For the roughly 6.9 million Americans living with Alzheimer’s disease, the arrival of drugs that target amyloid plaques in the brain was supposed to mark a turning point. After decades of failed trials, lecanemab and donanemab offered something new: treatments that actually clear the sticky protein deposits long thought to drive the illness. But a Cochrane systematic review (reference CD016297), led by Cochrane Dementia and Cognitive Improvement Group authors including first author Mandy Ryan, published in early 2025, has landed on a conclusion that many families will find difficult to hear. The benefits of these anti-amyloid monoclonal antibodies, while real, are limited, and the risks that come with them are not trivial.
What the strongest trial data actually show
The centerpiece of the evidence for lecanemab is the CLARITY-AD trial, a phase 3 randomized controlled study published in the New England Journal of Medicine in early 2023. Researchers enrolled nearly 1,800 people with early-stage Alzheimer’s, confirmed by amyloid PET scans or spinal fluid tests, and tracked them for 18 months. The primary measure was the Clinical Dementia Rating-Sum of Boxes (CDR-SB), a clinician-scored tool that rates how well a person manages memory, orientation, judgment, and daily tasks on an 18-point scale.
Patients on lecanemab declined by an average of 1.21 points. Those on placebo declined by 1.66 points. The difference: 0.45 points over a year and a half.
That gap was statistically significant. But on an 18-point scale, 0.45 points is small enough that most patients and caregivers would struggle to notice it in the rhythm of daily life. It is not the difference between a parent who remembers a grandchild’s name and one who does not. It is closer to a subtle shift on a clinical checklist, one that a trained rater can detect but that rarely shows up as a clear change at the dinner table.
The Cochrane review, applying standardized methods for assessing bias and certainty of evidence across multiple anti-amyloid antibodies, reached a pointed conclusion: these drugs produce a small slowing of cognitive and functional decline, but the clinical importance of that slowing remains uncertain. Cochrane reviews are widely considered the gold standard for evidence synthesis in medicine, and this one’s verdict carries weight in both clinical practice and policy debates.
The safety trade-off
Even a modest benefit might be worth pursuing if the treatment were low-risk. These drugs are not. The most serious concern is a set of side effects known as amyloid-related imaging abnormalities, or ARIA. These include brain swelling (ARIA-E) and small brain bleeds (ARIA-H), detected on routine MRI monitoring that patients must undergo throughout treatment.
In the CLARITY-AD trial, roughly 21% of lecanemab patients developed ARIA-E, compared with about 9% on placebo. Most cases were mild or asymptomatic, but some caused headaches, confusion, or visual disturbances. The risk is sharply higher for carriers of the APOE4 gene variant, particularly those who carry two copies. The FDA’s prescribing information for lecanemab includes warnings about ARIA and recommends genetic testing before treatment, a step that adds complexity and cost to the decision.
Donanemab, the competing anti-amyloid antibody from Eli Lilly that the Cochrane review also examined, showed a similar pattern: a modest CDR-SB benefit paired with elevated ARIA rates. Across the drug class, the review found that the safety signals are consistent enough to demand careful patient selection and ongoing monitoring.
Medicare coverage remains conditional
The U.S. government’s approach to paying for these drugs reflects the uncertainty in the evidence. The Centers for Medicare and Medicaid Services issued a national coverage determination (tracking number CAG-00460N) that ties reimbursement for anti-amyloid antibodies to a framework called coverage with evidence development. In practice, this means Medicare will pay for the drugs only when patients are enrolled in qualifying clinical studies or registries designed to generate real-world data on outcomes.
The policy creates a two-track system. Patients treated at academic medical centers or large health systems with active registries can access the drugs through Medicare. Those in rural areas, community clinics, or settings without registry infrastructure may face significant barriers. People with more advanced dementia or multiple serious health conditions, groups that were excluded from the pivotal trials, generally do not qualify at all.
As of spring 2026, results from the CMS-mandated registries have not been published. The data that could clarify whether the modest trial benefits hold up in everyday clinical practice, across a more diverse patient population, are still being collected.
The questions that remain open
Several gaps in the evidence are large enough to shape the entire debate.
Duration of benefit. The CLARITY-AD trial lasted 18 months. Eisai, the drugmaker, has presented open-label extension data at medical conferences suggesting continued separation between treated and untreated groups, but open-label extensions lack the rigor of randomized comparisons. No completed long-term randomized data are publicly available. Whether the 0.45-point advantage grows over three or five years, or whether it plateaus or narrows, is unknown.
Patient-perceived quality of life. CDR-SB is scored by clinicians, not patients. No large-scale registry data have yet confirmed whether people on these drugs, or their caregivers, perceive a tangible difference in independence, daily functioning, or quality of life. This is the question that matters most to families, and it does not yet have a clear answer.
Cost-effectiveness. Lecanemab carries an annual list price of approximately $26,500. The Institute for Clinical and Economic Review, an independent U.S. body that evaluates drug value, concluded in its 2023 assessment that the drug’s price would need to be substantially lower to align with the clinical benefit it delivers. When infusion costs, serial MRI monitoring, genetic testing, and specialist visits are factored in, the total per-patient expense climbs well beyond the sticker price. For a Medicare system already under financial strain, the question of value for money is not academic.
What this means for families weighing treatment
None of this evidence says these drugs are worthless. What it says is that they are not the breakthroughs that early headlines suggested. Anti-amyloid antibodies like lecanemab appear to slow cognitive decline modestly in carefully selected people with early Alzheimer’s disease. They do not halt the illness. They do not reverse it. And they come with a treatment burden, biweekly intravenous infusions, regular MRI scans, and the possibility of brain swelling or bleeding, that is far from negligible.
For clinicians, the Cochrane review reinforces the importance of shared decision-making. Patients and families deserve a frank conversation about the size of the expected benefit, the risks of ARIA, and the practical demands of treatment, before a first infusion is scheduled. For policymakers, the conditional coverage framework that CMS has put in place looks well-calibrated to the current state of the science: it allows access while insisting on the generation of better data.
For the millions of families touched by Alzheimer’s, the honest message is that science has taken a genuine but small step forward. The drugs work, in a narrow, measurable sense. Whether that narrow effect adds up to more good days, more moments of recognition, more months of independence, is something the data cannot yet confirm. Until they can, these treatments are best understood as incremental tools in a fight that still lacks a decisive weapon.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
