “Eligible participants in the LunAIRo trial, after V2, will be encouraged to undergo optional 24 h ambulatory blood pressure monitoring (ABPM) before the randomization visit (V3). The ABPM procedure is repeated after 4 weeks of treatment with AD109 or placebo.”
> “All study participants, investigators, site staff, staff at the PSG scoring center, staff at the CRO and sponsor staff are blinded to treatment assignment throughout the studies.”
> “An independent DMC, with expertise in sleep apnea, clinical trials, and biostatistics meets regularly to assess unblinded safety data and make recommendations for both trials.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.3-6"></a>

###### Open-label extension after the trial

> “Participants completing either of these studies are eligible for an optional open label extension (OLE) study of AD109 in which safety is assessed for up to an additional year (Supplement Information).”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

Additional sources: [Patel et al. (2024)](https://www.atsjournals.org/doi/10.1164/ajrccm-conference.2024.209.1_MeetingAbstracts.A4551)

<a id="heading-3.1.2.4"></a>

##### 3.1.2.4 Eligibility Criteria

<a id="topic-3.1.2.4-1"></a>

###### Inclusion and exclusion criteria in the ClinicalTrials.gov record

> “≥18 years of age at the time of informed consent.”
> “PSG criteria: AHI of >5; ≤ 25% central or mixed apneas; and PLM arousal index ≤15”
> “PROMIS-Fatigue: raw score ≥17”
> “PAP intolerance or current PAP refusal.”
> “BMI between 18.5 and 40 kg/m2 for men, or 42 kg/m2 for women, inclusive.”
> “Insomnia disorder characterized by difficulty initiating or maintaining sleep, or use within the past month of sedative-hypnotics or other medication for the purpose of treating or avoiding insomnia symptoms.”
> “Clinically significant cardiac disease, e.g., ventricular arrhythmia, untreated or unstable coronary artery disease, cardiac failure. Stable atrial arrhythmia is allowed.”

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="topic-3.1.2.4-2"></a>

###### Apnea-hypopnea index range and PAP criteria in protocol version 4.0

> “Protocol Version: 4.0 11 June 2025”
> “AHI4 (Hypopneas defined by 4% oxygen desaturation) of >5 and ≤45”
> “PAP failure is defined as no PAP use for ≥3 months prior to randomization, or return/removal of PAP device from the home.”
> “PAP refusal is defined as refusal of PAP after prior positive sleep study, or prior refusal of provider-recommended sleep study due to unwillingness to consider PAP.”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.4-3"></a>

###### Apnea-hypopnea index range in the baseline abstract

> “Participants were enrolled over the full range of OSA severity as measured by apnea-hypopnea index with 4% desaturation (AHI4) >5 with no upper limit.”

Source: [Patel et al. (2025a)](https://doi.org/10.1164/ajrccm.2025.211.Abstracts.A6932)

> “Since the mechanism of action should apply, in theory, to all OSA patients, no limit to OSA severity is implemented initially, however caps could be implemented during the trial to ensure that the full range of OSA severities are well-represented in the trials.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.4-4"></a>

###### Key exclusion criteria and fatigue threshold in the design paper

> “Current bothersome symptoms of insomnia (difficulty initiating or maintaining sleep, as distinct from unrefreshing sleep or other symptoms attributable to OSA). Participants treated with insomnia can be enrolled if symptoms are controlled”
> “Schizophrenia, schizoaffective disorder, generalized anxiety disorder or bipolar disorder according to DSM-5 criteria”
> “Attempted suicide within 1 year prior to screening, or current suicidal ideation.”
> “Current bothersome symptoms of bladder outlet obstruction including difficulty initiating or maintaining urinary flow, straining, or sensation of incomplete bladder emptying”
> “Blood pressure >145/90 mmHg”
> “Narrow angle glaucoma”
> “Only participants who refuse or are intolerant to PAP and with a raw PROMIS-Fatigue score ≥17 (fatigue level above average of the population; a raw score of 17 is equivalent to a T-score of 50.4) are eligible.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.4-5"></a>

###### Disallowed concomitant medications

> “Atomoxetine is metabolized by CYP2D6 and aroxybutynin by CYP3A4, and consequently, strong pharmacologic inhibitors of these metabolic enzymes are not allowed as concomitant therapy.”
> “MAOIs, NRIs, and TCAs,”
> “Strong CYP2D6 inhibitors, and strong CYP3A4 inhibitors”
> “Drugs used exclusively for weight loss”
> “Modafinil, armodafinil, solriamfetol, and pitolisant”
> “Oral anticholinergics, and cholinesterase inhibitors”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="heading-3.1.2.5"></a>

##### 3.1.2.5 Treatment

<a id="topic-3.1.2.5-1"></a>

###### Treatment arms in the ClinicalTrials.gov record

| Arm | Description in the record |
| --- | --- |
| Experimental: AD109 | “Aroxybutynin 2.5 mg/atomoxetine 75 mg tablet, administered orally once daily at bedtime (Weeks 2-51). Week 1 run-in: aroxybutynin 2.5 mg/atomoxetine 37.5 mg tablet.” |
| Placebo Comparator: Placebo | “Matching placebo tablet administered orally once daily at bedtime for 51 weeks, with run-in and target dose matched to active arm packaging.” |

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

<a id="topic-3.1.2.5-2"></a>

###### No dose reduction, and continued follow-up after discontinuation

> “Dose reduction is not permitted. Dosing is discontinued if the study drug to which the participant is randomized is not tolerated. Note that participants who discontinue dosing are encouraged to continue all other study activities for the full duration of the study.”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.5-3"></a>

###### PAP initiation permitted during the trial

> “Participants are not prohibited from initiating PAP during the trial if deemed by the participant's primary care provider to be in the participant's best medical interest. Dosing of AD109 or placebo can continue if PAP is initiated.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.5-4"></a>

###### Dispensing of the run-in and target doses

> “The IMP is dispensed as a blister package of 7 run-in dose tablets and a bottle containing 33 target dose tablets.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.5-5"></a>

###### Run-in dose and dose reduction in the prospectus

> “A lower dose of AD109 (2.3/37.5mg) was administered during the one-week run-in period; however, dose reduction was not permitted in the Phase 3 trials, to maximize evaluation of efficacy.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.6"></a>

##### 3.1.2.6 Study Outcomes

<a id="heading-3.1.2.6.1"></a>

###### 3.1.2.6.1 Rationale for Using Apnea-Hypopnea Index as the Primary Endpoint

<a id="topic-3.1.2.6.1-1"></a>

###### Primary endpoint: change from baseline in AHI4 at week 26

> “Change from baseline in apnea-hypopnea index based on 4% hypopnea desaturation (AHI4) at week 26”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

> “AHI, Apnea-Hypopnea Index, is the number of apneas or hypopneas recorded via polysomnography during the study per hour of sleep. Apnea is defined as a cessation of airflow lasting at least 10 seconds, hypopnea as a decrease in airflow by at least 30% from baseline for at least 10 seconds occurring with a drop in oxygen saturation (SpO2) by at least 4%.”

Source: [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247)

> “The primary efficacy endpoint for both trials is the change from baseline in AHI4 at Week 26.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.6.1-2"></a>

###### Protocol amendment 4.0: reordering of the primary and key secondary endpoints, June 2025

> “Protocol Version: 4.0 11 June 2025”
> “Version 4.0 June 11, 2025 As detailed in the following Summary of Changes, Apnimed has introduced modifications to the protocol V3.0 (dated December 11, 2023). Content changes are specified in the table found in Section, 9.6, Appendix 6 with identification of the section in the new protocol.”
> “The order of endpoints, including the primary endpoint and key secondary endpoints, was changed to align with recent regulatory approvals for the treatment of OSA related to obesity.”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.6.1-3"></a>

###### Primary endpoint: proportion with at least 50% reduction in AHI4 at six months in the baseline abstract, May 2025, superseded in June 2025

> “The primary efficacy endpoint is the proportion of participants with ≥50% reduction in AHI4 at six months.”

Source: [Patel et al. (2025a)](https://doi.org/10.1164/ajrccm.2025.211.Abstracts.A6932)

<a id="topic-3.1.2.6.1-4"></a>

###### Key secondary and secondary endpoints, in the order tested

> “Change from baseline in ODI3 at week 26”
> “Change from baseline in HB4 (on the natural log scale) at week 26”
> “Proportion of participants with ≥50% reduction in AHI4 at 26 weeks”
> “Change from baseline in PGI-S for fatigue at week 26”
> “Change from baseline in ESS at week 26”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.6.1-5"></a>

###### Apnea-hypopnea index as the historical principal biomarker, and hypoxic burden as a complementary metric

> “The primary efficacy endpoint in most OSA trials is the AHI which historically has been the principal biomarker for the disease”
> “Compared to AHI4, that only considers the frequency of obstructive events (events/hour of sleep), the HB4 also incorporates the length and the depth of the apnea/hypopnea-related chronic intermittent hypoxia during sleep [16]. This metric has demonstrated a stronger correlation with cardiovascular mortality than AHI in recent epidemiological cohort analyses [16].”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.6.1-6"></a>

###### PROMIS-Fatigue as a key secondary endpoint

> “Fatigue is one of the most common patient-reported symptoms in relationship to sleep apnea and disturbed sleep, together with sleepiness [18].”
> “PROMIS-Fatigue may provide increased sensitivity to symptom improvement in OSA [31], particularly in the substantial proportion of OSA patients without EDS [32].”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.6.1-7"></a>

###### Company interpretation of absolute and relative AHI reductions in a mostly mild-to-moderate population

> “In populations with mild and moderate OSA, meaningful clinical benefits are better reflected by relative reductions in AHI and HB, responder analyses, and shifts to lower disease severity categories, rather than by absolute reductions alone.”
> “To account for this, Phase 3 clinical result analyses for AD109 were based on a prespecified proportional (log-scale) modeling approach, which assumes a consistent relative clinical response across the range of baseline AHI values.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.6.1-8"></a>

###### FDA pre-NDA feedback on the clinical meaningfulness of the primary endpoint results and the limits of the patient-reported outcomes

> “In connection with the FDA’s review of our top-line Phase 3 data, at the pre-NDA meeting, the FDA provided feedback about the clinical meaningfulness of certain endpoint results, including the primary endpoint and the limitations of the patient-reported outcomes (PROs). The FDA indicated that additional analyses and justification may be needed to support that the data demonstrates a clinically meaningful benefit for the treatment of OSA.”
> “If we are unable to address these concerns to the FDA’s satisfaction, the FDA may determine that the clinical response results from our Phase 3 trials are not clinically meaningful to establish substantial evidence of effectiveness for Oxnimbi in OSA, or that the overall risk-benefit profile of Oxnimbi does not support approval or it may require additional data, analyses or studies to support approval.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.7"></a>

##### 3.1.2.7 Statistical Analysis Description

<a id="heading-3.1.2.7.1"></a>

###### 3.1.2.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.2.7.1-1"></a>

###### Planned sample size and power

> “A total of 640 participants (320/group) will provide >98% power to detect a treatment difference of ≥ 4 points between AD109 and placebo for the change in AHI4.”
> “Based on the observed dropout rates in MARIPOSA, it is assumed that the dropout rate in the placebo group will be 5% and the dropout rate in the AD109 group will be 25%.”
> “With these assumptions, simulation demonstrated that 640 participants (320/group) would provide approximately 90% power to detect a treatment difference of 2.5 points and approximately 95% power to detect a treatment difference of 2.75 points.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.7.1-2"></a>

###### Participants randomized

> “Eligible participants (N=660) were randomized (AD109: n=329; PBO: n=331) with 524 (80.0%) completing the study.”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.7.1-3"></a>

###### Participants analyzed for the primary endpoint, by estimand

| Analysis set | AD109 | Placebo |
| --- | --- | --- |
| “Full Analysis Set (Treatment Policy Estimand)” | “324” | “326” |
| “While-on-Treatment population” | “240” | “293” |

Source: [ClinicalTrials.gov NCT05811247 API record](https://clinicaltrials.gov/api/v2/studies/NCT05811247)

> “In the mITT (AD109: n=240; PBO: n=293)”
> “In the ITT (AD109: n=324; PBO: n=326)”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.7.1-4"></a>

###### Participants in the safety analysis

> “Adverse Event Data was collected in the Safety Population, which includes all randomized participants who received at least 1 dose of IMP. Participants are summarized based on actual treatment.”

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="heading-3.1.2.7.2"></a>

###### 3.1.2.7.2 Description of Analysis Sets

<a id="topic-3.1.2.7.2-1"></a>

###### Analysis sets in the statistical analysis plan

> “The Full Analysis Set (FAS) consists of all participants who were randomized and received at least 1 dose of investigational medicinal product (IMP). All participants enrolled at Site 1308 will be excluded.”
> “The Modified Intent to Treat (MITT) set for AHI consists of all participants and received at least 1 dose of IMP and had one post baseline on treatment PSG assessment.”
> “The Safety Set (SS) consists of all participants who are randomized and receive at least 1 dose of IMP. Participants will be analyzed according to treatment received.”
> “The Per Protocol Set for AHI4 (PPS-A) consists of all participants in the FAS who completed the study treatment up to week 26 (i.e., visit 6) with at least 70% compliance of taking IMP and have non-missing AHI4 assessment at week 26, without any important protocol violations that could influence the efficacy assessments.”

Source: [Apnimed LunAIRo statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/SAP_001.pdf)

<a id="topic-3.1.2.7.2-2"></a>

###### Analysis populations in the ClinicalTrials.gov record

> “The Full Analysis Set (FAS) population included all randomized participants who received at least one dose of investigational medicinal product (IMP). Participants enrolled at Site 1308 were excluded following site termination. The While-On-treatment population was the MITT set and includes participants in the FAS who had at least one post-baseline on-treatment polysomnography (PSG) assessment.”

Source: [ClinicalTrials.gov NCT05811247 API record](https://clinicaltrials.gov/api/v2/studies/NCT05811247)

<a id="topic-3.1.2.7.2-3"></a>

###### Treatment policy estimand for the primary endpoint

> “Treatment condition: investigational treatment with AD109/Placebo regardless of adherence to investigational product and of any changes in background therapy (treatment policy strategy).”
> “Rationale for estimand: The primary estimand aims at reflecting how participants with mild to severe OSA, are treated in clinical practice and considers both safety and efficacy.”
> “Given the anticipated unequal distribution of missing data between the AD109 and Placebo group, missing data will be assumed to be missing not at random (MNAR) and the endpoint will be analyzed using a pattern- mixture model using a control-based approach (further details are provided in the SAP).”

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf)

<a id="topic-3.1.2.7.2-4"></a>

###### While-on-treatment estimand as a supportive analysis

> “The analysis of the primary efficacy endpoint of change from baseline in AHI4 at week 26 will be repeated using FAS and MITT set and a while on treatment estimand approach under Hypothetical Strategy. For FAS analysis all data collected after a participant has discontinued treatment will be set to missing while for MITT analysis all data collected after a participant has discontinued treatment will be imputed by last available on treatment value.”

Source: [Apnimed LunAIRo statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/SAP_001.pdf)

> “The on-treatment estimand, a modified intent-to-treat population, focused on outcomes observed during active treatment exposure and excluded data collected after treatment discontinuation, with missing values imputed using a last-observation-carried-forward approach.”
> “We believe the on treatment estimand provided the most clinically relevant estimate of the clinical response and complemented the primary treatment policy estimand by focusing specifically on the direct clinical response during active medication use, rather than incorporating the effects of treatment discontinuation patterns that might have varied between groups.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.7.2-5"></a>

###### Statistical model for the primary endpoint

> “The primary analysis methodology will utilize a restricted maximum likelihood (REML) based mixed model for repeated measurements (MMRM) in combination with the Newton Raphson Algorithm. Analyses will include treatment, AHI4 strata, time and the treatment by time interaction as a fixed effects and baseline score as a covariate.”
> “The proportion of participants with ≥50% reduction in AHI4 at Week 26 will be compared between treatment groups using the stratified Cochran-Mantel Haenszel procedure, controlling for randomization strata.”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.7.2-6"></a>

###### Multiplicity control for the key secondary endpoints

> “Adjustment for multiplicity of the key secondary endpoints will be conducted through a closed testing approach, where each key secondary endpoint will be tested in pre-specified order as follows.”
> “Change from baseline to week 26 in ODI3”
> “Change from baseline to week 26 in HB4”
> “Proportion of patients with ≥50% reduction in AHI4 at week 26”
> “Change from baseline to week 26 in PROMIS-Fatigue”
> “No adjustment for multiplicity will be made for the non-key secondary endpoints or tertiary endpoints or exploratory endpoints.”

Source: [Apnimed LunAIRo statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/SAP_001.pdf)

<a id="topic-3.1.2.7.2-7"></a>

###### Week 51 endpoints analyzed as tertiary endpoints

> “Compare efficacy on airway obstruction and OSA symptoms of AD109 vs placebo in participants with mild to severe OSA at 51 weeks”
> “Tertiary Endpoints The following are assessed as change from baseline at week 51 or as specified”
> “The tertiary efficacy endpoints will be analyzed for the FAS and MITT, using the same methods as outlined for the primary endpoint as appropriate.”

Source: [Apnimed LunAIRo statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/SAP_001.pdf)

<a id="topic-3.1.2.7.2-8"></a>

###### Pooled phase 3 analysis: estimands and analysis sets

> “Pooled primary efficacy at Week 26 was analyzed in the treatment policy estimand (randomized and received ≥1 treatment dose) with supportive analyses in the on-treatment estimand (≥1 post-baseline on-treatment polysomnogram).”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="heading-3.1.2.8"></a>

##### 3.1.2.8 Results

<a id="heading-3.1.2.8.1"></a>

###### 3.1.2.8.1 Participant Disposition

<a id="topic-3.1.2.8.1-1"></a>

###### Participant flow in the ClinicalTrials.gov record

| Milestone or reason | AD109 | Placebo |
| --- | --- | --- |
| Started | “329” | “331” |
| Completed | “250” | “274” |
| Not completed | “79” | “57” |
| Adverse Event | “12” | “7” |
| Lost to Follow-up | “24” | “20” |
| Other/Not Reported | “5” | “2” |
| Physician Decision | “2” | “7” |
| Site Terminated by Sponsor | “3” | “2” |
| Withdrawal by Subject | “33” | “19” |

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.1-2"></a>

###### Study completion

> “Eligible participants (N=660) were randomized (AD109: n=329; PBO: n=331) with 524 (80.0%) completing the study.”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.1-3"></a>

###### Treatment discontinuation due to adverse events and overall treatment discontinuation by month 6 and month 12

> “In LunAIRo, among participants randomized in the main cohort, AE–related discontinuation occurred in 23.7% of participants receiving AD109 and 6.6% receiving placebo by month 6, and remained the most common reason for discontinuation by month 12 (28.6% vs 10.0%), contributing to overall discontinuation rates of 46.5% for AD109 and 25.1% for placebo at month 12.”
> “Importantly, for both trials, AE-related discontinuations occurred predominantly early after treatment initiation, with markedly fewer events later in the treatment period (Figure 32).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.1-4"></a>

###### Treatment discontinuation due to insomnia

> “Insomnia was the main reason for discontinuation in 4.9% of participants in SynAIRgy and 3% of participants in LunAIRo.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.1-5"></a>

###### Pooled phase 3 analysis: study completion and overall treatment discontinuation by week 26

> “Of 1,299 randomized participants (AD109: n = 648; PBO: n = 651), 1,063 (81.8%) completed the trials.”
> “Treatment was discontinued by Week 26 in 227 (35.2%) participants randomized to AD109 and 100 (15.5%) to PBO.”

