Donanemab

Donanemab (brand name Kisunla; INN donanemab; development code LY3002813) is a humanized IgG1 monoclonal antibody developed by Eli Lilly targeting N-terminal pyroglutamate-modified amyloid-beta (pGlu-Aβ). It received FDA approval in July 2024 for early symptomatic Alzheimer’s disease (early AD), making it the second FDA-approved anti-amyloid immunotherapy after lecanemab (January 2023) and the first to demonstrate a defined-duration treatment paradigm — patients who clear their amyloid burden to below a threshold can discontinue dosing. Within the AmyloSENS framework, donanemab represents the second-generation of an extracellular-amyloid-clearance strategy targeting a post-translationally modified, plaque-enriched Aβ epitope.

Identity

• WHO INN: donanemab
• Brand name: Kisunla (Eli Lilly and Company)
• Development code: LY3002813; donanemab-azbt (USAN)
• Type: humanized IgG1 monoclonal antibody; biologic
• PubChem CID: null — not assigned for protein biologics needs-canonical-id
• ChEMBL ID: null — not confirmed; verification pending needs-canonical-id
• DrugBank ID: null — verification pending needs-canonical-id
• Molecular weight: ~150,000 Da (IgG1 class; exact MW not retrieved)
• Epitope: N-terminal pyroglutamate (pGlu3)-modified Aβ, present preferentially in deposited neuritic plaques; does not bind monomeric non-modified Aβ

Mechanism of action

Donanemab binds pGlu3-Aβ, a truncated and cyclized form of Aβ generated by N-terminal pyroglutamation (glutaminyl cyclase-mediated conversion of the Glu3 residue). This modification is:

1. Enriched in the cores of amyloid plaques compared to soluble pools
2. More prone to aggregation than full-length Aβ (pGlu3-Aβ acts as a seed for plaque growth)
3. Largely absent from monomeric/oligomeric soluble Aβ in plasma and CSF

By targeting pGlu3-Aβ, donanemab binds preferentially to deposited plaque rather than circulating Aβ species. After antibody–antigen engagement on the plaque surface, the Fc region of the IgG1 isotype recruits microglial Fc-gamma receptors (FcγRI and FcγRIII), triggering receptor-mediated phagocytosis of the antibody-tagged plaque deposit. This is the dominant proposed mechanism for plaque removal; complement-mediated lysis is a secondary contributor.

Donanemab does not bind monomeric full-length Aβ(1-42) or Aβ(1-40) and is inactive against soluble Aβ oligomers that lack the pyroglutamate modification. This epitope selectivity is proposed to explain the comparatively favorable ARIA-E rate versus earlier pan-Aβ antibodies (e.g., aducanumab), as reduced engagement with vascular amyloid may limit perivascular microglial activation ¹.

no-mechanism — The precise contribution of pGlu-Aβ seeding to total plaque burden and the therapeutic magnitude of pGlu3-Aβ-specific vs full-Aβ-plaque clearance remain incompletely characterized.

Regulatory history and approval status

FDA approval (July 2024)

Donanemab received FDA traditional approval in July 2024 for adults with early symptomatic Alzheimer’s disease (mild cognitive impairment or mild AD dementia), confirmed by amyloid-positive PET or CSF biomarkers. Key approval conditions:

• APOE ε4 homozygote restriction: The approved prescribing information excludes APOE ε4/ε4 homozygotes due to substantially elevated ARIA risk in this genotype ¹. APOE genotyping is required before initiation.
• MRI surveillance: Baseline MRI required; surveillance MRI required per label at defined intervals during treatment.
• Amyloid confirmation: Positive amyloid status required before treatment.

Context within the anti-amyloid class

Drug	Approval	Developer	Epitope	Year
Aducanumab (Aduhelm)	FDA accelerated approval 2021	Biogen	Pan-Aβ (N-terminal 3–7)	Withdrawn January 2024
Lecanemab (Leqembi)	FDA traditional approval January 2023	Eisai/Biogen	Aβ protofibrils	Active
Donanemab (Kisunla)	FDA traditional approval July 2024	Eli Lilly	N-terminal pGlu3-Aβ (deposited plaque)	Active

EMA submission status: not confirmed at time of seeding. unsourced

Clinical evidence

TRAILBLAZER-ALZ 2 — Phase 3 primary trial

Design: Multicenter, randomized, double-blind, placebo-controlled Phase 3 trial in early symptomatic AD; 76-week (18-month) treatment period ².

