Progression of Mild Cognitive Impairment Visualized on PET

Caroline Helwick

June 17, 2008

June 17, 2008 (New Orleans, Louisiana) — The use of molecular imagining might help clinicians monitor the progression of mild cognitive impairment (MCI) and help to identify people at risk for Alzheimer's disease (AD), according to a study presented here at the Society of Nuclear Medicine 2008 Annual Meeting.

"We can now visualize in vivo changes in the brain that correlate with clinical testing. Not all patients with MCI progress to AD. We think we will eventually be able to determine which persons will," Chester Mathis, PhD, professor of radiology at the University of Pittsburgh, in Pennsylvania, told Medscape Radiology.

He and his colleagues evaluated molecular changes in the brains of people with MCI using positron-emission tomography (PET) and the Pittsburgh compound-B (PiB) radiotracer. PiB binds to the beta-amyloid plaques believed to be responsible for the development of AD. PiB levels above a cut-off threshold of 1.44 to 1.48 distribution volume ratio, depending on the brain region affected, were considered to be elevated (that is, PiB positive).

The longitudinal study included 23 people diagnosed with MCI (mean age, 70 years). The study also included 5 mild- to moderate-AD patients and 33 elderly controls. Subjects underwent baseline PiB scans and were followed for 2 to 4 years, either clinically or with repeat scans.

"Our hypothesis was that MCI subjects with amyloid plaque (based on PiB elevations) would develop AD, and those without plaque would not," Dr. Mathis said.

At baseline, 13 of 23 (57%) MCI subjects were PiB positive, and 10 of 23 (43%) were PiB negative. Among the 13 PiB-positive patients, 5 progressed to AD. None of the PiB-negative patients progressed; in fact, 6 remained stable and 4 actually reverted to normal. "This was probably because there were other reasons for their MCI, such as depression or drug use, which change over time," he explained.

Amyloid Plaque Burden and AD

"We found that about 60% of the MCI subjects already had plaque loads comparable to AD subjects, whereas about 35% had no detectable plaque," he said. "This means that about 15% of MCI subjects per year converted to AD."

"These are slow changes in the amyloid plaque burden from year to year, barely above the test/retest variability. The changes conform to the hypothesis generated by postmortem studies, which is that the full transition from a situation of very little plaque to substantial plaque occurs over the course of about 10 years. This is exactly what our imaging told us," he said in his presentation.

In the elderly control group, about 25% had significant plaque deposition but were asymptomatic, which indicates future risk for AD. Of the 5 people with established AD (all PiB positive at baseline), only 2 had further increases in PiB levels, indicating that "there may be a ceiling to plaque deposition," he added.

Diagnostic Value of PiB

Daniel H. Silverman, MD, head of the neuronuclear imaging research group at the UCLA School of Medicine, in Los Angeles, California, and moderator of the session, told Medscape Radiology that he has concerns about the diagnostic and prognostic utility of PiB in MCI, as opposed to fluorodeoxyglucose (FDG), which has been found to be highly sensitive, even in young adults. Although there is "overwhelming evidence" that imaging with FDG-PET can accurately identify and assess people with MCI and dementia, the role of PiB is less established, he said.

For one thing, PiB depends on the presence of extracellular amyloid plaque, which can also be observed in cognitively intact people. Furthermore, most PiB studies have not been confirmed with postmortem findings, unlike those with FDG-PET, he pointed out.

"When you try to compare the diagnostic accuracy of PiB and FDG-PET without autopsy confirmation, you are not only on shaky ground, you have buried yourself under the rubble of the ensuing earthquake," he commented, adding that, for prognostic purposes, shorter follow-up, as in this study, is acceptable. In addition, the investigators remained appropriately conservative in terms of their conclusions, he noted.

"So far, the data are insufficient to determine whether this test has prognostic utility, let alone whether it is better than FDG-PET," he said. "But PiB could expedite the development of agents aimed at reducing the amyloid plaque burden, and be useful in monitoring in this setting."

This study was partially funded by GE Healthcare. Dr. Mathis disclosed that he is a coinventor of the PiB technology, which is free to academic investigators but is subject to GE Healthcare approval for commercial use. Dr. Silverman disclosed relationships with GE Healthcare, Siemens, Cardinal Health, and Syntermed. He is coinventor of NeuroQ, which is licensed to Syntermed.

Society of Nuclear Medicine 2008 Annual Meeting: Abstract 139. Presented June 16, 2008.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.