Anti-amyloids in early Alzheimer's: eligibility and ARIA

A 71-year-old woman comes to the memory clinic with her daughter. For a year and a half, she has forgotten appointments, repeats questions and has started writing everything down in a notebook, but she still cooks, drives near home and pays her own bills. Cognitive assessment confirms mild cognitive impairment. Positron emission tomography (PET) shows brain amyloid, and genetic testing finds one copy of the apolipoprotein E ε4 allele (APOE4). She has atrial fibrillation and takes an oral anticoagulant. The daughter read that there are new drugs that "clean the brain" and asks whether her mother can start. The scene is fictional and was written for teaching.

This text uses the case as a clinical reasoning scenario for students and health professionals. The daughter's question opens several others: what anti-amyloid antibodies do and how much they help, where they are approved, who is eligible, what ARIA is, how genotype changes the risk and what to do about the anticoagulant. The text does not replace the neurologist's assessment, the label in your country or your institution's protocol. If you play the case of the day on NEXO, you will recognize the challenge: weighing benefit and risk when no single clue decides on its own.

Key points

• Lecanemab and donanemab are monoclonal antibodies that remove beta-amyloid plaques and modestly slow decline in early Alzheimer's.
• The indication is mild cognitive impairment or mild dementia with amyloid confirmed by PET or cerebrospinal fluid.
• The FDA approved lecanemab through the traditional pathway on July 6, 2023, and donanemab on July 2, 2024.
• In the European Union, both were authorized in 2025 only for people with one or no copies of APOE4, under a controlled access program.
• ARIA is the main risk: edema or microhemorrhages seen on MRI, more common in APOE4 carriers, especially homozygotes.
• Anticoagulant use weighs against treatment in the use recommendations; thrombolytics should be avoided.

Why this is a reasoning scenario

Early Alzheimer's is when families most seek an answer and when the decision is hardest. Anti-amyloid antibodies brought the first option that acts on the pathology, but with modest benefit, real risk and requirements for tests and follow-up (van Etten and colleagues, 2026).

The scenario is good for training because each clue changes the reasoning. Mild impairment suggests a treatment window. PET confirms the pathology. APOE4 raises the risk. The anticoagulant creates a conflict. The post on signs, symptoms and history shows how to listen to what the family reports, and the post on diagnostic uncertainty discusses how to decide when the data do not add up.

What the scene teaches

The scene trains five habits. The first is to confirm the clinical stage and the presence of amyloid before talking about medication. The second is to know the real size of the benefit. The third is to check where and for whom each drug was approved. The fourth is to estimate the risk of ARIA with the genotype. The fifth is to treat the anticoagulant as a central clue, not as a detail.

What is not at stake

No one should stop the anticoagulant for atrial fibrillation to make the patient eligible without assessment by the cardiologist and the neurologist. The risk of embolic stroke remains. The decision is shared among the teams, the patient and the family.

What anti-amyloid antibodies do

Lecanemab and donanemab are monoclonal antibodies directed at beta-amyloid (Demir and Karakuş Dilbaz, 2026). Lecanemab is directed at soluble and insoluble aggregated forms of beta-amyloid (US Leqembi label, revised 7/2026). Donanemab is also directed at aggregated forms of amyloid and can be stopped when plaques fall to minimal levels on PET (US Kisunla label, revised 7/2025). For lecanemab, the label provides for maintenance after 18 months (US Leqembi label, revised 7/2026).

Removing plaque is not curing

Both drugs reduce amyloid on PET, and this reduction is accompanied by a slower clinical decline (Sims et al., 2023), (US Leqembi label, revised 7/2026). They do not restore lost memory or stop the disease. The goal is to gain time at the stage when the person still has autonomy.

How they are given

• Lecanemab intravenously: 10 mg/kg every 2 weeks, in an infusion of about 1 hour.
• Lecanemab subcutaneously (Iqlik autoinjector): 500 mg once a week.
• Lecanemab after 18 months: keep the starting dose or switch to 10 mg/kg intravenously every 4 weeks, or 360 mg subcutaneously per week.
• Donanemab intravenously every 4 weeks, in about 30 minutes: 350 mg, then 700 mg, then 1,050 mg and, from the fourth infusion on, 1,400 mg.
• Donanemab: consider stopping when PET shows plaques at a minimal level.

