Bret S. Stetka, MD


June 24, 2013

In This Article

Background and Biomarkers

AD is an epidemic, affecting over 5 million Americans and nearly 40 million people worldwide. AD afflicts 1 in 8 people aged 65 years or older and nearly half of those 85 years and older.[2] The reason for the staggering prevalence is simply that we're living longer. “The major risk factor for AD is advanced age,” noted Small. According to the Centers for Disease Control and Prevention, life expectancy in 1900 in the United States was around 47 years; in 2013, it's nearly 80 years.[3,4]

"So what is Alzheimer's disease?” asked Small rhetorically, before a brief history lesson. In 1906, German psychiatrist and neuropathologist Alois Alzheimer presented the first case of the condition that would bear his name. His initial patient died 4 years after her symptoms began, and on autopsy her brain contained the waxy protein fragments and twisted fibers now known to be amyloid plaque and tau tangle protein accumulations. Assumed to be a rare form of dementia, it wasn't until decades later that more progress was made. A 1968 paper[5] by Blessed, Tomlinson, and Roth correlated plaques and tangles with "senility," pathologizing cognitive dysfunction previously thought to be a normal part of aging and igniting AD awareness.

Diagnosing and monitoring AD initially proved tricky. The AD brain exhibits gross atrophy and prominent collections of plaques and tangles. But the "normal" brain can too, in lower concentrations. Plaques and tangles build up gradually as we age. Moreover, until recently, detecting such changes was difficult in living patients. Over the years, numerous potential AD biomarkers have been considered with varying degrees of success. Serum, blood, and cerebrospinal fluid (CSF) assays have proved useful, particularly in research settings, as have genomics, vascular risk factor assessments, and neuroimaging. The American Academy of Neurology now recommends a CT or MRI scan in cases of suspected AD to rule out other causes of impaired cognition such as stroke or tumor, as findings associated with AD such as generalized atrophy can be nonspecific. However, structural imaging[6,7] can contribute to AD identification. The hippocampus undergoes significant atrophy as dementia progresses, and though hippocampal volume reduction cannot confirm AD in an individual, it can distinguish AD in an aggregate patient sample.

Better yet, the application of PET scan technology in AD research has advanced the functional characterization of the AD-afflicted brain. Work by Small, Mosconi, and others[8,9,10] using PET with a radiolabeled glucose analogue (fluorodeoxyglucose [FDG]-PET) helped correlate specific patterns of glucose metabolism with different causes of dementia. These findings were, in 2004, enough to convince the Centers for Medicare & Medicaid Services that PET was useful in dementia diagnosis, making reimbursement possible. Other radio-tracers have since been developed that allow for the visualization of specific protein targets in the brain, namely Pittsburgh compound B (PiB) and the recently approved florbetapir, which bind to amyloid, and FDDNP, which binds to both amyloid and tau. FDDNP studies have correlated cognitive decline with the degree and location of amyloid and tau accumulation; amyloid preferentially accumulates in the lateral temporal region with tau more prominent in the medial temporal lobes.

Combining diagnostic markers may prove the most useful approach to AD diagnosis. A 2013 study[11] by Prestia and colleagues assessed for hippocampal atrophy, decreased CSF amyloid, and decreased brain glucose metabolism in 73 patients with mild cognitive impairment. Among those with no positive biomarkers, just 4% went on to develop AD; in patients positive for all 3, 100% ultimately progressed to AD. Still, while AD biomarker profiles are increasingly common in academic and research settings, the limited efficacy of current therapies has largely kept them from the clinic. At the moment, a more accurate AD diagnosis provides limited benefits to the patient.

"People come in wanting these fancy scans, and we can do that," noted Small. "But the question is: 'How is this going to change the course of treatment?' We need to find biomarkers that predict treatment responses, but we're not quite there yet."


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.
Post as: