A new class of Alzheimer’s drugs promised to be the first treatments that actually slow the disease rather than just manage symptoms. Millions of families watched as the FDA approved them. But a rigorous Cochrane systematic review, published in early 2025 and drawing renewed attention this spring, delivers a sobering assessment: for patients with early-stage Alzheimer’s, the cognitive benefits of anti-amyloid monoclonal antibodies are, in absolute terms, “absent or trivial.”
The finding puts a sharp point on a question that patients, caregivers, and physicians have been wrestling with since lecanemab (marketed as Leqembi) and donanemab (marketed as Kisunla) reached the market. These drugs can clear amyloid plaques from the brain. But does that translate into a difference anyone can actually feel?
What the Cochrane review found
The Cochrane review examined randomized controlled trials lasting at least 12 months that compared anti-amyloid monoclonal antibodies against placebo in adults with mild cognitive impairment or mild dementia due to Alzheimer’s disease. This is the population most likely to be offered these drugs in a real clinic, not patients in later stages of the disease.
The results were blunt. On standard clinical scales like the CDR-SB (Clinical Dementia Rating-Sum of Boxes), the drugs produced statistically detectable differences from placebo, but those differences fell well short of what clinicians consider meaningful in a patient’s daily life. Amyloid plaques were reduced. Cognitive decline was not reduced in a way that patients or caregivers could notice.
A separate, independent systematic review and meta-analysis focused on minimal clinically important difference thresholds (MCID), which represent the smallest change a patient would recognize as meaningful, reached a strikingly similar conclusion. Across lecanemab, donanemab, and aducanumab, cognitive and functional changes failed to exceed those MCID thresholds in any of the pooled analyses. Meanwhile, the harms were not trivial at all. Amyloid-related imaging abnormalities, known as ARIA, occurred at significant rates: brain swelling (ARIA-E) affected roughly 12 to 24 percent of treated patients in major trials, and microbleeds (ARIA-H) were even more common.
Why the FDA approved them anyway
The disconnect between these reviews and the regulatory pathway is not a contradiction so much as a reflection of two different questions being asked. The FDA converted Leqembi to traditional approval in 2023, and donanemab received approval in 2024. In both cases, the agency weighed the severity of Alzheimer’s disease, the near-total absence of effective alternatives, and the statistical evidence of slowed decline on clinical scales.
That framework is designed to answer a regulatory question: does the drug’s benefit outweigh its risk enough to justify market access? The Cochrane reviewers asked a different question: does the benefit cross the line into territory that patients can actually feel? A drug can clear both hurdles, or it can clear one and not the other. In this case, the independent evidence suggests these treatments meet the regulatory bar while falling short of the clinical meaningfulness bar.
The FDA’s approval of Leqembi did include a requirement for ApoE e4 genetic testing before treatment, acknowledging that carriers of this gene variant face substantially higher risk of ARIA. That safety measure signals the agency’s own recognition that these drugs carry real dangers for certain patients.
What patients and families should know
Several practical realities shape the decision facing anyone considering these treatments in 2026.
The benefit is small. On the CDR-SB scale, lecanemab slowed decline by about 0.45 points over 18 months compared to placebo in its pivotal trial. The minimal clinically important difference on that scale is generally considered to be 1.0 to 2.0 points. The drug’s effect, while real in a statistical sense, sits below the threshold most experts use to define a change a patient would notice.
The risks are not small. ARIA requires monitoring with repeated MRI scans. Brain swelling can be asymptomatic or can cause headaches, confusion, and in rare cases, serious complications. Patients on blood thinners or those who carry two copies of the ApoE e4 gene face elevated risk. Three deaths linked to brain bleeding occurred during the lecanemab trial, though the precise causal relationship remains debated.
The cost is significant. Leqembi carries a list price of roughly $26,500 per year, not including the cost of infusions, MRI monitoring, and genetic testing. The Centers for Medicare and Medicaid Services (CMS) agreed to cover the drug under certain conditions, including participation in a registry, but out-of-pocket costs for patients vary widely depending on insurance.
Long-term data remain thin. The pivotal trials ran 18 months. Open-label extension studies have provided some longer-term follow-up, but without a placebo comparison group, those results are harder to interpret. Whether the modest early benefit grows, stabilizes, or fades over years of treatment is genuinely unknown.
Where Alzheimer’s research goes from here
The modest results from amyloid-targeting drugs have not killed the amyloid hypothesis, but they have forced a reckoning with its limits. Clearing plaques from the brain appears to be, at best, a necessary but insufficient piece of a far more complex disease. Researchers are increasingly focused on combination approaches that target additional pathways: tau protein tangles, neuroinflammation, vascular damage, and metabolic dysfunction.
Several combination trials are now underway or in planning stages, pairing anti-amyloid antibodies with drugs that address these other mechanisms. Whether stacking interventions will produce the kind of meaningful cognitive preservation that amyloid clearance alone has not delivered is the central open question in the field.
For now, the evidence reviewed by two independent groups of researchers points in the same direction. These drugs do what they were designed to do biologically. They remove amyloid. But removing amyloid has not yet translated into the kind of benefit that changes a patient’s life. Families weighing treatment should have that honest conversation with their physicians, including a clear discussion of ApoE e4 status, ARIA risk, monitoring requirements, and what “slowing decline” actually means in concrete, everyday terms.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
