Correspondence

PET Scanning in Mild Cognitive Impairment

N Engl J Med 2007; 356:1175-1176March 15, 2007

Article

To the Editor:

The report by Small and colleagues (Dec. 21 issue)1 on the use of positron-emission tomography (PET) after injection of 2-(1-{6-[(2-[F-18]fluoroethyl)(methyl)amino]-2-naphthyl} ethylidene) malononitrile (FDDNP) to estimate the risk of Alzheimer's disease in persons with mild cognitive impairment raises the question of the clinical criteria for the diagnosis of this disease. The current criteria of the National Institute of Neurological Disorders and Stroke–Alzheimer's Disease and Related Disorder Association for diagnosing Alzheimer's disease clinically are 20 years old. If FDDNP-PET is validated, it will pose several questions.

First, will the terms “probable” and “possible” Alzheimer's disease become outdated? Will there be a new diagnostic category of “Alzheimer's disease without dementia” for asymptomatic people in whom imaging studies show evidence of plaques and tangles? Second, which diagnostic test or tests are best? Numerous tests — FDDNP-PET, Pittsburgh Compound-B PET, magnetic resonance spectroscopy, functional magnetic resonance imaging, cerebrospinal fluid biomarkers, and computerized memory tests — appear to help in the diagnosis,1,2 but few studies have systematically compared these tests. Third, how do we measure the cost-effectiveness of the markers? Is it fair to require that a marker for the diagnosis of early Alzheimer's disease improve patients' quality of life or long-term outcomes when we do not have drugs to prevent or halt the disease? Fourth, will we start to see drive-through amyloid-scanning centers? Finally, does amyloid even have a causal role in Alzheimer's disease? Regardless of these questions, however, Alois Alzheimer, were he alive today, would probably be pleased with the progress from pathological to in vivo diagnosis1 that inevitably leads to such questions.

P. Murali Doraiswamy, M.D.
Duke University Medical Center, Durham, NC 27710
dorai001@mc.duke.edu

Dr. Doraiswamy reports receiving an honorarium from General Electric and research grants and honoraria from Pfizer, Eisai, Elan, Forest, Novartis, Myriad, Jannsen, GlaxoSmithKline, and Lilly.

2 References

1
Small GW, Kepe V, Ercoli LM, et al. PET of brain amyloid and tau in mild cognitive impairment. N Engl J Med 2006;355:2652-2663
Full Text | Web of Science | Medline

2
Petrella J, Coleman RE, Doraiswamy PM. Neuroimaging and early diagnosis of Alzheimer disease: a look to the future. Radiology 2003;226:315-336[Erratum, Radiology 2003;227:613.]
CrossRef | Web of Science | Medline

To the Editor:

The method described by Small and colleagues could become an important and frequently used diagnostic aid. Accordingly, the safety of using intravenous FDDNP will be of great importance, yet there is little in their report that addresses questions about safety. Were there any apparent side effects? What is the fate of the brain-attached FDDNP? Are the molecular binding sites known? What effects do FDDNP and its metabolites have on the brain? How long does FDDNP remain attached to the targets, tau and amyloid, and does this interaction affect their degradation or metabolism or the integrity of the surrounding neurons and glia? Most important, did the cognitive status of the patients and controls deteriorate or improve during the days, weeks, or months after injection of FDDNP?

Gordon J. Gilbert, M.D.
500 Pasadena Ave. S., St. Petersburg, FL 33707
drgg22@tampabay.rr.com

Author/Editor Response

We appreciate Doraiswamy's speculation that Alois Alzheimer might have been pleased by the progress from pathological to in vivo diagnosis and agree that techniques such as FDDNP-PET could dramatically change our clinical approach to Alzheimer's disease. If a presymptomatic scan were found to predict a favorable response to a disease-modifying treatment, then screening “brain checks” might be used to identify candidates for treatments that would hold brain protein deposits at bay, in order to delay the onset of Alzheimer's disease, just as we use blood cholesterol levels to identify candidates for cholesterol-lowering drugs, in order to stave off stroke or heart disease. Validated surrogate markers do lead to new definitions of disease, such as hypertension or hyperthyroidism, and in the case of neurodegeneration, we may eventually abandon old terminology and technology for diagnosing Alzheimer's disease. We agree that further study will help to elucidate the efficacy and effectiveness of noninvasive techniques such as FDDNP-PET; however, in vivo brain imaging of amyloid or tau deposits may still be useful for diagnosis and for treatment monitoring, even if these proteins are not found to cause Alzheimer's disease.

Gilbert raises questions about safety. Only nanomole quantities of FDDNP are needed for detection by PET (typically labeled with 10 mCi or 300 ng per dose), and estimates of toxicity have confirmed the safety of this compound for these studies. Only four subjects had minor adverse events during PET scanning (two subjects had minor bruising at venipuncture sites, and two had transient headache), and these events did not differ from those observed in other PET studies. Approximately 3% of the injected dose of FDDNP reaches the brain, which is similar to the percentage of the dose of many experimental receptor-labeling probes used with PET. The short radioactive half-lives of these probes and the very low mass administered will most likely make them safe for use in humans. No effects on brain tissue in either animals or humans have been observed with any of these probes. As expected, FDDNP injections have had no known effect on cognitive status during the days, weeks, or months after administration.

Gary W. Small, M.D.
Vladimir Kepe, Ph.D.
Jorge R. Barrio, Ph.D.
David Geffen School of Medicine at UCLA, Los Angeles, CA 90024
gsmall@mednet.ucla.edu

Citing Articles (1)