Spring 2011

New Biomarkers for Alzheimer’s Disease

By Norbert Myslinski, PhD
Aging Well
Vol. 4 No. 2 P. 6

Recent research reported at the Society for Neuroscience Convention in November reveals new biomarkers for Alzheimer’s disease (AD), which is important because biomarkers have the potential of providing an early AD diagnosis, long before the cognitive symptoms become apparent and irreversible.

In the development of AD, there are initial changes in the brain without any symptoms. This advances to an intermediary condition known as amnestic mild cognitive impairment (AMCI) involving subtle signs. About one-half of those with AMCI develop dementia within five years. Currently, there is no method for differentiating AD from other forms of dementia except by the process of elimination. Extensive testing is needed to rule out other more treatable causes such as vascular disease, Parkinson’s disease, vitamin B12 deficiency, and depression. A diagnosis of “probable Alzheimer’s” is all that is possible until an autopsy reveals a buildup of amyloid plaques and neurofibrillary tangles in the brain.

There are no easy and reliable tests for AD biomarkers such as amyloid or tau that are suitable for the clinical setting. Investigators continue to search for such tests because disease-modifying drugs are believed to be most effective early in the disease process. Currently there are two promising ways to measure the amyloid and tau biomarkers: fluid sampling and brain imaging.

Blood sampling is easy but not very reliable. Cerebrospinal fluid from lumbar punctures produces more consistent results. Several studies have suggested that measuring three spinal fluid proteins may provide more accuracy. AD patients tend to exhibit a pattern consisting of low beta-amyloid-42, high total tau, and elevated phosphorylated tau. One study involving 1,500 subjects conducted at 12 sites in Europe and the United States followed patients with AMCI who progressed to AD. They found that 83% had the three-protein signature. Another study conducted by the Alzheimer’s Disease Neuroimaging Initiative found the three-protein signature in 90% of the AD subjects, 72% of the AMCI subjects, and 36% of the normal subjects.

Brain imaging is most promising when an agent known as Pittsburgh Compound-B is used to enhance images produced by PET scans. The compound binds to amyloid deposits in the brain, making them more visible on the scans. However, it has a short 20-minute half-life that restricts its use to research centers able to synthesize the compound on site.

A new ligand known as florbetapir (Amyvid) lasts longer and thus would allow this method to be used more widely by physicians and researchers. Christopher Clark of Eli Lilly presented data in a January article in The Journal of the American Medical Association that shows florbetapir retention correlates well with actual beta-amyloid deposits in the brain. He used florbetapir to visualize amyloid deposits in 29 volunteers near the end of their lives and then compared the image results to immunohistochemistry and silver stains performed at autopsy. Of the 15 who showed AD pathology at death, 14 were positive for AD by the florbetapir scan. The 14 without AD were negative for florbetapir.

Looking Forward
In late January, the FDA’s Peripheral and Central Nervous System Drugs Advisory Committee voted unanimously to recommend approval of florbetapir on the condition that Eli Lilly implements a reader training program that demonstrates reader accuracy and consistency. The committee indicated that more data are needed to show that the scans are accurate and beneficial. A negative scan with this tool would be helpful to clinicians in ruling out AD as a cause of dementia. However, a positive scan does not necessarily indicate AD because a patient can have amyloid buildup without having AD.

Brain imaging showing structural changes in the brain also holds the potential for the early detection of AD. At the Society for Neuroscience convention, Sarah Madsen, PhD, of UCLA’s Laboratory of Neuro Imaging reported that the caudate nucleus was significantly smaller in AD patients. The major function of the caudate is considered to be motor control, so this finding suggests that AD produces broader damage in the brain than previously thought. These results complement those of other studies demonstrating that AD patients have a smaller hippocampus, a part of the brain important for converting short-term memories to long.

Sarah George, a neuroscientist at Rush University, presented evidence of the thinning of certain cortical areas in incipient AD. She studied the substantia innominata (SI) containing cholinergic neurons of the nucleus basalis of Meynert, which innervate the entire cortical mantel and degenerate during the progression of AD. Her 47 subjects were all diagnosed with AMCI and were at a high risk of developing AD. In this longitudinal study, she separated her subjects into those who progressed to a diagnosis of AD (converters) and those who remained stable (nonconverters).

All subjects were followed annually with clinical evaluations and MRI scans. Over time, 22 converted to AD (mean age of 80), and 25 remained stable and did not convert (mean age of 77). Among converters, SI volume was found to be positively correlated with the thickness of the right inferior parietal lobe, the right superior frontal lobe, and the right precuneous. Cortical thinning was not correlated with SI volume in nonconverters. Therefore, SI’s cortical projection sites exhibited cortical thinning in incipient AD. However, SI volume did not differ between converters and nonconverters. This suggests that cortical thinning precedes SI volumetric changes. Therefore, MRI screening of certain cortical areas seems to be a strong candidate for an early AD biomarker.

All of these diagnostic procedures have the potential of helping the 40% of the United States population over the age of 85 who have probable AD. Worldwide, AD costs about $500 million annually in care and treatment expenses. It is anticipated that by 2050, there will be 80 million patients with AD.

Pharmacologic Prospects
Although there are drugs that alleviate AD symptoms in some patients, there is no cure for AD, nor any way to slow its progress. Most clinically approved AD drugs, such as  donepezil, galantamine, and rivastigmine, are cholinesterase inhibitors that only stimulate remaining acetylcholine neurons that are preferentially destroyed in AD. These drugs work in fewer than one-half of the people tested, showing modest and only temporary effects. Most compounds under investigation have failed in late-stage clinical trials. In August 2010, Eli Lilly had to stop research on a drug called semagacestat, a gamma-secretase inhibitor that blocks the formation of beta-amyloid. In this clinical trial, the condition of the subjects taking the drug became worse than those taking a placebo.

Despite the failures, research on AD has exploded. About 20 papers on the subject are published each week. Approximately 100 experimental drugs are currently being examined for potential benefits. Nobel Laureate Paul Greengard, PhD, of New York’s Rockefeller University, is working on inhibiting gamma-secretase-activating protein, which should be more selective and produce fewer side effects than gamma secretase inhibitors. Pfizer Inc and Johnson & Johnson are testing a monoclonal antibody called bapineuzumab, which attacks and clears beta-amyloid. Allon Therapeutics Inc is testing davunetide, a drug that reduces damage done by tau in patients with mild cognitive impairment. Charles Glabe, PhD, a professor at the University of California, Irvine, is working on a new vaccine that protects against memory problems associated with AD. Unlike previous AD vaccines that caused dangerous inflammation of the brain, this one uses nonhuman protein that is unlikely to cause the dangerous autoimmune response.

The diagnostic techniques mentioned in this article are used in research but are not yet ready for clinical use. When they are, more effective drugs and other therapies will be necessary to take advantage of the earlier detection.

— Norbert Myslinski, PhD, is a neuroscientist and faculty member in the Dental and Nursing Schools of the University of Maryland in Baltimore and director of the “Physiology of Aging” course. He founded the Baltimore Chapter of the Society for Neuroscience and currently serves as president.