Clinical trial endpoints in Alzheimer's disease, as of Q2 2026

Why endpoints matter

The first thing to know about an Alzheimer's trial is what it measured. The drug, the population, the duration - all matter. But the interpretation of the result rests on which endpoint the trial used and how that endpoint behaves.

This snapshot summarizes the endpoints that appear most often in late-stage Alzheimer's trials and how each is read.

Cognitive and functional composite endpoints

Clinical Dementia Rating - Sum of Boxes (CDR-SB). Composite measure across six domains - memory, orientation, judgment and problem-solving, community affairs, home and hobbies, and personal care. Range: 0 to 18, higher is worse. Sensitive to change in early symptomatic populations. Used as the primary endpoint in the lecanemab and donanemab pivotal trials. Discussed in our Signal on the CDR-SB clinical-meaningfulness debate.

Integrated Alzheimer's Disease Rating Scale (iADRS). Combines a cognitive sub-score (ADAS-Cog) with a functional sub-score (ADCS-iADL). Designed to be sensitive across the early symptomatic window. Used as a co-primary or secondary endpoint in several recent late-stage trials.

Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS-Cog). Long-standing cognitive battery. Multiple versions (ADAS-Cog 11, 13, 14). Less sensitive to change at the very earliest disease stages than CDR-SB or iADRS, but well-validated and broadly familiar.

Alzheimer's Disease Cooperative Study - Activities of Daily Living (ADCS-ADL) scale. Functional outcome measure capturing instrumental and basic activities of daily living. Used as a key secondary in most disease-modification trials.

Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA). Brief cognitive screens. Routine in clinical care, but not typically primary endpoints in registration-supporting trials because of ceiling and floor effects.

Biomarker endpoints

Amyloid PET (centiloid scale). Quantitative measure of brain amyloid burden. Used as both an inclusion criterion and an outcome - meaningful amyloid clearance is the central biomarker readout for amyloid-directed therapies and supports the donanemab finite-duration treatment paradigm.

Tau PET. Quantitative measure of tau pathology in defined brain regions. Increasingly used as both a stratification tool and an outcome measure in trials of tau-directed therapies.

Plasma p-tau217 and related plasma biomarkers. Used both for trial enrollment screening and as exploratory pharmacodynamic outcome measures. The role as a regulatory-supportive outcome continues to evolve.

Volumetric MRI - hippocampal and whole-brain. Used as supportive structural outcome measures. Interpretation interacts with the early volume changes seen in some anti-amyloid trial populations.

Patient-reported and quality-of-life outcomes

Patient-reported and caregiver-reported outcomes (e.g., QoL-AD, ZBI for caregiver burden) are increasingly incorporated into late-stage trial designs. They have not yet displaced clinician-rated composites as primary endpoints in registration-supporting trials, but they carry growing weight in payer and clinician interpretation of trial readouts.

How to read a readout

When a trial reports a result, three questions are load-bearing:

1. What was the primary endpoint, and what is the absolute size of the difference between treatment and control? A statistically significant result with a small absolute delta carries different operational and clinical weight than a result with a large delta.
2. Was the population early symptomatic, prodromal, or preclinical? Endpoint sensitivity and the meaning of any change are population-dependent.
3. How did the supporting endpoints - biomarker, functional, patient-reported - track with the primary? Concordance across endpoints strengthens the interpretation; divergence complicates it.

This snapshot is updated when material changes occur in commonly used endpoints or in the regulatory framing around them.