ARIA Monitoring
Amyloid-related imaging abnormalities (ARIA) are the most clinically significant adverse effect of anti-amyloid monoclonal antibody therapies for Alzheimer disease. First described in 2010, ARIA encompasses two radiologic phenomena: ARIA-E (edema/sulcal effusion) and ARIA-H (microhemorrhages/superficial siderosis). ARIA events are most common in the early months of therapy, occur more frequently in APOEε4 carriers (particularly homozygotes), and are usually detected on routine surveillance MRI before clinical symptoms emerge. A structured approach to baseline risk assessment, serial MRI monitoring, severity grading, and evidence-based management is essential for the safe delivery of lecanemab and donanemab.
Bottom Line
- Two types: ARIA-E (vasogenic edema/sulcal effusion on FLAIR) and ARIA-H (microhemorrhages and superficial siderosis on T2*/SWI) — they can co-occur
- Incidence: ARIA-E occurs in ~12.6% with lecanemab and ~24.4% with donanemab; most cases (70–75%) are asymptomatic
- APOEε4 risk: Homozygotes have the highest rates — ARIA-E up to 34.5% (lecanemab) and 41.7% (donanemab); genotyping is strongly recommended before treatment
- Timing: 71% of ARIA-E detected within the first 3 months; 92% within the first 6 months
- MRI monitoring: Mandatory at baseline, ~week 7–8, ~week 12–14, ~week 26–28, and at 1 year
- Management: Mild asymptomatic ARIA may allow continued treatment; moderate ARIA-E requires suspension; severe ARIA-H mandates permanent discontinuation
- Prognosis: 81% of ARIA-E resolves within 4 months; ARIA-H hemosiderin is permanent on imaging but usually clinically silent
Pathophysiology and Types
ARIA is believed to result from the mobilization of amyloid-beta (Aβ) from parenchyma and perivascular spaces during antibody-mediated plaque clearance, causing transient vessel wall inflammation and blood-brain barrier disruption.
ARIA-E (Edema/Effusion)
- Anti-amyloid antibodies bind to and disrupt amyloid plaques, releasing soluble Aβ fragments into the perivascular drainage pathways
- Transient increase in perivascular amyloid burden triggers vessel wall inflammation and disruption of the blood-brain barrier
- Fluid leaks into brain parenchyma (vasogenic edema) or subarachnoid space (sulcal effusion)
- Appears as hyperintense signal on T2-weighted and FLAIR sequences, typically cortical or subcortical
- Usually reversible — 81% of cases resolve within 4 months of treatment suspension
- ARIA-E may occur concurrently with ARIA-H; combined events require more aggressive management
ARIA-H (Hemorrhage/Siderosis)
- Amyloid-weakened small vessel walls rupture during the clearance process, closely linked to underlying cerebral amyloid angiopathy (CAA)
- Hemosiderin deposits appear as hypointense foci on T2* gradient-recalled echo (GRE) or susceptibility-weighted imaging (SWI)
- Microhemorrhages: ≤10 mm at greatest diameter — the most common form of ARIA-H
- Macrohemorrhages: >10 mm at greatest diameter — rare but carry the highest morbidity and mortality risk
- Superficial siderosis: Linear hemosiderin deposits along the cortical surface, representing prior cortical hemorrhage
- Unlike ARIA-E, hemosiderin deposits are permanent on imaging even after clinical resolution
- Lobar distribution (cortical/subcortical) of microhemorrhages favors CAA/ARIA etiology vs. deep gray matter (basal ganglia, thalamus) distribution in hypertensive microbleeds
APOEε4 and ARIA Risk
- APOEε4 is the strongest genetic risk factor for both AD and cerebral amyloid angiopathy, making vessel walls more vulnerable during plaque clearance
- APOEε4 homozygotes carry 2.19× higher risk of ARIA-E and 3.45× higher risk of ARIA-H vs. noncarriers
- ε3/ε3: ARIA-E ~5–15%; ARIA-H ~12–19% — lowest risk; standard monitoring
- ε3/ε4: ARIA-E ~11–21%; ARIA-H ~14–32% — moderate risk; enhanced counseling
- ε4/ε4: ARIA-E ~33–42%; ARIA-H ~40–50% — highest risk; detailed risk-benefit discussion required
- APOE genotyping is strongly recommended before anti-amyloid therapy; donanemab labeling specifically recommends it
Incidence by Drug and Genotype
ARIA incidence varies substantially between the two FDA-approved anti-amyloid antibodies. Donanemab generally demonstrates higher ARIA rates than lecanemab across all APOE genotypes, likely reflecting differences in antibody target (pyroglutamated Aβ vs. protofibrils), dosing intensity, and magnitude of amyloid clearance. The table below summarizes key ARIA rates from the pivotal Phase 3 trials.
