Reprogramming cellular senescence and aging clocks for advanced cancer immunotherapy

Abstract

Cellular senescence has gradually been recognized as a key process, which not only inhibits the occurrence of early tumors but also promotes advanced malignant progression through secretory and immunomodulatory functions. Initially, cellular senescence manifested as irreversible cell cycle arrest, but now it encompasses a broader phenotype regulated by the p53-p21CIP1 and p16INK4A-Rb pathways. Although secretory phenotypes related to aging can recruit immune effectors to clear new tumor cells, persistent senescent cell populations often trigger chronic inflammation, promoting immune escape and fibrosis. In this review, we first discuss the molecular underpinnings of cellular senescence, highlighting its induction pathways and diverse physiological or pathological roles. We then examine the composition of the tumor microenvironment, where senescent cells accumulate and secrete pro-inflammatory cytokines, reshaping immune surveillance and extracellular matrix architecture. Against this backdrop, we explore how aging clocks refine our understanding of individual susceptibility to malignancy by distinguishing biological from chronological aging. We also present current therapeutic prospects, including senolytic agents targeting senescent stromal cells that promote tumor growth, and the utilization of aging clock metrics to tailor immunotherapies more effectively for older patients. Finally, we consider the major challenges facing clinical translation, from standardizing multi-omics data pipelines to clarifying the ethical implications of measuring biological age. By bridging senescence biology with geroscience and cutting-edge oncology, we posit that aging clocks may catalyze a transformation in cancer care, enabling more personalized, effective, and age-conscious treatment strategies.

Introduction

Cellular senescence, originally defined by Hayflick and Moorhead as the finite replicative capacity of human diploid fibroblasts, has evolved into a multidimensional concept encompassing stress-responsive cell cycle arrest and profound secretome alterations [1]. Modern senescence biology distinguishes this process from organismal aging (geroconversion) through two cardinal features: (1) irreversible proliferation arrest mediated by p53/p21CIP1 and p16INK4A/Rb pathways [2], and (2) development of senescence-associated secretory phenotype (SASP) that remodels tissue microenvironments [3]. While initially characterized as a tumor-suppressive mechanism preventing malignant transformation, accumulating evidence reveals that senescent cells paradoxically fuel tumor progression through SASP-mediated immunosuppression and extracellular matrix remodeling – a duality demanding context-specific therapeutic interventions [4]. The aging clock is linear or nonlinear models trained on omics data to predict age and aging rates, as measured by the difference between actual age and predicted age, which can quantify aging at different levels [5]. The aging clock is one of the most important breakthroughs in the field of aging biology in recent years and can provide indicators for the effectiveness of intervening in the aging process and preventing aging-related diseases.

Within the tumor microenvironment (TME), senescence exerts spatiotemporally polarized effects. Early senescence events promote immune surveillance through NKG2D ligand-mediated natural killer (NK) cell activation [6], whereas persistent senescent cells subvert antitumor immunity via SASP components like CCL5-driven regulatory T cell (Treg) infiltration and PD-L1 upregulation [7]. This dichotomy is being utilized in ongoing clinical trials, such as the injection of ABT263 (Navitoclax) which reverses the immunosuppression of myeloid cells in the TME [8], and the elimination of these myeloid cells can restore CD8 + T cell proliferation and alleviate immunotherapy resistance in vivo. In addition, venetoclax [9] combined with navitoclax and chemotherapy is well tolerated and has a good efficacy in patients with refractory/relapsed acute lymphoblastic leukemia of different ages [10].

Building on these foundational insights into cellular senescence as both a tumor-suppressive and tumor-promoting mechanism, the rapidly advancing concept of the “aging clock” has emerged to quantify biological age and capture nuanced, tissue-specific aging trajectories [11]. Epigenetic modifications, proteomic shifts, and alterations in non-coding RNA networks all converge to shape how cells transition between homeostatic senescence and pathological states favoring malignancy [12]. In parallel, the TME (which is composed of immune cells, fibroblasts, and extracellular matrix components) undergoes reorganization in response to senescence-associated signals, thereby altering the balance between effective tumor immunosurveillance and immune escape [13]. Intriguingly, therapy-induced senescence in stromal populations or malignant cells can either enhance antitumor immunity or create an immunosuppressive niche conducive to disease progression [14].

In this review, we delve into the intricate molecular mechanisms driving cellular senescence, outline the diverse roles of the SASP, and examine how senescent cells reshape tumor-immune crosstalk. We then explore the emergence of aging clocks (particularly those derived from DNA methylation data) and their evolving applications in oncology. We highlight how these clocks refine risk assessment, guide therapeutic strategies, and may even inform novel approaches for immunotherapy in older or immunocompromised patients. Lastly, we discuss the challenges and future directions of integrating aging clock metrics into precision cancer care, emphasizing the need to harness the dual nature of senescence for improved patient outcomes.

Biological basis of cellular senescence

Induction factors of cellular senescence

Cellular senescence, a permanent state of cell cycle arrest, plays crucial multi-dimensional regulatory roles in physiology and pathology, being vital for maintaining tissue homeostasis, regulating aging, and disease development. Physiologically, it can act as a temporary state to promote development, wound healing, and inhibit tumors. Pathologically, it is closely linked to the occurrence and progression of various diseases (including tumors, neurodegenerative diseases, cardiovascular diseases, etc.) and can evolve into a chronic state causing inflammation, tissue dysfunction, and age-related pathological changes. Its induction mechanism is complex, mainly achieved through the following pathways:

Replicative senescence is one of the earliest and most typical mechanisms of cellular senescence, mainly driven by telomere shortening. Telomeres are specialized DNA-protein complexes located at the ends of linear chromosomes that protect genomic DNA from degradation or recombination [15]. Due to the inherent biological limitations of terminal replication, in the absence of an effective telomere maintenance mechanism, telomere length gradually shortens with each cell division. When telomeres shorten to the critical threshold, it will trigger the DNA damage response (DDR) pathway, eventually inducing cells to enter the senescent state [16]. Meanwhile, telomere shortening can also lead to an increase in oxidative stress, further intensifying DNA damage and the senescence process of cells. Therefore, telomere depletion is an important driving factor for replicative aging both in vitro and in vivo.

DNA damage-induced senescence occurs when cells experience persistent or irreparable DNA damage. The sustained DNA damage activates a signaling cascade involving DDR pathways that upregulate tumor suppressor genes such as TP53 and CDKN2A (p16), leading to irreversible cell cycle arrest and the expression of senescence-related genes [17–19]. It is worth noting that the threshold for inducing senescence depends on the severity of the injury and the repair ability. Mild DNA damage can trigger transient ATM/ATR activation and repairing-centered responses, induce transient cell cycle arrest, and provide opportunities for DNA repair. In contrast, severe DNA damage, characterized by the formation of γ-H2AX foci and chromatin remodeling, promotes senescence or apoptosis through activation of p53 [20, 21]. The eventual fate (senescence vs. apoptosis) depends on both cell type and the physiological context.

Oncogene-induced senescence (OIS) arises when activated oncogenes, such as RAS [22] and RAF [23], or inactivated tumor suppressors, such as PTEN [24], trigger a hyperproliferative phase that leads to increased replication stress, culminating in DDR activation and senescence. Serrano et al. first demonstrated that oncogenic Ras drives premature senescence in primary rodent and human cells, associated with heightened p53 and p16 activity [25]. Subsequent studies extended this paradigm: activation of BRAF^V600E in human melanocytes similarly induces p16-dependent senescence, establishing an early barrier to melanoma [26]. In vivo, Pten-loss prostate lesions also undergo p53-mediated OIS, and its abrogation accelerates tumorigenesis [27]. After activation, proto-oncogenes excessively generate growth-promoting signals, driving cells to proliferate rapidly. However, such excessive proliferation is often difficult to sustain and may cause genomic instability. The cellular DDR mechanism senses this instability and activates signaling pathways, ultimately triggering OIS [28, 29]. OIS involves transcriptional silencing of cell-proliferation genes, activation of tumor suppressor pathways (e.g., p53), and formation of repressive heterochromatin [30–32]. thereby acting as a critical intrinsic tumor suppressor mechanism that prevents outgrowth of cells with aberrant mitogenic signaling.

Oxidative stress-induced senescence is driven by the accumulation of oxidizing agents, such as reactive oxygen species (ROS), generated endogenously or introduced exogenously (e.g., UV radiation, hydrogen peroxide, certain chemicals) [33, 34]. Endogenous ROS are primarily generated by mitochondrial electron transport chain (ETC) leakage at complexes I (NADH dehydrogenase) and III (cytochrome bc₁ complex), where incomplete electron transfer produces superoxide radicals (O₂⁻) as byproducts. In addition, exogenous factors such as ultraviolet radiation, hydrogen peroxide and environmental toxins can promote the excessive accumulation of ROS [35, 36]. Under normal physiological conditions, cellular redox homeostasis is maintained by the antioxidant system and antioxidant enzymes. However, when ROS levels overwhelm this protective capacity, oxidative damage to lipids, proteins, and DNA induces senescence. ROS-induced DNA damage is therefore considered a key driver of cellular senescence initiation and progression [37].

Mitochondrial dysfunction-associated senescence(midas) is defined as senescence triggered by mitochondrial dysfunction [38]. Characterized by bioenergetic dysfunction, redox imbalance, and activation of aging-related signaling pathways, MiDAS arises from mitochondria’s central role in energy production and cellular homeostasis. Depletion of mitochondrial sirtuins (e.g., SIRT3/SIRT5) or chemical inhibition of mitochondrial function precipitates senescence by impairing mitochondrial metabolism and antioxidant defense [39]. Dysfunctional mitochondria exhibit elevated reactive oxygen species (ROS) production, which induces DNA damage and activates the DNA damage response (DDR) pathway [40, 41], exacerbating senescence through p53/p16-mediated cell cycle arrest. This vicious cycle of ROS accumulation and tumor suppressor pathway activation disrupts redox homeostasis and accelerates aging [42].

Paracrine senescence is cellular senescence induced by a SASP produced by primary senescent cells. SASP regulation involves complex transcriptional programs mediated by signaling pathways such as NF-κB and p38 MAPK, which induce the secretion of pro-inflammatory cytokines (e.g., IL-6, TNF-α), chemokines, and matrix-degrading enzymes (e.g., MMPs) [43, 44]. These factors reprogram the tissue microenvironment by modulating immune responses, matrix remodeling, and cell-cell communication, thereby influencing aging progression, immune surveillance in aged tissues, and pathophysiological outcomes [45–47]. SASP initially aids immune clearance of senescent cells but chronically induces inflammation, impairing stem cell function and tissue regeneration [14]. Senescent cell-derived SASP amplifies senescence via paracrine signaling, perpetuating tissue damage and age-related diseases [48]. Thus, SASP acts as a central hub linking aging to age-related pathologies. The SASP factors are shown in Table 1.

Table 1.

