Using a compound developed at the University of Pittsburgh, researchers in Sweden have been able to scan the brains of Alzheimer's patients and, for the first time, spot the amyloid plaques that are a hallmark of the disease.
The breakthrough could lead to early diagnosis and to the development of new treatments.
The first human trials of "Pittsburgh Compound B," or PIB, began in February in Uppsala, Sweden, and the findings from the first 14 patients are being presented today at the International Conference on Alzheimer's Disease and Related Disorders in Stockholm.
Until now, the tiny, abnormal clumps of amyloid protein associated with the disease could be identified in the brain only through an autopsy. Doctors consequently have had to diagnose Alzheimer's by excluding other explanations for a patient's dementia.
The Pittsburgh compound, which is labeled with an extremely short-lived radioisotope, is injected into patients, who are then scanned using positron emission tomography, or PET, imaging. In the Swedish studies, nine patients with Alzheimer's showed obvious concentrations of the compound in those areas of the brain where plaques typically form. On the PET image, those areas appear bright yellow, red, green, or blue, depending on the concentration.
But the same areas in five non-Alzheimer's patients appeared black on their PET scans -- a strong indication that no compound was present, Dr. William Klunk of Western Psychiatric Institute and Clinic said yesterday.
Human trials at the UPMC Health System might begin late this year or early next year. Additional patients will continue to be scanned by the researchers at Uppsala University PET Center.
Klunk, director of psychiatry at the Pittsburgh Alzheimer Disease Research Center, said the compound he developed with Pitt radiology professor Chester Mathis isn't much needed by patients with moderate to severe cases of dementia. Their problems are obvious. But it could prove critically important for diagnosing patients who have still-subtle symptoms or who may not be symptomatic at all.
"This is a tool to push diagnosis earlier and earlier," Klunk said. Early diagnosis, he explained, might allow more patients to start treatment when it is most effective.
The Pittsburgh compound is not a new way to treat Alzheimer's, Klunk emphasized, but it might well speed the development of new ones. In the near term, in fact, the primary use of PIB likely will be for evaluating the effectiveness of experimental treatments.
"Having the ability to quantify amyloid deposition in the brain will have a profound impact on our ability to monitor the progression of Alzheimer's, as well as gauge the effectiveness of medical treatments," said William Thies, vice president of medical and scientific affairs for the Alzheimer's Association.
Many therapies are designed to slow the accumulation of these amyloid proteins, but without a means of directly seeing if the drugs are working, evaluating their effectiveness can take much longer, Mathis said.
Klunk and Mathis have been working on the compound for more than a decade. They began with studying the same dyes that pathologists use to highlight amyloid plaques during brain autopsies. They needed to find a way to alter the dyes so that, once injected into the bloodstream, they could cross what's called the "blood-brain barrier" and enter the brain. The barrier, which lines the blood vessels in the brain, is a protective mechanism that keeps many drugs and other substances out of the brain.
Those early attempts were unsuccessful, so they turned to thioflavin-T, a poor-performing dye that now is seldom used by pathologists. But it proved to be perfect for their purposes. To help it cross the blood-brain barrier, they altered the positively charged molecule so that it was electrically neutral, Mathis said. And they tried attaching various chemical groups to the molecule to increase its affinity for amyloid protein.
They tried more than 100 chemical variations of thioflavin-T before finding one called 6-OH-BTA-1 that crossed the blood-brain barrier, attached to amyloid plaques and cleared itself out of normal brains.
The compound was tested locally in both mice and baboons to determine that the trace amounts that are injected -- a thousandth of a milligram -- were safe and that it effectively bound to amyloid plaques. Further microscopic studies with mice at Massachusetts General Hospital confirmed that the compound was indeed latching onto the plaques. Each plaque measures only about a tenth of a millimeter in diameter -- "not the head of a pin, but the point of a pin," Klunk said.
Mathis said it may take several more years of testing to confirm that the Pittsburgh compound is an accurate indicator of amyloid plaques and before the research community accepts it as a standard test.
He said he expects that the cost of the scans will be about the same as the PET scans now used for certain cancers -- between $1,700 and $2,000 each.
E-mail this story 