| Parameter | Lecanemab (CLARITY AD) | Donanemab (TRAILBLAZER-ALZ 2) |
|---|---|---|
| Overall ARIA | 21.3% vs. 9.3% placebo | 36.8% vs. placebo |
| ARIA-E | 12.6% vs. 1.7% placebo | 24.4% vs. placebo |
| Symptomatic ARIA-E | 2.8% | 5.8% |
| ARIA-H | 17.3% vs. 9.0% placebo | 31.3% vs. placebo |
| ARIA-E ε3/ε3 | 5% | 15% |
| ARIA-E ε3/ε4 | 11% | 21% |
| ARIA-E ε4/ε4 | 33% | 36% |
| ARIA-related deaths | Rare (post-marketing) | 3 in Phase 3 trial |
Additional Risk Factors
- Baseline cerebral microhemorrhages: Pre-existing microbleeds on SWI/GRE correlate with higher ARIA-H risk during treatment
- Higher amyloid plaque burden: Greater amyloid to clear creates more perivascular congestion and inflammation
- Anticoagulant use: May increase severity of hemorrhagic ARIA; associated with fatal events in post-marketing data
- Dose escalation: ARIA can emerge early in treatment or after dose increases
- Early treatment course: The first 3–6 months carry the greatest risk — most ARIA is detected on the first 2 surveillance MRIs
Anticoagulant and Antiplatelet Considerations
- Anticoagulants (warfarin, DOACs): NOT recommended during anti-amyloid therapy — risk of worsening ARIA-H; in the CLARITY AD trial, anticoagulant users were excluded
- Antiplatelet agents: May be considered with caution — aspirin ≤325 mg/day, clopidogrel, prasugrel, and ticagrelor at standard doses were permitted in trials
- Reassess antiplatelet therapy if ARIA-H develops; consult cardiology for patients requiring anticoagulation for atrial fibrillation, mechanical valves, or recent VTE
- Concurrent anticoagulant use was associated with fatal ARIA-related events in post-marketing reports
MRI Sequences and Protocol
| Sequence | Detects | Appearance | Notes |
|---|---|---|---|
| FLAIR | ARIA-E | Hyperintense parenchymal/sulcal signal | Most sensitive for vasogenic edema; compare side-by-side with prior |
| T2* GRE | ARIA-H microhemorrhages | Hypointense foci | Standard for microbleeds; less sensitive than SWI |
| SWI | ARIA-H (microhemorrhage + siderosis) | Hypointense foci/linear cortical signal | Higher sensitivity; preferred when available |
| DWI | Acute ischemia (differential) | Restricted diffusion | Distinguishes ARIA-E from acute infarction |
Practical MRI Protocol
- Minimum sequences: Axial FLAIR and axial T2* GRE or SWI
- Field strength: 1.5T or 3T (3T preferred for microhemorrhage sensitivity)
- Consistency: Use the same scanner, coil, and parameters for all serial exams at a single facility
- Slice thickness: ≤5 mm for FLAIR; ≤3 mm for SWI/GRE
- Radiologist communication: Inform that patient is on anti-amyloid therapy; provide ARIA monitoring protocol
MRI Monitoring Schedule
| Timepoint | Lecanemab | Donanemab | Purpose |
|---|---|---|---|
| Baseline | Within 12 months of start | Within 12 months of start | Eligibility screening; baseline comparison |
| ~Week 7–8 | Before 5th infusion | Before 3rd infusion | First surveillance — peak ARIA-E risk |
| ~Week 12–14 | Before 7th infusion | Before 4th infusion | Second surveillance |
| ~Week 26–28 | Before 14th infusion | ~6 months | Most ARIA-E detected by this point |
| ~Week 52 | Before 26th infusion | ~12 months | Annual; especially APOEε4 carriers |
| Symptom-driven | Any new neurologic symptom | Headache, confusion, seizure, gait change | |
| Post-ARIA | Every 2–4 months until resolution | Track ARIA-E resolution / ARIA-H stability | |
Severity Grading and Management
ARIA severity is graded separately for ARIA-E (based on size and extent of FLAIR abnormalities) and ARIA-H (based on number of new microhemorrhages and siderosis foci). Severity grading directly determines whether treatment should be continued, suspended, or permanently discontinued.