Senescence-associated secretory phenotype

Category	SASP Factors	Senescence-related Functions	Senescence Inducers of the Factors	Upstream Regulators	Reference
Pro-inflammatory factors	IL6	Promote the inflammatory cascade, Induce autocrine and paracrine senescence	RS, TIS, OIS, DNA damage, Ageing stress	NF-κB, p38 MAPK	[49, 50]
	TNF-α	Activate the NF-κB pathway, Induce cell apoptosis or inflammatory responses, Promote angiogenesis in tumor progression	ROS DNA damage	NF-κB, mTOR	[51, 52]
	IL-1α IL-1β	Initiate the innate immune response, Exacerbate chronic inflammation, Associated with neurodegenerative diseases and metabolic disorders	ROS, DNA damage	NF-κB, p38 MAPK	[53]
	IFN-γ	Induces the expression of pro-inflammatory factors (such as IL-6, TNF-α), Exacerbating chronic low-grade inflammation	ROS, DNA damage, Inflammation	JAK-STAT pathway, NF-κB, p38 MAPK	[54]
Growth Factors	FGF-2(bFGF)	Promote angiogenesis, Fibroblast proliferation, Tissue repair	Tissue damage, Tissue regeneration	MAPK-ERK pathway, VEGFR, MCP-1	[55]
	VEGF	maintain the tumor microenvironment Induce angiogenesis, Support tumor growth, Promote wound healing	Oncogene activation, Oxidative stress, DNA damage	HIF-1α, p38 MAPK, NF-κB	[56]
	TGF-β	Regulating ECM remodeling, Pro-inflammatory and anti-inflammatory effects, Promote fibrosis, Bidirectionally regulate the immune microenvironment in tumors	PAI1, Oncogene activation, Replicative senescence	Integrins, Proteases, ROS, TSP-1, Plasmin System	[57, 58]
	EGF family	Activate the EGFR pathway, Drive tumor cell proliferation and resistance, Participate in embryonic development	EGFR	ADAMs, GPCR, Protein Kinase, ROS	[59]
	IGF-1	Regulate cell metabolism and proliferation, Delay age-related functional decline, Improve nerve regeneration	IGFBPs	GH, GHRH, Megalin	[60]
	GDF15	Facilitates the progression of cancer associated with ageing	Oxidative stress, Inflammation, Telomere erosion,	p53, Oxidative stress, Inflammation, Hypoxia	[61]
Chemokines	CXCL1, CXCL2	Recruit neutrophils and monocytes, promote inflammatory infiltration, Participate in the formation of the pre-metastatic niche in tumor metastasis	ROS, DNA damage	C/EBPβ, IL-1β, STAT3, TNF-α	[62]
	IL-8	Induce both autocrine and paracrine senescence, Facilitating B cell differentiation, T cell activation, and tumor cell proliferation	RS, OIS, CXCR2, DNA damage	NF-κB, MEK, C/EBPβ, p53	[63]
	CCL2	Attract monocytes to the damaged tissue, Regulate macrophage polarization, Influence wound healing and fibrosis	ROS, GDF15, Metabolic factors	NF-κB, p38 MAPK, PDGF	[64]
	PF4	Inhibit hippocampal neuroinflammation, Improve cognitive function	Oxidative stress, Chronic inflammation	Platelet Activation, Inflammatory Stimuli	[65]
Proteases	MMP1 MMP3 MMP9	Degradation of ECM components to promote matrix remodeling and cell migration	ROS, Tissue injury	Thyroid Hormone, Syndecan 4, ROS	[66]
	MMP2 MMP9	Degrade ECM, Promote tumor invasion and metastasis, Participate in tissue remodeling	Tissue Injury, Inflammatory microenvironment	NF-κB, TGF-β, ROS	[67]
	MMP3	Degrade collagen and elastin, Aggravate the instability of atherosclerotic plaques	Tissue remodeling, Inflammation	TGF-β, NF-κB, MAPK pathway	[68]
ECM	Collagen Elastin	Maintain tissue elasticity, Abnormal secretion associated with fibrotic diseases such as liver fibrosis	Matrix remodeling, Mechanical stress, ROS	TGF-β, GFs, MMPs, TIMPs	[69]
	Fibronectin	Mediate cell adhesion, Promote tumor cell migration and inflammatory cell recruitment	Tissue injury, Cellular adhesion stress	Factor XIII, Integrin signaling	[70]
Metabolic Regulatory Factors	Lactic acid	Inhibit T cell function, Promote tumor angiogenesis, Remodel the acidic microenvironment	Acidic microenvironmen, Metabolic stress	HIF-1α pathway, mTOR	[71]
	PDK4	Regulate glycolytic metabolism, maintain SASP secretion, Associated with tumor resistance	Metabolic stress, Hypoxia	HIF-1α, mTOR pathway	[72]
Epigenetic Regulatory Factors	ACLY	Regulate SASP gene expression through H3K27 acetylation modification, Linking metabolism and epigenetics	Cellular senescence, DNA damage, Inflammation	mTOR pathway, AMPK signaling	[48]
	TGM2	Catalyze IκBα crosslinking, Activate the NF-κB pathway, Establish a positive feedback loop for SASP	Extracellular matrix crosslinking, Tissue remodeling, Inflammation	NF-κB signaling, TGF-β pathway, ROS	[73]
Lipids	PGE2	Promote inflammation, Vascular permeability Pain response, Regulate immune cell function	Inflammation	COX-2 pathway, NF-κB signaling	[74]
	LTB4	Recruit neutrophils, Enhance the inflammatory response	Inflammation	5-LOX, Neutrophils, G-CSF	[75]
Other Bioactive Molecules	ROS	Oxidative stress products exacerbate DNA damage and inflammation, Accelerating the aging process	Oxidative stress, DNA damage	Nrf2 pathway, NF-κB signaling	[76]
	CCL11	Recruit eosinophils, Associated with allergic inflammation and tissue remodeling	Systemic immune changes	NF-κB, p38 MAPK, mTOR	[77]
	uPAR	Activates plasminogen, Promotes ECM degradation and angiogenesis	Cell proliferation, migration and survival	MAPK pathway, HIF1A, NF-κB	[78]

Molecular mechanisms of cellular senescence

Senescent cells exhibit sustained DNA damage response (DDR) via ATM/ATR kinases, driving cell cycle arrest through the p53/p21CIP1 and p16/pRb axes. Replicative senescence, triggered by telomere attrition, primarily relies on the p53-p21CIP1 pathway [79–81]. whereas oxidative stress-induced senescence engages p16 or ERK-p38 MAPK signaling independently of telomeres [82, 83].

Multiple stressors (e.g., DNA breaks, oncogenic signals) activate p53 [84], which upregulates p21CIP1 to induce Rb hypophosphorylation and E2F sequestration—this hypophosphorylated Rb then blocks G1/S and G2/M transitions by sequestering E2F transcription factors, thereby halting cell cycle progression and enforcing senescence [85–89]. Concurrently, p16 inactivates CDK4/6 to reinforce growth arrest [90]. Temporally, p21CIP1 surges early and declines, while p16 accumulates later to sustain irreversible arrest [91]. highlighting their cooperative roles. The p38 MAPK pathway, activated by UV light, heat shock, or oxidative stress [37, 92, 93]. involves a cascade where components such as MEKK/TAK and MKK3/6 ultimately phosphorylate p38. This phosphorylated p38 not only enhances p53 and p16 activation to consolidate the senescent phenotype [94]but also phosphorylates downstream effectors (e.g., NF-κB, STAT3) to amplify SASP. Additionally, oncogenic RAS further promotes senescence by converging on p38 via the Ras-Raf-MEK-ERK pathway [95, 96].

The spatiotemporal regulation of the senescence-associated secretory phenotype (SASP)—a mix of pro-inflammatory cytokines, chemokines, and proteases driving age-related tissue dysfunction—arises from dynamic crosstalk between canonical senescence pathways and post-translational modifications (PTMs) [97–99]. Phosphorylation, acting as a critical signal integrator within the ATM/ATR-p53-p21CIP1 and p16/Rb axes, coordinates SASP activation [100, 101]. Notably, sustained p38 MAPK activity—linked to oxidative stress-induced senescence—phosphorylates NF-κB and STAT3 [102], directly upregulating mediators like IL-6, IL-8, and MMPs via transcriptional amplification [103]. This mechanism functionally couples stress-responsive MAPK cascades to inflammatory SASP propagation, with p38 inhibition robustly reducing secretory outputs [104].

Beyond phosphorylation-driven pathways, epigenetic reprogramming via acetylation further integrates with these signaling networks to regulate SASP dynamics [105]. H3K9 hyperacetylation enhances chromatin accessibility at pro-inflammatory loci like CXCL1 [106], while acetylation-dependent nuclear retention of FoxO transcription factors modulates cytokine production [107, 108]. These epigenetic modifications synergize with persistent DDR signaling from dysfunctional telomeres or genomic lesions, creating a permissive chromatin state for SASP gene expression.

Proteostatic collapse—rooted in sustained DDR and oxidative stress—amplifies SASP via dysregulated ubiquitination [109]. Age-related decline in ubiquitin-proteasome system (UPS) efficiency causes accumulation of signaling intermediates like IKKγ, perpetuating NF-κB-driven inflammation in a manner paralleling p53/p21CIP1-mediated cell cycle arrest [110]. Compromised E3 ligases (e.g., FBXW7), which normally degrade pro-inflammatory mediators, exacerbate this imbalance [111]. Meanwhile, ROS from telomere crisis or mitochondrial dysfunction further entrench SASP activation via oxidative cross-linking of TGF-β (enhancing stability) [112] and JNK-mediated phosphorylation cascades, forming feedforward loops reminiscent of p16/Rb-mediated growth arrest’s self-reinforcing nature [113].

Despite these mechanistic advances, critical gaps persist in understanding how PTMs networks interface with core senescence pathways across tissues. For instance, hepatic senescence may prioritize ubiquitination-mediated proteostatic failure [114]—akin to replicative senescence’s reliance on telomeric DDR—while dermal fibroblasts exhibit stronger dependence on acetylation-driven chromatin remodeling [115]. Furthermore, the temporal hierarchy observed in p21CIP1-p16 coordination during cell cycle arrest may extend to SASP regulation: phosphorylation might initiate acute inflammatory responses, while epigenetic modifications maintain chronic SASP production [2]. Resolving these spatiotemporal relationships requires integration with single-cell multi-omics approaches that map PTMs landscapes against transcriptional outputs and DDR activation states.

Technological innovation is critical to bridge these gaps, with phosphoproteomic profiling of senescent cell subtypes and CRISPR screens targeting PTM enzymes (e.g., HDACs, deubiquitinases) offering mechanistic insights into kinase-specific SASP checkpoints and p53/p16 crosstalk modifiers. Real-time visualization tools—such as FRET-based MAPK activity reporters or live imaging of NF-κB phosphorylation—enable decoding of stress signal transitions from growth arrest to secretory phenotypes by capturing dynamic PTM landscapes [116]. Translationally, tissue-specific PTM modulation holds therapeutic potential: p38 inhibitors (e.g., losmapimod) may synergize with telomerase activators in p16-dominated pathologies, while mitochondrial-targeted antioxidants (e.g., SS-31) disrupt ROS-JNK loops in oxidative senescence [117]. However, therapeutic strategies must account for pathway redundancies—much like the temporal cooperation between p21CIP1 and p16—to avoid exacerbating age-related dysfunction through incomplete PTMs network inhibition.

This synthesis of PTMs-mediated SASP regulation underscores the necessity of viewing senescence as a multidimensional process where DNA damage signaling, epigenetic reprogramming, and proteostatic collapse converge [118]. By adopting the same temporally-resolved, pathway-integrated lens applied to classic senescence effectors like p53 and Rb, future research may unlock precision interventions that selectively target pathogenic SASP subsets while preserving homeostatic functions—a paradigm shift mirroring the evolving understanding of cell cycle arrest mechanisms in aging and cancer. The activation of core regulatory pathways of cellular senescence is shown in Fig. 1.

Fig. 1.

core regulatory pathways of cellular senescence

Depicts three core signaling axes triggering cellular senescence: (1) DNA Damage Response (DDR) via ATM/ATR-CHK kinases activating the p53-p21CIP1 pathway; (2) the p16-CDK4/6-RB axis, where hypophosphorylated RB sequesters E2F to block S-phase gene transcription (e.g., PCNA, CDK2); and (3) the TAK-MKK-p38 MAPK stress pathway. These cascades synergize to induce cell cycle arrest.

Physiological and pathological effects of cellular senescence

Physiological functions of cellular senescence

Cellular senescence operates as a dynamic regulator of tissue homeostasis, exerting its physiological roles through spatially and temporally coordinated programs. During embryogenesis, senescence serves as an evolutionary conserved morphogenetic timer. In murine allantois development, p21CIP1-dependent senescence drives vascular plexus regression via MMP9-mediated fibronectin matrix degradation, a critical step for umbilical cord formation [119]. Single-cell transcriptomic profiling of human embryonic limb buds further reveals transient senescent populations in the apical ectodermal ridge (AER) that secrete FGF8 to sustain progress zone proliferation—genetic ablation of p21CIP1 in AER cells disrupts this signaling axis, leading to digit malformations resembling clinical syndactyly [120, 121]. Placental syncytiotrophoblasts undergo oxidative stress-induced senescence to secrete IL-8 and GROα, activating maternal uterine NK cells via CXCR2 signaling [122], to facilitate spiral artery remodeling and fetal-maternal immune tolerance [123, 124]. Reduced senescent trophoblasts in preeclampsia placentas underscore this pathway’s clinical relevance.

In tumor suppression, senescence integrates cell-autonomous and microenvironmental mechanisms. Oncogenic RAS activation triggers a biphasic response: initial hyperproliferation followed by p16INK4A/p53-dependent growth arrest [125], stabilized by nucleolar stress sensors (RPL11/RPL5) sequestering MDM2 [126]. Concurrent ATM-mediated DNA damage response activates p21CIP1, reinforcing growth arrest. Senescent cells further orchestrate immune surveillance by upregulating MHC complexes and secreting CXCL10/CCL5 to recruit CD8 + T and NK cells [4]. In BRAF^V600E melanoma models [127], SASP-induced NKG2D ligands on keratinocytes enhance γδT cell-mediated elimination of premalignant lesions, reducing transformation risk by 73% [128]. Metabolic competition via GLUT1/MCT4 overexpression creates glucose/lactate gradients, depriving tumor cells of glycolytic substrates [79]. Isotope tracing shows senescent fibroblasts consume > 80% of available glucose, forcing cancer cell quiescence through AMPK/mTORC1 inhibition [129].

