Lou-Ellen Barkan interviews
Dr. MONY de LEON

LOU-ELLEN BARKAN (LEB): Mony, there are so many exciting advances in the field and I know you have participated in many of these. How did you become interested and involved in Alzheimer’s research?

Mony de Leon (MDL): In 1974, as a City University of New York (Queens College) masters degree student, I worked one year in an anatomy lab in Brooklyn College, and began to consider brain aging and cognition as a field of interest. During an internship working for the City of Nethanya in Israel in the area of social gerontology, I became aware of the great numbers of older people who were unable to take care of themselves. The experts then did not have any good biological or medical explanations for this. These elderly had no recourse other than palliative care and ultimate institutionalization. The causes of dementia were unexplored. I decided to learn about what goes wrong with the brain.

There was a fortuitous development—the invention of the computerized axial tomography (CAT) scan, which became commercially available in the early 1970s. In New York, the first CAT scans organized for dementia evaluation were at NYU. The scans were obtained for their written clinical reports and used to screen subjects for clinical trials in order to rule out strokes and tumors and other nasty but identifiable things that happen to the brain. In 1976, as a Columbia University student, I contacted Drs. Steven Ferris and Sam Gershon of the NYU group and they very graciously gave me access to the x-ray films.

So, I started my career as a volunteer at NYU, making brain measurements and developing observation protocols so as to learn if it was possible to discriminate between people with and without a cognitive problem. In 1979, I had the opportunity to publish in The Lancet, the first paper demonstrating cortical atrophy in addition to ventricular enlargement in senile dementia of the Alzheimer type.

In 1980, I completed my doctoral studies at Columbia University and was offered postdoctoral fellowship training at NYU with mentoring in both Psychiatry and Radiology. To my delight, another great opportunity emerged. The first of five NIH Centers were funded in 1980 to develop and evaluate a new technique called positron emission tomography (PET). Thus, in collaboration with Drs. Steven Ferris, Barry Reisberg and Ajax George, a 27-year active collaboration was started between NYU faculty and the Brookhaven National Laboratory team led by Drs. Alfred Wolf, David Christman, and Joanna Fowler. As a post-doc, I became responsible for conducting the Alzheimer PET scan studies at Brookhaven. This is a role I hold today, currently with the able collaboration of Drs. Susan DeSanti, Henry Rusinek, Wai Tsui, Lisa Mosconi, and Joanna Fowler.

As with many new technologies, there was a tremendous initial growth period that focused on the use of the 18F-fluorodeoxyglucose (FDG), a radio-tracer for glucose metabolism with a PET camera, and the interpretation of the results. We spent the first years developing image analysis protocols and as the machines evolved,we refocused our experiments to new brain regions. Over the years, a body of work accumulated. In 1980, we wrote the first report on PET data in senile dementia. We used structural CAT images to map and to sample the metabolic image data and found widespread brain pathology.

While we didn’t realize it at the time, it was unusual for one organization to have interactive research projects using both CAT scans and PET scans. By 1982, both the structural and metabolic lines of work at NYU were funded by the NIH. We pursued both these complementary research objectives for many years examining white matter pathology, patterns of brain atrophic changes, and hormonal and genetic influences on brain metabolism. The advent of the new MRI technology compelled our moving the research from CAT to MRI scans about 1985. Our sustained emphasis has always been on developing improved and earlier diagnostic strategies for senile dementia, which was to be later called Alzheimer’s disease (AD).

MRI was instrumental in our figuring out a piece of a puzzle that enabled us to improve the early diagnosis. Based on the available pathology, we hypothesized that in the earliest stages of AD there is a brain structure (the hippocampus) that could be uniquely affected. Our idea was correct, using a special scanning technique we developed for CAT scans and anatomically validated on the MRI, the data showed that the hippocampus could be imaged and evaluated during life.

Today, after nearly 30 years, using even better tools, we are now asking:What is normality? How does one define what a normal and healthy older brain should look like? What are the causes of the anatomical and metabolic changes that occur with increasing age, and how do they differ from AD? Are there even earlier features of Alzheimer’s disease to be detected?

LEB: People ask us all the time: do I have Alzheimer’s or is it another form of dementia? Pick’s disease? Fronto-temporal dementia? Can you distinguish these after a diagnosis of dementia or before someone shows the signs of the disease?

MDL: That’s a very important question. The best time for doctors to accurately diagnose diseases is after they have enough information to be confident. But, the best time for the patient is well in advance of the symptoms, so that ideally something can be done to preempt the adverse process. Unfortunately, the capacity for an accurate and early diagnosis is at present a work in progress.

Diseases like Fronto-temporal dementia (FTD) and Lewy Body dementia (LBD) are difficult to evaluate unless there is the full-blown appearance of the expected behavioral symptoms. The early features are easily mistaken for other conditions, and to make matters more complex there can be overlapping pathology, such that one patient may have both Lewy Body dementia and Alzheimer’s. The current reliance on behavioral symptoms virtually rules out an early diagnosis.

