Reviewed by the Help Dementia Editorial Team — our editors review every article for accuracy against guidance from the National Institute on Aging, the Alzheimer’s Association, and peer-reviewed sources.
Spatial transcriptomics sits at the center of this dementia and brain health question.
Spatial transcriptomics is revolutionizing our understanding of Alzheimer’s disease by mapping where specific genes are expressed within the brain’s intact tissue architecture. This technology reveals not just which genes are active in Alzheimer’s brains, but precisely where they’re active—down to individual cell locations within different brain regions and layers. For example, a 2024 study published in Nature Genetics discovered that a particular inflammatory program controlled by glial cells (the brain’s immune cells) becomes abnormally active in the upper layers of the brain’s cortex in Alzheimer’s patients, a finding that would have been impossible to pinpoint using older methods that destroy tissue structure. What makes spatial transcriptomics fundamentally different from previous gene-mapping techniques is that it preserves the brain’s three-dimensional architecture while simultaneously measuring gene activity. Traditional approaches require scientists to grind up brain tissue, destroying all information about where genes are located relative to other cells and structures.
Spatial transcriptomics keeps the tissue intact, allowing researchers to see the complete cellular neighborhood around each gene expression change. This spatial context matters enormously because Alzheimer’s involves interactions between multiple cell types—neurons, glial cells, and immune cells—that cluster in specific regions of the brain. The practical impact is profound for dementia research. Multiple research teams working in 2024 and 2025 have begun using these spatial mapping techniques to understand how Alzheimer’s pathology spreads through the brain and which cell types drive different aspects of cognitive decline. This knowledge is already opening new avenues for potential treatments that could target the specific cell populations and brain regions driving disease progression.
Table of Contents
- How Spatial Transcriptomics Reveals Gene Expression Patterns in Alzheimer’s Brains
- Building Comprehensive Databases of Alzheimer’s Gene Expression Maps
- Understanding Cell-Type Specific Changes in Aging and Alzheimer’s
- Comparing Healthy Aging, Mild Cognitive Impairment, and Alzheimer’s Disease
- Limitations and Warnings About Translating Lab Discoveries to Clinical Practice
- Integrating Spatial Transcriptomics with Imaging and Biomarkers
- The Future of Spatial Transcriptomics in Dementia Research and Clinical Care
- Conclusion
How Spatial Transcriptomics Reveals Gene Expression Patterns in Alzheimer’s Brains
Spatial transcriptomics combines two complementary technologies: some methods use imaging to visualize where genes are expressed, while others use sequencing that retains spatial information from tissue sections. A 2025 study published in Cell Reports used a particularly innovative approach by integrating magnetic resonance imaging (MRI) data directly with spatial transcriptomics from the same brain tissue. The researchers discovered that areas showing reduced fractional anisotropy on MRI—a measure of how organized white matter is—correlated with changes in myelin-producing oligodendrocytes and accumulation of amyloid-beta protein, the hallmark of Alzheimer’s disease. This kind of integration allows doctors to potentially see on a patient’s MRI scan the exact microscopic changes that spatial transcriptomics reveals in tissue. The technology generates extraordinarily detailed maps.
A 2025 study in Nature Communications examined tissue from the temporal cortex and white matter of 40 individuals ranging from healthy to those with significant Alzheimer’s pathology. The researchers identified cell type-specific subclusters within neuronal and glial populations—meaning they didn’t just say “glial cells are affected,” but rather “specific subtypes of glial cells in specific cortical layers show these expression changes.” This level of granularity is a major shift from how researchers previously understood Alzheimer’s at the cellular level. One important limitation is that spatial transcriptomics remains technically demanding and expensive, available only in research settings at major medical centers. studies examining brain tissue rely on autopsy samples or rare donated tissue from living patients undergoing surgery. This means the findings, while scientifically robust, currently come from relatively small numbers of patients compared to large-scale studies of genetics or blood biomarkers.
Building Comprehensive Databases of Alzheimer’s Gene Expression Maps
The volume of spatial transcriptomics data has grown large enough that researchers have begun creating comprehensive databases. A 2024 Nature Communications article described ssREAD (a single-cell and spatial RNA-seq database for Alzheimer’s disease), which contains 1,053 samples from 67 different studies totaling over 7.3 million individual cells, plus 381 spatial transcriptomics datasets from 18 separate human and mouse brain studies. This consolidated resource allows researchers worldwide to ask questions about gene expression patterns across multiple studies without having to coordinate data sharing between labs. The existence of ssREAD and similar databases represents a paradigm shift in Alzheimer’s research. Previously, each lab would generate their own datasets, and comparisons across studies required painstaking efforts to account for technical differences.
