The Drugs, Honestly

Thomas Brennan is 74 and has a yellow legal pad with eleven questions on it. His wife Alice, 71, has added three more in the margins. They have been sitting in the waiting room of Dr. Karen Walsh’s neurology practice in Tucson for twenty minutes, and they have spent the twenty minutes the way they have spent the past two months: trying to reconcile two completely different stories about the same drug.

Thomas was diagnosed with early-stage Alzheimer’s disease eight months ago. Amyloid-positive, confirmed by PET scan. His neurologist has presented lecanemab as a treatment option. Thomas and Alice have read about it extensively. Some articles describe it as a breakthrough. Others describe it as a dangerous overpromise. They have found nothing in between that explains what 27% slowing of decline means in the life of a 74-year-old man who still does his own taxes and walks two miles every morning.

Dr. Walsh’s door opens. Thomas picks up the legal pad. The appointment is about to begin, and the questions on that pad are the right questions.

What Lecanemab and Donanemab Actually Do

#

Both drugs are monoclonal antibodies that clear amyloid-beta plaques from the brain. Amyloid plaques are protein deposits that accumulate between neurons in Alzheimer’s disease and are believed to contribute to the cascade of damage that kills brain cells. Lecanemab, marketed as Leqembi, received full FDA approval in July 2023. Donanemab, marketed as Kisunla, received FDA approval in July 2024.

The clinical trials showed that both drugs slow cognitive decline by approximately 27% to 35% over eighteen months in early-stage, amyloid-positive patients. The measurement uses a validated clinical scale called the CDR-SB, the Clinical Dementia Rating Sum of Boxes, which tracks function across six domains of daily life.

What the drugs do not do: repair neurons that have already died. Prevent the accumulation of tau protein, which is the other hallmark pathology of Alzheimer’s. Restore function that has already been lost. Reverse the disease. Cure the disease. These are not minor caveats. They define the boundary of what treatment means in 2026 for Alzheimer’s disease, and that boundary is narrower than most media coverage suggests.

What 27% Slowing Means in Daily Life

#

This is the section that matters most, because it is the section that almost nobody writes.

The CDR-SB scale runs from 0 to 18. A score of 0 indicates no impairment. The placebo group in the lecanemab trial progressed by approximately 1.66 points over eighteen months. The treatment group progressed by approximately 1.21 points. The difference is 0.45 points on an eighteen-point scale.

That difference is statistically significant. It met the trial’s primary endpoint. It is a real effect produced by a real drug. It is also, on a day-to-day basis, not something most families would notice. The 27% slowing does not mean the person is 27% better than they would have been. It means the rate of decline is 27% slower, measured over eighteen months, producing a difference of less than half a point on a clinical scale. Translation: a few more months of preserved function at the current level before the next measurable decline. Spread over a year and a half, that is real. It is not what most families imagine when they read the word “breakthrough.”

The honest framing: lecanemab and donanemab buy time. Not a lot of time. Enough time that, for some patients, the additional months of preserved function are worth the cost, the risk, and the infusion schedule. For other patients, they are not. That is a decision the patient and family make with their neurologist, and the quality of that decision depends entirely on understanding what 27% means in their specific life.

Who These Drugs Are For

#

Both drugs are approved for early-stage patients only. Mild cognitive impairment and mild dementia due to Alzheimer’s disease. Moderate and advanced stages are not eligible because the drugs target amyloid, and by the moderate stage, the damage has progressed beyond what amyloid clearance alone can meaningfully slow.

Amyloid-positive status is required. This means a PET scan or, in some cases, a lumbar puncture confirming that amyloid plaques are present. Not every person with Alzheimer’s symptoms has been tested for amyloid. Not every clinic has a PET scanner. The geographic access problem is real: rural patients and patients in communities without academic medical centers may not have practical access to the testing required to determine eligibility.

APOE4 status matters significantly. The APOE4 gene variant, carried by approximately 25% of the population and a known risk factor for Alzheimer’s, substantially increases the risk of the most serious side effect of these drugs. Patients who are APOE4 homozygotes, meaning they carry two copies of the variant, face the highest risk and require the most careful benefit-risk discussion with their neurologist.

