Verubecestat-Prodromal
Randomized Trial of Verubecestat for Prodromal Alzheimer's Disease
Clinical Question
Can verubecestat, a BACE-1 inhibitor that reduces amyloid-beta production, slow disease progression in patients with prodromal Alzheimer's disease (MCI with elevated brain amyloid)?
Bottom Line
Verubecestat did not improve clinical ratings of dementia among patients with prodromal Alzheimer's disease, and some measures suggested that cognition and daily function were worse among patients who received verubecestat than among those who received placebo, despite reducing brain amyloid levels
Major Points
- Phase 3 trial terminated early for futility after interim analysis
- Primary outcome not met: CDR-SB worsening was similar or greater with verubecestat vs placebo
- 40mg dose showed statistically worse outcome than placebo on CDR-SB (2.02 vs 1.58, p=0.01)
- Higher progression rate to dementia with verubecestat (HR 1.38 for 40mg vs placebo)
- Verubecestat effectively reduced brain amyloid on PET (decreased 0.04 SUVR vs increased 0.02 with placebo)
- Cerebrospinal fluid Aβ42 reduced >60% with verubecestat confirming target engagement
- Multiple secondary outcomes suggested worse cognition/function with verubecestat
- Cognitive worsening apparent by week 13 and did not progress further
- Adverse events more common with verubecestat including rash, sleep disturbance, weight loss, hair color change
Design
Study Type: Randomized, double-blind, placebo-controlled, parallel group phase 3 trial
Randomization: 1
Blinding: Double-blind; participants, investigators, and assessors blinded; audiorecorded assessments with central rater quality review
Enrollment Period: November 2013 to April 2018
Follow-up Duration: 104 weeks (planned); terminated early for futility
Centers: 238
Countries: United States, Canada, Europe (multiple), Australia, New Zealand, Japan
Sample Size: 1454
Analysis: Longitudinal ANCOVA with Bonferroni correction; hierarchical sequential testing; modified intention-to-treat
Inclusion Criteria
- Age 50-85 years
- Subjective decrease in memory for at least 1 year corroborated by informant
- RBANS Delayed Memory Index score ≤85 (≥1 SD below age/education-appropriate mean)
- Positive brain amyloid on visual inspection of amyloid-ligand PET
- MMSE score 24-30
- Did not meet criteria for dementia
- Diagnosis of prodromal Alzheimer's disease confirmed by independent review
- Stable dose of AChEI or memantine for ≥3 months if taking
Exclusion Criteria
- Meets criteria for dementia
- Other causes of cognitive impairment
Arms
| Field | Verubecestat 12mg | Verubecestat 40mg | Control |
|---|---|---|---|
| Intervention | Verubecestat 12mg orally once daily | Verubecestat 40mg orally once daily | Matching placebo orally once daily |
| Duration | 104 weeks (planned); terminated early | 104 weeks (planned); terminated early | 104 weeks (planned); terminated early |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Change in CDR-SB score from baseline to week 104 (range 0-18, higher scores indicate worse cognition and daily function) | Primary | 1.58 | 12mg vs placebo: 0.67; 40mg vs placebo: 0.01 (favoring placebo) | ||
| Progression to dementia due to Alzheimer's disease (events per 100 patient-years) | Secondary | 19.3 | 12mg: 24.5; 40mg: 25.5 | 12mg: HR 1.30 (97.51% CI 1.01-1.68); 40mg: HR 1.38 (97.51% CI 1.07-1.79) | |
| CCS-3D change at 104 weeks | Secondary | 0.77 | 12mg: 0.77; 40mg: 0.78 | ||
| Hippocampal volume change at 104 weeks (%) | Secondary | -6.1% | 12mg: -6.5%; 40mg: -6.7% | ||
| Cortical amyloid SUVR change at 104 weeks | Secondary | +0.02 | 12mg: -0.03; 40mg: -0.04 | ||
| ADCS-ADLMCI change at 104 weeks | Secondary | -4.1 | 12mg: -5.2; 40mg: -5.8 | ||
| Any adverse event | Adverse | 421 (87.0%) | 12mg: 441 (91.3%); 40mg: 446 (92.1%) | ||
| Serious adverse event | Adverse | 96 (19.8%) | 12mg: 124 (25.7%); 40mg: 101 (20.9%) | ||
| Death | Adverse | 3 (0.6%) | 12mg: 3 (0.6%); 40mg: 1 (0.2%) | ||
| Rash, dermatitis, urticaria | Adverse | 62 (12.8%) | 12mg: 96 (19.9%); 40mg: 101 (20.9%) | ||
| Depression | Adverse | 25 (5.2%) | 12mg: 32 (6.6%); 40mg: 50 (10.3%) | ||
| Anxiety | Adverse | 21 (4.3%) | 12mg: 33 (6.8%); 40mg: 44 (9.1%) | ||
| Sleep disturbance | Adverse | 22 (4.5%) | 12mg: 38 (7.9%); 40mg: 44 (9.1%) | ||
| Weight loss | Adverse | 10 (2.1%) | 12mg: 27 (5.6%); 40mg: 32 (6.6%) | ||
| Hair color change | Adverse | 0 (0%) | 12mg: 12 (2.5%); 40mg: 24 (5.0%) | ||
| Falls and injuries | Adverse | 100 (20.7%) | 12mg: 124 (25.7%); 40mg: 123 (25.4%) | ||
| Suicidal ideation | Adverse | 31 (6.4%) | 12mg: 33 (6.8%); 40mg: 45 (9.3%) |
Subgroup Analysis
Exploratory subgroup analyses did not suggest that the effect of verubecestat on CDR-SB was altered by baseline APOE4 gene-carrier status, age, sex, MMSE score, or PET SUVR. Cognitive worsening appeared by week 13 and did not progress further relative to placebo.
Criticisms
- Trial terminated early for futility limiting statistical power for some analyses
- Cognitive worsening may reflect BACE-1 inhibition effects on synaptic function beyond amyloid
- BACE-2 inhibition may contribute to adverse effects (uncertain physiologic functions)
- Disease-modifying effect requires longer follow-up than 104 weeks to fully assess
- Prodromal stage may already be too late in disease course for amyloid-lowering benefit
- Only 47-49% of patients completed the trial due to early termination
- Cerebrospinal fluid substudy limited to 5-6 patients per group
- Higher discontinuation rate in verubecestat groups due to adverse events
- Amyloid reduction achieved but did not translate to clinical benefit
Funding
Merck Sharp & Dohme
Based on: Verubecestat-Prodromal (New England Journal of Medicine, 2019)
Authors: Michael F. Egan, James Kost, Tiffini Voss, ..., David Michelson
Citation: N Engl J Med 2019;380:1408-20
Content summarized and formatted by NeuroTrials.ai.