Imaging Agents That Identify Alzheimer’s Disease on the Way

A definitive diagnosis requires a full brain biopsy, but imaging compounds may help identify the deposition of amyloid plaque, which is associated with the disease

Among the most dreaded outcomes of a long life, affecting 15 million people and costing over $200 billion annually in the United States, Alzheimer’s disease (AD) was first described in 1906 by the German psychiatrist and neuropathologist Alois Alzheimer. It is the most common of the degenerative forms of dementia and is a growing problem in the United States as the population ages.

The course of AD is highly variable and can be present in mild forms for long periods of time. In fact, detailed neuropsychological testing can detect mild cognitive impairment years before the patient fulfills a clinical diagnosis. Diagnosis is clinically diagnosed as “probable or possible” AD with cognitive and behavioral assessments. Definitive diagnosis of AD currently requires pathologic examination of brain tissue for the presence of plaques and neurofibrillary tangles.

Risk factors for AD include age (>65 years), heredity, sex (more likely in women), lifestyle, and predisposing conditions of diabetes, hypertension, and hypercholesteremia. The exact cause is unknown, although the disease is highly associated with beta-amyloid plaques (an insoluble fibrous protein) and neurofibrillary tangles in the brain. The most accepted hypothesis is that deposits of either amyloid beta or a related beta-amyloid protein are the basis of the disease. The observation that the gene for amyloid beta precursor protein is on chromosome 21 and those born with three copies of this gene, trisomy 21 (Down syndrome) almost universally develop AD by age 40, support this theory. Beta-amyloid deposition precedes the clinical presentation of AD by years; some experts suggest even decades.

Another strong association with AD (even stronger than beta-amyloid deposition) is the development of neurofibrillary tangles within nerve cell bodies caused by the deposition of a protein called tau. These tau deposits appear to cause the microtubules within the neurons to disintegrate, resulting in malfunction and eventual death of the neurons. The neurofibrillary tangles appear later in the disease process, near the time of clinically apparent dementia.

Developed in 1984, the diagnostic criteria for AD depend on documentation of mental decline. In mid-April 2011, the National Institute on Aging and the Alzheimer’s Association published recommendations to update this criterion. They sought to make the new recommendations flexible enough to be used by both general health care providers and investigators. These recommendations include the use of biomarkers. It is important to note that the core clinical diagnosis of AD must be met in order to diagnose AD; the presence of biomarkers alone cannot predict or diagnose AD with certainty. Another key point is that biomarkers represent a continuum of measurement and that diagnostic labels of “positive” or “negative” require “a line in the sand,” so to speak, in a continuum of possible biologic values. One of the proposed biomarkers is imaging of beta-amyloid in the brain.


The importance of direct imaging of the beta-amyloid led Avid Radiopharmaceuticals to develop an imaging compound to visualize amyloid plaques. Using radioactive fluorine attached to a targeting molecule (an analog of the Pittsburgh compound B), Amyvid binds with high affinity to beta-amyloid plaques. Amyvid has low affinity for CNS and cardiovascular receptors and monoamine transporters.

The FDA is still considering the approval of Amyvid. The clinical development program for florbetapir F18 (the generic name of Amyvid) involved six completed clinical trials involving 496 patients. The last trial, named A07, was considered the pivotal trial. A07 was an open-label, single-arm study designed to determine the relationship between measurements of brain beta-amyloid using the agent and true levels of beta-amyloid measured post mortem (autopsy cohort).

Obviously, this is a very difficult study to perform, as you must wait for the person to die to complete the database. In fact, only 29 of the initial 152 subjects enrolled in the autopsy cohort were included in the final efficacy analysis.


To demonstrate the specificity, florbetapir-PET imaging was also performed in a cohort of 47 subjects unlikely to have — and therefore assumed not to have — brain amyloid plaque based upon their age, normal cognitive exam, and lack of clinical findings and risk factors for AD (including genetic markers for known AD risk factors). This group had a mean age of 26, a median age of 27, and a range of 18-50.

Three radiologists independently reviewed each of the images in a randomized blinded fashion for the autopsy cohort and the control group. The primary clinical endpoint of the Avid studies was correlation between the Amyvid-PET image burden with an autopsy exam of brain tissue for amyloid deposits.

The efficacy data for the 29 subjects in the autopsy cohort who died and had histopathologic examination of their brain demonstrated a statistically significant correlation. But the FDA was critical of the sample size. The FDA was also clear in its analysis that although the “correlation was statistically significant, close exam of the data revealed that there was not an exact match between the amyloid levels established through reads of the PET images and levels determined by histopathology.” This led the FDA to question the sensitivity of the imaging process, something of which managed care companies will presumably take note.

Another concern of the FDA was the considerable inter-reader variability among the radiologists reading the PET images.

Managed care considerations

Alzheimer’s disease is clearly a public health issue. The ability to diagnose AD in its earliest stages would certainly aid in our understanding of it, but given that there are no proven preventive or therapeutic medications, the mere availability of a test to measure brain amyloid levels is likely to receive considerable attention by managed care decision makers.

Coupled with the fact that the presence of amyloid does not mean the subject has AD and does not predict AD, managed care companies are well within their purview to question coverage for this test, should it receive FDA approval.

The National Institute on Aging and the Alzheimer’s Association summarized their feelings by stating that biomarker technology as it exists today may be useful in three circumstances: in investigational studies, in clinical trials, and as an optional clinical tool when deemed appropriate by the clinician.

Hopefully, these tests will progress to the point where they really lead to an improvement in the diagnosis and treatment of AD, the true goal of Tomorrow’s Medicine!

Thomas Morrow, MD, is the immediate past president of the National Association of Managed Care Physicians. He has 24 years of managed care experience at the payer or health plan level. Contact him at
The author is a director in the value-based health department at Genentech. He has had no other industry affiliations in the past three years. The views expressed in Tomorrow’s Medicine are the author’s alone.

MANAGED CARE May 2011. ©MediMedia USA

Our most popular topics on