Cognition Therapeutics Aligns on Key Aspects of Pivotal Study for Zervimesine (CT1812) in DLB Psychosis

FDA Aligns with Cognition on Pivotal DLB Psychosis Study Design

Cognition Therapeutics, Inc. announced that it has received written feedback from the U.S. Food and Drug Administration (FDA), confirming alignment on key aspects of a pivotal Phase 3 trial for zervimesine (CT1812) in DLB psychosis. The FDA agreed that psychosis associated with dementia with Lewy bodies (DLB) could be an approvable outcome, supporting a New Drug Application (NDA). The registrational program is expected to begin in mid-2027, enrolling DLB patients with psychosis symptoms. Participants will be randomized to receive 100 mg of once-daily oral zervimesine or placebo for nine months, with the neuropsychiatric inventory (NPI) as the primary endpoint. This follows positive Phase 2 SHIMMER trial results where zervimesine slowed progression of hallucinations and delusions by 89%.

• Cognition Therapeutics has reached a significant agreement with the FDA, establishing that psychosis associated with dementia with Lewy bodies (DLB) is an approvable outcome. This is a critical development for a condition with no FDA-approved treatments, where current off-label antipsychotics carry substantial risks, highlighting a major unmet medical need for patients and caregivers.
• The planned pivotal Phase 3 study for zervimesine is anticipated to commence in mid-2027. It will enroll individuals with DLB experiencing psychosis symptoms, including hallucinations and delusions, and will allow for patients on stable background off-label antipsychotic medications. The trial design involves a nine-month treatment period with 100 mg of once-daily oral zervimesine or placebo, utilizing the neuropsychiatric inventory (NPI) as a novel primary endpoint.
• The design of the upcoming Phase 3 study is informed by encouraging data from Cognition's Phase 2 COG1201 ‘SHIMMER’ trial in mild-to-moderate DLB. In this earlier study, zervimesine demonstrated a notable impact on psychosis symptoms, specifically slowing the progression of hallucinations and delusions by 89% compared to placebo, as measured by the NPI. Further analyses from the Phase 2 study are scheduled for presentation at the Alzheimer’s Association’s International Conference (AAIC).

FDA Aligns on Pivotal Zervimesine Trial Design for DLB Psychosis

The clinical trial landscape for DLB psychosis encompasses a range of study designs — from randomized controlled trials to retrospective cohort analyses — evaluating antipsychotics, cholinesterase inhibitors (ChEIs), and other interventions. Endpoints have spanned both psychiatric efficacy measures and motor tolerability assessments, reflecting the dual challenge of managing psychosis without exacerbating parkinsonism.

• Quetiapine RCT (2007): A multicenter, randomized, double-blind, placebo-controlled parallel-group trial enrolled 40 patients (DLB: n=23; Parkinson's disease with dementia: n=9; Alzheimer's disease with parkinsonian features: n=8) over 10 weeks. The primary efficacy endpoint was change in the Brief Psychiatric Rating Scale (BPRS) from baseline; the primary tolerability endpoint was change in the UPDRS motor section. The trial was confounded by a protocol design change and incomplete recruitment, yielding no significant differences on primary or secondary efficacy measures. An unexpectedly large placebo effect, a mean quetiapine dose of only 120 mg/day, and insufficient statistical power were cited as contributing factors. Quetiapine did not worsen parkinsonism but was associated with a trend toward decline in daily functioning.

• Quetiapine vs. Pimavanserin retrospective cohort (2020): This study compared 47 quetiapine-treated and 45 pimavanserin-treated patients with psychosis in PD or DLB. The primary endpoint was time to treatment discontinuation, analyzed via Kaplan-Meier survival analysis. Pimavanserin demonstrated lower early discontinuation rates but higher late discontinuation rates relative to quetiapine (HR <1 before day 43; HR >1 after day 43; P=0.04). No between-group difference in mortality was observed (HR 0.37; 95% CI 0.06–2.45; P=0.88). Notably, 33% of the pimavanserin cohort carried a DLB diagnosis versus 11% in the quetiapine cohort (P=0.01).