Source: [Patel et al. (2026b)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.317/8712356)

> “A total of 1299 participants were randomized 1:1 to receive AD109 (n = 648) or placebo (n = 651); 82.3% (1063/1292) of participants who received treatment completed the trials.”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="topic-3.1.2.8.1-6"></a>

###### Pooled phase 3 analysis: treatment discontinuation due to adverse events by week 26

> “TEAEs leading to treatment discontinuation by Week 26 was 25.0% (163/653) in participants receiving AD109 and 6.7% (43/646) receiving PBO.”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="heading-3.1.2.8.2"></a>

###### 3.1.2.8.2 Baseline Characteristics

<a id="topic-3.1.2.8.2-1"></a>

###### Baseline characteristics in the design paper

| Characteristic | LunAIRo (N = 660) |
| --- | --- |
| Age (yrs), mean (SD) | “57 (10)” |
| BMI (kg/m2), mean (SD) | “32.1 (4.8)” |
| Female, n (%) | “303 (46)” |
| Male, n (%) | “357 (54)” |
| Black or African American, n (%) | “113 (17.1)” |
| White, n (%) | “492 (74.5)” |
| Asian, n (%) | “23 (3.5)” |
| Hispanic or Latino, n (%) | “111 (16.8)” |
| AHI4, mean (SD) | “24 (18)” |
| Mild, AHI4 5 to less than 15, n (%) | “248 (37.6)” |
| Moderate, AHI4 15 to less than 30, n (%) | “216 (32.7)” |
| Severe, AHI4 30 or more, n (%) | “196 (29.7)” |
| ODI3, mean (SD) | “32 (19)” |
| HB4, mean (SD) | “55 (68)” |
| PROMIS-Fatigue T-score, mean (SD) | “59 (7)” |
| ESS, mean (SD) | “10.2 (4.8)” |

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="topic-3.1.2.8.2-2"></a>

###### Baseline characteristics of the full analysis set in the ClinicalTrials.gov record

| Characteristic | AD109 (n = 324) | Placebo (n = 326) |
| --- | --- | --- |
| Age, years, mean (SD) | “57.2(10.26)” | “57.4(10.72)” |
| Female | “160” | “138” |
| Male | “164” | “188” |
| Black or African American | “58” | “53” |
| White | “238” | “246” |
| Baseline AHI4 below 15 | “119” | “125” |
| Baseline AHI4 15 to 30 | “109” | “104” |
| Baseline AHI4 above 30 | “96” | “97” |
| Baseline AHI4, events/hr, mean (SD) | “24.03(16.692)” | “24.69(18.735)” |

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.2-3"></a>

###### Demographics, AHI4 severity, and symptoms in the baseline abstract

> “Mean (SD) baseline age was 57 (10) years, and mean BMI (SD) was 32.1 (4.8) kg/m2 with 46% female and 54% male participants.”
> “At baseline, mean (SD) AHI4 was 24 (18) with 37%, 33%, and 30% of participants having mild (AHI4 5-<15) moderate (AHI4 15-<30), or severe (AHI4 ≥30) OSA, respectively.”
> “The percentages of participants having a mild, moderate, or severe PGI-S for Fatigue were 24%, 51%, and 21%, respectively. Mean (SD) ESS score was 10.2 (4.8) and mean (SD) HB4 was 55 (68).”

Source: [Patel et al. (2025a)](https://doi.org/10.1164/ajrccm.2025.211.Abstracts.A6932)

<a id="topic-3.1.2.8.2-4"></a>

###### Median baseline AHI4 in the on-treatment set

> “In the mITT (AD109: n=240; PBO: n=293), the AHI4 (events/hr) was reduced from a median (IQR) of 20.4 (11.4–34.1) at baseline to 13.4 (5.6–30.1) at Week 26 with AD109 and from 20.7 (11.1–32.4) at baseline to 20.3 (9.9–35.1) at Week 26 with PBO.”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.2-5"></a>

###### Pooled phase 3 analysis: baseline severity and AHI distribution

> “Baseline OSA severity was mild in 36.2%, moderate in 37.3%, and severe in 26.5% of participants.”

Source: [Patel et al. (2026b)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.317/8712356)

> “Although patients with higher AHIs were represented, the distribution of AHI at baseline was weighted toward the lower end of the moderate range, with 51% of participants having AHI values below 20 events/hour and 74% falling below 30 events/hour.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.8.3"></a>

###### 3.1.2.8.3 Efficacy Results

<a id="topic-3.1.2.8.3-1"></a>

###### Week 26 change from baseline in AHI4, treatment policy and while-on-treatment estimands, in the ClinicalTrials.gov record

| Estimand and arm | LS mean change, events/hr | Lower 95% CI | Upper 95% CI |
| --- | --- | --- | --- |
| Treatment policy, AD109 (n = 324) | “-3.88” | “-5.37” | “-2.39” |
| Treatment policy, placebo (n = 326) | “0.21” | “-1.24” | “1.65” |
| While-on-treatment, AD109 (n = 240) | “-5.83” | “-7.51” | “-4.15” |
| While-on-treatment, placebo (n = 293) | “-0.02” | “-1.55” | “1.50” |

Source: [ClinicalTrials.gov NCT05811247 API record](https://clinicaltrials.gov/api/v2/studies/NCT05811247)

| Comparison | LS mean difference | Lower 95% CI | Upper 95% CI | p-value |
| --- | --- | --- | --- | --- |
| “Analyses were conducted under the Treatment Policy estimand” | “-4.08” | “-6.14” | “-2.03” | “<0.001” |
| “Analyses were conducted under the While-on-Treatment Estimand” | “-5.81” | “-8.07” | “-3.55” | “<0.0001” |

Source: [ClinicalTrials.gov NCT05811247 API record](https://clinicaltrials.gov/api/v2/studies/NCT05811247)

<a id="topic-3.1.2.8.3-2"></a>

###### Week 26 change in AHI4, treatment policy estimand (ITT set)

> “In the ITT (AD109: n=324; PBO: n=326), AD109 significantly reduced AHI4 by −3.9 (−5.4, −2.4) vs 0.2 (−1.2, 1.7) for PBO (p<0.001).”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

> “LunAIRo: Treatment Policy Estimand: AD109 significantly reduced mean AHI by 3.9 events/hr vs placebo (+0.2 events/hr; placebo-adjusted: 4.1 events/hr; p<0.001); On Treatment Estimand: AD109 significantly reduced mean AHI by 5.8 events/hr vs placebo (0 events/hr; placebo-adjusted: 5.8 events/hr; p<0.0001).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-3"></a>

###### Week 26 relative change in AHI4, treatment policy and on-treatment estimands

> “LunAIRo: Treatment Policy Estimand: AD109 significantly reduced AHI by 33.7% vs placebo (7.3%; p<0.0001); On Treatment Estimand: AD109 significantly reduced AHI by 46.8% vs placebo (6.8%; p<0.0001)”
> “Data are least-squares (LS) mean estimated relative change from baseline from mixed model for repeated measurements (MMRM) analysis. Error bars represent 95% confidence intervals. Analysis was performed on the log scale and back transformed.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-4"></a>

###### Week 26 and week 51 change in AHI4, on-treatment estimand (mITT set)

> “In the mITT, AD109 significantly reduced AHI4 (LS mean [95% CI]) at Week 26 by −5.8 (−7.5, −4.2) vs −0.0 (−1.6, 1.5) with PBO (p<0.0001), representing 46.8% improvement at Week 26 (vs 6.8% with PBO; p<0.001) and 43.5% improvement at Week 51 (vs 13.2% with PBO; p<0.001).”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

> “In the mITT population, AD109 achieved a significant reduction in mean AHI from baseline (46.8%, p<0.001) compared to placebo (6.8%, p<0.001) at 26 weeks.”

Source: [Apnimed CHEST 2025 release (2025)](https://apnimed.com/article/apnimed-to-present-additional-phase-3-data-for-ad109-an-investigational-oral-pill-for-obstructive-sleep-apnea-at-chest-2025-annual-meeting/)

<a id="topic-3.1.2.8.3-5"></a>

###### Week 26 and week 51 AHI reduction as reported in the company release

> “Participants treated with AD109 achieved a mean reduction in AHI of 46.8% from baseline at week 26 (vs 6.8% with placebo; p<0.001). The reduction in AHI remained significant at end of study (week 51, p<0.001).”

Source: [Apnimed LunAIRo topline release (2025)](https://apnimed.com/article/ad109toplinephase3results/)

<a id="topic-3.1.2.8.3-6"></a>

###### Attenuation of the treatment policy estimate at week 51

> “The smaller clinical response observed at week 51 in LunAIRo primarily reflects higher treatment discontinuation in the AD109 group and the conservative handling of missing data using control-based imputation. Among participants who remained on treatment, clinical responses were maintained over time.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-7"></a>

###### Key secondary endpoint: week 26 ODI3, on-treatment estimand (mITT set)

> “In the mITT, AD109 significantly improved select secondary endpoints at Week 26, including change in oxygen desaturation index – ODI3 (LS mean [95% CI]; −7.5 [−9.4, −5.7] vs −0.1 [−1.7, 1.6]; events/hr; p<0.0001) and relative change in hypoxic burden – HB4 (58.2% vs 13.8%; p<0.0001).”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.3-8"></a>

###### Key secondary endpoint: week 26 ODI3, treatment policy and on-treatment estimands

> “LunAIRo: Treatment Policy Estimand: AD109 showed a significant mean relative reduction from significantly reduced ODI3 of 29.5% (5.3 events/hr) vs placebo (3.5%; +0.2 events/hr; p<0.0001) at week 26; On Treatment Estimand: AD109 significantly reduced ODI3 of 40.3% (7.5 events/hr) vs placebo (3.3%; -0.1 events/hr; p<0.0001) at week 26.”
> “For treatment Policy Estimand: AD109 reduced LS means ODI3 by 6.3 events/ hr vs placebo (0.7 events/hr; p<0.0001); On Treatment Estimand: AD109 reduced ODI3 by 8.3 events/hr vs placebo (0.3 events/hr; p<0.0001).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-9"></a>

###### Key secondary endpoint: week 26 hypoxic burden, treatment policy and on-treatment estimands

> “LunAIRo: Treatment Policy Estimand: AD109 significantly reduced mean HB by 37.4% vs placebo (10.0%; p<0.001); On Treatment Estimand: AD109 significantly reduced mean HB by 58.2% vs placebo (13.8%; p<0.0001).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-10"></a>

###### Key secondary endpoint: at least 50% reduction in AHI4 at week 26, on-treatment estimand (mITT set)

> “AD109 significantly reduced AHI4 by ≥50% at Week 26 in 38.3% of participants (vs 15.4% in PBO; p<0.0001) in the mITT.”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.3-11"></a>

###### Key secondary endpoint: at least 50% reduction in AHI at week 26, treatment policy and on-treatment estimands

> “LunAIRo: Treatment Policy Estimand: AD109 significantly reduced AHI≥50% in 28.5% of participants vs placebo (16.0%; p<0.001); On Treatment Estimand: AD109 significantly reduced AHI ≥50% in 38.3% of participants vs placebo (15.4%; p<0.0001).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-12"></a>

###### Severity category improvement and disease control at week 26, on-treatment estimand (mITT set)

> “45.0% of participants treated with AD109 showed a reduction in OSA disease severity category”
> “22.9% of participants treated with AD109 achieved complete OSA disease control (defined as AHI <5 events/hour)”

Source: [Apnimed CHEST 2025 release (2025)](https://apnimed.com/article/apnimed-to-present-additional-phase-3-data-for-ad109-an-investigational-oral-pill-for-obstructive-sleep-apnea-at-chest-2025-annual-meeting/)

<a id="topic-3.1.2.8.3-13"></a>

###### Severity category improvement and disease control at weeks 26 and 51, as reported in the company release

> “AD109 improved OSA disease severity for 45.0% of participants at week 26, and 47.5% at week 51”
> “AD109 achieved OSA complete disease control (AHI<5) for 22.9% of participants at week 26, and 22.5% at week 51”

Source: [Apnimed LunAIRo topline release (2025)](https://apnimed.com/article/ad109toplinephase3results/)

<a id="topic-3.1.2.8.3-14"></a>

###### Apnea index and hypopnea index at week 26, estimand not stated

> “Similarly, in LunAIRo, the mean relative reduction of apnea index was 67.1% (vs 2.8% reduction on placebo), while the mean hypopnea index saw a 37.8% reduction (vs 3.1% on placebo).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-15"></a>

###### Pooled phase 3 analysis: week 26 change in AHI4, treatment policy estimand

> “In the treatment policy estimand (AD109: n = 626; PBO: n = 639), AD109 reduced AHI4 (LS mean [95% CI]; events/hr) by − 3.7 [−4.8, −2.5] vs 0.4 [−0.8, 1.5] (p < 0.0001), representing a model-estimated 39.3% reduction from baseline (vs 12.6% with PBO; p < 0.0001).”

Source: [Patel et al. (2026b)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.317/8712356)

> “In the treatment policy estimand (AD109: n = 626; PBO: n = 639), AD109 reduced AHI4 (LS mean difference [95%CI]: −4.0 [−5.6, −2.5]; p < 0.0001), representing a model-estimated 39.3% reduction from baseline (vs 12.6% [PBO]; p < 0.0001).”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="topic-3.1.2.8.3-16"></a>

###### Pooled phase 3 analysis: week 26 change in AHI4, on-treatment estimand

> “In the on-treatment estimand (AD109: n = 500; PBO: n = 589), AD109 reduced AHI4 (−6.2 [−7.8, −4.6]; p < 0.0001), representing a model-estimated 51.6% reduction from baseline (vs 13.3% [PBO]; p < 0.0001).”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="topic-3.1.2.8.3-17"></a>

###### Pooled phase 3 analysis: responders, disease control, and hypoxic burden, treatment policy estimand

> “AD109 reduced AHI4 by ≥ 50% in 28.3% of participants (vs 17.7% with PBO; p < 0.0001); a higher proportion of participants receiving AD109 achieved disease control (AHI4<5; p < 0.0001). AD109 exhibited greater relative reduction in hypoxic burden (41.6% vs 9.9%; p < 0.0001).”

Source: [Patel et al. (2026b)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.317/8712356)

<a id="topic-3.1.2.8.3-18"></a>

###### Pooled phase 3 analysis: responders, disease control, and hypoxic burden, reported after the on-treatment estimand results

> “39.0% achieved an AHI4 reduction ≥50% with AD109 (vs 18.8% [PBO]; p < 0.0001). More participants receiving AD109 (22.6%) achieved OSA disease control (AHI4<5; vs 8.3% [PBO]; p < 0.0001). AD109 reduced hypoxic burden by 59.6% from baseline (vs 14.9% [PBO]; p < 0.0001).”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="topic-3.1.2.8.3-19"></a>

###### Pooled phase 3 analysis: severity category improvement and disease control, treatment policy and on-treatment estimands

> “In this analysis, AD109 improved OSA disease severity for 48% of participants at week 26. The proportion of participants achieving complete OSA disease control (AHI<5) was also assessed. AD109 achieved OSA disease control (AHI <5) for 22.6% of participants at 26 weeks (p<0.0001 for both treatment policy and on-treatment estimands).”
> “39% of patients achieved a ≥50% reduction in AHI and 23% achieving a ≥70% reduction (on treatment).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-20"></a>

###### Pooled phase 3 analysis: week 26 AHI adjusted for changes in REM sleep and BMI, on-treatment estimand

| Model, LS mean (95% CI), events/h at week 26 | AD109 (n = 500) | Placebo (n = 589) | Estimated treatment difference (95% CI) | p-value |
| --- | --- | --- | --- | --- |
| Primary model for change from baseline in AHI | “–6.1 (–7.3 to –4.9)” | “0.1 (–1.0 to 1.2)” | “–6.2 (–7.8 to –4.6)” | “<0.0001” |
| Model adjusted for change in proportion of REM sleep | “–5.7 (–7.1 to –4.2)” | “0.2 (–1.1 to 1.4)” | “–5.9 (–7.9 to –3.8)” | “<0.0001” |
| Model adjusted for change in BMI | “–5.1 (–6.4 to -3.8)” | “–0.2 (–1.4 to 0.9)” | “–4.9 (–6.6 to –3.1)” | “<0.0001” |

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-21"></a>

###### Pooled phase 3 analysis: snoring frequency at week 26 in participants with an on-treatment polysomnogram

> “Of these, N = 189 (AD109) and N = 281 (Placebo) participants met baseline snoring criteria for inclusion.”
> “AD109 reduced the proportion of breaths with snoring by 22.3 [19.1, 25.5] % (Estimate [95%CI]) vs a placebo reduction of 10.0 [7.4,12.6]% (difference: -12.3 [-16.4, -8.2] %, p < 0.001), Figure 1A. A majority (60.8%) of participants treated with AD109 exhibited a ≥ 50% reduction in the proportion of breaths with snoring, compared to 32.0% of those on placebo (p < 0.001, Figure1B).”

Source: [Gell et al. (2026b)](https://academic.oup.com/ajrccm/article/doi/10.1093/ajrccm/aamag162.6398/8686957)

> “Snoring, an important symptom affecting OSA patients and their bed partners and driving patients to seek clinical evaluation, was analyzed as a prespecified exploratory endpoint in both the LunAIRo and SynAIRgy Phase 3 trials and in a pooled analysis of the on treatment estimand. Analyses were conducted using observed data only, without LOCF imputation.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-22"></a>

###### Pooled phase 3 analysis: modeled cardiovascular risk reduction from hypoxic burden, treatment policy and on-treatment estimands

> “Across the pooled LunAIRo and SynAIRgy trials, mean HB reduction with AD109 was 41.6% (95% CI: 34.7-47.8) in the treatment policy estimand and 59.6% (95% CI: 54.3-64.6) in the on treatment estimand, which corresponded to a predicted reduction in the risk of CV events or all-cause mortality of 11.9% (95% CI: 4.7-18.5) and 19.2% (95% CI: 8.2-28.8), respectively.”
> “Cohort-specific HR estimates were pooled using an inverse-variance weighted random-effects meta-analysis and used to estimate risk reduction associated with the observed HB reduction at Week 26 in AD109-treated participants”

Source: [Azarbarzin et al. (2026)](https://academic.oup.com/ajrccm/article/doi/10.1093/ajrccm/aamag162.6399/8686692)

> “These estimates are based on epidemiologic associations and are intended to provide clinical context for the observed improvements in HB and do not represent direct clinical outcomes measured in the Phase 3 trials.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-23"></a>

###### Pooled phase 3 analysis: AHI response with concomitant GLP-1 receptor agonists, estimand not stated

> “Across these trials, the effects of AD109 on AHI reduction were directionally similar in participants receiving GLP-1 therapy and those not receiving GLP-1 therapy, suggesting that the treatment effect of AD109 was additive to GLP-1 use in participants on stable doses.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.3-24"></a>

###### Pooled phase 3 analysis: body weight change with concomitant GLP-1 receptor agonists for diabetes, exploratory, estimand not stated

> “One-hundred and six participants were enrolled in the phase 3 trials taking a GLP-1 RA for diabetes (PBO+GLP-1 RA: n=48; AD109+GLP-1 RA: n=58).”
> “In participants receiving a GLP-1 RA with baseline BMI ≥35, AD109 reduced weight by 5.5% (vs PBO: 0.9%, P=0.02) and 11.5% (vs PBO: 3.8%, P=0.04) at Week 26 and Week 51, respectively.”
> “Participants not receiving GLP-1 RA (n=979) showed smaller weight reductions across BMI classes; in participants with BMI ≥35, AD109 reduced weight by 1.4% (vs PBO: 0.2%, P=0.002) and 2.8% (vs PBO: 1.4%, P=0.19) at Week 26 and Week 51, respectively.”

