Inflammaging and clonal hematopoiesis interplay and their impact on human disease

Abstract

Clonal hematopoiesis of indeterminate potential (CHIP) is an aging-related condition wherein a substantial fraction of circulating leukocytes is descended from a single somatically mutated hematopoietic stem cell (HSC). CHIP increases the risk of hematological malignancies and multiple chronic diseases (for example, cardiovascular pathologies) and contributes to persistent, low-grade inflammation or inflammaging. Inflammaging, in turn, causes functional impairment of normal HSCs, including reduced self-renewal potential. In contrast, CHIP-mutant HSCs not only are resistant to inflammaging-induced functional decline but also gain a selective expansion advantage in an inflammatory environment. A recent surge of discoveries has expanded the clinical and mechanistic understanding of the CHIP–inflammaging interplay, highlighting its broader relevance to age-related diseases and reinforcing the need for timely synthesis of emerging evidence. In this Review, we discuss the current understanding of the molecular and cellular mechanisms that cause CHIP, its interplay with inflammaging, as well as the pathophysiological consequences and the translational implications for diseases of the elderly.

Introduction

Hematopoietic stem cells (HSCs) are the foundational cells of the blood and immune system. They reside in the bone marrow, where they self-renew and sustain lifelong production of all blood cell types, including white blood cells (leukocytes) — such as neutrophils, monocytes, and lymphocytes — red blood cells (erythrocytes), and platelets (thrombocytes)¹. Hematopoietic stem and progenitor cells (HSPCs) is a broader term encompassing both HSCs and multipotent progenitors (MPPs). MPPs represent a downstream stage in the hematopoietic hierarchy that may exhibit distinct lineage biases, such as myeloid- or lymphoid-skewed outputs^1,2. With advancing age, HSPCs in the bone marrow (BM) progressively acquire somatic mutations, some of which are associated with increased self-renewal, giving rise to an outsized fraction of mutant leukocytes^3–6. In the absence of apparent hematological malignancy, this aberrant expansion of mutant hematopoietic clones is known as clonal hematopoiesis of indeterminate potential (CHIP).

CHIP predisposes to increased risk for hematologic malignancies and cardiovascular disease (atherosclerosis and chronic ischemic heart failure [G]), conditions for which advanced age is a major risk factor [G]^7–12. In the last 2–3 years, CHIP has been associated with a higher risk for a growing list of inflammation-associated chronic disorders; these include chronic obstructive pulmonary disease, chronic kidney disease, chronic liver disease, osteoporosis, rheumatoid arthritis, and periodontitis^13–21, although cardiovascular disease remains the disorder most strongly linked with CHIP^12,22 (Figure 1).

Figure 1 |. CHIP as a mechanistic link of comorbidities.

Clonal hematopoiesis of intermediate potential (CHIP) is an aging-related phenomenon in which hematopoietic stem and progenitor cells (HSPCs) acquire somatic mutations resulting in a proliferative advantage (vs. normal HSPCs) and selective expansion of a genetically distinct subset of mutant leukocyte progeny, which thus comprise an outsized fraction of total leukocytes in the periphery^3–6. CHIP-mutant leukocytes exhibit altered phenotypes that not only contribute to systemic inflammation but also promote the pathogenesis of chronic diseases, including cardiovascular disease (CVD), the disorder most strongly linked with CHIP^{9,12,19,52–54,128,130}. CHIP-driven chronic disorders depicted here often coexist as comorbidities; it is thus possible that CHIP acts as a common mechanistic basis that contributes to their emergence in the same patients¹⁹. CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; MASLD, metabolic dysfunction-associated steatotic liver disease.

Aging is accompanied by chronic, low-grade inflammation termed ‘inflammaging’, which enhances the risk for chronic diseases²³. This condition is characterized by elevated concentrations of pro-inflammatory mediators and markers (for example, IL-1, IL-6, TNF, and C-reactive protein) in the blood and tissues, in the absence of overt infection^24,25. Inflammaging may arise from aging-related alterations, such as altered hematopoietic function including CHIP, immune cell dysregulation, cellular senescence-associated proinflammatory secretory phenotype (SASP)^26,27 [G], increased gut permeability and associated endotoxemia [G], enhanced oxidative stress (e.g. due to dysfunctional mitochondria), defective autophagy leading to increased inflammasome [G] activation, dysbiosis of the microbiota, as well as systemic metabolic and hormonal alterations and altered neuro-immune communication^23,28–31 (Figure 2a). Arising as a stochastic event of aging, CHIP can exacerbate inflammation in the elderly and thus contributes to inflammaging³². In turn, inflammation may contribute to the aberrant expansion of CHIP-mutant HSC clones, explaining, in part, their dominance over normal HSCs in the context of aging, as shown by studies in mice^33,34. Despite extensive evidence in experimental animals and humans that inflammation (and/or infection) can promote the expansion of CHIP-mutant clones^35–45, most human data are adjusted for age by applying statistical methods to remove the effect of age from the analysis. This adjustment is necessary to ensure fair comparisons between groups, since age strongly influences disease risk and could otherwise confound the results. Consequently, such studies are not well-suited to directly assess the specific contribution of inflammaging to CHIP development.

Figure 2 |. Multi-level drivers of aging-related chronic inflammation (inflammaging) and clonal hematopoiesis.

a | Major aging-related pathophysiological alterations (acting systemically or in the bone marrow or other tissues) that contribute to inflammaging, i.e., low-grade inflammation without evidence of obvious infection. The list focuses on aspects of relevance to this review and is not exhaustive; the reader is referred to specialized reviews on the topic^23,30,31. CHIP, clonal hematopoiesis of indeterminate potential; DAMPs; damage-associated molecular patterns; HSC, hematopoietic stem cell. b | Although aging is a primary driver of CHIP emergence, this condition is additionally promoted by various genetic, biological, and lifestyle risk factors, as indicated^{4,81,88,89,92–95}. Aging in itself is a complex factor that can exert multiple effects on the hematopoietic stem cells (HSCs) and their bone marrow (BM) niche at different levels, that collectively increase the risk of CHIP. c | Although FOXO activity supports adaptive responses to cellular stress and helps preserve viability, its prolonged activation can trigger cell-cycle arrest or apoptosis. TCL1A appears to enhance the activity of all AKT isoforms, which in turn suppress FOXO-mediated transcription. As a result, TCL1A-expressing HSCs exhibit reduced expression of FOXO target genes and increased expression of genes that promote cell division⁹⁰. By maintaining HSCs in a proliferative state while dampening stress-induced responses, TCL1A may play a critical role in clonal hematopoiesis, as illustrated in the next panel. d | Under normal conditions, the TCL1A promoter is inaccessible in HSCs, resulting in transcriptional repression (top). However, in the presence of driver mutations — such as those in TET2, ASXL1, or spliceosome components like SF3B1 and SRSF2— TCL1A becomes aberrantly expressed (potentially due to the CHIP-associated epigenetic remodeling) and promotes clonal expansion of the mutated HSCs (middle). Notably, the alternative (protective) allele of rs2887399 limits chromatin accessibility at the TCL1A promoter, thereby reducing TCL1A expression and neutralizing the clonal advantage conferred by CHIP-associated mutations (bottom).

Recent studies suggest that CHIP may act both as a driver and a consequence of inflammatory dysregulation. However, the possibility of reverse causality (that is, that inflammation promotes CHIP) requires further validation through prospective longitudinal studies [G]^{9,22,32,35,36,38–49}. A self-sustained feed-forward loop connecting CHIP and inflammaging would contribute to the chronicity of aging-related inflammatory diseases. The CHIP concept has transformed our view of how aging influences inflammation and disease susceptibility.

