The field of Alzheimer’s disease (AD) is seeing a heated debate about the best way to diagnose this devastating condition. On one side, the Alzheimer Association (AA) proposed new criteria that rely solely on biological markers—such as proteins seen in brain scans or found in the liquid bathing the brain and spine, called cerebrospinal fluid (CSF)—even in individuals who show no symptoms. These biomarkers, which include amyloid and tau proteins, are widely associated with the risk of developing AD. However, a group of international experts, known as the International Working Group (IWG), has proposed an alternative approach, suggesting that the disease should be diagnosed based on biological markers and clinical symptoms, rather than biomarkers alone.

The power and promise of biomarkers in Alzheimer’s diagnosis

The debate largely centers on the power of biomarkers, which have transformed Alzheimer’s research and opened the door to early detection. Biomarkers, such as amyloid and tau, are proteins associated with Alzheimer’s and can now be identified in living individuals through advanced brain scans and fluid tests. These biomarkers have proven valuable in research and clinical trials, where they can help track the disease’s progression or the effectiveness of new drugs.

Supporters of biomarker-based diagnosis see it as a breakthrough in early detection. It provides a way to identify people who may develop Alzheimer’s even before they show signs of memory loss or cognitive decline. By identifying Alzheimer’s in its earliest stages, they hope to intervene sooner and slow down or even prevent the onset of symptoms.

Despite this optimism, the IWG is cautious about how diagnoses based only on biomarkers might play out in real life. For example, a study has shown that 90 to 100 percent of adult individuals with Down syndrome develop AD dementia in their seventies, and the presence of fluid biomarkers can be detected as early as the third decade of life. However, the IWG stated that the same model of biomarker predictability for dementia cannot be assumed in cognitively normal individuals with sporadic AD (absence of genetic determinants leading to AD). In other words, the presence of biomarkers does not necessarily mean that a person has Alzheimer’s, just as high cholesterol doesn’t guarantee someone will have a heart attack. According to their view, a biomarker-positive result should instead indicate an increased risk, rather than a definitive diagnosis of Alzheimer’s.

Diagnosis without symptoms: Currently more harmful than beneficial?

One of the most significant criticisms of the AA’s biomarker-only approach is its potential impact on cognitively normal individuals—those who lack any memory or cognitive problems. Under the new AA criteria, a person who tests positive for the Core 1 panel of biomarkers could receive an Alzheimer’s diagnosis, even if they show no signs of memory loss or cognitive decline. Although the new AA criteria is not yet a clinical practice guideline, the industry influence in this AD diagnosis shift has been questioned.

The IWG argues that such diagnoses may create unnecessary worry and confusion for patients and their families, given that many biomarker-positive individuals may never develop Alzheimer’s symptoms in their lifetime. If people without symptoms receive a diagnosis based solely on biomarkers, they may face years of stress and anxiety over a condition that may never manifest.

Moving forward: Defining risk rather than disease

The IWG proposes that, instead of diagnosing Alzheimer’s in symptom-free people, cognitively normal individuals with positive amyloid and tau biomarkers should be classified as “at risk for Alzheimer’s disease” or “presymptomatic.” The latter category implies a very high risk of developing clinical cognitive symptoms. It is also associated with specific genetic variations, such as those in proteins of the amyloid cascade or a specific version of the gene coding for a lipid transport protein (APOE4 allele ε4). As it is acknowledged by the IWG, the number of presymptomatic individuals might increase as new risk of progression factors are identified by new cohort studies.

Only symptomatic patients with specific clinical presentations, including mild cognitive impairment and dementia, in combination with positive CSF or brain imaging AD biomarkers, should be diagnosed with AD. Such a classification system is based on the probabilistic amyloid cascade model, which highlights the role of other determinants of disease besides amyloid, such as lipid metabolism markers (APOE genes), lifestyle factors, and accompanying diseases.

Ultimately, the IWG recognizes the value of nondiagnostic biomarker studies in individuals with normal cognitive function to assess the risk of progression to clinical symptoms, and therefore to evaluate preventive interventions. However, in the clinical setting, the role of AD biomarkers should be supportive to, but not substitute the specific clinical assessment during AD diagnosis.

Laura García-Pupo is a medical writer.