Population: n = 1,736 randomized (donanemab n=860; placebo n=876); amyloid-positive (PET-confirmed); tau stratified into low/medium and high tau subgroups.

Primary endpoints:

Population	Endpoint	Donanemab	Placebo	Difference	p-value
Combined (low/med + high tau)	iADRS at 76 wk	-10.19	-13.11	+2.92 (95% CI 1.51–4.33)	<0.001
Combined	CDR-SB at 76 wk	1.72	2.42	-0.70 (95% CI -0.95 to -0.45)	<0.001
Low/medium tau subgroup	iADRS at 76 wk	-6.02	-9.27	+3.25 (95% CI 1.88–4.62)	<0.001
Low/medium tau subgroup	CDR-SB at 76 wk	1.20	1.88	-0.67 (95% CI -0.95 to -0.40)	<0.001

The primary success was in the combined population for iADRS; prespecified success in low/medium tau subgroup for CDR-SB. The high tau population showed smaller differences and did not reach the prespecified significance threshold for the co-primary endpoint.

Key interpretation: The ~29% slowing of CDR-SB progression in the combined population and ~36% slowing in the low/medium tau subgroup are statistically robust. The clinical meaningfulness of a 0.67–0.70-point CDR-SB difference at 76 weeks is debated; the scale range is 0–18 and the MCID (minimum clinically important difference) for CDR-SB is estimated at ~1 point in most analyses. contradictory-evidence — see Cochrane 2026 review.

Amyloid clearance (secondary endpoint): 80.1% (95% CI 76.1%–83.6%) of low/medium tau participants and 76.4% (95% CI 72.9%–79.6%) of combined-population participants who had a PET scan achieved amyloid clearance (<24.1 Centiloids) at 76 weeks in the donanemab group, vs 0.2% in the placebo group ².

Limited-duration treatment paradigm: Participants whose amyloid PET met pre-specified clearance criteria were switched to blinded placebo during the trial, operationalizing a finite-course treatment model unique to donanemab among currently approved anti-amyloid antibodies. Lecanemab is dosed continuously by comparison.

TRAILBLAZER-ALZ 2 Long-Term Extension (LTE)

Duration: 78-week double-blind extension following the 76-week main trial ³.

Key findings (vs weighted ADNI external controls):

• Early-start arm: -1.2 points CDR-SB difference from ADNI controls at 3-year assessment
• Delayed-start arm (initiated donanemab after placebo phase): -0.8 points CDR-SB difference at 76 weeks post-initiation
• Early vs delayed-start comparison: Early-start participants had significantly lower risk of disease progression on CDR-Global over 3 years (HR = 0.73; p<0.001)
• Amyloid reaccumulation: Among those who cleared amyloid, modeling predicted a median reaccumulation rate of ~2.4 Centiloids/year; the clinical significance of reaccumulation is ongoing

Interpretation limitation: ADNI is an observational cohort selected for research participation, not a true randomized comparator. External-control comparisons carry unmeasured confounding. needs-replication — randomized extension data without external-control reliance would strengthen confidence.

Posttreatment amyloid and outcomes correlation

A secondary analysis of TRAILBLAZER-ALZ 2 (n=1,582) found strong correlation between posttreatment amyloid levels and clinical outcomes ⁴:

• Lower posttreatment amyloid correlated with slower iADRS progression (R² = 0.73)
• CDR-SB correlation R² = 0.87
• p-tau217 correlation R² = 0.86; p-tau181 R² = 0.88; GFAP R² = 0.87
• No correlation with NfL (R² = 0.03)

This supports amyloid clearance as mechanistically upstream of clinical benefit — but correlation data from a non-randomized secondary analysis cannot establish causation.

TRAILBLAZER-ALZ 4 — head-to-head vs aducanumab

TRAILBLAZER-ALZ 4 was a Phase 3, open-label, randomized trial (n=148) directly comparing donanemab vs aducanumab ⁵:

Timepoint	Donanemab amyloid clearance rate	Aducanumab amyloid clearance rate
6 months	37.9%	1.6%
12 months	70.0%	24.6%
18 months	76.8%	43.1%

Median time to clearance: 359 days (donanemab) vs 568 days (aducanumab); p<0.001. ARIA-E rate: 23.9% (donanemab) vs 34.8% (aducanumab). This was not a head-to-head vs lecanemab — no such trial has been completed as of 2026-05-09. needs-replication — lecanemab vs donanemab head-to-head trial data not yet published.

