Spring 2010

Imaging Alzheimer’s Disease
By Beth W. Orenstein
Aging Well
Vol. 3 No. 2 P. 6

The cause of Alzheimer’s disease (AD) is unknown, but scientists believe it may involve a malfunction in the processing of a protein known as amyloid precursor protein, which is found in many cells, including brain neurons. It is known that patients with the disease have a buildup of amyloid protein aggregates or plaque in their brains.

Researchers at the University of Pittsburgh have developed a radiotracer known as Pittsburgh Compound-B (PiB). When used with positron emission tomography (PET), data suggest that PiB may be able to not only confirm AD but diagnose it in people who are presymptomatic. PiB is useful because it has the ability to cross the blood-brain barrier, bind specifically to amyloid plaques, and clear from normal brain tissue. It is a variation of a tissue dye (thioflavin-T) used to positively diagnose AD after death.

Since researchers unveiled images of controls and AD subjects using PiB and PET at a 2002 international meeting in Stockholm, Sweden, a number of groups have been doing AD research. “We think over 3,000 studies have been done ... on human subjects,” says Chester Mathis, PhD, a professor of radiology at the University of Pittsburgh and coinventor of the PiB technology. “PiB is by far the most widely used compound. There have been groups throughout the world who have picked this up and are moving fairly quickly with it.”

In the past, only a brain autopsy revealed a definitive AD diagnosis. However, Mathis says, the excitement is no longer just that an AD diagnosis can be made using a brain scan while a patient is still alive. The hope is that PET scans using PiB and other tracers can be used to assess the extent of amyloid deposits in patients’ brains years before AD symptoms appear, when the disease may still be preventable. “Today, the real excitement in the field is not at end stage where things are going to be very difficult to reverse, if it’s possible to reverse,” Mathis says. “It’s going to be earlier in the course of the disease, where you can stop the process before the irreversible damage is done.”

Several pharmaceutical companies are working on antiamyloid therapies and if they are successful, it would be crucial to be able to determine which patients with plaque are likely to develop AD, Mathis says. As a result, drugs could be administered to those who are at high risk of developing AD.

According to the Alzheimer’s Association, more than 5 million Americans have AD. As baby boomers age, that number is expected to rise significantly. By 2030, it is believed that 8 million people in the United States will be diagnosed with AD, and the number could double to 16 million by 2050.

Tracking Treatment
PET studies with PiB also could be used to track the success of the drug therapies, Mathis says. “If these drug therapies are developed, the prediction is they will be more efficacious because they will be able to help people whose brains have not yet been severely damaged,” he says.

Some studies have suggested that the characteristic signs of AD are present a decade or more before dementia sets in. So if the presence of amyloid deposits could be confirmed with brain scans, drug therapies could be administered in time to be useful, Mathis says.

With the scan technique, researchers are beginning to identify which patients are likely to develop AD. At the annual meeting of the Society of Nuclear Medicine in 2008, Mathis and his research group presented a longitudinal study investigating the progression of AD. They scanned 35 people at yearly intervals for four years. Of the 35, four had mild to moderate AD; 10 had mild cognitive impairment (MCI), believed to be a precursor to AD; and 21 were elder controls.

The research literature shows that 30% to 40% of people with MCI do not progress to AD during five to 10 years of follow-up testing. “Our hypothesis was that MCI subjects with brain plaque would develop AD and MCI subjects without plaque would not advance to AD,” Mathis says.

The researchers found that about 60% of the MCI subjects had plaque loads comparable with AD subjects, while about 35% of MCI subjects had no detectable plaque. After monitoring their participants for four years, the researchers found that only those with plaque progressed to the clinical diagnosis of AD. Mathis says that in AD subjects, “There appears to be a ceiling to a plaque deposition, and plaque concentration does not increase as the disease progresses from mild to advanced stages of AD.”