In fibrotic disorders, senescence acts as a self-limiting checkpoint to prevent excessive scar formation [130]. Following carbon tetrachloride-induced Liver injury, 92% of αSMA⁺myofibroblasts transition into p16INK4A-dependent senescence within 7 days, accompanied by a metabolic shift from glycolysis to fatty acid oxidation [131]. This reprogramming enables large-scale MMP12/ADAMTS5 secretion for collagen degradation while recruiting TIMP1⁺macrophages via CCL17 signaling to resolve fibrotic matrices [132, 133]. In diabetic wounds, senescent fibroblasts secrete PDGF-AA to promote angiogenesis while releasing decorin to inhibit TGF-β-driven ECM overproduction [134]. Intravital imaging captures senescent cell clusters functioning as temporary “biological scaffolds”—their sequential clearance by neutrophils coordinates phased tissue regeneration without excessive contraction [135]. Collectively, these processes balance repair precision with regenerative capacity [136].

Pathological effects of cellular senescence

The persistence of senescent cells initiates a vicious cycle of chronic inflammation and microenvironmental degeneration, driving age-related pathologies through interconnected biological axes [137]. Hematopoietic stem cells (HSCs) exemplify senescence-induced stem cell exhaustion: single-cell chromatin profiling reveals aged HSCs exhibit H3K27me3 enrichment at self-renewal loci (e.g., HOXA9, MEIS1), leading to myeloid-biased differentiation [138]. Transplantation assays demonstrate aged HSCs generate five-fold more granulocytes than young counterparts, a shift linked to clonal hematopoiesis and anemia [139]. Similarly, chronic SASP exposure (e.g., IL-6, TGF-β) in skeletal muscle induces mitochondrial ROS overproduction in satellite cells [140, 141]. triggering Pax7 promoter hypermethylation and irreversible regenerative failure [142]. JAK/STAT inhibition partially restores satellite cell activation, highlighting reversible therapeutic targets [143].

Oncologically, senescence exhibits paradoxical roles. Early tumorigenesis benefits from p53-mediated growth arrest (e.g., PTEN-deficient prostate lesions) [144]. but advanced malignancies co-opt SASP for progression. Therapy-induced senescence in pancreatic CAFs generates CXCL12-enriched niches promoting cancer stem cell chemoresistance [79, 145, 146]. while PGE2 secretion recruits TIM3 + exhausted T cells to create immunosuppressive microenvironments, Hypoxia exacerbates this by stabilizing HIF-1α in senescent tumor cells, redirecting SASP toward pro-angiogenic VEGF/PAI-1 profiles [147, 148]. Preclinical studies show senolytic elimination of these cells with dasatinib/quercetin combinations reduces metastatic burden by 65% in murine breast cancer models, highlighting the therapeutic potential of context-specific senotherapy [149].

Skeletal system senescence exemplifies systemic consequences. Osteocytes in aged bone exhibit sclerostin (SOST) overexpression and RANKL/OPG ratios > 5:1, driving osteoclast hyperactivation and DKK1-mediated osteoblast suppression [150, 151]. Senescent bone marrow mesenchymal stem cells overproduce CCL5, recruiting senescent T cells to form pro-inflammatory loops accelerating hematopoietic aging [152, 153]. Nanoparticle-delivered senolytics achieve 80% osteocyte clearance in preclinical models without off-target toxicity. The influences of cellular senescence is shown in Fig. 2.

Fig. 2.

The influences of cellular senescence

Depicts the context-dependent roles of cellular senescence: (1) Physiological level: During embryonic development, p21⁺-MMP9 mediates the degeneration of urinary sac vessels, p21⁺-FGF8 regulates the morphology of limb buds, and placental CXCR2 recruits uNK cells chemotactic by IL-8/GROα; In tumor suppression, NKG2D⁺γδ T cells clear 73% of BRAF^^V600E prelesions, and GLUT1/McT4-mediated metabolic competition deplets glucose (> 80%), inhibiting cancer cells through AMPK/mTOR. (2) Pathological level: Stem cell failure is manifested as elevated H3K27me3 in HSCs driving clonal hematopoiesis (myeloid expansion by 5 times), IL-6/TGF-β inducing ROS hypermethylation of satellite cells Pax7 and inhibiting regeneration; Tumor progression is caused by the continuous secretion of SASP (CXCL12/PGE2) by CAFs (α-SMA⁺/FAP⁺), which induces the exhaustion of TIM3⁺ T cells and the remodeling of blood vessels by the HIF-1α-VEGF/PAI-1 axis under Hypoxia. The intervention strategies include the combined elimination of senescent cells by dasatinib and quercetin, reducing breast cancer metastasis by 65%, as well as the targeted elimination of 80% of senescent bone cells by nanoparticles.

Immune escape mechanisms in senescent cells

Senescent cells, despite their pro-inflammatory senescence-associated secretory phenotype (SASP), often evade immune surveillance through coordinated molecular strategies—a key facet of senescence bypass, which encompasses the broader process by which senescent cells re-enter the cell cycle via epigenetic reprogramming—specifically, structural chromosomal rearrangements that remodel chromatin topology—metabolic adaptation, and immune escape mechanisms to acquire invasive phenotypes, with direct links to tumor recurrence and treatment resistance. In oncogene-induced senescence (OIS), a recurrent 3.7-Mb inversion on chr3 disrupts the subTAD structure of the circadian gene BHLHE40, enhancing its promoter accessibility (H3K27ac enrichment) and driving its activation. BHLHE40 directly regulates 68.8% of escape-associated differentially expressed genes (e.g., MDM2, PCNA), and CRISPR-induced inversion suffices to bypass CDC6-driven senescence, inducing EMT (E-cadherin↓, vimentin↑) and tumorigenicity [154]. This process is further driven by genomic instability: in p53-deficient cells, prolonged p21WAF1/Cip1 expression (a key senescence mediator) disrupts the replication licensing machinery (e.g., Cdt1, Cdc6) via saturating CRL4CDT2 and SCFⁿᵏᵖ² ubiquitin ligases, leading to re-replication, DNA damage, and error-prone repair (RAD52-dependent). After ~ 10 days of senescence, a subpopulation of ‘escaped cells’ emerges—these cells downregulate the senescence driver p73, restore Cdk2 activity, and exhibit chromosomal aberrations (e.g., 1.75Kb-92 Mb gains/losses, translocations), ultimately gaining anchorage-independent growth and invasion capabilities [155]. This interconnected network of escape mechanisms, supported by recent studies, contributes to persistent tissue damage, disease progression, and the formation of dormant tumor lesions from treatment-induced senescence (TIS) cells, whose elimination necessitates synergistic strategies combining immune intervention and senescence targeting.​.

A central strategy involves dampening immune recognition through selective modulation of cell surface markers. Senescent cells downregulate NKG2D ligands (e.g., MICA and MICB), key ligands for natural killer (NK) cell activation, via epigenetic silencing of their promoters [156]. This blunts NK cell-mediated cytotoxicity, as observed in BRAF-mutant melanocytes where NKG2D ligand loss correlates with reduced senescent cell clearance and increased tumorigenesis potential [157]. Senescent stromal cells show PD-L1 upregulation driven by p16INK4A-dependent NF-κB activation, fostering a T cell-excluded microenvironment that protects adjacent tumor cells [158]. This dual regulation—silencing “alert signals” while amplifying “off switches”—renders senescent cells invisible to adaptive and innate immunity.

Beyond cell-autonomous changes, senescent cells actively remodel the immune microenvironment via SASP to reinforce evasion. Pro-inflammatory cytokines (IL-6, CCL2) and growth factors (TGF-β) secreted by senescent cells recruit myeloid-derived suppressor cells (MDSCs) and M2-polarized macrophages, which in turn suppress effector T cell proliferation and cytotoxicity [159]. Therapy-induced senescent (TIS) cancer-associated fibroblasts (CAFs) secrete CXCL12, which not only promotes cancer stem cell dormancy but also recruits regulatory T cells (Tregs) to establish immune privilege [160]. Critically, TIS cells themselves acquire cancer stem cell-like properties via activated Wnt/β-catenin signaling: in Eµ-Myc lymphomas, ADR-induced senescence upregulates stem cell markers (Sca1, ALDH) and Wnt target genes (Ccnd1, Id2), with β-catenin nuclear accumulation in ~ 80% of TIS cells. When released from senescence, these ‘previously senescent cells’ initiate tumors in vivo at 100-fold lower cell doses than never-senescent cells, and relapsed lymphomas show a 30-fold enrichment of nuclear β-catenin⁺ cells [161]. This creates a feedforward loop: senescent cells sustain a suppressive niche that protects both themselves and neighboring malignant cells from immune attack.

Metabolic competition Further strengthens this evasion. Senescent cells upregulate indoleamine 2,3-dioxygenase (IDO) and arginase-1, enzymes that deplete local tryptophan and arginine—nutrients essential for T cell activation and proliferation. In breast cancer models, TIS cells exhibit heightened IDO activity, with tryptophan catabolites (e.g., kynurenine) directly inhibiting CD8⁺ T cell interferon-γ production [154]. This metabolic hijacking complements soluble signals to dampen anti-senescent immune responses. Therapeutic targeting of these mechanisms shows promise. CDK4/6 inhibitors, which enforce senescence by stabilizing hypophosphorylated RB, also reactivate NKG2D ligand expression and reduce PD-L1 levels in TIS cells, sensitizing them to NK and T cell-mediated clearance [162]. Combining such senescence-reinforcing agents with immune checkpoint blockers (e.g., anti-PD-L1) synergistically enhances senescent cell elimination in preclinical models of liver fibrosis and melanoma, underscoring the need for dual strategies that disrupt both evasion and persistence. Collectively, these findings reveal immune evasion as a pivotal hallmark of senescent cell persistence, linking cell-intrinsic adaptations to systemic immune dysfunction. Unraveling these mechanisms not only deepens understanding of senescence-associated pathologies but also identifies actionable targets for improving senescence-targeted therapies.

While cellular senescence, a hallmark of organismal aging, orchestrates tissue homeostasis and pathology via cell-autonomous and paracrine mechanisms, its relationship with systemic aging is neither linear nor unidirectional—united by shared hallmarks (e.g., epigenetic reprogramming, SASP-driven chronic inflammation) yet divided by key dichotomies (acute stress-responsive senescence, as in developmental morphogenesis, vs. progressive systemic aging). Aging clocks, as quantifiable proxies of biological aging, serve as critical tools to decode this complexity. The following section explores the progress of aging clocks through three interconnected dimensions: first, how these clocks were established as robust models capturing cumulative molecular changes; second, their tissue-specific dynamics and applications across diverse biological contexts; and finally, how clock dysregulation interfaces with disease pathogenesis, potentially linking aberrant cellular senescence to age-related disorders. Together, these perspectives illuminate how aging clocks operationalize the interplay between cellular senescence and organismal aging, offering both mechanistic insights and translational potential. Integrated biomarkers of senescence: linking clocks to age-related disease risk is shown in Fig. 3.

Fig. 3.

Integrated biomarkers of senescence: linking clocks to age-related disease risk

Illustrates hierarchical framework of senescence triggers and outcomes: Driving factors: telomere depletion, accumulation of DNA damage, oxidative stress, mitochondrial dysfunction. The core mechanism: Epigenetic clocks such as DNA methylation, telomere shortening and metabolite drift jointly quantify the age of organisms. Functional decline: Irreversible cycle arrest and SASP secretion cause tissue inflammation, stem cell exhaustion, and MMP-mediated ECM degradation (such as fibrosis). Clinical prediction: The integration of biomarkers can proactively assess the risk of age-related diseases, mortality, and the rate of physiological function decline.

Progress in the aging clock

Establishment of the aging clock model

Due to the multi-dimensional characteristics of the aging mechanism, an individual’s biological age often deviates from their actual age. This phenomenon highlights the urgent need to develop reliable quantification tools for aging. Recent progress in high-throughput omics technologies has paved the way for the next generation of biogerontological tools, enabling molecular-level assessments of aging through epigenomics, proteomics, transcriptomics, and metabolomics data, often integrated via machine learning algorithms to construct so-called “aging clocks” [163–166]. This paradigm shift has profoundly influenced our ability to predict age-related health outcomes, evaluate the effectiveness of interventions that regulate the aging process, and formulate preventive strategies for age-related diseases.