Sources for the list: (US Leqembi label, revised 7/2026) and (US Kisunla label, revised 7/2025). The 350, 700, 1,050 and 1,400 mg schedule is the one recommended today by the label; TRAILBLAZER-ALZ 2 used 700, 700, 700 and 1,400 mg (US Kisunla label, revised 7/2025), (Wang et al., 2025).

What the studies showed and what was approved

The benefit was measured on scales that combine cognition and function. It is important to read the numbers as group averages, not as a promise for one person.

Lecanemab

In the label, the main lecanemab study showed, at 18 months, a difference of 0.45 point on the CDR-SB in favor of the drug, about 27 percent less decline than placebo (US Leqembi label, revised 7/2026). The CDR-SB ranges from 0 to 18, and higher scores indicate more impairment.

Donanemab

In TRAILBLAZER-ALZ 2, Sims et al. randomized 1,736 participants with early symptomatic Alzheimer's (Sims et al., 2023). At 76 weeks, in the population with low or intermediate tau, the iADRS scale fell 6.02 points with donanemab and 9.27 with placebo; in the total population, 10.19 and 13.11 (Sims et al., 2023). On the CDR-SB, the slowing was 36 percent in the population with low or intermediate tau (Sims et al., 2023). In the label, the difference in the total population was 2.92 points on the iADRS and 0.70 on the CDR-SB (US Kisunla label, revised 7/2025).

Stopping when the plaque clears

In TRAILBLAZER-ALZ 2, those who reached sufficient amyloid reduction switched to placebo in a blinded way. According to the label, this happened to 17 percent of those treated at week 24, 47 percent at week 52 and 69 percent at week 76 (US Kisunla label, revised 7/2025). This shows a practical difference: donanemab may have a limited duration.

Aspect	Lecanemab	Donanemab
Route and interval	Intravenous every 2 weeks or subcutaneous weekly	Intravenous every 4 weeks
Duration	Continuous, with maintenance after 18 months	Can stop with minimal plaque on PET
FDA approval	Traditional on 07/06/2023	07/02/2024
Authorization in the European Union	04/15/2025	09/24/2025
European restriction by genotype	One or no copies of APOE4	One or no copies of APOE4
Follow-up MRIs in the label	After 1, 2, 3 and 6 months of treatment	Before infusions 2, 3, 4 and 7

Sources for the table: (US Leqembi label, revised 7/2026), (US Kisunla label, revised 7/2025), (FDA, 2023), (FDA, 2024), (EMA, Leqembi) and (EMA, Kisunla).

Regulatory status: FDA and EMA

The regulatory status differs on the two sides of the Atlantic, and the most important difference is genotype.

United States. On July 6, 2023, the FDA converted the accelerated approval of Leqembi (lecanemab) into traditional approval, based on the confirmatory study, and added the boxed warning about ARIA (FDA, 2023). On July 2, 2024, the FDA approved Kisunla (donanemab) for adults with early symptomatic Alzheimer's, also with a boxed warning about ARIA (FDA, 2024). The labels in force on the date of this text were revised in 7/2026 (Leqembi) and 7/2025 (Kisunla) (US Leqembi label, revised in 7/2026), (US Kisunla label, revised in 7/2025). In the United States, genotype does not exclude treatment, but the label calls for APOE4 testing before starting and a discussion of the risk (US Leqembi label, revised in 7/2026), (US Kisunla label, revised in 7/2025).

European Union. Leqembi received marketing authorization in the European Union on April 15, 2025, after a CHMP opinion on November 14, 2024, for adults with mild cognitive impairment or mild dementia due to Alzheimer's who have one or no copies of APOE4 (EMA, Leqembi). Kisunla had a negative opinion in March 2025, a favorable re-examination in July 2025 and authorization on September 24, 2025, with the same genotype restriction (EMA, Kisunla). Both require a controlled access program, European registry and educational materials, including a patient card in the case of donanemab (EMA, Leqembi), (EMA, Kisunla).

Other countries. Outside these two regions, the indication depends on the local agency. The patient in the scene, with one copy of APOE4, would not be excluded by genotype in either the United States or the European Union. What weighs in her case is another clue.

Who can receive it: eligibility

The appropriate use recommendations for lecanemab, by Cummings and colleagues, and for donanemab, by Rabinovici and colleagues, translate the trial criteria into practice (Cummings and colleagues, 2023), (Rabinovici and colleagues, 2025).

