In search of Alzheimer's disease

Content provided by the Faculty of the Harvard Medical School

In search of Alzheimer's disease

Does the diagnosis of dementia come years too late? There's plenty of evidence that the processes leading to Alzheimer's disease and other dementing illnesses begin as early as age 30 or 40. Moderate dementia, according to the standard Clinical Dementia Rating Scale, implies difficulty performing acts like dressing, bathing, and toileting. And yet it's estimated that more than two-thirds of people with Alzheimer's disease are already moderately demented by the time they receive a diagnosis, and more than half of those now suffering from dementia have never been diagnosed by a physician.

Some may wonder whether it matters, since as yet there is no way to cure or prevent the most common type of dementia. Fortunately, though, research is producing both more potential ways to recognize the symptoms at an early stage and more reasons for wanting to do so.

Looking at the symptoms

For now, the diagnosis still depends mainly on familiar changes in memory, mood, and behavior. The Clinical Dementia Rating Scale, for example (see Resources), evaluates everyday thinking and behavior with detailed questions about memory, orientation, judgment and problem-solving, community activities, home and hobbies, and personal hygiene. Five stages of impairment are counted, from none to severe. The Alzheimer's Association provides a briefer list of warning signs (see box below).

A sign that's often undervalued is increasing worry about memory loss. Subjective complaints are often dismissed as mere overreactions to the "senior moments" almost all of us experience as we age. But it's been shown that the more often and more strongly a person complains about failing memory, the more likely he or she is to be on the path to Alzheimer's.

Researchers have developed many ways to look for dementia. The most widely used screening method is the Mini-Mental State Examination, a very simple test of memory, orientation, calculation, language, and visual-spatial skills, on which most normal adults get a perfect or nearly perfect score. Another brief test is the Seven-Minute Screen, which includes such tasks as stating the date and day of the week, drawing a clock that shows a certain time, and naming objects in a certain category, such as animals.

Unfortunately, this screening is probably not of much use for the early detection of Alzheimer's disease. In one study, Indiana physicians screened more than 3,000 people over age 65 and found that the cost was nearly $4,000 for a single diagnosis of dementia. The United States Preventive Services Task Force, an expert panel sponsored by the Department of Health and Human Services, has found that there is not enough evidence to recommend routine screening for dementia.

Earlier detection

Many experts now regard a state called Mild Cognitive Impairment as the earliest phase of Alzheimer's disease. By a common definition, it involves serious memory complaints (preferably corroborated by others) and objectively impaired memory (taking into account age and education), with normal overall intellectual capacity and adequate daily functioning, except possibly in complex matters like handling finances. The American Academy of Neurology recommends neuropsychological tests and, if possible, brain scans every six months for people determined to be suffering from these symptoms.

The National Institute on Aging has developed a test for mild cognitive impairment (and possibly for even earlier stages of cognitive decline). Called the Ten Word List, it's a more difficult version of similar items included in the Mini-Mental State Examination and other screening tests. Patients are read a list of words and asked to remember it after a few minutes and some intervening distractions. This particular task — delayed verbal recall — seems to be the best indicator of future Alzheimer's disease. In a study published in 2005, a delayed verbal recall test predicted (when the cutoff point was properly adjusted) whether a healthy elderly person would develop Alzheimer's within the next 10 years with a sensitivity of 83% and a specificity of 74%. That means 83% of people who eventually fell victim to Alzheimer's had failed the test, and 74% of those who passed the test did not develop the disease.

Researchers are look­ing for other ways to identify the risk of dementia earlier and, in particular, to distinguish which persons with memory problems will continue to deteriorate. Brain scans are now showing results — in fact, they are about as accurate as psychological tests in predicting the course of the illness. The main changes observed on brain images occur in the hippocampus, a re­gion crucial for forming and consolidating mem­ories. A mag­netic reso­nance imaging (MRI) study showed that atrophy in this region predicted which persons over age 60 would suffer intellectual decline in the following six years. A positron emission tomography (PET) study showed that the hippocampus in people with mild cognitive impairment consumes up to 14% less energy than it does in matched healthy controls. Other research has found that low activity in the hippocampus predicts Alzheimer's disease three years in advance, and in another study, both mild cognitive impairment and Alz­heimer's up to nine years in advance.

