A landmark Cochrane systematic review published in 2026 delivers a sobering verdict on the class of anti-amyloid monoclonal antibodies once promoted as transformative for Alzheimer's disease. Analyzing 17 randomized controlled trials involving more than 20,000 participants with mild cognitive impairment or early dementia, the high-quality meta-analysis found that while the drugs—including lecanemab and donanemab—robustly cleared amyloid plaques on PET imaging, they produced no clinically meaningful improvements in cognition, function, or quality of life over 18 months. This directly challenges the amyloid cascade hypothesis that has dominated research funding and drug development for three decades.

The Cochrane review stands as the most rigorous synthesis to date, prioritizing patient-centered clinical endpoints over surrogate biomarkers. Individual trials such as the 2023 NEJM CLARITY AD study (n=1,795, RCT) for lecanemab reported a 27% relative slowing of decline on CDR-SB, yet the absolute difference was approximately 0.45 points on an 18-point scale—widely viewed by independent experts as below the threshold for noticeable patient benefit. Similarly, the TRAILBLAZER-ALZ 2 trial for donanemab showed marginal gains accompanied by high rates of amyloid-related imaging abnormalities (ARIA), with brain swelling or microhemorrhages affecting 24-36% of participants, particularly APOE4 carriers. A third source, a 2024 critical review by Richard and colleagues in Alzheimer's & Dementia, anticipated these findings by highlighting how prior anti-amyloid failures (bapineuzumab, solanezumab) followed the same pattern: biomarker success without clinical translation.

Mainstream coverage, including the source article, largely missed the deeper historical pattern and opportunity costs. Since John Hardy's seminal 1992 Science paper proposing the amyloid hypothesis, over $30 billion in R&D has flowed into this single target, crowding out alternative mechanisms. What coverage often got wrong was framing statistical significance on surrogate markers as therapeutic victory while downplaying conflicts of interest—Hardy himself has consulted for Eli Lilly, Biogen, and Eisai, manufacturers of the featured drugs. The review's decision to pool data is defensible precisely because all agents share the same primary target: amyloid-beta proteins, regardless of exact binding differences.

This episode reveals a systemic flaw in neurodegenerative drug development: the over-reliance on biomarkers that may reflect epiphenomena rather than causal drivers in late-onset sporadic Alzheimer's. Amyloid accumulation is a feature of normal aging as well as disease; genetic evidence strongly supports its role in rare early-onset forms, but observational studies and failed trials increasingly point to neuroinflammation, vascular dysfunction, insulin resistance, and tau propagation as more proximal levers of cognitive decline. The FINGER trial (a 2-year multi-domain lifestyle RCT, n=1,260) demonstrated clinically relevant slowing through exercise, diet, and cognitive training—benefits larger than those from anti-amyloid agents yet receiving far less attention.

UK and French health authorities have correctly declined routine coverage given the marginal efficacy, high cost ($26,000+ per year), and serious safety risks. The persistent hype around plaque clearance has delayed progress on multifactorial and preventive strategies. As co-author Edo Richard stated, future efforts must target other pathways. For patients and families, the gap between biomarker headlines and lived experience has never been clearer: removing amyloid is not enough. Alzheimer's research now stands at an inflection point where genuine analysis must replace reflexive allegiance to a单一 hypothesis that, after 30 years and billions spent, has not delivered meaningful relief.