Sleep apnoea pill cut breathing events but left fatigue unchanged

Continuous positive airway pressure works for obstructive sleep apnoea, but plenty of people cannot tolerate wearing a mask night after night, and some refuse it outright. That leaves a large group with untreated interrupted breathing and nothing much to fall back on. AD109 is an investigational fixed-dose oral combination of aroxybutynin 2.5 mg and atomoxetine 75 mg, designed to target the neuromuscular dysfunction behind the airway collapsing during sleep. The obvious question is whether a tablet can meaningfully reduce breathing events over months rather than a single night, and whether people feel any different if it does.

SynAIRgy enrolled adults with mild to severe obstructive sleep apnoea who were intolerant of positive airway pressure therapy or had refused it, then assigned them at random to AD109 or a matching placebo for 26 weeks across 69 centres. Neither participants nor investigators knew who was taking what. In total 646 people were randomised, with a median age of 58 years, 49.3% female, a median body mass index of 32.4 kg/m2 and a median starting apnoea-hypopnoea index of 19.6 events an hour; 35% had mild, 42% moderate and 23% severe disease. The main outcome was the change in apnoea-hypopnoea index from baseline to week 26, with oxygen desaturation index, hypoxic burden, fatigue and sleep impairment questionnaires and the share of people halving their apnoea-hypopnoea index as key secondary measures.

At week 26 the difference between groups in breathing events was 4.0 events an hour in favour of the drug, which the model translated into a 44.1% decrease from baseline versus 17.6% on placebo. Oxygen desaturation index and hypoxic burden also improved compared with placebo, so the oxygen picture moved alongside the event count. The fatigue questionnaire told a different story: there was no statistically significant difference between the drug and placebo. Tolerability was a real issue, with 21.2% of people on AD109 discontinuing because of adverse events against 3.1% on placebo, most commonly dry mouth, nausea, insomnia and urinary hesitation. No serious adverse events were judged treatment related.

Drug treatment for obstructive sleep apnoea has long been the field's missing option, with devices, positional therapy, weight change and surgery doing the work instead. This trial shows a tablet can shift the standard breathing measures over six months in exactly the group that cannot use a mask, which is new territory. What it does not yet show is symptom benefit: the gap in fatigue scores was not significant, so the link between fewer events and feeling better remains open. Longer follow-up, harder outcomes such as blood pressure or cardiovascular events, and independent replication away from the manufacturer are all still needed.

For anyone who has abandoned a CPAP machine, this is a reason to think a drug option may eventually exist, not evidence that one is ready to solve the problem. The improvement in breathing events and oxygen measures was real in this trial, yet the average difference was modest and the fatigue that often drives people to seek treatment did not clearly budge. Around one in five participants stopped the drug because of side effects, which matters for anyone weighing tolerability against a mask. Nothing here changes existing treatment for people who get on with their machine.