Dec 12, 2006
WASHINGTON — Scientists appear to have found a fingerprint of Alzheimer's disease lurking in patients' spinal fluid, a step toward a long-awaited test for the memory-robbing disease that today can be diagnosed definitively only at autopsy.
Researchers at New York's Weill Cornell Medical College discovered a pattern of 23 proteins floating in spinal fluid that, in very preliminary testing, seems to identify Alzheimer's — not perfectly, but with pretty good accuracy.
Far more research is needed before doctors could try spinal-tap tests in people worried they have Alzheimer's, specialists caution.
But the scientists already are preparing for larger studies to see if this potential “biomarker” of Alzheimer's, reported Tuesday in the journal Annals of Neurology, holds up.
“We're looking to an era in which the kinds of uncertainties that many patients and their families face about the diagnosis of Alzheimer's disease will no longer be a problem,” predicts Dr. Norman Relkin, a neurologist and the study's senior researcher.
About 4.5 million Americans have Alzheimer's, a toll expected to more than triple by 2050 as the population greys. The creeping brain disease gradually robs sufferers of their memories and ability to care for themselves, eventually killing them. There is no known cure; today's drugs only temporarily alleviate symptoms.
Currently, doctors diagnose Alzheimer's mainly by symptoms. That makes early diagnosis particularly difficult, and even more advanced disease can be confused with other forms of dementia. Nor is there a good way to track the disease's progression, important both for decisions about patient care as well as in testing the effectiveness of new drugs.
Major research is under way to try to change that, including a $60 million study now under way to give brain scans to 800 older Americans and try to pin down the earliest brain changes associated with Alzheimer's.
At the same time, scientists also are hunting what they call biomarkers — signs of the disease in areas other than hard-to-test brain tissue.
“A valid biomarker for Alzheimer's disease is sorely needed,” said Dr. Sam Gandy, a neuroscientist at Philadelphia's Thomas Jefferson University and spokesman for the Alzheimer's Association.
But the new protein pattern requires “rigorous validation” by other researchers to make sure it really is linked to Alzheimer's, he cautioned.
By hunting for one protein at a time, scientists have discovered a few other biomarker candidates in cerebrospinal fluid. But Dr. Relkin and colleagues at Cornell University expanded the hunt: Using a technology called proteomics, they simultaneously examined 2,000 proteins found in the spinal fluid of 34 people who died with autopsy-proven Alzheimer's, comparing it to the spinal fluid of 34 non-demented people.
What emerged were 23 proteins, many that by themselves had never been linked to Alzheimer's but that together formed a fingerprint of the disease.
Then the researchers looked for that protein pattern in the spinal fluid of 28 more people — some with symptoms of Alzheimer's or other dementia, some healthy. The test indicated Alzheimer's in nine of the 10 patients that doctors suspect have it, and incorrectly fingered three people.
What's next? That huge brain-scanning study is collecting spinal fluid samples from some participants, and Dr. Relkin has begun talks with those researchers about testing his results. At his own hospital, he's using the protein test in a study of an experimental Alzheimer's treatment to see if changes in the fingerprint may predict when the drug does or doesn't work.
Scientists believe that Alzheimer's begins its insidious brain attack years, even decades, before forgetfulness appears — and if so, there should be evidence of those changes in the spinal fluid, Dr. Relkin explained.
“The spinal tap gives people pause,” he acknowledged, agreeing that a blood test would be easier. But, “in expert hands ... it's not much more traumatic than having blood taken.”