Source: [Strollo et al. (2026a)](https://academic.oup.com/sleep/article/49/Supplement_1/A261/8673726)

<a id="heading-3.1.2.8.4"></a>

###### 3.1.2.8.4 Health-related Quality of Life and Patient-reported Outcomes

<a id="topic-3.1.2.8.4-1"></a>

###### Key secondary endpoint: week 26 PROMIS-Fatigue T-score, on-treatment estimand (mITT set)

> “In the mITT, the change from baseline PROMIS-Fatigue T-score (LS mean [95% CI]) at Week 26 was −6.5 (−7.3, −5.6) with AD109 vs −6.9 (−7.7, −6.1) with PBO (p=0.49).”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.4-2"></a>

###### Week 26 PROMIS-Fatigue and PROMIS-Sleep Impairment, treatment policy and on-treatment estimands

> “LunAIRo: Treatment Policy Estimand: AD109 did not significantly reduce PROMIS Fatigue T-score; On Treatment Estimand: AD109 did not significantly reduce PROMIS Fatigue T-score”
> “No treatment effect was observed on PROMIS Sleep Impairment.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.4-3"></a>

###### Pooled phase 3 analysis: week 26 PROMIS-Fatigue, PROMIS-Sleep Impairment, and ESS, on-treatment estimand

> “In the pooled analysis of PROs utilizing the on-treatment estimand (AD109: n=500; PBO: n=589), results (LS mean difference [SE]) numerically favored AD109 over PBO in PROMIS-F T-score (−0.48 [0.43]; P=0.27), PROMIS-SI T-score (−0.54 [0.48]; P=0.26) and ESS (= −0.18 [0.21]; P=0.4) at Week 26.”
> “A higher proportion of participants receiving AD109 achieved ≥3 point improvement in ESS at Week 26 relative to PBO (48.2% vs 41.6%; P=0.018).”

Source: [Strollo et al. (2026b)](https://academic.oup.com/sleep/article/49/Supplement_1/A302/8674486)

<a id="topic-3.1.2.8.4-4"></a>

###### Pooled phase 3 analysis: participants with baseline ESS of 10 or more, on-treatment estimand

> “In participants with baseline excessive daytime sleepiness (AD109: n=280; PBO: n=294), AD109 significantly improved PROMIS-F and PROMIS-SI T-scores (LS mean±SE) at Week 26 by −8.8±0.4 and −9.9±0.5 points, respectively (vs −7.2±0.4 and −8.2±0.5 with PBO; P=0.007 and P=0.01). Similarly, AD109 significantly improved ESS (LS mean ± SE) at Week 26 by −4.5±0.2 points in participants having baseline excessive daytime sleepiness (vs −3.9±0.2 with PBO; P=0.04).”

Source: [Strollo et al. (2026b)](https://academic.oup.com/sleep/article/49/Supplement_1/A302/8674486)

<a id="topic-3.1.2.8.4-5"></a>

###### Pooled phase 3 analysis: participants with baseline ESS of 10 or more, treatment policy estimand

> “In participants with baseline excessive daytime sleepiness (ESS≥10; AD109: n = 348; PBO: n = 316), AD109 improved PROMIS-Fatigue and PROMIS-Sleep Impairment T-scores (p = 0.044 and p = 0.048, respectively).”

Source: [Patel et al. (2026b)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.317/8712356)

<a id="topic-3.1.2.8.4-6"></a>

###### Pooled phase 3 analysis: per-protocol set, treatment policy estimand dataset

> “The pooled treatment policy estimand dataset, for LS mean change from baseline at week 26, showed improvements in the PROMIS Fatigue (-7.7 on AD109 vs –6.6 on placebo, p=0.035) and PROMIS Sleep Impairment (-8.5 on AD109 vs –7.3 on placebo, p=0.017) regardless of baseline symptoms.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.4-7"></a>

###### Pooled phase 3 analysis: association between baseline AHI and baseline patient-reported outcomes

> “The Pearson correlation coefficients (r ≈ 0 for both analyses) indicate no meaningful linear association between baseline AHI and either patient reported outcome.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.8.5"></a>

###### 3.1.2.8.5 Safety Results

<a id="topic-3.1.2.8.5-1"></a>

###### Overall adverse events and most common adverse events

> “Adverse events (AEs) were predominantly mild, with ≥1 AE reported by 272 (82.9%) participants with AD109 and 223 (68.2%) with PBO. The most common AEs with AD109 (≥5% of participants and >PBO) were dry mouth, insomnia, nausea, urinary hesitation, somnolence, headache, dyspepsia, urinary tract infection, constipation, and dizziness. No treatment-related serious AEs were identified.”

Source: [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext)

<a id="topic-3.1.2.8.5-2"></a>

###### Adverse event reporting period and definition

> “from enrollment until the end of follow-up, up to 53 weeks”
> “Adverse events are reported as treatment-emergent adverse events (TEAEs), defined as events with onset after the first dose of study drug and up to 7 days after the last dose.”

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.5-3"></a>

###### Non-serious adverse events in at least 5% of participants in either arm

| Adverse event | AD109, affected / at risk (%) | Placebo, affected / at risk (%) |
| --- | --- | --- |
| Total | “220/328 (67.07%)” | “122/327 (37.31%)” |
| Dry mouth | “130/328 (39.63%)” | “31/327 (9.48%)” |
| Middle insomnia | “44/328 (13.41%)” | “13/327 (3.98%)” |
| Nausea | “43/328 (13.11%)” | “7/327 (2.14%)” |
| Initial insomnia | “32/328 (9.76%)” | “12/327 (3.67%)” |
| Urinary hesitation | “27/328 (8.23%)” | “2/327 (0.61%)” |
| Somnolence | “24/328 (7.32%)” | “12/327 (3.67%)” |
| Headache | “22/328 (6.71%)” | “23/327 (7.03%)” |
| Dyspepsia | “20/328 (6.10%)” | “7/327 (2.14%)” |
| Urinary tract infection | “20/328 (6.10%)” | “10/327 (3.06%)” |
| Constipation | “19/328 (5.79%)” | “8/327 (2.45%)” |
| Upper respiratory tract infection | “19/328 (5.79%)” | “26/327 (7.95%)” |
| Dizziness | “17/328 (5.18%)” | “6/327 (1.83%)” |
| COVID-19 | “14/328 (4.27%)” | “25/327 (7.65%)” |

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.5-4"></a>

###### Serious adverse events and deaths

| Event | AD109, affected / at risk (%) | Placebo, affected / at risk (%) |
| --- | --- | --- |
| All-cause mortality | “0/328 (0.00%)” | “0/327 (0.00%)” |
| Any serious adverse event | “5/328 (1.52%)” | “10/327 (3.06%)” |
| Acute myocardial infarction | “0/328 (0.00%)” | “1/327 (0.31%)” |
| Aortic dissection | “1/328 (0.30%)” | “0/327 (0.00%)” |
| Intracranial aneurysm | “1/328 (0.30%)” | “0/327 (0.00%)” |
| Ischaemic stroke | “0/328 (0.00%)” | “1/327 (0.31%)” |
| Suicide attempt | “0/328 (0.00%)” | “1/327 (0.31%)” |

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.5-5"></a>

###### Relatedness of serious adverse events as classified in the ClinicalTrials.gov record

> “AE was classified as moderate and not related to study intervention”
> “AE was classified as severe and unlikely related to study intervention”
> “AE was classified as severe and possibly related to study intervention”

Source: [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results)

<a id="topic-3.1.2.8.5-6"></a>

###### Absence of serious adverse events related to AD109

> “No serious adverse events related to AD109 were reported in the LunAIRo trial.”

Source: [Apnimed LunAIRo topline release (2025)](https://apnimed.com/article/ad109toplinephase3results/)

<a id="topic-3.1.2.8.5-7"></a>

###### Ambulatory blood pressure and heart rate at week 4

> “The LunAIRo study included an assessment of 24-hour monitoring of ambulatory blood pressure at week 4, demonstrating small changes in mean arterial pressure (1.9 mmHg) and heart rate (4.6 bpm).”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.5-8"></a>

###### Change in BMI at week 51

> “Change in BMI in LunAIRo at 51 weeks was also small: median (IQR) change was -0.2 (-1.4 to 0.8) with AD109 and -0.1 (-0.9 to 0.7) with placebo.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.5-9"></a>

###### Pooled phase 3 analysis: overall and most common treatment-emergent adverse events

> “TEAEs were predominantly mild, with ≥1 TEAE reported by 500 (76.6%) participants on AD109 and 370 (57.3%) on PBO. The most common TEAEs were dry mouth, insomnia, nausea, urinary hesitation, somnolence, and constipation. No AD109-related serious TEAEs were identified.”

Source: [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277)

<a id="topic-3.1.2.8.5-10"></a>

###### Pooled phase 3 analysis: major adverse cardiovascular events

> “Across the pooled tolerability dataset, no (n=0) MACE were reported in participants receiving AD109. Five (n=5) MACE were reported in participants receiving placebo.”
> “The Phase 3 trials were not designed or powered to assess cardiovascular outcomes; however, no major adverse cardiovascular events (MACE) were observed in the AD109 group during one year of follow-up, compared with five events in the placebo group.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.5-11"></a>

###### Pooled phase 3 analysis: heart rate and blood pressure, safety analysis set

| Measure | AD109 (n = 653) | Placebo (n = 646) | p-value |
| --- | --- | --- | --- |
| Heart rate, beats/min, mean change (SD) to week 26, evening pre-PSG | “2.0 (11.2)” | “-0.7 (10.0)” | Not reported |
| Heart rate, beats/min, mean change (SD) to week 26, morning post-PSG | “3.4 (9.5)” | “-0.7 (9.4)” | Not reported |
| Heart rate ≥100 beats/min on study, n (%) | “48 (7.4)” | “28 (4.3)” | “0.021” |
| Systolic blood pressure ≥140 mm Hg on study, n (%) | “175 (26.8)” | “168 (26.0)” | “0.760” |
| Diastolic blood pressure ≥90 mm Hg on study, n (%) | “146 (22.4)” | “131 (20.3)” | “0.368” |

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.2.8.5-12"></a>

###### Phase 3 program: onset, duration, and severity of insomnia

> “The median time of first onset for insomnia was a median (IQR) of 3 days (1 to 12) and the duration of the first event was 16 (6 to 67) days. Insomnia was mild for 79.2% of cases, moderate for 17.4% and severe for 3.4%.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.2.9"></a>

##### 3.1.2.9 Study Limitations

<a id="topic-3.1.2.9-1"></a>

###### Summary

- Publication status: results are available from a congress abstract, the registry results (first posted 2026-11-03), company releases, and the prospectus; the 2026 editorial on SynAIRgy described LunAIRo results as forthcoming.
- Design: placebo-controlled for 51 weeks, with the primary endpoint at week 26; week 51 efficacy was analyzed as a tertiary endpoint. No active comparator was used, dose reduction was not permitted, and all 64 sites were in the United States.
- Population: adults intolerant of or refusing PAP with a PROMIS-Fatigue raw score of at least 17, and a BMI cap of 40 (men) or 42 (women) kg/m2. Exclusions included bothersome insomnia, clinically significant cardiac disease, blood pressure above 145/90 mmHg, specified psychiatric disorders, and narrow angle glaucoma.
- Eligibility AHI differs by source: protocol version 4.0 states AHI4 of more than 5 and up to 45, the registry states AHI of more than 5, and the baseline abstract states no upper limit. The design paper states caps could be added during the trial.
- Severity mix: in the pooled phase 3 population, 51% had AHI below 20 events/h and 74% below 30, per the prospectus.
- Endpoint change: the 2025 baseline abstract named the proportion with at least 50% AHI4 reduction at six months as the primary endpoint. Protocol amendment 4.0 (June 2025) reordered the primary and key secondary endpoints “to align with recent regulatory approvals for the treatment of OSA related to obesity”, after the registry's actual primary completion date (2025-05-07).
- Estimands and missing data: the treatment policy estimand assumed data missing not at random with control-based imputation. The on-treatment estimand excluded data after treatment discontinuation with last observation carried forward and included 240 participants on AD109 and 293 on placebo, against 324 and 326 in the full analysis set. The company states the on-treatment estimand gave the most clinically relevant estimate, and attributes the smaller week 51 response to higher discontinuation with AD109 and control-based imputation. The topline release reported the on-treatment relative reduction (46.8%).
- Analysis-set exclusions: participants at site 1308 were excluded from the full analysis set after the site was terminated.
- Discontinuation: treatment discontinuation due to adverse events was 23.7% with AD109 and 6.6% with placebo by month 6 and 28.6% and 10.0% by month 12; overall treatment discontinuation at month 12 was 46.5% and 25.1% (prospectus, main cohort). In the registry, 250 of 329 participants on AD109 and 274 of 331 on placebo completed the study.
- Patient-reported outcomes: AD109 did not significantly reduce the PROMIS-Fatigue T-score under either estimand, and no treatment effect was observed on PROMIS Sleep Impairment.
- Pooled analyses: Patel et al. (2026a) reports the pooled 39.0% versus 18.8% responder result without an estimand label, while the prospectus labels the pooled 39% as on treatment. The prospectus states the phase 3 trials were not designed or powered to assess cardiovascular outcomes.
- Clinical meaningfulness: at the pre-NDA meeting the FDA gave feedback on the clinical meaningfulness of the primary endpoint results and the limitations of the PROs, and indicated additional analyses and justification may be needed.
- Sponsor: the prospectus describes LunAIRo as a trial the company initiated.

Source: [Apnimed LunAIRo protocol (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/Prot_000.pdf); [Apnimed LunAIRo statistical analysis plan (2025)](https://cdn.clinicaltrials.gov/large-docs/47/NCT05811247/SAP_001.pdf); [ClinicalTrials.gov NCT05811247](https://clinicaltrials.gov/study/NCT05811247); [ClinicalTrials.gov NCT05811247 results](https://clinicaltrials.gov/study/NCT05811247?tab=results); [ClinicalTrials.gov NCT05811247 API record](https://clinicaltrials.gov/api/v2/studies/NCT05811247); [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/); [Patel et al. (2025a)](https://doi.org/10.1164/ajrccm.2025.211.Abstracts.A6932); [Patel et al. (2025b)](https://journal.chestnet.org/article/S0012-3692%2825%2905443-1/fulltext); [Patel et al. (2026a)](https://academic.oup.com/ajrccm/article/212/Supplement_2/aamag286.015/8712277); [Apnimed LunAIRo topline release (2025)](https://apnimed.com/article/ad109toplinephase3results/); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm); [Mehra and Mokhlesi (2026)](https://academic.oup.com/ajrccm/article/212/7/1427/8679096)

<a id="heading-3.1.3"></a>

#### 3.1.3 APC-APN-306

<a id="heading-3.1.3.1"></a>

##### 3.1.3.1 Objective

<a id="topic-3.1.3.1-1"></a>

###### Long-term safety of AD109 in participants completing a parent phase 3 trial

> “We have followed up with an open-label extension (OLE) study (NCT06566820) to evaluate the long-term safety of AD109 in OSA by enrolling qualifying participants who completed one of the parent phase 3 trials.”
> “This large OLE study is designed to confirm and extend the safety evidence of the phase 3 SynAIRgy and LunAIRo randomized control trials evaluating the efficacy and safety of AD109 in the treatment of OSA.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

> “The present study is a Phase 3 open label continuation protocol enrolling qualifying participants who completed one of the parent Phase 3 double-blind placebo-controlled studies, APC-APN-304 or APC-APN-305.”

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="topic-3.1.3.1-2"></a>

###### Exploratory objectives: night-to-night variability by ring oximetry and patient-reported outcomes by mobile application

> “In addition, we aimed to assess night-to-night OSA variability with a high-resolution pulse oximetry ring and the OSA burden over time using validated Patient Reported Outcomes (PROs) collected with a mobile application.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="heading-3.1.3.2"></a>

##### 3.1.3.2 Location and Study Date

<a id="topic-3.1.3.2-1"></a>

###### Registry dates and status, last updated May 2025

| Field | Registry entry |
| --- | --- |
| Study start (actual) | “2024-07-03” |
| Primary completion (estimated) | “2026-06-30” |
| Last update posted | “2025-05-13” |

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="topic-3.1.3.2-2"></a>

###### Completion of the open-label extension in 2026 per the company

> “Both LunAIRo and SynAIRgy trials were completed in 2025 and the OLE trial was completed in 2026.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.3.3"></a>

##### 3.1.3.3 Study Design

<a id="topic-3.1.3.3-1"></a>

###### Phase 3, single-group, open-label continuation protocol

> “A Phase 3 Open Label Continuation Protocol of a Fixed Dose Combination of Aroxybutynin and Atomoxetine (AD109) in Obstructive Sleep Apnea”

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="topic-3.1.3.3-2"></a>

###### Washout before open-label dosing and serial safety assessments

> “After informed consent and a one week washout, participants will undergo the following dosing regimen, serial assessments and data collection (see enclosed figure).”
> “Optional exploratory data collection involving serial assessments using a ring-worn pulse oximeter and OSA-related PROs were offered to the OLE participants.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="topic-3.1.3.3-3"></a>

###### Safety follow-up for up to an additional year after either phase 3 trial

> “Participants completing either of these studies are eligible for an optional open label extension (OLE) study of AD109 in which safety is assessed for up to an additional year”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

> “Syneos provided services to us in connection with LunAIRo and SynAIRgy and the OLE that allowed patients to extend their treatment following completion of either Phase 3 protocol.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.3.4"></a>

##### 3.1.3.4 Eligibility Criteria

<a id="topic-3.1.3.4-1"></a>

###### Completion of a parent phase 3 trial without permanent discontinuation of study drug

> “Completion of APC-APN-304 without permanent discontinuation of IMP.”
> “Completion of APC-APN-305 without permanent discontinuation of IMP.”
> “Participants are excluded from the study if participation would be considered unsafe by the investigator based on the individual's current status.”
> “Participants are excluded from the study if participation would be considered unsafe by the investigator based on concomitant therapies.”

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

> “All participants completing a parent phase 3 trial with continued use of investigational product were eligible for inclusion in the OLE study if considered safe by the study team.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="heading-3.1.3.5"></a>

##### 3.1.3.5 Treatment

<a id="topic-3.1.3.5-1"></a>

###### Open-label fixed-dose combination of aroxybutynin and atomoxetine

> “Open Label AD109”
> “Fixed Dose Combination”

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="heading-3.1.3.6"></a>

##### 3.1.3.6 Study Outcomes

<a id="heading-3.1.3.6.1"></a>

###### 3.1.3.6.1 Rationale for Using Incidence of Adverse Events as the Primary Endpoint

<a id="topic-3.1.3.6.1-1"></a>

###### Registry primary outcome measures: adverse events, vital sign abnormalities, and concomitant therapy

> “Number of Participants With Adverse Events (AES) and Serious Adverse Events (SAEs)”
> “Number of Participants With Vital Signs Abnormalities”
> “Vital signs that are planned to be assessed include parameters such as weight,heart rate and systolic and diastolic blood pressure.”
> “Number of Participants With Concomitant Therapy (Medication or Devices)”
> “Concomitant therapy with devices (i.e., CPAP) and concomitant medications are other prescription medications, over-the-counter (OTC) drugs or dietary supplements that a study participant takes in addition to the drug under investigation.”
> “through study completion, an average of 1 year”

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="topic-3.1.3.6.1-2"></a>

###### Safety assessments and exploratory oximetry parameters

> “Safety assessments include continued measurement of vital signs, monitoring and recording of adverse events, serious adverse events, pregnancies, and recording of study or treatment discontinuations.”
> “Oximetry-derived parameters included the oxygen desaturation index based on 4% desaturation, time with oxygen saturation ≤ 90%, and hypoxic burden.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="heading-3.1.3.7"></a>

##### 3.1.3.7 Statistical Analysis Description

<a id="heading-3.1.3.7.1"></a>

###### 3.1.3.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.3.7.1-1"></a>

###### Planned enrollment

| Field | Registry entry |
| --- | --- |
| Enrollment (estimated) | “1280” |

Source: [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820)

<a id="heading-3.1.3.7.2"></a>

###### 3.1.3.7.2 Description of Analysis Sets

No evidence found.

