Alzheimer’s Tests on the Horizon?

Three new studies suggest it may become possible to diagnose Alzheimer’s before symptoms appear.

According to researchers, patients with Alzheimer's may be affected by the disease long before symptoms like memory loss appear. Early diagnosis could slow its progression and improve treatments.


Can Alzheimer's disease be predicted? And if it could, would you want to take the test?

Those questions are being sparked by three new studies, each of which presents a potential method for foretelling, years in advance, the memory loss and cognitive impairment that are the hallmarks of the debilitating brain disease.

Researchers have already identified genetic risk factors that give some people higher odds of one day contracting Alzheimer's. The new studies are different: They propose that certain "biomarkers" of the disease may be detectable after it has begun to affect the brain, but before symptoms have become apparent.

The three studies target different biomarkers in different places: one in the blood, one in the cerebrospinal fluid, and one in the brain itself, as seen in PET scans. None of the tests are close to being available in doctors' offices and clinics. All will require more years of research.

What they share is the potential to "help us understand the early stages of the disease," says Dean Hartley, director of science initiatives for the Alzheimer's Association—and to improve treatment.

A Slow, Invisible Progression

Researchers believe that for perhaps as long as two decades before symptoms appear, the disease follows a steady, damaging course. It's characterized by a buildup in the brain of clumps, or plaques, of beta-amyloid protein and somewhat later by twisted tangles of another protein called tau protein. By the time the Alzheimer's diagnosis is made, the brain damage is already extensive.

The hope is that findings gleaned from earlier detection and diagnosis will also help encourage and speed the development of new therapies that can intervene that much earlier. "The earlier you are able to treat, the more you will be able to do to help, and the prognosis will probably be much better," Hartley says.

For now, however, the options for treating Alzheimer's are limited. The Food and Drug Administration has approved five drugs. All offer temporary relief for some people from the symptoms of Alzheimer's, but none treat the underlying disease.

That's the rub with the idea of earlier diagnosis. On the one hand, the possibility of a test for a disease that currently affects about 35 million people worldwide, including some five million Americans—numbers that are expected to grow larger as the population ages—heralds hope for even more research breakthroughs. On the other hand, with treatment options so limited, potential patients may be leery of foreseeing their future.

"If someone told me that there is a great test for someone like me, I wouldn't want it," says Jason Karlawish, professor of medicine, medical ethics, and health policy at the University of Pennsylvania Perelman School of Medicine. "It would be knowledge that would add to my level of existential anxiety."

Harbinger in the Blood

Perhaps the simplest and least invasive test would be the blood test currently under development by researchers led by Howard Federoff, professor of neurology and executive dean of the Georgetown University School of Medicine. Federoff and his colleagues from six other medical institutions collected blood samples and administered cognitive and memory tests to 525 people aged 70 and over. They then repeated the tests annually.

At the start of the study, 46 participants had some cognitive impairment; by the third year, another 28 had developed symptoms. The researchers discovered that the people with impairment had lower levels of ten blood lipids, even at the beginning of the study, than those who had remained cognitively healthy. The blood lipid levels could distinguish, with an accuracy of over 90 percent, who would remain cognitively normal and who would develop symptoms of Alzheimer's or of mild cognitive impairment over the next few years.

Blood lipids might predict Alzheimer's even farther in advance than that, Federoff says; to find out, he and his team are hoping to get access to archival blood samples from an older longitudinal study.

The researchers don't fully understand why blood lipids would be such good predictors of Alzheimer's. They'll need to understand more, and their study will need to be replicated, before there's any chance of a blood test for Alzheimer's becoming publicly available.

Spinal Taps and PET Scans

The other two potential tests are more involved. Claudio Soto, director of the Mitchell Center for Alzheimer's Disease and Related Brain Disorders at the University of Texas Medical School at Houston, has developed a test that detects misfolded beta-amyloid protein in tiny quantities in the cerebrospinal fluid of Alzheimer's patients—but not in patients with other neurological diseases.

"What we have shown so far is that we can find these particles and you can differentiate" Alzheimer's from non-Alzheimer's patients, Soto says. His study suggests these tiny beta-amyloid particles may be precursors of the plaques in the brain that are characteristic of Alzheimer's. He also wants to see if his technology can be applied to blood and urine samples—much easier to collect than cerebrospinal fluid, which requires a painful spinal tap.

PET scans can already detect beta-amyloid plaque in the brains of Alzheimer's patients. In the third new study, Murali Doraiswamy, professor of psychiatry and director of the neurocognitive disorders program at Duke University, showed that PET scans using the radioactive dye Amyvid can predict a high risk for developing Alzheimer's symptoms in patients that don't yet have them.

"If there are no plaques, the dye has nothing to stick to, so the PET scan is negative," Doraiswamy explains. "But if there are a lot of plaques," they will show up on the scan.

At the start of the study, 152 adults aged 50 and over were given PET scans as well as cognitive tests. Three years later they repeated the cognitive tests. Those with positive PET scans had a much greater rate of cognitive decline and were more likely to have been diagnosed with dementia than those whose scans had been negative.

But even a positive scan is not a perfect predictor, Doraiswamy cautions: "It does not mean you're going to get Alzheimer's."

To Know or Not To Know?

Once such tests become available, how do you decide whether to get one? "Counseling would be in order," says Federoff. It would focus, he says, on establishing whether the patient is ready for the potential results of a test—and also prepared to act on them. Getting treatment is only one possible action; planning financially and for the care that will be needed are others.

Soto sees the development of reliable diagnostic tools as essential for the development of effective treatments. "The main reason there is not good treatment is that there is not good early diagnosis," he says. "When you start [treatment] when the brain is destroyed, the treatment is less likely to work."