Leqembi and Kisunla: Who Actually Qualifies

Two drugs now change the course of early Alzheimer's — modestly. Here is who they are for, what the risks really are, and why the window closes.

What these drugs actually do

They clear amyloid plaque from the brain and modestly slow the rate of decline. They do not restore memory or stop the disease.

Lecanemab (brand name Leqembi, approved 2023) and donanemab (Kisunla, 2024) are monoclonal antibodies that bind to amyloid beta and help the body clear it. In their pivotal trials, both slowed the rate of cognitive and functional decline by roughly a quarter to a third over about 18 months compared with placebo.

That is a real effect, and it is the first time any treatment has changed the underlying course of the disease rather than easing symptoms. It is also modest. Nobody gets better. A person on treatment still declines — the hope is that they decline more slowly, and that the months bought are good ones.

Who qualifies

Treatment is meant to begin at mild cognitive impairment or mild dementia, with amyloid confirmed and a brain MRI that clears you to start.

In practice, most programs work through a checklist something like this:

• Stage. Mild cognitive impairment or mild dementia due to Alzheimer's. This is the stage at which both labels say treatment should be initiated, because it is the population studied. Moderate and severe dementia were not studied, and treatment is not started there.
• Confirmed amyloid. An amyloid PET scan, a spinal fluid test, or in some settings an appropriate blood biomarker test. A clinical impression of Alzheimer's is not sufficient.
• ApoE ε4 genotype. Both labels recommend testing before treatment, because carrying two copies substantially raises the risk of ARIA.
• A baseline MRI. Certain findings — including a significant number of pre-existing microhemorrhages, superficial siderosis, or a prior large brain bleed — generally rule treatment out.
• Blood thinners and clotting risk. Anticoagulant use is a serious consideration and is often a reason not to treat.
• Practical access. A prescriber, an infusion or dispensing pathway, and an MRI-capable center close enough to attend repeatedly. For rural families this, rather than the drug itself, is frequently the real barrier.

A great many people who want these drugs turn out not to qualify. That is worth knowing before you get your hopes up — and it is the single strongest argument for not delaying an evaluation, because stage is the one criterion that only moves in one direction.

ARIA: the risk families most need to understand

ARIA means swelling or small bleeds in the brain. It is common, usually silent, occasionally serious, and it is why the MRI schedule exists.

Both drugs carry a boxed warning — the FDA's strongest — for amyloid-related imaging abnormalities. Most cases are picked up on a routine scan and cause no symptoms at all. Some cause headache, confusion, dizziness, nausea, visual changes or seizures. Rarely, ARIA has been fatal.

Risk depends heavily on genetics. Across the labels, ARIA of any kind occurred in roughly 45% of people carrying two copies of APOE4 on lecanemab and 55% on donanemab, compared with about 13% and 25% of non-carriers. Symptomatic brain swelling in the two-copy group ran around 8–9%.

That is why ApoE testing comes first, and it is a genuinely difficult conversation — the result also tells adult children something about their own risk that they may not have asked to know.

What treatment actually involves

Regular dosing, four monitoring MRIs in the first year, and a schedule that takes over the calendar.

Lecanemab has historically meant an intravenous infusion every two weeks; donanemab is a monthly infusion. Both require a baseline MRI and then monitoring scans before the 3rd, 5th, 7th and 14th doses. The FDA added that earlier third-dose scan in August 2025 after reviewing serious early cases of brain swelling.

In July 2026 the FDA approved a subcutaneous autoinjector for lecanemab as a starting dose, adding to the maintenance version approved in 2025 — so it is now possible to begin treatment at home rather than at an infusion center. Initiation is 500 mg once weekly, given as two 250 mg injections; after 18 months the dose drops to 360 mg weekly. US availability is expected from late August 2026.

That removes the biggest logistical burden of this treatment. It does not remove the MRI schedule, and it moves the injection itself onto a family member — which is a real change worth discussing before you agree to it.

Donanemab has a defined endpoint that lecanemab does not: it can be stopped once amyloid has been cleared, often around 12 to 18 months, rather than continuing indefinitely. Trials of a once-yearly maintenance dose are ongoing but that approach is not approved.

What it costs, and who pays

Medicare covers both for people who meet criteria, but what you pay out of pocket depends on which part of Medicare the drug falls under.

Intravenous infusions are generally billed under Medicare Part B, where beneficiaries typically pay 20% coinsurance after the Part B deductible, with no annual out-of-pocket ceiling. Many people rely on Medigap or a Medicare Advantage plan's cap to limit that exposure.

The at-home injection is dispensed by a specialty pharmacy, so it is expected to fall under the Part D pharmacy benefit — where 2026 brings a $2,100 annual out-of-pocket cap. That could be a large difference for some households, but coverage varies by plan and this is new enough that it is worth confirming with your specific plan rather than assuming.

The list price is $385 per autoinjector. Note that a starting dose uses two autoinjectors each week, so the weekly list cost at initiation is roughly double the per-pen figure.

Separately, budget for the scans, the specialist visits, and the travel. For families outside metropolitan areas, the repeated MRI requirement is often the largest hidden cost.

What actually happens in the real world

A meaningful minority of people stop early — and that is useful to know going in.

An analysis of 458 patients who began lecanemab between January 2023 and September 2025, presented in April 2026, found that about 63% were still adherent at six months. Around 7% stopped after only one or two doses. Among those with imaging data available, roughly 91% received follow-up imaging within six months.

People stop for all the usual human reasons: an ARIA finding, an infusion reaction, the burden of the schedule, a decline that outpaces the drug, or a decision that it simply isn't worth it. None of those are failures.

Going in with clear eyes — about the size of the benefit, the real risks, and how much of your life the schedule will take — is what makes the decision yours rather than something that happens to you.

Why this belongs in a conversation about early diagnosis

Stage is the one eligibility criterion that never improves while you wait.

Families wait an average of three and a half years between first symptoms and a diagnosis. For someone whose disease begins in mild cognitive impairment, that delay can be the entire treatment window.

You may decide these drugs are not right for your person. Plenty of thoughtful families do — the benefit is modest, the risks are real, and the schedule is demanding. But that should be a decision you get to make, not one that gets made for you by a year spent trying to get someone to write down a diagnosis.