A sweeping new Cochrane systematic review has reignited the long-simmering 'amyloid wars,' delivering a rigorous evidence synthesis that calls into question the scientific foundation for recently approved Alzheimer's monoclonal antibodies including lecanemab (Leqembi), aducanumab (Aduhelm), and donanemab. While the STAT newsletter item frames this as another round of debate, it underplays the review's devastating implications for a 30-year therapeutic paradigm that has consumed an estimated $42 billion in R&D spending with strikingly little clinical return.

The Cochrane review aggregates data from multiple phase 3 RCTs totaling over 12,000 participants. These high-quality, double-blind, placebo-controlled trials consistently show that anti-amyloid antibodies achieve their surrogate endpoint—substantial plaque reduction on PET imaging—yet deliver only modest slowing of cognitive decline. In the landmark CLARITY AD trial (n=1,795, industry-funded RCT with multiple authors holding equity in Eisai/Biogen), lecanemab produced a 0.45-point difference on the 18-point CDR-SB scale over 18 months. This equates to roughly a 5–7 month delay in progression, an effect size many independent neurologists deem below the threshold of minimal clinically important difference. ARIA-related brain swelling and microhemorrhages occurred in 12–21% of treated patients, with higher rates in APOE4 carriers.

Mainstream coverage, including the referenced STAT piece, largely misses three critical dimensions. First, the foundational science itself is shaky. A landmark 2022 Science investigation by Charles Piller exposed apparent data manipulation in key 2006–2008 papers from the University of Minnesota that helped cement the amyloid cascade hypothesis; subsequent replication attempts have faltered. Second, regulatory capture is evident: the FDA's 2021 accelerated approval of aducanumab overruled its own advisory committee's 10–0 near-unanimous rejection, later prompting congressional scrutiny and resignations. Third, the opportunity costs are staggering. While anti-amyloid programs dominated NIH and venture funding, multifactorial approaches received short shrift despite stronger signals.

A 2023 Lancet Neurology editorial and the FINGER trial (n=1,260, multicenter RCT) demonstrated that multidomain lifestyle interventions targeting vascular risk, diet, exercise, and cognitive training produced effect sizes on cognition roughly double those of lecanemab, with no serious adverse events. Large observational cohorts such as the UK Biobank (n>500,000) further link midlife hypertension, insulin resistance, and chronic inflammation to Alzheimer's risk far more robustly than amyloid burden alone. No major conflicts of interest were declared in the Cochrane authors, contrasting sharply with the financial entanglements pervasive in the original drug trials.

The pattern is now unmistakable: repeated failure of anti-amyloid agents (bapineuzumab, solanezumab, gantenerumab) followed by incremental adjustments to trial design, endpoint selection, and regulatory thresholds rather than genuine scientific reevaluation. The Cochrane findings expose how surrogate endpoints decoupled from meaningful patient benefit became acceptable currency, allowing high-priced therapies ($26,500–$56,000 annually) to reach market despite questionable real-world impact.

This moment demands more than another inconclusive trial. It requires redirecting resources toward neuroinflammation, metabolic dysfunction, tau propagation, and vascular health—domains with converging evidence from both RCTs and robust observational data. The amyloid hypothesis did not fail because the drugs were imperfect; it failed because the underlying causal model was oversimplified. Until mainstream journalism moves beyond press-release cycles to interrogate these deeper structural and scientific failures, patients will continue to be offered marginal therapies at enormous societal cost while more promising avenues remain underfunded.