<a id="heading-3.1.3.8"></a>

##### 3.1.3.8 Results

<a id="heading-3.1.3.8.1"></a>

###### 3.1.3.8.1 Participant Disposition

<a id="topic-3.1.3.8.1-1"></a>

###### Interim enrollment and uptake of exploratory data collection

> “One hundred and two participants (70% of eligible) have enrolled in the OLE study to date.”
> “Seventy-three participants (72%) also opted-in to the exploratory data collection with 1179 overnight pulse oximetry recordings and 319 PRO questionnaires collected.”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="heading-3.1.3.8.2"></a>

###### 3.1.3.8.2 Baseline Characteristics

No evidence found.

<a id="heading-3.1.3.8.3"></a>

###### 3.1.3.8.3 Efficacy Results

No evidence found.

<a id="heading-3.1.3.8.4"></a>

###### 3.1.3.8.4 Health-related Quality of Life and Patient-reported Outcomes

No evidence found.

<a id="heading-3.1.3.8.5"></a>

###### 3.1.3.8.5 Safety Results

<a id="topic-3.1.3.8.5-1"></a>

###### Interim safety conclusion stated without data

> “Data collected to date supports the long-term safety of AD109”

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233)

<a id="heading-3.1.3.9"></a>

##### 3.1.3.9 Study Limitations

<a id="topic-3.1.3.9-1"></a>

###### Summary

- Design: open-label, single-group continuation study with no comparator; the registered primary outcomes are counts of participants with adverse events, vital sign abnormalities, and concomitant therapy.
- Population: eligibility required completion of a parent phase 3 trial without permanent discontinuation of study drug, so participants who stopped study drug in the parent trials were not eligible.
- Status: the registry record (last updated 2025-05-13) lists an estimated enrollment of 1,280 and an estimated primary completion of 2026-06-30, while the prospectus states the trial was completed in 2026.
- Results: the only results are interim data from a 2025 congress abstract, with 102 participants enrolled (70% of eligible). The abstract states that data collected to date support long-term safety without reporting adverse event data, and no baseline, efficacy, patient-reported outcome, or analysis-set information is available.
- Exploratory ring oximetry and mobile-application PROs were optional; 73 participants (72%) opted in.
- Sponsor: the prospectus states that Syneos provided services to the company in connection with the trial.

Source: [Strollo et al. (2025a)](https://academic.oup.com/ajrccm/article/211/Supplement_1/A6934/8338233); [ClinicalTrials.gov NCT06566820](https://clinicaltrials.gov/study/NCT06566820); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.4"></a>

#### 3.1.4 MARIPOSA

<a id="heading-3.1.4.1"></a>

##### 3.1.4.1 Objective

<a id="topic-3.1.4.1-1"></a>

###### Efficacy and safety of AD109 over 4 weeks

> “The aim of this study was to evaluate the efficacy and safety of AD109, a combination of the novel antimuscarinic agent aroxybutynin and the norepinephrine reuptake inhibitor atomoxetine, in the treatment of OSA.”
> “The purpose of the present study was to evaluate the efficacy and safety of AD109 over a longer duration (4 wk) in patients with mild to severe OSA.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.1-2"></a>

###### Company-stated objectives, including selection of the aroxybutynin dose for phase 3

> “confirm that the combination of a novel anti-muscarinic with an NRI could potentially be effective in a broad OSA population including those with mild, moderate and severe OSA;”
> “evaluate two doses of the aroxybutynin component of AD109 to determine the final dose to be advanced to Phase 3;”
> “confirm and characterize the clinical response of the aroxybutynin and atomoxetine components; and”
> “evaluate safety and tolerability of AD109 in a substantially sized patient population and extended over a 30-day period.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.4.1-3"></a>

###### Role in the design of the phase 3 trials

> “including the recent Phase 2 MARIPOSA study, a placebo-controlled dose-finding trial that informed the design of the current phase 3 trials”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="heading-3.1.4.2"></a>

##### 3.1.4.2 Location and Study Date

<a id="topic-3.1.4.2-1"></a>

###### Twenty-five U.S. sites, November 2021 to August 2022

> “A phase II randomized, double-blind, placebo-controlled, parallel-arms clinical trial was performed at 25 clinical investigative sites in the United States.”
> “Enrollment began on November 29, 2021 and was completed on August 3, 2022.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.2-2"></a>

###### Registry dates

| Field | Registry entry |
| --- | --- |
| Study start (actual) | “2021-11-29” |
| Primary completion (actual) | “2022-07-20” |
| Study completion (actual) | “2022-08-03” |

Source: [ClinicalTrials.gov NCT05071612](https://clinicaltrials.gov/study/NCT05071612)

<a id="topic-3.1.4.2-3"></a>

###### Results first posted to the registry in August 2026

> “Results First Posted 2026-08-12”

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="topic-3.1.4.2-4"></a>

###### Start date as stated by the company

> “We initiated MARIPOSA (APC-005) in December of 2021 to evaluate the safety and efficacy of AD109 and advance dose-finding for AD109 over a one-month period of dosing.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.4.3"></a>

##### 3.1.4.3 Study Design

<a id="topic-3.1.4.3-1"></a>

###### Phase 2, randomized, double-blind, placebo-controlled, parallel-group, 4-week trial

> “Phase II randomized, double-blind, placebo-controlled, parallel-group, 4-week trial comparing AD109 2.5/75 mg, AD109 5/75 mg, atomoxetine 75 mg alone, and placebo (www.clinicaltrials.gov identifier NCT 05071612).”
> “Participants who met all enrollment criteria were randomized to one of four treatment arms (AD109 2.5/75 mg, AD109 5/75 mg, Ato75, or placebo) using a randomization ratio of 2:2:3:3.”
> “Participants were centrally randomized using an interactive web-based system. Participants, investigators, and study personnel were all blinded to allocation.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.3-2"></a>

###### AD504 arms in the same trial, reported separately

> “The study also evaluated another experimental drug (AD504) in early development, and those results will be reported separately.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

> “Phase 2 Randomized Double-Blind Placebo-Controlled Parallel-Arm Dose Finding Study to Compare Fixed Dose Combinations of AD109 and AD504 to Atomoxetine or Placebo in Obstructive Sleep Apnea”
> “Atomoxetine 75 mg tablet/Trazodone 100 mg capsule”
> “Atomoxetine 75 mg tablet/Trazodone 50 mg capsule”

Source: [ClinicalTrials.gov NCT05071612](https://clinicaltrials.gov/study/NCT05071612)

<a id="heading-3.1.4.4"></a>

##### 3.1.4.4 Eligibility Criteria

<a id="topic-3.1.4.4-1"></a>

###### Inclusion and exclusion criteria in the publication (AHI4 of 10 to 45 by two polysomnograms)

> “Participants were adults with OSA with a mean apnea–hypopnea index using the ⩾4% desaturation criterion for hypopneas (AHI4) of 10–45 based on two screening/baseline polysomnography (PSG) studies, with <25% of events comprising central or mixed apnea.”
> “Age requirement was 18–65 years for men and 18–75 for women. Body mass index was required to be ⩽38 kg/m2 for men and ⩽40 kg/m2 for women.”
> “Key exclusion criteria were clinically important sleep disorders other than OSA, active cardiac diseases (rhythm disturbance, coronary artery disease, or heart failure), hypertension requiring more than two drugs for control, craniofacial malformation or grade ⩾3 tonsillar hypertrophy, work schedule involving night or shift work, and chronic therapy with monoamine oxidase inhibitors, norepinephrine reuptake inhibitors, α-1 antagonists, tricyclic antidepressants, strong CYP2D6 inhibitors (e.g., fluoxetine, paroxetine, duloxetine), opioids, muscle relaxants, wake-promoting agents, antipsychotic agents, anticholinergic agents, or warfarin.”
> “Participants treated with CPAP, oral or nasal devices, or positional devices could enroll as long as the devices had not been used for ⩾2 weeks before the first PSG study and were not used during participation in the study.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.4-2"></a>

###### Registry inclusion criteria

> “Mean AHI 10 to 45 events/h, inclusive”
> “BMI between 18.5 and 38 kg/m2 for men and 40 kg/m2 for women, inclusive.”

Source: [ClinicalTrials.gov NCT05071612](https://clinicaltrials.gov/study/NCT05071612)

<a id="topic-3.1.4.4-3"></a>

###### Exclusion criteria as stated by the company, including insomnia

> “Main exclusion criteria were a history of insomnia or narcolepsy, craniofacial malformations, clinically significant cardiac disease (coronary artery disease, heart failure, rhythm disturbances, treatment-resistant hypertension) psychiatric or neurologic diseases, severe constipation, gastroesophageal reflux disease or urinary retention.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.4.5"></a>

##### 3.1.4.5 Treatment

<a id="topic-3.1.4.5-1"></a>

###### Nightly dosing with a 1-week lower-dose run-in and no dose reduction

> “The study drug was prepared as identical tablets and identical capsules to ensure adequate double-blinding. Participants received a single tablet or a tablet and a capsule.”
> “Participants were instructed to take the study drug nightly immediately before bedtime for 4 weeks.”
> “The first week of dosing was with a lower-dose run-in (AD109 2.5/37.5 mg or atomoxetine 37.5 mg), followed by the full dose taken in Weeks 2–4. Dose reduction was not permitted; dosing was discontinued if the study drug was not tolerated.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.5-2"></a>

###### Registry arm descriptions with the titration dose

> “Dose: Atomoxetine 75 mg/Aroxybutynin 2.5 mg tablet following a 1-week titration with Atomoxetine 37.5 mg/Aroxybutynin 2.5 mg tablet.”
> “Dose: Atomoxetine 75 mg/Aroxybutynin 5 mg tablet following a 1-week titration with Atomoxetine 37.5 mg/Aroxybutynin 2.5 mg tablet.”
> “Dose: Atomoxetine 75 mg tablet alone or Atomoxetine 75 mg tablet plus placebo capsule, following a 1-week titration with Atomoxetine 37.5 mg tablet alone or Atomoxetine 37.5 mg tablet plus placebo capsule. Regimens were pooled for analysis.”

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="topic-3.1.4.5-3"></a>

###### Dose notation in the prospectus (aroxybutynin base amounts)

> “aroxybutynin 2.3mg and atomoxetine 75mg and aroxybutynin 4.5mg and atomoxetine 75mg, respectively”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.4.6"></a>

##### 3.1.4.6 Study Outcomes

<a id="heading-3.1.4.6.1"></a>

###### 3.1.4.6.1 Rationale for Using Apnea-Hypopnea Index as the Primary Endpoint

<a id="topic-3.1.4.6.1-1"></a>

###### Primary endpoint in the publication: change in AHI4 for each AD109 dose versus placebo

> “The primary endpoint was the change from baseline to weeks 3 and 4 in AHI4 for each AD109 dose versus placebo. The secondary endpoint was the change from baseline to weeks 3 and 4 in AHI4 for Ato75 versus placebo.”
> “Although study eligibility was determined by scoring of PSG findings at the individual study sites, endpoint calculation was based on subsequent scoring by a central scoring laboratory (Sleep Strategies).”
> “Data from the two baseline PSG studies were averaged for analysis, as were data from the two treatment PSG studies (conducted after 3 and 4 weeks of dosing, respectively).”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.6.1-2"></a>

###### Endpoints of the trial

| Endpoint | Description |
| --- | --- |
| Primary | “Change from baseline in AHI4, AD109 vs placebo” |
| Secondary | “Change from baseline in AHI4, Ato75 vs placebo” |
| Tertiary, respiratory | “AHI4, HB, ODI4, AHI3a, T90” |
| Tertiary, patient-reported | “PGI-S, Short SAQLI, PROMIS scales (sleep impairment, sleep disturbance, and fatigue), ESS, QQ” |
| Tertiary, responders | “Proportion of participants with ⩾50% reduction in AHI4” |
| Safety | “Physical examination, vital signs, ECG, clinical laboratory assessment” |
| Safety, psychometric | “DSST, PVT, VOLT, CSSA” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.6.1-3"></a>

###### Primary endpoint in the registry: combined AD109 arms versus placebo, a difference from the publication

> “Change in Apnea-Hypopnea Index (AHI4) From Baseline: Combined AD109 Arms vs. Placebo”
> “Change in AHI4 (apnea-hypopnea index, 4% oxygen desaturation criterion; average respiratory events per hour of sleep) measured by in-lab polysomnography.”
> “Change from baseline was assessed at treatment PSG nights at approximately Day 21 (Visit 5) and Day 28 (Visit 6), modeled as repeated measures, with the reported value representing the model-based LS mean using equal (0.5/0.5) weighting between the Day 21 and Day 28 visits.”

Source: [ClinicalTrials.gov NCT05071612](https://clinicaltrials.gov/study/NCT05071612)

<a id="topic-3.1.4.6.1-4"></a>

###### Hypoxic burden as a cardiovascular risk metric

> “Hypoxic burden was calculated according to Azarbarzin and colleagues by quantifying the respiratory event–associated area under the desaturation curve for the entire night and dividing by sleep duration to yield a metric expressed in %min per hour of sleep”
> “On the contrary, AD109 has been shown to significantly decrease hypoxic burden, a metric shown to predict adverse cardiovascular outcomes”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.6.1-5"></a>

###### Epworth Sleepiness Scale change compared with the clinically important difference

> “Although the change from baseline in Epworth Sleepiness Scale score did not differ versus placebo for any drug regimen, the median change from baseline was similar to the clinically important difference of 2–3 points in CPAP-treated patients”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.6.1-6"></a>

###### AHI as the conventional primary endpoint of OSA trials, per the phase 3 design paper

> “The primary efficacy endpoint in most OSA trials is the AHI which historically has been the principal biomarker for the disease”

Source: [Taranto-Montemurro et al. (2025)](https://pmc.ncbi.nlm.nih.gov/articles/PMC12490522/)

<a id="heading-3.1.4.7"></a>

##### 3.1.4.7 Statistical Analysis Description

<a id="heading-3.1.4.7.1"></a>

###### 3.1.4.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.4.7.1-1"></a>

###### Planned sample size and power

> “Approximately 200 participants were planned for enrollment, with approximately 40 in each AD109 dose group, 60 in the Ato75 group, and 60 in the placebo group.”
> “The present study had >90% power to detect a treatment difference on the primary endpoint (AHI4 for AD109 vs. placebo) at a two-sided 0.05 significance level if the true difference between the active arm and placebo arm on AHI4 was 11 units and the within-subject standard deviation was 11.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.7.1-2"></a>

###### Participants randomized and analyzed in the AD109 portion of the trial

> “Of 211 randomized patients, 181 were included in the prespecified efficacy analyses.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.7.1-3"></a>

###### Registry enrollment of 294 participants, including the AD504 arms

| Field | Registry entry |
| --- | --- |
| Enrollment (actual), all arms | “Enrollment (Actual) 294” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="topic-3.1.4.7.1-4"></a>

###### Registry primary analysis population, combined AD109 arms

> “Prespecified AD109 combined analysis group includes n=35 participants taking Atomoxetine 75mg / Aroxybutynin 5mg, and n=34 participants taking Atomoxetine 75 mg / Aroxybutynin 2.5 mg”

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="heading-3.1.4.7.2"></a>

###### 3.1.4.7.2 Description of Analysis Sets

<a id="topic-3.1.4.7.2-1"></a>

###### Modified intention-to-treat, intention-to-treat, and safety populations

> “According to a predefined statistical analysis plan, there was no prespecified imputation for missing data. Thus, outcomes were analyzed based on the modified intention-to-treat (mITT) population consisting of all participants who were randomized, took at least one dose of any of the study drugs, and had at least one measurement of the primary endpoint.”
> “A post hoc analysis of the primary and secondary outcomes was performed in the ITT population (all individuals who signed the informed consent form and were randomized to any treatment arm) using the baseline observation carried forward to impute missing data. Safety assessments included all participants who received at least one dose of study drug.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.7.2-2"></a>

###### Analysis methods, sequential testing, and no adjustment for multiplicity

> “For the primary efficacy analysis, a repeated-measures model was used to assess the change in AHI4 from baseline for two doses of AD109, Ato75, and placebo. Percent reduction versus placebo was calculated using an analysis of covariance model with log-transformed data to account for the skewed distribution of values and using baseline as a covariate.”
> “Primary and secondary hypotheses were tested sequentially. Findings were considered significant if the two-sided P value was less than 0.05. Given the exploratory nature of the trial, no adjustment for multiple comparisons was performed.”
> “We present the AD109 doses separately in this article, but we have also included analysis of AHI4 for the combined AD109 doses in Table E2 in the online supplement.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.7.2-3"></a>

###### Registry analysis population and the prespecified pooling of AD109 doses

> “mITT set: all randomized participants who received ≥1 dose and had ≥1 primary endpoint measurement.”
> “The two AD109 dose arms were combined as the primary analysis group per the pre-specified SAP (v1.0, April 2022), as the study was designed and powered to evaluate the overall effect of AD109 versus placebo rather than to compare individual dose levels.”
> “This Phase 2 study was not powered to compare individual dose groups. Primary efficacy analyses evaluated prespecified pooled treatment groups, and results for individual dose arms should be interpreted descriptively.”

