Anti-amyloid mAbs (lecanemab / donanemab / aducanumab)
Cognitive · Amyloid-beta-targeting monoclonal antibody
Tier D-
High-risk evidence profile
Bottom line
Read Off Label grades Anti-amyloid mAbs (lecanemab / donanemab / aducanumab) as D- (2.6/10) based on weak-moderate evidence, low-moderate benefit magnitude, and a high-risk safety profile.
Its high-risk evidence profile reflects a Safety score of 0/10; reported benefits and harms can differ by indication, dose, route, duration, and population.
Cochrane 2026 (Nonino et al, n=20342 across 17 RCTs): the entire class produces only trivial cognitive (SMD -0.
Studied, labeled, or reported-use context: Lecanemab 10 mg/kg IV q2 weeks; donanemab 700 mg IV q4 weeks for 3 doses then 1400 mg IV q4 weeks — Rx; not individualized guidance.
Frontier potential — editorial judgment, not evidence
This is a separate axis from the grade above. It asks a different question: if this worked in humans, how much would it matter? It is our judgment, it is not part of the composite score, and it cannot raise a grade. A high number here is a statement about the idea, not about whether it works or whether anyone should take it.
What the ceiling rests on
The first therapies to change the underlying pathology of Alzheimer's disease rather than symptoms — amyloid clearance is unambiguous on imaging, and lecanemab and donanemab both showed statistically significant slowing of decline in large Phase 3 trials, which is a genuine scientific milestone after decades of failure.
Why it may not get there
The clinical effect sizes are small and their real-world meaningfulness is contested — the differences sit near or below commonly cited thresholds for a noticeable change. ARIA (amyloid-related imaging abnormalities) causes brain oedema and haemorrhage, with deaths reported, and APOE4 homozygotes are at markedly higher risk. Cost, infusion burden, and MRI monitoring are substantial. This is proof the pathway is modifiable more than proof it is the right lever.
What would change this Longer-term data on whether the divergence widens or plateaus.
What the evidence says
Cochrane 2026 (Nonino et al, n=20342 across 17 RCTs): the entire class produces only trivial cognitive (SMD -0.11) and dementia-severity (SMD -0.12) benefit while consistently increasing ARIA-E. Authors conclude 'successful removal of amyloid does not seem to be associated with clinically meaningful effects' and 'future research should focus on other mechanisms of action.' Costly (~$26K/year for lecanemab); requires APOE genotyping (4/4 homozygotes excluded or carefully monitored due to ARIA risk); MRI surveillance every dose. 2026 regulatory refinements: FDA approved a modified donanemab titration (TRAILBLAZER-ALZ 6) that significantly lowers ARIA-E incidence and severity (most benefit in APOE4 carriers), and accepted a lecanemab subcutaneous maintenance autoinjector (LEQEMBI IQLIK) for weekly home dosing - convenience plus a somewhat improved titration-safety story, though the trivial-clinical-benefit conclusion is unchanged. Indirect comparisons show lecanemab carries lower ARIA risk than donanemab.
Mechanism
Monoclonal antibodies bind aggregated amyloid-beta protein in brain (lecanemab targets soluble protofibrils; donanemab targets pyroglutamate amyloid plaques; aducanumab targets fibrillar Aβ) — drive amyloid clearance via Fc-mediated microglial phagocytosis. First disease-modifying drugs for Alzheimer's based on the amyloid hypothesis
Studied, labeled, or reported dose & route
Lecanemab 10 mg/kg IV q2 weeks; donanemab 700 mg IV q4 weeks for 3 doses then 1400 mg IV q4 weeks
This records doses and routes described in studies, approved labeling, clinical practice, or documented use. It is not individualized guidance and does not establish safety; context changes with indication, formulation, health history, monitoring, and other medicines.
Citations
A link means the source informed this entry; it does not mean every claim on the page comes from that source. If a citation looks wrong or you want a claim re-checked, send a correction.
Common questions
- What does the evidence show about Anti-amyloid mAbs (lecanemab / donanemab / aducanumab)'s effects?
- Read Off Label rates the evidence for Anti-amyloid mAbs (lecanemab / donanemab / aducanumab) as Weak-Moderate and the benefit magnitude as low-moderate, producing an overall grade of D- (2.6/10). Cochrane 2026 (Nonino et al, n=20342 across 17 RCTs): the entire class produces only trivial cognitive (SMD -0.
- What safety findings are reported for Anti-amyloid mAbs (lecanemab / donanemab / aducanumab)?
- Anti-amyloid mAbs (lecanemab / donanemab / aducanumab) has a high risk profile in the database. High (ARIA-E [edema] in ~25% of patients on lecanemab; ~107 more per 1000 vs placebo across class; ARIA-H [hemorrhage]; rare fatal cerebral hemorrhages especially with anticoagulants or APOE4/4 homozygotes; infusion reactions) Legal status: Rx (lecanemab/Leqembi FDA-approved Jan 2023; donanemab/Kisunla FDA-approved July 2024; aducanumab/Aduhelm WITHDRAWN by Biogen Jan 2024).
- What does the low Safety score for Anti-amyloid mAbs (lecanemab / donanemab / aducanumab) represent?
- The Safety score is 0/10. It records the substantial harms described for one or more use contexts; the Benefit score, indication, dose, route, duration, population, and legal status remain separate parts of the evidence profile.
- What doses or routes are reported for Anti-amyloid mAbs (lecanemab / donanemab / aducanumab)?
- Lecanemab 10 mg/kg IV q2 weeks; donanemab 700 mg IV q4 weeks for 3 doses then 1400 mg IV q4 weeks This is context from studies, approved labeling, clinical practice, or documented use; it is not individualized guidance and does not establish safety.
- How does Anti-amyloid mAbs (lecanemab / donanemab / aducanumab) work?
- Monoclonal antibodies bind aggregated amyloid-beta protein in brain (lecanemab targets soluble protofibrils; donanemab targets pyroglutamate amyloid plaques; aducanumab targets fibrillar Aβ) — drive amyloid clearance via Fc-mediated microglial phagocytosis. First disease-modifying drugs for Alzheimer's based on the amyloid hypothesis
This is an independent synthesis of published research by a non-clinician. Scores are opinions supported by citations, not prescriptions. See the full disclaimer and methodology for how this score was produced and what it does and doesn't mean.