Using scanners to differentiate degenerative diseases of the brain offers promise, but not without problems. The MRI and PET machines provide a brain map. The fundamental basic observation for structural MRI imaging is the water bound in tissues as contrasted with the free water that pools in brain cavities. Brain atrophy is defined by a loss of tissue that is reflected in decreased tissue volume and increased water accumulation in the brain’s cavities and sulci (spaces between the folds).

With MRI, the objective in the imaging diagnostic examination of patients with cognitive deficits is to identify other possible specific diseases such as stroke and tumors that account for symptoms and may be treatable. Failing to find such focal causal lesions, MRI is then used to describe the different spatial distributions of atrophic changes that occur with degenerative diseases such as FTD or LBD, against the background of aging. That is to say, the MRI is used to provide a description of the distribution and severity of the atrophy. This provides clues for the diagnostic evaluation, i.e.—in FTD there is more atrophy in the frontal and temporal lobes than in the parietal and occipital lobes, and in AD more atrophy is found in the hippocampus and temporal and parietal lobes. Clinicians look for this type of MRI evidence in support of the clinical observations. However, like the behavioral changes, the atrophy patterns afforded by MRI in the early stages of degenerative diseases are not very useful for distinguishing between AD, FTD and LBD degenerative causes.

Recently,PET was approved by the FDA for the purpose of distinguishing between FTD and Alzheimer’s disease. However, the utility of the technique is still undergoing testing, more research is especially needed for the early stages of disease. PET diagnoses, like MRI, rely on patterns of brain change in glucose metabolism, the brain’s primary fuel. But, alterations in metabolism are typically more sensitive to disease than structural atrophic changes. Thus, PET is considered a more sensitive technique for detecting degenerative diseases like AD, PD, and FTD than is MRI. Basically, any condition that affects fuel consumption will influence the outcome of the PET study.

In the diagnostic process using PET, clinicians examine the brain for metabolic compromise in the same regions that are studied in MRI. With the PET there is the additional advantage of diagnosing possible LBD, by its involvement of the occipital lobe, a feature not found with MRI. However, reliance on such patterns is not without error as there are other diseases and conditions that could also affect that anatomy. Nevertheless, FDG-PET has been approved for reimbursement through Medicare, and within limits, is available in certain cases for the differential diagnosis between FTD and AD.

Overall, neither brain atrophy, as reflected in MRI water distributions, nor FDG-PET glucose metabolism alterations are unique features of any one disease. They are secondary markers of damage and, as such, an index measure of the magnitude of damage. Having said that, both MRI and the PET are sensitive to neurodegenerative brain diseases and valuable contributors to the clinical diagnostic process. With imaging of degenerative diseases clinicians can point to an anatomical locus of a problem, however, the cause will not necessarily be ascertained.

Therein lies the basis of the more recent work that we’ve been doing; trying to improve the specificity of the descriptions by studying measures that are directly related to the pathology of AD. About 10 years ago, we started looking at cerebrospinal fluid (CSF) for various proteins and molecules that are coming from the brain and which could deliver clues about the biological identity of cognitive decline. Today at NYU, with my colleagues Drs. Les Saint Louis, Kenneth Rich, Miroslaw Brys, Lidia Glodzik, Remi Switalski, and Martin Sadowski, we routinely examine the CSF for diagnosis.

The spinal fluid, which acts as a waste disposal system, collects the waste and gets rid of it via the venous blood. By evaluating the CSF one gets evidence for both the normal and abnormal functions that are going on in the brain. The brain discards unused and degraded products and clears them out via CSF and blood. The Alzheimer’s disease brain is characterized by abnormalities in neuronal tau proteins which become hyperphosphorylated tau proteins, causing tangles and excess amyloid beta peptide accumulating into amyloid plaques. Biological evidence for these events is also cleared out via the CSF and can be detected. Therefore, one can measure both the by-products of the good and bad guys, and even get some direct evidence for the bad guys themselves. This potential has brought the diagnosis field closer to making specific diagnoses for Alzheimer’s disease.

LEB: You are describing a procedure that confirms a diagnosis once there is an indication that someone is ill. This is very different from a genetic scan that can predict what is likely to happen as I age.

MDL: MRI and PET scans are not scanning the genome. They are looking for existing abnormalities, that is, evidence that the suspected pathology is already underway. You are capturing evidence for change. But the question remains, how early can one predict clinical symptoms? Genetic evaluation has revealed several early onset genes for AD that enable prediction of disease and even age of onset. Nevertheless, there are no predictive genes for the much more common late-onset AD. Consequently, the only prediction we have for late-onset AD is that provided by detecting a process already started. The evidence suggests that one can sustain many years of damage prior to the symptoms of AD. The brain can tolerate a lot of damage given its reserve and its redundant organizational features before symptoms appear.

LEB: Are you finding indications earlier—before symptoms become apparent to clinicians?