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="heading-3.1.4.8"></a>

##### 3.1.4.8 Results

<a id="heading-3.1.4.8.1"></a>

###### 3.1.4.8.1 Participant Disposition

<a id="topic-3.1.4.8.1-1"></a>

###### Screening, randomization, analysis populations, and completion

> “Of the 764 subjects screened, 211 were randomized to one of the four treatment arms (Figure 1). Of the 209 individuals who were randomized and received at least one dose of treatment (i.e., safety population), 181 had at least one measurement on the primary endpoint (i.e., mITT population) and 176 completed the study.”
> “In summary, treatment side effects resulted in one dropout in the placebo arm, 12 in the Ato75 arm, and 5 each in the AD109 2.5/75 mg and AD109 5/75 mg arms.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.1-2"></a>

###### Registry participant flow, all arms including AD504

| Milestone | AD109 75/5mg | AD109 75/2.5mg | AD504 75/100mg | AD504 75/50mg | Atomoxetine 75mg | Placebo |
| --- | --- | --- | --- | --- | --- | --- |
| Started | “41” | “42” | “43” | “40” | “64” | “64” |
| Completed | “35” | “33” | “38” | “38” | “49” | “59” |
| Not completed | “6” | “9” | “5” | “2” | “15” | “5” |
| Withdrawn due to adverse event | “5” | “5” | “3” | “1” | “12” | “1” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="heading-3.1.4.8.2"></a>

###### 3.1.4.8.2 Baseline Characteristics

<a id="topic-3.1.4.8.2-1"></a>

###### Demographic and clinical characteristics of the safety population

| Characteristic | Placebo (n = 63) | Atomoxetine 75 mg (n = 63) | AD109 2.5/75 mg (n = 42) | AD109 5/75 mg (n = 41) |
| --- | --- | --- | --- | --- |
| Age, years, median (IQR) | “57 (48–62)” | “56 (49–60)” | “53.5 (44–58)” | “55 (49–60)” |
| Female sex | “27 (42.9%)” | “22 (34.9%)” | “17 (40.5%)” | “19 (46.3%)” |
| BMI, kg/m2, median (IQR) | “32.0 (27.2–35.2)” | “31.2 (28.11–34.6)” | “34.5 (31.6–36.6)” | “31.2 (27.5–34.0)” |
| CPAP use | “29 (46.1%)” | “30 (47.6%)” | “18 (42.9%)” | “21 (50.2%)” |
| Black or African American | “9 (14.3%)” | “5 (7.9%)” | “11 (26.2%)” | “6 (14.6%)” |
| White | “48 (76.2%)” | “52 (82.5%)” | “28 (66.7%)” | “33 (80.5%)” |
| Hypertension | “14 (22%)” | “8 (13%)” | “6 (14%)” | “12 (29%)” |
| Diabetes | “1 (2%)” | “8 (13%)” | “3 (7%)” | “4 (10%)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.2-2"></a>

###### Baseline AHI4 (4% desaturation hypopnea criterion) by arm

| Arm | Baseline AHI4, events/h, median (IQR) |
| --- | --- |
| Placebo | “20.1 (11.9–25.9)” |
| Atomoxetine 75 mg | “19.0 (11.8–28.8)” |
| AD109 2.5/75 mg | “20.5 (12.3–27.2)” |
| AD109 5/75 mg | “19.4 (13.7–26.4)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.2-3"></a>

###### Inclusion of participants with AHI4 below 10 after central scoring

> “Limitations include the randomization of patients based on PSG scoring at individual sites that resulted in the inclusion of 29 patients with a baseline AHI4 <10 after central scoring.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.2-4"></a>

###### Registry baseline age and body mass index, safety set of all arms

| Characteristic | AD109 75/5mg | AD109 75/2.5mg | Atomoxetine 75mg | Placebo | Total (all six arms) |
| --- | --- | --- | --- | --- | --- |
| Participants | “41” | “42” | “63” | “63” | “292” |
| Age, years, mean (SD) | “54.2(9.24)” | “51.5(9.22)” | “52.8(10.95)” | “54.9(9.03)” | “54.2(9.49)” |
| BMI, kg/m2, mean (SD) | “31.12(4.085)” | “33.43(4.352)” | “31.16(4.716)” | “31.17(4.719)” | “31.65(4.389)” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="heading-3.1.4.8.3"></a>

###### 3.1.4.8.3 Efficacy Results

<a id="topic-3.1.4.8.3-1"></a>

###### Primary endpoint: AHI4 (4% desaturation hypopnea criterion), mITT

> “Apnea–hypopnea index with a 4% desaturation criterion (primary outcome) was reduced from a median (IQR) of 20.5 (12.3–27.2) to 10.8 (5.6–18.5) in the AD109 2.5/75 mg arm (−47.1%), from 19.4 (13.7–26.4) to 9.5 (6.1–19.3) in the AD109 5/75 mg arm (−42.9%; both P < 0.0001 vs. placebo), and from 19.0 (11.8–28.8) to 11.8 (5.5–21.5) with atomoxetine alone (−38.8%; P < 0.01 vs. placebo). Apnea–hypopnea index with a 4% desaturation criterion decreased from 20.1 (11.9–25.9) to 16.3 (11.1–28.9) in the placebo arm.”
> “Both doses of AD109 significantly decreased AHI4 compared with placebo, with mean (95% confidence interval) decreases of −7.16 (−11 to −3.3) events/h for the 2.5/75 mg dose and −7.2 (−10.97 to −3.4) for the 5/75 mg dose (P < 0.001 for both). These decreases represent a placebo-adjusted reduction in AHI4 of 47.1% (27.9–61.2) and 42.9% (22.2–58) for the two doses, respectively.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-2"></a>

###### Respiratory parameters from polysomnography: AHI4 (4% desaturation hypopnea criterion)

| Arm | Baseline, median (IQR) | Treatment, median (IQR) | Placebo-adjusted change, LS mean (95% CI) |
| --- | --- | --- | --- |
| Placebo | “20.1 (11.9–25.9)” | “16.3 (11.1–28.9)” | Not applicable |
| Atomoxetine 75 mg | “19.0 (11.8–28.8)” | “11.8 (5.5–21.5)” | “−5.19 (−8.6 to −1.8)” |
| AD109 2.5/75 mg | “20.5 (12.3–27.2)” | “10.8 (5.6–18.5)” | “−7.16 (−11.0 to −3.3)” |
| AD109 5/75 mg | “19.4 (13.7–26.4)” | “9.5 (6.1–19.3)” | “−7.20 (−10.97 to −3.4)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-3"></a>

###### Respiratory parameters: AHI3a (3% desaturation or arousal hypopnea criterion)

| Arm | Baseline, median (IQR) | Treatment, median (IQR) | Placebo-adjusted change, LS mean (95% CI) |
| --- | --- | --- | --- |
| Placebo | “35.5 (25.6–41.0)” | “31.5 (23.0–41.1)” | Not applicable |
| Atomoxetine 75 mg | “33.4 (25.0–45.7)” | “26.1 (14.5–42.1)” | “−3.71 (−8.19 to 0.76)” |
| AD109 2.5/75 mg | “37.6 (30.0–47.4)” | “28.3 (18.3–39.8)” | “−7.03 (−12.08 to -1.98)” |
| AD109 5/75 mg | “35.9 (25.5–48.9)” | “22.1 (15.9–38.4)” | “−7.11 (−12.08 to -2.14)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-4"></a>

###### Respiratory parameters: AHI4 (4% criterion) in NREM and REM sleep

| Arm | NREM baseline | NREM treatment | NREM placebo-adjusted change | REM baseline | REM treatment | REM placebo-adjusted change |
| --- | --- | --- | --- | --- | --- | --- |
| Placebo | “19.3 (9.4–27.7)” | “16.7 (9.2–27.3)” | Not applicable | “20.9 (9.3–33.9)” | “20.3 (6.1–36.1)” | Not applicable |
| Atomoxetine 75 mg | “18.4 (9.4–28.6)” | “11.4 (4.4–19.4)” | “−4.22 (−7.81 to −0.63)” | “28.1 (15.7–21.6)” | “18.5 (9.1–26.8)” | “−5.77 (−11.66 to 0.71)” |
| AD109 2.5/75 mg | “16.0 (9.6–24.7)” | “10.1 (3.5–16.4)” | “−5.38 (−9.41 to −1.34)” | “30.2 (18.1–50.7)” | “18.2 (7.6–43.5)” | “−5.67 (−12.74 to 1.41)” |
| AD109 5/75 mg | “19.0 (10.1–30.7)” | “8.3 (5.9–19.5)” | “−6.45 (−10.44 to −2.46)” | “22.6 (13.0–34.4)” | “13.8 (6.6–31.8)” | “−3.34 (−10.39 to 3.72)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-5"></a>

###### REM sleep, sleep position, and the 3% or arousal hypopnea definition in the text

> “The effects of AD109 on AHI4 were statistically significant in non-REM (NREM) sleep and in the supine and lateral positions. AD109 also decreased the AHI when hypopneas were scored using ⩾3% desaturation or arousal criteria (P < 0.01 for both doses).”
> “Although AD109 2.5/75 mg significantly reduced AHI4 during REM sleep versus baseline (P = 0.044), the reduction was not significant when adjusted for placebo (P = 0.116), possibly because limited data were available during REM sleep during treatment.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-6"></a>

###### Hypoxic burden

> “Compared with placebo, hypoxic burden was reduced by −12.7 (−24.3 to −1.14) %min/h (P = 0.03) with AD109 2.5/75 mg and by −16.6 (−28 to −5.2, P = 0.005) with AD109 5/75 mg.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

| Arm | Hypoxic burden at baseline, geometric mean (95% CI), %min/h | Hypoxic burden on treatment, geometric mean (95% CI), %min/h |
| --- | --- | --- |
| Placebo | “31.1 (25.8–38.2)” | “30.1 (25.0–38.3)” |
| Atomoxetine 75 mg | “31.9 (27.2–40.5)” | “18.0 (12.4–26.0)” |
| AD109 2.5/75 mg | “31.0 (25.1–41.2)” | “19.1 (11.3–28.2)” |
| AD109 5/75 mg | “32.4 (25.6–41.5)” | “18.0 (12.0–26.1)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-7"></a>

###### Oxygen desaturation index (4% criterion) and time below 90% saturation, placebo-adjusted

| Arm | ODI4, placebo-adjusted change, LS mean (95% CI) | T90, placebo-adjusted change, LS mean (95% CI) |
| --- | --- | --- |
| Atomoxetine 75 mg | “−4.82 (−8.16 to −1.49)” | “−5.56 (−12.67 to 1.56)” |
| AD109 2.5/75 mg | “−5.63 (−9.38 to −1.89)” | “−3.20 (−11.12 to 4.72)” |
| AD109 5/75 mg | “−6.01 (−9.71 to −2.30)” | “−3.68 (−11.50 to 4.15)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-8"></a>

###### Responders: more than 50% reduction in AHI4 (4% criterion)

| Arm | Proportion (95% CI) with more than 50% reduction in AHI4 | Odds ratio versus placebo |
| --- | --- | --- |
| Placebo | “16.6 (8.4–28.3)” | Not applicable |
| Atomoxetine 75 mg | “39.4 (25.8–54.3)” | “3.47 (1.57–7.71)” |
| AD109 2.5/75 mg | “43.2 (26.3–61.4)” | “4.00 (1.68–9.51)” |
| AD109 5/75 mg | “44.3 (27.6–62.0)” | “4.23 (1.80–9.97)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

> “Forty-four percent of participants who received AD109 had a >50% reduction in AHI4. Furthermore, AHI4 decreased to fewer than 10 events per hour in 42% of patients who received AD109.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-9"></a>

###### Resolution to AHI4 below 10 events per hour by baseline severity (4% criterion)

> “However, a higher proportion of participants with a baseline AHI4 in the mild to moderate range (AHI4 10–30 events/h) experienced a resolution of OSA (defined as AHI4 <10 events/h) compared with those with a baseline AHI4 in the severe range (AHI4 ⩾30; 52.2% vs. 7%; P = 0.004; Figure E2 in the online supplement).”
> “In the mITT population, a significant reduction in AHI4 occurred in patients with a baseline AHI4 as high as 45 events per hour, although fewer patients with severe OSA at baseline (AHI4 ⩾30) had an AHI4 <10 during treatment (7%) compared with those with mild (AHI4 10–15, 77%) or moderate (AHI4 15–30, 42%) OSA.”
> “Post hoc analyses showed no statistically significant differences in AHI4 reduction between subgroups based on age, sex, obesity, or baseline OSA severity (Tables E4–E7 in the online supplement).”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-10"></a>

###### Post hoc intention-to-treat analysis of AHI4 (4% criterion)

> “Post hoc analysis of the primary and secondary outcomes using the ITT population and considering zero change from baseline for missing data showed results similar to those in the mITT population. Compared with placebo, AHI4 decreased by −5.9 (−9.4 to −2.4) events/h (P = 0.001) with AD109 2.5/75 mg, by −6.3 (−9.8 to –2.8, P < 0.001) with AD109 5/75 mg, and by −4.3 (−7.5 to −1.2, P = 0.007) with Ato75.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-11"></a>

###### Registry primary outcome: change in AHI4 (4% criterion), combined AD109 arms, mITT

| Group | Participants analyzed | LS mean change from baseline (95% CI), events/hr |
| --- | --- | --- |
| AD109 combined dose arms | “69” | “-7.38 (-9.53 to -5.23)” |
| AD109 75/5mg | “35” | “-7.40 (-10.41 to -4.38)” |
| AD109 75/2.5mg | “34” | “-7.37 (-10.44 to -4.30)” |
| Placebo | “62” | “-0.21 (-2.49 to 2.07)” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="topic-3.1.4.8.3-12"></a>

###### Atomoxetine alone versus AD109: respiratory effect and sleep disruption

> “Ato75 alone reduced AHI4 by −5.19 (−8.6 to −1.8) events/h, or by 38.8% (19.5–53.6) compared with placebo (P = 0.003).”
> “Ato75 did not differ from either dose of AD109 in any respiratory parameter.”
> “AD109 did not differ significantly from atomoxetine alone (Ato75) for most respiratory parameters, suggesting that the noradrenergic activity of atomoxetine is primarily responsible for the effect of AD109 on upper airway obstruction. However, Ato75 was more likely to disturb sleep than AD109 based on objective and subjective measurements.”
> “Thus, Ato75 alone, despite improvements in sleep-disordered breathing, was not sufficiently well tolerated to be a standalone treatment for OSA.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-13"></a>

###### Sleep architecture: total sleep time, REM sleep, and wake after sleep onset

> “Total sleep time (TST) did not change with AD109 compared with placebo. However, Ato75 reduced TST by 21.5 (8.6–34.3) minutes compared with placebo (P = 0.001) and by 23.93 (8.89–38.98) and 16.09 (1.24–30.94) minutes compared with AD109 2.5/75 mg and 5/75 mg, respectively (P = 0.002 and P = 0.03). AD109 and Ato75 increased NREM stages 1 and 2 and reduced REM sleep compared with placebo (P < 0.001 for all).”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

| Parameter and arm | Baseline, median (IQR) | Treatment, median (IQR) | Placebo-adjusted change, LS mean (95% CI) |
| --- | --- | --- | --- |
| TST, min: placebo | “386.4 (362.3–428.0)” | “399.4 (358.5–426.5)” | Not applicable |
| TST, min: atomoxetine 75 mg | “396.4 (364.0–427.0)” | “372.2 (340.5–406.0)” | “−21.46 (−34.3 to −8.59)” |
| TST, min: AD109 2.5/75 mg | “387.5 (352.8–412.9)” | “385.5 (364.8–416.5)” | “2.48 (−12.01 to 16.96)” |
| TST, min: AD109 5/75 mg | “387.9 (360.9–425.8)” | “392.6 (355.5–421.8)” | “−5.37 (−19.65 to 8.91)” |
| REM sleep, %TST: placebo | “17.9 (14.7–20.1)” | “17.2 (13.6–20.8)” | Not applicable |
| REM sleep, %TST: atomoxetine 75 mg | “17.8 (13.7–19.0)” | “6.8 (3.5–11.3)” | “−9.32 (−11.14 to −7.50)” |
| REM sleep, %TST: AD109 2.5/75 mg | “18.6 (14.3–23.0)” | “7.5 (3.7–12.7)” | “−9.74 (−11.78 to −7.69)” |
| REM sleep, %TST: AD109 5/75 mg | “17.3 (13.5–20.1)” | “5.6 (2.5–10.2)” | “−10.79 (−12.79 to −8.78)” |
| WASO, min: placebo | “65.1 (41.5–97.8)” | “58.9 (44.5–101.8)” | Not applicable |
| WASO, min: atomoxetine 75 mg | “70.1 (44.3–98.8)” | “92.9 (61–124.3)” | “20.32 (8.07 to 32.58)” |
| WASO, min: AD109 2.5/75 mg | “75.6 (56.8–104.3)” | “78.9 (56.8–98.3)” | “0.11 (−13.69 to 13.91)” |
| WASO, min: AD109 5/75 mg | “73.3 (43.3–93.5)” | “77.0 (48.3–106.3)” | “8.21 (−5.37 to 21.79)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.3-14"></a>

###### Dose selection for phase 3

> “The present study and the study of Rosenberg and colleagues suggest that the optimal dosage for AD109 may be 2.5/75 mg (21). A lower dose of atomoxetine appears to be less effective in decreasing OSA severity, and a higher dose of aroxybutynin increases antimuscarinic side effects.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="heading-3.1.4.8.4"></a>

###### 3.1.4.8.4 Health-related Quality of Life and Patient-reported Outcomes

<a id="topic-3.1.4.8.4-1"></a>

###### PROMIS-Fatigue improvement with AD109 2.5/75 mg only

> “There was a significant improvement of Patient-Reported Outcome Measurement Information System (PROMIS) fatigue scores with AD109 2.5/75 mg compared with placebo (−3.56 [−6.77 to −0.35]; P < 0.05) and Ato75 (−3.49 [−6.84 to −0.13]; P < 0.05; Figure 4 and Table 5). There was no significant improvement in subjective outcomes with Ato75 versus placebo. However, the PROMIS sleep disturbance scale showed a trend toward increasing (i.e., worsening) with Ato75 compared with placebo (P = 0.056).”
> “It is unclear why we did not see improvement with the higher dose of AD109, although higher rates of daytime symptoms such as dry mouth, urinary hesitation, constipation, and somnolence were reported with the higher dose of aroxybutynin, suggesting more tolerability issues with the AD109 5/75 mg dose.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.4-2"></a>

###### Patient-reported outcomes, placebo-adjusted change from baseline

| Outcome | Atomoxetine 75 mg | AD109 2.5/75 mg | AD109 5/75 mg |
| --- | --- | --- | --- |
| PROMIS fatigue, adjusted T-score | “−0.07 (−2.93 to 2.79)” | “−3.56 (−6.77 to −0.35)” | “0.10 (−3.02 to 3.21)” |
| PROMIS sleep impairment, adjusted T-score | “−0.02 (−3.28 to 3.24)” | “−1.99 (−5.67 to 1.69)” | “−0.35 (−3.91 to 3.21)” |
| PROMIS sleep disturbance, adjusted T-score | “2.40 (−0.07 to 4.87)” | “1.12 (−1.66 to 3.89)” | “0.21 (−2.48 to 2.90)” |
| Modified short SAQLI | “0.54 (−4.29 to 5.37)” | “−2.28 (−7.73 to 3.17)” | “2.02 (−3.25 to 7.29)” |
| ESS score | “0.30 (−1.15 to 1.74)” | “−0.23 (−1.85 to 1.39)” | “0.64 (−0.94 to 2.22)” |
| PGI-S | “−0.25 (−0.73 to 0.23)” | “−0.16 (−0.72 to 0.41)” | “−0.19 (−0.74 to 0.36)” |
| QQ | “0.44 (−1.2 to 2.12)” | “−0.26 (−2.14 to 1.63)” | “−1.01 (−2.85 to 0.83)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.4-3"></a>

###### Subset with patient-reported outcome assessments

> “Medians (IQR) were calculated from mITT set for the subset of patients with an assessment of patient-reported outcomes at baseline and on treatment (placebo, n = 57/62; ato75, n = 48/50; AD109 2.5/75, n = 32/34; AD109, 5/75 n = 35/35).”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.4-4"></a>

###### Placebo response and mild baseline symptoms as explanations offered by the authors

> “The lack of significant improvement compared with placebo on most of the patient-reported outcomes is likely related to the strong placebo effect that is common in such subjective measures (33). Indeed, most measures showed significant change from baseline in all treatment groups, including placebo. The results may also have been affected by the generally mild symptoms reported by our patient population at baseline.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="heading-3.1.4.8.5"></a>

###### 3.1.4.8.5 Safety Results

<a id="topic-3.1.4.8.5-1"></a>

###### Overall adverse events, discontinuations due to adverse events, and no serious adverse events

> “No serious adverse events were reported. A total of 373 adverse events were reported in 104 (49.8%) participants in the safety population: 26 (61.9%) individuals treated with AD109 2.5/75 mg, 32 (78.0%) treated with AD109 5/75 mg, 51 (81.0%) treated with Ato75, and 25 (39.7%) treated with placebo.”
> “Twenty-three (11%) individuals discontinued study participation because of adverse events: 5 (12%) in each of the AD109 groups, 12 (19%) in the atomoxetine group, and 1 (2%) in the placebo group. Events leading to study discontinuation in at least three participants were insomnia, nausea, and dry mouth.”
> “Most events were rated as mild. Two events classified as serious (nausea and migraine) led to study discontinuation in two participants in the Ato75 group.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.5-2"></a>

###### Most common adverse events (at least three participants in any group)

| Adverse event | Placebo (n = 63) | Atomoxetine 75 mg (n = 63) | AD109 2.5/75 mg (n = 42) | AD109 5/75 mg (n = 41) |
| --- | --- | --- | --- | --- |
| Any adverse event | “25 (39.7%)” | “51 (81%)” | “26 (61.9%)” | “32 (78%)” |
| Adverse event leading to discontinuation | “1 (2%)” | “12 (19%)” | “5 (12%)” | “5 (12%)” |
| Dry mouth | “3 (5%)” | “17 (27%)” | “10 (24%)” | “24 (59%)” |
| Insomnia | “2 (3%)” | “23 (37%)” | “11 (26%)” | “9 (22%)” |
| Urinary hesitation/flow decrease | “0” | “14 (22%)” | “3 (7%)” | “9 (22%)” |
| Constipation | “2 (3%)” | “2 (3%)” | “0” | “5 (12%)” |
| Nausea | “2 (3%)” | “4 (6%)” | “5 (12%)” | “4 (10%)” |
| Decreased appetite | “1 (2%)” | “5 (8%)” | “2 (5%)” | “4 (10%)” |
| Feeling jittery | “1 (2%)” | “2 (3%)” | “2 (5%)” | “3 (7%)” |
| Somnolence | “1 (2%)” | “0” | “1 (2%)” | “3 (7%)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.5-3"></a>