Clinical criteria

For lecanemab, the recommendations call for mild cognitive impairment or mild dementia due to Alzheimer's, with a score of 22 to 30 on the Mini-Mental State Examination (MMSE) (Cummings and colleagues, 2023). For donanemab, they call for age 60 to 85 years and an MMSE of 20 to 30 (Rabinovici and colleagues, 2025). In both, the presence of amyloid must be confirmed by PET or CSF testing before starting (Cummings and colleagues, 2023), (Rabinovici and colleagues, 2025), (US Leqembi label, revised in 7/2026).

Criteria on MRI

The recommendations follow the trial exclusions: cerebral hemorrhage larger than 10 mm, more than 4 microhemorrhages, superficial siderosis, vasogenic edema, extensive white matter lesions, multiple lacunar infarcts or a large-territory stroke (Cummings and colleagues, 2023). The baseline MRI must be recent (US Leqembi label, revised in 7/2026), (US Kisunla label, revised in 7/2025).

Medications that weigh against

Both recommendations exclude people using anticoagulants, such as warfarin, direct oral anticoagulants and heparin, and recommend avoiding thrombolytics (Cummings and colleagues, 2023), (Rabinovici and colleagues, 2025). This is the point that blocks the patient's case, and it is covered in its own section below.

What the team checks before proposing

• Clinical stage: mild impairment or mild dementia, with an MMSE in the range of the recommendations.
• Amyloid confirmed by PET or CSF.
• Recent MRI without the exclusion findings.
• APOE4 genotype, discussed with the patient before testing.
• Medication list, with attention to anticoagulants and antiplatelet agents.
• Ability to attend the infusions and the follow-up MRIs.
• The patient's and family's expectations about a modest benefit.

Sources for the list: (Cummings and colleagues, 2023), (Rabinovici and colleagues, 2025), (US Leqembi label, revised in 7/2026) and (US Kisunla label, revised in 7/2025).

ARIA: what it is, who is at higher risk and how to monitor

ARIA stands for amyloid-related imaging abnormalities. There are two forms: ARIA-E, with edema or effusion, and ARIA-H, with microhemorrhages or superficial siderosis (van Etten et al., 2026). The hypothesis is that removal of amyloid from the vessels damages the blood-brain barrier and inflames the vessel wall (Schlemm et al., 2026).

How frequent

In TRAILBLAZER-ALZ 2, ARIA-E occurred in 24.0 percent of those treated with donanemab and in 2.1 percent with placebo; 52 participants had ARIA-E with symptoms (Sims et al., 2023). There were 3 deaths considered related to treatment (Sims et al., 2023). Infusion reactions occurred in 8.7 percent (Sims et al., 2023). Zimmer and colleagues showed that symptomatic ARIA tends to be more severe on imaging and that most events occur early in treatment (Zimmer et al., 2025).

Genotype changes the risk

The prescribing information shows a clear gradient. With lecanemab, ARIA occurred in 45 percent of APOE4 homozygotes, 19 percent of heterozygotes and 13 percent of non-carriers; symptomatic ARIA-E, in 9, 2 and 1 percent (US Leqembi prescribing information, revised 7/2026). With donanemab, ARIA occurred in 55 percent of homozygotes, 36 percent of heterozygotes and 25 percent of non-carriers; symptomatic ARIA-E, in 8, 7 and 4 percent (US Kisunla prescribing information, revised 7/2025). The Leqembi prescribing information notes that about 15 percent of the study patients were homozygotes (US Leqembi prescribing information, revised 7/2026).

The modified titration

In TRAILBLAZER-ALZ 6, Wang and colleagues compared donanemab starting regimens; the 350, 700, 1,050 and 1,400 mg regimen reduced ARIA-E by 41 percent in relative risk compared with the original regimen, while maintaining amyloid reduction (Wang et al., 2025). In a real-world study, Hakim and colleagues observed ARIA in 8.1 percent with the modified titration, about three times less than with the original regimen (Hakim et al., 2026).

