What the New Drugs Actually Do

They are sitting in a neurologist’s office, married forty-three years. He is seventy-one. The PET scan confirmed what they suspected: amyloid plaques in his brain, the signature of Alzheimer’s disease. His cognitive symptoms are mild. He forgets appointments, loses the thread of conversations, asks the same questions twice. He still drives, still manages the finances, still knows exactly who she is.

The neurologist is explaining the options. There are new drugs now, she says. For the first time, treatments that can slow the disease rather than just manage symptoms. She describes the infusions, the monitoring, the risks, the costs. She is careful and thorough.

He listens, then asks the question that matters most: “Will this fix it?”

The pause before she answers is the most important moment in this conversation. What she says next will determine whether this family trusts her, trusts the treatment, trusts the process ahead. The answer is not yes. The answer is not no. The answer is complicated, and it deserves to be told straight.

What These Drugs Are

For the first time in history, FDA-approved treatments can slow the progression of Alzheimer’s disease. This is not hype. It is not a misreading of the data. After decades of failed trials and false starts, two drugs have demonstrated, in large rigorous studies, that they modestly slow cognitive decline in people with early-stage Alzheimer’s.

The drugs are lecanemab (brand name Leqembi) and donanemab (brand name Kisunla). Both are monoclonal antibodies, laboratory-engineered proteins designed to target and help clear amyloid beta from the brain. Amyloid beta is the protein that aggregates into the plaques that have been associated with Alzheimer’s for over a century. Whether amyloid causes the disease or is a byproduct of it has been debated for decades. What the trials showed is that clearing it, in early-stage patients, correlates with slower decline.

Lecanemab targets a form of amyloid called protofibrils, which are thought to be particularly toxic to brain cells. Donanemab targets a modified form called pyroglutamate amyloid, found in mature plaques. Both drugs are given by intravenous infusion. Both require confirmation of amyloid pathology before treatment can begin. Both work only in early-stage disease.

The mechanism is straightforward: bind to amyloid, flag it for removal by the brain’s immune system, reduce the plaque burden over time. The clinical question was whether reducing plaques would translate into preserved cognition. For decades, that question went unanswered or answered in the negative. These drugs changed that.

What “Slowing Decline” Actually Means

The pivotal trial for lecanemab, called CLARITY AD, enrolled nearly 1,800 people with early-stage Alzheimer’s. Half received the drug; half received placebo. After eighteen months, the treated group showed 27 percent less decline on a standard cognitive and functional scale compared to placebo.

Twenty-seven percent sounds significant. What does it mean in practice?

Cognitive decline in Alzheimer’s is measured on scales that track memory, orientation, judgment, and the ability to perform daily activities. Both groups in the trial declined. The placebo group declined more. At the end of eighteen months, the treated group had declined to roughly the point the placebo group had reached five to seven months earlier. Another way to say this: lecanemab bought approximately five to seven months of cognitive time over the course of treatment.

Five to seven months is not a cure. It is not a halt. It is not a reversal. But for a family trying to hold onto every clear conversation, every recognized face, every moment of presence, five to seven months is not nothing. It is time. Whether that time is worth the costs and risks of treatment is a question each family must answer for themselves.

Donanemab showed similar results in its pivotal trial, TRAILBLAZER-ALZ 2. Participants who started treatment earlier in the disease course showed greater benefit. Long-term extension data presented in 2025 suggested that the benefit continued to accrue over three years, with treated patients maintaining a meaningful gap over what would have been expected without treatment.

Donanemab has one feature lecanemab does not: the possibility of stopping treatment once amyloid is sufficiently cleared. In the trials, the majority of patients achieved amyloid-negative status on PET scans within about eighteen months. The hypothesis is that once plaques are cleared, ongoing treatment may not be necessary, though long-term data on what happens after stopping is still being gathered.

What These Drugs Do Not Do

Clarity requires saying what these treatments cannot accomplish.

They do not restore lost function. If someone has already lost the ability to manage finances or remember recent events, these drugs will not bring those abilities back. They may slow further loss. They do not reverse what has already happened.

They do not work in moderate or advanced Alzheimer’s disease. The trials enrolled only people with mild cognitive impairment or mild dementia due to Alzheimer’s. People whose disease has progressed further were excluded. This is not arbitrary. The biology suggests that once neuronal loss has reached a certain point, removing amyloid does not help because the damage is done. The window for these treatments is early, and it closes.

