These are incredibly important studies. I have a C3 mutation that prevents shutdown of the innate immune system's complement cascade. From the inflammation and damage I've experienced after a concussion and then long COVID, I find this research so compelling. Great job!
there have been studies of ameliorating adverse CV CHIP mutation effects--in any of this cardiovascular literature with CHIP series has the issue of memory decline and specific CHIP mutations been examined? The point being is the CHIP mutation reflecting the degree of peripheral blood replacement and its inflammatory diathesis...
in both studies, we find that individuals with CHIP are less likely to be diagnosed with Alzheimer's disease, and less likely to have high levels of Alzheimer's associated neuropathology (like amyloid and tau) in their brains. These protective associations are strongest in individuals with the largest CHIP clones, consistent with a potential causal link.
Would be ironical and extraordinary if JAK2 TET2 or ASXL1 causing the highest CV risk were protective of alzheimer dementia--more likely the other way around..?
Thanks, yes in fact they are! All major CHIP drivers including TET2 and ASXL1 trend protective when associated with Alzheimer's disease, although low sample size limits the statistical power once we start analyzing individual mutations, and we haven't had enough JAK2 samples to specifically analyze JAK2. These data are in Figure 3c here: https://pmc.ncbi.nlm.nih.gov/articles/PMC10353941/
One of our hypotheses is that some of the inflammatory pathways active in macrophages that can worsen cardiovascular disease may have beneficial roles in Alzheimer's disease. For example, if the mutant microglia can help clear amyloid aggregates out of the brain via phagocytosis, that is one mechanism that could be protective in Alzheimer's disease.
It is important to note that these associations are for *large* CHIP clones only, we don't see any association between AD and smaller CHIP clones. Smaller clones have been investigated in separate work from Huang et al in Cell this year (which Eric also touched on in his Ground Truths post).
that helps--and if the phagocytosis story is TET2 specific, then DNMT3A never protecting is more than just a null. Makes me wonder whether the action is at a particular allele inside it..
Kim (Cell Rep 2025) has hematopoietic DNMT3A R882H turning monocyte-derived microglia pathogenic--motor deficits more like atypical parkinsonism than dementia, --Gozdecka (Nature 2025) finds metformin suppresses that clone specifically--R882 CH OR 0.49 (0.32-0.74) in UKB, nothing for non-R882 DNMT3A and nothing for any of the other drivers. a second group got to the same point on their own.
so speaking of ironies and inversions, this inverts in a funny way--the clone metformin takes out is the one with actual CNS toxicity, and is within the gene, your data say never protected the brain anyway. nothing protective actually gets subtracted..
and your point about the large clones makes this all the more interesting-if the effect tracks clone size then shrinking VAF is doing something dose-like, (which cuts in opposite directions depending on which clone you happen to be shrinking..)
Two things I am curious about: has anyone pulled R882 apart from the rest of DNMT3A against AD or parkinsonism? Because they behave like separate animals in marrow, no idea whether they do in brain. and NCT07188740 is enrolling R882 carriers on metformin with VAF as the primary endpoint--if Kim is right the thing to watch might be motor rather than memory, gait or UPDRS-III. One could do an ancillary at baseline and six months and it would cost almost nothing and would be the first look at whether shrinking a clone actually moves anything in the brain..?
❣️✨Fun fascinating independent studies with complementary findings
https://bsky.app/profile/kenaiseasky.bsky.social/post/3mudnyab4h22s
These are incredibly important studies. I have a C3 mutation that prevents shutdown of the innate immune system's complement cascade. From the inflammation and damage I've experienced after a concussion and then long COVID, I find this research so compelling. Great job!
Eric, Julia, Nathan
there have been studies of ameliorating adverse CV CHIP mutation effects--in any of this cardiovascular literature with CHIP series has the issue of memory decline and specific CHIP mutations been examined? The point being is the CHIP mutation reflecting the degree of peripheral blood replacement and its inflammatory diathesis...
yes, we have been investigating the role of CHIP in Alzheimer's disease and cognitive decline, both in the present study and our prior 2023 work: https://www.nature.com/articles/s41591-023-02397-2
in both studies, we find that individuals with CHIP are less likely to be diagnosed with Alzheimer's disease, and less likely to have high levels of Alzheimer's associated neuropathology (like amyloid and tau) in their brains. These protective associations are strongest in individuals with the largest CHIP clones, consistent with a potential causal link.
Would be ironical and extraordinary if JAK2 TET2 or ASXL1 causing the highest CV risk were protective of alzheimer dementia--more likely the other way around..?
Thanks, yes in fact they are! All major CHIP drivers including TET2 and ASXL1 trend protective when associated with Alzheimer's disease, although low sample size limits the statistical power once we start analyzing individual mutations, and we haven't had enough JAK2 samples to specifically analyze JAK2. These data are in Figure 3c here: https://pmc.ncbi.nlm.nih.gov/articles/PMC10353941/
One of our hypotheses is that some of the inflammatory pathways active in macrophages that can worsen cardiovascular disease may have beneficial roles in Alzheimer's disease. For example, if the mutant microglia can help clear amyloid aggregates out of the brain via phagocytosis, that is one mechanism that could be protective in Alzheimer's disease.
It is important to note that these associations are for *large* CHIP clones only, we don't see any association between AD and smaller CHIP clones. Smaller clones have been investigated in separate work from Huang et al in Cell this year (which Eric also touched on in his Ground Truths post).
the irony might run the other way..?
that helps--and if the phagocytosis story is TET2 specific, then DNMT3A never protecting is more than just a null. Makes me wonder whether the action is at a particular allele inside it..
Kim (Cell Rep 2025) has hematopoietic DNMT3A R882H turning monocyte-derived microglia pathogenic--motor deficits more like atypical parkinsonism than dementia, --Gozdecka (Nature 2025) finds metformin suppresses that clone specifically--R882 CH OR 0.49 (0.32-0.74) in UKB, nothing for non-R882 DNMT3A and nothing for any of the other drivers. a second group got to the same point on their own.
so speaking of ironies and inversions, this inverts in a funny way--the clone metformin takes out is the one with actual CNS toxicity, and is within the gene, your data say never protected the brain anyway. nothing protective actually gets subtracted..
and your point about the large clones makes this all the more interesting-if the effect tracks clone size then shrinking VAF is doing something dose-like, (which cuts in opposite directions depending on which clone you happen to be shrinking..)
Two things I am curious about: has anyone pulled R882 apart from the rest of DNMT3A against AD or parkinsonism? Because they behave like separate animals in marrow, no idea whether they do in brain. and NCT07188740 is enrolling R882 carriers on metformin with VAF as the primary endpoint--if Kim is right the thing to watch might be motor rather than memory, gait or UPDRS-III. One could do an ancillary at baseline and six months and it would cost almost nothing and would be the first look at whether shrinking a clone actually moves anything in the brain..?
CAN WE HAVE THE LINKS TO THE ARTICLES PLEASE?
here are the papers!
https://www.nature.com/articles/s41586-026-10939-0
https://www.science.org/doi/10.1126/science.adt8307
thx