In this Review we discuss CHIP in the context of aging and inflammaging, with an emphasis on its causes and the molecular and cellular mechanisms involved. To guide the reader through this complex interplay, we first examine how aging, inflammation, and oxidative stress affect HSCs, processes that are related to CHIP. We then discuss the molecular mechanisms and incidence of CHIP, followed by lifestyle and biological factors that promote its development. Subsequent sections explore how inflammation drives the expansion of CHIP-mutant clones and contributes to chronic disease, including cancer. We also consider potentially protective aspects of CHIP in certain contexts, offering a nuanced view of its biological impact. Finally, we examine its relationship with the microbiota and epigenetic ageing, and conclude with translational implications and therapeutic opportunities. A surge of recent discoveries have expanded our understanding of the interplay between CHIP and inflammaging and its clinical significance. CHIP has been increasingly linked to a broader spectrum of chronic diseases, including not only cardiovascular disease but also chronic obstructive pulmonary disease, chronic kidney disease, chronic liver disease, osteoporosis, rheumatoid arthritis, and periodontitis among other^13–21. These associations, highlighted in large-scale cohort studies, underscore CHIP as a systemic risk factor for inflammatory disorders beyond hematologic malignancies. Concurrently, new observations and mechanistic insights indicate a bidirectional relationship between CHIP and inflammaging^{9,14–16,19,20,39–41,42 ,44,45,50–54}. Notably, IL-1 signaling has emerged as a key driver not only of the dysfunction of normal HSCs, but also of the selective expansion of CHIP-mutant clones and the downstream pathological consequences associated with CHIP^9,33,55–60. This has spurred growing interest in targeting inflammation in CHIP carriers to mitigate disease risk. Agents like canakinumab (IL-1β inhibitor) and other anti-inflammasome therapies are under investigation, particularly in individuals with TET2 mutations^{9,33,55,58–60}. All these new developments necessitate an even deeper understanding of the mechanisms that underlie the expansion of CHIP-mutant clones and the CHIP-chronic disease associations, knowledge that will be crucial for developing personalized preventive and therapeutic strategies, an area also explored herein.

Impact of aging, inflammation and oxidative stress on HSCs

To better understand how aging and inflammaging may influence the biology and function of CHIP-mutant clones, it is instructive to first examine their effects on normal HSCs. In the BM, HSPCs are responsible for steady-state and stress-adapted hematopoiesis. Regarding the latter, HSCs can sense systemic inflammation affecting the BM by means of Toll-like or other receptors for pathogen-associated molecular patterns (PAMPs) [G], as well as receptors for cytokines and growth factors (e.g., IL-1β, IL-6, M-CSF)^61–63 (Figure 3a). In response to such stimuli, HSCs can undergo self-renewing divisions and multi-lineage differentiation to ultimately generate increased numbers of mature myeloid cells to replenish those that were consumed in peripheral tissues under emergency conditions (emergency myelopoiesis [G])^1,64,65. However, chronic inflammatory exposure and/or punctuated inflammatory insults (e.g., periodic flares in rheumatoid arthritis) can impair the function of HSCs^6,24. These impairments include reduced self-renewal capacity and a bias toward myeloid differentiation, changes that closely resemble those seen in aged HSCs^3,66,67.

Figure 3 |. Inflammatory and aging-associated drivers of HSC functional alterations.

a | Hematopoietic stem cells (HSCs) express Toll-like receptors (TLRs) that directly respond to pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharide (LPS) that is detected by TLR4. Moreover, HSCs express receptors for cytokines, including IL-1β, and for growth factors, such as M-CSF. These stimuli can drive the proliferation and differentiation of HSCs and downstream progenitors, such as multipotent progenitors (MPPs) and granulocyte-monocyte progenitors (GMPs), leading to enhanced myelopoiesis^61–63. This myeloid-biased differentiation can be protective in certain contexts like acute infections, but chronic inflammatory exposure to these and other infection-associated or inflammatory stimuli can cause functional impairment in HSCs^1,24,63. b | With aging, HSCs undergo functional changes including altered lineage output. This age-associated skewing favors differentiation toward the myeloid and megakaryocytic lineages — resulting in increased production of neutrophils, monocytes/macrophages, and platelets — while diminishing lymphoid output (B and T lymphocytes)^68,69. The myeloid bias contributes to a pro-inflammatory systemic environment, often referred to as ‘inflammaging’, while the decline in lymphopoiesis compromises adaptive immunity. c | In old age, HSCs not only exhibit altered lineage output (panel b), but also show impaired self-renewal capacity and increased vulnerability to apoptosis or cellular senescence⁷⁶. This functional decline is in great part attributed to intrinsic aging-associated alterations within HSCs. Such changes include upregulated pro-inflammatory signalling and stress responses — marked by elevated levels of reactive oxygen species (ROS) and activation of the mechanistic target of rapamycin (mTOR) pathway⁷³ — as well as a diminished ability to perform autophagy⁷⁴. Additionally, aged HSCs demonstrate reduced efficiency in repairing DNA damage⁷⁵ and exhibit epigenetic dysregulation due to aberrant chromatin remodelling^78,79. d | With age, HSCs may acquire somatic mutations that enhance self-renewal, leading to clonal expansion of mutant leukocytes—a condition known as clonal hematopoiesis of indeterminate potential (CHIP). These mutations, often loss-of-function, arise from missense changes in critical domains, nonsense mutations, or small insertions/deletions (top panel). The most frequently affected genes encode the enzymes DNMT3A (DNA methyltransferase 3A), TET2 (Ten-eleven translocation methylcytosine dioxygenase-2), and ASXL1 (Additional sex combs-like 1). DNMT3A and TET2 act on DNA and regulate cytosine methylation and demethylation, respectively, whereas ASXL1, as part of the polycomb repressive complex-2, modifies histones by mediating trimethylation of histone H3 at lysine 27 (H3K27) (lower panels)^5,147. Given that the affected genes encode epigenetic regulators, their inactivation in HSCs disrupts transcriptional control, promoting clonal dominance, altering leukocyte phenotypes, and increasing the risk of age-related diseases.

With aging, HSCs exhibit impaired self-renewal and skewed differentiation — favoring myelopoiesis [G] and megakaryopoiesis [G] at the expense of lymphopoiesis [G]^68,69 (Figure 3b) — and are thus less able than their younger counterparts to restore normal hematopoiesis, e.g., upon transplantation^70–72. Moreover, the number and proportion of HSCs (relative to the total number of cells in the BM) increases with aging and this might, to some degree, compensate for their reduced capacity to regenerate blood cells in old age. Aging-associated intrinsic alterations to the HSCs that contribute to their functional decline include upregulated pro-inflammatory signaling and stress responses (such as increased abundance of reactive oxygen species (ROS) and mechanistic target of rapamycin (mTOR) activation)⁷³, impaired capacity to perform autophagy⁷⁴ or to repair DNA damage^75–77 and epigenetic dysregulation due to aberrant chromatin remodeling^78,79 (Figure 3c). DNA damage accumulates in HSCs with aging and may ultimately lead not only to impaired self-renewal capacity, but also to increased susceptibility to apoptosis or cellular senescence (Box 1)⁷⁶. However, certain DNA damage-induced somatic mutations may become fixed and expand within the aged HSC compartment, giving rise to mutant clones that dominate the hematopoietic output⁸⁰, thereby contributing to clonality, as discussed below.

Box 1 |. Cellular senescence and HSCs.

Cellular senescence is a state of permanent cell-cycle arrest induced by intrinsic or extrinsic stressors^25,183. It represents a stress response to aging-associated challenges, such as telomere attrition, genomic instability/DNA damage, oxidative stress, inflammation, epigenetic dysregulation and oncogene activation^27,184,185. Senescent cells acquire a senescence-associated secretory phenotype (SASP), characterized by secretion of high amounts of pro-inflammatory cytokines, chemokines, growth factors, and metalloproteinases. These mediators can act in a paracrine manner to induce senescence of neighboring healthy cells and spill into the circulation, contributing to inflammaging²³. This reciprocal reinforcement between senescence and inflammation has significant implications for hematopoiesis. Inflammation-induced HSC senescence can destabilize the integrity of the HSC population¹⁸⁶ and persistent infection/inflammation could even deplete the HSC pool¹⁸⁷. Beyond SASP, senescent cells exhibit increased resistance to apoptosis and impaired mitophagy, leading to mitochondrial dysfunction^183,188. This dysfunction may further promote SASP via the release of mitochondrial damage-associated molecular patterns (DAMPs), such as mitochondrial DNA, which can activate different inflammatory signaling pathways, including TLR9, the NLRP3 inflammasome and the cGAS-STING pathway [G]¹⁸⁹. Moreover, disruption of the electron transport chain elevates production of ROS, causing DNA damage that reinforces cell-cycle arrest in a feed-forward loop that links mitochondrial dysfunction and cellular senescence¹⁹⁰. Senescence preferentially affects HSCs rather than Sca-1⁻ (LS⁻K) hematopoietic progenitors, which are thought to be less sensitive to oxidative stress^110,191,192. Besides general pro-senescence factors (inflammation, oxidative stress, physiological aging), other factors that promote HSC senescence include mutations in the ATM gene – which is involved in DNA repair – and serial transplantation^109,112. In response to ROS-induced p38 MAPK activation, HSCs upregulate the senescence marker p16^Ink4a (cyclin-dependent kinase inhibitor), leading to impaired self-renewal and potential exhaustion¹¹². Inflammation-induced premature senescence in HSCs is associated with enhanced ROS-mediated DNA damage¹⁰⁹, increased senescence-associated β-galactosidase (SA-β-gal), and upregulation of tumor suppressor p53 and p16^INK4A109. While p53 and the p21 cyclin-dependent kinase inhibitor contribute to initiation of senescence, their expression wanes as p16^INK4A expression rises, eventually leading to irreversible cell cycle arrest. Notably, deletion of p16^INK4A in old mice diminishes HSC functional decline, thus promoting self-renewal and repopulation capacity¹¹¹. Although senescent HSCs exhibit reduced clonogenicity and defective replicative capacity, they retain the ability to differentiate into progenitors and generate mature progeny before their eventual exhaustion^110,193.

In summary, persistent, recurring or chronic inflammation or other types of stress may impair the self-renewal capacity of HSCs leading to their terminal exhaustion [G]. In contrast, CHIP-mutant clones resist inflammation-induced proliferative stress and gain a selective expansion advantage in an inflammatory environment.

Molecular mechanisms and incidence of CHIP

The emergence of CHIP-mutant clones roughly coincides with the development of inflammaging in the elderly. CHIP affects >10% of individuals older than 65 with a threshold of 2% of variant allele frequency (VAF) [G] of mutations^7,8,80,81.

CHIP-associated mutations typically result in loss-of-function effects, which arise from missense mutations [G] in critical domains, nonsense mutations [G], or base-pair insertions/deletions. The mutations most commonly affect three genes encoding epigenetic modifier enzymes: DNMT3A (DNA methyltransferase 3A), TET2 (Ten-eleven translocation methylcytosine dioxygenase-2), and ASXL1 (Additional sex combs-like 1). DNMT3A and TET2 control cytosine methylation and demethylation, respectively, and mutations in these genes account for almost 65% of acquired CHIP mutations. ASXL1 is a component of the polycomb repressive complex-2 [G] that mediates trimethylation of H3K27, a repressive epigenetic modification (Figure 3d).

Besides commonly affected genes (DNMT3A, TET2 and ASXL1), CHIP mutations are also found in genes encoding splicing machinery factors (U2AF1, SRSF2), DNA damage repair genes (PPM1D, TP53, CHEK2), and inflammatory signaling molecules (JAK2, STAT3, MYD88)⁴⁷. Blood DNA sequencing from ~50,000 individuals followed by population genetic modeling revealed that age-related accumulation of mutations are associated with fitness advantages, thus implying that positive selection, rather than genetic drift [G], drives clonal hematopoiesis⁸².

The prevalence of CHIP increases further with increasing age and can eventually affect >60% of individuals ≥80 years old⁸³. Highly sensitive sequencing technology, which could detect CHIP mutations as rare as 0.03% VAF, showed that almost all (95%) individuals 50-to-60 years of age exhibit CHIP, predominantly involving DNMT3A and TET2 mutations. Although the clinical significance of such very small clones is questionable^84,85, patients with chronic ischemic heart failure exhibit impaired survival associated with even significantly lower VAFs for DNMT3A and TET2 mutations than the classical CHIP definition (VAF ≥2%)⁸⁶.

Lifestyle and biological factors that promote CHIP

In the absence of overt disease, aging is the primary factor for developing CHIP⁴ (Figure 2b). The aging-related increased accumulation of mutations in HSCs may in part be attributed to their increased resistance to DNA damage-induced apoptosis, as compared to young HSCs⁷⁷. In other words, aged HSCs can suffer DNA damage without necessarily undergoing apoptotic death, thereby accumulating mutations⁷⁷. Although the mechanisms whereby aging fosters CHIP are poorly understood, it is conceivable that aging-associated factors, such as cellular senescence, metabolic alterations including mitochondrial dysfunction, and epigenetic drift [G], might alter the HSC niche in a manner that favors the somatic evolution and/or expansion of CHIP-mutant HSCs^11,87.

Heredity also influences the acquisition of CHIP mutations. For instance, several inherited polymorphic genomic loci, such as a common deletion (rs34002450) mapping to the telomerase reverse transcriptase (TERT) locus, predispose to elevated risk of CHIP^81,88,89. Intriguingly, a common inherited polymorphism (rs2887399) in the promoter of the T cell lymphoma/leukemia 1A gene (TCL1A) is associated with reduced growth rate of HSC clones carrying CHIP mutations in TET2, ASXL1, or spliceosome factors (SF3B1, SRSF2)⁹⁰. This effect is not observed in DNMT3A-mutated HSC clones which, along with WT HSCs, do not express TCL1A. TCL1A expression confers a fitness advantage in clones carrying the above-mentioned mutated CHIP-driver genes because TCL1A activates the AKT pathway and suppresses FOXO-driven transcription, helping HSPCs resist stress and maintain proliferation⁹⁰ (Figure 2c). The protective variant appears to block accessibility to the promoter of TCL1A, which is otherwise aberrantly de-repressed in HSCs downstream of TET2, ASXL1, SF3B1, and SRSF2 mutations⁹⁰ (Figure 2d). Beyond the influence of, or irrespective of, genetics and age, chronic diseases with an inflammatory component, chronic infections, chemotherapy/radiation, and lifestyle behavior (e.g., smoking, unhealthy diet and obesity) may further accelerate CHIP^91–95 (Figure 2b).

Exposure to chemotherapy or radiation treatment for a primary malignancy modulates the fitness landscape of CHIP by selecting for mutations in DNA damage response (DDR) genes (e.g., PPM1D and TP53)^96,97. Sequential blood sampling from patients with solid tumors who underwent chemotherapy or radiation therapy showed that clones with mutations in DDR genes outcompete clones with non-DDR CHIP mutations⁹⁷. Consistently, truncating PPM1D mutations not only confer chemotherapy resistance, but also cause selective expansion of PPM1D-mutant hematopoietic cells (vs. WT controls) during exposure to chemotherapy in vitro (human cells) or in vivo (mice transplanted with PPM1D-mutant or control hematopoietic cells)⁹⁸. Similarly, in hematopoietic competition assays in mice, HSPCs with heterozygous Trp53 missense mutations exhibit a competitive fitness advantage, relative to control HSPCs, upon DNA damage (sublethal gamma-irradiation)⁹⁹. Thus, PPMID- and TP53-mutant HSPCs appear to acquire relative competitive fitness under the selective pressure of genotoxic stress owing to increased resistance to DNA damage-induced apoptosis.

Smoking can act as a mutagenic factor in CHIP by causing DNA damage and/or compromising DNA repair, thus explaining, at least in part, its epidemiological association with CHIP^89,96,100. A recent Mendelian randomization analysis showed that smoking is causal for overall CHIP and for specific CHIP-mutant clones (ASXL1 and DNMT3A)⁹⁵. Moreover, exposure to cigarette smoke promotes the expansion of CHIP-mutant clones in transplanted mice over time, relative to controls exposed to ambient air¹⁰¹. Mechanistically, this effect may be driven by cigarette smoke-induced low-grade inflammation and oxidative stress¹⁰². In brief, smoking can enhance CHIP both by augmenting mutagenesis and contributing to the expansion of the mutant clones.

Individuals with obesity display a higher prevalence of CHIP-associated mutations (most commonly in DNMT3A), likely due to obesity-induced inflammation — including within the BM — that may promote clonal expansion⁹⁴. Supporting a causal relationship, studies in obese mice with CHIP (driven by heterozygosity of Tet2, Dnmt3a, Asxl1, or Jak2) showed that the presence of ‘fatty BM’ (FBM) enhances the expansion of CHIP clones relative to lean controls. The obesity-induced enhanced growth of mutant clones was suppressed by anti-inflammatory treatments⁹⁴. Moreover, both human and mouse HSCs carrying mutated DNMT3A display increased self-renewal in FBM environment, which is counteracted by anti-IL-6 neutralizing antibodies, thus suggesting that FBM-derived inflammatory signals promote DNMT3A-driven CHIP via a paracrine mechanism¹⁰³. The age-associated decline in sex hormones can indirectly influence CHIP, in part through changes in FBM, the accumulation of which is negatively correlated with the serum levels of sex hormones in both men and women¹⁰⁴. Consistently, castration in male mice¹⁰³ and ovariectomy in females¹⁰⁵ leads to increased FBM, an effect that is reversed by estrogen supplementation in mice¹⁰⁵ and mirrored in postmenopausal women on estrogen treatment¹⁰⁶. A prospective 20-year study showed that the size (VAF) of CHIP clones increased with age in individuals with obesity receiving standard care, but not in those who underwent bariatric surgery¹⁰⁷, a procedure known to attenuate obesity-related sequelae, including low-grade chronic inflammation¹⁰⁸. Notably, the rate of clone growth correlated positively with insulin resistance and negatively with HDL-cholesterol levels¹⁰⁷.

Inflammation and growth of CHIP-mutant clones

As discussed above, chronic or recurrent inflammatory insults impair the fitness of normal HSCs, in part by inducing p53-mediated apoptosis or cell-cycle arrest^{63,72,109–113}. In such settings, HSCs are repeatedly driven to self-renew, proliferate, and differentiate, rendering them particularly vulnerable to replication-associated DNA damage and mutations. Inflammation-induced proliferative stress, often accompanied by ROS generation, may result in DNA lesions that are not properly repaired. Over time, this selective pressure may favor the emergence and expansion of mutant HSCs that are better adapted to inflammatory environments and can outcompete normal HSC clones (Figure 4a).

Figure 4 |. Differential effect of inflammation on CHIP-mutant and normal HSCs.

a | Depending on intensity and chronicity, systemic inflammation may act on hematopoietic stem cells (HSCs) and promote functional decline (‘aging-associated hematopoietic phenotype’), including possible exhaustion^24,112. In contrast, CHIP-mutant clones can resist the inflammatory stress and acquire selective advantage for survival and expansion at the expense of normal clones^{35,50,120,181}. Thus, inflammation decreases the fitness of normal HSCs and increases the fitness of CHIP-mutant HSCs. b | IL-6–driven inflammation promotes the survival of TET2-mutant HSPCs by activating SHP2 and STAT3 signaling pathways. This activation enhances the expression of the anti-apoptotic protein BCL2 and the long non-coding RNA MORRBID, which in turn suppresses the pro-apoptotic molecule BIM, thereby preventing apoptosis (normal HSPCs are suppressed under the same inflammatory conditions)³⁶.

Inflammation promotes self-renewal in CHIP-mutant clones

Whereas normal human HSCs in a CHIP-mutant setting display inflammatory and aging-related transcriptomic signatures, in the same samples these detrimental effects are attenuated in HSCs with mutations in TET2 or DNMT3A¹¹⁴. In the setting of TET2-driven CHIP in mice, the expansion of Tet2^+/– clones increases with aging in an IL-1R-dependent manner and in parallel with elevated IL-1α and IL-1β inflammation in the BM³³. IL-1β increases the self-renewal capacity of TET2-deficient HSPCs by enhancing the expression of genes that promote HSPC maintenance while restricting their differentiation (e.g., Erg, Meis1, and Egr1)⁵⁶. In contrast, IL-1β-induced stress rather exhausts WT HSPCs⁵⁷. In normal HSCs exposed to chronic IL-1β inflammation, the transcription factor PU.1 limits protein synthesis and cell-cycle activity, in an effort to prevent aberrant HSC expansion¹¹⁵. PU.1-deficient HSCs lack this regulatory mechanism and overexpress cell-cycle and protein synthesis genes and undergo excessive expansion in response to chronic IL-1β signaling. Interestingly, TET2 mutations can cause aberrant methylation at PU.1-binding sequences¹¹⁶. This alteration might interfere with the function of PU.1 and thus enable TET2-mutant HSCs to engage in aberrant protein synthesis and cell-cycle activity, thereby explaining, in part, their disproportionate expansion under inflammatory stress^9,33.

Inflammation-driven clonal expansion through cell death suppression

Another mechanism whereby TET2-deficient HSCs can outcompete normal HSC clones in an inflammatory BM environment is by upregulating anti-apoptotic pathways in response to inflammatory cytokines, such as TNF and IL-6^35,36. Relative to WT controls, Tet2^−/− HSPCs cultured for 24 days under semisolid conditions in the presence of TNF express increased mRNA levels of anti-apoptotic molecules (Bcl2 and Birc2) and decreased mRNA levels of pro-apoptotic molecules (Tnfrsf1a, Tnfrsf1b, Fas, Casp3, and Casp8)³⁵. Moreover, IL-6 stimulation of Tet2^−/− HSPCs activates non-receptor protein tyrosine phosphatase SHP2− and signal transducer and activator of transcription 3-mediated (STAT3)−mediated signaling. This leads to enhanced expression of BCL2 and the long non-coding RNA MORRBID, which in turn suppresses the expression of the pro-apoptotic molecule BIM³⁶ (Figure 4b). TET2-deficient HSPCs display enhanced activation of the cGAS-stimulator of interferon genes (STING) pathway, which in turn contributes to their clonal expansion³⁷. Contrastingly, normal HSCs express an endogenous cGAS inhibitor (circular RNA antagonist for cGAS) in an effort to downregulate the cGAS-STING pathway and hence potential exhaustion from cGAS-STING–mediated type I interferon responses¹¹⁷.

Global methylation changes were associated with decreased stress-induced apoptosis and increased self-renewing capacity of Dnmt3a^−/− HSCs following infection-driven inflammation³⁸. In fact, not only does loss of Dnmt3a promote self-renewal, but also confers essentially indefinite longevity in affected HSCs as determined in serial (twelve) transplantations over 60 months, i.e., surpassing the lifespan of the mice from which they were originally derived¹¹⁸. Similarly, the loss-of-function DNMT3A R878H mutation endows the mutant HSCs with resistance to necroptosis [G]¹¹⁹. Specifically, Dnmt3a^R878H/+ HSCs suppress RIPK1-RIPK3-MLKL (receptor interacting protein kinase 1–receptor interacting protein kinase 3–mixed lineage kinase domain-like)–mediated necroptosis in response to inflammatory (TNF) or proliferative (transplantation) stress¹¹⁹. Consistently, the selective expansion of Dnmt3a^R878H/+ HSCs (vs. WT HSCs) upon their transplantation into middle-aged (13–15 month-old) recipient mice was dependent on TNF signaling, since the effect was abrogated in Dnmt3a^R878H/+ Tnfrsf1a^−/− HSCs, i.e., which carry the CHIP mutation and additionally lacking the TNFR1 receptor¹²⁰.

Infection-associated IFN signalling fuels clonal expansion.

IFNγ signaling fuels the expansion of Dnmt3a^−/− HSCs in mice with chronic mycobacterial infection³⁸. Whereas IFNγ signaling can be detrimental to the fitness of WT HSC, Dnmt3a-mutant HSC (carrying the heterozygous R878H mutation) resist IFNγ-induced apoptosis and sustain their quiescence by upregulating the thioredoxin-interacting protein (Txnip)-p53-p21 pathway¹²¹. Specifically, the defective function of mutated DNMT3A causes loss of DNA methylation at the Txnip promoter leading to upregulation of TXNIP which stabilizes p53, in turn upregulating p21 that helps prevent IFNγ-induced proliferation and apoptosis of HSCs¹²¹.

Although the evidence that inflammation favors the expansion of CHIP-mutant clones is primarily derived from in vitro mechanistic models and experimental animal studies, it aligns well with clinical observations. For instance, the finding that IFNγ promotes the expansion of DNMT3A mutant clones³⁸ is consistent with observations in ulcerative colitis patients, who not only have increased prevalence of CHIP but also display elevated serum IFNγ levels associated with the presence of DNMT3A mutant CHIP clones⁵⁰.