ARIA-E Severity
| Grade | FLAIR Findings | Symptoms | Management |
|---|---|---|---|
| Mild | <5 cm; single location | Usually asymptomatic | May continue with close MRI monitoring |
| Moderate | 5–10 cm or multifocal | Headache, confusion, dizziness | Suspend treatment; MRI in 2–4 months |
| Severe | >10 cm, confluent; mass effect | Seizure, focal deficits | Discontinue; hospitalize; MRI in 1–2 months |
ARIA-H Severity
| Grade | Imaging Findings | Management |
|---|---|---|
| Mild | ≤4 new microhemorrhages; no siderosis | May continue with monitoring |
| Moderate | 5–9 new microhemorrhages; 1–2 foci siderosis | Suspend; serial MRI |
| Severe | ≥10 new microhemorrhages; >2 siderosis foci; any macrohemorrhage | Permanently discontinue |
Asymptomatic ARIA Management
- Mild ARIA-E/H (asymptomatic): Treatment may continue at clinician discretion with repeat MRI in 2–4 months
- Moderate ARIA-E (asymptomatic): Suspend treatment; reinitiate only after complete or near-complete radiographic resolution
- Moderate ARIA-H (asymptomatic): Suspend; repeat MRI in 2–4 months to confirm stability before considering reinitiation
- Most asymptomatic cases resolve without intervention beyond treatment suspension
Symptomatic ARIA — Urgent Management
- Common symptoms: Headache, confusion, nausea/vomiting, visual disturbances, gait instability, seizure
- Immediate: Suspend anti-amyloid therapy; obtain urgent brain MRI (FLAIR + SWI/GRE + DWI); rule out stroke, infection
- Treatment: Analgesics, antiemetics, antiepileptic drugs as needed; consider corticosteroids for severe ARIA-E (evidence limited)
- Hospitalize for seizure, altered consciousness, focal deficits, or mass effect
- Reinitiation: Only after complete ARIA-E resolution (typically 4–16 weeks); permanently discontinue for any macrohemorrhage or recurrent symptomatic ARIA
Baseline MRI Eligibility Criteria
Patients must undergo a screening brain MRI within 12 months of treatment initiation to assess eligibility. The MRI must include FLAIR and SWI or T2* GRE sequences. The following findings on baseline MRI generally exclude patients from anti-amyloid therapy:
| Exclusionary Finding | Threshold | Rationale |
|---|---|---|
| Cerebral microhemorrhages | >4 (≤10 mm each) | Significant CAA; high ARIA-H risk |
| Macrohemorrhage | ≥1 (>10 mm) | Severe vascular fragility |
| Superficial siderosis | Any | CAA hallmark |
| Vasogenic edema | Any at baseline | Possible active CAA-related inflammation |
| Lacunar infarcts | ≥2 or major territory stroke | Significant cerebrovascular burden |
| White matter hyperintensities | Fazekas ≥3 | Extensive small vessel disease |
| Aβ angiitis / CAA-ri | Any evidence | Active inflammatory vascular process |
Treatment Reinitiation After ARIA
- ARIA-E fully resolved on FLAIR: May reinitiate at the same dose with close monitoring; schedule additional MRI 2–4 weeks after resumption
- ARIA-E partially resolved: Extend observation an additional 2–4 months before reinitiation; weigh clinical benefit vs. risk
- ARIA-H stable (no new microhemorrhages): May reinitiate with caution if total microhemorrhage count remains ≤4
- Recurrent ARIA-E on reinitiation: Strongly consider permanent discontinuation
- Any macrohemorrhage (>10 mm): Do not reinitiate treatment
- Severe ARIA of either type: Permanent discontinuation is recommended
Rare Serious and Fatal Cases
Although most ARIA events are self-limited, rare serious and fatal outcomes have been reported:
- Three deaths were attributed to ARIA-related events in the donanemab TRAILBLAZER-ALZ 2 Phase 3 trial