Now researchers can identify which expression changes appear consistently across studies (suggesting they’re truly fundamental to the disease) versus which appear only in specific datasets (suggesting they may be technical artifacts or study-specific variations). A gene expression change observed in multiple independent studies using different spatial transcriptomics methods is far more likely to represent a genuine biological target for drug development. However, consolidating datasets introduces a significant challenge: technical biases from different sequencing platforms and tissue preparation methods can create apparent gene expression differences that don’t reflect actual biology. Researchers must carefully account for these “batch effects” when analyzing combined datasets. A gene that appears highly expressed in samples from one laboratory using one technology might appear differently in samples from another laboratory using slightly different protocols.
Understanding Cell-Type Specific Changes in Aging and Alzheimer’s
One of the most striking discoveries from spatial transcriptomics is that Alzheimer’s disease affects different cell types in the brain in fundamentally different ways. A 2025 study published in Molecular Neurodegeneration used spatial transcriptomics to reveal that microglia—the brain’s resident immune cells—show remarkable regional heterogeneity and functional diversity across the brain. Rather than all microglia responding to Alzheimer’s pathology the same way, different populations of microglia in different brain regions activate different gene programs. The study also documented dynamic interactions between microglia, astrocytes (star-shaped support cells), neurons, and oligodendrocytes, showing how changes in one cell type ripple through the entire cellular ecosystem. This cell-type specificity has real implications for understanding disease progression.
Microglia in the frontal cortex, which handles executive function and decision-making, may respond to amyloid-beta accumulation differently than microglia in the hippocampus, which is critical for memory formation. Understanding these regional differences could explain why different patients experience different patterns of cognitive decline—some losing language abilities early, others losing memory first. A therapeutic drug that activates a beneficial microglia response in one brain region might paradoxically worsen inflammation in another region if not carefully designed. The cellular interactions revealed by spatial transcriptomics are far more complex than previously appreciated. Neurons don’t simply accumulate amyloid and die in isolation; they send molecular signals to nearby glial cells, which respond with their own gene expression changes, which in turn affect other neurons. Spatial transcriptomics captures these relationships by showing which cell types are located near amyloid plaques and what gene programs they activate in response.
Comparing Healthy Aging, Mild Cognitive Impairment, and Alzheimer’s Disease
The Nature Communications study examining 40 individuals at different disease stages provided a crucial comparison: researchers could directly contrast gene expression patterns in healthy older brains, brains showing early amyloid accumulation without cognitive symptoms, and brains with full Alzheimer’s disease with severe cognitive decline. This comparison revealed that certain gene expression changes appear very early, before symptoms develop, while others correlate more closely with actual cognitive decline. Some glial genes, for instance, begin changing their expression when amyloid first accumulates but before tau tangles (another Alzheimer’s hallmark) form extensively. Understanding these sequential changes matters because it suggests potential intervention windows. If a particular gene expression change happens years before cognitive decline appears, interventions targeting that change might have time to prevent disease progression.
Conversely, if a gene expression change only appears in the late stages of Alzheimer’s when neurons are already dying extensively, targeting that change might come too late. Spatial transcriptomics provides the necessary spatial and temporal roadmap to distinguish early-opportunity targets from late-stage changes. The comparison also reveals heterogeneity in disease pathology. Not all Alzheimer’s brains look identical at the gene expression level, even at similar disease stages. Some patients show prominent neuroinflammatory changes, others show more extensive myelin damage, and still others show primary neuronal loss. This molecular heterogeneity likely explains why patients respond differently to the same treatments—they may have fundamentally different underlying cellular mechanisms driving their cognitive decline.
Limitations and Warnings About Translating Lab Discoveries to Clinical Practice
An important limitation worth understanding: most spatial transcriptomics studies examine postmortem brain tissue, which means the expression patterns reflect the final stage of disease, not the living, dynamic process of disease development. A gene that appears highly expressed in a deceased Alzheimer’s patient’s brain may have been activated decades earlier in response to early amyloid accumulation, then changed again as disease progressed, and finally ended up at this expression level just before death. Studies in mouse models provide some dynamic information, but mouse brains and human brains differ in fundamental ways—mice don’t naturally develop Alzheimer’s disease the way humans do. Another warning: identifying a disease-associated gene expression pattern doesn’t automatically mean that pattern is the cause of disease rather than a consequence. Inflamed microglia might activate inflammatory genes in response to amyloid-beta (cause and effect), or amyloid might be accumulating because the inflammatory microglia response failed (opposite causality), or both might be responding to some deeper problem.