CHIP and chronic disease

Causal connections were initially established between CHIP and cardiovascular disease, followed by additional experimental studies implicating CHIP as a contributor to the pathology of other chronic diseases^12,22. The underlying mechanisms have been investigated in mouse models. It is noteworthy that most of the discussed experimental studies rely on competitive BM transplantation following myeloablation [G]. This model has inherent limitations, such as induction of vascular damage¹²², which may complicate the study of cardiovascular disease, but may also injure the bone marrow niche¹²³ and thus affect the hematopoietic process. Nevertheless, competitive BM transplantation following myeloablation has recapitulated important aspects of human CHIP. For example, the clonal expansion advantage of CHIP-mutant clones is accompanied by a modest bias toward myelopoiesis in Tet2-driven CHIP⁹, while no specific hematopoietic lineage is preferentially influenced by Dnmt3a mutations^19,124. Moreover, as discussed below, the model has shown predictive validity, and treatments used in mouse CHIP are relevant in human CHIP^{9,58–60,81,125,126}. For additional information on the utility and limitations of animal models of CHIP, the reader is referred to a recent review¹²⁷.

Cause-and-effect links between CHIP and chronic disease

Studies modeling human TET2 loss-of-function mutations showed that BM transplantation from Tet2^−/− mice into low-density lipoprotein receptor-deficient (Ldlr^−/−) mice enhances atherosclerosis^9,128. Increased plaque formation was attributed to NLRP3 inflammasome-mediated IL-1β secretion by Tet2-deficient macrophages⁹. TET2 CHIP carriers also show elevated serum IL-1β⁸¹. In both mice and humans, TET2 deficiency synergizes with hyperlipidemia to activate NLRP3 and promote atherosclerosis¹²⁹. Other CHIP mutations causally linked to cardiovascular disease in mice, include those affecting Dnmt3a⁵⁴, Ppm1d, a cancer therapy-related CHIP-driver gene¹³⁰ and Jak2⁵³. The Jak2^V617F gain-of-function mutation accelerates atherosclerosis in Ldlr^−/− mice, and this effect is attenuated by the IL-1 receptor antagonist Anakinra, which inhibits necrotic core formation in atherosclerotic lesions⁵³. Additionally, Jak2^V617F-driven CHIP promotes platelet activation and enhances arterial thrombosis¹³¹. CHIP-mutant myeloid cells may promote heart failure via interactions with cardiac fibroblasts, as shown for DNMT3A mutations¹³². DNMT3A-silenced monocytes release heparin-binding EGF-like growth factor and activate fibroblasts. Consistently, in a Dnmt3a-driven CHIP model, mice subjected to experimental myocardial infarction develop heart failure accompanied by increased cardiac interstitial fibrosis [G]¹³². Tet2-driven CHIP also worsens experimental heart failure via the NLRP3/IL-1β axis⁵².

DNMT3A-driven CHIP has been associated with increased prevalence of periodontitis in humans and causally implicated in inflammatory bone loss in mouse models of periodontitis and arthritis, driven by heightened neutrophil and Th17 responses and osteoclastogenesis¹⁹. These findings suggest CHIP as a shared mechanism underlying comorbidities such as periodontitis and rheumatoid arthritis^133,134 (Figure 1). CHIP was causally associated with additional arthritic and bone loss disorders, namely gout¹³⁵ and osteoporosis¹⁸, in studies using TET2-driven and DNMT3A-driven models of CHIP, respectively. Similar mutant TET2-driven and/or mutant DNMT3A-driven mouse models have causally linked CHIP to chronic obstructive pulmonary disease (COPD)¹⁴, metabolic dysfunction-associated steatohepatitis¹⁵ and kidney fibrosis¹³⁶, supporting epidemiological associations of CHIP with the respective chronic disease in the lung¹⁴, liver¹⁵ and kidney¹⁶.

Bidirectional links between CHIP and chronic disease

Mathematical modeling and empirical data show that increased HSC division in atherosclerosis accelerates somatic evolution by raising the likelihood of CHIP-driver mutations³⁹. In line with this, hypercholesterolemia-induced atherosclerosis in a mouse model of TET2-driven CHIP increased HSC proliferation and expanded the mutant fraction of monocytes and neutrophils³⁹. The model may also explain shorter telomeres in CHIP leukocytes⁸⁹, though it relies on assumptions that may not fully capture the complexity of the hematopoietic system.

A 6-year longitudinal study in healthy middle-aged individuals found that CHIP (mainly DNMT3A or TET2 mutations) increased risk of de novo femoral atherosclerosis⁴⁹. This argues against reverse causality and adds temporal insight beyond cross-sectional studies. However, findings may not generalize to other vascular beds, and lack of clone expansion does not exclude effects on de novo somatic mutagenesis⁴⁹. Supporting this, hypercholesterolemia failed to expand Tet2-mutant clones in mice⁹. In contrast, myocardial infarction promoted Jak2^V617F clone expansion⁴⁸, suggesting that the presence or absence of reverse causality may be context dependent and influenced by the severity of inflammation.

Consistent with this context-dependent effect, CHIP is disproportionately prevalent in systemic sclerosis patients under 50 (25% vs. 4% in controls), suggesting disease-driven acceleration⁴⁴. Similarly, CHIP arises early in systemic lupus erythematosus, including in women of childbearing age when it is typically rare⁴⁵. Though uncommon under age 40, some young IBD patients (≤45 years) show myeloid CHIP (e.g., TET2, ASXL1) with high (>20%) VAF, which increases with age⁴⁰. Importantly, anti-TNF therapy is inversely associated with myeloid CHIP prevalence⁴⁰. Patients with Vacuoles, E1 enzyme, X-linked, Autoinflammatory, Somatic (VEXAS) syndrome exhibit enriched CHIP mutations, likely due to chronic inflammation⁴¹. HIV-positive individuals have higher CHIP incidence and faster clonal expansion than HIV-negative controls^42,43, consistent with an inflammatory environment that promotes CHIP-mutant growth. The role of IL-6 in promoting the fitness of Tet2−/− clones^36,137 is consistent with genetic evidence that the IL6R missense variant p.Asp358Ala specifically hinders TET2 clonal expansion⁵¹. Collectively, these findings support reverse causality, warranting validation through prospective studies using sensitive targeted sequencing.

The notion of reverse causality is strongly supported by experimental studies. Inflammation exerts selective pressure that promotes the aberrant expansion of CHIP-mutant HSPCs^35,72,120, creating a permissive environment for clonal hematopoiesis. This is further reinforced by mutant myeloid progeny that sustain or amplify the inflammatory response. For example, in Tet2^H1881R-driven CHIP, aged mice gradually develop an inflammatory BM milieu with elevated IL-1β and other cytokines, coinciding with the emergence of MHC-II^hi Tet2 mutant monocytes³⁴. With advancing age, this model progresses to myeloid malignancy, with high mortality by 100 weeks³⁴. In the BM stromal niche, mesenchymal stromal cells (MSCs) exposed to Dnmt3a^R878H/+ HSPCs upregulate senescence markers (e.g., SA-β-gal and BCL-2) both in vivo and ex vivo via a contact-independent mechanism replicable by inflammatory cytokines (IL-6 and TNF)⁸⁷. In vivo depletion of senescent MSCs by senolysis [G] attenuates inflammation and limits the expansion of Dnmt3a^R878H/+ hematopoietic cells⁸⁷.

Together with clinical observations linking systemic inflammation or infection to enhanced CHIP^{40–45,50,51}, and preclinical and clinical studies in the preceding sub-section showing that CHIP-mutant clone expansion exacerbates cardiovascular and other chronic diseases^{9,14–16,18,19,49,52–54,128,130}, these findings support a bidirectional relationship between CHIP and inflammation (Figure 5). Thus, not only does CHIP promote inflammation, but inflammation may also amplify CHIP-mutant clone expansion, potentially establishing a self-sustaining feed-forward loop that contributes to the onset and chronicity of inflammatory comorbidities in the elderly.

Figure 5 |. Feed-forward loop connecting CHIP and inflammation.

The expansion of CHIP-mutant clones causally contributes to inflammatory diseases, whereas systemic inflammation associated with these disorders may accelerate somatic evolution and clonal hematopoiesis as a consequence of increased HSC proliferation^{35,38–41,44,45,50}. The resulting increased generation of CHIP-mutant and hyper-inflammatory leukocytes can further exacerbate chronic disease^{9,19,53,54,130}, thereby generating a feed-forward loop, in which CHIP and chronic disease are reciprocally reinforced.

