Can MRI Predict Alzheimer's?

Mild cognitive impairment (MCI) is a term that is used to describe the loss of memory, language or other mental functions. It is a descriptive term, based on a patient’s symptoms, and not an actual diagnosis. Some patients with MCI will remain at the same level of cognition for many years. A subset of patients with MCI, however, go on to develop Alzheimer’s disease (AD), a relentlessly and irreversibly progressive disorder in which vulnerable brain cells lose their normal function and die. The loss of brain cells in AD ultimately leaves a patient unable to carry out even the simplest tasks. MCI thus represents a period of significant uncertainty and fear for patients and their families, during which there is no way to determine the likelihood of progressing to dementia.

A number of research groups around the country have focused efforts on developing reliable methods for assessing the risk that an individual patient with MCI has for developing AD. Magnetic resonance imaging (MRI), currently the premier method for imaging the brain’s anatomy and function, has traditionally fallen short in this role. Why? AD differs significantly from other neurological disorders like brain tumors, multiple sclerosis and stroke on MRI. The latter disorders result in changes in the normal distribution of water in the brain; the disruptions are of a sufficient magnitude to alter the appearance of the brain on MRI. The death of neurons in AD, in contrast, is an insidious process that takes place over many years. Changes on MRI images in AD, seen only as regional atrophy of the brain, are not reliably detected by visual inspection until late in the disease process.

The Alzheimer’s Disease Neuroimaging Initiative (ADNI) – a multi-site collaborative effort launched in 2003 by the NIH, the FDA, private pharmaceutical companies and non-profit organizations – is the largest trial to date for evaluating the potential role of MRI in the diagnosis of AD. The UCSF Department of Radiology is the hub of operations for this effort, guided by principal investigator Michael Weiner, M.D. Using data from ADNI, a number of groups* have now shown that quantitative assessment of the brain’s structure – measurements of the thickness or volume of specific brain regions derived using sophisticated computer algorithms – can be used to stratify the risk that a patient with MCI has for going on to develop AD. Radiologists at UCSF are currently working together with experts from the UCSF Memory and Aging Center to evaluate the role for using quantitative MRI approaches in the evaluation of patients presenting with cognitive complaints.

*[1] Alzheimer disease: quantitative structural neuroimaging for detection and prediction of clinical and structural changes in mild cognitive impairment. McEvoy LK, Fennema-Notestine C, Roddey JC, Hagler DJ Jr, Holland D, Karow DS, Pung CJ, Brewer JB, Dale AM; Alzheimer's Disease Neuroimaging Initiative. Radiology 2009

*[2] Automated MRI measures predict progression to Alzheimer's disease. Desikan RS, Cabral HJ, Settecase F, Hess CP, Dillon WP, Glastonbury CM, Weiner MW, Schmansky NJ, Salat DH, Fischl B; Alzheimer's Disease Neuroimaging Initiative. Neurobiol Aging 2010