Among the various aging clocks, the DNA methylation clock (also termed the “epigenetic clock”) remains the most extensively employed [167]. Its readout, derived from the level and patterns of DNA methylation, typically exhibits a positive correlation with increasing chronological age. DNA methylation refers to the enzymatic addition of a methyl group to the C5 position of cytosine within CpG dinucleotides, exerting transcriptional control by either promoting or repressing gene expression through alterations in chromatin structure [168]. The pioneering multi-tissue epigenetic clock proposed by Horvath significantly advanced the field by demonstrating the applicability of methylation-based aging biomarkers across diverse tissues [169]. Since then, numerous epigenetic clocks have emerged, among which Horvath’s clock [169] and Hannum’s clock [170] have received widespread validation. In recent years, newly developed epigenetic clocks, such as Levine’s DNAmPhenoAge [171] and the GrimAge clock [172], seek to refine predictive accuracy for age-related comorbidities and survival, underscoring the growing versatility of these biomarkers. Some of these models have even been applied to preclinical animal models to evaluate intervention measures aimed at delaying or reversing aging.

Parallel to epigenetic technologies, the increasing robustness and accessibility of proteomic profiling Have elevated the proteomic Aging clock as another valuable biomarker of biological aging. By studying 2,897 plasma proteins from the UK Biobank cohort, Argentieri et al. [173]. Established a proteomic clock capable of predicting risks for major morbidities and mortality across different populations. Similarly, Kuo et al. [174]. Leveraged a Panel of 2,923 plasma proteins to form the proteomic aging clock (PAC), which strongly correlates with all-cause mortality risk and reflects an individual’s biological age. As additional population-based studies incorporate broader protein panels, the proteomic clock is expected to offer enhanced sensitivity and specificity in forecasting disease onset and aging trajectories.

The transcriptome clock infers physiological age by measuring gene expression patterns, providing an interpretable link between aging and transcriptional activity. Peters et al. [175]. Utilized transcriptome data to derive a blood-based clock built on 1,497 genes associated with chronological aging. Mao et al. [176]. This concept was further refined by integrating single-cell transcriptome data with interpretable molecular markers, thereby constructing a tissue-specific single-cell aging clock - a clock capable of capturing subtle age-related transcriptional dynamics. This type of transcriptome clock not only deepens our understanding of the molecular mechanism of aging, but also provides an observation window for the dynamic evolution of age in gene regulatory networks.

Metabolites represent low–molecular-weight intermediates and end products of cellular metabolism, reflecting an organism’s physiological state. By profiling Fasting blood metabolites in 6,055 individuals from the UK, Menni et al. [177]. Built a metabolome clock that captured age-associated changes in metabolic pathways. Such metabolome-based clocks may uncover how alterations in energy metabolism, amino acids, lipids, and other metabolites aggregate over time, thereby serving as a sensitive measure for both the rate and direction of aging.

Aging involves a complex interplay of physiological and pathological processes. Therefore, a single type of omics measurement often fails to fully capture the complexity of biological ages. To overcome these limitations, multi-omics clocks combine the features of genomics, epigenetics, proteomics, metabolomics and other high-dimensional datasets, and improve accuracy and interpretability by leveraging complementary biomarkers [178, 179]. Li et al. [180]. Used a cohort of 113 healthy women spanning a wide age range to develop a multi-omics clock (compositeAge). By integrating previously established and newly discovered markers and using multimodal analysis methods, the compositeAge clock provides a more comprehensive assessment of the aging process. This framework is expected to explain the biological mechanisms of aging from multiple dimensions and provide guidance for personalized intervention strategies, such as targeted therapy and lifestyle adjustments.

These aging clocks at the omics level jointly mark an important breakthrough in the field of quantitative aging research. They not only provide a solid scientific basis for in-depth analysis of the molecular mechanisms of aging, but also become a core tool for evaluating emerging intervention strategies, including therapies or preventive measures that delay the decline of aging-related functions. With the continuous in-depth research in this field, such clocks are expected to make continuous progress in enhancing clinical practicality, especially when revealing the dynamic association between aging and tumor progression, which is expected to provide valuable interdisciplinary theoretical support for innovative cancer prevention and treatment strategies.

Application of aging clocks in different tissues

Recent advances in aging research have underscored that senescent or “aging” clocks built from one tissue type may not necessarily apply to another, highlighting the importance of tissue specificity. For instance, some clocks that accurately track biological aging in blood may prove unreliable when transferred to oral epithelial tissue, and the same holds true for other tissue pairings [181]. Indeed, Apsley et al. [182]. Demonstrated a relatively low correlation between epigenetic age estimates from blood samples and oral tissue. Furthermore, when the epigenetic inheritance model of blood origin was extrapolated to oral tissues, predictive biases occurred. The mismatch between this model and the actual data highlights the core challenges faced when applying the aging clock in clinical or research scenarios: samples and measurement data may originate from multiple tissue types, and there are significant differences in the cellular composition and epigenetic characteristics of different tissues inter-organizational heterogeneity brings difficulties to the universality of the model, and more refined analytical methods are needed to improve its accuracy in cross-organizational applications.

In the research of lung cancer (a recognized age-related disease), typical examples of tissue-specific aging clocks Have been discovered. In 2015, Levine et al. [171]. Used data From 2,029 women drawn from the Women’s Health Initiative to assess whether baseline levels of intrinsic epigenetic age acceleration (IEAA) in blood tissues might predict subsequent lung cancer risk. Strikingly, they discovered that individuals exhibiting higher IEAA scores were indeed more likely to develop lung cancer, marking the first evidence that a blood-based DNA methylation clock could offer a window into future malignancy. Although these data point toward an exciting application of senescence clocks in oncology, it is essential to appreciate that a clock optimized for blood tissue may not yield the same utility in analyzing lung biopsies. Researchers continue to refine and validate these blood-based models, aiming to integrate them with imaging data, circulating tumor DNA analyses, and traditional clinical risk factors for improved lung cancer screening and prognostication.

Beyond the study of lung cancer, there has been a surge of interest in characterizing tissue-specific epigenetic age in the brain, particularly in the context of neurodegenerative disorders such as Alzheimer’s disease (AD) [183, 184]. Grodstein et al. [185]. Leveraged longitudinal data from the Religious Orders Study and the Rush Memory and Aging Project to compare three widely used epigenetic clocks—Horvath, Hannum, and PhenoAge—as well as a newer DNA methylation clock trained on cortical tissue. Their findings revealed that brain-derived clocks can diverge significantly from blood-based clocks. This distinction is highly relevant to AD research, as the brain’s cellular composition, metabolic demands, and susceptibility to inflammatory changes may yield different epigenetic signals over time. Of particular note is the emerging connection between cellular senescence and AD progression, including evidence for tau-dependent neurodegeneration and the buildup of senescent cells in vulnerable brain regions [186, 187]. Levine et al. [171]. Extended these observations by measuring DNAm PhenoAge in the dorsolateral prefrontal cortex (DLPFC), finding that individuals diagnosed with AD displayed cortical tissue that appeared epigenetically “older” by roughly one year when compared to non-AD controls, even after adjusting for chronological age. Moreover, the degree of epigenetic age acceleration corresponded positively with core neuropathological features. These findings suggest that tissue-specific clocks, particularly those derived from postmortem brain samples, may one day become integral to monitoring AD progression or even identifying individuals at heightened risk before clinical symptoms manifest.

In parallel, research on senescence in breast tissue Has provided additional insight into how clocks can vary across organs and cell types. In 2024, a team at Westlake University constructed a single-cell transcriptomic-based aging clock in mouse mammary gland stem cells spanning a broad age range (2 to 29 months) [188]. By delineating cell-specific transcriptional trajectories across different developmental stages, the investigators explored how stem cell populations evolve under aging pressures and how these changes might predispose to malignancies such as breast cancer. Their work raises the hypothesis that tracking transcriptional aging signatures in mammary glands could uncover early alterations that precede tumor initiation. Meanwhile, Hofstatter et al. [189]. Investigated normal breast tissue samples (96 samples From 88 individuals) using epigenetic aging clock techniques. They identified a significant link between accelerated epigenetic aging in histologically normal breast tissue and a higher risk of developing breast cancer, thus indicating that earlier shifts in the local epigenetic landscape might set the stage for tumorigenesis. Although more in-depth research is needed on the specific causal mechanisms, these findings have highlighted the great potential of tissue-specific aging clocks as predictive or stratified tools in oncology.

Another organ where tissue-specific aging clocks have drawn attention is the kidney, especially in the context of chronic kidney disease (CKD). CKD is a significant global health burden, and accelerated aging is often implicated in its pathophysiology [190–192]. Responding to this need, Ognian and colleagues [190] developed a refined DNA methylation clock to characterize the relationship between Aging and kidney dysfunction. This study covered 47 patients who underwent kidney transplantation one Year later or began dialysis treatment, and was compared with 48 healthy controls. The results show that, compared with the healthy group of the same age, the biological age of patients with renal failure shows obvious signs of aging. This discovery suggests that the specific epigenetic clock of the kidneys may be used to assess the risk of an individual developing to end-stage renal failure, thereby providing crucial guidance for formulating treatment strategies to delay or even reverse renal damage. Considering that factors such as dialysis, immunosuppressive drugs and coexisting diseases may all have an impact on epigenetic characteristics, this tissue-specific model still needs to be further optimized and improved based on more abundant longitudinal data and a larger sample size.

It is worth noting that although “internal organs” such as the lungs, brain, breasts and kidneys have always been the research focus in the development of new aging clocks, the skin, as a surface tissue of the body, also has non-negligible scientific value in aging research. As the part of the body that comes into direct contact with the external environment, the skin is highly susceptible to ultraviolet radiation, environmental toxins and mechanical damage, making it an ideal model for analyzing the aging process, the regulation of circadian rhythms and the interaction mechanism of the tissue microenvironment [193–196]. The recent research by Duan et al. [197]. Explored the association between the disruption of the circadian rhythm of skin cells and accelerated aging as well as the pathological mechanisms of skin diseases. The study found that the imbalance of skin circadian rhythm homeostasis and the activation of aging-related pathways may synergically intensify inflammatory responses, inhibit cell renewal, and promote the remodeling of skin structure. This study reveals the specific role of the circadian rhythm aging regulatory network in skin tissue, providing a new perspective for clarifying the molecular mechanism of skin aging induced by external stress. Integrating circadian metrics into senescence clock models may not only provide insight into age-associated skin disorders but also yield generalizable principles relevant to other tissues, especially those subject to cyclical environmental or metabolic regulation [198, 199].

However, the latest research reveals that the tissue-specific limitations of the aging clock are far more complex than expected: Mildred Min et al. [200] review indicates that universal clocks can Have a systematic error of 20–30% in organs such as the lungs, kidneys, and skin. In the cerebral cortex, only when the dedicated Cortical Clock is used, the correlation between the pathological score of Alzheimer’s disease and biological age reaches 0.83 [185]. To make matters worse, clocks often struggle to distinguish between “normal aging” and “pathological processes” for instance, the oral tissue clock may misinterpret the methylation drift driven by periodontitis as aging, thereby systematically overestimation biological age [181]. Furthermore, the BLSA longitudinal study emphasizes that cross-cohort validation is difficult to replicate due to differences in measurement costs and dimensions [201]: The iCAS-DNAmAge developed by the Chinese team based on a cohort of 10,000 Han people Has a MAE reduction of 1.7 years in the East Asian sample compared to the Horvath clock. Reverse verification shows that the error in the European sample has increased instead [202]. Erin Macdonald-Dunlop et al. [203] research Found that the correlation of the same NMR metabolome clock between the Scottish and British biobanks was 0.74 vs. 0.21, respectively, suggesting that the lifestyle differences between island and mainland populations are sufficient to undermine the stability of the model. This further highlights the urgency of formulating a standardized, multi-organization, and multi-population verification framework.

Altogether, the application of senescent or epigenetic clocks across a range of tissues has revolutionized our grasp of age-related diseases and opened new avenues for preventative medicine and therapeutic intervention [204, 205]. The observed tissue specificity makes it clear that a “one-size-fits-all” approach to aging clocks is inadequate, given the substantial heterogeneity in cellular composition, environmental exposures, and regulatory networks across different organs. As the field advances, researchers are increasingly testing multi-tissue frameworks and leveraging machine learning techniques to integrate tissue-specific data into composite age estimates. These advanced new-generation models not only greatly enhance the accuracy of disease diagnosis, but also hold the promise of uncovering novel molecular targets that can intervene in or delay age-related diseases such as lung cancer, Alzheimer’s disease and chronic kidney disease. It is worth noting that optimizing and validating these models relies on large-scale longitudinal cohort studies, standardized data acquisition systems, and in-depth interdisciplinary collaboration between basic research teams and clinicians. Ultimately, if the aging clock can be comprehensively examined from a systematic perspective - that is, fully recognizing and utilizing the subtle differences in the aging characteristics of different organs - it will lay a solid methodological foundation for precise geriatric science. This can not only effectively delay age-related pathological processes, but also comprehensively enhance the healthy lifespan and quality of life of the elderly worldwide.

Association of the aging clock with disease

Recent research indicates that aging clocks can serve as powerful tools for quantifying aging trajectories, monitoring the effectiveness of therapeutic interventions, and identifying systemic changes that occur during both normal and pathological aging. For instance, Han et al. [206]. reported that these clocks effectively gauge improvements from anti-aging and rejuvenation therapies while helping to measure the metabolic impact of certain interventions. Their findings suggest that clinical and experimental protocols aiming to slow down or reverse age-related functional decline may draw on aging clock readouts as an objective endpoint, guiding the development of more targeted interventions.

Building on this concept, Li et al. [180]. conducted a detailed multi-omics study in 113 Healthy women aged 20 to 66 years, proposing that an integrative aging clock system—derived by combining data sets from various omics layers—can serve as a meaningful reference to evaluate individuals’ aging rates. The use of simultaneous epigenetic, transcriptomic, proteomic, and metabolomic profiling helps dissect broader population-level trends, such as chronic low-grade inflammation, endocrine imbalances, or metabolic dysregulation. Over time, these factors may converge to drive tissue degeneration and associated age-related diseases. Such multi-omics clock frameworks could potentially allow for more personalized medical interventions, as they highlight the specific molecular drivers of aging relevant to each individual.

At the molecular level, epigenetic patterns—most notably, DNA methylation—represent a key mechanism by which aging clocks track cellular division and senescence. Johnstone et al. [207]. illuminated how global hypomethylation may function as a “mitotic clock,” a process that accumulates with repeated somatic cell divisions. Their study linked hypomethylation rates to both the onset of cellular senescence and the potential for malignant transformation. From a translational perspective, pharmacological or genetic manipulations that either prevent excessive methylation loss or selectively clear hypomethylated cells may help avert age-related pathologies, including cancer. This line of investigation parallels the broader push to modulate epigenetic marks as a therapeutic strategy in geroscience.

Aside from epigenetics, single-cell transcriptomics also has emerged as a cutting-edge method for advancing our understanding of aging biology. In a pivotal study, Matthew et al. [208]. generated single-cell RNA sequencing data from the subventricular zone (SVZ)—a neurogenic region in 28 mice spanning multiple age groups. They built aging clocks based on transcriptomic changes in neural progenitor populations, thereby quantifying how regenerative capacity in the brain diminishes over time. Notably, their findings highlight the concept of “heterochronic parabiosis,” wherein tissues can exhibit younger or older molecular states when exposed to different systemic environments, suggesting that interventions such as blood factor exchanges or exogenous growth factors might one day be harnessed to slow age-related brain decline.

Further evidence of the power of single-cell analyses comes from Zhu et al. [209]., who examined blood single-cell RNA sequencing data in seven supercentenarians (SCs). These individuals, who surpassed 110 years of age, presented a unique window into human longevity. Notably, their cells showed increased ribosome-associated transcripts and reduced pro-inflammatory signaling pathways. This distinct molecular signature suggests that a heightened capacity for protein synthesis, coupled with a subdued chronic inflammatory state, might underpin exceptionally slow aging processes. Collectively, these data offer fresh targets—such as ribosome assembly or inflammatory mediators—for therapies aiming to foster healthy aging and ward off frailty. Proteomic strategies likewise provide insight into age-related vulnerabilities at the systems level. The proteomic age clock formed by Argentieri et al. [173]. harnesses large-scale plasma protein data to build models that predict chronological age while illuminating disease risk. By surveying thousands of plasma proteins, their clock can pinpoint potential molecular drivers behind multimorbidity. In the future, researchers hope to refine proteome-based models to identify specific protein targets that are amenable to modulation—through either pharmaceuticals or lifestyle changes—to forestall age-related changes and reduce the burden of chronic diseases.

Taken together, these multifaceted studies reinforce that aging clocks are rapidly evolving beyond mere academic curiosity. They now offer practical applications for gauging biological age, assessing disease susceptibility, guiding intervention strategies, and parsing the underpinnings of exceptional longevity. As research continues to integrate new high-throughput technologies—from multi-omics profiling to machine learning-based pattern recognition—aging clocks possess the transformative potential to redefine how we detect, manage, and prevent the myriad disorders that accompany the aging process. Large-scale and longitudinal validation, coupled with an emphasis on mechanistic insights into how specific interventions regress aging clock metrics, will be essential for ushering these tools into routine clinical practice. With continued refinement, aging clocks may become cornerstone methodologies in precision medicine, enabling clinicians and researchers alike to intervene more effectively in the biology of aging. Aging Clock: Multi-dimensional aging assessment from molecular to single-cell is shown in Fig. 4.

Fig. 4.

Aging Clock: Multi-dimensional aging assessment from molecular to single-cell

Depicts multi-omics frameworks for quantifying biological aging: (1) Bulk omics layers: Epigenetics: DNA methylation clocks (e.g., Horvath clock) track mitotic age and cancer risk. Transcriptomics: Gene expression shifts (e.g., CDKN2A/p16↑) reflect cellular stress responses. Proteomics/Metabolomics: Inflammatory cytokines (IL-6) and metabolite profiles predict disease susceptibility. (2) Single-cell resolution: scRNA-seq reveals cell-type-specific transcriptome changes (e.g., neural progenitor depletion → 58% decline in brain regeneration capacity). Cross-omics integration identifies inflammation-metabolism crosstalk (e.g., SASP-metabolic reprogramming).

Composition of the tumor-immune microenvironment in cellular senescence process

The tumor microenvironment (TME) is a complex and dynamic network composed of various cellular components and extracellular matrix. In this microenvironment, tumor cells constantly exchange signals with immune cells, stromal fibroblasts and extracellular matrix (ECM), and these signal exchanges can both promote and inhibit the progression of tumors. With the increase of age, the accumulation of senescent cells and the increase of inflammatory mediators jointly disrupt the balance of the TME, creating an environment conducive to tumor development. Specifically, a comprehensive analysis of the cellular composition in the TME (including innate immune cells and adaptive immune cells, tumor-associated fibroblasts) and the structural changes of the ECM is helpful for understanding how the aging process and cellular aging regulate the tumor’s response mechanism to treatment.

Aging-related changes of immune cells

The immune system undergoes profound remodeling with aging, characterized by functional decline and phenotypic shifts across multiple cell populations. These changes collectively impair anti-tumor immunity and contribute to the heightened susceptibility to malignancy observed in older individuals. Key aging-related alterations in immune cells are summarized in Table 2.

Table 2.

Aging-related phenotypic and functional changes in immune cells and their implications for tumor immunity

Cell Type	Major Subtypes/Functional Roles	Aging-Associated Alterations	Impact on Tumor Immunity	References
Lymphocytes	T cells (CD4⁺ helper T cells, CD8⁺ cytotoxic T cells) B cells Natural killer (NK) cells	T cells: Exhaustion phenotype (reduced proliferation, diminished cytokine secretion; downregulated CD28, upregulated PD-1, TIM-3, and KLRG1) B cells: Decreased immunoglobulin diversity, impaired high-affinity antibody production NK cells: Dysregulated expression of activating/inhibitory receptors, reduced tumor cytotoxicity	Compromised adaptive immune surveillance; Impaired antibody-mediated tumor clearancez; Enhanced tumor immune evasion	[210–213]
Macrophages	M1 (pro-inflammatory, tumoricidal) M2 (immunosuppressive, pro-angiogenic)	Shift toward “inflammaging” (chronic low-grade inflammation) Preferential polarization to M2-like phenotypes in the tumor microenvironment	Promotion of tumor progression via inflammatory mediators; Support for angiogenesis and immune escape	[214–219]
Neutrophils	Heterogeneous subsets (anti-tumor vs. pro-tumor [N2])	Reduced functional capacity (impaired chemotaxis, diminished oxidative burst) Skewing toward pro-tumor N2 subsets in older individuals	Diminished tumor clearance Enhanced immunosuppression and tumor growth	[220–223]
Eosinophils/Basophils	Mediators of allergic/parasitic responses; immune modulation via cytokine secretion	Altered cytokine secretion profiles (limited data in aging)	Potential dysregulation of tumor-associated inflammatory networks	[224, 225]
Dendritic Cells (DCs)	Professional antigen-presenting cells (bridge innate and adaptive immunity)	Reduced capacity for cross-presentation of tumor antigens Impaired T cell priming	Blunted anti-tumor T cell responses	[226–229]
Mast Cells	Mucosal/connective tissue residents; regulate vascular permeability and immune infiltration	Altered activation thresholds and cytokine secretion patterns in aging	Context-dependent effects (may promote or suppress tumor growth via immune modulation)
Other Blood Cells	Red blood cells (RBCs): Immune complex clearance Hematopoietic stem cells (HSCs): Progenitors of all immune lineages	RBCs: Impaired immunoadhesion function HSCs: Reduced self-renewal capacity and lineage differentiation potential	Compromised clearance of immune complexes; Impaired long-term immune competence	[230, 231]

These age-related immune cell changes disrupt the equilibrium between protective and pathogenic immune responses, creating a microenvironment that favors tumor initiation and progression. Senescent immune populations typically exhibit reduced cytotoxic activity, heightened immunosuppressive signaling, and chronic inflammatory tone—features that collectively weaken anti-tumor surveillance. Elucidating these shifts is essential for developing targeted strategies to restore immune competence in the aging host.

Recent studies, however, highlight significant challenges in characterizing immune cell senescence, particularly in distinguishing true senescence from other dysfunctional states and accounting for context-dependent heterogeneity [232, 233]. A primary hurdle is the lack of specific biomarkers: widely used markers such as CD28 loss, CD57 upregulation, and KLRG1 expression are not exclusive to senescence. For example, CD28⁻ CD8⁺ T cells retain limited proliferative capacity and may resist Treg-mediated suppression, blurring the line between senescence and terminal differentiation [234, 235]. Similarly, CD57, often linked to T cell senescence, marks highly cytotoxic terminally differentiated NK cells without senescent features, complicating cross-cell type comparisons [236].

Overlap with exhaustion and anergy further complicates characterization. Senescent T cells share markers like SA-β-Gal and p16^(INK4a) with exhausted T cells, while both exhibit impaired cytotoxicity and cytokine secretion [237, 238]. This overlap is exemplified by p21CIP1 upregulation, which occurs in senescence, exhaustion, and even beneficial immune tolerance responses, making it an unreliable standalone marker [239]. Lipofuscin accumulation—reflecting macromolecular damage and metabolic deregulation—emerges as a more specific marker, as it is absent in exhausted or anergic cells [233, 240], but its detection requires specialized tools like the fluorophore-conjugated reagent GLF16.

Immune cell senescence also exhibits remarkable heterogeneity, driven by tissue microenvironments and stress stimuli. For instance, senescent-like macrophages in the TME may express p16^(INK4a) and SA-β-Gal but retain functional plasticity, while alveolar macrophages in aged lungs show distinct senescence phenotypes linked to GM-CSF signaling [241]. In T cells, tumor-derived cAMP or Treg-mediated glucose competition induces senescence via ATM-dependent DNA damage, whereas MDSCs transfer exosomal GPR84 to activate p53 signaling—highlighting context-specific induction mechanisms [148, 242]. This “senescence mosaic” challenges the generalizability of findings from single tissues or stimuli.

Finally, model systems often fail to recapitulate in vivo senescence. In vitro induction via radiation or repeated division does not replicate the gradual, multi-factorial aging observed in tissues [243], while mouse models lack key features like CMV-driven T cell exhaustion [244]. Additionally, circulating immune cells poorly reflect tumor-infiltrating populations: senescence markers in blood T cells may underestimate TME-specific senescence due to tissue-specific epigenetic imprints, such as H3K27me3 modifications in aged T cells [245].

The role of the tumor-associated fibroblasts

Fibroblasts are mesenchymal cells fundamentally responsible for producing extracellular matrix components and orchestrating tissue repair [246, 247]. In the context of cancer, fibroblasts that become activated by tumor-derived factors are called CAFs. CAFs frequently arise from local or circulating precursor cells that adopt a myofibroblast-like phenotype, characterized by elevated α-smooth muscle actin (α-SMA) expression, enhanced contractility, and high-level production of growth factors and matrix proteins [248–250].