Mechanisms of inflammation amplification by CHIP-mutant leukocytes

CHIP-mutant leukocytes can amplify inflammation via mechanisms that are not necessarily dependent on the loss of catalytic activity of the mutated epigenetic modifier enzymes. In wild-type macrophages, independently of its role in DNA methylation, TET2 recruits histone deacetylase (HDAC)-2 to the IL-6 promoter resulting in suppressed Il6 transcription through histone deacetylation. The absence of this regulatory mechanism in Tet2^−/− macrophages enhances IL-6 production upon activation¹³⁸ (Figure 6a). TET2 employs a similar mechanism at the IL-1β promoter to suppress Il1b transcription⁹. Moreover, TET2 regulates the expression of the NLRP3 inflammasome, thereby controlling the secretion of mature IL-1β by activated macrophages. In contrast, macrophages lacking TET2 or harboring loss-of-function mutations fail to exert these regulatory controls, resulting in excessive production of IL-1β in response to infectious or inflammatory stimuli⁹ (Figure 6a). Nevertheless, lack of catalytic activity alone in Tet2 mutant clones (missense mutation H1881R) is sufficient to lead to enhanced inflammation relative to normal clones³⁴. TET2 and DNMT3A maintain the integrity of mitochondrial DNA (mtDNA) by controlling the expression of the mtDNA-binding protein TFAM (transcription factor A mitochondria) via a mechanism that is unrelated to DNA methylation. TET2 or DNMT3A loss of function in human macrophages, as occurs in CHIP, leads to mtDNA release and activation of cyclic GMP-AMP synthase (cGAS) signaling, which induces Type-I IFNs¹³⁹. Additionally, induced T regulatory cells (iTregs) harboring the heterozygous Dnmt3a R878H mutation display significantly impaired immunosuppressive activity relative to Dnmt3a^+/+ iTregs, presumably leading to unrestrained T cell effector responses¹⁹.

Figure 6 |. CHIP-mutant cells promote inflammation via both cell-intrinsic and paracrine effects.

a | In wild-type macrophages, TET2 facilitates the recruitment of histone deacetylase (HDAC) to the IL-6 promoter, thereby suppressing Il6 transcription through histone deacetylation¹³⁸. A similar mechanism operates at the IL-1β promoter, where TET2 downregulates Il1b transcription⁹. Additionally, TET2 limits the expression and activity of the NLRP3 inflammasome, reducing the release of mature IL-1β by activated macrophages. Contrastingly, TET2-deficient macrophages or those harboring TET2 loss-of-function mutations lack these regulatory mechanisms and thus IL-6 and IL-1β are overproduced in response to infection or inflammation, exacerbating inflammatory conditions such as atherosclerosis⁹. Ac denotes acetylated histone. b | Activated CHIP-mutant cells release cytokines and/or chemokines that can activate gene transcription in non-mutant cells within proximity. c | Activated CHIP-mutant macrophages secrete the indicated Th17-inducing cytokines resulting in enhanced release of IL-17 from WT Th17 cells. IL-17 in turn activates downstream targets, such as WT fibroblasts¹⁴⁰. d | Paracrine activation of WT platelets by TXA2-secreting Jak2^V617F platelets resulting in increased arterial thrombosis¹³¹. LPS, lipopolysaccharide; RANK, receptor activator of nuclear factor-κB ligand; RANKL, RANK ligand, TXA2, thromboxane A2.

Bioinformatic analyses of intercellular communication networks in humans¹⁴⁰ or mice¹⁹, supported by in vitro and in vivo findings^19,140, indicate that CHIP-mutant cells can promote inflammation not only in a cell-intrinsic manner but also by activating neighboring wild type cells via paracrine effects (Figure 6a). For instance, DNMT3A-silenced macrophages induce activation of naive CD4⁺ T cells leading to increased IL-17 release that can subsequently activate cardiac fibroblasts¹⁴⁰ (Figure 6b), a contributory mechanism to enhanced fibrosis and heart failure (see above). A study in a mouse model of JAK2^V617F-driven CHIP demonstrated paracrine activation of WT platelets by thromboxane A2 secreted from mutant (Jak2^V617F) platelets¹³¹ (Figure 6c). Moreover, ex vivo osteoclastogenesis assays with Dnmt3a^–/– BM cells, or with Dnmt3a^R878H/+ BM cells enriched for osteoclast precursors (CD11b^–/loLy6C^hi cells¹⁴¹), yielded significantly more osteoclasts (than WT control cultures) attributable to paracrine secretion of inflammatory/pro-osteoclastogenic mediators^18,19. Overall, paracrine activation mechanisms can amplify pro-inflammatory and pro-osteoclastogenic circuits involving both mutant and WT cells and, therefore, inflammatory tissue damage in CHIP-driven disorders may not be mediated solely by cell-intrinsic effects of the CHIP mutations.

Inflammation as a link between CHIP and cancer

CHIP is a strong risk factor for subsequent development of blood cancers⁸. In this context, inflammation is a key driver of the progression of CHIP to myeloproliferative neoplasms (MPNs), myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). Although these interrelated disorders can arise in a stepwise fashion, CHIP is not necessarily a requirement. Thus, MPNs, MDS and AML can also develop de novo from normal hematopoiesis, and CHIP may progress directly to AML¹⁴².

Chronic immune stimulation, as suggested by history of infectious or autoimmune diseases, is associated with increased risk of CHIP-associated conditions, such as AML and MDS¹⁴³. Moreover, patients with MDS and chronic myelomonocytic leukemia have increased prevalence of antecedent diseases of inflammaging (chronic cardiovascular, metabolic, pulmonary and bone and joint disorders) as compared to healthy individuals or patients with solid malignancies¹⁴⁴. Thus, chronic inflammation may not only promote the expansion of CHIP clones, but may also act as a trigger for MDS and/or AML development.

CHIP appears to exert an adverse impact on the survival from solid (non-hematological) tumors⁹⁶. The underlying mechanism(s) are not well understood although CHIP-mutant clones could influence cancer progression (or recurrence) possibly through interactions with tumor cells^96,145,146. Consistently, CHIP-associated mutations (such as those affecting DNMT3A and TET2) are frequently detected in the leukocytes that infiltrate primary breast cancers¹⁴⁵. Such interactions might include cell-cell contact or paracrine release of cytokines and chemokines¹⁴⁷, thus the clonally derived effector cells may create an environment that is conducive for cancer growth. In this regard, a recent study showed that the infiltration of CHIP-mutant leukocytes into the microenvironment of solid tumors leads to dysregulated inflammatory gene expression and adverse clinical outcomes¹⁴⁸.

Mimicking the effect of CHIP mutations that impair DNMT3A function, inflammaging causes declined expression of DNMT3A, in turn leading to enhanced IL-1⍺ production and promotion of solid tumor growth¹⁴⁹. Inhibition of IL-1 receptor signaling using the antagonist anakinra early during tumor development in mice attenuated the growth of lung, colonic, and pancreatic cancers¹⁴⁹, suggesting that a similar anti-tumor approach may be relevant in the context of DNMT3A-driven CHIP.

Potentially protective effects of CHIP

CHIP may also provide host protection in certain contexts. In elderly individuals with impaired hematopoiesis due to the functional decline of normal HSCs, the enhanced expansion capacity of CHIP-mutant HSCs may lead to sufficient production of hematopoietic cells, thereby compensating for the failure of senescent or dysfunctional HSCs¹⁵⁰. Analysis of ‘common’ (60–89 years) and ‘longevous’ (≥90 years) groups of elderly individuals suggested that certain CHIP mutations (e.g., in TET2 but not DNMT3A) might help prolong life¹⁹. Allogeneic hematopoietic stem-cell transplantation from donors with CHIP appears to reduce relapse/progression rates in recipients with hematological malignancies, possibly due to enhanced graft-versus-leukemia responses mediated by donor-derived CHIP-mutant cells¹⁵¹. Consistently, TET2-deficient CAR-T cells are more effective in tumor cell killing than TET2-sufficient controls¹⁵². CHIP was intriguingly associated with reduced risk of Alzheimer’s disease and lower abundance of neuritic plaques¹⁵³. This connection may be causal according to Mendelian randomization analysis, although the underlying mechanism is not clear¹⁵³.