• ChEI meta-analysis (search cutoff April 2022): A systematic search across PubMed/MEDLINE, Embase, and PsychInfo identified 34 eligible placebo-controlled RCTs evaluating donepezil, rivastigmine, or galantamine in AD, PD, or DLB. Individual participant data were obtained from 17 trials (n=6,649 participants; mean age 75.0 ± 8.2 years), with a 2-stage random-effects meta-analysis performed. Primary endpoints were hallucinations and delusions; secondary endpoints covered all other neuropsychiatric subdomains and total neuropsychiatric score. ChEI treatment was associated with reductions in both delusions and hallucinations in AD and PD subgroups; individual participant data for DLB were not available.

• ECT retrospective study in DLB/MCI-LB: Twelve patients with DLB and 13 with MCI with Lewy bodies who underwent ECT for psychiatric symptoms were included. Diagnoses were confirmed via dopamine transporter SPECT and I-MIBG myocardial scintigraphy. Psychiatric symptom severity was assessed using the Clinical Global Impressions Severity Scale (CGI-S) before and approximately one week after the final ECT session; improvement was rated with the CGI-I. Cognitive function, dementia severity, and adverse events were also evaluated. ECT significantly improved psychiatric symptoms (CGI-I: 60% very much improved; 20% much improved; 16% minimally improved; 4% no change) and parkinsonism (Hoehn and Yahr: 1.76 ± 1.2 vs. 1.04 ± 0.7; P<0.001), alongside improvement in dementia severity (CDR; P=0.037). Adverse events included delirium in 24% and amnesia in 4% of patients.

• DLB therapeutic trial synthesis (ClinicalTrials.gov and PubMed): A structured review identified 35 trials on ClinicalTrials.gov and 14 on PubMed, with 21 trials ultimately analyzed. Of 11 completed trials with reported results, two agents demonstrated positive outcomes: zonisamide (phases 2 and 3) showed improvements in parkinsonian syndrome, and neflamapimod (phase 2) showed improvements in cognition and ambulation.

Addressing Critical Unmet Needs in DLB Psychosis Treatment

Treatment of psychosis in dementia with Lewy bodies (DLB) is complicated by a convergence of pharmacological vulnerabilities, diagnostic complexity, and a near-absence of approved therapeutic options. The core paradox is that the agents most commonly deployed to manage psychosis — antipsychotics — carry substantial risk of severe, potentially fatal adverse reactions in this population specifically because of the underlying pathophysiology of DLB.

• Neuroleptic sensitivity as a defining and life-threatening barrier: Approximately 50% of DLB patients exhibit neuroleptic sensitivity, a risk significant enough to be included among the clinical diagnostic criteria for the disease. This sensitivity — driven by DLB patients' heightened vulnerability to antidopaminergic and anticholinergic mechanisms — manifests as extrapyramidal symptoms (EPS), accelerated cognitive and physical decline, and increased mortality. Critically, this risk extends to both typical and atypical antipsychotics, severely narrowing the therapeutic window.

• Agent-specific limitations across the antipsychotic class: Individual atypical antipsychotics each present distinct risk profiles. Olanzapine (10 mg) produced motor worsening despite psychosis control, and three of eight patients could not tolerate it even at the lowest available doses (2.5–7.5 mg); in 2004, olanzapine and risperidone were found to confer a threefold elevated stroke risk in dementia patients. Risperidone carries a high risk of neuroleptic malignant syndrome (NMS). Clozapine remains controversial due to its potent anticholinergic action and agranulocytosis risk. Notably, a case of quetiapine-induced NMS has been documented in DLB, underscoring that even agents with relatively favorable EPS profiles are not without serious risk. Ziprasidone has been associated with substantial motor worsening, and long-acting injectable paliperidone palmitate, while achieving psychosis remission in one case, induced iatrogenic akinetic-rigid syndrome.

