Anti-amyloid drugs are a class of therapies designed to remove sticky protein clumps called amyloid-beta plaques from the brains of people with Alzheimer’s disease. The first of these drugs, lecanemab, received full FDA approval in 2023, followed by donanemab in 2024, making them the first treatments to meaningfully slow the biological progression of the disease rather than just managing symptoms. In clinical trials, these antibody-based drugs slowed cognitive decline by roughly 30% over a year and a half, a real but modest effect that has sparked both hope and heated debate about what counts as clinically meaningful improvement.
How Amyloid Plaques Build Up in the Brain
The brain naturally produces a protein called amyloid precursor protein (APP), which sits in the membranes of brain cells. APP gets chopped up by a series of enzymes. When it’s cut by one set of enzymes (beta-secretase followed by gamma-secretase), the result is a small fragment called amyloid-beta. In healthy brains, amyloid-beta is produced in small amounts and cleared away. In Alzheimer’s disease, this clearance system breaks down, and amyloid-beta fragments stick together into clumps that grow from small soluble clusters (called oligomers and protofibrils) into large insoluble plaques that accumulate between neurons.
1PubMed Central. Amyloid precursor protein processing and Alzheimer’s diseaseThere’s actually a protective pathway too. When a different enzyme (alpha-secretase) cuts APP first, it prevents amyloid-beta from being formed at all and instead generates a fragment called sAPPalpha, which appears to support neuron health.
2PubMed Central. Regulation of β cleavage of amyloid precursor protein The balance between these two pathways is part of what goes wrong in Alzheimer’s, and it’s this imbalance that anti-amyloid drugs aim to address, albeit from the downstream end by clearing amyloid after it has already formed.
How the Approved Drugs Work
Lecanemab, donanemab, and the earlier (and now withdrawn) aducanumab are all monoclonal antibodies. They’re lab-designed proteins that bind to amyloid-beta in the brain and flag it for removal by the immune system. But they don’t all target the same form of amyloid. Lecanemab was designed to preferentially bind protofibrils, the intermediate-size clumps thought to be especially toxic. Donanemab targets a modified form of amyloid-beta found in established plaques. Aducanumab was designed to target both soluble and insoluble forms of amyloid.
These distinctions sound important in theory, but laboratory evidence complicates the picture. A recent study measured how strongly lecanemab and aducanumab bind to amyloid aggregates extracted from actual human Alzheimer’s brains and found no statistically significant difference in their preference for soluble versus insoluble amyloid forms. The researchers concluded that lecanemab is unlikely to have a meaningfully greater preference for protofibrils than aducanumab does, at least in the context of real brain tissue rather than synthetic preparations.
3PubMed Central. Anti‐amyloid antibody equilibrium binding to Aβ aggregates from human Alzheimer’s disease brain This finding suggests that the clinical differences among these drugs may depend less on which exact form of amyloid they prefer and more on factors like dosing, how quickly amyloid gets cleared, and how the immune response plays out in each patient.
What the Clinical Trials Actually Showed
The pivotal trial for lecanemab (called CLARITY-AD) enrolled people with early Alzheimer’s, meaning mild cognitive impairment or mild dementia with confirmed amyloid on brain scans. After 18 months, patients on lecanemab showed a decline of 1.21 points on a key clinical rating scale (CDR-SB), compared to 1.66 points in the placebo group, a difference of about 0.45 points. On amyloid PET scans, lecanemab reduced brain amyloid by about 55 centiloids, while the placebo group actually gained a few centiloids, reflecting the natural accumulation of the disease.
4PubMed Central. Anti-Amyloid Therapies for Alzheimer’s Disease: Progress, Pitfalls, and the Path AheadA meta-analysis pooling results across lecanemab trials confirmed that the drug slowed cognitive decline across multiple measurement tools and significantly reduced amyloid burden on brain imaging.
5PubMed. Monoclonal therapy with lecanemab in the treatment of mild Alzheimer’s disease: A systematic review and meta-analysis There’s also striking autopsy evidence from one patient treated with aducanumab. Brain regions where amyloid had been successfully cleared showed lower levels of tau (the other protein involved in Alzheimer’s) and less brain shrinkage than regions where amyloid remained, suggesting that removing amyloid can protect downstream brain structures.
