A new 2026 Cochrane Database of Systematic Reviews analysis delivers a decisive blow to the dominant amyloid hypothesis that has shaped Alzheimer's research for more than three decades. This high-quality systematic review and meta-analysis synthesized data from 17 randomized controlled trials enrolling 20,342 participants with mild cognitive impairment or mild dementia. It concludes that anti-amyloid monoclonal antibodies produce no clinically meaningful benefit on cognition or daily function, despite successfully clearing amyloid-beta plaques. Absolute changes on validated scales such as ADAS-Cog and CDR-SB fell well below established minimum clinically important difference thresholds, confirming that statistical signals in earlier trials never translated into patient-relevant outcomes.

The review also documents increased risks of amyloid-related imaging abnormalities (ARIA), including brain edema and microhemorrhages, occurring in a substantial proportion of treated patients. While many ARIA events were radiologically detected without immediate symptoms, inconsistent long-term symptom reporting across trials leaves their clinical significance uncertain. Lead author Francesco Nonino and senior author Edo Richard emphasize that continued investment in this pathway is unlikely to yield meaningful therapies.

Mainstream coverage, including the MedicalXpress summary, accurately reported the null clinical findings but missed critical context and historical patterns. It failed to highlight how the field repeatedly confused surrogate biomarker reduction with therapeutic success. This same gap plagued the 2021 FDA accelerated approval of aducanumab (Aduhelm), which relied on amyloid PET clearance despite one pivotal trial failing its primary clinical endpoint and an advisory committee voting overwhelmingly against approval. Coverage also underplayed the modest absolute effect sizes in subsequent drugs: lecanemab's 2023 CLARITY AD trial (NEJM; phase 3 RCT, n=1,795) produced only a 0.45-point difference on CDR-SB over 18 months against a background of continued decline, while donanemab's TRAILBLAZER-ALZ 2 (JAMA 2023; phase 3 RCT, n=1,736) showed similarly small slowing that still fell short of clinical meaningfulness for most patients.

The amyloid hypothesis, formulated in the early 1990s, funneled an estimated $20-30 billion in public and private funding into a single mechanistic pathway while sidelining alternative lines of inquiry. Multiple prior agents including solanezumab, bapineuzumab, and gantenerumab generated similar biomarker success paired with clinical failure, yet the pattern was dismissed as 'too little, too late' or 'wrong patients.' Industry-sponsored trials frequently featured authors with direct financial conflicts; independent syntheses like this Cochrane review, which minimizes such bias, are therefore especially valuable.

What others have missed is the opportunity cost. While resources poured into anti-amyloid monoclonals costing tens of thousands per patient annually, multimodal lifestyle intervention trials such as the FINGER study (Lancet 2015; RCT, n=1,260) demonstrated clinically relevant cognitive benefits through diet, exercise, cognitive training, and vascular risk management. Growing evidence also points to neuroinflammation, tau propagation, metabolic dysfunction, and cerebrovascular pathology as equally or more promising targets. The disconnect between plaque removal and clinical benefit now appears robust enough to reclassify Alzheimer's as a syndrome driven by multiple pathways rather than a single proteinopathy.

This Cochrane evidence should prompt regulators, funders, and journals to raise the bar for surrogate endpoints and demand patient-centered outcomes in neurodegenerative trials. Billions spent and years of research bandwidth lost cannot be recovered, but redirecting future efforts toward inflammation, vascular health, and prevention offers genuine hope for the millions facing this devastating disease.