The Rapidly Evolving Front in Alzheimer’s Diagnosis

Imaging technologies may pave the way toward early diagnosis and the development of disease modifying therapies.

With Ronald C. Petersen, MD, PhD, a Q&A

The development and use of biomarkers represents one of the most promising areas of development in Alzheimer’s disease over the last several years. Identifying tau tangles and amyloid plaques in the brain may not only lead to earlier detection of the disease but may also offer clues into its pathology and lead to the development of impactful disease modifying therapies.

Ahead, Ronald Petersen, MD, PhD, Professor of Neurology and Clinical Distinguished Investigator at the Mayo Clinic in Rochester, MN, shares expert perspective on biomarkers and how they may shape the future of Alzheimer’s disease diagnosis and research.

In light of recent innovations in Alzheimer’s disease imaging, how would you assess the current state of assessment and diagnosis?

Key towards understanding the current moment in Alzheimer’s disease research is an appreciation of recent history on disease diagnosis and pathology, according to Dr. Petersen. “Until very recently, Alzheimer’s disease was characterized as a clinical pathologic entity, meaning that people had to fulfill the criteria for the clinical syndrome,” says Dr. Petersen. The criteria were based on cognitive impairment (usually involving memory and other cognitive domains) and sufficient severity to constitute a change in functional activity: If it had a typical gradual, progressive history, most likely that was underlying Alzheimer’s disease. But Dr. Petersen points out that the best clinicians could do was call it “probably” Alzheimer’s disease: “The only way it could be made definite is if the person was followed to death, and plaques and tangles were identified at autopsy.”

That set of criteria guided clinical practice for over 25 years, until 2011 when the National Institute on Aging and the Alzheimer’s Association convened a group of experts to revise the criteria. According to Dr. Petersen, the revision was needed, “given the biologic advancements that had been made over that two and a half decades, and the fact that we now had biomarkers on the horizon that could really characterize this underlying pathology.”

In 2011, the disease was broken into three clinical phases: dementia, mild cognitive impairment, and pre-clinical Alzheimer’s disease. While many individuals were already fairly familiar with the dementia phase, the others stood as new classifications for AD. “Mild cognitive impairment means that people may have had a slight cognitive impairment, usually a memory impairment, but were otherwise functionally preserved.” Many folks with mild cognitive impairment still function normally in the community, but the memory problem is beyond what we would expect for aging and is pathological. Dr. Petersen refers to this phase as a sort of “in-between” state.

The pre-clinical Alzheimer’s disease classification is reserved for people who are cognitively normal but harbor some of the underlying biologic features of Alzheimer’s disease; “That is, with biomarkers that could now determine whether a patient is amyloid-positive by cerebrospinal fluid (CSF) or by amyloid PET scanning,” says Dr. Petersen.


Advances in biomarkers are carving out new paths in the diagnosis of Alzheimer’s disease and in the development of disease modifying therapies. As imaging technology continues to be refined, the clinical skills of neurologists remain the foundation of successful diagnosis and management. Ultimately, Alzheimer’s is just one of the players that influence cognitive function and therefore combination approaches are likely to form the foundation of therapeutic intervention for Alzheimer’s in the years ahead.

Although the distinction between these states appears to run afoul of the more nuanced view of Alzheimer’s disease as a continuum, Dr. Petersen argues that this clinical characterization is very useful for the clinician to help identify the stages and to communicate with patients and then serve as a research background. The classifications also included a variety of qualifiers like intermediate MCI due to Alzheimer’s disease or definite high likelihood due to Alzheimer’s disease if all the biomarkers line up.

Published criteria for each classification gave the clinical characteristics for that stage, and then under it there were combinations of biomarkers, which came in two classes: the presence or absence of amyloid and neurodegeneration. “Here neurodegeneration meant that we could see perhaps atrophy on an MRI scan, either hippocampal atrophy or cortical atrophy, or you could demonstrate a certain pattern of hypometabolism on FDG PET scan,” Dr. Petersen explains. “So you could have either one of those would constitute evidence for neurodegeneration.”