Spatial transcriptomics is exceptionally good at mapping “what’s different,” but determining “what’s causing what” requires additional experiments and careful interpretation. Clinical translation also faces a timing problem. The spatial transcriptomics discoveries being reported in 2024-2025 examined brain tissue from patients who developed symptoms years or decades ago under different environmental and medical conditions. By the time these findings mature into clinical trials, the patients currently developing Alzheimer’s will have lived through different decades with different exposures and interventions. A therapeutic target validated based on 2024 spatial transcriptomics data might prove less important for 2035 Alzheimer’s patients with different lifestyles or preventive treatments.
Integrating Spatial Transcriptomics with Imaging and Biomarkers
The integration of spatial transcriptomics with MRI imaging, mentioned in the Cell Reports 2025 study, represents one of the most clinically relevant advances. By directly comparing what spatial transcriptomics reveals at the microscopic cellular level with what MRI shows at the macroscopic brain structure level, researchers can create a bridge between research-grade molecular understanding and clinical imaging available in hospitals. If MRI patterns can reliably predict the spatial transcriptomics changes that precede cognitive decline, MRI might eventually serve as a non-invasive proxy for the underlying molecular pathology.
Current research is exploring whether combinations of blood biomarkers—proteins that can be measured in routine blood tests—might reflect the spatial transcriptomics changes visible in brain tissue. If a blood biomarker pattern correlates with a specific spatial transcriptomics signature of neuroinflammation or myelin damage, it could theoretically allow clinicians to infer something about the underlying cellular changes without ever examining brain tissue. This remains speculative, but the ssREAD database and similar resources provide the foundation for exploring these connections systematically.
The Future of Spatial Transcriptomics in Dementia Research and Clinical Care
A 2024 Frontiers in Neurology review noted that spatial transcriptomics methods continue advancing rapidly, with improvements in resolution, speed, and accessibility. Newer methods can map smaller and smaller regions of tissue with higher and higher precision. These advances suggest that within a few years, spatial transcriptomics might move beyond research studies to become a tool for understanding individual patients’ disease biology—similar to how genetic sequencing moved from research to clinical practice.
The comprehensive databases like ssREAD will likely accelerate discovery by allowing researchers to form hypotheses from consolidated data and then test them in new studies. This virtuous cycle—learn from existing data, generate and test new hypotheses, add results back to the database—could accelerate the pace of discovery compared to the traditional model where each lab worked independently. For dementia patients and families, this acceleration matters enormously because each year of delay in understanding disease mechanisms and developing treatments represents substantial disability and lost quality of life.
Conclusion
Spatial transcriptomics has transformed Alzheimer’s research by revealing not just what genes are expressed abnormally in affected brains, but exactly where these expression changes occur and how they interact across different cell types and brain regions. Recent studies from 2024-2025 have identified specific neuroinflammatory programs in upper cortical layers, myelin damage coupled with amyloid accumulation, and regional heterogeneity in how glial cells respond to disease. These discoveries are being consolidated into comprehensive databases containing millions of cells and hundreds of spatial transcriptomics datasets, creating a foundation for understanding the molecular mechanisms underlying individual variation in Alzheimer’s disease.
The path from these molecular discoveries to improved clinical care remains a multi-year process requiring careful validation, understanding of causality versus correlation, and translation of laboratory findings into interventions that can be tested and delivered in living patients. However, the spatial mapping of gene expression in Alzheimer’s brains has already revealed previously unknown cellular interactions and disease mechanisms that could not have been detected using older technologies. For individuals and families affected by dementia, these advances represent genuine scientific progress toward understanding what goes wrong in aging brains and how that understanding might lead to prevention or treatment strategies.
You Might Also Like
- Ultrasensitive Immunoassay Platforms Detect Trace Alzheimer’s Biomarkers
- Telepresence Robots Enable Remote Family Visits for Alzheimer’s Patients
- Population Cohort Studies Track Alzheimer’s Risk Factors Over Decades
For more, see Alzheimer’s Association — medical tests.