The researchers also had assumed that the elder control group would have little PiB retention, which would indicate no amyloid deposits. However, they found that 25% of the control group aged 65 to 80 had significant deposits of amyloid plaque in their brains. Therefore, Mathis says, the researchers hypothesized that the control subjects were in a presymptomatic at-risk state that eventually will lead to the development of AD.

One issue with PiB retention indicating whether a patient will develop AD is that it can be difficult to determine how much amyloid is indicative of disease, Mathis says. “It’s not that one day you’re negative for PiB and the next day you’re positive. It’s a continuous variable. We see this in the elderly compared to young controls. They start creeping up with age in terms of amyloid in the brain. The older one grows, the more amyloid in the brain in about 25% to 35% of the elderly. For many years, this has been known as the result of postmortem studies. So there comes a point at which you can say, ‘That subject is positive or negative with respect to the extent of amyloid deposition.’”

Mathis’ research team has developed a scale to suggest who would develop AD. At one end of the scale would be someone who is young and healthy and would have no PiB retention. At the opposite end, someone with at least a certain amount of PiB retention would be considered to have AD. PiB levels above a cutoff threshold of 1.44 to 1.48 distribution volume ratio, depending on the brain region affected, were considered to be elevated (PiB positive). “However, the cutoff is just a cutoff, and amyloid deposition is a continuous variable. We don’t know what it really means with respect to absolute amount of amyloid in the brain because we have autopsy data only on AD subjects at the end of the amyloid deposition spectrum,” Mathis says. “But it is useful for dichotomizing subjects into groups and comparing groups for research purposes.”

While different research teams have different cutoff points, most are within a few percentage points of each other with respect to where they would draw the line, Mathis adds.

William Jagust, MD, a professor of public health and neuroscience at the University of California, Berkeley, is another researcher who has been using imaging and PiB to track amyloid buildup in the brain and determine its relationship to AD. “It’s a real breakthrough in our ability to track this protein,” he says of the technique.

However, Jagust says, he doesn’t believe there will ever be a cutoff point where a physician could say with certainty that someone is likely to develop AD and someone else is not. “Drawing a line is arbitrary,” he says. “It may be useful and an important predictor of whether someone is likely to develop AD, but I don’t think it’s going to turn out to be quite that simple.”

Having amyloid plaque in the brain is a necessary condition for an AD diagnosis, but it is not sufficient, Jagust says. “Something else has to happen. The levels necessary to develop AD may be different with different people. One person may require high levels of plaque and another not. That may have to do with other factors that can cause one to get AD. It’s real complicated. It’s not too complicated to attack or to think about, but I don’t think it’s going to be as simple as drawing a line and saying above it ‘yes’ and below it ‘no,’” he says.

Jagust is optimistic that researchers will someday find a biomarker that could be used in conjunction with PET scanning to determine who is likely to develop AD. The biomarker may be a blood test similar to prostate-specific antigen for prostate cancer or it may be something different, he says.

Mathis has been working on developing a noninvasive brain-imaging method since the mid-1980s. He began his work at Lawrence Berkeley Laboratory and in the mid-1990s met with William E. Klunk, MD, PhD, a professor of psychiatry and neurology who also had been working on a method to quantify amyloid deposition in the brain, prompting him to move to the University of Pittsburgh. They began collaborating and together invented the PiB compound.

The PET scan with PiB is straightforward and a relatively simple procedure to perform, Mathis says. The patient is injected intravenously with the radiotracer while outside the scanner. About 40 minutes after the radiotracer is injected, the patient is placed in the PET scanner. The PET scan takes about 20 minutes. The patient is taken from the scanner and goes home within a few hours.

Mathis says the development of the PiB compound will make a difference if and when there are therapies to treat or at least slow the progression of AD in people who are shown to have amyloid plaque buildup in their brains. “Until then,” he says, “it’s interesting, but the field will get very, very excited about it when there is an effective treatment for AD and the two can be used together to help patients.”

— Beth W. Orenstein is a freelance medical writer in Northampton, Pa.