They do not address the other pathologies involved in Alzheimer’s. Amyloid plaques are one part of the disease. Tau tangles, the other hallmark pathology, are not directly targeted by these drugs. Neuroinflammation, synaptic dysfunction, and neuronal death are not directly addressed. Clearing amyloid may slow the cascade, but it does not stop every element of it.

They do not work for everyone who takes them. The 27 percent slowing is an average across the trial population. Some patients showed more benefit; some showed less; some showed none that could be distinguished from placebo. There is currently no reliable way to predict who will respond well.

They are not cures. This word should not appear in any honest conversation about these drugs. They are disease-modifying treatments that offer modest slowing of progression in a subset of patients. That is genuinely new. It is not a cure.

Who Qualifies

Eligibility for lecanemab and donanemab is limited to people with early-stage Alzheimer’s disease and confirmed amyloid pathology.

Early-stage means mild cognitive impairment (MCI) due to Alzheimer’s or mild Alzheimer’s dementia. These are clinical designations based on cognitive testing and functional assessment. Someone who scores in the moderate or severe range on standard scales is not a candidate. Someone whose primary diagnosis is a different dementia (vascular, Lewy body, frontotemporal) is not a candidate, even if they also have some amyloid pathology.

Confirmed amyloid pathology traditionally required an amyloid PET scan, which involves injecting a radioactive tracer and imaging the brain. This is expensive (typically $3,000 to $6,000), requires specialized equipment, and is not available everywhere. The FDA-cleared blood biomarker test, discussed in the first installment of this series, offers an alternative pathway: a positive blood test can support the diagnosis, though many clinicians still recommend confirmatory PET imaging before starting treatment.

Before initiating treatment, patients must also undergo genetic testing for APOE ε4 status. This gene variant is the strongest known genetic risk factor for late-onset Alzheimer’s. It is also a risk factor for the primary side effect of these drugs. People who carry two copies of ε4 (homozygotes) face substantially higher risk of brain swelling and bleeding during treatment. This does not automatically disqualify them, but it changes the risk-benefit calculation and requires careful discussion.

What Treatment Looks Like

Treatment with these drugs is not simple. It requires infrastructure, time, and sustained engagement with the healthcare system.

Lecanemab is given by intravenous infusion. The initial protocol required infusions every two weeks for eighteen months. In 2025, the FDA approved maintenance dosing options: once-monthly IV infusion or weekly subcutaneous injection that patients can administer at home after initial training. The subcutaneous option reduces the infusion center burden but still requires regular medical oversight.

Donanemab is given by monthly IV infusion. A modified dosing protocol approved in 2025 starts with lower doses and titrates up, which reduced the rate of the primary side effect from 24 percent to 14 percent while maintaining efficacy. Treatment continues until amyloid is cleared, typically twelve to eighteen months, after which patients may be able to stop.

Both drugs require regular MRI monitoring to detect ARIA (amyloid-related imaging abnormalities, discussed below). The minimum is five to seven MRIs in the first year of treatment. In 2025, the FDA updated monitoring recommendations for lecanemab to include earlier MRIs following reports of serious adverse events. Each MRI requires scheduling, travel, time, and often a copay.

The time burden falls heavily on caregivers. Someone must drive the patient to infusions, wait during the procedure, drive them home, and monitor for side effects afterward. For a family already stretched thin by caregiving demands, adding biweekly or monthly medical appointments is not trivial.

The Risks

The primary safety concern with both drugs is ARIA: amyloid-related imaging abnormalities. This term covers two related phenomena: ARIA-E (edema, or brain swelling) and ARIA-H (hemorrhage, or microbleeds in the brain).

ARIA occurs because clearing amyloid from blood vessel walls can temporarily destabilize those vessels. In most cases, ARIA is detected on routine MRI monitoring and causes no symptoms. Patients are unaware it is happening. In some cases, it causes headache, confusion, dizziness, nausea, or vision changes. In rare cases, it is serious. In very rare cases, it is fatal.

The numbers: In the lecanemab trial, approximately 21 percent of treated patients experienced ARIA of some kind on MRI. Most cases were asymptomatic. By late 2025, the FDA’s adverse event reporting system had recorded 101 serious ARIA cases with lecanemab, including 6 deaths. These numbers require context. Millions of doses have not yet been administered; the drugs are still early in their rollout. The rate of serious events may be higher or lower than trial data suggested as real-world experience accumulates.