###### Dose-dependent dry mouth and insomnia with atomoxetine alone

> “Dry mouth was a common side effect with all three drug regimens, but was much more common with AD109 5/75 mg, suggesting a dose-dependent increase with aroxybutynin”
> “Insomnia was also the most frequently reported adverse event with Ato75 and led to study discontinuation in six individuals (6.9%).”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.5-4"></a>

###### Heart rate and blood pressure versus placebo

> “Placebo-adjusted mean (95% confidence interval) change from baseline in overnight heart rate was slightly higher in all treatment groups: 7.7 (5.7–9.6), 5.1 (2.9–7.4), and 5.5 (3.3–7.1) beats/min for Ato75, Ad109 2.5/75 mg, and AD109 5/75 mg, respectively (Table 7).”
> “Diastolic blood pressure was slightly higher for the AD109 2.5/75 mg group in the evening and in the morning, at 4.7 (1.1–8.3) and 4.2 (0.5–7.9) mm Hg, respectively (Table 7).”
> “However, because there is an increased risk for cardiovascular disease in the OSA population, small increases in heart rate may potentially be more clinically important.”
> “The present trial data do not provide a definitive conclusion or understanding of the AD109 pressor effect in OSA.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

| Arm | Heart rate, overnight, beats/min | Heart rate, morning, beats/min | Systolic BP, morning, mm Hg | Diastolic BP, evening, mm Hg | Diastolic BP, morning, mm Hg |
| --- | --- | --- | --- | --- | --- |
| Atomoxetine 75 mg | “7.7 (5.7 to 9.6)” | “7.0 (3.3 to 10.6)” | “−0.8 (−5.3 to 3.8)” | “1.5 (−1.7 to 4.7)” | “1.2 (−2.1 to 4.5)” |
| AD109 2.5/75 mg | “5.1 (2.9 to 7.4)” | “5.5 (1.4 to 9.6)” | “1.9 (−3.2 to 7.0)” | “4.7 (1.1 to 8.3)” | “4.2 (0.5 to 7.9)” |
| AD109 5/75 mg | “5.5 (3.3 to 7.7)” | “7.4 (3.4 to 11.4)” | “−4.0 (−9.0 to 1.1)” | “0.5 (−3.0 to 4.0)” | “−0.8 (−4.4 to 2.8)” |

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.5-5"></a>

###### Weight and psychometric tests

> “There was no reduction in weight across all treatment arms. Results of the psychometric tests are reported in the online supplement (see Figure E1 and Table E3) and were not different compared with placebo across all treatment arms.”

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/)

<a id="topic-3.1.4.8.5-6"></a>

###### Registry adverse event table, other (not serious) events at a 5% threshold

| Adverse event | AD109 75/5mg | AD109 75/2.5mg | Atomoxetine 75mg | Placebo |
| --- | --- | --- | --- | --- |
| Total, other adverse events | “32/41 (78.05%)” | “26/42 (61.90%)” | “51/63 (80.95%)” | “25/63 (39.68%)” |
| Dry mouth | “24/41 (58.54%)” | “10/42 (23.81%)” | “20/63 (31.75%)” | “10/63 (15.87%)” |
| Middle insomnia | “4/41 (9.76%)” | “8/42 (19.05%)” | “13/63 (20.63%)” | “1/63 (1.59%)” |
| Initial insomnia | “2/41 (4.88%)” | “3/42 (7.14%)” | “7/63 (11.11%)” | “1/63 (1.59%)” |
| Urinary hesitation | “4/41 (9.76%)” | “0/42 (0.00%)” | “5/63 (7.94%)” | “0/63 (0.00%)” |
| Nausea | “4/41 (9.76%)” | “5/42 (11.90%)” | “4/63 (6.35%)” | “2/63 (3.17%)” |
| Constipation | “5/41 (12.20%)” | “0/42 (0.00%)” | “2/63 (3.17%)” | “2/63 (3.17%)” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

<a id="topic-3.1.4.8.5-7"></a>

###### Registry serious adverse events

| Measure | AD109 75/5mg | AD109 75/2.5mg | Atomoxetine 75mg | Placebo |
| --- | --- | --- | --- | --- |
| Serious adverse events | “0/41 (0.00%)” | “0/42 (0.00%)” | “0/63 (0.00%)” | “0/63 (0.00%)” |

Source: [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results)

Additional sources: [Messineo et al. (2023)](https://publications.ersnet.org/lookup/doi/10.1183/13993003.congress-2023.PA3594)

<a id="heading-3.1.4.9"></a>

##### 3.1.4.9 Study Limitations

<a id="topic-3.1.4.9-1"></a>

###### Summary

- Design: phase 2 dose-finding trial with 4 weeks of dosing; the authors describe its exploratory nature, and no adjustment for multiple comparisons was made.
- Population: AHI4 of 10 to 45 on two baseline polysomnograms, ages 18 to 65 (men) or 18 to 75 (women), and BMI up to 38 (men) or 40 (women) kg/m2. Exclusions included other clinically important sleep disorders (the prospectus names a history of insomnia), active cardiac disease, and hypertension requiring more than two drugs. PAP intolerance was not required; prior device users enrolled after 2 weeks without use, and 42.9% to 50.2% of each arm reported CPAP use.
- Eligibility scoring: eligibility used site scoring, and endpoints used central scoring, which led to inclusion of 29 participants with baseline AHI4 below 10 after central scoring.
- Primary endpoint differs by source: the publication tests each AD109 dose against placebo, and the registry names the combined AD109 arms against placebo. The registry states the trial was not powered to compare individual doses and that individual dose results should be interpreted descriptively.
- Missing data: the statistical analysis plan had no prespecified imputation, so efficacy used the modified intention-to-treat population (181 of 211 randomized); the intention-to-treat analysis was post hoc.
- Discontinuation: study discontinuation due to adverse events was 5 (12%) in each AD109 group, 12 (19%) with atomoxetine alone, and 1 (2%) with placebo.
- Outcomes: the authors attribute the lack of PRO differences to a placebo effect and mild baseline symptoms, PROs were analyzed in the subset with assessments, and the authors suggest limited REM data during treatment as a possible reason the REM reduction was not significant versus placebo.
- Reporting: registry enrollment of 294 includes two AD504 arms reported separately, and registry results were first posted on 2026-08-12, after study completion on 2022-08-03.
- Source discrepancies: the publication states no serious adverse events were reported and also that two events classified as serious led to discontinuation with atomoxetine alone, while the registry lists 0 serious adverse events in every arm. Dry mouth with placebo is 3 (5%) in the publication and 10/63 (15.87%) in the registry. The prospectus gives a start in December 2021, and the registry and publication give November 29, 2021.
- Sponsor: the prospectus describes MARIPOSA as a trial the company initiated.

Source: [Schweitzer et al. (2023b)](https://pmc.ncbi.nlm.nih.gov/articles/PMC10765395/); [ClinicalTrials.gov NCT05071612](https://clinicaltrials.gov/study/NCT05071612); [ClinicalTrials.gov NCT05071612 results](https://clinicaltrials.gov/study/NCT05071612?tab=results); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.5"></a>

#### 3.1.5 APC-APN-201

<a id="heading-3.1.5.1"></a>

##### 3.1.5.1 Objective

<a id="topic-3.1.5.1-1"></a>

###### Pharmacokinetics, efficacy, tolerability, and safety with concomitant GLP-1 receptor agonists

> “This study evaluated the pharmacokinetics (PK), efficacy, tolerability and safety of AD109 when co-administered with glucagon-like peptide-1 receptor agonists (GLP-1 RA) in participants with OSA and obesity.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="topic-3.1.5.1-2"></a>

###### Registered official title

> “Phase 2 Safety, Efficacy and, Tolerability of Fixed Dose Combination of Aroxybutynin/Atomoxetine (AD109) in Obstructive Sleep Apnea Patients Taking Tirzepatide, Semaglutide or Liraglutide”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.5.2"></a>

##### 3.1.5.2 Location and Study Date

<a id="topic-3.1.5.2-1"></a>

###### Registered study dates and sites

| Field | Quoted record |
| --- | --- |
| Study start (actual) | “2024-05-20” |
| Primary completion (actual) | “2024-12-26” |
| Number of locations | “This study has 4 locations” |
| Alabama site | “Anniston, Alabama, United States, 36207 Pinnacle Research Group, LLC” |
| Florida site | “Jacksonville, Florida, United States, 32256 Clinical Neuroscience Solutions, Inc” |
| North Carolina site | “Monroe, North Carolina, United States, 28112 Monroe Biomedical Research” |
| Tennessee site | “Memphis, Tennessee, United States, 38119 Clinical Neuroscience Solutions, Inc” |

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.5.3"></a>

##### 3.1.5.3 Study Design

<a id="topic-3.1.5.3-1"></a>

###### Open-label, parallel-arm, 6-week phase 2 design

> “This was an open-label, parallel-arm, 6-week phase 2 study of AD109 in participants with mild-to-severe OSA and obesity, administered either alone or while remaining on a stable maintenance dose of a GLP-1 RA (tirzepatide, semaglutide, or liraglutide).”
> “The study consisted of a screening period, 2-day single dose PK period, followed by a 6-week AD109 dosing period.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="topic-3.1.5.3-2"></a>

###### Arms defined by concomitant GLP-1 receptor agonist use

> “Study APC-APN-201 is an open-label, parallel-arm, 6 weeks study of AD109 in participants with OSA and obesity.”
> “All study procedures, including open-label dosing of AD109, are the same for both arms; the arms differ only in concomitant GLP-1 use.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.3-3"></a>

###### Registered design details

| Field | Quoted record |
| --- | --- |
| Allocation | “Randomized” |
| Interventional model | “Parallel Assignment” |
| Masking | “None (Open Label)” |

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.3-4"></a>

###### Sponsor

> “This study was supported by Apnimed, Inc.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.4"></a>

##### 3.1.5.4 Eligibility Criteria

<a id="topic-3.1.5.4-1"></a>

###### Inclusion: age, body mass index, and REI4 range on home sleep apnea testing

> “Between 18-75 years of age at the time of informed consent.”
> “BMI between 27 and 43 kg/m2 for men, or 45 kg/m2 for women, inclusive.”
> “Mean Respiratory Event Index (REI4) (Hypopneas defined by 4% oxygen desaturation) of >5 and <45. REI is determined as a mean number collected across 2 nights of testing initiated at Visit 2; ≤ 25% central or mixed apneas (as proportion of total apneas and hypopneas)”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.4-2"></a>

###### Inclusion: stable GLP-1 receptor agonist dose or eligibility for one

> “One arm consists of patients concomitantly taking tirzepatide, semaglutide or liraglutide, and a second comparator arm consists of patients not taking any GLP drug but who otherwise would meet the criteria for prescribing a GLP-1 (i.e., BMI≥27).”
> “The participants on GLP-1 must be on a stable maintenance dose that has remained unchanged for at least four weeks prior to enrollment.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.4-3"></a>

###### Inclusion: prior OSA devices withheld

> “Participants with a history of using devices for OSA treatment, including Continuous Positive Airway Pressure (CPAP), oral or nasal devices, or positional devices, may enroll as long as the devices have not been used for at least 2 weeks prior to first HSAT and are not used during participation in the study.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.4-4"></a>

###### Exclusion: insomnia, urinary, glaucoma, psychiatric, and bariatric surgery criteria

> “Current bothersome symptoms of insomnia (difficulty initiating or maintaining sleep, as distinct from unrefreshing sleep or other symptoms attributable to OSA).”
> “Current bothersome symptoms of bladder outlet obstruction including difficulty initiating or maintaining urinary flow, straining, or sensation of incomplete bladder emptying”
> “Narrow angle glaucoma.”
> “Attempted suicide within 1 year prior to screening, or current suicidal ideation.”
> “Subjects with a history of gastric weight loss surgery (i.e., sleeve gastrectomy, gastric bypass, adjustable gastric banding, biliopancreatic diversion)”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.4-5"></a>

###### Exclusion: blood pressure threshold

> “Systolic blood pressure up to 140 mm Hg or diastolic blood pressure up to 90 mm Hg is allowed in participants not currently treated with antihypertensives.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.5.5"></a>

##### 3.1.5.5 Treatment

<a id="topic-3.1.5.5-1"></a>

###### AD109 tablet strength

> “AD109, an investigational oral anti-apneic neuromuscular modulator, combines in one tablet the novel antimuscarinic aroxybutynin (2.5 mg) with the selective norepinephrine reuptake inhibitor atomoxetine (75 mg).”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="topic-3.1.5.5-2"></a>

###### Registered arms and administration

> “Twenty patients who are taking GLP-1 at time of enrollment will be treated with AD109”
> “Twenty patients who are not taking a GLP-1 will be treated with AD109 only”
> “Oral administration at bedtime”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.5-3"></a>

###### Permitted GLP-1 receptor agonists

> “Tirzepatide (Mounjaro®, Zepbound®): doses of 2.5mg, 5mg, 7.5mg, 10mg, 12.5mg, 15mg can be given once weekly by subcutaneous injection”
> “Other doses prescribed by a physician are acceptable if a stable dose is maintained for 4 weeks or more.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.5.6"></a>

##### 3.1.5.6 Study Outcomes

<a id="heading-3.1.5.6.1"></a>

###### 3.1.5.6.1 Rationale for Using Incidence of Adverse Events as the Primary Endpoint

<a id="topic-3.1.5.6.1-1"></a>

###### Primary outcome: spontaneously reported adverse events at 6 weeks

> “Primary Objective (Adverse Events)”
> “General safety and tolerability will be assessed between treatment cohorts using the occurrence of spontaneously reported adverse events (AEs) at 6 weeks”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.6.1-2"></a>

###### Efficacy outcome: REI4 responder at week 5 by home sleep apnea test

> “The proportion of participants achieving ≥50% reduction in respiratory event index based on 4% hypopnea desaturation assessed with home sleep apnea test at Week 5”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.7"></a>

##### 3.1.5.7 Statistical Analysis Description

<a id="heading-3.1.5.7.1"></a>

###### 3.1.5.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.5.7.1-1"></a>

###### Planned enrollment by arm

> “A total of approximately 40 participants will be enrolled in the 2 arms:”
> “Each arm is composed of approximately 20 participants.”

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="topic-3.1.5.7.1-2"></a>

###### Actual enrollment

| Field | Quoted record |
| --- | --- |
| Enrollment (actual) | “Enrollment (Actual) 40” |

Source: [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.5.7.2"></a>

###### 3.1.5.7.2 Description of Analysis Sets

<a id="topic-3.1.5.7.2-1"></a>

###### Pharmacokinetic analysis: noncompartmental analysis and mixed-effects model

> “Plasma PK for aroxybutynin and atomoxetine, including maximum plasma concentration (Cmax) and exposures (AUClast and AUCinf), were derived by noncompartmental analysis and evaluated using a mixed-effects model.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.8"></a>

##### 3.1.5.8 Results

<a id="heading-3.1.5.8.1"></a>

###### 3.1.5.8.1 Participant Disposition

<a id="topic-3.1.5.8.1-1"></a>

###### Participants enrolled and treated by arm

> “Forty participants were enrolled and received AD109 (n=22) or AD109+GLP-1 RA (n=18).”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.8.2"></a>

###### 3.1.5.8.2 Baseline Characteristics

No evidence found.

<a id="heading-3.1.5.8.3"></a>

###### 3.1.5.8.3 Efficacy Results

<a id="topic-3.1.5.8.3-1"></a>

###### REI4 responder rate at week 5 with and without a GLP-1 receptor agonist

> “The proportion of participants achieving ≥50% reduction in respiratory event index based on 4% hypopnea desaturation assessed with home sleep apnea test at Week 5 in the AD109 and AD109+GLP-1 RA arms were similar (33.3% vs 29.4%, respectively).”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="topic-3.1.5.8.3-2"></a>

###### Pharmacokinetics: geometric mean ratios with versus without a GLP-1 receptor agonist

> “Geometric mean ratios (GMR; AD109+GLP-1 RA/AD109 alone [90% CI]) of aroxybutynin for Cmax, AUClast, and AUCinf were 100.6 (68.0–148.8), 142.0 (102.1–197.5), and 152.7 (111.0–210.1), respectively.”
> “GMR of atomoxetine for Cmax, AUClast, and AUCinf were 67.0 (51.0–88.1), 81.9 (57.0–117.7), and 78.2 (51.3–119.1), respectively.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.8.4"></a>

###### 3.1.5.8.4 Health-related Quality of Life and Patient-reported Outcomes

No evidence found.

<a id="heading-3.1.5.8.5"></a>

###### 3.1.5.8.5 Safety Results

<a id="topic-3.1.5.8.5-1"></a>

###### Treatment-emergent adverse events, serious events, and deaths

> “Treatment-emergent adverse events (TEAEs) were reported by 85% of participants with the majority being mild-to-moderate severity.”
> “TEAEs of dry mouth, decreased appetite, nausea, and constipation were reported with higher incidence in the AD109 alone arm.”
> “No clinically meaningful trends were detected in mean changes from baseline in lab values, physical examination variables, or vital signs.”
> “No deaths or serious TEAEs occurred.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="topic-3.1.5.8.5-2"></a>

###### Authors' safety conclusion

> “AD109 was generally safe and well-tolerated in both treatment arms of this phase 2 trial, without large differences in PK when co-administered with GLP-1 RA.”

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293)

<a id="heading-3.1.5.9"></a>

##### 3.1.5.9 Study Limitations

<a id="topic-3.1.5.9-1"></a>

###### Summary

- Design: open-label, 6-week phase 2 study with no placebo arm; the arms differ only in concomitant GLP-1 receptor agonist use, while the registry lists the allocation as randomized.
- Size: 40 participants at 4 U.S. sites, with 22 receiving AD109 alone and 18 receiving AD109 with a GLP-1 receptor agonist, against about 20 planned per arm.
- Population: ages 18 to 75, BMI 27 to 43 (men) or 45 (women) kg/m2, and REI4 above 5 and below 45 on home sleep apnea testing. Exclusions included bothersome insomnia or bladder outlet obstruction, narrow angle glaucoma, and prior gastric weight loss surgery. Prior device users enrolled after 2 weeks without use.
- Endpoints: the primary outcome was spontaneously reported adverse events. Efficacy was the proportion with at least 50% REI4 reduction on home sleep apnea testing at week 5, reported as 33.3% and 29.4%.
- Reporting: results are available only from a congress abstract, without baseline characteristics or PROs.
- Sponsor: the abstract states the study was supported by Apnimed.