6JAMA. Clinicopathologic Evaluation of Amyloid Clearance in Alzheimer DiseaseThe “Clinically Meaningful” Debate
The numbers above are where the controversy lives. A 0.45-point difference on the CDR-SB scale over 18 months is statistically significant, but whether patients or their families can actually perceive that difference in daily life is a separate question. A rapid review of what constitutes a “minimal clinically important difference” on the CDR-SB found that for people with mild cognitive impairment, a change of about 1 point is considered meaningful, and for those with mild Alzheimer’s, the threshold is about 2 points.
7PubMed Central. Minimal clinically important difference in Alzheimer’s disease: Rapid review By that yardstick, the treatment benefit from lecanemab at 18 months falls short of what clinicians typically consider noticeable to patients.
Proponents argue that the benefit likely grows over longer treatment durations since the drug is slowing an ongoing process, and that the 18-month snapshot underestimates the lifetime value of the treatment. Critics counter that extrapolating from short trials is speculative, particularly given the safety risks. This tension is genuine, and it hasn’t been resolved. The field is essentially betting that clearing amyloid early enough and long enough will produce meaningful clinical benefit, but the definitive long-term data don’t exist yet.
ARIA and Other Safety Risks
The most serious side effect of anti-amyloid antibodies goes by the name amyloid-related imaging abnormalities, or ARIA. There are two forms: ARIA-E, which involves brain swelling (edema), and ARIA-H, which involves tiny bleeds (microhemorrhages) in the brain. Most cases are mild and resolve on their own, but severe cases have led to hospitalizations and, in rare instances, deaths during clinical trials and early real-world use.
The mechanism behind ARIA appears to involve the drug’s interaction with blood vessels in the brain. In Alzheimer’s, amyloid often accumulates not just in brain tissue but also in the walls of small blood vessels, a condition called cerebral amyloid angiopathy (CAA). When antibodies clear amyloid from these vessel walls, they can temporarily increase vascular vulnerability. The immune response against amyloid near blood vessels increases permeability, allowing fluid and blood products to leak through damaged vessel walls.
8PubMed Central. Amyloid-related imaging abnormalities (ARIA): radiological, biological and clinical characteristics In other words, the drug’s core mechanism of action and its primary side effect are two sides of the same coin: clearing amyloid from brain vasculature disrupts the vessels in the process.
9PubMed Central. Amyloid-related imaging abnormalities (ARIA) in Alzheimer’s disease: from pathophysiology to individualized risk assessmentWho Faces the Highest Risk
The single strongest predictor of ARIA risk is a person’s APOE genotype. The APOE4 variant, which is already the most common genetic risk factor for Alzheimer’s itself, substantially increases the risk of developing ARIA during treatment. People who carry two copies of APOE4 (homozygous carriers) face the highest risk. A case report documenting ARIA in a woman with APOE4 homozygosity treated with lecanemab emphasized that prescribing decisions for this group require especially careful consideration.
10Clinical and Experimental Neuroimmunology. Amyloid‐related imaging abnormalities in a woman with apolipoprotein E ε4 homozygotes treated with lecanemab for Alzheimer’s diseaseBecause of this risk, APOE genotyping before starting treatment and regular brain MRI monitoring during treatment are now standard recommendations.
11PubMed Central. Lecanemab: Appropriate Use Recommendations by Korean Dementia Association Patients on blood thinners face additional risk, and those with significant pre-existing CAA on imaging may be poor candidates. This creates a paradox: the people at highest genetic risk for Alzheimer’s are also the ones who face the greatest danger from the treatment.
Blood Tests Are Changing Who Gets Diagnosed
Until recently, confirming amyloid in the brain required either a PET scan (expensive, limited availability, involves radiation) or a spinal tap to measure cerebrospinal fluid. This bottleneck meant that many potential candidates for anti-amyloid therapy were never properly evaluated. Blood-based biomarkers are now changing this landscape. Among these, a blood test measuring phosphorylated tau-217 (p-tau217) has emerged as a particularly reliable indicator of brain amyloid buildup.
12PubMed Central. P-tau217 as a Reliable Blood-Based Marker of Alzheimer’s DiseaseResearch has shown that combining p-tau217 with another blood marker (the ratio of amyloid-beta 42 to 40) can detect abnormal brain amyloid status with very high accuracy, even in people who don’t yet have cognitive symptoms. These plasma markers also predict future amyloid accumulation on PET scans, suggesting they could eventually identify candidates for treatment before symptoms appear.