Donanemab’s trial showed a 24 percent rate of ARIA-E with standard dosing. The modified titration protocol reduced this to 14 percent. Donanemab also showed a higher rate of serious ARIA events in APOE ε4 homozygotes, leading to careful consideration before treating this population.

Risk factors for ARIA include APOE ε4 carrier status (especially homozygotes), use of anticoagulant medications, and the presence of cerebral amyloid angiopathy (amyloid deposits in blood vessel walls). Patients on blood thinners require particularly careful evaluation, as anticoagulation may increase bleeding risk if ARIA-H occurs.

The informed consent conversation matters enormously here. Patients and families need to understand that these drugs carry real risks, that monitoring exists to catch problems early, that most ARIA cases resolve without lasting harm, and that serious events, while rare, do occur. The decision to start treatment should be made with clear eyes, not with desperation that accepts any risk for any chance of benefit.

The Cost

Lecanemab costs approximately $26,500 per year for the drug itself. Donanemab is priced similarly. But the drug cost is not the total cost.

Add the amyloid PET scan required for diagnosis: $3,000 to $6,000. Add the MRIs required for monitoring: several hundred dollars each after insurance, multiplied by five to seven in the first year. Add the infusion center fees, the specialist visits, the genetic testing. Add the travel costs for families who do not live near an infusion center. Add the lost wages for caregivers who take time off work to drive to appointments.

Total first-year costs can exceed $40,000. In subsequent years, costs decrease if patients transition to maintenance dosing or complete treatment, but they remain substantial.

Medicare covers both drugs following their traditional FDA approvals. This was not guaranteed; Medicare initially limited coverage during the accelerated approval period, expanding it only after confirmatory trial data emerged. Coverage does not mean free. Medicare Part B requires a 20 percent copay for infused drugs. For a $26,500 drug, that is over $5,000 per year out of pocket, assuming no supplemental coverage.

For the family in the neurologist’s office, cost is part of the calculation. Can they afford the copays? Can the caregiver afford the time off work? Is there an infusion center within reasonable driving distance? These are not abstract questions. They determine whether the treatment that exists on paper exists in practice.

The Access Gap

The infrastructure required to deliver these treatments does not exist everywhere.

Treatment requires an amyloid PET scanner or access to the new blood biomarker tests with confirmatory imaging capability. It requires an infusion center staffed to administer monoclonal antibodies and monitor for reactions. It requires neurologists experienced with anti-amyloid therapy, who know how to interpret ARIA on MRI and when to pause or stop treatment. It requires MRI availability sufficient for the monitoring schedule.

In major metropolitan areas with academic medical centers, this infrastructure largely exists. In rural America, it largely does not. A family in rural Mississippi or Montana may be hours from the nearest neurologist, let alone one experienced with these specific drugs. The new blood tests may eventually bring diagnosis closer to home, but treatment still requires infusion infrastructure that small communities cannot support.

Racial disparities compound geographic ones. Black and Hispanic Americans face the highest risk of Alzheimer’s disease. They are also the least likely to have access to the diagnostic testing, specialist evaluation, and treatment infrastructure these drugs require. They are underrepresented in the clinical trials that generated the efficacy data. They are more likely to be diagnosed late, after the treatment window has closed. The drugs exist. The access is unequal.

The cost-access paradox is stark: the most expensive treatments in Alzheimer’s history are arriving in a healthcare system that was already failing to serve the people who need them most. Approving a drug does not deliver it. Coverage does not equal access. The gap between what is possible and what is available defines much of the American healthcare experience, and it is nowhere more visible than here.

What Comes Next

The current drugs are first-generation treatments. The pipeline behind them is active, and some of what is coming may matter more than what is already here.

Trontinemab, in development by Roche, uses a technology called Brainshuttle that enhances the antibody’s ability to cross the blood-brain barrier. Early-phase data showed 91 percent of participants achieving amyloid-negative status by week 28, faster than current drugs, with ARIA-E rates below 5 percent, substantially lower than current drugs. Phase III trials are underway, with data expected in 2026 or 2027. If the results hold, trontinemab could offer faster amyloid clearance with a better safety profile. That is a meaningful improvement, though it remains to be proven in larger trials.

