← Back

Lecanemab Phase 2b

Year of Publication: 2021

Journal: Alzheimer's Research & Therapy

Link: https://doi.org/10.1186/s13195-021-00813-8

PDF: https://alzres.biomedcentral.com/article...195-021-00813-8

Clinical Question

Does lecanemab slow cognitive decline and reduce brain amyloid in patients with early Alzheimer's disease (MCI and mild AD)?

Bottom Line

Study 201 (N=854, 18 months) tested 5 doses of lecanemab vs placebo in early AD. The 10 mg/kg biweekly ED90 dose missed the 12-month Bayesian primary endpoint (64% probability of >25% ADCOMS benefit vs 80% threshold), but at 18 months reduced brain amyloid by -0.306 SUVr units and showed 27-30% ADCOMS, 47-56% ADAS-Cog14, and 26-33% CDR-SB slowing vs placebo. Established proof-of-concept for 10 mg/kg biweekly lecanemab, informing the successful Clarity AD phase 3 trial that led to FDA approval (2023).

        Major Points
        Phase 2b multicenter randomized double-blind placebo-controlled Bayesian adaptive dose-ranging trial (BAN2401-G000-201, Swanson 2021)
854 patients with early AD (MCI due to AD or mild AD dementia), amyloid-PET or CSF Aβ-positive, CDR 0.5-1
6 arms: placebo + 5 lecanemab doses (2.5 mg/kg biweekly; 5 mg/kg monthly; 5 mg/kg biweekly; 10 mg/kg monthly; 10 mg/kg biweekly)
Response-adaptive randomization preferentially allocated to most-likely ED90 dose
12-month Bayesian primary endpoint: 64% probability of >25% ADCOMS benefit vs 80% threshold — missed
18-month prespecified secondary: brain amyloid reduced by -0.306 SUVr units at 10 mg/kg biweekly
ADCOMS slowed 27-30%; ADAS-Cog14 slowed 47-56%; CDR-SB slowed 26-33% at 18 months
CSF biomarkers: Aβ42 increased, p-tau/t-tau decreased; less hippocampal atrophy on MRI
ARIA-E 9.9% at 10 mg/kg biweekly; dose-related; higher in APOE ε4 homozygotes
Established 10 mg/kg biweekly as ED90 dose for phase 3 Clarity AD
Clarity AD phase 3 (N=1795) was positive in 2023 — led to FDA full approval of lecanemab (Leqembi) July 2023
First anti-amyloid antibody with unambiguous NEJM-published phase 3 CDR-SB slowing in early AD

      

Design

Study Type: Phase 2b multicenter randomized double-blind placebo-controlled Bayesian adaptive dose-ranging trial

Randomization: 1

Blinding: Double-blind

Follow-up Duration: 18 months

Sample Size: 854

Analyzed: 854

Analysis: Bayesian response-adaptive with 80% posterior probability threshold; frequentist secondary

Baseline Characteristics

Characteristic	Control	Active
N	245	609
Age mean	~71	~71
CDR-SB	2.9	2.9
Amyloid-PET positive	Required	Required
APOE ε4 carrier	~70%	~70%

Arms

Field	Control	Lecanemab 2.5 mg/kg biweekly	Lecanemab 5 mg/kg monthly	Lecanemab 5 mg/kg biweekly	Lecanemab 10 mg/kg monthly	Lecanemab 10 mg/kg biweekly (ED90)
N	245	52	51	92	253	161
Intervention	Placebo IV biweekly	Lecanemab IV every 2 weeks	Lecanemab IV monthly + placebo on alternate biweekly visits	Lecanemab IV every 2 weeks	Lecanemab IV monthly + placebo alternate	Lecanemab IV every 2 weeks
Duration	18 months	18 months	18 months	18 months	18 months	18 months

Outcomes

Outcome	Type	Control	Intervention	P-value
12-month Bayesian analysis of ADCOMS (Alzheimer's Disease Composite Score) change from baseline for the ED90 dose	Primary	Reference (natural decline)	10 mg/kg biweekly showed trend toward slowing	Primary missed; prespecified secondary analyses reached significance
Brain amyloid (18-month, PET SUVr)	Secondary	Stable/slight increase	-0.306 SUVr units (10 mg/kg biweekly)	Highly significant
ADCOMS (18 months, Bayesian/frequentist)	Secondary	Reference	Slowed by 27%/30% vs placebo	Significant by frequentist
ADAS-Cog14 (18 months)	Secondary	Reference	Slowed by 56%/47%	Significant in Bayesian/frequentist
CDR-SB (18 months)	Secondary	Reference	Slowed by 33%/26%	Significant in Bayesian/frequentist
CSF Aβ42	Secondary	No change	Increased	Supportive
CSF p-tau/t-tau	Secondary	No change	Decreased	Supportive
Hippocampal volume (MRI)	Secondary	More atrophy	Less atrophy	Supportive
ARIA-E (amyloid-related imaging abnormalities-edema) at 10 mg/kg biweekly	Adverse	0.8%	9.9%	Dose-related
ARIA-H (microhemorrhage/superficial siderosis)	Adverse	Low rate	Similar rate	No major excess
ARIA-E in APOE ε4 homozygotes	Adverse	Low	Markedly higher (later recognized)	Class effect
Infusion reactions	Adverse	Uncommon	More common with lecanemab	Expected for IV biologic
Headache	Adverse	Common	Common	Similar
Falls	Adverse	Similar	Similar	No difference
Any TEAE	Adverse	High rate	High rate	Comparable overall
Discontinuation for AE	Adverse	Low	Slightly higher with active	ARIA-related primarily
Death	Adverse	Few	Few; not drug-related in 2b	No excess mortality

Subgroup Analysis

Dose-response was clear across doses: higher exposure with biweekly dosing produced greater amyloid reduction and clinical slowing. APOE ε4 carriers had higher ARIA rates but similar cognitive benefit. The 10 mg/kg biweekly dose was identified as the ED90 and carried forward to Clarity AD (CLARITY-AD, N=1795, published NEJM 2023, primary endpoint CDR-SB positive — 27% slowing with HR 0.73, p<0.001). FDA granted accelerated approval Jan 2023, full approval July 2023.

Criticisms

12-month primary Bayesian endpoint missed — technically a failed phase 2; reliance on 18-month prespecified secondary analyses for efficacy conclusions
Adaptive design with small individual arms complicates interpretation and power calculations
ARIA-E rate 9.9% at 10 mg/kg biweekly requires intensive MRI monitoring during titration — substantial clinical burden; higher in APOE ε4 homozygotes (~30% later in phase 3)
Modest absolute slowing (0.45 point CDR-SB difference over 18 months) — clinical meaningfulness debated
Open-label extension data and APOE-stratified analyses matured after 2b — phase 3 Clarity AD provided definitive efficacy/safety
Cost (~$26,500/year) and infusion burden limit real-world access despite approval

Funding

Eisai / BioArctic (developer of lecanemab)

Based on: Lecanemab Phase 2b (Alzheimer's Research & Therapy, 2021)

Content summarized and formatted by NeuroTrials.ai.

Lecanemab Phase 2b

(2021)

Objective

Lecanemab (BAN2401) — humanized IgG1 monoclonal antibody preferentially targeting soluble Aβ aggregates (oligomers and protofibrils) — evaluated in a Bayesian adaptive design across 5 doses vs placebo in early Alzheimer's disease.

Study Summary

• Lecanemab 10 mg/kg biweekly reduced brain amyloid by -0.306 SUVr units at 18 months (placebo stable/increased) — robust target engagement.
• 12-month Bayesian primary endpoint missed: 64% probability of >25% ADCOMS benefit vs 80% threshold.
• 18-month prespecified analyses showed 27-30% ADCOMS, 47-56% ADAS-Cog14, and 26-33% CDR-SB slowing vs placebo.
• CSF biomarkers (Aβ42, p-tau, t-tau) improved with lecanemab; hippocampal volume less atrophy.
• ARIA-E incidence 9.9% at 10 mg/kg biweekly — manageable with monitoring; lower in APOE ε4 noncarriers.
• Proof-of-concept supported the phase 3 Clarity AD trial (published 2023, positive), leading to lecanemab FDA approval.

Intervention

5 lecanemab doses (2.5 mg/kg biweekly; 5 mg/kg monthly; 5 mg/kg biweekly; 10 mg/kg monthly; 10 mg/kg biweekly) vs placebo IV every 2 weeks, 18-month duration. Bayesian adaptive response-based randomization to preferentially allocate to the simplest dose achieving ≥90% of maximum effect (ED90).

Inclusion Criteria

Adults with early AD (mild cognitive impairment due to AD or mild AD dementia), MMSE ≥22, CDR 0.5-1, amyloid-PET or CSF Aβ-positive.

Study Design

Arms: Placebo vs 5 Lecanemab doses (see above)

Patients per Arm: Placebo 245; Lecanemab total 609 (adaptive allocation across 5 doses); 10 mg/kg biweekly largest active arm

Outcome

• 12-month primary Bayesian ADCOMS analysis: missed (64% probability of ≥25% benefit vs 80% threshold) — trial did not succeed per Bayesian criteria
• 18-month prespecified analyses (Bayesian/frequentist): brain amyloid -0.306 SUVr units; ADCOMS slowed 27%/30%; ADAS-Cog14 slowed 56%/47%; CDR-SB slowed 33%/26%
• CSF biomarker support: Aβ42 increased, p-tau/t-tau decreased
• ARIA-E at 10 mg/kg biweekly: 9.9% (higher in APOE ε4 carriers)
• Hippocampal volume: less atrophy with lecanemab (not a primary endpoint)

Bottom Line

Major Points

Phase 2b multicenter randomized double-blind placebo-controlled Bayesian adaptive dose-ranging trial (BAN2401-G000-201, Swanson 2021)
854 patients with early AD (MCI due to AD or mild AD dementia), amyloid-PET or CSF Aβ-positive, CDR 0.5-1
6 arms: placebo + 5 lecanemab doses (2.5 mg/kg biweekly; 5 mg/kg monthly; 5 mg/kg biweekly; 10 mg/kg monthly; 10 mg/kg biweekly)
Response-adaptive randomization preferentially allocated to most-likely ED90 dose
12-month Bayesian primary endpoint: 64% probability of >25% ADCOMS benefit vs 80% threshold — missed
18-month prespecified secondary: brain amyloid reduced by -0.306 SUVr units at 10 mg/kg biweekly
ADCOMS slowed 27-30%; ADAS-Cog14 slowed 47-56%; CDR-SB slowed 26-33% at 18 months
CSF biomarkers: Aβ42 increased, p-tau/t-tau decreased; less hippocampal atrophy on MRI
ARIA-E 9.9% at 10 mg/kg biweekly; dose-related; higher in APOE ε4 homozygotes
Established 10 mg/kg biweekly as ED90 dose for phase 3 Clarity AD
Clarity AD phase 3 (N=1795) was positive in 2023 — led to FDA full approval of lecanemab (Leqembi) July 2023
First anti-amyloid antibody with unambiguous NEJM-published phase 3 CDR-SB slowing in early AD

Study Design

Study Type: Phase 2b multicenter randomized double-blind placebo-controlled Bayesian adaptive dose-ranging trial
Randomization: Yes
Blinding: Double-blind
Sample Size: 854
Follow-up: 18 months

Primary Outcome

Definition: 12-month Bayesian analysis of ADCOMS (Alzheimer's Disease Composite Score) change from baseline for the ED90 dose

Control	Intervention	HR/OR	P-value
Reference (natural decline)	10 mg/kg biweekly showed trend toward slowing	- (Bayesian credibility interval; did not meet pre-specified success criterion)	Primary missed; prespecified secondary analyses reached significance

Limitations & Criticisms

12-month primary Bayesian endpoint missed — technically a failed phase 2; reliance on 18-month prespecified secondary analyses for efficacy conclusions
Adaptive design with small individual arms complicates interpretation and power calculations
ARIA-E rate 9.9% at 10 mg/kg biweekly requires intensive MRI monitoring during titration — substantial clinical burden; higher in APOE ε4 homozygotes (~30% later in phase 3)
Modest absolute slowing (0.45 point CDR-SB difference over 18 months) — clinical meaningfulness debated
Open-label extension data and APOE-stratified analyses matured after 2b — phase 3 Clarity AD provided definitive efficacy/safety
Cost (~$26,500/year) and infusion burden limit real-world access despite approval

Citation

View Original Publication →

Lecanemab Phase 2b

Clinical Question

Bottom Line

Major Points

Design

Baseline Characteristics

Arms

Outcomes

Subgroup Analysis

Criticisms

Funding

Ask about Lecanemab Phase 2b