- Post-marketing fatalities have occurred with lecanemab, often in patients on concurrent anticoagulants or with extensive underlying CAA
- Fatal cases have involved massive cerebral edema with herniation, large intracerebral hemorrhage, or status epilepticus
- Key risk factors for fatal ARIA include: APOEε4 homozygosity, concurrent anticoagulation, extensive pre-existing CAA, and multiple prior ARIA events
- The overall ARIA-attributable mortality in clinical trials is estimated at <0.5%, but this risk must be clearly communicated during the informed consent process
Red Flags Requiring Urgent Evaluation
- New-onset seizure or witnessed convulsive activity
- Acute confusion or altered consciousness not explained by other causes
- Focal neurologic deficits (weakness, visual field loss, aphasia)
- Severe, persistent headache unresponsive to analgesics
- Gait instability or unexplained falls
- Any of these in a patient on anti-amyloid therapy → immediate treatment hold and urgent brain MRI
Differential Diagnosis
Not all new imaging abnormalities during anti-amyloid therapy represent ARIA. Key differential considerations include:
- Acute ischemic stroke: Restricted diffusion on DWI distinguishes from ARIA-E vasogenic edema
- CAA-related inflammation (CAA-ri): May mimic ARIA-E with extensive white matter FLAIR abnormalities; can be pre-existing
- Posterior reversible encephalopathy syndrome (PRES): Similar FLAIR pattern with posterior predominance; typically associated with hypertension
- CNS infection: Clinical context and CSF analysis distinguish meningoencephalitis from ARIA
- Hypertensive microhemorrhages: Deep gray matter location (basal ganglia, thalamus) favors hypertensive etiology vs. lobar location for CAA/ARIA
Clinical Case
A 72-year-old man with mild Alzheimer disease, APOEε4 heterozygous, began anti-amyloid monoclonal antibody therapy. After the 4th infusion, he reported headache, nausea, and vomiting partially improved by over-the-counter analgesics and not interfering with daily functioning. FLAIR MRI revealed parenchymal edema measuring 6 cm at the widest point in the right occipital lobe. This was classified as symptomatic, moderate ARIA-E (6 cm lesion with clinical symptoms). Treatment was suspended, and follow-up MRI at 12 weeks showed complete resolution. Treatment was reinitiated with enhanced MRI monitoring.
References
- Risacher SL. Neuroimaging in dementia. Continuum (Minneap Minn). 2024;30(6):1761-1789.
- Geldmacher DS. Treatment of Alzheimer disease. Continuum (Minneap Minn). 2024;30(6):1823-1844.
- van Dyck CH, Swanson CJ, Aisen P, et al. Lecanemab in early Alzheimer's disease. N Engl J Med. 2023;388(1):9-21.
- Sims JR, Zimmer JA, Evans CD, et al. Donanemab in early symptomatic Alzheimer disease. JAMA. 2023;330(6):512-527.
- Cummings J, Apostolova L, Rabinovici GD, et al. Lecanemab: appropriate use recommendations. J Prev Alzheimers Dis. 2023;10(3):362-377.
- Hampel H, Elhage A, Cho M, et al. Amyloid-related imaging abnormalities (ARIA): radiological, biological and clinical characteristics. Brain. 2023;146(11):4414-4424.
- Sperling R, Salloway S, Brooks DJ, et al. Amyloid-related imaging abnormalities in patients with Alzheimer's disease treated with bapineuzumab. Lancet Neurol. 2012;11(3):241-249.
- Jack CR, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer's disease. Alzheimers Dement. 2024;20(8):5143-5169.
- Greenberg SM, Charidimou A. Diagnosis of cerebral amyloid angiopathy: evolution of the Boston criteria. Stroke. 2018;49(2):491-497.
- Cummings J. Anti-amyloid monoclonal antibodies are transformative treatments that redefine Alzheimer's disease therapeutics. Drugs. 2023;83(7):569-576.