Bristol Myers Squibb (BMS) has announced a delay in the anticipated data readout for its drug Cobenfy, which is being studied for use in treating psychosis related to Alzheimer’s disease. The results from the Phase III ADEPT-2 trial are now expected in 2026 after BMS identified “irregularities” at a small number of study sites involved in the research.
The company did not disclose details about the nature of these irregularities or which specific sites were affected. It also did not explain how these issues might impact the overall integrity of the data collected so far.
According to BMS, data from the compromised sites have been excluded from analysis. After discussions with both the U.S. Food and Drug Administration and an external monitoring committee, BMS decided to enroll additional patients into ADEPT-2. The company stated it remains blinded to all study data.
There had been previous speculation among analysts regarding possible difficulties with ADEPT-2. Leerink Partners noted on October 30 that BMS executives were questioned during their third-quarter earnings call about whether enrollment had been reopened at certain sites. At that time, a spokesperson told Leerink Partners: “there has been no clinical trial update on ADEPT-2, and any reopening of enrollment would be a significant development.”
Leerink Partners had previously expected results from ADEPT-2 by late 2025. However, BMS said Wednesday that “additional trial results from the ADEPT program,” including studies ADEPT-1 and ADEPT-4 as well as ADEPT-2, are now projected for release by year-end 2026.
Investor concerns have grown due to shifting communications from BMS management about the status of this trial. In a note issued Wednesday morning, analysts at BMO Capital Markets referenced investor unease over potential failure of ADEPT-2 but added that compared to outright failure, news of a delay "is likely to be viewed a positive."
Cobenfy is a muscarinic antipsychotic acquired by BMS through its $14 billion purchase of Karuna Therapeutics in December 2023. Less than a year later, Cobenfy received FDA approval for schizophrenia treatment in September 2024. Since then, however, progress has slowed.
In April 2025, disappointing results emerged from another Phase III study called ARISE. This trial tested Cobenfy as an add-on therapy alongside atypical antipsychotics for schizophrenia symptoms but failed to reach statistical significance despite some observed improvements.
Following this setback, Leerink Partners wrote on April 23 that Cobenfy could have "far less potential than we originally anticipated." Although BMS indicated plans to discuss future steps with health regulators after ARISE’s outcome, no such talks had occurred as of late July.