Microbiota, hematopoietic inflammaging and CHIP

Studies in young mice have shown that intestinal microbiota-derived structural components and metabolites (e.g., bacterial DNA, LPS, short-chain fatty acids, peptidoglycan, lactate) reach the BM via the systemic circulation and contribute to the maintenance of myelopoiesis and the HSPC pool size^154–158. However, alterations to the gut epithelial barrier, the function of which is compromised in old age in both mice¹⁵⁹ and humans¹⁶⁰, may render the gut microbiota potentially detrimental for the hematopoietic system. Disruption of the gut epithelial barrier leads to increased translocation into the circulation of microbe-derived compounds, which elevate systemic inflammation¹⁵⁹ (Figure 7). Thus, compared to young (2-month old) specific-pathogen free (SPF) WT mice, aged (2-year old) SPF WT mice have increased blood levels of PAMPs at steady state²⁴. This aging-related increase in circulating PAMPs drive increased production in the BM of IL-1α and IL-1β, which in turn act on HSCs and promote their aging-associated hematopoietic phenotype (e.g., impaired regenerative capacity and myeloid-biased differentiation)²⁴.

Figure 7 |. Impact of disrupted gut epithelial barrier on hematopoiesis.

The gut epithelial barrier may become impaired in old age or due to intestinal infection and/or microbial dysbiosis^159,160,182. A defective barrier may in turn lead to increased leak into the circulation of microbe-derived compounds, which cause systemic inflammation^137,159. Systemic inflammation is sensed by the bone marrow and may lead – especially when persistent or chronic) – to hematopoietic inflammaging, in the case of normal HSCs^1,24. On the other hand, CHIP-mutant HSCs thrive in an inflammatory environment which exerts a selective pressure that drives their aberrant expansion^35,72,120.

Under conditions of impaired intestinal barrier integrity, the gut microbiome substantially influences the expansion of CHIP-mutant clones (Figure 7). In Tet2^−/− mice, TET2-deficient proinflammatory myeloid cells disrupt the gut barrier integrity and lead to disseminated bacteria-induced systemic inflammation that amplifies the expansion of TET2-deficient clones, which is reversed by anti-IL-6 treatment¹³⁷. The myeloproliferation in Tet2^−/− mice is also counteracted by antibiotic treatment or by rederiving and maintaining Tet2^−/− mice under GF conditions^137,161. Aging-associated alterations to gut permeability and the gut microbiota, therefore, act as upstream extrinsic factors that contribute to hematopoietic inflammaging. In summary, in old age, the microbiota not only can aggravate HSC aging but may also contribute to the selective expansion of CHIP-mutant HSC clones^24,35,50,72.

CHIP and epigenetic aging

CHIP has been associated with accelerated epigenetic aging^162,163, i.e., a higher biological than chronological age, as estimated by algorithms based on the methylation state of specific CpG sites in the genome¹⁶⁴. A recent study of 4,370 CHIP-carriers has associated DNA methylation biomarkers of aging with the CHIP clonal expansion rate¹⁶⁵. Whereas such associations lack directionality, it is conceivable that CHIP-associated inflammation may accelerate biological aging, which in turn may exacerbate CHIP. As a biomarker of biological aging, epigenetic age has been associated with age-related morbidities and all-cause mortality^166,167. Importantly, CHIP interacts with epigenetic aging to increase all-cause mortality¹⁶³. Thus, combined information on CHIP and epigenetic aging may help identify individuals at high risk for aging-related pathologies and guide therapeutic interventions¹⁶³. In this regard, among individuals with CHIP, only a subset (40%) of those with accelerated epigenetic aging also had elevated cardiovascular disease risk¹⁶³.

Translational implications and potential for therapeutic approaches

Screening for CHIP in elderly patients may facilitate precision-medicine approaches to reduce the risk for inflammatory comorbidities. Such individuals may benefit not only from immunological interventions that suppress the pathogenic activity of CHIP-mutant effector cells^{9,15,52,53,58}, but also from preventive treatments that block the aberrant expansion of CHIP-mutant clones^{19,33,56,125,126,168} or promote the function of epigenetic modifier enzymes affected by CHIP^169–172.

Inhibition of downstream pathological effects of CHIP

NLRP3 inflammasome-driven IL-1β release by macrophages promotes the formation of atherosclerotic lesions in animal models of CHIP, suggesting that inhibitors of inflammasomes or the cytokine (or its receptor) may provide a therapeutic benefit in human patients with CHIP. As a proof-of-concept, administration of the NLRP3 inhibitor MCC950 in Ldlr^−/− mice with TET2-driven CHIP caused significant reduction (by ~50%) of the atherosclerotic plaque size⁹. Treatment with the anti-inflammatory drug colchicine inhibited the overproduction of IL-1β and the acceleration of atherosclerosis linked to TET2-driven CHIP⁵⁸. Consistently, patients prescribed colchicine display weaker association between myocardial infarction and CHIP driven by mutations in TET2, albeit not other CHIP-driver genes⁵⁸. Similarly, the ability of anti-IL-1β (canakinumab) to reduce the incidence of major cardiovascular events and comorbidities (arthritis, gout and osteoarthritis) in patients with previous myocardial infarction⁵⁹ is more pronounced in a subset of patients with CHIP-associated TET2 mutations⁶⁰. In this context, the ongoing TECTONIC trial is evaluating whether IL-1β inhibition with canakinumab in individuals with TET2-driven CHIP and coronary artery disease can reduce vascular inflammation (ClinicalTrials.gov identifier: NCT06691217).

Blockade of CHIP-mutant clone expansion

The mechanistic target of rapamycin (mTOR), which primes quiescent stem cells for cell-cycle entry, is upregulated in Dnmt3a mutant HSPCs due to DNA hypomethylation of the mTOR gene¹⁶⁸. In a study that linked DNMT3A-driven CHIP to inflammatory bone loss, the FDA-approved drug rapamycin blocked the aberrant clonal expansion of Dnmt3a^R878H/+ hematopoietic cells and prevented periodontal inflammation and bone loss¹⁹. Similarly, the expansion of ASXL1-mutant HSCs depends on overactive mTOR signaling¹⁷³, suggesting a shared mechanism underlying CHIP progression. Rapamycin might exert similar effects in humans, since the drug also suppresses the proliferation of leukemic cells carrying the DNMT3A^R882H mutation¹⁶⁸. Dnmt3a mutant HSCs also display elevated mitochondrial respiration, which is essential for their enhanced expansion compared to WT controls. This competitive advantage is suppressed by the anti-diabetic drug metformin^125,126, an inhibitor of complex I of the electron transport chain¹⁷⁴. Mechanistically, metformin enhances the methylation capacity in DNTM3A-mutant HSPCs and reverses their aberrant DNA CpG and histone methylation profiles¹²⁵. Consistently, individuals on metformin exhibit decreased prevalence of CHIP driven by R882-mutant DNMT3A¹²⁶. Another potentially promising therapeutic is fedratinib, an FDA-approved JAK2 inhibitor, which was shown to restrain HSPC expansion and myelopoiesis as well as atherosclerosis development in western diet-fed Apoe^−/− mice¹⁷⁵.

Consistent with findings that inflammatory cytokines associated with inflammaging, including IL-1 and IL-6, promote the expansion of CHIP-mutant clones in animal models^33,137, genetic or pharmacological ablation of interleukin-1 receptor 1 (IL1R1) blocks the aberrant expansion of Tet2^+/− clones during aging³³. Interestingly, the above-mentioned TECTONIC trial (ClinicalTrials.gov identifier: NCT06691217) also includes longitudinal assessment of changes in TET2 clonal VAF as a proxy for clonal expansion or contraction, thereby testing whether targeting inflammation can influence not only the downstream consequences of CHIP but also the trajectory of development of clonal hematopoiesis itself. A subset of carriers of large TET2- or DNMT3A-associated CHIP clones (i.e, VAF >10%) with an inhibitory IL-6 receptor gene variant (IL6R p.Asp358Ala) display reduced risk for cardiovascular events, as compared to carriers of large CHIP clones without this IL6R genetic variant¹⁷⁶. Importantly, the presence of IL6R p.Asp358Ala was not associated with attenuated risk of cardiovascular events in subjects without CHIP¹⁷⁶. Therefore, inhibition of IL-6 signaling might selectively benefit patients in which cardiovascular disease is exacerbated by large CHIP clones.