• Dopaminergic therapy: limited efficacy and psychosis exacerbation risk: Dopaminergic treatment for the motor features of DLB requires a difficult risk-benefit calculation. Motor benefit — defined as greater than 10% improvement over baseline UPDRS Part III score — occurred in only one-third of subjects. Among those who did achieve motor benefit, one-third subsequently developed worsened hallucinations or psychosis. Consequently, only approximately 22% of DLB subjects achieved motor benefit without psychosis exacerbation, reflecting the narrow therapeutic margin of dopaminergic agents in this disease context.

• Cholinesterase inhibitor limitations in practice: Although cholinesterase inhibitors represent a relatively safer option — with cholinergic deficits linked to visual hallucinations and evidence that these agents can improve apathy, anxiety, attention, hallucinations, delusions, and sleep disturbance — real-world tolerability remains a concern. In a seven-patient donepezil study (5–10 mg once daily), only three patients completed the full 8-week therapy; three were discontinued prematurely due to insufficient response and/or adverse events. Worsening of extrapyramidal motor features is rare but has been reported.

• Broader drug-class hypersensitivity beyond antipsychotics: DLB-associated hypersensitivity is not confined to neuroleptics. Gabapentinoids have been reported to induce psychosis in DLB patients even at low doses, suggesting a generalised susceptibility to CNS-active agents that further constrains prescribing options. Modafinil administration has been documented to exacerbate agitation and hallucinations in DLB, with psychotic symptoms remitting upon discontinuation.

• Diagnostic under-recognition and absence of approved therapies: DLB is frequently misdiagnosed as Alzheimer's disease or Parkinson's disease-related dementia, delaying appropriate management and increasing exposure to contraindicated agents. No medications are currently approved in the USA specifically for psychosis in dementia. Among emerging agents, pimavanserin shows promise in Lewy body dementia, but the nelotanserin programme has been suspended, and only NYX-458 remains under active investigation for cognitive impairment. The evidence base supporting clozapine, quetiapine, and pimavanserin for psychosis control remains limited, and further studies in patients with severe, intractable neuropsychiatric symptoms are explicitly needed.

Zervimesine's Pivotal Step Towards Addressing DLB Psychosis

The recent FDA alignment on the pivotal Phase 3 trial design for zervimesine (CT1812) marks a critical juncture for patients grappling with psychosis in dementia with Lewy bodies (DLB). This neurodegenerative condition, encompassing DLB and Parkinson's disease dementia, presents a tremendous burden, particularly due to debilitating non-motor symptoms like psychosis. The FDA's confirmation that psychosis itself can serve as an approvable outcome for a New Drug Application provides invaluable regulatory clarity, significantly de-risking the development pathway for Cognition Therapeutics.

This strategic move positions zervimesine to potentially fill a substantial unmet medical need. While the drug was previously noted in early research as a potential disease-modifying therapy, the current focus on symptomatic treatment for psychosis highlights a pragmatic approach to bring a much-needed therapy to patients sooner. The positive Phase 2 SHIMMER trial results, indicating an 89% slowing of hallucinations and delusions, offer a strong foundation for the upcoming registrational program. However, the journey ahead is not without its challenges. The larger Phase 3 trial, set to begin in mid-2027 and utilizing the neuropsychiatric inventory (NPI) as its primary endpoint, must robustly demonstrate sustained efficacy and a favorable safety profile over nine months.

The therapeutic landscape for LBD psychosis is evolving. While some investigational therapies like nelotanserin have faced setbacks, others, such as pimavanserin, show promise. Zervimesine will need to clearly differentiate itself in this competitive environment. A successful outcome in Phase 3 would not only establish zervimesine as a vital symptomatic treatment but could also pave the way for further exploration of its broader potential, including its earlier-identified disease-modifying properties. This development underscores the ongoing strides in addressing the complex challenges of Lewy body dementia, offering a beacon of hope for patients and their caregivers.