Since the publication of these guidelines, Dr. Petersen points out that further research has clarified that neurodegeneration was perhaps too non-specific—since neurodegeneration is present in other non-Alzheimer’s pathologies—but was nevertheless thought to be an important feature of progressive disease.

Since the new criteria have been published, how have biomarkers amplified or perhaps altered understanding of the clinical pathology of Alzheimer’s disease?

Over the last several years, the biomarker field has progressed even further, according to Dr. Petersen. “Now we’re able to do PET scanning for the tau protein, which comprises the neurofibrillary tangles,” he observes. “This now enables us an opportunity to actually determine whether a living person has the underlying biologic features that define Alzheimer’s disease, namely, the presence of the amyloid plaques and the presence of the tau tangle.” If people have the clinical syndromes and have those features, physicians can be pretty certain that those clinical syndromes are due to underlying Alzheimer’s disease.

With the new imaging technology, physicians do not have to wait until autopsy to confirm Alzheimer’s disease. Some studies have already shown that amyloid presence at the time of autopsy closely match the amyloid seen on the scan, according to Dr. Petersen. He expects the same results in studies looking at tau PET imaging and autopsy as well that are in development right now.


A thorough history-taking can help characterize cognitive impairment as a likely degenerative process, which narrows down the possibilities when it comes to identifying a likely underlying pathology.

In light of these changes, Dr. Petersen points out that another expert panel has been convened to now look at the criteria from 2011 to determine whether they were on track regarding amyloid versus neurodegeneration. The panel hasn’t yet come to any final determination, but Dr. Petersen expects we may see some change as a result. “They’re leaning toward a terminology shift such that Alzheimer’s disease will be classified solely on the basis of the biomarkers independent of any clinical syndrome.”

He notes further that new criteria could incorporate biomarker results for a more refined classification. “If you have amyloid present, you would have what’s called Alzheimer pathophysiology. If you have evidence of amyloid and tau, you would then be labeled as having Alzheimer disease, independent of the clinical syndrome,” says Dr. Petersen. “You could be cognitively normal, you could have mild cognitive impairment, or you could have dementia, but that is characterized separately.” These are the two parallel continua that would help to characterize the clinical progression, but the underlying pathophysiology would determine whether it’s Alzheimer’s disease or not.

Dr. Petersen stresses that much of this remains conjecture until the panel releases its findings. Nevertheless, it offers a broad portrait of how the field has evolved and offers clues about future directions. In terms of where all this work leads, Dr. Petersen is hopeful it may impact the therapeutic spectrum. “As we develop therapies, drugs, vaccines, and antibodies, the thinking is that the sooner we intervene, the better, and that’s where the biomarkers become critically important.”

While the panel will provide useful information about the direction of biomarker use, Dr. Petersen emphasizes that their recommendations will likely not lead to an immediate revision of criteria or guidelines. “We’re still pretty much at the proof of concept stage. We’re really posing a research framework, or a conceptualization of the field. It’s a very good framework, but it’s going to need to be validated before it would then evolve into guidelines or criteria.” Nevertheless, this framework may very well guide the direction of future research, he adds.

How will biomarkers continue to be refined, and what impact do you see them having on the field moving forward?

While the current amyloid biomarkers have great utility right now, particularly F18 tracers, Dr. Petersen sees how they could be refined to a higher degree. “These compounds could be tuned up such that there’s less of non-specific binding or off-target binding, and give you improved demarcation of the amyloid presence in the cortical ribbon,” he notes. Most of the sophistication or the improvement in tracers likely needs to be done on the tau side, according to Dr. Petersen. “Tau PET imaging is relatively new, and most of us use one tracer called AV1451, and it labels 3 and 4 repeat tau, which is what we see in Alzheimer’s disease.” However, it does not label primary tauopathies very well. “Other diseases that have tau as a primary component, like frontotemporal lobar dementia and progressive supranuclear palsy, these are primary tauopathies, and this particular tracer does not label them very well.” He points out that the tracer does a reasonably good job for Alzheimer’s disease, but that it, too, has off-target binding. “It binds at several other structures that can be distracting when you’re looking at an image, and consequently that could certainly be improved upon in white matter and grey matter distinctions, and the like.”