One of the core features of CAFs is their remarkable ability to modulate the immune environment. By releasing immunosuppressive cytokines such as TGF-β or IL-10, CAFs can limit the infiltration and function of cytotoxic T cells and NK cells. They can also recruit immunosuppressive cell subsets, including Treg cells and MDSCs, thereby establishing a permissive niche for tumor growth [251]. Aging can further exacerbate this process: as fibroblasts accumulate DNA damage or oncogenic signals over time, they may adopt a senescent CAFs phenotype marked by a potent SASP. This SASP not only alters matrix architecture but also fosters chronic inflammation that paradoxically can promote tumor progression. In some scenarios, senescent CAFs rigidify the extracellular matrix, creating a barrier that limits drug or immune-cell penetration into the tumor core [252].

Because CAFs hold central influence over both the mechanical stroma and the immunological milieu, they have emerged as a prime therapeutic target. Specific strategies include inhibiting key signaling pathways (e.g., TGF-β blockade) to reprogram CAFs from a tumor-promoting to a tumor-inhibitory state [253]. Additionally, senolytic drugs that selectively eliminate senescent CAFs are being investigated to disrupt the fibrotic stroma that hinders immunotherapies. However, contrasting studies show that certain CAFs subsets can also inhibit invasive tumor spread. Hence, the precise role of CAFs may be context-dependent, demanding robust biomarker profiling to ensure that interventions do not inadvertently enhance tumor aggressiveness. Regulation of the immune system by aging Fig. 5.

Fig. 5.

Regulation of the immune system by aging

Defines molecular cascades driving senescence: (1) Cycle arrest: Carcinogenic stress/DNA damage activates DDR→ p53 -p21CIP1 and p16INK4A→ inhibits CDK4/6, causes low phosphorylation of Rb and impairs E2F, blocking S-phase genes (PCNA, CDK2, CCNE1). (2) SASP amplification: The cytoplasmic DNA fragment triggers cGAS-STING, the TLR/IKK pathway phosphorylates IκB, promoting the nuclear translocation of NF-κB (p50/p65), and inducing the transcription of IL-6, TGF-β, VEGF, etc. (3) Immune escape: Chronic SASP upregulates inhibitory molecules such as PD-L1 and TIM-3, leading to T cell dysfunction and resulting in “blocked infiltration”.

Role of the extracellular matrix in the tumor-immune microenvironment

The ECM provides structural integrity, mechanical cues, and biochemical signals that together shape cell behavior. In solid tumors, ECM proteins (including collagen, fibronectin, laminin, and proteoglycans) often become dysregulated in composition, organization, and crosslinking [254]. This remodeling of the ECM can influence everything from angiogenesis to T cell infiltration and tumor cell migration.

High ECM density or enhanced crosslinking can raise interstitial fluid pressure, impeding effective drug delivery [255]. A densely fibrotic stroma may thereby serve as a physical barrier to immune cells, preventing effector T cells and NK cells from contacting malignant cells [256]. Moreover, ECM fragments generated by excessive proteolysis (e.g., by matrix metalloproteinases) can act as signaling ligands, activating immunosuppressive pathways and further fueling tumor invasion [257]. In older patients, age-related ECM stiffness—resulting from continuous collagen deposition and crosslinking—may deepen these barriers, exacerbating tumor aggressiveness and diminishing immunotherapeutic efficacy [258]. ECM molecules also orchestrate crosstalk between various stromal cells and tumor-infiltrating immune cells. For instance, changes in the ECM composition can promote macrophage polarization towards an anti-inflammatory M2 phenotype, thus impairing T cell–mediated tumor rejection [259]. In addition, tumor-associated fibroblasts produce ECM components that facilitate interactions with tumor-associated macrophages (TAMs), as highlighted by Yu et al. [260]. These interactions can potentiate immunosuppression within the TME, diminishing the effectiveness of checkpoint inhibitors and other immunotherapies [261]. Hence, therapeutic strategies that modulate ECM composition—such as collagenase-based degradation or inhibition of lysyl oxidase (LOX) crosslinking enzymes—are being explored to “soften” the stroma, increase immune-cell infiltration, and improve the efficacy of chemotherapy or immunotherapy [262].

Interaction of cellular senescence with the tumor immune microenvironment

Cellular senescence refers to the permanent stagnation state of the cell cycle induced by oncogene activation, telomere depletion or other stress factors. This mechanism was initially regarded as a tumor suppression strategy that could exert a blocking effect in the early stage of abnormal cell proliferation. However, as organisms age, senescent cells gradually accumulate, and they secrete an aging-related secretory phenotype (SASP) containing pro-inflammatory and growth factors. At this point, the aging mechanism may “reverse” and contribute to tumorigenesis. In the tumor microenvironment, apart from tumor cells themselves, numerous cell types such as immune cells, endothelial cells, and stromal fibroblasts may also enter the senescent state. Their combined actions reconfigure local and systemic immune responses to cancer.

Effect of senescent cells on immune cells

Senescent cells can release complex extracellular vesicle (EV) populations (commonly called exosomes or microvesicles). These senescent cell–derived EVs (SEVs) ferry microRNAs, proteins, and lipids that alter the behavior of immune cells [263–265]. Monocytes, neutrophils, and B cells internalize SEVs through endocytic pathways, sometimes undergoing functional reprogramming. Pretreating monocytic THP-1 cells with SEVs was shown to boost LPS-induced TNF-α production, pushing these cells into a more pro-inflammatory mode [266, 267]. While a transient inflammatory surge can be protective, persistent stimulation can drive chronic inflammation (“inflammaging”), which supports immunosuppressive remodeling within tissues, including tumors [268, 269]. Thus, SEVs highlight the dualistic nature of senescent cells in shaping the TME: they may initially trigger heightened immune surveillance, but prolonged exposure can eventually dampen vigilance and promote immune exhaustion [270]. In addition to EVs, senescent cells also secrete soluble factors such as IL-6, IL-1β, TGF-β, and chemokines, etc. These factors impair the function of the immune system through multiple mechanisms. They not only inhibit the antigen presentation ability of dendritic cells and hinder the recruitment of naive T cells, but also promote the clonal expansion of regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs) [271–273]. This reconstruction of the immune cell lineage will further deteriorate the immunosuppressive state of the tumor microenvironment and greatly weaken the body’s ability to clear tumor cells. Furthermore, senescent endothelial cells can selectively mediate the tissue infiltration of immunosuppressive cells by up-regulating adhesion molecules (such as ICAM-1 and VCAM-1), making the tumor microenvironment more difficult to achieve effective tumor control [45, 274–276]. To sum up, the aging-related secretory phenotype (SASP) exerts extensive and complex regulatory effects on both the innate immune and adaptive immune systems. In the field of geriatric oncology, how to selectively inhibit the harmful components in the aging-related secretory phenotype (SASP) while retaining its physiological protective function has become a key scientific problem that urgently needs to be overcome in this field.

Accumulation of senescent cells in the tumor microenvironment

Senescent cells tend to accumulate in aged tissues due to reduced clearance by the immune system and repeated exposure to genotoxic stresses (e.g., chemotherapy, radiotherapy) [277, 278]. In the tumor setting, oncogenic signaling and paracrine factors from cancer cells can also accelerate stromal senescence [8, 279]. As Krtolica et al. [132] demonstrated, senescent fibroblasts can stimulate malignant traits in neighboring epithelial cells, an example of evolutionary “antagonistic pleiotropy.” While senescence of potentially oncogenic cells is protective early in life, the persistent presence of senescent stromal cells in older organisms can paradoxically drive tumor growth [128]. Senescent fibroblasts, endothelial cells, and immune cells in the TME commonly secrete proteases and ECM-remodeling enzymes that restructure the tumor niche, enhancing invasiveness. They can also release pro-angiogenic factors like VEGF, amplifying the vascular supply necessary for tumor expansion [280]. In certain breast cancer models, senescent endothelial cells were associated with increased vascular permeability, facilitating both tumor-specific immune infiltration and, paradoxically, metastatic dissemination [281]. Intriguingly, such observations underscore the context dependence of senescence within the tumor microenvironment. In some scenarios, senescent stromal cells help immune cells detect nascent tumors; in others, they set the stage for immune evasion and metastatic spread [4].

The aging paradox: the dual phase role of SASP in tumor immunity

The relationship between cellular senescence and the tumor immune microenvironment is not unidirectional but participates in the entire process of tumor occurrence and development through a dynamic balance mechanism. Among them, the proposal of the senescence paradox provides a key perspective for analyzing this relationship: Aging significantly increases the incidence of cancer, but cellular senescence itself has the potential to activate immune surveillance and suppress cancer. This seemingly contradictory phenomenon actually stems from the spatiotemporal dynamics of the aging process - short-term acute aging can inhibit tumors, while long-term chronic aging promotes tumor progression by reshaping the microenvironment.

SASP, as the core hub connecting cellular senescence and the immune microenvironment, exhibit a significant spatiotemporal dependent double-edged sword effect in their functions. In young individuals or in the early stage of tumor occurrence, SASP activates the NF-κB pathway through IL-6/TNF-α, and the cGAS-STING signaling triggered by the coplasmic DNA fragment recruits natural killer cells (NK cells) and cytotoxic T cells to clear malignant cells, exerting an evolutionally conserved “immune surveillance alarm” role [282–284]. However, with the increase of age, the clearance efficiency of senescent cells decreases, leading to their accumulation in tissues, driving SASP to shift from an acute defense signal to a chronic pathological stimulus [285]. This chronic transformation reshapes the tumor microenvironment through a triple cascade reaction: (1) extracellular matrix degradation mediated by matrix metalloproteinases (MMPs); (2) Abnormal angiogenesis induced by vascular endothelial growth factor (VEGF); (3) The depletion of T cells caused by the upregulation of immune checkpoint molecules such as PD-L1 eventually establishes an immunosuppressive ecological niche [286]. This spatio-temporal evolution process perfectly interprets Williams’ antagonistic multiplicity hypothesis [287] - the tumor suppressor function of SASP during the reproductive period is retained by natural selection, but in the post-reproductive period lacking selection pressure, its cumulative damage leads to an exponential increase in cancer incidence with age, forming an evolutionary paradox of “guardian in youth and rebel in old age“ [288].

Aging and tumor immune escape mechanisms

The efficacy of antitumor immunity hinges upon a coordinated cascade encompassing antigen release, APC-mediated processing, T cell priming, trafficking, and target cell elimination—a process systematically compromised by aging and cellular senescence [289]. Senescent dendritic cells exhibit impaired cross-presentation, while aged T cells display elevated inhibitory checkpoints (PD-1, CTLA-4, LAG-3) that drive functional exhaustion [290–292]. This erosion of immune competence synergizes with tumor-intrinsic mechanisms: malignant cells exploit checkpoint pathways to establish “cold” microenvironments devoid of effector T cells [238], further exacerbated by age-related declines in NK/T cell cytotoxicity [248]. Notably, in non-small cell lung cancer, proliferative defects and dysregulated cytokine secretion cripple memory T cell surveillance, with tumor-derived immunosuppressive factors amplifying immune paralysis [293, 294]. Consequently, age-associated immunosenescence fosters an immune-privileged niche conducive to metastasis.

Emerging tools now quantify this pathophysiological nexus. DNA methylation-based aging clocks reveal that accelerated epigenetic aging—indexed by elevated “senescence reprogramming scores”—correlates with key immunosuppressive features: expanded MDSCs/Tregs and constrained TCR diversity in elderly patients [8]. Mechanistically, senescence-associated epigenetic modifiers like EZH2 catalyze H3K27me3 deposition at inflammatory loci (e.g., IFNG), silencing antitumor immunity [9]. Thus, while acute senescence may transiently bolster immune surveillance via SASP-mediated alarm signals (e.g., cGAS-STING activation), chronic senescence ultimately dominates in aged hosts, sculpting protumorigenic microenvironments through stromal remodeling and immune dysfunction [13, 295]. Therapeutically, this dynamic equilibrium necessitates precision strategies: senolytics to eliminate pathologic senescent cells, SASP modulators to attenuate chronic inflammation, and epigenetic reprogramming to reverse immune cell exhaustion. Integrating aging clocks as biomarkers could stratify patients for checkpoint blockade or cytokine therapy, particularly in immunocompromised geriatric cohorts. Harnessing senescence biology—augmenting its early tumor-suppressive functions while neutralizing late-stage protumor effects—represents a frontier in translational oncology.

Prospect of the aging clock in cancer therapy

There is a close biological association between aging and cancer: Aging is not only a core risk factor for the occurrence of various malignant tumors, but also elderly patients generally have a lower tolerance to standard chemotherapy regimens. To address this dual challenge, researchers are focusing on the “aging clock” - a comprehensive assessment tool that can quantify the aging process of organisms (rather than simply timing age). By unveiling how senescence mechanisms evolve within our tissues, aging clocks promise to refine risk assessments, therapeutic targeting, and potentially even preventative measures. In recent years, multiple lines of evidence—from large-scale population studies to experimental work in model organisms—underscore the usefulness of these clocks in clinical oncology.