MDL: Yes, we are finding very early evidence for pathology. Our efforts are what is referred to as pre-symptomatic evaluation. Today we are evaluating normal volunteers for evidence of degenerative brain diseases well before they are clinically recognizable to their doctors and even to the patients and their families. This is now an important line of work for us—one in which we invest the majority of our efforts.

There are three technologies that have demonstrated progress in this area. Foremost are the two prime scanning modalities I mentioned earlier: the MRI for brain structural change and the FDG PET for glucose metabolism reductions. With these scanning modalities, the hippocampus and related hippocampal formation areas are central in the evaluation. Both techniques reveal early damage and both predict future vulnerability. They are telling you key brain regions are already starting to change before the person shows behavioral signs.

The third effective modality for disease detection is cerebrospinal fluid (CSF) evaluation. The CSF biological markers that have been most informative are markers that relate to tau and amyloid proteins,which are normal proteins until they become misfolded through mechanisms that are not known. Then like the protein in cooked eggs, they are irrevocably altered. These proteins in both the normal and abnormal states can be detected in the CSF. Very recently there have been reports that some AD-related information may also be available from the blood. These exciting possibilities are very preliminary and await replication testing, which our laboratory is undertaking.

Together, the three lines of information— the structural imaging, the metabolic imaging, and the biochemistry—represent the vanguard of research into early diagnosis. For discussion purposes, let’s say a person takes an MRI and suspicion of disease is raised. If there were an effective therapy—particularly one that is risky—before being prescribed, one would want to make sure that the person was clearly at risk. In this scenario, perhaps a second test, possibly one even more demanding such as CSF collection or more expensive such as PET could be used to complement the MRI in order to provide additional and confirmatory information. With two or more tests pointing in the same direction, we have seen that the probability is increased that the person is destined to develop AD. To date, the evidence strongly points to a future where the disease is recognized early across multiple modalities and hopefully treated early.

LEB: Some of our clients who have the disease in the family—particularly the early-onset form of the disease—don’t want to be tested. They don’t want to know, because as far as they are concerned, there’s nothing that they can do about it.

MDL: That is a legitimate and commonly asked question. What is the advantage for one to know? If, for example, one knew,would he or she torment herself every time she forgets the name of somebody at a cocktail party? Why would he/she want to have that reflex built into their daily life? Knowledge from testing doesn’t make a lot of sense for the individual, unless there are appropriate courses of action. Aside from getting one’s estate in order, if one could participate in an NIH or industry-sponsored prevention trial that would be justification for some. Currently, prevention trials are not available, but they are in planning stages.

Genetic information pointing to a mutation causing AD that would be likely expressed by a relatively certain age would be a terrible thing to learn and would warrant psychological counseling. However, testing without the individual learning the results can still be of benefit to others and to future generations that will be affected. My suggestion would be for one to stay involved but at a level one is comfortable with. In the final analysis, participation in research is essential for those with symptoms and for those at risk. But it is vital also for those that are currently aging normally. This is where the bulk of Alzheimer’s disease originates.

Part II of our Reflections interview with Dr. Mony de Leon will be in
the Spring 2008 issue.

Dr. Mony de Leon is Professor of Psychiatry at the New York University (NYU) School of Medicine and Scientist at the New York State, Nathan Kline Institute. He received his Gerontology doctorate from Columbia University in 1980. His published doctoral dissertation described, for the first time, the cortical atrophy of Alzheimer’s disease in living patients. For over 25 years, he has continued to develop imaging and biomarker approaches for the early diagnosis of Alzheimer’s disease (AD). Among his qualifications, he has published over 200 papers in AD research and founded both the NYU Neuroimaging Laboratory and the NYU School of Medicine, Center for Brain Health (CBH). The CBH is an interdisciplinary clinical research center, funded by the NIH, with a team of 25 clinicians and scientists. Dr. de Leon is a reviewer and/or editorial board member for 22 journals and has served on both national and international NIH advisory panels in the area of the early diagnosis of AD. In 2006, he was voted “World’s pioneer in the brain imaging of Alzheimer’s disease” at the Alzheimer centennial in Tubingen, Germany.

Among the research highlights of CBH scientists are: 1980, the first report of brain glucose metabolism reductions in AD using FDG-PET; 1989, the first study showing that the transition between Mild Cognitive Impairment (MCI) and AD could be accurately predicted by estimating hippocampal atrophy on CAT scan; 2001, the first report that entorhinal cortex glucose metabolism reductions in normal elderly predicted 3-years in advance, future MCI; 2003, using NYU developed MRI software, the first 3-year MRI prediction of future MCI in normal volunteers; 2005, the advance prediction of MCI was extended to 8-years using automated PET image analysis software developed at NYU; and in 2007 it was shown that cerebrospinal fluid biomarkers contribute to MRI imaging in the 3-year advance prediction of the transition between MCI and AD.

Current objectives entail using the early imaging and biomarker diagnostic capacity to define presymptomatic candidates for Alzheimer prevention studies.

Previous | Next