Retatrutide for Sleep Apnea: The NEJM Trial Results

Short answer: in two placebo-controlled trials published in the New England Journal of Medicine on October 20, 2026, retatrutide reduced the apnea-hypopnea index (AHI), the number of breathing pauses or shallow breaths per hour of sleep, by about 32 to 34 events per hour at the 9 mg and 12 mg doses in adults with obesity and moderate-to-severe obstructive sleep apnea (OSA), against about 10 on placebo, over 80 weeks. In people who also had type 2 diabetes the reduction was smaller, about 25 events per hour against 10. Retatrutide is not approved for sleep apnea or anything else; Lilly has said it will include OSA in its first US application.

What obstructive sleep apnea is

In OSA the upper airway repeatedly narrows or closes during sleep, so breathing stops and restarts many times a night. The US National Heart, Lung, and Blood Institute notes that untreated sleep apnea “increases the risk for stroke, heart attack, and other serious problems.” Severity is measured by AHI on a sleep study: an AHI of 15 or more events per hour is moderate OSA and 30 or more is severe. Excess weight, especially around the neck and abdomen, is the main risk factor that can be changed, which is why weight-loss drugs are being tested for it.

How the trials were designed

The OSA trials were nested inside two of the main Phase 3 obesity trials: TRIUMPH-1-OSA in people without diabetes (part of TRIUMPH-1) and TRIUMPH-2-OSA in people with type 2 diabetes (part of TRIUMPH-2). Both were randomised, double-blind and placebo-controlled, lasted 80 weeks, and compared retatrutide 4 mg, 9 mg or 12 mg once weekly with placebo.

• Who could join: adults with a BMI of 27 or more and a diagnosis of OSA with an AHI of at least 15 on a screening sleep study.
• CPAP use: people not using positive airway pressure (PAP, including CPAP) had to be unable or unwilling to use it. People already using PAP could join but stopped it for about a week before each sleep study, so the AHI measured the airway without the machine.
• Main outcome: change in AHI at 80 weeks for the 9 mg and 12 mg doses.
• Not allowed: wakefulness drugs, sleeping pills, opioids and oral appliances for OSA during the trials.

The results

Change in AHI at 80 weeks (events per hour):

Difference from placebo (95% confidence interval):

These figures use the trials’ main “hybrid” analysis, which treats stopping the drug because it was not working or not tolerated as a treatment failure. The intention-to-treat results, which count everyone as randomised, were almost identical (for example -24.7 and -22.5 for 9 mg and 12 mg in TRIUMPH-1-OSA).

Weight: participants on retatrutide lost 12.5% to 22.2% of body weight in TRIUMPH-1-OSA and 12.1% to 18.5% in TRIUMPH-2-OSA, against 4.8% and 5.6% on placebo.

Side effects: the most common were gastrointestinal, consistent with the rest of the retatrutide program (see our side effects overview). The paper’s full safety tables were not publicly available when this page was written. The wider results of both trials are summarised in Retatrutide’s Published Trials, Explained.

How to read these numbers

The 9 mg dose did as well as 12 mg. In people without diabetes, 9 mg produced the largest AHI drop, and the two confidence intervals overlap almost entirely. Earlier reports, including Lilly’s June 2026 announcement of reductions of “up to 60.6%,” did not make the dose clear; the published paper shows the benefit plateaued between 9 mg and 12 mg.

People with diabetes improved less. This matches the weight data: people with type 2 diabetes lose less weight on incretin drugs, and OSA improvement tracks weight loss. A drop of about 15 events per hour beyond placebo is still clinically meaningful when the starting point is 56.

Most participants started with severe OSA. An average fall of 32 to 34 events per hour from about 59 still leaves many people with residual OSA. The paper’s abstract does not report how many reached normal or mild levels (an AHI under 5 or under 15), and those results will matter to patients and sleep physicians.

How it compares with tirzepatide

Tirzepatide (Zepbound), also made by Lilly, became the first drug approved for OSA, on December 20, 2024, “for the treatment of moderate to severe obstructive sleep apnea (OSA) in adults with obesity.” Its approval rested on the SURMOUNT-OSA trials.

These are different trials, of different lengths, with different populations and analysis methods, so the comparison is indicative only. On these numbers retatrutide’s effect on AHI is in the same range as tirzepatide’s, perhaps somewhat larger. No head-to-head OSA trial exists. In SURMOUNT-OSA, 42% of participants not using PAP reached an AHI under 5, or 5 to 14 with little daytime sleepiness, against 16% on placebo; the equivalent retatrutide figure has not been reported in the abstract.

For the wider comparison see retatrutide vs Zepbound.

Does this replace CPAP?

No. CPAP opens the airway mechanically from the first night and remains the standard treatment for moderate-to-severe OSA. A drug like retatrutide, if approved, would treat the underlying excess weight over many months. In both retatrutide and tirzepatide trials, people with OSA still had residual events on average at the end. Anyone with OSA should keep using their prescribed therapy and only change it after a repeat sleep study and a discussion with their sleep physician.

What we still do not know

• How many participants no longer had moderate or severe OSA, or no OSA at all, at 80 weeks.
• Effects on oxygen levels during sleep (hypoxic burden), daytime sleepiness and sleep-related quality of life, which the trials registered as secondary outcomes.
• Side effects within the OSA groups, including dysesthesia and treatment discontinuation, which were not in the public abstract.
• Whether the benefit lasts after the drug is stopped.

Frequently asked questions

Does retatrutide cure sleep apnea?

Not on average. In the trials it reduced AHI substantially, but participants started with severe OSA and many still had sleep apnea at 80 weeks. The proportion whose OSA resolved has not yet been reported publicly.

Which dose worked best for sleep apnea?

In people without diabetes, 9 mg reduced AHI by 34.3 events per hour and 12 mg by 31.7, a difference well within the trials’ margin of error. In people with diabetes the two doses were nearly identical.

Is retatrutide approved for sleep apnea?

No. Retatrutide is not approved for any condition. Lilly has said its first US application will include obstructive sleep apnea, alongside obesity and knee osteoarthritis pain. Tirzepatide (Zepbound) is the only approved drug for OSA in adults with obesity.

Should I stop my CPAP if I am losing weight?

No. Do not change OSA treatment without a repeat sleep study and your doctor’s advice. Weight loss can reduce OSA severity, but the trials show residual disease is common.

This page is for education and is not medical advice. Retatrutide is an investigational drug that is not approved by the FDA or any other regulator.