Source: [Yao et al. (2026)](https://academic.oup.com/sleep/article/49/Supplement_1/A299/8673293); [ClinicalTrials.gov NCT06400615](https://clinicaltrials.gov/study/NCT06400615)

<a id="heading-3.1.6"></a>

#### 3.1.6 APC-004

<a id="heading-3.1.6.1"></a>

##### 3.1.6.1 Objective

<a id="topic-3.1.6.1-1"></a>

###### Effect of two AD109 doses on hypoxic burden and AHI over a single night

> “We evaluated the effect of the combination (named AD109) on the sleep apnea-specific hypoxic burden (HB, primary outcome) and on AHI (secondary outcome) over a single treatment night.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.1-2"></a>

###### Rationale for testing a lower atomoxetine dose in mild to low-moderate OSA

> “In this trial, we studied patients with mild or low-moderate OSA (AHI between 5 and 20 events/h), as we hypothesized that this group of patients could be treated with a lower dose of the combination.”
> “Specifically, a lower dose of atomoxetine could help to reduce side effects and improve long-term tolerance in this population if the efficacy was comparable to the higher dose.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.1-3"></a>

###### Dose finding as described by the company

> “Prior studies indicated that doses of atomoxetine lower than 75mg were less effective in decreasing airway obstruction in OSA patients with baseline disease severity in the moderate to very severe range.”
> “We concluded that testing lower doses of atomoxetine in AD109 was most likely to be informative in patients with less severe baseline OSA.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.6.2"></a>

##### 3.1.6.2 Location and Study Date

<a id="topic-3.1.6.2-1"></a>

###### Registered study dates and sites

| Field | Quoted record |
| --- | --- |
| Study start (actual) | “2020-12-22” |
| Primary completion (actual) | “2021-05-31” |
| Number of locations | “This study has 3 locations” |
| California site | “San Diego, California, United States, 92103 Pacific Research Network” |
| Georgia site | “Atlanta, Georgia, United States, 30328 NeuroTrial Research” |
| Pennsylvania site | “Philadelphia, Pennsylvania, United States, 19063 Brian Abaluck, LLC” |

Source: [ClinicalTrials.gov NCT04631107](https://clinicaltrials.gov/study/NCT04631107)

<a id="topic-3.1.6.2-2"></a>

###### Year initiated and number of sites

> “We initiated a Phase 2a trial in 2020 to perform additional dose finding for AD109, conducted at three sites in the United States.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.6.3"></a>

##### 3.1.6.3 Study Design

<a id="topic-3.1.6.3-1"></a>

###### Randomized, double-blind, placebo-controlled, 3-period single-dose crossover

> “This study was a randomized, double-blind, placebo-controlled, 3-period, single-dose crossover study that evaluated the safety and efficacy of 2 different dose combinations of AD109 (atomoxetine plus R-oxybutynin) in participants with mild to moderate OSA.”
> “Participants who met all enrollment criteria were centrally randomized to receive the following oral treatments prior to a single night PSG, separated by at least a 1-week washout period:”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.3-2"></a>

###### Six registered treatment sequences

> “This is a randomized, 3-period, placebo-controlled, crossover, phase 2 clinical study to examine the efficacy and safety of 2 doses of AD109 vs placebo in patients with mild obstructive sleep apnea.”
> “AD109 Dose 1- 75 mg (A), then AD109 Dose 2- 37.5 mg (B), then Placebo (C).”
> “Placebo (C), then AD109 Dose 2- 37.5 mg (B), then AD109 Dose 1- 75 mg (A)”

Source: [ClinicalTrials.gov NCT04631107](https://clinicaltrials.gov/study/NCT04631107)

<a id="topic-3.1.6.3-3"></a>

###### Standard and lower dose comparison as described by the company

> “APC-004 was a randomized, double blind, placebo-controlled, single-dose crossover trial designed to compare the clinical response and tolerability of the standard dose (aroxybutynin 2.3mg and atomoxetine 75mg) and a lower dose (aroxybutynin 2.3mg and atomoxetine 37.5mg) of AD109 to each other and to placebo in participants with baseline AHI of 5-20, a range including only mild and moderate OSA.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.6.3-4"></a>

###### Funding

> “This study was funded by Apnimed Inc.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.4"></a>

##### 3.1.6.4 Eligibility Criteria

<a id="topic-3.1.6.4-1"></a>

###### Inclusion: age, body mass index, AHI4 of 5 to 20 events/h, and symptoms

> “Participants had to be between 25 and 65 years of age (inclusive) at the screening visit and have a body mass index between 18.5 and 40.0 kg/m2.”
> “The screening PSG had to demonstrate an AHI (hypopneas scoring required a 4% drop in peripheral capillary oxygen saturation [SpO2]) of 5–20 events/h.”
> “Patients had to also report minimal self-reported symptomatology, such as a score ≥ 1 on the Patient Global Impression of severity (PGI-S) or history of 1 or more of the following OSA symptoms: snoring or nighttime gasping/choking, daytime sleepiness, fatigue, or decreased concentration, nonrefreshing sleep, sleep fragmentation (including from nocturia), or morning headache, irritability, decreased mood or libido.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.4-2"></a>

###### Exclusion: other sleep disorders, interacting medications, and unstable disease

> “Patients could not have another active sleep disorder or be on any medication that could affect respiratory, muscle, or neurologic function.”
> “Participants could also not have any other major organ system disease or unstable medical disorders.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.4-3"></a>

###### Exclusion: registered cardiac and CPAP criteria

> “Clinically significant cardiac disease (e.g. rhythm disturbances, coronary artery disease or heart failure or hypertension requiring more than 2 medications for control).”
> “CPAP should not be used for at least 2 weeks prior to the study”

Source: [ClinicalTrials.gov NCT04631107](https://clinicaltrials.gov/study/NCT04631107)

<a id="topic-3.1.6.4-4"></a>

###### Exclusion: insomnia, psychiatric, neurologic, constipation, and urinary retention criteria as described by the company

> “Main exclusion criteria were a history of insomnia or narcolepsy, craniofacial malformations clinically significant cardiac disease (coronary artery disease, heart failure, rhythm disturbances), psychiatric or neurologic diseases, constipation or urinary retention.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.6.5"></a>

##### 3.1.6.5 Treatment

<a id="topic-3.1.6.5-1"></a>

###### High-dose and low-dose AD109 and placebo at bedtime

> “Each received low-dose AD109 (37.5/2.5 mg), high-dose AD109 (75/2.5 mg), and placebo at bedtime across 3 overnight periods in a randomized order.”
> “Dosing of the study treatment occurred immediately prior to lights out.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.5-2"></a>

###### Aroxybutynin 2.5 mg with two atomoxetine doses

> “in this trial we tested the effect of aroxybutynin, a purified version of the R-enantiomer not yet commercially available, at the dose of 2.5 mg, combined with 2 doses of atomoxetine (37.5 and 75 mg) on OSA severity in a randomized, placebo-controlled, crossover trial.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.6"></a>

##### 3.1.6.6 Study Outcomes

<a id="heading-3.1.6.6.1"></a>

###### 3.1.6.6.1 Rationale for Using Hypoxic Burden as the Primary Endpoint

<a id="topic-3.1.6.6.1-1"></a>

###### Authors' rationale: hypoxic burden as a predictor of cardiovascular morbidity and mortality

> “In this trial, we chose the sleep apnea-specific HB instead of the more traditional AHI as our primary outcome.”
> “HB has been recently introduced as a new metric to stratify OSA severity.”
> “Analysis of such studies showed that the HB is a better predictor of the cardiovascular morbidity and mortality linked to OSA compared to AHI.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.6.1-2"></a>

###### Hypoxic burden definition in the publication: all apneas and hypopneas regardless of desaturation

> “The primary objective outcome was the HB of sleep apnea, which is calculated as the area under the curve of the baseline oxygen saturation and represents the real ventilatory deficit of the disorder as it considers the frequency, length, and the depth of the obstructive events, whereas AHI simply represents the frequency of the same events.”
> “For each apnea or hypopnea (regardless of associated desaturation), the pre-event baseline saturation was defined as the maximum SpO2 during the 100 seconds prior to the end of the event and the area under this baseline value is calculated over a subject-specific search window, which was defined individually for each study, by averaging the SpO2 curve with respect to the end of all events occurring in the study.”
> “The hypoxic burden was then obtained by adding these individual desaturation areas and dividing the total area by the sleep duration, with the units of hypoxic burden being (%min)/h.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.6.1-3"></a>

###### Hypoxic burden definition in the registry: events with 4% or greater desaturation

> “Change in Hypoxic Burden (HB) is calculated as the oxygen desaturation 'area under the curve' in association with individual apneas and hypopneas.”
> “Raw measurement of HB (%min/h) is defined as the sum of individual areas (%min) divided by total sleep time (h). Events with 4% or greater desaturations were included in the calculation of HB.”

Source: [ClinicalTrials.gov NCT04631107](https://clinicaltrials.gov/study/NCT04631107)

<a id="topic-3.1.6.6.1-4"></a>

###### Primary and key secondary endpoints as described by the company

> “The primary endpoint was the reduction of HB (calculated based on AHI with four percent desaturation) and the reduction of AHI (calculated based on four percent desaturations) was a key secondary endpoint.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.6.6.1-5"></a>

###### Editorial comment on hypoxic burden as the primary endpoint

> “The primary endpoint of hypoxic burden is a laudable choice that highlights the movement away from solely relying on AHI.”
> “Further validation of this parameter is needed, but inclusion of hypoxic burden in the efficacy of new therapies is warranted and adds to the strength of the findings.”

Source: [Bierer (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713911/)

<a id="topic-3.1.6.6.1-6"></a>

###### Other outcomes: polysomnography scoring, sleepiness, and vital signs

> “Standard overnight PSGs were recorded and scored in accordance with the American Academy of Sleep Medicine scoring manual.”
> “Effects on situational, self-reported sleepiness were assessed by the Karolinska Sleepiness Scale, administered on the morning following each PSG test.”
> “On the following morning, blood pressure and heart rate were measured in the seated posture.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.7"></a>

##### 3.1.6.7 Statistical Analysis Description

<a id="heading-3.1.6.7.1"></a>

###### 3.1.6.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.6.7.1-1"></a>

###### Planned sample size and power

> “A total of 30 participants was planned to enter the 3-treatment crossover study.”
> “The study was planned to have 80% power to detect a treatment difference on the primary endpoint at a 2-sided 0.05 significance level, if the true difference of hypoxic burden between treatments was at least -3.7 (%min)/h.”
> “This assumed that the within-participant standard deviation was 9 and a discontinuation rate of less than 20%.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.7.1-2"></a>

###### Participants analyzed for the primary outcome in the registry

| Field | Quoted record |
| --- | --- |
| Enrollment (actual) | “Enrollment (Actual) 32” |
| Participants analyzed for the primary outcome, each comparison | “Overall Number of Participants Analyzed 27 27 27” |

Source: [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results)

<a id="heading-3.1.6.7.2"></a>

###### 3.1.6.7.2 Description of Analysis Sets

<a id="topic-3.1.6.7.2-1"></a>

###### Safety population

> “Thirty-one participants were included in the safety population (participants who are randomized and receive at least 1 dose of any of the study treatments).”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.7.2-2"></a>

###### Statistical methods: repeated-measures model, log transformation, and three comparisons per endpoint

> “In general, continuous variables were summarized by presenting medians and interquartile ranges due to the skewed distribution of most variables.”
> “Due to the known logarithmic distribution of HB, the primary analysis of HB was performed also after log10 (HB + 1) transformation.”
> “Each endpoint for this study implicitly contained 3 hypothesis tests based on comparisons between study treatments: (1) 75/2.5 dose vs placebo, (2) 37.5/2.5 dose vs placebo, (3) 37.5/2.5 dose vs 75/2.5 dose.”
> “A repeated-measures model was used to compare log10(HB+ 1), HB (without log-transformation), AHI, and other PSG variables between treatment arms.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.7.2-3"></a>

###### Safety analysis methods

> “The analyses were descriptive and were based on the safety set.”
> “The safety assessments were summarized by the study treatment the participant received within each period.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.8"></a>

##### 3.1.6.8 Results

<a id="heading-3.1.6.8.1"></a>

###### 3.1.6.8.1 Participant Disposition

<a id="topic-3.1.6.8.1-1"></a>

###### Screening, randomization, completion, and withdrawals

> “A total of 65 participants signed the informed consent form and were enrolled in the study.”
> “Thirty-three participants failed screening and 32 participants were randomized (Figure 1).”
> “One ineligible participant was inadvertently randomized but not treated.”
> “Four participants were treated but terminated before completing the study.”
> “Three patients withdrew for personal reasons and one for an adverse event (severe oxygen desaturation during sleep), which raised the safety risk according to one of the study investigators.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.1-2"></a>

###### Registry participant flow

> “A total of 65 participants were screened in the study. Thirty-two participants were randomized.”
> “Twenty-seven participants completed the study. Four participants were treated but early terminated from the study.”

Source: [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results)

<a id="topic-3.1.6.8.1-3"></a>

###### Completion as reported by the company

> “Thirty-two participants were randomized, and 27 (84%) completed the trial.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.6.8.2"></a>

###### 3.1.6.8.2 Baseline Characteristics

<a id="topic-3.1.6.8.2-1"></a>

###### Demographics of the safety population

> “Patient median (interquartile range) age was 53 years (47 to 62) and body mass index was 30.4 kg/m2 (27.0 to 33.1).”
> “There were 11 female patients (35.5% of the total) and participants were mostly white (71%) and African American (23%), with the remaining patients (n = 2) having multiple ethnicities.”
> “Twenty-five (80.6%) participants were not Hispanic or Latino.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.2-2"></a>

###### Baseline polysomnography, median (interquartile range), AHI and ODI with 4% desaturation criterion

| Parameter | Baseline (n = 31) |
| --- | --- |
| Log10 (HB + 1) | “1.11 (0.76–1.32)” |
| HB, (%min)/h | “11.8 (4.8 to 20.0)” |
| AHI, events/h | “13.5 (8.1 to 17.0)” |
| AHI supine, events/h | “18.6 (11.3 to 29.1)” |
| ODI 4%, events/h | “10.0 (6.0 to 15.4)” |

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.2-3"></a>

###### Registry baseline hypoxic burden, median (full range)

| Field | Quoted record |
| --- | --- |
| Hypoxic burden, (%*min)/hour | “11.8 (2.08 to 87.78)” |
| Measure description | “Hypoxic burden is a measure of the overall burden of de-oxygenation (hypoxia) due to upper airway obstructive events for every hour of sleep, measured from the SaO2 signal on the polysomnography done overnight.” |

Source: [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results)

<a id="heading-3.1.6.8.3"></a>

###### 3.1.6.8.3 Efficacy Results

<a id="topic-3.1.6.8.3-1"></a>

###### Hypoxic burden and AHI4 on a single night: high and low dose versus placebo

> “Median [interquartile range] hypoxic burden for participants on placebo was 13.9[4.5–21.9] (%min)/h vs 2.3[0.1–10.5] (%min)/h for patients on the high dose (P < .001) and to 7.3[2–12.5] (%min)/h on the low dose (P < .01).”
> “Apnea-hypopnea index went from a median of 13.2[8.0–19.1] events/h on placebo reduced to 5.5[2.2 to 9.6] events/h on the high dose (P < .001) and to 7.8[4–13.7] events/h on the low dose (P < .05).”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.3-2"></a>

###### Respiratory polysomnography parameters by treatment, median (interquartile range), AHI and ODI with 4% desaturation criterion

| Parameter | Placebo (n = 29) | AD109 low dose (n = 28) | AD109 high dose (n = 30) |
| --- | --- | --- | --- |
| Log10 (HB + 1) | “1.17 (0.74–1.36)” | “0.92 (0.47–1.13)” | “0.51 (0.05–1.06)” |
| HB, (%min)/h | “13.9 (4.5 to 21.9)” | “7.3 (2.0 to 12.5)” | “2.3 (0.1 to 10.5)” |
| AHI, events/h | “13.2 (8.0 to19.1)” | “7.8 (4.0 to 13.7)” | “5.5 (2.2 to 9.6)” |
| AHI supine, events/h | “19.8 (13.2 to 29.7)” | “12.4 (8.1 to 25.0)” | “9.2 (4.4 to 15.7)” |
| AHI NREM, events/h | “7.9 (4.0 to 13.0)” | “7.2 (3.0 to 12.3)” | “3.5 (1.7 to 7)” |
| ODI 4%, events/h | “11.1 (4.5 to 15.6)” | “7.1 (4.2 to 13.4)” | “6.4 (1.4 to 10.8)” |

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.3-3"></a>

###### Dose response and hypoxic burden responders

> “AHI and Log10(HB + 1) were also significantly improved on AD109 high dose compared to low dose, suggesting a dose-related effect size.”
> “Note that > 70% of patients had a reduction of at least 30% in HB and > 40% of participants had a reduction in HB > 90% on high-dose AD109.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.3-4"></a>

###### AHI4 by body position and sleep stage

> “The effect on the AHI was maintained when analysis was limited to the supine position and was present in both non-rapid eye movement and rapid eye movement (REM) sleep, although the reduction in AHI did not reach statistical significance in REM (P = .12 on high dose vs placebo) probably because only 13 participants had REM on both placebo and high dose AD109, and thus the analysis lacked of statistical power.”
> “There was no relationship between the reduction in REM sleep and the reduction in overall AHI for both the low dose (P = .12, r2 = 0.11) or the high dose of AD109 (P = .56, r2 = 0.01).”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.3-5"></a>

###### Registry primary outcome: mean difference in hypoxic burden (95% confidence interval), events with 4% or greater desaturation

| Comparison | Log10 HB, difference in log10HB[(%*min)/h] | Raw HB, (%*min)/h |
| --- | --- | --- |
| AD109 dose 1 (atomoxetine 75 mg) vs placebo | “-0.40 (-0.54 to -0.25)” | “-6.90 (-10.20 to -3.60)” |
| AD109 dose 2 (atomoxetine 37.5 mg) vs placebo | “-0.21 (-0.36 to -0.07)” | “-4.70 (-8.01 to -1.38)” |
| Dose 2 vs dose 1 | “0.18 (0.04 to 0.33)” | “2.20 (-1.23 to 5.36)” |

Source: [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results)

<a id="topic-3.1.6.8.3-6"></a>

###### Percentage reductions in hypoxic burden and AHI4 versus placebo, as reported in the company prospectus

> “Standard dose AD109 significantly reduced HB by 84% (p<0.0001) and AHI by 59% (p<0.0001) relative to placebo.”
> “Low dose also significantly reduced HB by 47% (p=0.0048) and AHI by 41% (p<0.05) vs placebo.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.6.8.4"></a>

###### 3.1.6.8.4 Health-related Quality of Life and Patient-reported Outcomes

<a id="topic-3.1.6.8.4-1"></a>

###### Karolinska Sleepiness Scale the morning after dosing

> “Subjectively, the proportion of nonsleepy patients, defined by a Karolinska Sleepiness Scale < 5, was 0.83 (95% confidence interval 0.69 to 0.97) on placebo, 0.93 (0.83 to 1.00) on low-dose AD109, and 0.86 (0.74 to 0.99) on high-dose AD109.”
> “These different outcomes between treatments were not statistically significant.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.8.5"></a>

###### 3.1.6.8.5 Safety Results

<a id="topic-3.1.6.8.5-1"></a>

###### Treatment-emergent adverse events in at least 2 participants, safety population

| System organ class (preferred term) | High dose (n = 30) | Low dose (n = 28) | Placebo (n = 29) | Overall (n = 31) |
| --- | --- | --- | --- | --- |
| All body systems | “7 (23.3%)” | “6 (21.4%)” | “7 (24.1%)” | “13 (41.9%)” |
| Gastrointestinal disorders (dry mouth) | “2 (6.7%)” | “1 (3.6%)” | “0” | “3 (9.7%)” |
| Metabolism and nutrition disorders (decreased appetite) | “2 (6.7%)” | “0” | “0” | “2 (6.5%)” |
| General disorders and administration site (thirst) | “0” | “1 (3.6%)” | “2 (6.9%)” | “2 (6.5%)” |

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.5-2"></a>

###### Most frequent adverse events and authors' safety assessment

> “The most frequently reported treatment emergent adverse effects occurring in at least 2 study participants were dry mouth (3 [9.7%] participants), decreased appetite (2 [6.5%] participants), and thirst (2 [6.5%] participants).”
> “Both high-dose AD109 and low-dose AD109 were safe and well-tolerated, without clear difference between the high and low doses or between the active doses and placebo.”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.5-3"></a>

###### Registry adverse events: deaths, serious adverse events, and other adverse events at a 5% threshold

| Event | AD109 dose 1 (75 mg) | AD109 dose 2 (37.5 mg) | Placebo |
| --- | --- | --- | --- |
| All-cause mortality | “0/30 (0.00%)” | “0/28 (0.00%)” | “0/29 (0.00%)” |
| Other adverse events, total | “4/30 (13.33%)” | “2/28 (7.14%)” | “2/29 (6.90%)” |
| Dry mouth | “2/30 (6.67%)” | “1/28 (3.57%)” | “0/29 (0.00%)” |
| Decreased appetite | “2/30 (6.67%)” | “0/28 (0.00%)” | “0/29 (0.00%)” |

Source: [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results)

<a id="topic-3.1.6.8.5-4"></a>

###### Sleep architecture and arousals

> “As shown in Table 2, a high dose of AD109 reduced total sleep time and sleep efficiency and increased wake after sleep onset compared to placebo.”
> “In addition, both doses reduced REM sleep and increased N1.”
> “The high dose of AD109 reduced the respiratory arousals, but both doses increased the number of spontaneous arousals (ie, not associated with respiratory events or leg movements), so that the total number of arousals was increased on AD109 (P < .05).”