Pharmacological interventions targeting aging mechanisms: advances in clinical research and potential relevance for tumor therapy

When clocks detect accelerated senescence in key compartments (e.g., immunosenescence via epigenetic signatures [296]), senolytics emerge as a cornerstone strategy to selectively eliminate senescent cells, directly rejuvenating the tumor microenvironment (TME [100]). This approach synergizes with conventional oncology: preclinical studies demonstrate senolytics enhance radiotherapy and immune checkpoint inhibitors by reducing immunosuppressive SASP factors (TGF-β, IL-6) and increasing CD8 + T-cell infiltration. Clinically, ongoing trials validate this paradigm—Dasatinib-quercetin + SBRT in NSCLC (NCT04733534, Phase Ib [297]) shows 70% improved progression-free survival, while navitoclax-atezolizumab in TNBC (NCT04123366, Phase II) achieves 38% objective response rate.

Notably, traditional systemic senolytics (e.g., free dasatinib, navitoclax) cause off-target toxicities (doxorubicin-induced cardiotoxicity, navitoclax-related thrombocytopenia), limiting clinical use. Three preclinically validated targeted strategies address this:

GalNPs [298]: ~100 nm silica nanoparticles coated with 6-mer β(1,4)-galacto-oligosaccharides, encapsulating cytotoxics (doxorubicin, navitoclax). They leverage senescent cells’ 2–3-fold higher SAbGal activity—SAbGal hydrolyzes the coating to release drugs selectively. In palbociclib-treated SK-MEL-103 xenografts, GalNP(dox) reduced tumors by 47% (vs. 22% free doxorubicin) and eliminated cardiotoxicity; in bleomycin-induced Fibrosis, it cut collagen by 58% and normalized lung function without hepatic/renal damage.

mGL392 [299]: PEO-b-PCL micelles carrying a lipofuscin-binding domain (LBD) and dasatinib (ester-linked). Lipofuscin (senescent-exclusive) anchors mGL392, and esterases release dasatinib. It had a selectivity index > 8.3 (vs. 1.82 free dasatinib) in vitro, eliminated 76% SA-β-gal⁺ cells (vs. 42% free dasatinib) in B16 xenografts, and kept liver/muscle markers normal.

TRX-CBI + DPP4 sorting [300]: TRX-CBI (Fe(II)-activated prodrug) releases CBI in senescent cells (2.1-fold higher Fe(II)). DPP4 sorting enriches paracrine senescent cells (40%→85%), and TRX-CBI killed 78% of them (vs. <15% non-senescent toxicity).

In addition to senescent cell scavengers, other complementary intervention measures include circadian modulators (e.g., REV-ERB agonists) to restore immune rhythms, mitophagy inducers (e.g., urolithin A) for mitochondrial rejuvenation, and caloric restriction mimetics (e.g., metformin [301]) to attenuate nutrient-sensing dysregulation. Equitable implementation requires tiered pricing for accessible senolytics (e.g., generic fisetin), portable epigenetic clocks, and open-access biomarker platforms. Ultimately, deploying senolytics as first-line aging interventions shifts cancer care from tumor-centric to host-directed therapy, concurrently targeting malignant cells and age-permissive microenvironments to intrinsically delay biological aging [302].

Ongoing clinical investigations into pharmacological interventions targeting aging mechanisms have yielded promising yet preliminary findings in several key therapeutic candidates [303]. Rapamycin, a mechanistic target of rapamycin (mTOR) inhibitor, demonstrated acceptable safety profiles in healthy elderly populations (70–95 years) during an 8-week randomized controlled trial (1 mg daily dose). Although cognitive and metabolic parameters remained unaltered, this intervention exhibited hematological modulatory effects through decreased hemoglobin levels and induced immunophenotypic alterations in lymphocyte subsets [304]. Its structural analog, everolimus (RAD001), displayed enhanced potential in mitigating immunosenescence during influenza vaccination responses. Dual dosing regimens (0.5 mg daily vs. 5 mg weekly) achieved a 20% elevation in antibody titers and significant downregulation of PD-1 expression in both CD4 + and CD8 + T cell populations [305].

Metformin, as a traditional First-line treatment drug for type 2 diabetes, has become a highly potential anti-aging candidate drug due to its multipotent regulatory effects on aging-related pathways. Preclinical studies have shown that this drug can extend the healthy Lifespan of rodents by up to 30%. Mechanistically, it inhibits mitochondrial complex I, activates AMP-activated protein kinase (AMPK), reduces insulin/IGF-1 signaling, and attenuates the senescence-associated secretory phenotype (SASP). The landmark Targeting Aging with Metformin (TAME) trial, a proposed 6-year randomized controlled trial enrolling 3,000 non-diabetic individuals aged 65–79, aims to assess its effects on delaying age-related diseases. Notably, the UK Prospective Diabetes Study (UKPDS) reported a 36% reduction in all-cause mortality among metformin-treated patients over a 10-year follow-up, outperforming conventional diabetes therapies. These benefits are hypothesized to originate from dual modulation of nutrient sensing (mTOR/IGF-1 inhibition) and inflammatory pathways (reduced IL-6, TNF-α secretion), as supported by preclinical data showing decreased senescent cell accumulation and improved mitochondrial function. While short-term trials confirm safety and metabolic improvements, ongoing research seeks to validate its role in slowing biological aging through biomarkers such as GDF15 and cystatin C, positioning metformin as a leading candidate for translating geroscience into clinical practice [306].

In a 6-week randomized double-blind placebo-controlled crossover trial in healthy mid-to-older adults (55–79 years, n = 24 completers), nicotinamide riboside (NR, 500 mg bid), a NAD + precursor, was well-tolerated, with mild adverse events in < 10% and no serious safety issues. Primary outcomes revealed a significant ~ 60% increase in peripheral blood mononuclear cell NAD + levels and a near five-fold elevation in nicotinic acid adenine dinucleotide (NAAD), a marker of NAD + metabolism. Exploratory analysis showed that systolic blood pressure and carotid femoral pulse wave velocity showed a decreasing trend, and this effect was more significant in individuals with elevated baseline blood pressure. Although glucose regulation and exercise capacity were unaffected, NR’s modulation of NAD+-dependent pathways aligns with preclinical evidence of vascular protection via sirtuin activation and oxidative stress reduction. These findings support NR as a candidate for calorie restriction-mimetic interventions in aging, justifying larger trials to validate cardiovascular benefits and underlying mechanisms [307].

Sodium-glucose cotransporter 2 (SGLT2) inhibitors, initially developed For managing type 2 diabetes, have emerged as promising candidates for targeting aging-related pathways, demonstrating benefits beyond glycemic control. Preclinical studies in murine models, such as TA-1887 in diabetic mice and canagliflozin in genetically heterogeneous mice, show extended lifespan (a 14% median survival increase in males) and reduced age-related pathologies. Clinical trials, including EMPA-REG OUTCOME (22% reduction in cardiovascular death, HR = 0.78, 95%CI 0.64–0.97) and DAPA-HF (26% lower heart failure hospitalization risk, HR = 0.74, 0.65–0.85). These effects are hypothesized to involve dual mechanisms: systemic metabolic reprogramming (promoting ketogenesis, fatty acid oxidation) and direct cellular actions (inhibiting NLRP3 inflammasome, mitigating endothelial senescence, restoring mitochondrial dynamics). Observational data and meta-analyses further support their role in lowering pro-inflammatory cytokines (e.g., IL-6) and improving markers of vascular and renal health. While generally well-tolerated, long-term studies are required to fully characterize their impact on lifespan and age-related diseases, positioning SGLT2 inhibitors as a novel class for geroscience interventions targeting core aging processes [308].

S-nitrosoglutathione reductase (GSNOR), a key enzyme regulating protein S-nitrosylation, declines with age in primary cells, mouse tissues, and human peripheral blood mononuclear cells, driven by epigenetic downregulation via TET1-mediated promoter methylation. This decline leads to excessive S-nitrosylation of mitochondrial proteins such as Drp1 and Parkin, disrupting mitochondrial dynamics (fission-fusion balance) and impairing mitophagy, resulting in mitochondrial fragmentation, reduced transmembrane potential, and dysfunction—hallmarks of cellular senescence. GSNOR-deficient mice exhibit accelerated aging phenotypes, including protein aggregation and motor dysfunction, while exceptionally long-lived humans maintain high GSNOR and TET1 expression, avoiding age-related epigenetic silencing. Mechanistically, Gsnor-mediated nitrosation stress damages mitochondrial quality control, thereby establishing the association between S-nitrosylation and the theory of mitochondrial free radical aging. These findings establish the status of GSNOR as a key longevity regulator and emphasize that targeting the TET1/GSNOR axis to restore mitochondrial homeostasis may become a potential strategy for intervening in age-related diseases [309].

Bisphosphonates, primarily used to treat osteoporosis by inhibiting osteoclast-mediated bone resorption [310], have emerged as potential Lifespan-modulating agents, with growing evidence Linking their use to reduced mortality in both skeletal and non-skeletal contexts. The HORIZON trial demonstrated a 28% reduction in all-cause mortality (HR = 0.72, 95% CI 0.56–0.93) in Hip Fracture patients treated with zoledronate over 3 years, along with reduced fracture risk, prompting exploration of their broader anti-aging effects. Observational studies and meta-analyses further support this benefit, with nitrogen-containing bisphosphonates (e.g., alendronate, risedronate) showing stronger associations with survival improvement, particularly in women and individuals with higher baseline mortality risk. In cancer populations, bisphosphonates reduce skeletal-related events and demonstrate survival benefits in breast and colon cancer, possibly via tumor microenvironment modulation and enhanced anti-tumor immune responses. The dual role of bisphosphonates in preserving bone health and exerting systemic anti-aging effects positions them as promising candidates for interdisciplinary interventions targeting age-related diseases [311].

Notwithstanding these advances, critical challenges persist in anti-aging therapeutic development. Current limitations include constrained sample sizes in early-phase trials, insufficient longitudinal safety data, and the inherent complexity of quantifying multidimensional aging phenotypes. Future research priorities should emphasize large-scale randomized controlled trials with composite healthspan metrics as primary endpoints, complemented by integrative multi-omics biomarker profiling to optimize intervention strategies and personalize therapeutic approaches in aging populations. Table 3 Anti-Aging Drugs in Clinical Trials: Preclinical vs. Clinical Evidence Classification. The relationship between tumor therapy and cell aging is shown in Fig. 6.

Table 3.

Anti-aging drugs in clinical trials: preclinical vs. clinical evidence classification

Drug Class	Representative	Senolytic Strategy	Preclinical Evidence (Aging/Cancer)	Clinical Evidence (Aging/Cancer)	Trial Phase	Cancer-Related Evidence	Reference
mTOR inhibitors	Rapamycin	Indirect (autophagy)	Lifespan extension in yeast/mice; tumor growth suppression	Improved vaccine response ↓inflammation in elderly	I/II	Tumor suppression (preclinical only)	[304, 312]
NAD + boosters	Nicotinamide riboside	No	Improved mitochondrial function in aged mice	↑ PBMC NAD+; ↓BP in hypertensive	II	No direct evidence	[308, 313]
Biguanides	Metformin	Indirect (SASP)	lifespan extension in mice; chemoprevention	↓all-cause mortality (UKPDS); TAME trial ongoing	III (TAME)	↓Cancer incidence in diabetics (clinical)	[306]
SGLT2 inhibitors	Empagliflozin	No	14% lifespan extension in male mice	22% ↓CVD death (EMPA-REG); ↓renal decline	III	↓Cancer mortality (post-hoc analyses)	[308, 314]
GLP-1 RAs	Semaglutide	No	Inhibited tumor growth in pancreatic models	↓AD risk; renal protection	III	Preclinical tumor inhibition	[315]
Senolytics(Non-targeted)	Dasatinib + Quercetin	Direct (PI3K/AKT)	Reduced chemo-induced senescence	↓senescent cells in IPF; improved exercise endurance	I/II	Ongoing in solid tumors (NCT04733534)	[316, 317]
	Fisetin	Direct (LMP)	Reduced senescent glial cells	↓Inflammatory indices; cognitive benefits in MCI	II	Chemosenescence clearance (preclinical)	[318]
Senolytics(Targeted)	GalNPs (doxorubicin/navitoclax)	Direct (SAbGal-mediated drug release)	Abrogated doxorubicin cardiotoxicity; 47% ↓tumor volume (SK-MEL-103 xenografts); 58% ↓lung collagen (fibrosis models)	No systemic toxicity (hepatic/renal); ongoing in pulmonary fibrosis	Preclinical (translational)	Eliminated senescent tumor cells; avoided systemic toxicity (preclinical)	[298]
	mGL392 (dasatinib conjugate)	Direct (lipofuscin-targeted; esterase-mediated release)	Selectivity index > 8.3 (vs. 1.82 for free dasatinib); 76% ↓SA-β-gal⁺ cells (B16 xenografts)	No liver/muscle toxicity; ongoing in melanoma	Preclinical (translational)	62% ↓tumor volume; no off-target cell damage (preclinical)	[299]
	TRX-CBI + DPP4 sorting	Direct (Fe(II)-activated; DPP4-enriched paracrine senescent cells)	78% ↓paracrine senescent cells; blocked tertiary senescence; avoided ABT-263 resistance	< 15% toxicity to non-senescent cells; ongoing in vascular aging	Preclinical (translational)	Cleared therapy-induced senescent cells (preclinical); prevented relapse	[300]
Bisphosphonates	Zoledronic acid	No	Reduced metastasis in breast cancer models	28% ↓all-cause mortality (HORIZON); ↓breast cancer recurrence	III	↓Recurrence in breast cancer (clinical)	[319]
Telomerase activators	TA-65	No	Delayed cellular senescence in vitro	↑Telomere length; improved immune function	II	No direct evidence	[320]
SERMs	Raloxifene	No	Reduced mammary tumorigenesis in models	33% ↓MCI risk; ↑bone density	III	Chemoprevention (preclinical)	[321]

Fig. 6.

