Subscribe to eTOC

News from the AAN Annual Meeting
Ronald C. Petersen, MD, PhD, FAAN: On Why Experts Are Considering a Non-Clinical Definition of Alzheimer's Disease


In the Wartenberg Lecture at the AAN Annual Meeting, Ronald C. Petersen, MD, PhD, FAAN, provided an overview of new directions in Alzheimer's disease research and an increasing focus on defining the disease by its pathophysiology alone.

BOSTON — With the availability of technology to assess the brain for signs of Alzheimer's disease (AD) — and the understanding that clinical symptoms tend to arise only years after pathological ones — AD might eventually come to be defined by pathophysiology alone, Ronald C. Petersen, MD, PhD, FAAN, professor of neurology at the Mayo Clinic in Rochester, MN, said here in April during the Robert Wartenberg Lecture at the AAN Annual Meeting.

Such a move would be a major shift, but this concept is being discussed, at least informally, among leaders in the field, he said.

“These ideas are being tossed about with regard to defining Alzheimer's disease as the presence of the pathophysiology, plaques and tangles, irrespective of the clinical picture,” Dr. Petersen said. “The clinical spectrum obviously is very important, but that may vary, depending upon the underlying pathology. Again, these are concepts being considered.”

He said that the discussions “are ongoing right now,” while adding from the lectern during his talk that “we had some discussions just yesterday in the Alzheimer's community,” while at the annual meeting.

“The analogy here, of course, is cancer,” Dr. Petersen said. “We talk about the cancer biopsy, diagnostic characteristics of the underlying disease, irrespective of where the clinical progression is at the time....This is just theoretical, but I'm trying to give you a feel for where the thinking is in the field moving forward.”


Regardless of how the disease is officially defined, the landscape of Alzheimer's disease is quickly changing, Dr. Petersen said.

The pathophysiological understanding of the disease using imaging now can be broken down into three main baskets, he said: structural, with magnetic resonance imaging; functional, with fluorodeoxyglucose-positron emission tomography (FDG-PET); and molecular, with testing for amyloid-beta and tau proteins.

“Previously, Alzheimer's disease was defined only definitively at the time of autopsy,” he said. “We now have pretty good measures of looking at these pathologic indices in life. And that has given us a tremendous advantage.”

In the Mayo Clinic Study on Aging — a population-based study of Olmsted County in Minnesota in which more than 5,000 people have undergone a variety of testing over the last 13 years — researchers have tried to better understand how those with no cognitive impairment, those with mild cognitive impairment (MCI), and those with dementia may be able to be grouped based on their biomarkers. The study — led in part by Clifford R. Jack, MD, and David S. Knopman, MD, FAAN, at the Mayo Clinic — has shown the possibility of predicting disease course based on biomarkers detected well before clinical symptom onset. These findings have been published over the last few years.

Researchers are grouping patients' symptom severity and seeing where they fall based on their permutations of pathophysiological markers — whether a combination of amyloid-positive and neurodegeneration-negative, or amyloid-positive and neurodegeneration-positive, and so on. They are now adding tau protein imaging in how they characterize this combination of markers.


DR. RONALD C. PETERSEN: “Previously, Alzheimers disease was defined only definitively at the time of autopsy. We now have pretty good measures of looking at these pathologic indices in life. And that has given us a tremendous advantage.”

About 2,800 people have under FDG-PET and amyloid scans, and 1,100 have undergone PET for tau proteins, Dr. Petersen said.

Among those with no clinical symptoms, researchers have found that the prevalence of the group that is both amyloid-negative and neurodegeneration-negative stays very close to 100 percent through their 50s and early 60s, but then begins a sharp decline as people hit their mid-60s.

Dr. Petersen was quick to point out a finding that points to the challenge of ascertaining the cause of clinical dementia symptoms: Suspected non-Alzheimer's disease-pathology (SNAP) — which was unanticipated at the outset of the study — is seen in more than 20 percent of people in the general population.


The field is also undergoing a reconsideration of mild cognitive impairment (MCI), the clinical stage between pre-clinical and full-blown dementia. The AAN is sponsoring an evidence-based medicine review of the literature, asking, “What is the prevalence of MCI? What does it mean for progression? Can we do anything about it?” Dr. Petersen said.

But as AD comes to be defined and diagnosed, it will have to be considered only one part of a constellation of factors that can contribute to dementia symptoms, Dr. Petersen said. “So, any particular individual likely has that whole array of complementing pathologies contributing to the clinical spectrum.”