How to monitor

For lecanemab, the prescribing information calls for a recent MRI before starting and further MRIs after 1 month of treatment (before the 3rd intravenous or 5th subcutaneous dose), 2 months, 3 months and 6 months, with reinforced clinical monitoring in the first 14 weeks (US Leqembi prescribing information, revised 7/2026). For donanemab, the MRI is done before the 2nd, 3rd, 4th and 7th infusions (US Kisunla prescribing information, revised 7/2025). In the presence of new symptoms, such as headache, confusion, visual change, dizziness, nausea or difficulty walking, the MRI is brought forward, and the dose is suspended or maintained according to the severity of ARIA on imaging and the symptoms, per the tables in the prescribing information (US Leqembi prescribing information, revised 7/2026), (US Kisunla prescribing information, revised 7/2025). Most events are asymptomatic and only appear because the MRI was done (van Etten et al., 2026).

Anticoagulant and antiamyloid: the point of the case

The patient in the scene takes an anticoagulant for atrial fibrillation. That is the clue that weighs most.

What the recommendations and prescribing information say

The appropriate use recommendations exclude people taking anticoagulants from treatment with lecanemab and donanemab (Cummings et al., 2023), (Rabinovici et al., 2025). The update from the Alzheimer's Association ARIA working group maintains caution and recalls that giving thrombolytics to treated people carries a very high risk, with reported deaths (van Etten et al., 2026). The Leqembi prescribing information records intracerebral hemorrhage in 2.5 percent (2 of 79) of patients taking an anticoagulant, and the Kisunla one describes 2 fatal intracerebral hemorrhages after thrombolytic (US Leqembi prescribing information, revised 7/2026), (US Kisunla prescribing information, revised 7/2025). Both prescribing information documents call for caution with anticoagulants (US Leqembi prescribing information, revised 7/2026), (US Kisunla prescribing information, revised 7/2025).

What the meta-analysis says

Schlemm and colleagues pooled four clinical trials and one real-world study and found no increase in ARIA-E or ARIA-H with oral anticoagulant (Schlemm and colleagues, 2026). But major bleeding occurred in 1.4 percent of those using an anticoagulant and in 0.23 percent of those not using one (Schlemm and colleagues, 2026). The authors call for caution in interpretation, because the absolute numbers are small and the studies were not designed for that question (Schlemm and colleagues, 2026).

How to reason through the case

With one copy of APOE4, the patient would not be excluded by genotype, but her ARIA risk is higher than that of a non-carrier (US label for Leqembi, revised 7/2026), (US label for Kisunla, revised 7/2025). Added to the anticoagulant, the picture falls outside the recommendations for appropriate use (Cummings and colleagues, 2023), (Rabinovici and colleagues, 2025). An honest conversation includes the small average benefit, the bleeding risk and the fact that stopping the anticoagulant for atrial fibrillation also carries risk. Alternatives for preventing stroke without anticoagulation are a matter for the cardiologist. Meanwhile, memory follow-up, control of vascular risk factors and support for the family continue.

If the patient were on treatment

If someone using an antiamyloid reaches the emergency department with ischemic stroke, the team needs to know about the treatment before considering thrombolysis (van Etten and colleagues, 2026). That is why the patient card required in the European Union is more than a formality (EMA, Kisunla).

How NEXO helps train this reasoning

In NEXO, you try the diagnosis by searching for the code in ICD-10 and receive proximity feedback in five levels, which shows whether the guess landed on the exact code, the same category, the same group, the same chapter or a different chapter. This format fits the theme of this text: Alzheimer's disease, mild cognitive impairment, atrial fibrillation and intracerebral hemorrhage sit in different chapters, but meet in the same decision. The post about ICD-10 shows how to find the right code.

To use the case of the day as training, try three habits. Before the first guess, ask which clue changes management. With each clue, ask whether it raises or lowers the risk of what you are thinking of doing. After the result, note the clue you treated as a detail.

Three exercises with cases

Find the conflict. In each case with a new drug, look for the old drug that can make the new one dangerous.

Use the genotype. Say how the risk would change if the patient were homozygous for APOE4.

Check the agency. Say whether the patient would be eligible under the US label and under the European authorization.

The post about differential diagnosis with clues shows how to organize hypotheses, the post about case-based learning in clinical practice shows how cases help fix these habits, and the post about clinical deterioration on the ward reminds you that new confusion and severe headache are warning signs.

The limits of the game's use are described on the NEXO economy page. If you research teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

Do these drugs cure Alzheimer's disease?

No. They modestly slow decline in the early stage (Sims and colleagues, 2023), (US label for Leqembi, revised 7/2026).

Can someone with two copies of APOE4 receive them?

In the European Union, no (EMA, Leqembi), (EMA, Kisunla). In the United States, the label allows it after discussion of the risk, which is much higher (US label for Leqembi, revised 7/2026), (US label for Kisunla, revised 7/2025).