According to Dr. Petersen, the tracer development industry is an active area, and physicians should expect continued development. “There certainly can be improvement so that we’re more specific with regard to positive and negative scans, as well as specificity. We also would like them to be maximally sensitive, and to pick up the very early features of the development in the brain of amyloid and tau.”

Can you discuss the concept of early diagnosis of Alzheimer’s disease, whether it’s feasible, and the impact it might have on treatment?

Assessment tools such as the Montreal Cognitive Assessment (MOCA) don’t specifically characterize somebody with Alzheimer’s disease but instead suggest a possible need for more in-depth evaluation. Therefore, their utility is limited but specific, according to Dr. Petersen. “We still advocate the use of them now because it’s important that people characterize what might be the cause of forgetfulness or cognitives issues,” he says. It can also help answer questions about the severity, whether there is a treatable component of it, and what it means for the future, for planning, financial issues, and helping the family. The real game-changer, though, will be when a disease modifying therapy is developed. “Then it really becomes critical to be able to identify the process as early as possible.”

Amyloid or Tau? Let Cerebrospinal Spinal Fluid Decide

Despite the utility of imaging technologies for the assessment of Alzheimer’s disease, they are not widely used due to their immense cost. On a public health basis, this isn’t practical, says Dr. Petersen. “If a disease modifying treatment were to become available, one can imagine that the FDA and the pharmaceutical company will require that one of the targets of the therapy be present in individuals. If you’ve got an anti-amyloid therapy and the drug was developed using biomarkers to characterize individuals, then the labeling of the drug may require you to demonstrate the presence of amyloid in the person before that person becomes eligible for the drug.” PET scanning is one way to determine that, but it costs thousands of dollars, which may be too burdensome for patients.

While costs may come down in the future, Dr. Petersen notes that there is another way of determining presence of amyloid or tau: Cerebrospinal fluid (CSF). “A spinal tap, lumbar puncture, is relatively inexpensive compared to imaging,” he says. While there may be some discomfort associated with a lumbar puncture, they can be done quite expertly and painlessly. Additionally, they come at a fraction of the cost of imaging procedures. “I’m not sure how you would characterize the cost of a lumbar puncture, but probably in the hundreds of dollars rather than in the thousands of dollars,” says Dr. Petersen. “If, in fact, it becomes necessary to determine whether a person has amyloid or tau in the brain, one could think about a lumbar puncture as opposed to the scan.”

As for the idea of truly diagnosing the disease in the early stages, there are promising directions, but much work remains. “There is work underway looking at other perhaps biological measures: Might there be a blood signal to measure amyloid in the blood? Yes, but it’s not been validated yet.” Similar tests have been developed for tau, but these, too, need to be validated. Another area of interest is the genetic characterization of the disease, suggesting that the field is moving earlier and earlier in terms of diagnosis.

Dr. Petersen emphasizes that the motivation for these technologies would be to address the disease itself therapeutically. “Clearly if a disease-modifying therapy that was shown to be effective and was relatively safe, risk profile was tolerable, and hopefully relatively inexpensive, then you certainly would want to go as early as possible,” he says. “But there are a lot of ‘ifs’ in there that haven’t come to be yet.”

As for the public perception of Alzheimer’s disease, Dr. Petersen believes that it is important to impress upon people that it’s a very gradually progressing entity. “Some of these biologic features of the disease like early deposition of amyloid may happen 10 to 20 years before people become symptomatic,” says Dr. Petersen. “The good news is that there might be a window of time to intervene with a disease-modifying therapy that would prevent the clinical symptoms down the road.” Nevertheless, he says, “the public needs to appreciate that this is a very insidious process that evolves and may play out at different points in the lifespan depending on your particular family.”

In terms of the here and now, what advice can you offer regarding diagnosis of Alzheimer’s disease?

Although improved imaging technology and the growing utility of biomarkers may aid in the assessment and intervention of Alzheimer’s disease, Dr. Petersen believes that the clinical skills of neurologists remain the foundation of successful diagnosis and management. “The main feature of this disease is still cognitive impairment, and that needs to be assessed by the practicing neurologist,” he observes.

