Large-scale brain imaging studies have revealed that the pathological hallmarks of Alzheimer’s disease—including amyloid plaques and tau tangles—appear differently across racial and ethnic groups, suggesting that disease progression and presentation may vary based on genetic and biological factors beyond what standard clinical assessments capture. A comprehensive analysis of thousands of brain imaging scans has documented these disparities, raising important questions about whether current diagnostic criteria adequately identify Alzheimer’s in all populations and whether treatment approaches need to account for these biological differences.
For families and clinicians navigating dementia care, this research underscores that a diagnosis based solely on cognitive decline may miss important context about the underlying brain pathology occurring at different rates and in different patterns across demographic groups. These findings emerge from neuroimaging data that tracks amyloid and tau accumulation—the two proteins most associated with Alzheimer’s disease pathology—and shows that the relationship between pathology levels and cognitive symptoms is not uniform across all groups. Some individuals show significant brain pathology without proportional cognitive decline, while others show cognitive changes with less advanced pathological changes visible on imaging, and these patterns appear to correlate with racial and ethnic background.
Table of Contents
- What Brain Imaging Reveals About Pathological Differences
- The Challenge of Cognitive Reserve and Disease Expression
- How Systemic Health Disparities May Influence Brain Pathology
- Implications for Diagnosis and Treatment Decisions
- The Critical Gap in Neuropathology Autopsy Verification
- Genetic Variation and APOE Status Across Populations
- Advancing Equitable Neuroimaging Research and Clinical Practice
What Brain Imaging Reveals About Pathological Differences
Brain imaging studies using positron emission tomography (PET) scans and other advanced neuroimaging techniques measure the buildup of amyloid-beta and tau proteins in the brain, providing a window into the biological stage of Alzheimer’s disease independent of memory or thinking test scores. These imaging biomarkers have become central to Alzheimer’s research because they can show disease progression years before cognitive symptoms appear. However, when researchers analyzed imaging data across large, diverse populations, they discovered that the amount of pathology visible on scans did not predict cognitive decline equally well across racial and ethnic groups—meaning that a person from one background might show substantial amyloid burden on imaging yet maintain normal thinking, while someone from another background with similar imaging results might experience noticeable cognitive changes.
This mismatch has profound implications for diagnosis and research enrollment. If clinical trials predominantly enroll based on cognitive symptoms and a positive amyloid scan, but the relationship between pathology and symptoms differs by race, then research populations may systematically exclude certain groups at earlier disease stages while over-representing others at similar pathological stages, potentially leading to treatments validated in less diverse populations. The reasons for these differences are not yet fully understood. Factors may include genetic variations in how different populations process or accumulate these proteins, differences in educational background and cognitive reserve (the brain’s ability to compensate for damage), variation in vascular health across groups related to systemic health disparities, or differences in how cognitive decline is measured and interpreted across different populations.
The Challenge of Cognitive Reserve and Disease Expression
Cognitive reserve—the concept that education, occupational complexity, and lifetime intellectual engagement build a buffer against cognitive decline—may play a different protective role across demographic groups, potentially explaining why some individuals show substantial brain pathology without proportional memory loss. Research has long suggested that people with higher educational attainment and more cognitively demanding careers show some resilience to cognitive symptoms despite measurable brain changes, but the relationship between reserve and Alzheimer’s pathology may not be identical across all populations, particularly when educational and occupational opportunities have been unequally distributed historically. This creates a diagnostic dilemma: standard cognitive testing relies on detecting decline from a person’s baseline, which assumes that baseline is similar across groups.
However, if some individuals start with higher cognitive reserve due to differences in education or lifetime opportunity, their cognitive decline from a higher baseline might not be detected by tests designed with average populations in mind, leading to underdiagnosis of Alzheimer’s in highly educated individuals from certain backgrounds, while others might show cognitive test score reductions that reflect normal variation rather than disease. A person might score in the average range on a memory test yet be experiencing decline from their personal cognitive baseline, or conversely, score low without actual decline from their own baseline. An important limitation of current research is that cognitive reserve is difficult to measure objectively—it is typically estimated through proxy measures like years of education or occupational history, both of which have been affected by systemic inequities. This means that apparent differences in cognitive reserve across racial and ethnic groups may partly reflect measurement artifacts rather than true biological differences in how the brain compensates for damage.
