BMS ADEPT-2 Continues Cobenfy Alzheimer’s Trial

Key Takeaways

• ADEPT-2 (NCT06126224) is Phase 3 for psychosis associated with Alzheimer’s disease dementia; BMS has not claimed endpoint success.
• BMS excluded data from a small number of sites with trial-execution irregularities before database lock.
• After FDA consultation, an independent interim analysis led the DMC to recommend continued enrollment to the original target.
• ADEPT-2, ADEPT-1, and ADEPT-4 readouts are expected by the end of 2026; Cobenfy remains approved only for adult schizophrenia.

Why does the “meets endpoint” framing fail?

Earlier headlines implied Cobenfy already cleared ADEPT-2. That is not what BMS disclosed. The December 2025 corporate update describes study continuation after irregularities, exclusion of affected site data, and ongoing blinding. For investors and medical affairs teams, treating ADEPT-2 as a positive readout would mis-state the Bristol Myers Squibb Alzheimer's trial status and overstate near-term label expansion risk.

Primary wording is in the BMS news release dated December 3, 2025.

What did Bristol Myers Squibb announce about ADEPT-2?

Following a blinded review, BMS identified irregularities due to clinical trial execution at a small number of sites. Before database lock, the company decided to exclude patient data from those sites from the primary analysis. After consulting the FDA, an independent party ran an interim efficacy and safety analysis that the Data Monitoring Committee reviewed.

The DMC recommended continuing the study by enrolling additional patients to reach the original target population. BMS said it agrees with that path, will recruit more participants, and remains blinded. Laura Gault, head of neuroscience drug development, framed the exclusion decision as protecting study integrity in an area of high unmet need.

How is the Bristol Myers Squibb Alzheimer's trial designed?

ADEPT-2 is a multicenter, randomized, double-blind, placebo-controlled Phase 3 trial of KarXT/Cobenfy in adults with psychosis associated with Alzheimer’s disease dementia. Per ClinicalTrials.gov NCT06126224, the primary objective is efficacy versus placebo on the NPI-C Hallucinations and Delusions score.

• Primary endpoint: change in NPI-C: H+D (maximum combined score 45; higher is worse)
• Key secondary endpoint: Clinical Global Impression-Severity (CGI-S)
• Population: roughly ages 55–90 with mild to severe AD and moderate to severe psychosis
• Status context: recruiting / enrollment expansion after site exclusions

What is already approved for Cobenfy?

Cobenfy (xanomeline and trospium chloride) is approved for schizophrenia in adults. The FDA press announcement describes it as the first antipsychotic for schizophrenia that targets cholinergic receptors rather than dopamine receptors. The agency’s Drug Trials Snapshot for COBENFY lists the original approval date as September 26, 2024, based on two five-week trials in adults with schizophrenia.

That schizophrenia label does not authorize Alzheimer’s psychosis use. Any ADEPT success would be a separate indication pathway requiring its own efficacy, safety, and regulatory package. Until an Alzheimer’s psychosis approval exists, schizophrenia remains the only marketed claim.

Strategic implications if ADEPT later succeeds

If ADEPT-2 and companion ADEPT studies ultimately succeed, Cobenfy could become a first-in-class muscarinic approach for Alzheimer’s-related psychosis, expanding beyond dopamine-blocking antipsychotics that carry black-box warnings in dementia-related psychosis. For Bristol Myers Squibb, a positive ADEPT package would also test whether the schizophrenia launch narrative can translate into a broader neuropsychiatry franchise.

That upside is conditional. Until unblinded Phase 3 data are public, valuation models should treat Alzheimer’s psychosis as an option value, not a de-risked franchise extension. Near-term diligence should track enrollment pace after the site exclusions, whether ADEPT-1 and ADEPT-4 share affected sites, and how FDA dialogue evolves around data integrity before any labeling discussion.

Operational diligence checklist for analysts

Teams covering the Bristol Myers Squibb Alzheimer's trial should separate three facts that often get conflated in secondary coverage. First, ADEPT-2 is ongoing and expanding enrollment. Second, BMS remains blinded and has not published primary-endpoint results. Third, Cobenfy’s approved use is schizophrenia, not dementia-related psychosis. Keeping those layers distinct prevents false-positive signals from entering portfolio models or medical-affairs briefs.

Practical monitoring sources remain the BMS corporate newsroom, ClinicalTrials.gov status changes on NCT06126224, and future FDA labeling updates if an Alzheimer’s psychosis sNDA is filed. Wire reprints that claim “endpoint met” without quoting BMS topline language should be treated as errors.

What remains unproven

BMS has not released positive ADEPT-2 topline results. The nature and patient count of the site irregularities were not quantified in the December release. Endpoint success, safety in elderly dementia patients, caregiver burden outcomes, and FDA approvability for Alzheimer’s psychosis all remain open. Competing narratives that “Cobenfy met the endpoint” should be discarded until a primary disclosure says so.

Related NovaPharma coverage

• Roche Phase 3 trontinemab in Alzheimer’s disease
• AbbVie Cerevel neuroscience pipeline expansion
• Amgen MariTide Phase 2 obesity data

Frequently Asked Questions

Primary Sources

1. BMS: ADEPT-2 study continuation — Dec 3, 2025
2. ClinicalTrials.gov NCT06126224 — ADEPT-2
3. FDA: Cobenfy schizophrenia approval announcement