The diagnostic and prognostic power of Alzheimer's disease will grow even stronger — though more complicated as well — with greater understanding of the proteins alpha-synuclein, TAR DNA-binding protein 43, as well as vascular disease, he said.

“I think we'll be able to, in the future, characterize a person with a profile of underlying biomarkers,” Dr. Petersen said. “Why? Because we'd like to think that the therapies will be specific for the underlying pathologies. So, we may develop amyloid therapies, hopefully sooner than later, as well as tau therapies. And certainly therapies for these other proteinopathies as well.”


Commenting on the lecture by Dr. Petersen, Anne Fagan, PhD, professor of neurology at the Hope Center for Neurological Diseases at Washington University and investigator at the Washington University Alzheimer's Disease Research Center, said: “As AD researchers, we have been thinking about this for years, especially those [of us] in the biomarker trenches,” she said. “And this is a huge issue.”

It would help distinguish AD from the many other potential causes of cognitive impairment symptoms, said Dr. Fagan, who has spent two decades researching the biomarkers of AD, and supports moving toward more reliance on the pathophysiology in making a diagnosis.

“You could have somebody who is taking multiple drugs and has a drug interaction that impairs cognition,” she said. “You could have somebody who's depressed. That will impact their cognition and will have absolutely nothing to do with Alzheimer's disease. In that regard, having biomarkers through the cerebrospinal fluid [CSF] or imaging is going to be very useful for sorting these things out.”

One complication in moving toward more reliance on pathophysiology in the definition of AD, she said, has to do with unreasonable fears of lumbar puncture for collecting CSF. She said concerns over the risk of paralysis and the anxiety over pain from the punctures can sometimes be overblown, especially among primary care providers and those outside the neurology community.

“I think there needs to be better overall education,” she said. “There's a lot of misinformation out there that needs to be dispelled.”

Perhaps even more important, there is no effective treatment for AD even if pathophysiology of the disease is found.

“In the very early symptomatic stages and definitely in the asymptomatic stage, say that somebody comes in and they're found to be biomarker-positive,” she said. “The first thing they're going to ask is: ‘OK, so when are my symptoms going to start? When am I going to start putting bread in the freezer instead of the bread box? Every time they misplace the keys, they're freaking out. Is this a public service?”

The approach to diagnosis is likely to change more dramatically once more effective treatments are developed, because there will be more urgency to detect the disease and begin treatment earlier, she said.

“Once a disease-modifying therapy comes on the market, I think you're going to see a humongous change.”


In Practice Current, Neurology® Clinical Practice (NCP) invites neurologists worldwide to explore controversial clinical topics that do not have sufficient diagnostic or therapeutic evidence. Readers can share their practice with fast online surveys on “hot topics,” see real-time results displayed on an interactive world map, compare their practice with peers, and access opinions from internationally-recognized experts.

The current survey on treatment of epilepsy in pregnancy is available at: Neurologists and epileptologists, including trainees, and advanced practice providers are encouraged to participate.

A commentary with concise background and available evidence on the topic will be published in the August 2017 issue of NCP. Three experts from diverse backgrounds will weigh in what they do in their practice: Augustina Charway-Felli, MD, PhD (Ghana), Cynthia L. Harden, MD (United States), and Torbjörn Tomson, MD, PhD (Sweden).

New in 2017, Practice Current is featured every four months in the Neurology® podcast. Section Editor, Luca Bartolini, MD brings experts to a forum and allows a moderated discussion on the results of a Practice Current survey. The June 1 podcast discussing antiepileptic drug discontinuation for epilepsy in remission features Gregory D. Cascino, MD, FAAN (United States), Ley Sander, MD, PhD, FRCP (United Kingdom), and Ali A. Asadi-Pooya, MD (Iran). The second podcast on treatment of neuromyelitis optica will be published in October 2017.


• Sign up to watch the video of the Wartenberg lecture.
    • Knopman DS, Jack CR Jr., Wiste HJ, et al. 18F-fluorodeoxyglucose positron emission tomography, aging, and apolipoprotein E genotype in cognitively normal persons Neurobiol Aging 2014; 35(9):2096–2108.
    • Vermuri P, Lowe VJ, Knopman DS. Tau-PET uptake: Regional variation in average SUVR and impact of amyloid deposition Alzheimers Demen (Amst) 2016;6:21–30.