Clinical Diagnosis of Alzheimer’s Disease Depends on Both Clinical and Biomarker Evidence

A clinical diagnosis of Alzheimer’s disease should require two components: having one of the characteristic clinical syndromes and positive amyloid β and tau markers. So advises the International Working Group (IWG) in a recent paper published online ahead of print in Lancet Neurology (2021 Apr 29; S1474-4422(21)00066-1).

The expert group also notes that while biomarker testing is not advised for patients without cognitive impairment in clinical settings, unimpaired patients who are tested and found to have positive biomarkers should be categorized as “asymptomatic at risk of progression” rather than as having Alzheimer’s disease.

“Defining Alzheimer’s disease based on both clinical and biological criteria is critical for accurately identifying the disease and best serving our patients,” says neurologist Marwan Sabbagh, MD, Director of Translational Research at Cleveland Clinic Lou Ruvo Center for Brain Health. Dr. Sabbagh is a member of the IWG and one of the article’s authors.

Earlier recommendations led to challenges

The IWG, the U.S. National Institute on Aging (NIA) and the Alzheimer’s Association (AA) have issued recommendations related to Alzheimer’s disease diagnosis multiple times over the past decades. The role of biomarkers has increased progressively as the field has advanced.

The IWG’s newly published recommendation — intended for both clinical and research settings — is in response to the 2016 IWG-AA and 2018 NIA-AA statements, which advised that, for research purposes, the presence of positive amyloid β and tau markers is sufficient to diagnose Alzheimer’s disease regardless of clinical state. This allowed Alzheimer’s disease to be studied across its clinical spectrum, from patients having no symptoms to those with advanced dementia. However, the guidance led to problems in the clinical arena, where the biomarker-only definition was also sometimes adopted.

Biomarkers alone are inadequate for predicting future cognition

The new IWG guidance points out that a purely biological definition of Alzheimer’s disease has low prognostic accuracy, as positive amyloid β and tau markers do not always predict that dementia — or even mild cognitive impairment — will occur in a patient’s lifetime.

The IWG also notes that Alzheimer’s disease biomarkers are commonly found in other neurodegenerative diseases (most often dementia with Lewy bodies), which can potentially lead to overlooking the primary cause of a patient’s symptoms.

The new guidance specifies that the diagnosis of Alzheimer’s disease can be established with the presence of both amyloid β and tau markers (from either cerebrospinal fluid or positron emission tomography) plus one of the following Alzheimer’s disease phenotypes:

• Amnestic variant of the hippocampal type (typical syndrome with memory problems predominating)
• Posterior cortical atrophy variant (visual symptoms predominating)
• Logopenic variant primary progressive aphasia (language problems predominating)

For patients with an atypical clinical presentation or unknown or negative biomarker status, further investigation is needed, including evaluation for nondegenerative syndromes.

The article emphasizes that clinical determination of an Alzheimer’s disease syndrome or the need for biomarker testing should not be based on patient self-reports of memory problems or cognitive decline in the setting of normal cognitive testing. Self-reports have been found to be poorly correlated with Alzheimer’s disease development, although such reports by an informant tend to be more predictive.

Use – and misuse – of biomarker testing

The IWG paper states that biomarker testing is not recommended for patients who are cognitively unimpaired, as the presence of positive biomarkers alone is unreliable in predicting an individual’s clinical course. In fact, the authors write that “the follow-up of cognitively unimpaired biomarker-positive individuals suggests that the majority of these individuals do not progress over time.”

Cognitively unimpaired patients who undergo biomarker testing anyway, either for research purposes or because they desire it, should be considered “at risk for progression to Alzheimer’s disease” if amyloid β and tau markers are positive.

Dr. Sabbagh emphasizes that despite the limitations of biomarkers, they are critical for making an accurate Alzheimer’s disease diagnosis. Basing the diagnosis on the clinical picture alone, he says, leads to an inaccurate diagnosis about 25% of the time.

He notes that biomarker testing will become even more important with the potential imminent FDA approval of new Alzheimer’s disease monoclonal antibody therapy, as qualification for treatment will likely be contingent on a definitive diagnosis.

“Biomarkers are essential to researchers and clinicians for diagnosis, choosing appropriate therapy and disease monitoring,” he observes. “We expect them to become even more important in the future as they become better characterized and with the discovery of new markers and more-specific therapies.”