Cochrane 2026 meta-analysis

A 2026 Cochrane systematic review of 7 anti-amyloid antibodies (17 trials, n=20,342) including one donanemab trial (TRAILBLAZER-ALZ 2) concluded that anti-amyloid antibodies produce “trivial” effects on cognitive function and “little to no difference” in dementia severity and functional ability at 18 months ⁶. The review found: “successful removal of amyloid from the brain does not seem to be associated with clinically meaningful effects in people with mild cognitive impairment or mild dementia due to Alzheimer’s disease.” ARIA-E absolute risk difference was +107 per 1,000 treated.

contradictory-evidence — The Cochrane 2026 conclusion of “little to no meaningful difference” in clinical outcomes stands in direct tension with the statistically significant endpoints in TRAILBLAZER-ALZ 2 (Sims 2023) and the FDA’s traditional approval decision. The discordance reflects a genuine dispute over whether the measured effect sizes (CDR-SB ~0.67–0.70 difference at 76 weeks) are clinically meaningful to patients and caregivers, not whether the trials were methodologically valid. Both framings should be disclosed to readers.

Safety — ARIA

ARIA (Amyloid-Related Imaging Abnormalities) is the primary safety concern for donanemab. ARIA-E rate of 24.0% (205/853) and ARIA-H rate of 31.4% (268/853) are from the TRAILBLAZER-ALZ 2 safety set (Sims 2023 ²). The 24.4% ARIA-E figure in the row below is from the combined TRAILBLAZER-ALZ + TRAILBLAZER-ALZ 2 secondary analysis ¹; the two figures differ because the combined dataset includes TRAILBLAZER-ALZ (Phase 2) participants. Data from TRAILBLAZER-ALZ and TRAILBLAZER-ALZ 2 combined placebo-controlled trials ¹:

ARIA type	Donanemab rate	Placebo rate
ARIA-E (edema/effusion)	24.4% (Zimmer 2025 combined trials)	1.9%
Symptomatic ARIA-E	6.1% (of donanemab arm; 52/853)	—
Serious ARIA-E	1.5% (13/853)	—
ARIA-H (microhemorrhage/hemosiderosis)	31.4% (268/853)	13.6% (119/874)

APOE ε4 genotype effect: APOE ε4 allele number is an independent predictor of ARIA risk. APOE ε4/ε4 homozygotes are excluded from the approved indication due to substantially elevated ARIA-E incidence (specific rates not extracted at abstract-verification level; full PDF verification recommended) ¹.

Symptomatic ARIA management: The most common ARIA-E symptoms are headache and confusional state. ARIA events typically resolve with dose modification (pause or discontinuation) ± corticosteroids.

Deaths: No deaths directly attributed to ARIA were found in the available abstract data from TRAILBLAZER-ALZ 2 ¹. The package insert and full FDA review document carry the full mortality accounting.

long-term-unknown — ARIA rates beyond the 76-week trial period are not well characterized. Reaccumulation may necessitate retreatment, which could compound ARIA risk.

Dosing and administration

Per the FDA-approved indication (early symptomatic AD, amyloid-positive):

• Induction: 700 mg IV infusion every 4 weeks × 3 doses (first 3 months)
• Maintenance: 1,400 mg IV infusion every 4 weeks thereafter
• Duration: Treatment continues until PET-confirmed amyloid clearance (<24.1 Centiloids) or until clinical judgment indicates discontinuation
• Prerequisite: APOE genotyping required; APOE ε4/ε4 homozygotes excluded
• MRI surveillance: Baseline + pre-dose MRI per label schedule; treatment held or discontinued based on ARIA severity grade

The limited-duration paradigm distinguishes donanemab from lecanemab’s continuous dosing: patients who achieve amyloid clearance can discontinue and enter a monitoring phase.