13JAMA Neurology. Plasma Phosphorylated Tau 217 and Aβ42/40 to Predict Early Brain Aβ Accumulation in People Without Cognitive Impairment That said, about 10-20% of patients show discordant results between PET scans and fluid biomarkers, meaning the two methods occasionally disagree about whether someone has pathological amyloid levels.
14Wiley Online Library (Alzheimer’s & Dementia). Considerations in the clinical use of amyloid PET and CSF biomarkers for Alzheimer’s disease When that happens, clinicians have to use judgment and sometimes additional testing to determine whether treatment is appropriate.
Why Two Decades of Failure Preceded These Approvals
The recent approvals might make it seem like the field suddenly cracked the problem, but anti-amyloid drug development has a brutal track record. For roughly 20 years before lecanemab’s success, nearly every amyloid-targeting drug that entered large clinical trials failed. Reviews of this track record have identified several recurring problems: many trials enrolled patients whose disease was too advanced for amyloid removal to make a difference, some drugs didn’t reach sufficient doses to meaningfully clear amyloid, and the field’s understanding of Alzheimer’s pathology was too simplistic to account for the disease’s complexity.
15PubMed Central. Reasons for Failed Trials of Disease-Modifying Treatments for Alzheimer Disease and Their Contribution in Recent ResearchThere were also methodological shortcomings. An analysis of the clinical trial failures found that rational drug development principles were not always followed, raising the question of whether some failed compounds might have worked if tested more carefully.
16PubMed Central. Alzheimer’s Disease: Key Insights from Two Decades of Clinical Trial Failures The shift to treating earlier-stage patients, requiring biomarker confirmation of amyloid, and using more sensitive clinical endpoints appears to be what finally made the difference. Whether this vindicates the amyloid hypothesis or simply shows that removing amyloid provides a small benefit in a disease driven by many factors is still being argued.
Getting the Drug Into the Brain More Efficiently
One of the underappreciated challenges with antibody therapies is that very little of the injected drug actually reaches the brain. The blood-brain barrier, the tight lining of cells that protects the brain from substances circulating in the bloodstream, blocks most large molecules. Antibodies are large molecules, and only a tiny fraction of each infusion crosses into brain tissue. This means patients need high systemic doses and frequent infusions (lecanemab requires biweekly intravenous infusions), which contributes to both side effects and cost.
Researchers are working on “brain shuttle” technology to overcome this limitation. One approach fuses the anti-amyloid antibody to a fragment that binds the transferrin receptor, a protein that naturally transports molecules across the blood-brain barrier. An early version of this technology achieved brain concentrations roughly 80-fold higher than an unmodified antibody, reaching levels comparable to small-molecule drugs.
17PubMed Central. Bivalent Brain Shuttle Increases Antibody Uptake by Monovalent Binding to the Transferrin Receptor More recent work showed that an aducanumab variant equipped with a transferrin receptor shuttle achieved comparable amyloid plaque reduction at one-fifth the dose, with more even distribution across brain regions and less problematic concentration around blood vessels.
18PubMed Central. Transferrin receptor-binding blood-brain barrier shuttle enhances brain delivery and plaque-clearing efficacy of a therapeutic anti-Aβ antibody If this technology translates to humans, it could make anti-amyloid treatment both more effective and safer by reducing the high systemic doses that contribute to ARIA.
Small Molecules as an Alternative to Antibodies
Antibodies aren’t the only way to target amyloid. Researchers have long explored small-molecule compounds that could prevent amyloid-beta fragments from clumping together in the first place, rather than clearing plaques after they’ve formed.
19PubMed Central. Small molecule inhibitors of amyloid β peptide aggregation as a potential therapeutic strategy for Alzheimer’s disease This approach has several theoretical advantages: small molecules cross the blood-brain barrier more easily than antibodies, could be taken as pills rather than infusions, and would likely be far cheaper to manufacture.
The challenge is that amyloid-beta aggregation involves multiple microscopic steps, and disrupting the wrong step can sometimes make things worse by producing more toxic intermediate-size clumps. One research group has developed a strategy using chemical kinetics to identify compounds that target specific steps in the aggregation cascade, zeroing in on molecules that inhibit the earliest stages of clumping without increasing the production of toxic intermediates.
