JAMA August 08, 2023

Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial

John R. Sims, Jennifer A. Zimmer, Cynthia D. Evans, Ming Lu, Paul Ardayfio, JonDavid Sparks, Alette M. Wessels et al.

Bottom Line

In patients with early symptomatic Alzheimer disease, treatment with the anti-amyloid monoclonal antibody donanemab significantly slowed clinical and functional decline over 76 weeks but was associated with elevated rates of amyloid-related imaging abnormalities (ARIA).

Key Findings

1. In the primary analysis low/medium tau population (n=1,182), the least-squares mean change in iADRS score at 76 weeks was -6.02 for donanemab versus -9.27 for placebo (difference of 3.25 [95% CI, 1.88-4.62]; P < .001), representing a 35% slowing of clinical decline.
2. In the combined low/medium and high tau population (n=1,736), the iADRS change at 76 weeks was -10.2 for donanemab versus -13.1 for placebo (difference of 2.92 [95% CI, 1.51-4.33]; P < .001).
3. Donanemab slowed disease progression on the Clinical Dementia Rating-Sum of Boxes (CDR-SB) by 36% in the low/medium tau population (difference of -0.67; P < .001).
4. Nearly half (47%) of participants in the donanemab group had no clinical progression on the CDR-SB at 1 year, compared with 29% of those on placebo.
5. Amyloid-related imaging abnormalities with edema or effusion (ARIA-E) occurred in 24.0% of donanemab-treated patients compared to 1.9% of placebo-treated patients, with symptomatic ARIA-E occurring in 6.1% of the donanemab group.
6. Severe ARIA events included three treatment-related deaths in the donanemab arm.

Study Design

Design
Randomized Clinical Trial
Double-Blind
Sample
1,736
Patients
Duration
76 wk
Median
Setting
Multicenter, global
Population Participants aged 60 to 85 years with early symptomatic Alzheimer disease (mild cognitive impairment or mild dementia) and confirmed amyloid and tau pathology via PET imaging.
Intervention Donanemab administered intravenously every 4 weeks (700 mg for first 3 doses, then 1400 mg) for up to 72 weeks. Patients were switched to placebo if prespecified amyloid plaque clearance criteria were met.
Comparator Matching placebo administered intravenously every 4 weeks for 72 weeks.
Outcome Change in the integrated Alzheimer Disease Rating Scale (iADRS) score from baseline to 76 weeks in the low/medium tau population.

Study Limitations

The trial cohort lacked racial and ethnic diversity (91.5% White), limiting generalizability across different populations.
The limited follow-up duration of 76 weeks prevents conclusions regarding the long-term safety and maintenance of clinical benefit.
The protocol's limited-duration dosing approach (switching to placebo upon plaque clearance) introduced variability in the total amount and duration of donanemab exposure among participants.

Clinical Significance

TRAILBLAZER-ALZ 2 demonstrated that robust clearance of brain amyloid plaques can meaningfully modify the trajectory of early Alzheimer's disease. The trial's innovative dosing regimen, allowing patients to stop therapy once amyloid was cleared, introduced a paradigm shift in AD treatment. However, the high incidence of ARIA necessitates rigorous MRI monitoring, particularly for APOE ε4 carriers.

Historical Context

Following decades of failed trials targeting the amyloid cascade, the Alzheimer's field recently saw momentum shift with the approval of lecanemab (Clarity AD) and the accelerated approval of aducanumab. TRAILBLAZER-ALZ 2 solidified the viability of anti-amyloid monoclonal antibodies as disease-modifying therapies, validating that deep and rapid amyloid clearance yields significant downstream clinical and functional benefit in properly selected early-stage patients.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Donanemab specifically targets an N-terminally truncated, pyroglutamate-modified form of amyloid beta (N3pG) found almost exclusively in established brain plaques. Based on this mechanism of rapid plaque clearance, why is donanemab associated with Amyloid-Related Imaging Abnormalities (ARIA), and what are the two primary distinct subtypes of ARIA seen on MRI?

Key Response

This question connects basic pharmacology to clinical pathophysiology. Students must understand that clearing amyloid from cerebral vasculature compromises blood-brain barrier integrity, leading to ARIA-E (edema/effusion) and ARIA-H (microhemorrhages or superficial siderosis), which are the defining safety considerations of anti-amyloid monoclonal antibodies.

Resident
Resident

A 72-year-old patient with mild cognitive impairment is being evaluated for donanemab initiation. Before starting therapy, why is it critical to obtain both a baseline MRI and APOE genotyping, and how would an APOE e4/e4 homozygous result specifically alter your management and counseling?

Key Response

Residents must grasp the practical workup for anti-amyloid therapy. Baseline MRI is required to rule out excessive pre-existing microhemorrhages, while APOE genotyping is crucial because e4 homozygotes face a significantly higher risk of severe and symptomatic ARIA, requiring more rigorous shared decision-making and potentially altering the risk-benefit calculus for the patient.