Comparison with lecanemab

Feature	Donanemab (Kisunla)	Lecanemab (Leqembi)
Epitope	N-terminal pGlu3-Aβ (plaque-enriched)	Aβ protofibrils (aggregated)
Selectivity	Deposited plaque > soluble Aβ	Aggregated/oligomeric > monomeric
Dosing	700 mg × 3 → 1400 mg q4w IV; discontinue on clearance	10 mg/kg q2w IV; continuous
ARIA-E rate (Phase 3)	24.4%	~12–13%
Phase 3 CDR-SB slowing	~29% (combined population)	~27% (CLARITY-AD)
Head-to-head vs. each other	Not yet completed	Not yet completed
APOE ε4/ε4 exclusion	Yes (FDA label)	Not excluded; additional monitoring

Key practical difference: Donanemab’s q4w dosing is less burdensome than lecanemab’s q2w. The defined-duration paradigm may improve patient experience. However, donanemab’s ARIA-E rate (~24%) is approximately 2× that of lecanemab (~12–13%) — a clinically relevant safety difference that complicates direct comparison given different patient populations. No prospective head-to-head randomized trial comparing the two has been reported as of 2026-05-09.

Aging context — AmyloSENS

Within the AmyloSENS framework, extracellular amyloid-β accumulation in the aging brain represents a category of damage driven by the progressive failure of proteostasis maintenance (inefficient Aβ clearance via the glymphatic system, blood-brain barrier efflux, and microglial phagocytosis). Donanemab represents the second-generation AmyloSENS therapeutic — it physically clears the existing aggregate burden rather than reducing production (production-reduction approaches via β-secretase/BACE inhibition failed in Phase 3 trials with cognitive harm signals, and were abandoned).

Within the loss-of-proteostasis hallmark framework, Aβ plaque accumulation maps to extracellular proteostasis failure, distinct from intracellular misfolded protein clearance (the chaperone and UPS/autophagy-mediated mechanisms).

Current AmyloSENS therapeutic landscape (2026):

Agent	Target	Approval	Notes
Lecanemab (Leqembi)	Aβ protofibrils	FDA 2023	Continuous dosing; lower ARIA-E than donanemab
Donanemab (Kisunla)	pGlu3-Aβ (plaque)	FDA July 2024	Defined-duration; higher ARIA-E
Aducanumab (Aduhelm)	Pan-Aβ	Withdrawn Jan 2024	Controversial accelerated approval; Biogen discontinued
Remternetug	pGlu3-Aβ (same epitope as donanemab)	Phase 3	Subcutaneous formulation in development

See also: lecanemab, aducanumab (both implicit stubs as of 2026-05-09).

Limitations and gaps

1. Modest effect size: The 29–36% slowing of CDR-SB progression translates to an absolute difference of ~0.67–0.70 points at 76 weeks on a 0–18 scale (paper-reported slowing: 36.0% for low/medium tau CDR-SB; 28.9% for combined CDR-SB ²). The Cochrane 2026 review rates the class effect as “trivial” on cognition and “little to no difference” on dementia severity ⁶. The clinical significance is disputed.

2. High ARIA-E burden: At 24.4% ARIA-E (vs 12–13% for lecanemab), donanemab’s safety profile requires aggressive MRI surveillance and may limit use in patients with vascular comorbidities, anticoagulant use, or those who cannot tolerate periodic brain MRI. long-term-unknown

3. APOE ε4 homozygote exclusion: The most genetically high-risk AD population (APOE ε4/ε4, ~2% of the general population but enriched among early-onset AD cases) is excluded from the approved indication. Translation to this subgroup is blocked. needs-human-replication

4. Only Alzheimer’s disease: No evidence in other amyloidopathies (ATTR, IAPP, α-synuclein). Translation to aging-as-disease (prevention in cognitively normal amyloid-positive individuals) not yet established.

5. Amyloid-reaccumulation after discontinuation: Modeling predicts ~2.4 Centiloids/year reaccumulation after amyloid clearance ³. The clinical implications — and whether retreatment restores benefit — are unknown.

6. No lecanemab head-to-head RCT: All comparative claims between donanemab and lecanemab come from cross-trial comparisons with different populations, not a direct randomized trial. TRAILBLAZER-ALZ 4 compared donanemab to aducanumab (withdrawn), not lecanemab.

7. Preclinical/prevention gap: TRAILBLAZER-ALZ 3 (NCT05738486) is evaluating donanemab in preclinical AD (amyloid-positive, cognitively unimpaired) — baseline data published 2025 ⁷; primary results pending.

