Screening and enrollment are under way in the two phase 3 substudies of RADIANT, the trial programme testing remlifanserin (ACP-204) in hallucinations and delusions associated with Alzheimer's disease, months before the programme's phase 2 dose finding result is due. Acadia Pharmaceuticals, the sponsor, said on July 20, 2026 that the FDA had granted the drug fast track designation for that indication, that phase 2 enrollment was complete, and that topline phase 2 results were now expected in September to October 2026; the company repeated both points in its second quarter report on August 4, 2026. There is no treatment approved in the United States specifically for psychosis in Alzheimer's disease.
The trial registration (NCT06159673, sponsor protocol ACP-204-006) describes a master protocol covering three independent, randomized, double-blind, placebo-controlled substudies that enroll seamlessly one after another, with a planned total of 1,074 participants, a start date of November 14, 2023 and an estimated completion date of February 29, 2028. Substudy 1, the phase 2 part, compares remlifanserin 30 mg and 60 mg against placebo; substudies 2A and 2B are the phase 3 confirmatory studies of one or both doses. Eligibility runs from 55 to 95 years of age, in the community or in institutional care, with Alzheimer's disease by 2011 NIA-AA criteria and psychosis by the revised International Psychogeriatric Association criteria. Each substudy has a screening period of up to 49 days, six weeks of blinded treatment and a 30-day safety follow-up for those not entering an open label extension. A design abstract published in July 2025 by the sponsor's investigators puts the phase 2 part at about 318 participants and each phase 3 part at about 378, and gives the primary measure as change from baseline at week six in the total score of the hallucinations and delusions subscales of the Scale for the Assessment of Positive Symptoms.
That same abstract notes that roughly 30 percent of people with Alzheimer's disease have hallucinations or delusions, and that they are often managed with antipsychotics used off label, drugs the authors describe as carrying considerable safety concerns. Acadia has been here before: in April 2021 the company disclosed that the FDA had issued a complete response letter for its application to extend pimavanserin, already approved for Parkinson's disease psychosis, to hallucinations and delusions in dementia-related psychosis, citing among other things a lack of statistical significance in some dementia subgroups and too few patients with less common subtypes.
No efficacy or safety results from RADIANT have been reported. Fast track designation is a procedural mechanism intended to speed development and review of drugs for serious conditions with unmet need; it is not a finding that a drug works. The seamless design means the confirmatory studies begin before the dose response data from phase 2 are known, and the case for the mechanism, selective 5-HT2A inverse agonism without direct dopamine D2 blockade, rests on pharmacology and nonclinical work rather than on any published human result for this compound in this population. A separate phase 2 study of the same drug in Lewy body dementia psychosis (NCT07029581) began recruiting on August 6, 2025 and has an estimated completion date of March 2028.