A Cochrane review published around April 16, 2026 looked at 17 trials involving 20,342 people. The drugs being tested, including lecanemab and donanemab, did exactly what they were designed to do: they cleared amyloid plaques from the brain. And then, for most patients, nothing meaningful happened. Cognition barely moved. Daily function barely moved. The plaques were gone. The disease kept going.
Francesco Nonino, the lead author from the IRCCS Institute of Neurological Sciences in Bologna, put it plainly: "Unfortunately, the evidence suggests that these drugs make no meaningful difference to patients." Edo Richard, a neurology professor at Radboud University who sees Alzheimer's patients every week, said he wishes he had an effective treatment. He does not believe these drugs are it.
The drug companies pushed back fast. Eisai, which makes lecanemab, called the Cochrane analysis "scientifically deeply flawed" for mixing failed drugs with approved ones. The UK Dementia Research Institute made a fair point: lumping together drugs that successfully remove amyloid with ones that never did is a methodological problem. That criticism has merit. The Cochrane review does blur important distinctions.
But here is what the criticism does not change. Even the approved drugs, lecanemab and donanemab, showed only modest slowing of decline in their Phase 3 trials. Not reversal. Not stabilization. Slowing. And that slowing comes with brain swelling or bleeding, called ARIA, in somewhere between 10% and 20% of patients, requiring regular MRI monitoring. The drugs cost more than $26,000 per year. They work only in early-stage cases with confirmed amyloid. Most people with Alzheimer's do not qualify.
When Clearing the Plaque Is Not Enough
The amyloid hypothesis has driven Alzheimer's research for over 30 years. The idea is straightforward: amyloid plaques build up in the brain, they cause the disease, so clear the plaques and you treat the disease. Billions of dollars in US funding have followed that logic. The problem is that the logic keeps failing the test of patient outcomes.
BACE1 inhibitors, an earlier class of drugs, reduced amyloid aggressively in trials. Some patients got cognitively worse. Aducanumab got FDA approval in 2021 under controversial circumstances and was eventually pulled from the market. Now lecanemab and donanemab show modest effects in carefully selected early patients, and the Alzheimer's Association's real-world ALZ-NET data suggests those modest effects hold outside of trials. That is genuinely something. It is not nothing.
But modest effects in a narrow patient population, at $26,000 a year, with serious side effect risks, after 30 years of research and billions in investment, is a poor return. And the opportunity cost matters. Every dollar chasing amyloid is a dollar not going toward tau research, neuroinflammation, or neuroprotection approaches that the Cochrane authors specifically urged the field to pursue.
What Should Actually Change
If you have a family member in early-stage Alzheimer's with confirmed amyloid, talking to a neurologist about lecanemab or donanemab is reasonable. The real-world data is not nothing. But do not expect a turnaround. Expect, at best, a slower slide.
The bigger ask is for the researchers and funders. The National Institutes of Health, the Alzheimer's Association, and the pharmaceutical companies funding the next wave of trials need to seriously redirect toward non-amyloid targets. Not as a supplement to amyloid research. As the priority.
Thirty years is long enough to know when a hypothesis needs a rival.
