NEW YORK – An international working group of Alzheimer's disease experts has pushed back against the idea asserted by the Alzheimer's Association in recently issued guidelines that the neurodegenerative disease can be defined and diagnosed based only on biomarkers.
The working group, comprising nearly four dozen experts from institutions around the world, published their own recommendation to consider Alzheimer's as a clinical-biological construct in JAMA Neurology this month, updating a recommendation from 2021.
The authors of the JAMA Neurology paper evaluated the revised criteria for diagnosing and staging Alzheimer's that the Alzheimer's Association published this summer, in which the group recommended moving away from relying on clinical symptoms to diagnose and stage Alzheimer's and move toward incorporating biological criteria.
The Alzheimer's Association's guidelines established a set of core beta-amyloid and tau biomarkers as tools for the diagnosis, staging, and determining the prognosis of Alzheimer's and the effect of treatment. Using those guidelines, an abnormality in one of these core biomarkers would be sufficient to diagnose stage 1 presymptomatic Alzheimer's, even if an individual doesn't have any clinical symptoms, like dementia. While the Alzheimer's Association working group said that the condition can be diagnosed in asymptomatic patients using biomarkers, the group urged caution and stopped short of recommending using them to clinically diagnose the disease in presymptomatic patients.
The Alzheimer's Association's guidelines coincide with the market approval of drugs like Biogen and Eisai's Leqembi (lecanemab) and Eli Lilly's Kinsula (donanemab), which are designed to target and clear beta-amyloid in the brain. A shift toward a purely biomarker-based diagnosis of Alzheimer's could make a lot more people eligible for these drugs and other beta-amyloid- and tau-targeting drugs under development. The Alzheimer's Association accepts funding from the pharmaceutical industry, and Eisai, Lilly, and Biogen were among its top contributors in 2023.
The international working group in the JAMA Neurology paper pushed back against the "movement" to define Alzheimer's as a "purely biological entity based on biomarker findings." These experts emphasized that patients who do not have symptoms, but who have beta-amyloid pathology, should not be labeled as having Alzheimer's and should instead be considered at risk for Alzheimer's. They noted that for patients with these biomarkers, based on available evidence, it's unclear when, if ever, neurological symptoms will develop.
While these biomarkers are related to Alzheimer's disease progression, they don't sufficiently account for the various underlying mechanisms driving the disease, and only represent risk factors, the international working group said. "In the clinical setting, disclosing a diagnosis of AD to cognitively normal people with only core 1 AD biomarkers represents the most problematic implication of a purely biological definition of the disease," the authors wrote.
They suggested that a combination of clinical phenotypes and beta-amyloid and tau biomarkers should be used to establish an Alzheimer's diagnosis, which is similar to the guidelines put out by the Alzheimer's Association for cognitively impaired patients.
But in patients without symptoms, they recommended against diagnosing based only on the presence of beta-amyloid and tau, since while these biomarkers may increase a person's lifetime risk of progression to cognitive impairment, they are not deterministic. They cited research that lifetime risk of Alzheimer's dementia in a 65-year-old man with beta-amyloid positivity is 22 percent, just 1.7 times higher than someone without the biomarker.
There are some cases in which cognitively normal patients with a specific pattern of biomarkers that are closely associated with developing the disease could be referred to as presymptomatic Alzheimer's, such as those who have autosomal dominant mutations in APP, PSEN1, and PSEN2 that are associated with Alzheimer's. Mutations in these genes are known to cause a form of early-onset familial Alzheimer's, characterized by patients experiencing dementia before age 65.
"We foresee being able to add new biomarker profiles within this presymptomatic grouping," the authors added.
The authors also drew a distinction between research and clinical settings, and stressed that they support research to identify risk factors for the disease. However, diagnostic testing for Alzheimer's biomarkers should not be performed in cognitively normal patients outside of research settings at this time, they maintained.
Additionally, they raised concerns that emerging biomarkers could result in a future where a cognitively normal person could be diagnosed with multiple neurodegenerative diseases based on evaluations of beta-amyloid, tau, alpha-synuclein, and other biomarkers that are associated with processes that contribute to Alzheimer's in different ways.
"We argue that biomarkers alone should remain markers of pathological processes and not markers of a specific disease," the international working group wrote. "Furthermore, the contribution of biomarkers in the clinical setting depends on the context of use and, importantly, should differ between the assessment of individuals with and without cognitive impairment."