The more significant shift may come from drugs targeting tau rather than amyloid. Tau tangles, the other hallmark pathology of Alzheimer’s, correlate more closely with cognitive decline than amyloid plaques do. The hypothesis is that clearing amyloid may be necessary but not sufficient, and that addressing tau could provide additional benefit.

Several anti-tau drugs are in clinical development. BIIB080, an antisense oligonucleotide from Biogen, reduces tau production rather than clearing existing tangles. Phase II data is expected in 2026. Etalanetug, from Eisai, targets a specific region of the tau protein involved in its spread from cell to cell. Bristol Myers Squibb and Merck have their own tau programs. Results from these trials, expected between 2026 and 2028, will begin to answer whether the tau hypothesis translates into clinical benefit.

The combination therapy future is the destination many researchers envision. Effective treatment may require addressing amyloid, tau, and neuroinflammation together, the way cancer treatment often requires multiple drugs attacking different pathways. No combination trials have reported results yet. This is a five-to-ten-year horizon at minimum.

GLP-1 agonists, the drugs originally developed for diabetes and now used widely for weight loss (semaglutide is marketed as Ozempic and Wegovy), are being tested for Alzheimer’s. The connection runs through metabolism, inflammation, and vascular health. Phase III data is pending. If positive, these drugs could offer a very different treatment profile: oral or injectable medications already widely available, already covered by insurance, already manufactured at scale. That is a long way from proven, but it is worth watching.

Prevention trials are testing whether treatment before symptoms begin can delay or prevent cognitive decline entirely. The AHEAD study is evaluating lecanemab in people with elevated amyloid but no symptoms. TRAILBLAZER-ALZ 3 is doing the same with donanemab. Results are years away. If they are positive, the entire paradigm shifts from treatment to prevention. That is the hope. It is not today’s reality.

The honest timeline: effective combination therapy that meaningfully halts or reverses Alzheimer’s disease is likely a decade or more away. Treatments that do more than modestly slow decline are not imminent. Planning for today’s realities is not pessimism. It is the only responsible approach.

What This Means at the Kitchen Table

For the family in the neurologist’s office, and for every family facing this decision, here is what matters:

If you or your loved one has been diagnosed with early-stage Alzheimer’s with confirmed amyloid pathology, you have options that did not exist three years ago. Those options are imperfect. They are also real. A treatment that modestly slows decline is a treatment that preserves time. Whether the benefits outweigh the costs and risks is a decision that belongs to you, not to anyone else.

Questions to ask your neurologist: What is my APOE status, and what does it mean for my risk? What does the treatment schedule look like, and can we manage the logistics? What are the realistic expectations for benefit in my specific case? What happens if ARIA develops? What are the signs I should watch for at home? How will we know if the treatment is working?

For people not yet at the treatment decision, the existence of these drugs strengthens the case for early detection. The treatments only work in early-stage disease. A diagnosis that comes late is a diagnosis that forecloses options. The blood biomarker tests discussed in the first installment of this series are relevant here: they can identify amyloid pathology earlier, more accessibly, and at lower cost than PET imaging. Early detection is not just about knowing. It is about preserving the option to act.

For people with moderate or advanced Alzheimer’s disease, these drugs are not currently an option. That is a hard truth, and it deserves to be said directly rather than softened. The treatments approved today do not work in later-stage disease. What is available: symptomatic medications (cholinesterase inhibitors, memantine) that may provide modest benefit, clinical trials testing new approaches, and care strategies that maximize quality of life and support for caregivers. The Alzheimer’s Association maintains a clinical trial finder at TrialMatch. The next installment in this series addresses what happens to the people providing care.

For everyone: the landscape is changing faster than at any point in the history of Alzheimer’s research. Drugs that did not exist five years ago are now treating patients. Drugs that do not exist today may be available in five years. The appropriate response is neither false hope nor despair. It is clear-eyed attention to what is real now and what may be coming.

The neurologist in the opening scene pauses before answering. Then she says something like this: “These drugs will not fix it. Alzheimer’s is not something we can fix yet. What they may do is slow it down. Buy time. Keep you closer to where you are now for longer than you would be otherwise. The benefits are real but modest. The risks are real but manageable for most people. The decision is yours. I can tell you what the evidence shows. I cannot tell you what the right choice is for your family.”

That pause, and that honesty, is what this audience deserves. One overpromise and you lose their trust forever. One honest conversation and you might keep it.