Mutations that accumulate in blood cells with age and raise the risk of leukemia and lymphoma can also colonize the brain’s resident immune cells, where they appear to stoke the kind of inflammation tied to Alzheimer’s disease. A peer-reviewed study in Cell used ultra-deep sequencing of postmortem brains from the ROSMAP cohort and found somatic variants in well-known cancer-driver genes enriched specifically in microglia-like cells of people who had Alzheimer’s. The finding forces a rethinking of how blood-cancer biology and neurodegeneration overlap, because earlier work in large cardiovascular cohorts had linked the same class of mutations in blood to a lower risk of Alzheimer’s, not a higher one.
Why blood-cancer mutations in brain immune cells change the Alzheimer’s equation
Clonal hematopoiesis of indeterminate potential, or CHIP, describes the slow expansion of blood cells carrying somatic mutations in genes such as TET2, DNMT3A, and ASXL1. These mutations are best known for raising the odds of blood cancers, but they also alter immune-cell behavior. A 2023 study in cardiovascular cohorts, drawing on TOPMed populations including the Framingham Heart Study and the Cardiovascular Health Study with ADSP replication, reported that CHIP in the blood was associated with protection from Alzheimer’s disease. The National Institute on Aging highlighted that finding for its potential public-health significance, because CHIP is common in older adults and can be detected with straightforward sequencing of peripheral blood.
The new Cell study flips part of that narrative. According to the research, when the same cancer-driver variants show up not in circulating blood cells but in microglia-like cells inside the brain, they push those cells into inflammatory and proliferative states. Microglia serve as the brain’s first line of immune defense, clearing debris and responding to injury. When they acquire somatic mutations that drive unchecked activation, the result is chronic neuroinflammation, a process increasingly recognized as a direct contributor to Alzheimer’s pathology rather than a bystander effect. The contrast between apparently protective effects in the bloodstream and damaging effects in brain-resident immune cells raises the possibility that tissue context may completely reshape the consequences of a given mutation.
One plausible mechanism connects these two findings. CHIP clones circulating in the blood could, under certain conditions, cross a weakened blood-brain barrier and seed themselves in brain tissue. In mid-life, as vascular integrity declines, mutant blood-derived cells might engraft among resident microglia. A clone that behaves protectively in the periphery, perhaps by modulating systemic immune tone or dampening maladaptive inflammation, could become destructive once it takes up residence in the tightly regulated environment of the central nervous system. This hypothesis remains untested in living patients, but the postmortem evidence is striking enough to demand longitudinal follow-up and careful attention to how the same genetic lesion can play out differently in distinct cellular neighborhoods.
Postmortem brain sequencing and the ROSMAP cohort data
The Cell study, identified on PubMed as PII S0092867426003417, applied ultra-deep sequencing to postmortem brain samples and isolated microglia-like cells for variant calling. The researchers found that somatic variants in the clonal hematopoiesis genes TET2, DNMT3A, and ASXL1 were enriched in these brain immune cells in Alzheimer’s cases compared to controls. Variant allele fractions pointed to clonal expansions rather than rare, scattered mutations, implying that a subset of microglia had undergone selective growth driven by these cancer-associated changes.
A separate line of evidence strengthens the case that blood and brain clones are connected. Mosaic chromosomal alterations, another form of clonal change in blood cells, have been linked to Alzheimer’s risk in large genetic datasets. In that work, microglia-annotated brain cells were found to carry the same mCAs detected in blood from the same individuals. That blood-brain match suggests a shared clonal origin rather than independent mutation events arising in each tissue. It also provides a mechanistic bridge between epidemiological observations of mCAs and the cellular pathology seen in affected brains.
An independent peer-reviewed analysis of microglia clonal dynamics in Alzheimer’s, published separately, provided cross-study validation that clonally expanded, mutation-bearing microglia are enriched in affected brains and linked to neuroinflammatory phenotypes. Across datasets, mutation-positive microglia tended to express genes involved in cytokine signaling, phagocytosis, and antigen presentation, consistent with a chronically activated state. Together, these findings build a case that Alzheimer’s disease involves not just amyloid plaques and tau tangles but a distinct disorder of brain immune-cell clones, one that borrows its genetic toolkit from blood-cancer biology and deploys it in the central nervous system.
Unresolved gaps between blood protection and brain pathology
The central tension remains unresolved. CHIP in blood appears protective against Alzheimer’s in large epidemiological cohorts, yet the same mutations in brain microglia correlate with disease. No study has yet tracked mutations longitudinally from the moment they arise in blood through their potential migration into brain tissue. All human evidence so far comes from postmortem samples, which capture a single snapshot at the end of disease rather than the sequence of events that led there. Direct functional assays in living human microglia carrying patient-derived TET2 or DNMT3A variants do not exist; the mechanistic work relies on mouse models and cell-culture systems that cannot fully replicate human brain conditions.
Whether the protective effect in blood and the harmful effect in the brain reflect the same clones acting differently in two environments, or whether they represent distinct biological processes that happen to share a genetic signature, is the question that will shape the next round of research. One possibility is that peripheral CHIP alters systemic inflammation or vascular health in ways that indirectly buffer the brain from Alzheimer’s pathology, even as a minority of those clones eventually seed destructive microglial populations. Another is that different mutations within the same genes, or different combinations of co-occurring variants, determine whether a clone ultimately helps or harms neural circuits.
For the millions of people living with or at risk for Alzheimer’s, the practical consequence is this: blood-based biomarkers that capture CHIP or mosaic chromosomal alterations may not have a simple interpretation. A mutation detected in a blood test could signal a lower risk of dementia on a population level yet still mark a small subset of individuals in whom mutant cells have already infiltrated the brain. Future clinical studies will need to integrate deep blood sequencing with neuroimaging, cerebrospinal fluid markers, and, where possible, single-cell analyses of brain tissue to disentangle these competing effects.
Therapeutically, the emerging picture argues for caution. Strategies that aim to enhance or preserve CHIP-like clones because of their apparent protective association in blood might inadvertently worsen brain inflammation if those same clones are capable of colonizing microglia. Conversely, aggressive attempts to eradicate CHIP for cardiovascular or oncologic reasons could remove a systemic influence that, in some people, helps restrain Alzheimer’s pathology. Any intervention targeting clonal hematopoiesis will need to be evaluated not only for cancer and heart-disease outcomes but also for long-term cognitive trajectories.
What is clear is that Alzheimer’s cannot be fully understood through plaques and tangles alone. The disease also unfolds as a story of aging immune cells, somatic evolution, and clonal competition across tissues. As ultra-deep sequencing and single-cell technologies move from research settings into more routine clinical use, they may reveal that the same mutation can wear multiple faces: a blood-borne guardian in one context, and, once it crosses into the brain, a driver of slow-burning neurodegeneration. Recognizing and mapping that duality will be essential for turning genetic insights into safe, targeted treatments.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