The ARIA Risk, Explained

#

ARIA, amyloid-related imaging abnormalities, is the primary safety concern. It encompasses two types of brain changes visible on MRI: edema (brain swelling) and microhemorrhages (small bleeds). In the lecanemab trial, ARIA occurred in approximately 21% of treated patients. Most cases were asymptomatic, detected on routine monitoring MRIs, and resolved without intervention.

In a subset of patients, ARIA is symptomatic: headache, confusion, visual disturbances, nausea. These cases require monitoring and sometimes treatment interruption. In rare cases, ARIA is serious. Three deaths were reported in the lecanemab trial; the relationship between the drug and the deaths is contested. The FDA’s label reflects the risk.

APOE4 carriers face higher ARIA rates. APOE4 homozygotes face the highest rates. The monitoring protocol requires regular MRIs during treatment, typically before infusions four, seven, and twelve, and additional scans if symptoms appear. The infusion schedule itself requires an IV infusion every two weeks for lecanemab, which means proximity to an infusion center and the physical capacity to tolerate biweekly medical appointments.

What It Costs and Who Can Get It

#

Lecanemab costs approximately $26,500 per year. Donanemab costs approximately $32,000 per year. Medicare covers both drugs with prior authorization, which requires confirmation of amyloid-positive status and enrollment in a registry to collect real-world evidence. Supplemental insurance may cover the copay, but out-of-pocket costs vary.

The cost is not just the drug. It includes the PET scan to confirm amyloid status, the biweekly infusion visits, the monitoring MRIs, and the time. Thomas and Alice live twelve miles from an infusion center. A patient in rural Wyoming might live two hundred miles from one. The geographic barrier is not theoretical. For the full account of how these drugs were developed and what the evidence shows, see What the New Drugs Actually Do on Blue Gray Matters.

The equity dimension deserves a sentence of its own. The patients most likely to benefit from early treatment are the patients most likely to be diagnosed early. Patients diagnosed early are disproportionately white, insured, and urban. The drug itself does not discriminate. The system that determines who receives it does.

What Is Coming

#

The pipeline beyond lecanemab and donanemab is active. Anti-tau therapies, targeting the other hallmark pathology of Alzheimer’s, are in advanced clinical development. If amyloid clearance slows the disease by 27%, the hypothesis is that adding tau clearance may slow it further. No anti-tau drug has yet demonstrated efficacy in a Phase 3 trial.

GLP-1 receptor agonists, the class of drugs that includes semaglutide and liraglutide, are in active Alzheimer’s trials. The emerging connection between insulin resistance and Alzheimer’s pathology, documented in the diabetes-dementia research literature, has made this drug class a credible candidate. Results from major trials are expected within three to five years.

Subcutaneous injection formulations of lecanemab are in late-stage trials. If approved, they would replace the biweekly IV infusion with a self-administered injection, substantially reducing the access barrier for patients who do not live near infusion centers. The honest timeline for meaningful new treatment options beyond what exists today: three to five years, not three to five months.

Thomas and Alice’s Decision

#

Thomas proceeds with lecanemab. He is amyloid-positive, early stage, APOE4 heterozygote (one copy, not two, which places his ARIA risk in the moderate range), and he lives close enough to an infusion center to make the biweekly schedule manageable. The 27% slowing is real and, in his specific clinical situation, worth pursuing.

Alice understands what 27% means now. She understood it before the appointment, approximately. She understands it precisely now, because Dr. Walsh spent twenty minutes on that question alone. The number is not hope. It is not despair. It is a measured effect that translates to a few additional months of preserved function, and those months have specific value because Thomas intends to use them for the planning and the conversations and the decisions that BML-04.C1 describes. The treatment is buying time. Thomas is spending the time well.

The legal pad has no more questions on it. The decision was theirs to make, and they made it with the full information that two months of reading had not provided and forty minutes with a neurologist who answered every question did. The drugs, honestly, are modest and real. The decisions they enable are the patient’s to make. The clarity to make them well is the thing that was missing, and it was never a pharmacological problem.