Related pages

• lecanemab — companion anti-amyloid antibody; different epitope, lower ARIA-E, continuous dosing (implicit stub)
• aducanumab — withdrawn 2024; historical AmyloSENS context (implicit stub)
• amyloid-beta — target protein (implicit stub)
• loss-of-proteostasis — hallmark targeted
• sens-damage-categories — AmyloSENS category
• alzheimers-disease — clinical indication (implicit stub)

Footnotes

Footnotes

1. doi:10.1001/jamaneurol.2025.0065 · Zimmer JA, Ardayfio P, Wang H, Khanna R, Evans CD, Lu M, Sparks J, Andersen S, Lauzon S, Nery ESM, Battioui C, Engle SE, Biffi A, Svaldi D, Salloway S, Greenberg SM, Sperling RA, Mintun M, Brooks DA, Sims JR · JAMA Neurology 2025 May;82(5):461–469 · secondary analysis · n=combined TRAILBLAZER-ALZ + ALZ 2 · model: early symptomatic AD, APOE-stratified · abstract-only verification 2026-05-09 — PDF download failed; quantitative ARIA claims sourced from PubMed abstract only; verify before relying on exact figures. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

2. doi:10.1001/jama.2023.13239 · Sims JR, Zimmer JA, Evans CD, Lu M, Ardayfio P, Sparks J, Wessels AM, Shcherbinin S, Wang H, Monkul Nery ES, Collins EC, Solomon P, Salloway S, Apostolova LG, Hansson O, Ritchie C, Brooks DA, Mintun M, Skovronsky DM (TRAILBLAZER-ALZ 2 Investigators) · JAMA 2023 Aug 8;330(6):512–527 · n=1,736 · rct · p<0.001 (iADRS + CDR-SB combined population) · model: early symptomatic AD adults, amyloid-positive by PET; 76-week blinded treatment · PDF available locally in a local paper archive; quantitative claims should be verified against full text. ↩ ↩² ↩³ ↩⁴

3. doi:10.1016/j.tjpad.2025.100446 · Zimmer JA, Sims JR, Evans CD, et al. · J Prev Alzheimers Dis 2026;13(2):100446 · open-label extension · n=TRAILBLAZER-ALZ 2 eligible participants; ADNI external controls · 78-week extension following 76-week main trial · abstract-only verification 2026-05-09 — PDF download pending. ↩ ↩²

4. doi:10.1001/jamaneurol.2025.3869 · Lu M, Kim MJ, Collins EC, Shcherbinin S, Ellinwood AK, Yokoi Y, Brooks DA, Hansson O, Knopman DS, Sims JR, Mintun MA · JAMA Neurology 2025 Dec;82(12):1251–1256 · secondary analysis of TRAILBLAZER-ALZ 2 · n=1,582 (766 donanemab, 816 placebo) · model: early symptomatic AD adults, 76-week follow-up · abstract-only verification 2026-05-09 — PDF download pending. ↩

5. doi:10.1002/alz.70293 · Salloway S, Pain A, Lee E, Papka M, Ferguson MB, Wang H, Hu H, Lu M, Oru E, Ardayfio PA, Hoban DB, Collins EC, Brooks DA, Sims JR · Alzheimers Dement 2025 May;21(5):e70293 · Phase 3 open-label rct · n=148 · model: early symptomatic AD, donanemab vs aducanumab, amyloid PET primary · PDF available locally in a local paper archive. ↩

6. doi:10.1002/14651858.CD016297 · Nonino F, Minozzi S, Sambali L, et al. · Cochrane Database Syst Rev 2026;4(4):CD016297 · meta-analysis · n=20,342 (17 trials, 7 anti-amyloid antibodies including donanemab TRAILBLAZER-ALZ 2) · systematic review · abstract-only verification 2026-05-09 — not in local archive; citations sourced from PubMed abstract (PMID 41985900). ↩ ↩²

7. doi:10.1002/alz.70662 · Yaari R, Holdridge KC, Williamson M, et al. · Alzheimers Dement 2025;21(9):e70662 · baseline-data publication (screening cohort, not primary results) · model: preclinical AD (amyloid-positive, cognitively unimpaired); TRAILBLAZER-ALZ 3 · abstract-only verification 2026-05-09. ↩