Early Detection of Alzheimer's Disease:
How Early and by What Means?

The research frontier in treatment for Alzheimer's disease (AD) is being defined by the search for disease-modifying therapies—therapies that would alter the course of the disease by delaying its onset, slowing it down, or stopping its progression. Such treatments will likely be expensive and carry some risk, so it is ever more critical to be able to predict who is going to get the disease and when.

Redefining Early Detection

How early is "early" in a disease that strikes the elderly? Richard J. Caselli, M.D., a Mayo Clinic neurologist and his colleagues in the Arizona Alzheimer's Disease Center, are finding out that "early" may be as young as middle age in individuals at risk. In the first longitudinal investigation of its type, they have discovered subtle cognitive changes that predate clinical signs of AD by as many as 10 to 15 years in carriers of apolipoprotein e4.

Apolipoprotein (APOE) is a gene isoform or protein, present in all humans, that helps transport lipids. Three types or alleles—e2, e3, and e4—can be inherited in any of 6 combinations. While APOE e2 appears to be protective against AD, APOE e4 is a serious risk factor. In people with 2 copies of APOE e4, the risk of developing AD is 12 times greater than the risk in the general population. Approximately 25% of the population has at least 1 copy of APOE e4, and the presence of APOE e4 is thought to explain about 50% of cases of AD.

In the mid 1990s, Dr. Caselli and collaborators began conducting extensive behavioral and neuroimaging studies every 2 years on APOE e4 homozygotes (those with 2 copies), heterozygotes (those with 1 copy), and noncarriers. At the same time, Ronald C. Petersen, Ph.D., M.D., and colleagues at Mayo Clinic in Rochester, Minnesota, began to characterize a condition called mild cognitive impairment (MCI), subtle cognitive changes that predate clinical symptoms of AD by 3 to 5 years. Now Dr. Caselli and colleagues have found that in those known to be at extremely high risk (ie, APOE e4 homozygotes) subtle changes in cognitive status occur much earlier and may represent a pre-MCI stage of AD. Substantiation of these findings could alter the timetable for early therapeutic interventions in those with established risk factors. As Dr. Caselli says, "Alzheimer's research has focused on the elderly, but it appears that we need to look earlier in the lifespan. We are now including people as young as 20 years old in our samples."

Defining Risk

Knowing whom to treat is as important as knowing when to intervene. Early detection will likely depend on identifying combinations of risk factors. At Mayo Clinic's 3 locations, investigators are uncovering a wide array of AD predictors and methods of detection.

Two of the 29 Alzheimer's Disease Centers funded by the National Institutes of Health (NIH) are at Mayo Clinic. The Mayo Clinic Alzheimer's Disease Research Center, a partnership between Mayo Clinic in Rochester, Minnesota, and Jacksonville, Florida, is directed by Dr. Petersen in Minnesota with codirectors Neill R. Graff-Radford, M.D., and Steven G. Younkin, M.D., Ph.D., in Florida. The Arizona Alzheimer's Disease Center is a consortium of Arizona institutions, of which Dr. Caselli is the Clinical Core Director. In addition to basic science and clinical trials research, coordinated research at these 2 centers and Mayo's 3 sites is fostering discovery of reliable biomarkers for risk. Mayo Clinic in Arizona also has state funding to conduct regularly scheduled tissue sample analysis, neuroimaging, and neuropsychological testing across the lifespan. The Mayo Clinic Study on Aging in Minnesota, supported by a separate NIH grant, conducts similar studies on more than 2,000 subjects without dementia who have been enrolled to date.

"The ultimate goal of Alzheimer's research is to prevent the disease, and the only way to prevent it is by knowing which patients are at risk before they get it," states Dr. Graff-Radford. "To do that we need to stratify the population into levels of risk," adds Dr. Petersen, "which is what we're doing throughout Mayo Clinic's AD research programs." Biomarkers under investigation at Mayo include structural changes in the brain, genetic predisposition, changes in blood chemistry, altered cognitive status, and other behavioral signs of early onset.

Neuroimaging

Identifying correlations between behavioral changes and shifts in the volume, density, and activity of brain structures is a major step forward in defining risk before symptom onset in AD. Mayo's innovation in MRI is rapidly expanding the methods of detecting presymptomatic AD and other dementias. Led by neuroradiologist Clifford R. Jack, M.D., who won the American Academy of Neurology's Potamkin Prize for Research in Pick's, Alzheimer's, and Related Diseases in 2008, researchers at Mayo Clinic in Rochester are using voxel-based (volumetric) morphology, diffusion tensor imaging, and various other new analytical techniques to identify volumetric shifts in white matter as well as in specific brain structures that signal early signs of AD.

