Safety profiles of lecanemab, an anti-amyloid-β antibody for the treatment of early Alzheimer’s disease (AD), remain uncertain and may vary between randomized controlled trials (RCTs) and real-world evidence (RWE) studies.

This systematic review and meta-analysis aimed to evaluate the safety, tolerability, and acceptability of lecanemab based on findings from both RCTs and emerging RWE studies.

We systematically searched major databases and clinical trial registries from their inception to June 2025. Random-effects meta-analyses were performed to estimate the pooled incidence of key safety outcomes, including amyloid-related imaging abnormalities (ARIA), infusion-related reactions (IRRs), and treatment discontinuation (due to ARIA, adverse events [AEs], or any cause). The risk of ARIA according to the ApoE4 genotype was assessed via relative risk (RR). This study was registered with PROSPERO (No. CRD420251110679).

A total of two RCTs and five RWE studies encompassing 1576 patients were included. The pooled ARIA incidence was 19% (95% CI: 16%-23%), which was significantly modulated by ApoE4 status (RR 1.45 for heterozygotes, 3.54 for homozygotes vs noncarriers) and the pooled symptomatic ARIA incidence was 3% (95% CI: 2%-4%). IRRs occurred in 26% (95% CI: 19%–34%), with heterogeneity reduced in patients receiving specific pre-infusion prophylaxis. The pooled rate of discontinuation due to AEs was 8% (95% CI: 5%-11%), with discontinuation due to ARIA occurring in 5% (95% CI: 3%-7%) of patients in RWE studies.

Lecanemab-related ARIA demonstrates a clear ApoE4 gene-dose effect, supporting routine ApoE4 genotyping before treatment. Standardizing pre-infusion prophylaxis may reduce variability in IRRs incidence, while prompt recognition and management of ARIA are critical for improving treatment tolerability. These findings provide important evidence to support the safe clinical use of lecanemab.