Image: Courtesy of Dan Skovronsky
When actor Charlton Heston announced in August that he is "suffering symptoms consistent with Alzheimer's disease," he used qualified language because diagnosis is possible only postmortem. The lack of a clear set of symptoms and biomarkers is not only frustrating for families, but is hindering the search for treatments.
"Guidelines for treating Alzheimer's focus on symptom relief, and these are not sufficient to develop drugs to delay or prevent symptoms," said Neil Buckholtz, head of the Dementias of Aging branch of the National Institute on Aging (NIA) at "Imaging Alzheimer's and other Neurodegenerative Diseases," a symposium held recently at the General Electric Global Research Center in Schenectady, NY. Investigators lack measures with which to monitor the efficacy of drug candidates. "The holy grail is to develop direct in vivo measurements of the plaque and tangle burden in the brain. But until a valid biomarker is available, better indirect measures of Alzheimer progression are needed," said Clifford Jack, director of the Alzheimer's Disease Research Center at the Mayo Clinic.
Those biomarkers may be coming, thanks to the NIA's Alzheimer's Disease Neuroimaging Initiative, slated to begin by 2004 with $10 million-plus (US) from the National Institutes of Health, the Food and Drug Administration, the pharmaceutical industry, academic centers, device manufacturers, the Alzheimer's Association, and the Institute for the Study of Aging. Meetings between NIA and pharmaceutical companies led to the program. "The conclusion was that very little imaging has been done in clinical trials, and in most studies, the imaging quality wasn't very good," related Buckholtz.
The project will collect magnetic resonance imaging (MRI) and positron emission tomography (PET) serial images and clinical measures in 100 normal individuals and in 400 people experiencing mild cognitive impairment, which can precede Alzheimer disease. And in the spirit of the human genome project, the data will be freely available.
Speakers at the meeting presented variations on the MRI theme already under way. Arterial spin labeling, which measures local cerebral blood flow, and blood oxygen level-dependent (BOLD) MRI paint complementary portraits of the human brain. "We perturb the inflowing nuclear spins in arterial blood, wait until the perturbed spins enter tissue, and see changes in the venous outflow. The technique takes naturally existing water in blood and uses magnetic fields to label it. Then we image the brain to show the cerebral perfusion," explained David Alsop, associate professor of radiology at Harvard Medical School. Ebbing blood flow correlates with decline in mental state.
Reisa Sperling, an assistant professor of neurology at Harvard Medical School, explained BOLD. "It is an indirect measure of neuronal activity. Deoxyhemoglobin is a paramagnetic substance, and it decreases the MRI signal. We use it as an intrinsic contrast." BOLD-MRI scan results differ in patients with early Alzheimer versus controls when challenged with a face-name association task. In the patients, chaotic signals came from the hippocampus, the memory center, compared to a regular pattern in controls, she reported. Paradoxically, the hippocampus is more active in impaired patients. "It may be a 'use every neuron you've got' compensatory activity," she speculated.
A more direct view of the Alzheimer brain is plaque imaging. "Imaging atrophy and changes in blood flow, biochemistry, or glucose metabolism are indirect. Plaque imaging is more direct, because the amyloid burden is the primary pathological feature," said Jack.
Plaque imaging relies on mice transgenic for human beta-amyloid whose stained brains can be sliced and observed. Ideal probes for beta-amyloid must cross the blood-brain barrier, selectively target pathogenic amyloid, and be quantitatively detected with noninvasive imaging, related Dan Skovronsky, a research associate at the Center for Neurodegenerative Disease Research at the University of Pennsylvania.
Skovronsky compared iodine-modified versions of two standard stains for amyloid. A Congo-red derivative zeroed in on the tangles, made of tau protein, and a nonpathogenic variant of beta-amyloid. But a thioflavin T derivative had the desired specificity, homing in on beta-amyloid in plaques. "There was a good correlation between the amount of fluorescent binding and the number of plaques," Skovronsky said.
Jack unveiled preliminary data on a "smart" molecule that consists of the organic molecule putrescine, normally found in rotting flesh, linked to gadolinium and beta-amyloid. The putrescine pulls the molecule through the blood-brain barrier, gadolinium provides contrast, and beta-amyloid is taken up by forming plaques. "The construct binds specifically to plaques, leading to detectable changes in the MRI," he reported.
With these eclectic new views of the Alzheimer-altered brain, pharmaceutical companies may soon have the long- needed tools to effectively monitor this disease that affects so many millions--and, eventually, to intervene.
Ricki Lewis (firstname.lastname@example.org) is a contributing editor.