Restoration of epigenetic enzyme function

Vitamin C is a co-factor of Fe2(+)/α-KG–dependent dioxygenases, such as the TET family enzymes. In this role, vitamin C can promote DNA demethylation by increasing the activity of the functionally intact TET2 enzyme from the non-mutant allele^170,177. Specifically, the ability of vitamin C to boost TET2 activity enhances 5-hydroxymethycytosine levels and reverses aberrant HSC expansion in the mouse BM^170,177. Moreover, vitamin C promotes DNA demethylation in carriers of germline TET2 mutations, suggesting that it could mitigate the methylation defect associated with TET2-driven CHIP¹⁷⁸. Thus, in principle, vitamin C could restore TET2 function in TET2-CHIP carriers and serve as a preventive therapy.

Conclusions and perspectives

The emergence of CHIP results in an outsized fraction of CHIP-mutant mature leukocytes, which are phenotypically altered in ways that contribute to inflammaging, a common denominator of aging-related comorbidities^{4,6,32,179,180} (Figure 1). Although inflammaging impairs the fitness of normal HSCs, it provides a selective advantage to CHIP-mutant clones, which thereby outcompete normal clones^{33–36,38,39,46,47,87,120} (Figure 4). Together, these studies suggest that CHIP may be both a cause and a consequence (reverse causality) of inflammation, forming a self-reinforcing loop that underlies the chronicity of inflammatory comorbidities (Figure 5). However, owing to conflicting data obtained in different disease or experimental contexts, the validity of reverse causality warrants further testing in longitudinal studies^{9,38–45,48,49}.

Recent advances have shed light on the factors that foster the emergence of CHIP (Figure 2b) and the mechanisms whereby CHIP increases the risk of or exacerbates chronic diseases. Yet, key questions remain — particularly regarding how CHIP mutations affect leukocyte function and which signaling pathways link inflammaging to the development and expansion of CHIP-mutant clones. From a medical viewpoint, it is crucial to determine the specific risks that CHIP poses for specific chronic disorders. Moreover, do hematopoietic stem cell transplants containing CHIP-mutant clones increase the risk of cardiometabolic or inflammatory diseases in the recipients?

Given the high prevalence of CHIP in the elderly, there is urgent need to develop clinical interventions—that can block the expansion of CHIP-mutant clones and/or prevent their downstream pathological effects^{9,15,19,33,52,53,56,125,126,168} or even restore the function of epigenetic modifier enzymes affected in CHIP^{169–172,177}. Achieving this objective will require multidisciplinary collaboration among immunologists, biochemists, hematologists, and physicians specializing in chronic inflammatory and autoimmune diseases and cancer. The ultimate goal is to break the vicious cycle of mutually reinforcing interactions between CHIP and inflammaging.

Glossary

CELLULAR SENESCENCE-ASSOCIATED PROINFLAMMATORY SECRETORY PHENOTYPE (SASP)

The characteristic secretion profile of senescent cells, which includes proinflammatory cytokines, chemokines, growth factors, and proteases that can alter the tissue microenvironment, promote chronic inflammation, and adversely influence neighboring cells.

cGAS-STING pathway

An innate immune signaling cascade in which the sensor cGAS (cyclic GMP-AMP synthase) detects cytosolic DNA and activates STING (stimulator of interferon genes), leading to the production of type I interferons and other inflammatory mediators to defend against infections and cellular damage.

CHRONIC ISCHEMIC HEART FAILURE

A long-term condition in which the heart’s ability to pump blood is impaired due to reduced blood flow from narrowed or blocked coronary arteries, typically resulting from atherosclerosis, leading to symptoms such as fatigue, shortness of breath, and fluid retention.

EMERGENCY MYELOPOIESIS

Accelerated proliferation and enhanced myeloid differentiation of progenitors in the bone marrow in response to systemic infection (and other stresses, such as inflammation and cancer), leading to generation of mature myeloid cells to replenish neutrophils and other phagocytes that are consumed in large quantities during systemic infections.

ENDOTOXEMIA

A pathological condition characterized by the presence of endotoxins (specifically lipopolysaccharides from the outer membrane of Gram-negative bacteria) in the bloodstream, where they trigger a systemic inflammatory response by activating immune cells and releasing proinflammatory cytokines, potentially leading to sepsis, multi-organ dysfunction, or chronic inflammatory diseases.

EPIGENETIC DRIFT

Gradual alterations in DNA methylation with aging arising from epigenetic errors that occur from repeated HSC division induced by inflammatory stress or other insults.

EXHAUSTION (OF HSCS)

The progressive decline in the ability of HSCs to self-renew, proliferate, and differentiate into mature blood cells, typically due to intrinsic factors like DNA damage, epigenetic alterations, and telomere shortening, as well as extrinsic stressors such as chronic inflammation and changes in the bone marrow niche; this dysfunction impairs hematopoiesis and contributes to immunosenescence, anemia, and increased susceptibility to hematologic malignancies with aging.

FIBROSIS

The formation of excess fibrous connective tissue in an organ or tissue, typically as a reparative response to injury or linked with chronic inflammation, which can impair normal structure and function.

GENETIC DRIFT

The change in frequency of a given gene variant in the population owing to chance rather than a selective pressure.

INFLAMMASOME

A multiprotein complex formed in response to cellular stress or infection that activates inflammatory caspases, such as caspase-1, leading to the maturation and secretion of proinflammatory cytokines such as IL-1β and IL-18, and often triggering a form of programmed cell death called pyroptosis.

LYMPHOPOIESIS

The process by which hematopoietic stem and progenitor cells in the bone marrow differentiate into lymphoid lineage cells — B cells, T cells, and natural killer (NK) cells — which play essential roles in immune defense: B cells produce antibodies, T cells regulate and execute adaptive immune responses, and NK cells provide rapid innate immunity by targeting infected or malignant cells.

MEGAKARYOPOIESIS

The process by which hematopoietic stem and progenitor cells differentiate into megakaryocytes, large bone marrow cells responsible for producing platelets essential for blood clotting.

MISSENSE MUTATIONS

Single nucleotide changes in DNA that result in the substitution of one amino acid for another in the encoded protein, potentially altering its function.

MYELOABLATION

Medical (or experimental) procedure that involves the complete or near-complete destruction of bone marrow activity (typically through high-dose chemotherapy or total body irradiation) to eliminate existing hematopoietic cells. Often used as a preparatory step for hematopoietic stem cell transplantation, myeoablation results in severe myelosuppression, reducing red and white blood cells and platelets.

MYELOPOIESIS

The process by which hematopoietic stem and progenitor cells in the bone marrow differentiate into myeloid-lineage cells, including granulocytes (neutrophils, eosinophils, basophils), monocytes, macrophages, and dendritic cells, essential for innate immunity and tissue homeostasis.

NECROPTOSIS

A regulated form of cell death, induced by various inflammatory stimuli prominently including TNF. Unlike apoptosis, necroptosis represents an explosive form of cell death typified by swelling followed by rupture of the plasma membrane.

NONSENSE MUTATIONS

Point mutations in DNA that convert a codon encoding an amino acid into a premature stop codon, leading to truncated, often nonfunctional proteins.

PATHOGEN-ASSOCIATED MOLECULAR PATTERNS (PAMPS)

Evolutionarily conserved molecular structures found on microbes (such as bacterial lipopolysaccharides, viral RNA, or fungal β-glucan) that are recognized by pattern recognition receptors (PRRs) of the innate immune system, triggering immune responses to eliminate the invading pathogens.

POLYCOMB REPRESSIVE COMPLEX-2

A multi-protein epigenetic regulator that catalyzes the trimethylation of histone H3 on lysine 27 (H3K27me3), leading to transcriptional silencing of target genes involved in cellular differentiation, development and cell identity.

PROSPECTIVE LONGITUDINAL STUDIES

Clinical research designs that follow a group of individuals over time, collecting data at multiple time points to examine how exposures or risk factors influence future outcomes or disease development.

RISK FACTOR

A characteristic, condition, or behavior that increases the likelihood of developing a disease or adverse health outcome.

SENOLYSIS

The therapeutic process of selectively eliminating senescent cells (cells that have permanently exited the cell cycle but remain metabolically active and often secrete proinflammatory factors) to reduce chronic inflammation, restore tissue function, and potentially delay age-related diseases.

VARIANT ALLELE FREQUENCY (VAF)

The proportion of mutant DNA, defined as the ratio of sequence reads that bear the mutant allele to the total number of reads at a given genetic locus. Assuming heterozygosity and a diploid state, a VAF of 2% means that 4% of circulating blood cells are derived from a single mutant clone.