The relationship between tumor therapy and cell aging

Spatial mapping of senotherapy efficacy versus toxicity: Horizontal axis (therapeutic axis) - Immune activation: PD-1 inhibitors enhance the clearance of NKG2D⁺NK cells, and IL-7/IL-15 improve the mitochondrial metabolism of T cells to delay exhaustion; SASP inhibition: JAK blocks IL-6 signaling, and HMGB1 releases to pre-stimulate T cell responses. Vertical axis (adverse reactions) - Fibrosis: MMP mediates collagen deposition in the lungs/skin; Autoimmunity: Off-target attack damaging normal stem cells (uPAR-CAR T-induced nephrotoxicity risk); Crosstalk hazard: Mitochondrial damage leads to mtDNA leakage →cGAS-cGAMP-STING-NF-κB cascade amplifies inflammation and deteriorates in synergy with oxidative stress.

The aging clock serves as a biomarker

DNA methylation patterns are among the most thoroughly investigated aging biomarkers. Numerous studies show that specific CpG sites across the genome undergo predictable methylation shifts with advancing age, thus forming the foundation for “epigenetic clocks” [322, 323]. It is notable that the aging clock based on DNA methylation has significantly better efficacy than the evaluation model based on chronological age in predicting health outcomes such as mortality, cardiovascular diseases and specific cancers. Accelerated epigenetic aging (i.e., the age calculated by the epigenetic clock being greater than the actual age) is usually significantly associated with an increased risk of cancer and a reduced survival rate of patients. In the field of clinical oncology, detecting a patient’s “methylation age” can be used to assess whether the aging process of tissues is abnormally accelerated. This technique is expected to reveal the biological weaknesses exploited by tumor cells or the key vulnerable sites that affect the treatment response. From the perspective of the mechanism of action, DNA methylation achieves precise regulation of gene expression by regulating the chromatin accessibility of the promoter region or reshaping the enhancer function. The existence of consistent age-associated methylation patterns across tissues indicates that these epigenetic shifts are responsive to both intrinsic processes (e.g., DNA damage, telomere attrition) and extrinsic factors (diet, inflammation, or environmental toxins). As bioinformatics tools evolve, multi-tissue methylation clocks have been developed, lending greater precision to evaluate systemic versus tissue-specific aging [11]. This stratification means clinicians and researchers may one day combine epigenetic clock metrics with imaging, circulating tumor DNA (ctDNA), or single-cell sequencing to gain deeper insights into tumorigenesis and therapy optimization.

Beyond DNA methylation, histone PTMs—such as acetylation, methylation, phosphorylation, and ubiquitination—significantly influence chromatin structure and gene transcription. Shifts in histone marks (e.g., H3K27me3, H4K16ac) have been implicated in aging and tumor development. While standardized “histone clocks” remain less mature than DNA methylation-based clocks, emerging data suggest these marks also exhibit age-specific patterns that could be harnessed for clinical purposes. Proteomics-based clocks, meanwhile, measure changes in the abundance of selected signaling molecules, enzymes, structural proteins, and other functional proteins in biological fluids or tissues [173]. The loss of protein homeostasis, as a classic marker of aging, is manifested as protein aggregation/misfolding, abnormal proteasome degradation function and imbalance of the protein homeostasis network. These proteomic alterations are not only molecular manifestations of age-related pathology, but also directly promote the progression of various diseases, including cancer. With the development of high-throughput proteomics technology, dynamic monitoring of age-related proteomic characteristics is expected to provide real-time molecular indicators for disease severity assessment and treatment response monitoring. In addition, this technology is also expected to reveal novel molecular interaction networks - such as key pathways related to immune regulation imbalance or metabolic stress in the tumor microenvironment - providing potential intervention targets for the development of targeted drugs.

Non-coding RNAs (ncRNAs), notably miRNAs and lncRNAs, also participate in shaping gene expression programs that underlie aging and cancer progression [324–326]. Some micrornas (miRNAs) can be involved in the molecular regulation of the aging process by inhibiting the DNA repair pathway or regulating cell cycle regulatory proteins. It is worth noting that the dysregulation of non-coding RNA (ncRNAs) expression is widespread in tumors. It promotes the malignant progression of tumors by affecting the metastatic potential and immune escape mechanism of tumor cells. Researchers are attempting to construct a multi-omics aging clock by integrating ncRNA expression characteristics, epigenetic modification profiles, and proteomics data. This aging assessment model based on multi-dimensional molecular features can capture the biological complexity of aging at multiple levels, including post-transcriptional regulation, epigenetic modification, and protein function. This type of multi-omics model is expected to further improve the prediction accuracy and clinical application value of aging-related diseases by internally mapping the synergy and redundancy relationship of each data layer.

Future directions and challenges

Before aging clocks can be fully integrated into routine cancer treatments, robust validation across diverse geographic, ethnic, and generational cohorts is essential [327]. Standardizing protocols—from sample collection to bioinformatic modeling—will facilitate reproducibility. Moreover, integrating various data types requires advanced machine learning pipelines capable of interpreting multi-omics information while accounting for confounding factors like comorbidities, medication histories, and lifestyle. Such thorough validation will help persuade regulatory bodies to consider using aging clock measures in clinical decision trees, akin to well-established biomarkers like prostate-specific antigen (PSA) or BRCA mutational status [328].

The use of aging clocks in cancer therapy raises ethical questions and practical perspectives that warrant deeper exploration, particularly regarding treatment decisions and health equity. These tools, while offering potential for more precise care, introduce complexities that could reshape clinical practices and exacerbate existing disparities.

The integration of aging clocks into clinical oncology necessitates rigorous ethical scrutiny, particularly regarding their impact on life-sustaining intervention decisions such as Do-Not-Attempt-Resuscitation (DNAR) orders. A biologically older profile in a younger advanced-cancer patient might inadvertently justify restrictive DNAR directives, potentially denying interventions that align with patient goals. Conversely, a “biologically younger” older adult might face pressure to pursue aggressive therapies against their wishes. Thus, aging clocks must be contextualized within holistic frameworks that prioritize patient autonomy, symptom burden, and quality-of-life preferences rather than replacing them [329].

The economic implications of aging clocks extend beyond clinical costs to systemic inequities. Large-scale multi-omics profiling and continuous monitoring remain financially prohibitive, risking exclusion of socioeconomically disadvantaged groups from precision oncology benefits. Worse, insurance providers might leverage “accelerated aging” readings to deny coverage or inflate premiums, creating a novel form of biological discrimination. Regulatory safeguards—akin to the Genetic Information Nondiscrimination Act (GINA)—are urgently needed to prohibit such misuse. Notably, studies confirm that epigenetic age reversibility is achievable through interventions like weight management, underscoring the need to communicate results as dynamic and modifiable rather than deterministic.

To mitigate these risks, ethical guardrails must be reinforced through three interconnected pillars. First, rigorous validation across diverse ethnic, geographic, and generational cohorts is essential to minimize algorithmic biases, coupled with standardized protocols from sample collection (e.g., controlling for circadian effects on DNA methylation [329]) to bioinformatic modeling to ensure reproducibility. Second, clinician training should embed ethical communication frameworks that explicitly articulate biological age as a dynamic indicator of current physiological stress—such as cancer-related inflammation transiently accelerating epigenetic clocks—rather than a fixed lifespan determinant; tools like the AAHPM Shared Decision-Making Toolkit can operationalize this by bridging technical data with patient-centered values. Third, policy-driven equity measures are critical, including subsidized access in public health systems and need-based (rather than payment-driven) inclusion criteria; proactive policies must prevent the misuse of aging clocks as exclusion tools, as evidenced by the UK DNAR controversy where crisis-driven rationing amplified health inequities. Beyond ethical considerations, the integration of aging clocks with geroscience enables precision-targeted interventions against organ-specific aging.

Conclusion and future prospect

Cellular senescence stands as both a guardian and a potential saboteur in cancer control. Early in life, senescence halts unwanted proliferative signals; later on, senescent cells accumulate and shape a microenvironment that can foster tumorigenesis, fibroblast activation, and immunosuppression. Aging clocks, which measure subtle molecular alterations associated with senescence progression, can help clinicians and researchers parse out when and where senescence transitions from protective to pathological [45]. Beyond mere quantification, these clocks pull back the curtain on the biochemical crosstalk driving tumor-immune interactions, clarifying how immune surveillance may falter with age.

Advances in single-cell multi-omics, improved computational power, and machine learning algorithms will allow more precise delineation of tissue-specific versus systemic aging processes. Robust clocks that integrate epigenetic, transcriptomic, proteomic, and ncRNA-based data could act as early warning systems, signaling escalations in aging-driven oncogenic pathways and prompting precautionary interventions. SASP can be seen as a “double-edged sword.” It may recruit and activate immune cells to eradicate nascent tumorigenic cells or, conversely, sustain chronic inflammation and immunosuppression. Disentangling these opposing effects requires deeper structural and functional analyses of SASP components across cancer stages, potentially unveiling druggable targets to tilt the SASP balance from pro-tumor toward anti-tumor. On the therapeutic front, synergy between aging clocks, targeted drugs, and immunotherapies holds great promise. By integrating the “biological age” index of patients into clinical decision-making algorithms, doctors can precisely optimize the timing of immunotherapy, rationally apply senescent cell clearance therapy, or target and regulate the inflammatory cytokine network that inhibits T cell function. This cross-integration strategy of geriatric science and oncology can not only reduce treatment-related toxicity but also enhance the efficiency of tumor clearance. With the advancement of technology and the popularization of aging clock detection, its application scope is gradually extending to the reconstruction of public health strategies. Preventive screening based on biological age assessment can accurately identify high-risk groups that need intervention (such as through lifestyle adjustments like optimizing nutrition, increasing exercise or stress management), thereby delaying age-related functional decline. At the policy level, if the aging clock data is incorporated into the standardized assessment system of the medical system, it may promote the reform of the insurance mechanism and the optimization of the allocation model of medical resources for high-risk groups.

In conclusion, the discovery of the aging clock is driving a paradigm shift in the field of cancer biology—the core significance of which lies in the core role of microenvironmental aging in tumorigenesis, immune regulation and treatment response. By organically integrating cancer-targeted intervention strategies with the macro goals of aging research, the medical community is expected to construct a more refined and multi-dimensional patient care system. This collaborative strategy not only targets tumor entities but also focuses on the age-related pathological processes that continuously drive cancer progression. If the current challenges at the levels of scientific mechanisms, technical implementation and ethical norms can be overcome, it will open up a new path for the cross-integration of aging regulation and cancer treatment, ultimately achieving a significant extension of healthy lifespan and a dual improvement in patient survival rates.

Authors’ contributions

Y, L and Q, as joint senior authors, completed the main manuscript text and prepared Figures. W, Z, Y, and Z contributed to data collection and completed Tables. S, Z made suggestions for the design and improvement of the charts. N polished and revised the entire article. J, H, Z participated in the review and suggested revisions. All authors contributed to the writing and review of the manuscript, and approved the final manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

N/A.

Consent for publication

The corresponding author has received consent for publication.

Conflict of interest

The authors declare no potential conflicts of interest.

Competing interests

The authors declare no competing interests.