The Arizona Alzheimer's Disease Center is investigating MRI changes in brain density and volume, in white matter and gray, and in the integrity of neurovasculature in APOE carriers and noncarriers. They have also used positron emission tomography (PET) to study these patients. Numerous PET scan studies have shown that AD is associated with an abnormally low glucose metabolic rate in regions of the precuneus and posterior cingulate, the parietotemporal, and the frontal cortex. Dr. Caselli and his colleagues found that reduced glucose metabolism in these same brain regions correlated with APOE e4 gene dose (ie, the number of e4 alleles in a person's APOE genotype). They and others have concluded that PET may be a useful diagnostic tool and a quantifiable measure of the effects of gene-modifying therapies in the future.

Preclinical Cerebral Metabolic Rates of Glucose (CMRgl) Changes in APOE e4 Carriers

Enlarge

Mayo Clinic in Rochester has recently begun using a molecular-based imaging technique to detect pathologic alterations in the brain. The imaging tracer being used is called Pittsburgh compound B (PiB). It detects a form of the amyloid protein that is deposited in the brain. The proteins amyloid beta 42 (Ab42) and amyloid beta 40 (Ab40) are major components of the plaques associated with AD and a major part of the disease process.

A Blood Test for AD?

Amyloid protein increases with age, and in people with AD it is deposited in the brain as plaques, generating a decline in the level of certain forms of protein in the blood. Investigators in the Mayo Clinic Alzheimer's Research Center have developed a blood test that may Dr.amatically improve presymptom detection of AD. The test measures a ratio of amyloid proteins Ab42 to Ab40 in the blood. In a longitudinal study of 563 elderly subjects, they found that the 53 participants who developed MCI and/or AD had decreased levels of Ab42 at 3 to 5 years before symptom onset. Levels of Ab40 either increased or decreased at a slower rate. Those with the lowest ratio—low levels of Ab42 and high levels of Ab40—were 3 times more likely to develop AD or MCI. The research team, led by Dr.s Younkin and Graff-Radford in Jacksonville and Dr. Petersen, Bradley F. Boeve, M.D., and David S. Knopman, M.D., in Rochester are in the process of recruiting 3,000 people to complete the study in the next 10 years. Although the test would not be the sole predictor, in combination with other known risk factors, it could revolutionize the process of early detection.

Genetic Risk Factors

Mayo Clinic in Jacksonville has characterized the amyloid protein and the genetics of AD and other dementias. Few genes are as powerful as APOE e4 in signaling risk for AD. Genetic risk for the approximately 50% of AD not explained by it will likely be defined by gene combinations and by gene-plus-environmental factors. Scientists at Mayo Clinic in Jacksonville are conducting genome-wide association studies in search of just such gene combinations.

Behavioral Predictors and Environmental Risk Factors

In both Alzheimer's Disease Centers, researchers are looking for behavioral changes that predate AD. For example, they have found that difficult tests of verbal memory such as the Auditory Verbal Learning Test are particularly valuable in detecting both MCI and pre-MCI cognitive decline (Figure 2).

Performance of APOE e4 Homozygotes and Heterozygotes on the Auditory Verbal Learning Test

Enlarge

Cognitive changes are the obvious, but not the only, behavioral predictor of dementia. For example, rapid eye movement sleep disturbance, of which Dr.eam enactment behavior is the defining clinical feature, has been associated with dementia with Lewy bodies (LBD). Although dementia with LBD differs from AD, Dr. Caselli and colleagues in the Arizona Alzheimer's Disease Center looked at this behavior relative to risk for AD in a small sample of healthy APOE carriers and noncarriers. They found that those with Dr.eam enactment behavior had reduced glucose uptake in areas of the brain known to be similarly affected by LBD as measured by fluorodeoxyglucose (FDG) PET scans.

Finally, lifestyle factors may also influence risk. For example, Dr. Caselli and his team are finding that fatigue has a greater effect on memory in APOE homozygotes than it does in noncarriers. And, while correlations between memory impairment and cardiovascular status are weak in heterozygotes, they are robust in homozygotes. As Dr. Caselli states, "Thus, it may be that in people at high risk for AD, exercise and cardiovascular health are particularly critical."