For decades, the "amyloid hypothesis" — the theory that clearing sticky beta-amyloid plaques from the brain is the key to treating Alzheimer's disease — has served as the organizing principle of pharmaceutical research into neurodegeneration. Billions of dollars and years of clinical trials have been staked on this single premise. Now, a new meta-analysis from Cochrane, a global authority on evidence-based medicine, has reached a blunt conclusion: the current class of amyloid-targeting antibody drugs does not work.
The Cochrane review, which synthesizes data across multiple clinical trials to assess the overall weight of evidence, dismissed the efficacy of these treatments in terms that left little room for ambiguity. The response from the research community was swift and sharp. Critics of the review argue that it minimizes the modest but measurable cognitive benefits observed in recent high-profile trials and that it applies an inappropriately rigid framework to a disease where any deceleration of decline carries meaning for patients and families.
A hypothesis under perpetual trial
The amyloid hypothesis has occupied a singular and contested position in neuroscience for more than three decades. Its logic is straightforward: beta-amyloid proteins aggregate into plaques in the brains of Alzheimer's patients, and removing those plaques should slow or halt cognitive deterioration. The hypothesis survived a long string of clinical failures through the 2010s, with each negative trial attributed to flawed drug design, wrong dosing, or intervention too late in the disease's progression rather than to a flaw in the underlying theory.
The landscape shifted when a newer generation of monoclonal antibodies — notably lecanemab and donanemab — demonstrated statistically significant, if modest, reductions in cognitive decline in large Phase III trials. Regulatory approvals followed, and the field treated the results as vindication. For proponents, these drugs represented the first credible evidence that targeting amyloid could alter the trajectory of the disease, even if the effect size remained small.
Cochrane's meta-analysis challenges that narrative directly. By pooling trial data and weighing clinical significance alongside statistical significance, the review concluded that the magnitude of cognitive benefit does not justify the risks associated with treatment — particularly amyloid-related imaging abnormalities (ARIA), a category of side effects that includes brain swelling and microbleeds. The distinction Cochrane draws is not trivial: a drug can produce a statistically detectable effect and still fail to deliver a benefit that patients or clinicians would recognize as meaningful.
The deeper fault line
The backlash against the review reveals a fault line that runs deeper than any single data set. Many researchers in the amyloid camp do not dispute that current effect sizes are small. Their argument is sequential: clearing amyloid is a necessary first step, not a sufficient one. In this framing, the modest results of lecanemab and donanemab are proof of concept rather than finished therapies, and combination approaches — targeting tau, neuroinflammation, or synaptic dysfunction alongside amyloid — may eventually yield larger benefits.
Cochrane's methodology, by design, evaluates what the evidence shows now, not what it might show in the future. That epistemological gap is where the tension concentrates. The review applies the standards of evidence-based medicine as they exist: measurable clinical benefit weighed against measurable harm. The research establishment, meanwhile, is asking for tolerance of uncertainty — a willingness to treat incomplete progress as a foundation rather than a failure.
The stakes extend well beyond academic debate. Regulatory agencies face pressure to reconcile these competing interpretations when making approval and reimbursement decisions. Health systems must decide whether to fund treatments whose benefits may be real but remain difficult to detect at the individual patient level. And patients navigating an Alzheimer's diagnosis confront a landscape where the most prominent therapies are simultaneously described as breakthroughs and as ineffective, depending on who is interpreting the data.
What the Cochrane review has not done is close the amyloid debate. What it has done is force a sharper articulation of the question at its center: how much clinical ambiguity is acceptable when the alternative is no treatment at all, and who gets to draw that line?
With reporting from Endpoints News.
Source · Endpoints News
