Auvelity earns approval, but why did an earlier version of the drug fail?
A New Approval
On April 30, 2026, the FDA approved Auvelity, a combination of dextromethorphan with bupropion, for agitaiton in dementia.
This is the second approval for this condition, following the antipsychotic brexpiprazole (Rexulti).
Building on Failed Attempts
This is not the first attempt to approve dextromethorphan for this use. In 2015, Avanir Pharmaceuticals produced positive results with its dextromethorphan-quinidine combo, Neudexta, in a large, randomized trial. The company then developed a different version of the drug, deudextromethorphan-quinidine (replacing hydrogen with deuterium), to extend dextro’s short half-life.
However, this new version (AVP-786) failed to reduce agitation, suggesting deuteration also neutered its pharmacodynamic effects.
The New Trials
The FDA cites two trials in its decision:
- ADVANCE-1 (NCT 03226522), a 5-week trial
- ACCORD-2 (NCT 04947553), a long-term preventative trial that lasted up to 1.5 years
But wait. There are five trials at ClinicalTrials.gov, so let’s dig deeper:
- Two large 5-week trials
- Three long-term preventative trials
One of those long-term trials was terminated early, leaving only four that we hope will see publication. Among the acute, 5-week trials, only one was positive:
- POSITIVE: ADVANCE-1 366 randomized to Auvelity, bupropion, or placebo
- NEGATIVE: ADVANCE-2 408 randomized to Auvelity or placebo
None of these are published, but thanks to Kristen Ward and Leslie Citrome we have the details on ADVANCE 1, the positive trial. Here, only the combo (Auvelity) separated from placebo, and it did so by two weeks. Bupropion alone made no difference. Mean agitation scores (CMAI) fell by 15.4 points with Auvelity, versus 10.0 with bupropion and 11.5 with placebo. 73% responded to Auvelity vs 57% with placebo.
Somnolence (8.2%) and dizziness (6.3%) were the most common side effects exceeding placebo rates. Unlike an earlier study of dextromethorphan-ququinidine, falls were lower with Auvelity (2.5%) than with bupropion (14.3%) or placebo (1.9%).
Prevention and Long-Term Safety
Auvelity preventated agitation in the two ACCORD trials, although the FDA only cited the larger ACCORD-2. Both began by treating Alzheimer’s and agitation with open-label Auvelity in either 178 (ACCORD-1) or 295 (ACCORD-2) patients. Approximately 60% responded, and they were randomized to continue the medication or switch to placebo.
Both trials were successful, with relapse rates of 7.5% vs 25.9% in ACCORD-1 and 8.5% vs 28.6% in ACCORD-2. There was no cognitive decline, and adverse effects were similar for med and placebo.
Why Not Nuedexta?
Auvelity and Nuedexta both have only one positive acute phase-trial for agitation in dementia, so why is only one approved?
Auvelity also worked in randomized withdrawal trials, but that is not so meaningful. Stopping an antidepressant and a glutamatergic could cause agitation even if they don’t treat it.
Nuedexta’s early promise was followed by failure, but the failures concerned a different version of the drug.
The most likely explanation is that the new FDA requires only one well-designed trial for approval instead of two. To its discredit, Auvelity failed in its larger, phase III trial, while passing in the phase II/III trial
Practice Implications
- Auvelity and Nuedexta are reasonable options for agitation in dementia, and both are likely safer than antipsychotics.
- Auvelity contains an antidepressant, which may be beneficial for depressed patients with agitation, but it did not work for agitation and is known to increase the fall risk in older adults.
— Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