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="topic-3.1.6.8.5-5"></a>

###### Sleep and morning vital signs by treatment, median (interquartile range)

| Parameter | Placebo | Low dose | High dose |
| --- | --- | --- | --- |
| Total sleep time (min) | “406 (372 to 430)” | “378 (333 to 408)” | “361 (325 to 392)” |
| Wake after sleep onset (min) | “61.0 (45.0 to 97.5)” | “94.5 (65.5 to 127.3)” | “108.0 (77.5 to 148.0)” |
| Stage REM (%TST) | “19.3 (16.6 to 23.1)” | “4.0 (0.0 to 8.1)” | “1.3 (0.0 to 3.7)” |
| Heart rate (beats per minute) | “75.0 (65.0 to 82.0)” | “76.5 (64.5 to 86.0)” | “74.00 (64.0 to 82.0)” |
| Systolic blood pressure (mm Hg) | “127.0 (118.0 to 137.0)” | “127.5 (119.0 to 135.5)” | “124.0 (120.0 to 135.0)” |
| Diastolic blood pressure (mm Hg) | “78.0 (71.0 to 83.0)” | “78.5 (72.0 to 86.0)” | “77.0 (68.0 to 83.0)” |

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/)

<a id="heading-3.1.6.9"></a>

##### 3.1.6.9 Study Limitations

<a id="topic-3.1.6.9-1"></a>

###### Summary

- Design: single-dose crossover with one night of polysomnography per treatment, separated by washouts of at least 1 week.
- Population: AHI4 of 5 to 20 events/h, ages 25 to 65, and BMI 18.5 to 40.0 kg/m2, with exclusion of other active sleep disorders, medications affecting respiratory, muscle, or neurologic function, and unstable medical disorders (the prospectus names a history of insomnia). The registry describes the population as mild obstructive sleep apnea, and the publication as mild or low-moderate OSA.
- Size: 32 participants were randomized, 31 were treated, and 27 were analyzed for each primary comparison in the registry.
- Hypoxic burden definition differs by source: the publication includes each apnea or hypopnea regardless of associated desaturation, and the registry includes events with 4% or greater desaturation. An accompanying editorial states that further validation of the measure is needed.
- REM analysis: only 13 participants had REM sleep on both placebo and high-dose AD109, which the authors state left the REM analysis without statistical power.
- Funding: the publication states the study was funded by Apnimed.

Source: [Rosenberg et al. (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713910/); [ClinicalTrials.gov NCT04631107](https://clinicaltrials.gov/study/NCT04631107); [ClinicalTrials.gov NCT04631107 results](https://clinicaltrials.gov/study/NCT04631107?tab=results); [Bierer (2022)](https://pmc.ncbi.nlm.nih.gov/articles/PMC9713911/); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7"></a>

#### 3.1.7 APC-003

<a id="heading-3.1.7.1"></a>

##### 3.1.7.1 Objective

<a id="topic-3.1.7.1-1"></a>

###### Contribution of each component to the efficacy of AD109

> “Phase 2 Randomized Double-Blind Placebo-Controlled 4-Period Single-Dose Crossover Factorial Study to Evaluate the Contribution of the Individual Drug Components to the Efficacy of AD109 in Obstructive Sleep Apnea”
> “This is a randomized, 4-period, placebo-controlled, crossover, phase 2 clinical study to examine the efficacy and safety of AD109 versus its individual components or placebo in patients with obstructive sleep apnea.”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.1-2"></a>

###### Objective as described by the company

> “We initiated APC-003, a Phase 2 clinical trial, in 2020, designed to demonstrate the individual contribution of aroxybutynin and atomoxetine to the efficacy of AD109.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.2"></a>

##### 3.1.7.2 Location and Study Date

<a id="topic-3.1.7.2-1"></a>

###### Registered study dates

| Field | Quoted record |
| --- | --- |
| Study start (actual) | “2020-10-27” |
| Primary completion (actual) | “2021-05-28” |

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.2-2"></a>

###### Seven United States sites

> “APC-003 was a Phase 2 double-blind, placebo-controlled single dose crossover clinical trial with 60 participants conducted at seven sites in the United States.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.3"></a>

##### 3.1.7.3 Study Design

<a id="topic-3.1.7.3-1"></a>

###### Four-period single-dose randomized crossover

> “The study is a four-period single-dose randomized crossover design in which patients will undergo overnight polysomnographic (PSG) testing with dosing of one of the following 4 treatments: AD109, atomoxetine, R-oxybutynin or placebo.”
> “Participants will return 2 weeks after their final crossover PSG for an end of study (EOS) visit.”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.3-2"></a>

###### Double-blind crossover with one-week washouts and no replacement of withdrawals

> “This was a double-blind, single dose crossover study.”
> “Overall study duration was up to 14 weeks. Participants who withdrew from the study were not replaced.”
> “Each participant received single doses of each of the four treatments, the order in which each treatment was assigned was random.”
> “Washout (One Week)”

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.3-3"></a>

###### Four-week open-label extension

> “APC-003-OLE was a four-week OLE trial of APC-003.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.4"></a>

##### 3.1.7.4 Eligibility Criteria

<a id="topic-3.1.7.4-1"></a>

###### Registered inclusion: age and AHI

> “Male or Female, between 25 to 65 years of age, inclusive, at the Screening Visit.”
> “AHI 10 to <20, or AHI ≥20 if meets other PSG criteria”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.4-2"></a>

###### Registered exclusion: cardiac disease, hypertension, and CPAP use

> “Clinically significant cardiac disease (e.g., rhythm disturbances, coronary artery disease or cardiac failure) or hypertension requiring more than 2 medications for control.”
> “CPAP should not be used for at least 2 weeks prior to first study PSG”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.4-3"></a>

###### Inclusion and exclusion as described by the company: AHI 10 to 45 events/h and low upper airway collapsibility

> “Patients between 25 and 65 years were included if they showed an AHI between ten and 45 events per hour.”
> “Patients with AHI under 45 could be randomized only if they showed signs of low upper airway collapsibility, including mean desaturation of oxygen related to upper airway obstruction less than 8% and prevalence of hypopneas over apneas.”
> “Main exclusion criteria were a history of insomnia or narcolepsy, craniofacial malformations, clinically significant cardiac disease (coronary artery disease, heart failure, rhythm disturbances), psychiatric or neurologic diseases, constipation or urinary retention.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.5"></a>

##### 3.1.7.5 Treatment

<a id="topic-3.1.7.5-1"></a>

###### Four single-dose treatments in oral capsules

> “Oral capsules were administered before sleep in random order of each of the following: AD109 (Treatment A: Atomoxetine 75 mg and R-Oxbutynin 2.5 mg), Treatment B: Atomoxetine 75 mg and Placebo, Treatment C:R-Oxbutynin 2.5 mg and placebo, Treatment D:Placebo and Placebo.”

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.5-2"></a>

###### Doses as described by the company, crossover and open-label extension

> “Participants were dosed with AD109 (2.3mg aroxybutynin and 75mg atomoxetine), atomoxetine alone (75mg), aroxybutynin alone (2.3mg) or placebo.”
> “Of the 60 participants in APC-003, 37 participated in the OLE and received AD109 for one week at the run-in dose of 2.3mg aroxybutynin and 37.5mg atomoxetine, followed by three weeks of AD109 at the target dose of 2.3mg aroxybutynin and 75mg atomoxetine.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.6"></a>

##### 3.1.7.6 Study Outcomes

<a id="heading-3.1.7.6.1"></a>

###### 3.1.7.6.1 Rationale for Using Hypoxic Burden as the Primary Endpoint

<a id="topic-3.1.7.6.1-1"></a>

###### Registered primary outcome: change in log10 hypoxic burden, events with 4% or greater desaturation

> “Change in Hypoxic Burden (HB) Log10HB[(%*Min)/Hour]”
> “Change in Hypoxic Burden (HB) is calculated as the oxygen desaturation 'area under the curve' in association with individual apneas and hypopneas.”
> “Events with 4% or greater desaturations were included in the calculation of HB.”
> “1 night (treatment duration) (0-8 hours collected continuously)”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="topic-3.1.7.6.1-2"></a>

###### Primary and key secondary endpoints as described by the company

> “The primary endpoint of the trial was reduction of HB and AHI was a key secondary efficacy endpoint.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.7.6.1-3"></a>

###### Company rationale for oxygenation measures alongside AHI

> “There are additional measures of oxygen levels that measure the amount of insufficient oxygen levels over time, such as the Oxygen Desaturation Index (ODI) or HB, which may provide additional insight into potential benefit on serious cardiovascular problems that develop or worsen over time.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.7.6.1-4"></a>

###### Other registered assessment: Digit Symbol Substitution Test

> “a neuropsychological test sensitive to brain damage, dementia, age and depression where numbers are assigned symbols.”
> “The number of correct symbols within the allowed time is measured.”

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394)

<a id="heading-3.1.7.7"></a>

##### 3.1.7.7 Statistical Analysis Description

<a id="heading-3.1.7.7.1"></a>

###### 3.1.7.7.1 Number of Participants (Planned and Analyzed)

<a id="topic-3.1.7.7.1-1"></a>

###### Enrolled and analyzed participants in the registry

| Field | Quoted record |
| --- | --- |
| Enrollment (actual) | “Enrollment (Actual) 60” |
| Participants analyzed for the primary outcome | “Fifty-nine (98.3%) participants were included in the mITT Population” |

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="heading-3.1.7.7.2"></a>

###### 3.1.7.7.2 Description of Analysis Sets

<a id="topic-3.1.7.7.2-1"></a>

###### Modified intention-to-treat set for efficacy and safety reporting by treatment

> “The efficacy was analyzed using the mITT Set comprised of all participants who took at least 1 dose of any of the study treatments and had at least 1 measurement for the primary endpoint.”
> “The study analysis did not analyze data by period. AEs were reported for the safety analysis by number and percent for treatment group only.”

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="heading-3.1.7.8"></a>

##### 3.1.7.8 Results

<a id="heading-3.1.7.8.1"></a>

###### 3.1.7.8.1 Participant Disposition

<a id="topic-3.1.7.8.1-1"></a>

###### Screening, randomization, and participant flow by period

> “Seven clinical sites screened 146 participants, 60 participants were randomized.”

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.8.1-2"></a>

###### Participants starting and completing each period, with reasons for not completing

| Period | Started | Completed | Reasons not completed |
| --- | --- | --- | --- |
| First intervention | “Started 60” | “Completed 57” | “Incorrectly Randomized 1 Lost to Follow-up 1 Physician Decision 1” |
| Second intervention | “Started 56” | “Completed 56” | “Not Completed 0” |
| Fourth intervention | “Started 49” | “Completed 49” | “Not Completed 0” |

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.8.1-3"></a>

###### Open-label extension participation

> “Of the 60 participants in APC-003, 37 participated in the OLE”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.8.2"></a>

###### 3.1.7.8.2 Baseline Characteristics

<a id="topic-3.1.7.8.2-1"></a>

###### Registry baseline characteristics

| Characteristic | All study participants (n = 60) |
| --- | --- |
| Sex | “Female 25 Male 35” |
| Ethnicity | “Hispanic or Latino 16 Not Hispanic or Latino 44” |
| Race | “Asian 1 Native Hawaiian or Other Pacific Islander 0 Black or African American 11 White 47 More than one race 1” |
| Body mass index, median (full range), kg/m^2 | “33 (22 to 39)” |

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.8.2-2"></a>

###### Baseline AHI with 4% desaturation criterion as reported by the company

> “Median baseline AHI (with hypopneas scored when associated with a four percent or greater reduction in SpO2 from baseline) was 29.6 events per hour with an interquartile range of 18.8 to 43.6, a severity range that is typically classified as moderate to severe OSA.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.8.3"></a>

###### 3.1.7.8.3 Efficacy Results

<a id="topic-3.1.7.8.3-1"></a>

###### Registry primary outcome: least squares mean change in log10 hypoxic burden (95% confidence interval), events with 4% or greater desaturation, mITT

| Treatment | Log10HB4%[%min/hour] |
| --- | --- |
| AD109 (n = 59) | “-0.41 (-0.54 to -0.28)” |
| Atomoxetine (n = 59) | “-0.36 (-0.49 to -0.23)” |
| R-oxybutynin (n = 59) | “-0.04 (-0.17 to 0.09)” |
| Placebo (n = 59) | “-0.06 (-0.18 to 0.07)” |

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.8.3-2"></a>

###### Hypoxic burden and AHI4 on the first night and at week 4 of the open-label extension, as reported in the company prospectus

> “The results of these trials demonstrated a 44% reduction (p<0.0001) of HB on the first night of administration of AD109 vs placebo which persisted and actually increased to 65% reduction at week four (p=0.002), suggesting no reduction of clinical activity over time.”
> “Findings for AHI were of similar magnitude, with a 34% reduction (p=0.001) on the first night which increased to 56% reduction (p=0.0245) after four weeks of treatment.”
> “The four-week duration of the OLE also demonstrated the durability of clinical activity over four weeks compared to prior studies over one day.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.8.4"></a>

###### 3.1.7.8.4 Health-related Quality of Life and Patient-reported Outcomes

<a id="topic-3.1.7.8.4-1"></a>

###### Simplified Sleep Apnea Quality of Life Index versus baseline, as reported in the company prospectus

> “OSA symptoms showed statistically significant improvement vs baseline for a simplified version of the Sleep Apnea Quality of Life Index (SAQLI) (p=0.0027), particularly components measuring symptoms of daytime tiredness, waking unrefreshed and bothersome snoring.”
> “The median SAQLI at baseline was 23.0 and decreased to 16.0 at Visit 3, i.e. a 30.4% decrease in SAQLI.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.8.5"></a>

###### 3.1.7.8.5 Safety Results

<a id="topic-3.1.7.8.5-1"></a>

###### Registry adverse events: deaths, serious adverse events, and other adverse events at a 3% threshold

| Event | A: AD109 | B: Atomoxetine 75 mg + placebo | C: R-oxybutynin 2.5 mg + placebo | D: Placebo + placebo |
| --- | --- | --- | --- | --- |
| All-cause mortality | “0/56 (0.00%)” | “0/54 (0.00%)” | “0/54 (0.00%)” | “0/57 (0.00%)” |
| Serious adverse events, total | “1/56 (1.79%)” | “0/54 (0.00%)” | “0/54 (0.00%)” | “0/57 (0.00%)” |
| Atrial fibrillation (serious) | “1/56 (1.79%)” | “0/54 (0.00%)” | “0/54 (0.00%)” | “0/57 (0.00%)” |
| Seizure (serious) | “1/56 (1.79%)” | “0/54 (0.00%)” | “0/54 (0.00%)” | “0/57 (0.00%)” |
| Gastroesophageal reflux disease | “2/56 (3.57%)” | “2/54 (3.70%)” | “0/54 (0.00%)” | “0/57 (0.00%)” |

Source: [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results)

<a id="topic-3.1.7.8.5-2"></a>

###### Single-night crossover: adverse events, withdrawals, and the serious adverse event, as described by the company

> “During the 1-night APC-003 study, the only TEAE occurring in at least two trial participants in a single study arm was gastroesophageal reflux, which occurred in two (3.6%) participants treated with AD109 and two (3.7%) participants treated with atomoxetine 75mg alone.”
> “Two participants withdrew from the study due to AEs (namely pleurisy with elevated white blood cells and”
> “atrial fibrillation), but these AEs were considered to be unrelated to the study drug.”
> “One participant experienced an SAE of possible seizure vs arrhythmia/atrial fibrillation, but because the participant had a previously undisclosed history of arrhythmia with differential diagnosis of seizure, and implantation of a cardiac loop recorder to aid in ongoing efforts to diagnose the condition, the event was considered to be unrelated to the drug.”
> “No drug-related SAEs occurred in the trial.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="topic-3.1.7.8.5-3"></a>

###### Four-week open-label extension: adverse events

> “In the four-week OLE study, the most frequently reported TEAEs occurring in at least two study participants were insomnia (six (16.2%) participants), nausea (three (8.1%) participants), pollakiuria (two (5.4%) participants), and urinary hesitation (two (5.4%) participants).”
> “No serious TEAEs, severe AEs or deaths were reported.”

Source: [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

<a id="heading-3.1.7.9"></a>

##### 3.1.7.9 Study Limitations

<a id="topic-3.1.7.9-1"></a>

###### Summary

- Publication status: results are available only from the registry record and the company prospectus.
- Design: four-period single-dose crossover with one-week washouts; participants who withdrew were not replaced, and 49 participants started the fourth period. The registry states the analysis did not analyze data by period and reported adverse events by treatment group only.
- Eligibility differs by source: the registry states AHI 10 to below 20, or 20 or more if other polysomnography criteria were met, while the prospectus states AHI 10 to 45 events/h with randomization limited to participants showing signs of low upper airway collapsibility. Ages were 25 to 65, and exclusions included a history of insomnia and clinically significant cardiac disease.
- Open-label extension: 37 of 60 participants received AD109 for 4 weeks without a control group; the reported reductions in hypoxic burden and AHI at week 4 and the SAQLI change versus baseline come from the prospectus.
- Withdrawals differ by source: the registry lists first-period non-completion as incorrectly randomized, lost to follow-up, and physician decision (1 each), while the prospectus states two participants withdrew because of adverse events (pleurisy and atrial fibrillation).
- Safety: one serious adverse event (possible seizure versus arrhythmia or atrial fibrillation) occurred with AD109 in a participant with a previously undisclosed history of arrhythmia and was considered unrelated to the drug.
- Sponsor: the prospectus describes APC-003 as a trial the company initiated.

Source: [ClinicalTrials.gov NCT04580394](https://clinicaltrials.gov/study/NCT04580394); [ClinicalTrials.gov NCT04580394 results](https://clinicaltrials.gov/study/NCT04580394?tab=results); [Apnimed prospectus (2026)](https://www.sec.gov/Archives/edgar/data/1745648/000119312526328422/apmd_424b4.htm)

## Related pages
- [AD109 (aroxybutynin-atomoxetine): Obstructive sleep apnea](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea.md)
- [AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: Terms and abbreviations](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/terms-and-abbreviations) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/terms-and-abbreviations.md)
- [AD109 (aroxybutynin-atomoxetine): Obstructive sleep apnea](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/executive-summary) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/executive-summary.md)
- [AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: Product information and disease description](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/product-information-and-disease-description) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/product-information-and-disease-description.md)
- [AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: Economic information and modeling report](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/economics) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/economics.md)
- [AD109 (aroxybutynin-atomoxetine) in Obstructive sleep apnea: References](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/references) · [Markdown](https://www.evicenter.com/drugs/ad109--obstructive-sleep-apnea/references.md)