Because of these findings, in Sep­tember 2004 Medicare approved pay­ments for PET scans of patients with "probable" Alzheimer's disease and some participants in clinical drug trials.

A technique in the experimental stages makes use of a tracer dye that is injected into the bloodstream and travels to the brain, where it sticks to beta-amyloid, the stuff of the protein plaques that are thought to be a major cause of Alzheimer's disease. PET scans then highlight the plaques, indicating their density and location. Researchers are looking for a tracer that can be used with single photon emission computed tomography (SPECT), a less expensive and more widely available form of brain imaging. And others have been considering a tracer that works with MRI, which requires no exposure to radioactivity.

It would be useful to have a method of detection more closely related to disease pathology than psychological tests, but less expensive and complicated than a brain scan. Bodily fluids — blood, spinal fluid, and urine — are the first choice. The most promising substances to test for are beta-amyloid, the source of plaques, and tau protein, the stuff of the neurofibrillary tangles (dead neurons) that are an important biological marker of Alzheimer's. In mild cognitive impairment and Alzheimer's disease, tau protein levels in the spinal fluid rise as neurons die, while beta-amyloid levels fall because the pathological form of the protein is insoluble and remains in the brain. Like all present biological and psychological tests, these are imperfect. Many people with plenty of tangles and plaques never develop symptoms of Alzheimer's disease.

Other promising test substances are homocysteine, an amino acid (protein constituent); substances called isoprostanes, which may be markers of the damage done by oxygen free radicals, destructive products of metabolic activity; and C-reactive protein, a sign of inflammation. Researchers are working on ways to detect small changes in the levels of these substances and others in the blood and spinal fluid. Maybe someday the results will produce a formula that predicts Alzheimer's disease more accurately and at an earlier stage than anything available now.

The chief known genetic risk factor for the most common form of Alzheimer's disease is the E4 variant of the gene that codes for the manufacture of apolipoprotein (APO) E. In the United States, about 80% of people with Alzheimer's inherit a copy of this gene from at least one parent (though not all who carry the gene develop Alzheimer's). Another genetic variant has also been discovered that is associated with Alzheimer's, but only if it is inherited from both parents. This gene codes for the protein ubiquitin, which, like apolipoprotein E, is involved in the circulation and metabolism of cholesterol. One study suggests that a group of cholesterol-related gene variants, taken together, is also associated with susceptibility to Alzheimer's disease.

In recognition of all these developments and in hopes of further progress, the National Institute on Aging is sponsoring a five-year study called the Alzheimer's Disease Neuroimaging Initiative. The study, which began in the spring of 2005, will recruit hundreds of participants with Alzheimer's disease or mild cognitive impairment, along with healthy controls. Its aim is to track structural and functional brain imaging results, genetic and other biological markers, neuropsychological tests, and clinical symptoms in order to chart and correlate all these aspects of the transition from brain health to dementia. It should provide a large body of knowledge for use in early-stage prevention and treatment.

Why it matters

Given the lack of an effective treatment, early detection could have some drawbacks. Surveys show that most people don't want to take a test for future Alzheimer's disease unless it is more than 90% accurate — a level still unreachable. A study found that children of Alzheimer's patients did not become depressed or anxious when they were told that they carried the APO E4 genetic variant. But they were six times more likely to buy long-term care insurance, which is both good and bad news. Federal law restricts the use of genetic information to deny health insurance, but the law does not apply to long-term care insurance. If the number of people carrying this insurance who develop Alzheimer's increases, the cost will rise for everyone.

That problem requires a political solution, but otherwise knowing the risk has obvious advantages. Patients and families can make long-term care plans, formulate medical directives, choose a proxy for making legal decisions, decide whether to participate in clinical research, and start considering such issues as driving and handling money. People who learn they are not going to develop Alzheimer's will be able to rest easier. Those who turn out to have other causes of dementia (for example, depression or vascular problems) will have a better chance of getting proper treatment.

Most important for research, earlier detection of the signs of dementia would make clinical trials easier to conduct, and the development of drugs and other treatments would become quicker, safer, and more effective. Here two lines of research converge and reinforce each other. The earlier we can detect Alzheimer's disease, the better chance we have of finding ways to treat or prevent it; and the more we learn about risk factors and potential treatments, the more important early detection will become.

Last updated:	September 05, 2008