How Systemic Health Disparities May Influence Brain Pathology
The accumulation of amyloid and tau proteins in the brain does not occur in isolation from overall health; cardiovascular disease, diabetes, hypertension, and chronic inflammation all correlate with worse Alzheimer’s pathology and may accelerate the disease process. Racial and ethnic groups in many countries experience higher rates of these comorbid conditions due to differences in healthcare access, neighborhood environmental factors, economic stress, and other social determinants of health. For example, Black Americans have higher prevalence of hypertension and diabetes, conditions that are also associated with accelerated amyloid and tau accumulation, potentially explaining some of the observed differences in brain pathology patterns between racial groups.
The relationship between vascular health and Alzheimer’s pathology means that interventions targeting cardiovascular risk factors might have different impacts across populations depending on how widespread vascular disease is in that group. Someone with well-controlled hypertension and stable diabetes might show slower pathology accumulation, while someone with poorly managed cardiovascular conditions might show faster pathology progression, creating an apparent biological disparity that is actually driven by differences in disease management and healthcare access. These findings highlight that brain pathology does not develop in a vacuum—it reflects not just individual genetic susceptibility but also the cumulative biological burden of systemic health inequities that create different exposures to chronic disease across demographic groups.
Implications for Diagnosis and Treatment Decisions
Current diagnostic criteria for Alzheimer’s disease have been refined to incorporate imaging biomarkers alongside cognitive and functional changes, creating a more biologically grounded approach to diagnosis than relying solely on memory tests. However, if the relationship between pathology on imaging and clinical symptoms varies by race and ethnicity, then applying standardized diagnostic criteria universally may miss disease in some populations while over-diagnosing in others. A 65-year-old with moderate amyloid on PET scan and subtle cognitive changes might meet diagnostic criteria for Alzheimer’s in one diagnostic framework, while someone with similar imaging but different cognitive patterns from a different background might not, despite having similar pathology.
This has direct consequences for treatment decisions. New disease-modifying drugs that target amyloid or tau have shown modest cognitive benefits in some populations studied, but if these medications were tested primarily in groups with one pattern of pathology-to-symptom relationship, their effectiveness in other populations with different relationships may differ. Someone might be offered an amyloid-targeting drug based on imaging findings, but without knowing how pathology translates to symptoms in their particular population, predicting individual benefit becomes uncertain. The comparison between the benefit shown in clinical trials and the expected benefit in a specific individual from an underrepresented group becomes complex.
The Critical Gap in Neuropathology Autopsy Verification
Autopsy studies—which remain the gold standard for confirming Alzheimer’s pathology, as they allow direct examination of brain tissue rather than inference from imaging—have historically included limited racial and ethnic diversity, meaning that much of our understanding of what specific pathology patterns look like in different populations comes from relatively homogeneous autopsy cohorts. This creates a troubling circularity: imaging studies are validated against autopsy findings, but if autopsy samples don’t represent all populations, then imaging interpretation may be calibrated to pathology patterns that are unrepresentative of some groups. An imaging marker that correlates well with autopsy-confirmed pathology in one population might perform differently in another population with different pathological presentations.
The limited diversity in autopsy cohorts is not accidental—it reflects historical disparities in healthcare access and research participation, as well as practical factors like geography and research infrastructure. Expanding autopsy research to include more diverse populations would directly improve our ability to interpret imaging in those groups, but requires sustained investment in research recruitment and community partnership. Without this verification, recommendations about who should undergo imaging, when, and how to interpret results may inadvertently miss disease in some populations or lead to unnecessary testing in others.
Genetic Variation and APOE Status Across Populations
The APOE4 gene variant is the strongest known genetic risk factor for late-onset Alzheimer’s disease, and the frequency of this variant differs across racial and ethnic populations, which may contribute to observed differences in amyloid and tau accumulation patterns. However, the relationship between APOE4 status and disease risk is not uniform—carrying an APOE4 allele increases risk substantially in some populations but shows weaker predictive value in others. This genetic variation means that genetic risk stratification based on APOE status might be more or less predictive depending on someone’s ancestry, suggesting that personalized risk assessment cannot use a one-size-fits-all approach.
Advancing Equitable Neuroimaging Research and Clinical Practice
Addressing disparities in Alzheimer’s brain imaging requires intentional efforts to recruit and retain diverse participants in longitudinal neuroimaging studies, to validate imaging biomarkers across multiple populations simultaneously rather than sequentially, and to involve diverse communities in designing research questions and interpreting findings. Clinical practice should incorporate acknowledgment that cognitive and imaging findings may need to be interpreted with awareness of demographic factors and that standard cutoff scores for cognitive testing may perform differently across groups. For individual patients and families, this research suggests that conversations with clinicians about what cognitive decline actually represents should account for personal baseline, education, health history, and other individual factors rather than assuming standard diagnostic criteria will capture disease progression equally well for everyone.
