Type: Clinician-administered cognitive outcome measure
Purpose: Quantify cognitive function in Alzheimer's disease clinical trials
Developer: Rosen, Mohs, and Davis (1984)[@rosen1984]
Current use: Primary endpoint in LY-3372689 MAGNOLIA trial, Phase 3 anti-amyloid trials (Aducanumab), and most AD therapeutic trials
Type: Clinician-administered cognitive outcome measure
Purpose: Quantify cognitive function in Alzheimer's disease clinical trials
Developer: Rosen, Mohs, and Davis (1984)[@rosen1984]
Current use: Primary endpoint in LY-3372689 MAGNOLIA trial, Phase 3 anti-amyloid trials (Aducanumab), and most AD therapeutic trials
Overview
Mermaid diagram (expand to render)
The Alzheimer's Disease Assessment Scale — Cognitive Subscale (ADAS-Cog) is the gold-standard cognitive outcome measure in Alzheimer's disease clinical trials. It was developed in 1984 by Rosen et al. as a comprehensive assessment tool specifically for AD trials["@rosen1984"]. The scale measures multiple domains of cognition including memory, language, praxis, and visuospatial function.
ADAS-Cog has become the most widely used primary endpoint in AD drug trials, with a large body of validation data across diverse populations. Its sensitivity to change over time and ability to distinguish between AD patients and healthy controls has made it the benchmark for regulatory approval studies["@mohs2000"].
Versions
ADAS-Cog 11 (Original)
The original 11-item version contains:
Word recall (3 trials, 10 words each): Immediate word recall score
Naming: Identifies objects and fingers
Commands: Follows written commands
Constructional praxis: Draws geometric figures
Ideational praxis: Reproduces sequences
Orientation: Date, day, time, place
Word recognition: 12-word delayed recognition
Language: Spontaneous speech, comprehension
Maximum score: 70 (higher = more impairment)
ADAS-Cog 13
Adds two items to improve sensitivity:
Delayed word recall: Separate delayed recall trial
Number cancellation: Attention and visual scanning
Maximum score: 85
ADAS-Cog 14 (Most commonly used in recent trials)
Further additions:
Concentration/distractibility item
Improves ceiling/floor properties
Maximum score: 90
Scoring
Interpretation
0–5: Normal cognitive function
5–12: Mild cognitive impairment range
12–30: Mild-to-moderate AD range
30–50+: Moderate-to-severe AD range
Minimum Clinically Important Difference (MCID)
1–3 points: Often cited as MCID in mild AD
4+ points: Generally considered clinically meaningful change
Trial designs typically power for 25–50% slowing of decline vs. placebo
Advantages
Widely validated across hundreds of AD trials
Strong regulatory acceptance (FDA, EMA)
Sensitive to change in mild-to-moderate AD (MMSE 10–26)
Detailed cognitive profile across multiple domains
Limitations
Ceiling effects in very mild AD (preclinical)
Floor effects in moderate-to-severe AD
Administration time: 30–45 minutes
Training required for reliable administration
Cultural/language adaptations needed for global trials
Domains Assessed
| Domain | Items | Max Points | Description | |--------|-------|------------|-------------| | Memory | Word recall, recognition | 25 | Immediate and delayed verbal memory | | Language | Naming, commands, comprehension, speech | 25 | Verbal abilities and comprehension | | Praxis | Constructional, ideational | 10 | Visuospatial and motor execution | | Orientation | Time, place | 8 | Temporal and spatial awareness | | Concentration | Number cancellation | 5 | Attention and processing speed |
Use in OGA Inhibitor Trials
MAGNOLIA Trial (LY-3372689 in Early AD)
The MAGNOLIA trial (NCT05063565) used ADAS-Cog 13 as the primary efficacy endpoint[@kaye2024]:
Population: Early symptomatic AD (amyloid-confirmed)
Primary endpoint: Change from baseline in ADAS-Cog 13 at 18 months
Expected change: Treatment group expected to show ~30–40% slowing of decline
Context: OGA inhibition aims to reduce tau phosphorylation, potentially slowing cognitive decline by protecting synaptic and neuronal function
Why ADAS-Cog for OGA inhibitor trials
OGA inhibitors target tau pathology, which correlates with cognitive decline progression
ADAS-Cog captures the cognitive domains most affected by tau pathology (memory, executive function)
Strong baseline-to-follow-up sensitivity in the mild AD population being studied
Regulatory precedent: FDA has accepted ADAS-Cog as primary endpoint for tau-targeting therapies
MMSE: Brief screening tool for baseline stratification
CDR-SB: Clinical staging and global change
CSF/blood biomarkers: Tau phosphorylation, A-beta, neurodegeneration
OGA Inhibitor Context
The OGA inhibitor therapeutic approach targets tau hyperphosphorylation by increasing [O-GlcNAcylation](/mechanisms/protein-o-glcna-cylation-pathway) on tau proteins. This is based on the [Yin-Yang Hypothesis](/mechanisms/yin-yang-hypothesis) — that O-GlcNAcylation and phosphorylation compete for the same serine/threonine sites on tau.
If OGA inhibition successfully reduces tau phosphorylation:
Less tau aggregates into [neurofibrillary tangles](/diseases/alzheimers-disease#neurofibrillary-tangles)
Preserved neuronal/synaptic function
Slower cognitive decline as measured by ADAS-Cog
The key question in MAGNOLIA is whether sufficient target engagement (CSF O-GlcNAc increase) translates into measurable cognitive benefit at 18 months — a question that will inform the entire OGA inhibitor field.