20PubMed Central. Systematic development of small molecules to inhibit specific microscopic steps of Aβ42 aggregation in Alzheimer’s disease Computational screening methods are also being applied to accelerate the identification of promising candidates.
21PubMed. In Silico Screening of Small Molecule Inhibitors for Amyloid-β Aggregation None of these small molecules are in late-stage clinical trials yet, but they represent a fundamentally different path to targeting amyloid that could eventually complement or even replace antibody infusions.
Why Combination Therapy May Be Necessary
The roughly 30% slowing of disease progression that anti-amyloid antibodies achieve on their own has pushed the field toward a conclusion that most cancer researchers reached decades ago: complex diseases rarely yield to a single drug targeting a single pathway.
22PubMed Central. Alzheimer Combination Therapies: Overview and Scenarios Alzheimer’s involves not just amyloid but also abnormal tau protein, chronic brain inflammation, synaptic loss, and often co-occurring pathologies like alpha-synuclein deposits and TDP-43 accumulations.
An international task force convened in late 2024 agreed that combination therapies are both necessary and urgent. As of early 2024, about 13% of all Alzheimer’s clinical trials were testing combination approaches. The combination of anti-amyloid and anti-tau therapies has emerged as a particularly promising focus, since amyloid and tau appear to drive neurodegeneration through connected but distinct mechanisms. Combinations targeting inflammation alongside amyloid removal are also being explored.
23PubMed Central. Challenges and opportunities for novel combination therapies in Alzheimer’s disease: a report from the EU/US CTAD Task Force The hope is that attacking multiple pathways simultaneously will push that 30% benefit figure substantially higher.
The Practical Burden of Treatment
Even setting aside the scientific questions, delivering anti-amyloid therapy in the real world is daunting. Lecanemab requires intravenous infusions every two weeks, ongoing MRI monitoring for ARIA, confirmation of amyloid status through PET scans or biomarker testing, APOE genotyping, and coordination among neurologists, radiologists, infusion centers, and sometimes cardiologists for patients on blood thinners. Building the infrastructure to support all of this has required healthcare systems to essentially create new clinical programs from scratch.
24PubMed Central. A Framework for the Administration of Anti-amyloid Monoclonal Antibody Treatments in Early-Stage Alzheimer’s DiseaseReports from healthcare systems that have implemented lecanemab programs describe the challenge in blunt terms: it requires dedicated project management, custom electronic medical record tools, close cooperation among specialty services, and careful patient eligibility screening.
25PubMed. Developing Treatment Models for the Delivery of the Antiamyloid Therapy, Lecanemab: Considerations for Implementation of Lecanemab in Healthcare Systems For patients, the treatment commitment is significant: regular trips to infusion centers, multiple MRIs, blood draws, and the anxiety of monitoring for a potentially dangerous side effect. Subcutaneous formulations that patients could potentially self-administer at home are in development and could ease this burden considerably, but they are not yet available for the currently approved drugs.
Immunogenicity and Long-Term Questions
As with any biologic therapy, the body can develop antibodies against the drug itself, known as anti-drug antibodies (ADA). In lecanemab’s clinical program, about 41% of patients in the main trial developed detectable ADA, though the levels were very low and detailed analysis showed no meaningful impact on how the drug worked, how much amyloid was cleared, or safety outcomes.
26Alzheimer’s & Dementia. Lecanemab (BAN2401) Infusion Reactions and Immunogenicity: Results from Randomized Phase 2 Study and an Open‐Label Extension (OLE) This is reassuring for now, but the data cover treatment periods of roughly two to three years. Whether immunogenicity becomes a problem over longer treatment periods, potentially spanning a decade or more for patients starting early, remains unknown.
Other unanswered long-term questions include how long treatment needs to continue once amyloid is cleared, whether amyloid will reaccumulate after stopping, and whether the structural brain changes associated with prolonged ARIA have consequences that only emerge years later. These are not abstract concerns. If anti-amyloid drugs are prescribed to people with mild cognitive impairment in their sixties or seventies, treatment courses of five to ten years are plausible. The safety and efficacy data supporting that duration simply don’t exist yet.