Activated Microglia Imaged in Alzheimer Brain

Abstract & Commentary

Source: Cagnin A, et al. Lancet. 2001;358:461-467.

A radio-labelled marker of microglial activation has been found to bind more extensively to the brains of Alzheimer patients than age-matched normals, and could provide a means of monitoring brain inflammation in patients with mild cognitive impairment and dementia. Cagnin and colleagues used positron emission tomography (PET) to quantitate the distribution of carbon-11-labelled PK11195 in the brains of 15 normal patients, 8 Alzheimer’s disease (AD) patients, and 1 with mild cognitive impairment (MCI). They used PK11195 for this purpose, a ligand for the peripheral benzodiazepine receptor that binds to brain macrophages and activated microglia. Binding of PK11195 has been shown to increase in the aftermath of stroke and other processes that lead to the activation of microglia and the passage of peripheral macrophages into the brain. Its imagability with PET makes it suitable for measuring active inflammation in the living brain on a regional basis.

Normal subjects, ranging in age from 32-80, showed no significant increases in PK11195 binding, except in the thalamus. Cagnin et al speculate that this constitutive binding could be explained by the proximity of the thalamus to areas in which the blood-brain barrier is leaky. In contrast, Alzheimer patients showed increases in multiple brain areas relative to normals. Significant elevations were found in the entorhinal cortex, temporoparietal area, and cingulated cortex. The AD patients in this investigation were also studied using serial volumetric MRI imaging to quantitate regional brain atrophy. Areas that showed the most significant increases in PK11195 binding were also the regions with the highest rates of brain atrophy. This suggests that the inflammatory markers may reflect active areas of neurodegeneration.

One subject with MCI was examined in this study. This patient showed normal brain glucose metabolism on 18-FDG PET and did not have evidence of significant cognitive decline over a 23-month follow-up period. However, serial MRIs demonstrated atrophy in inferior and medial temporal areas that was predicted by the pattern of PK11195 binding when the patient was initially studied. Although this patient was not documented to develop frank AD in the course of the study, the findings suggest that PK11195 binding may increase during the preclinical stages of AD and could be an early marker of neurodegenerative changes.


Activated microglia play an important role in the clearance of amyloid deposits in the brain, and may contribute to AD pathology by mediating the conversion of diffuse amyloid deposits into neuritic plaques. A plethora of markers of inflammation have been found in the brains of AD patients postmortem, and inflammatory mechanisms have been targeted in several clinical trials aimed at preventing or slowing the course of AD. The in vivo imaging technique used in this study provides a novel window on the inflammatory component of AD pathology, and could prove extremely useful in the development of anti-inflammatory therapies and preventatives for AD in the future.

The finding that inflammatory changes in the brains of living AD patients correlated with subsequent brain atrophy could be interpreted as evidence that microglial activation is a harbinger of subsequent neural degeneration. However, in direct analogy to stroke, this marker of inflammation could reflect microglia clearance of cellular debris following neuronal death rather than active causation of neuronal pathology. The issues of whether inflammation is part of the problem or part of the solution in AD remains to be resolved.

Epidemiologic evidence suggests that anti-inflammatory medications may decrease the risk of AD, but do not provide evidence that they alter the course of existing dementia. It is currently uncertain whether the concept of treating demented patients with anti-inflammatory medication represents "too little, too late." Recent attempts to use conventional anti-inflammatory treatments in mild-to-moderate AD patients have yielded discouraging results. Low-dose prednisone failed to improve cognition in AD patients, and more recently, the antimalarial anti-inflammatory hydroxychloroquine failed to slow the rate of decline in mild AD (Van Gool WA, et al. Lancet. 2001;358:455-460). Preliminary studies of immunotherapy for AD have yet to establish the effectiveness of this approach in humans or establish whether peripheral immunization exerts its effects via the activation of brains microglia. At the present time, the use of anti-inflammatories such as NSAIDs in the treatment of AD patients is not recommended, although further work in this area is clearly justified. —Norman R. Relkin

Relkin, MD, PhD, Associate Professor of Clinical Neurology and Neuroscience, New York Presbyterian Hospital-Cornell Campus, is Assistant Editor of Neurology Alert.