August 16, 2010

The Alzheimer's Revolution?

First a seemingly arcane debate at a medical conference about the criteria for diagnosing Alzheimer’s disease made news; then an article in a neurology journal was widely touted as showing that a spinal fluid test was “100% accurate in predicting Alzheimer’s.”


Are we on the cusp of a revolution? Is there truly a new way to think about this formidable disease—and is a cure finally at hand?

What we know now is that sometime in their mid- or late-fifties, individuals who are destined to develop Alzheimer’s disease begin accumulating a protein called amyloid-beta in their brains. This material clumps together to form plaques, the very same plaques seen by Alois Alzheimer under the microscope back in 1905, when he established early criteria for the disease. In the past, such plaques could only be identified by brain biopsy. Today, their presence can be inferred from “biomarkers” such as a low level of amyloid-beta in the cerebrospinal fluid (CSF) or specific changes on a PET or MRI scan.

The proposed new diagnostic criteria for Alzheimer’s disease reflect these biochemical findings. They do not require that a person have a special scan or a lumbar puncture to make the diagnosis. In fact, they define “probable Alzheimer’s disease dementia” based on the insidious onset of symptoms, a clear-cut history of worsening cognition, and certain cognitive deficits present on history and examination. The role of biomarkers—low CSF amyloid beta, elevated CSF tau, or specified abnormalities on either PET or MRI scans)—is solely to “enhance” the diagnosis, that is to slightly increase its certainty.

More noteworthy than the suggested diagnostic criteria is the recommendation that a syndrome of “preclinical Alzheimer’s disease” be defined. This recognizes the reality that Alzheimer’s disease is years in developing and that the opportunity for intervening to prevent the disease will almost surely arise during this developmental phase. Preclinical AD is defined based solely on biomarker evidence of amyloid beta accumulation or of early neurodegeneration. It requires either measurement of CSF biomarkers or a specialized scan.

In the setting of these proposed definitions, the new study from the Archives of Neurology is very intriguing. It reports on a test of the CSF fluid that measures both amyloid and tau and combines the results in a way that helps predict whether the patient has AD,  or is healthy but likely to develop AD later on. The researchers conducted the study as part of a larger “Alzheimer’s Disease Neuroimaging Initiative” in which they followed a large group of individuals over time, some who were clinically normal, some who had Mild Cognitive Impairment(MCI--a state that often progresses to full-blown dementia), and some who were clinically diagnosed as having Alzheimer’s disease. The subset of this population who agreed to have a lumbar puncture formed the study sample for the Archives work: a total of 114 clinically normal individuals, 200 with MCI, and 102 with clinical evidence of Alzheimer’s. Based on this group, the scientists were able to define a “biomarker mixture” that was abnormal in 90% of those with dementia, 72% of those with Mild Cognitive Impairment, and 36% of those who were cognitively normal. When they tested their model in 2 other groups—one group of 68 people with autopsy-confirmed Alzheimer’s and one group of 57 patients with MCI who were followed for 5 years—they found that the biomarker mixture correctly identified 94% of those in the first group and all of those in the second group who progressed to full-blown dementia.

What all this means is not that patients with a clinical diagnosis of Alzheimer’s disease or of MCI should have a lumbar puncture or a special scan. It decidedly does not mean that cognitively intact older individuals should have their biomarkers measured to see if they are likely to get Alzheimer’s. What it does mean is that we now have the opportunity to do research on healthy people and on those with MCI to see if there is a drug that can prevent the progressive deposition of amyloid that appears to cause full blown Alzheimer’s.

The only problem with this approach is that currently no such drugs have been definitively identified. There are a host of putative disease-modifying agents: vaccines that stimulate the body to develop antibodies against amyloid (which sort of work but with the side effect of causing encephalitis, a brain inflammation), drugs that bind amyloid (which so far have been found to decrease the level of biomarkers but not to affect cognition), and chemicals that interfere with the enzyme that cleaves the abnormal amyloid from a larger precursor protein (which also do what they’re supposed to do biochemically but have not improved cognition).

The hope is that the newly defined preclinical syndrome will encourage researchers to dare to intervene before individuals have any abnormal symptoms. Putting the focus of research squarely on influencing the prodromal period may allow us to prevent or at least delay the progression of Alzheimer’s.

Is this a revolution? It is certainly a conceptual revolution. It embodies our best hope for intervening in the cascade of biochemical events that produces the leading form of dementia. Will it succeed? That remains to be seen.