New theory of Alzheimer's explains drug failures

Brain plaques, long considered the chief killer of brain cells and the cause of Alzheimer's disease, may actually play a protective role under a new theory that is changing the way researchers think about the disease.

Instead of sticky plaques, free-floating bits of a toxic protein called amyloid beta may be what's killing off brain cells in Alzheimer's patients, U.S. researchers say.

If the theory is right, then drugs that target plaque, including bapineuzumab -- being developed by Pfizer, Johnson & Johnson and Elan -- may be aiming at the wrong target, they say.

"The plaque is not the main culprit in terms of toxicity," said Dr. Scott McGinnis of Harvard Medical School and Brigham and Women's Hospital in Boston, who treats Alzheimer's patients and runs clinical trials testing new Alzheimer's drugs.

For more than two decades, the prevailing plan of attack for researchers and drug companies has been to find a way to remove sticky clumps of a protein called amyloid beta from the brain.

But several recent studies in mice and rats now suggest that floating pieces of amyloid beta called oligomers are the real bad actors in Alzheimer's disease.

And instead of being the chief toxin, several teams suspect, the plaques may be the body's way of trapping and neutralizing oligomers.

"If you say Alzheimer's, everyone immediately thinks that it's the plaques that actually cause the disease. That couldn't be further from the truth," Andrew Dillin, of the Salk Institute in California and the Howard Hughes Medical Institute, told reporters in London this week at a conference on aging.

"The data actually suggest these plaques are a form of protection that the body tries to put on. So this is a sign that your brain was trying to do something very useful and helpful to you, and the remnant was the formation of amyloid plaques," Dillin said.

A Good Thing?

Adrian Ivinson, who directs the Harvard NeuroDiscovery Center in Boston, a drug discovery center affiliated with Harvard Medical School working on new Alzheimer's drugs, said scientists are beginning to think plaque is a good thing.

"It actually sequesters all of that amyloid," he said, adding that oligomers are "the really toxic substance."

In the latest study, a team led by Dr. Sam Gandy of the Alzheimer's Disease Research Center at Mount Sinai School of Medicine in New York genetically engineered mice that form only oligomers, but never brain plaques.

They found these mice developed the same level of memory and thinking problems as genetically engineered mice that get both plaques and oligomers.

And when the team added a gene that converted the oligomers to plaques, the mice got no worse.

"That suggests that plaques were not necessary and the addition of plaque did not make the oligomer-induced memory problems any worse," said Gandy, whose study appeared last month in the Annals of Neurology.

The findings may help explain the stunning failure of drugs designed to remove plaques from the brain of patients, which do little to improve thinking in Alzheimer's patients.

Alzheimer's is the most common form of dementia in which patients progressively lose their ability to think and care for themselves. Current drugs only treat symptoms.

Lack of Effects

Gandy points to a recent imaging study in Lancet Neurology looking at the drug bapineuzumab -- now in late-stage clinical trials.

The team used an imaging agent called Pittsburgh Compound B or PiB that can be used in brain scans to identify amyloid plaques. Using these scans in 28 patients, the team found that bapineuzumab shrank brain plaques by 25 percent, but Gandy said the drug had no effect on patients' ability to think and reason.

"We don't know whether bapineuzumab sees oligomers or not," Gandy said in a telephone interview.

And because PiB can only see amyloid deposits and not floating clumps of oligomers, there is no way to know whether the drug is having any effect.

Gandy said the Lancet Neurology study may simply mean that patients need to be treated longer to benefit from bapineuzumab. Or, it may mean that the drug -- an engineered immune-system molecule called a monoclonal antibody -- is targeting the wrong thing.

Bapineuzumab has had mixed results in a mid-stage clinical trial, and some researchers were encouraged by the Lancet Neurology study because it reduced plaque levels in patients.

But Dillin said the drug, like several others aimed at trying to stop plaques from forming, is destined to fail.

"This hypothesis is actually completely wrong, and we need a new way to start looking at this disease. This is actually not a viable therapeutic avenue," Dillin said.

Pfizer this month said results of its U.S. phase 3 trials would be released in mid-2012 and the European phase 3 trials would be done in 2014, a bit later than analysts had expected.

Many investors have already written off bapineuzumab, but since Alzheimer's afflicts 26 million people worldwide, any success could mean billions of dollars in revenue.