For Sleep Apnea, a Once-Nightly Pill Reduced Breathing Interruptions

CATEGORY: SLEEP & RECOVERY

Can a pill do part of what a CPAP machine does? That’s the question the first completed phase 3 trial of a sleep apnea drug set out to answer. Over six months, a once-nightly capsule called AD109 reduced breathing interruptions by a model-estimated 44%, against 18% with placebo, in 646 adults with obstructive sleep apnea who had refused CPAP or given up on it.¹ The only drug approved for sleep apnea so far works indirectly, through weight loss. This one acts on the breathing machinery itself, and it has now been submitted to the FDA.

A Drug Aimed at the Throat’s Night Shift

In obstructive sleep apnea, the muscles that hold the throat open slacken during sleep and the airway narrows or closes, over and over, all night. Picture a tent whose guy ropes go loose every time the camper nods off. CPAP fixes that with air pressure. AD109 tries chemistry instead: it combines atomoxetine, a drug long used for attention deficit disorder, with aroxybutynin, and together they keep the brain’s daytime signal to those airway muscles partly switched on overnight. The tone stays, so the tent stays up more of the night. Earlier, smaller studies had suggested the combination starts working from the first night of dosing. This trial asked whether the effect holds up over six months of real use.

Four Fewer Interruptions an Hour, and Steadier Oxygen

The trial ran at 69 sites in the United States and Canada and enrolled a broad slice of the apnea population: median age 58, about half women, and a spread of disease from mild (35%) through moderate (42%) to severe (23%). Each of them had been offered CPAP and had either turned it down or abandoned it. Participants started with a median of about 20 breathing interruptions per hour of sleep. Neither they nor their doctors knew who had the real capsule. After 26 weeks, interruptions fell by a mean of 3.3 events per hour on the drug while rising 0.7 on placebo, a gap of 4.0 events per hour (95% CI 1.6 to 6.4, p=0.001). In percentage terms, the model-estimated reduction was 44.1% with AD109 and 17.6% with placebo.¹ Oxygen told the same story: the rate of oxygen dips fell more on the drug, and a measure of total nighttime oxygen shortfall improved markedly. Among people still taking their capsules, 39.6% on the drug achieved at least a halving of their interruptions, versus 22.3% on placebo.¹ When everyone randomized was counted, including those who had quit the capsules, that responder gap narrowed and was no longer statistically clear. Staying on the pill mattered.

What the Pill Didn’t Deliver

Breathing improved. Feeling better wasn’t demonstrated. On the trial’s fatigue questionnaire the drug group’s edge was small and fell just short of statistical significance, and a sleep-impairment score showed no clear difference either.¹ The authors are careful on this point: under the trial’s own statistical rules, once one endpoint misses, the later ones can’t be claimed as proven. Side effects were the other cost. About 21% of people on AD109 stopped because of them, most often dry mouth, nausea, insomnia or trouble urinating, compared with 3% on placebo.¹ The trial reported no serious treatment-related harms.

Where This Leaves the Mask

CPAP remains the standard treatment, and nothing here changes that for people who tolerate it. The point of this trial was everyone else. The researchers designed AD109 for what they describe as a large unmet need: people whose apnea goes untreated because they can’t or won’t sleep in a mask.¹ Untreated apnea isn’t just loud nights and groggy mornings. The repeated oxygen dips strain the heart and blood vessels, which is why the oxygen findings in this trial matter beyond the event counts. Whether fewer interruptions on a pill translate into fewer heart problems down the road is a question this six-month study wasn’t built to answer, and the authors don’t claim otherwise. For people who have already walked away from the mask, a pill that cuts interruptions nearly in half, without yet proving it makes mornings feel different, is a genuine first and an incomplete one. Regulators are weighing the application now.

Key Takeaways

• This was the first completed phase 3 trial of a drug treating obstructive sleep apnea itself, in people who had refused or abandoned CPAP.
• Breathing interruptions fell 44% on the once-nightly pill versus 18% on placebo over six months, with parallel gains in nighttime oxygen levels.
• Symptom scores for fatigue and sleep impairment didn’t clearly improve, so whether the pill makes people feel more rested remains unproven.
• About one user in five stopped over side effects such as dry mouth, nausea and insomnia.

References

1. Strollo PJ Jr, Farkas R, Taranto-Montemurro L, et al. Aroxybutynin and atomoxetine (AD109) for obstructive sleep apnea: a randomized phase 3 trial (SynAIRgy). American Journal of Respiratory and Critical Care Medicine. 2026;212(7):1569–1584. link