- Design
- Systematic review and random-effects meta-analysis of 21 phase 3 randomised trials
- Population
- 12,610 people with early Alzheimer disease on anti-amyloid monoclonal antibodies
- Primary outcome
- Incidence, severity and risk factors of ARIA-E and ARIA-H
- Effect
- ARIA-E 5.4%, ARIA-H 10.27%; ApoE e4 homozygotes OR 5.12 (ARIA-E) and 4.65 (ARIA-H)
This systematic review and meta-analysis pooled 21 phase 3 randomised trials reported in 20 articles, with 12,610 people with early Alzheimer disease receiving anti-amyloid monoclonal antibodies.
Pooled incidence was 5.4% for ARIA-E (oedema or effusion) and 10.27% for ARIA-H (haemosiderin deposition). Most cases were asymptomatic and radiographically mild to moderate.
For ARIA-E, ApoE e4 homozygosity (OR 5.12), heterozygosity (OR 1.91), higher antibody dose (OR 2.0) and baseline microhaemorrhages (OR 1.43) were associated with higher risk. For ARIA-H, risk was higher in ApoE e4 heterozygotes (OR 1.65) and homozygotes (OR 4.65). Symptomatic and radiographically severe ARIA-E were also more likely with e4 genotypes.
The authors do not report separate estimates for each agent in the abstract, and pooling trials with different drugs and doses limits how far the headline figures apply to any one antibody.
- Check ApoE genotype before considering anti-amyloid therapy; e4 homozygotes carried the highest ARIA risk.
- Review baseline MRI for microhaemorrhages, which raised ARIA-E risk.
- Plan MRI surveillance as the product information requires.
- Discuss that most ARIA was asymptomatic but some was symptomatic or severe.
Why it matters
It gives a way to tell which patients face the highest imaging-safety risk before treatment starts.
Don't overread it
Pooled data across different antibodies cannot give the risk for a specific drug, and long-term consequences of ARIA remain unknown.
The statistics, in plain English
A pooled incidence mixes different drugs and doses, so it is an average rather than the risk for any single product. An odds ratio of 5.12 does not mean 5 times the percentage risk; it means 5 times the odds, which exaggerates the risk when events are common.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free