Fellow
Fellow

TRAILBLAZER-ALZ 2 utilized a unique finite dosing model, halting donanemab treatment once patients achieved pre-defined amyloid clearance on PET, and also stratified patients by tau PET burden. How does the differential clinical response between the low/medium tau and high tau cohorts inform our understanding of the optimal therapeutic window for disease-modifying Alzheimer therapies?

Key Response

Fellows need to grapple with advanced trial design and disease staging. The trial demonstrated that clearing amyloid is significantly more clinically effective in patients with low-to-medium tau burdens, reinforcing the hypothesis that anti-amyloid interventions must occur early, before the cascade of widespread, autonomous tau-mediated neurodegeneration takes over.

Attending
Attending

Donanemab statistically slowed functional decline by roughly 22 to 35 percent on standard cognitive scales over 76 weeks, which translates to delaying clinical progression by approximately 4 to 7 months. In the context of shared decision-making, how do you frame the concept of 'slowing decline' versus 'improvement' to hopeful families, balancing this modest temporal delay against the heavy logistical burden of MRI monitoring and the risk of fatal ARIA?

Key Response

Attendings must expertly contextualize trial statistics for clinical practice. The critical teaching point is managing patient expectations—emphasizing that the drug does not restore memory or halt the disease entirely, but slightly flattens the curve. Weighing a few months of delayed progression against frequent infusions, anxiety-provoking MRI scans, and severe safety risks is the hallmark of wise, patient-centered care.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

Given that roughly 24 percent of the donanemab group experienced ARIA-E compared to only 2 percent in the placebo group, functional unblinding of both patients and clinical investigators is a profound methodological threat. How might this differential rate of a highly monitored adverse event bias subjective primary outcomes like the CDR-SB, and what specific blinding protocols should be mandated in future trials to mitigate this effect?

Key Response

Experts in research methodology must recognize how distinct, monitorable side effects can compromise double-blind designs. Because ARIA necessitates protocol-driven MRI discussions and safety interventions, many participants effectively become unblinded. This can artificially inflate the perceived treatment effect on subjective scales, highlighting the need for completely independent, blinded centralized raters who are strictly firewalled from safety data.

Journal Editor
Journal Editor

The screening cascade for the TRAILBLAZER-ALZ 2 trial was exceptionally rigorous, requiring precise cognitive scores, amyloid positivity, and specific tau PET thresholds, which resulted in a massive screen failure rate and a predominantly white study population with limited vascular comorbidities. As an editor critically appraising the external validity of this manuscript, how does this highly enriched 'unicorn' cohort limit the generalizability of the safety and efficacy data to real-world, diverse, multimorbid clinical populations?

Key Response

A seasoned editor would flag the homogeneity and strict selection criteria of the study population as a major threat to external validity. The rationale highlights the tension between a perfectly controlled experimental environment and the messy reality of clinical practice, questioning whether the observed risk-benefit ratio will hold true when the drug is deployed in broader, more diverse populations with mixed dementia pathologies.

Guideline Committee
Guideline Committee

Current guidelines for anti-amyloid therapies like lecanemab generally support continuous, chronic administration. Given the TRAILBLAZER-ALZ 2 data showing sustained clinical benefit after a 'treat-to-target' cessation of donanemab once amyloid plaque is cleared, should clinical practice guidelines formally transition to recommending finite dosing for this class of drugs, and what specific widely-available biomarker thresholds should guidelines mandate to define 'clearance' in community practice?

Key Response

Guideline committees must operationalize trial data into actionable, standard-of-care recommendations. This question challenges the committee to evaluate if the paradigm should shift from chronic therapy to induction-and-maintenance or finite dosing. It also forces consideration of healthcare economics, practical implementation, and how to define biomarker clearance (e.g., Centiloid levels on PET) for community practitioners who lack access to research-grade imaging analysis.

Clinical Landscape

Noteworthy Related Trials

2021

TRAILBLAZER-ALZ Trial

n = 257 · NEJM

Tested

Donanemab 700 to 1400 mg IV every 4 weeks

Population

Early symptomatic Alzheimer's disease patients with tau and amyloid pathology

Comparator

Placebo

Endpoint

Change in iADRS score at 76 weeks

Key result: Donanemab resulted in a significantly better composite score for cognition and daily functioning compared to placebo.
2022

EMERGE Trial

n = 1,638 · JPAD

Tested

Aducanumab up to 10 mg/kg monthly

Population

Patients with early Alzheimer's disease and confirmed amyloid pathology

Comparator

Placebo

Endpoint

Change in CDR-SB score at 78 weeks

Key result: High-dose aducanumab significantly reduced clinical decline on the CDR-SB, leading to the controversial accelerated FDA approval of the drug.
2023

CLARITY AD Trial

n = 1,795 · NEJM

Tested

Lecanemab 10 mg/kg biweekly

Population

Patients with early Alzheimer's disease

Comparator

Placebo

Endpoint

Change in CDR-SB score at 18 months

Key result: Lecanemab significantly reduced clinical decline on the CDR-SB scale by 27 percent compared to placebo.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis