Real-world lecanemab: ARIA in about one in seven, one in three for APOE4 homozygotes

Design

Systematic review and meta-analysis of real-world observational studies

Population

37 studies of patients with Alzheimer disease treated with lecanemab

Primary outcome

ARIA, discontinuation, adverse events, cognitive scores and amyloid PET

Effect

Any ARIA 15.1%; APOE4 homozygotes 34.1% vs non-carriers 11.0%; symptomatic ARIA 1.3%

A systematic review and meta-analysis in the Journal of Neurology, published on 21 October, pooled 37 real-world studies of lecanemab for Alzheimer disease.

Any amyloid-related imaging abnormality (ARIA) occurred in 15.1%, oedema-type in 6.5% and haemosiderin-type in 11.5%; symptomatic ARIA was 1.3%. Risk tracked APOE4 dose: 34.1% in homozygotes, 14.5% in heterozygotes, 11.0% in non-carriers. Baseline microbleeds were linked to more haemosiderin-type ARIA (19.3% vs 4.2%). Rates were lower in Asian than in US cohorts. Infusion reactions affected 18.5%, and 12.2% stopped treatment.

Amyloid PET fell as expected, but cognitive scales showed no significant change apart from a small early MoCA difference. Without control groups, that neither confirms nor refutes benefit.

The practical message is patient selection: genotype and baseline MRI shape the risk more than disease stage does. Lecanemab's availability in India is limited and costly; the selection principles apply wherever it is used.

• Discuss APOE4 status before treatment: homozygotes had about three times the ARIA rate of non-carriers.
• Count baseline microbleeds on MRI; they were linked to more haemosiderin-type ARIA.
• Expect infusion reactions in nearly one in five patients.
• Most ARIA was asymptomatic; scheduled MRI monitoring is how it is found.
• Do not quote real-world cognitive results as proof of benefit; they had no comparison group.

Why it matters

It confirms that trial-era ARIA risk holds in routine care and sharpens who is at highest risk.

Don't overread it

The cognitive outcomes were uncontrolled, so this cannot show whether lecanemab works in practice.

The statistics, in plain English

The pooled ARIA rates are proportions across studies of different sizes and settings, so the regional differences may partly reflect how intensively MRI was done. Subgroup rates by genotype are more informative than the overall 15%.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.