Autopsies on seven patients who died of Alzheimer's during the study showed that nearly all of the sticky beta-amyloid protein thought to be dangerous had been removed. But all patients still had severe dementia.
"It may be that these toxic plaques trigger the neurodegeneration, but don't have an ongoing role," said Clive Holmes of the University of Southampton, lead author, in a press statement. The study was published Friday in the medical journal, The Lancet.
The study was paid for by the Alzheimer's Research Trust, a British charity.
Alzheimer's disease is the most common cause of dementia and affects about 25 million people worldwide.
Other experts said that the study's findings pointed to a major gap in our understanding of the disease. Doctors have never been sure whether the brain plaques are the cause of Alzheimer's disease or just a side effect.
"We still don't have enough understanding of what we should target," said Dr. Bengt Winblad, director of the Alzheimer's Centre at Sweden's Karolinska Institute. Winblad was not connected to the study.
Aside from the plaque build-up, scientists also think that tangles of another brain protein called tau play a major role in Alzheimer's. Because those tangles form later than the plaque, some experts think they should be the focus instead.
"It may be harder to get a response from targeting plaque because that forms years before people actually have Alzheimer's," said Dr. Simon Lovestone, professor of Old Age Psychiatry at King's College in London. "By the time you do something, it may be too late."
Winblad said there was a better connection between brain tangles and Alzheimer's symptoms, but that no studies so far had looked at whether removing tangles might improve or even reverse Alzheimer's disease in patients.
Still, experts say that attacking toxic plaque in the brain shouldn't be abandoned just yet, since the formation of such plaques might be what sparks Alzheimer's disease in the first place.
"Removal of the initial motor for the disease might slow progression," wrote Peter H. St. George-Hyslop and John C. Morris of the University of Cambridge and the University of Toronto in an accompanying commentary in the Lancet.
