Imaging Reveals Alzheimer’s Clues both Before and After Disease Develops


by Carl Sherman

June, 2009

Years of research have failed to generate effective treatments for Alzheimer’s disease (AD). Now, scientists are employing diverse neuroimaging and analytical approaches to examine brain structure and function years to decades before symptoms might emerge, in an effort to develop treatments that could stave it off.

“We have all the evidence we need that pathology starts well before clinical symptoms appear,” says William Klunk, a professor of psychiatry and neurology at the University of Pittsburgh. “We need to use techniques that can show us that pathology, identify people who are going to develop the disease and treat them before the symptomatic phase, when it may be too late.”

To this end, researchers are employing imaging in brains diseased and healthy, old and younger.

MCI or Early Alzheimer’s?

 Much of this research has focused on people with mild cognitive impairment (MCI), mental decline including memory loss that is noticeable but not severe enough to interfere with life. Research indicates that 10 to 15 percent of people with MCI will develop Alzheimer’s disease within a year.

Researchers working to identify which cases of MCI are actually Alzheimer’s disease at an early stage have found clues in the loss of brain tissue recorded by magnetic resonance imaging (MRI). Most have focused on the hippocampus, a memory center hit early by Alzheimer’s. But one recent study, reported in the April 2009 Radiology, found that widespread atrophy—a pattern of tissue loss in temporal, cingulate and orbitofrontal regions—clearly differentiated Alzheimer’s patients from healthy control subjects.

“This pattern could be specific to AD,” says Linda McEvoy, a researcher in the multimodal imaging laboratory at University of California, San Diego, and an author of the paper. “There are other disorders that affect the hippocampus, but with these multiple cortical areas [involved], it’s more likely to be Alzheimer’s.”

When McEvoy examined people with MCI, she found the Alzheimer’s pattern of atrophy in more than half. A year later, their mental status had worsened significantly, on average, and the atrophy had progressed, while those with healthy-looking brains had remained stable. Of the 33 participants with MCI who were diagnosed with AD during that year, 26 had had the telltale brain loss pattern when first tested.

Other researchers have analyzed chemicals in cerebrospinal fluid (CSF). In a paper in the April 2009 Annals of Neurology, a team led by Leslie Shaw of the University of Pennsylvania reported a distinctive “biomarker signature” in the CSF of people with mild AD, which involved two proteins strongly associated with the disease: beta-amyloid and tau. The researchers found a similar concentration of these proteins in the CSF of 33 of 37 people with MCI who went on to develop AD by one year later.

“This is a very important paper,” says Neil Buckholtz, chief of the dementias branch at the National Institute on Aging. The changing concentrations of beta-amyloid fragments and tau protein in CSF, he says, may reflect the actual disease process.

Imaging a Telltale Protein

Since 2002, amyloid in the living human brain has been visible thanks to a tracer called Pittsburgh compound B (PiB), tested in part via research funded by the Dana Foundation. PiB binds to amyloid deposits and lights up on positron emission tomography (PET) scans. But because the isotope with which PiB is labeled, carbon-11 (11C), has only a 20-minute half-life, research with the tracer has been limited to institutions with a cyclotron on site.

Amyloid-binding compounds now under development should extend the reach of this technique. These compounds employ fluoride-18 (18F), whose half-life is 110 minutes, allowing use by scientists in any lab within a two-hour radius of a cyclotron, says Dean Wong, professor of psychiatry, neuroscience and environmental health sciences at Johns Hopkins University.

“The widespread availability of 18F amyloid imaging agents would be paradigm-shifting,” predicts Michael Weiner, director of the Center for Imaging of Neurodegenerative Diseases at the San Francisco Veterans Affairs Medical Center and principal investigator for the Alzheimer’s Disease Neuroimaging Initiative (see box). Beyond Alzheimer’s, “it will reveal a lot of new things about role of amyloid in normal aging, late-life depression and the impact on the brain of traumatic injury and diseases like diabetes.”

Spotting Alzheimer’s Before It Happens

Scientists ultimately hope to be able to identify the Alzheimer’s disease process before any symptoms occur—when the brain is presumably undamaged and treatment would have the best chance of success. Research toward this goal has focused on people who carry a gene, apoE4, strongly associated with increased risk. Those with one copy of apoE4 are twice as likely as noncarriers to develop AD; those with two copies are 10 times as likely.

In research reported in the April 21 issue of Proceedings of the National Academy of Sciences, Eric Reiman of the University of Arizona and colleagues used PiB PET to examine 28 cognitively normal people with an average age of 64. They found more beta-amyloid, on average, in the brains of those who carried one apoE4 allele than in noncarriers, and even more in those who had two alleles.

Another study, published one week later in the same journal, used functional MRI to examine brain activity in healthy young adults ages 20 to 35. A memory task caused greater hippocampal activation in those who carried at least one copy of the apoE4 allele than those who had none—an indication, the authors say, that the genetic variant affects brain function long before symptoms of neurodegeneration would appear.

How (or even whether) changes in brain function at age 30 are linked to the development of Alzheimer’s many years later is unclear; nor is it certain that apoE4 carriers with extra amyloid are the ones who will develop the disease.

“The time line is a long one,” says Mony de Leon, director of the Center for Brain Health in the department of psychiatry at New York University. “What will ultimately be needed are studies targeting normal individuals in high-risk groups that take observations [of several types] forward over an extended period.”

On the Cusp of Change

Neuroimaging and biomarker findings have given researchers a coherent, if tentative, map of how Alzheimer’s develops, says Ronald Petersen, director of the Alzheimer’s Disease Research Center at the Mayo Clinic. Amyloid deposition is detectable decades before clinical symptoms appear. Metabolic decline in the cortex starts to show up on scans as the disease begins to affect brain function. Areas of atrophy on MRI represent further and possibly irreversible deterioration, which grows more pronounced as the disease progresses.

“Different imaging tests might have utility at different points in the clinical spectrum,” he says.

On a practical level, the research has produced “a boatload of data, probably enough to establish revision of diagnostic criteria of AD and make [the diagnosis] earlier,” says de Leon.

Discussions about how neuroimaging and biomarkers might be integrated into diagnostic protocols have begun, says Buckholtz, of the National Institute on Aging. “If we get agreement within the scientific community, this could happen in the next couple of years.”

On the research front, “more and more companies and academic investigators are incorporating imaging and biomarkers in clinical trials,” he says. By making it possible to track therapeutic effects more quickly and precisely, “this data may clear the way for developing drugs with disease modifying effects.”