He also stresses the importance of continually refining those clinical skills, including and especially history-taking. “It’s still very important to understand the temporal course of these disorders of cognitive impairment,” he explains. “We often like to have the history corroborated by someone who knows the person well.” After talking to the patient, Dr. Petersen suggests asking the caretaker or spouse if they have been noticing the same thing or if the pattern is different or perhaps interpreted differently.

The reason history-taking is so essential, according to Dr. Petersen, is that it can be helpful with characterizing this as a likely degenerative process, and that narrows it down. “Are we dealing with an Alzheimer’s process or another neurodegenerative disease such as dementia with Lewy bodies or frontotemporal lobar degeneration? These other disorders present differently, and as a clinician you can get a pretty good index as to what’s likely the underlying pathology.”

Regarding examinations in the office, Dr. Petersen observes that brief mental exams can be useful for putting a metric in the chart. “It may not be enough and sometimes it may require that the clinician actually become a little bit more sophisticated with regard to his or her own armamentarium of cognitive assessment.” Assessment tools help give clinicians an index of how the person is doing, he suggests. Similarly, in other cognitive domains—language, problem-solving, attention, concentration, visual-spatial skills, you know—assessment tools might be very useful to characterize the clinical syndrome that is operating, according to Dr. Petersen.

These tools help physicians take the clinical course as far as possible can with regard to identifying the clinical features, says Dr. Petersen. It also allows you to track changes over time and provides a good index for helping them with regard to counseling.

As the biomarker field evolves, Dr. Petersen is hopeful they will become less expensive and more accessible to patients. “You don’t throw everything at a person, but you may choose one or two,” he says. “Most of us still like to do something like an MRI scan early on, so that we get a picture of the structure of the brain to make sure we’re not being fooled by some other process—perhaps a stroke, a tumor, or a hydrocephalus.” Also potentially useful is neuropsychological testing, which further helps to characterize the clinical profile of patients. “Those types of features are still going to be absolutely vital and probably a requisite before engaging in these more expensive and invasive measures that biomarkers may require,” Dr. Petersen implores. “Clinical acumen is still going to be at the forefront of characterizing these diseases.”

Conclusion: The Bigger Cognitive Picture

As the Alzheimer’s research spectrum continues to see new innovations and technologies, the hope is that we are moving closer to the end goal of disease modifying therapies. These advances are already yielding new ways of understanding the broader spectrum of Alzheimer’s disease and cognitive changes associated with aging, according to Dr. Petersen. “Alzheimer’s disease is very important but it is only part of the picture,” he says. “Plaques and tangles are very important, maybe the most important and maybe the driving force, but there are other pathologies that commonly appear in aging, such as lewy bodies made up of alpha-synuclein protein. A protein called PDP43 is commonly seen in aging, and then there’s vascular disease and vascular changes in the brain.” The most common explanation for abnormal cognitive changes in aging likely is a combination of those features. “Alzheimer’s disease is basically two components—it’s amyloid and tau—but there are other features that contribute.”

As imaging technologies become more refined, Dr. Petersen stresses that the role of the neurologist is increasingly important in teasing apart the clinical syndromes that may present with different pathologic features. “In a perfect world, we may end up prescribing combination therapy for our older patients who are having cognitive difficulties, and that profile of therapies may be dependent upon the clinical profiles from both the neurologists and the biomarker, allowing us tailor our therapies accordingly,” says Dr. Petersen.

Ronald C. Petersen, MD, PhD, is Professor of Neurology, the Cora Kanow Professor in Alzheimer’s Disease Research, and a Clinical Distinguished Investigator at the Mayo Clinic in Rochester, Minnesota.


Contact Info

For advertising rates and opportunities:
Wendy Terry

About Practical Neurology

Launched in 2002, Practical Neurology is a publication uniquely dedicated to presenting current approaches to patient management, synthesis of emerging research and data, and analysis of industry news with a goal to facilitate practical application and improved clinical practice for all neurologists. Our straightforward articles give neurologists tools they can immediately put into practice.