Does ARIA always cause symptoms?

No. Most cases are asymptomatic and appear only on the follow-up MRI (van Etten and colleagues, 2026).

Can someone using an anticoagulant receive them?

Appropriate use recommendations exclude these people, and the prescribing information calls for caution (Cummings et al., 2023), (Rabinovici et al., 2025), (US Leqembi prescribing information, revised 7/2026), (US Kisunla prescribing information, revised 7/2025).

Is donanemab forever?

Not necessarily. The prescribing information advises considering stopping when plaque reaches a minimal level on PET (US Kisunla prescribing information, revised 7/2025).

Final checklist

• I confirmed mild cognitive impairment or mild dementia and an MMSE in the recommended range.
• I confirmed amyloid by PET or CSF.
• I reviewed a recent MRI for microhemorrhages, siderosis and other exclusion findings.
• I discussed APOE4 testing before ordering it and explained the risk according to genotype.
• I checked anticoagulants and antiplatelet agents and involved the cardiologist when needed.
• I checked the country's regulatory status and the date of the decision.
• I agreed on the MRI schedule and the symptoms that require seeking care.
• I advised the family to report the treatment at any emergency care visit.

Want to practice your reasoning with progressive clues? Play the case of the day.

References

1. Cummings J, Apostolova L, Rabinovici GD, et al. Lecanemab: Appropriate Use Recommendations. J Prev Alzheimers Dis. 2023. PMID 37357276.
2. Demir ÜF, Karakuş Dilbaz F. Anti-Amyloid Monoclonal Antibodies in Early Alzheimer Disease: Lecanemab and Donanemab. Neurol Int. 2026. PMID 42784125.
3. Eisai Inc. LEQEMBI® (lecanemab-irmb) injection, for intravenous or subcutaneous use. Prescribing Information. Revised 7/2026. Accessed 24/10/2026. Official source.
4. Eli Lilly and Company. KISUNLA (donanemab-azbt) injection, for intravenous use. Prescribing Information. Revised 7/2025. Accessed 24/10/2026. Official source.
5. European Medicines Agency. Kisunla. EPAR. Accessed 24/10/2026. Official source.
6. European Medicines Agency. Leqembi. EPAR. Accessed 24/10/2026. Official source.
7. Hakim R, Harris K, Merrill L, et al. Real-world evidence for amyloid-targeted therapies: Lecanemab and donanemab. Alzheimers Dement (Amst). 2026. PMID 42661728.
8. Rabinovici GD, Selkoe DJ, Schindler SE, et al. Donanemab: Appropriate use recommendations. J Prev Alzheimers Dis. 2025. PMID 40155270.
9. Schlemm E, Gauthier S, Magnus T, et al. Risk of amyloid-related imaging abnormalities associated with anticoagulant therapy in patients with Alzheimer's disease treated with anti-amyloid monoclonal antibodies: a systematic review and meta-analysis. J Neurol Neurosurg Psychiatry. 2026. PMID 41448897.
10. Sims JR, Zimmer JA, Evans CD, et al. Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial. JAMA. 2023. PMID 37459141.
11. U.S. Food and Drug Administration. FDA approves treatment for adults with Alzheimer’s disease. 02/07/2024. Official source.
12. U.S. Food and Drug Administration. FDA Converts Novel Alzheimer’s Disease Treatment to Traditional Approval. 06/07/2023. Official source.
13. van Etten ES, Mahinrad S, Grill JD, et al. Amyloid-related imaging abnormalities (ARIA) in anti-amyloid therapies for Alzheimer's disease: An update from the Alzheimer's Association ARIA workgroup. Alzheimers Dement. 2026. PMID 42015334.
14. Wang H, Nery ESM, Ardayfio P, et al. The effect of modified donanemab titration on amyloid-related imaging abnormalities with edema/effusions and amyloid reduction: 18-month results from TRAILBLAZER-ALZ 6. J Prev Alzheimers Dis. 2025. PMID 40619285.
15. Zimmer JA, Ardayfio P, Wang H, et al. Amyloid-Related Imaging Abnormalities With Donanemab in Early Symptomatic Alzheimer Disease: Secondary Analysis of the TRAILBLAZER-ALZ and ALZ 2 Randomized Clinical Trials. JAMA Neurol. 2025. PMID 40063015.