The distinction matters because epigenetic age is increasingly measured with DNA methylation-based models known as epigenetic clocks. These clocks estimate biological age from molecular patterns rather than elapsed time, and they have revealed that individuals and tissues of the same chronological age can age at different rates. But the review stresses that neither biological nor epigenetic age has a universally accepted measure applicable to all tissues and disease contexts. Histone modifications and chromatin remodeling also change with age, yet they are not interchangeable with conventional methylation clocks. That context dependence, the authors argue, is both a limitation and an opportunity: it complicates simple measurement but offers a window into how genetic and environmental factors sculpt human aging at the level of individual cells.

Tumors add a second layer of complexity. Cancer cells routinely bypass normal limits on replication, and their epigenetic trajectories can differ substantially from those of non-malignant neighbors. A tumor sample may appear epigenetically older or younger depending on which clock is applied, which tissue trained the model, and the underlying biology of the malignancy. Clocks calibrated on healthy tissue therefore cannot be read automatically as measures of functional aging in malignant cells. Bulk-tissue measurements compound the problem by mixing methylation profiles from cancer cells, infiltrating immune cells, fibroblasts and endothelial cells into a single average; differences in tumor purity or cellular composition can shift the estimated epigenetic age without any single population having aged faster. The authors also caution that accessibility or transcriptional signatures associated with aging have not yet been validated as epigenetic clocks, and that senescence itself should be assessed with multiple complementary features, such as persistent growth arrest, DNA damage responses and secretory changes, rather than any single marker.

Among immune cells, CD8+ T cells are the most extensively studied example of age-associated epigenetic remodeling. Aging CD8+ T cells progressively lose proliferative reserve and shift from self-renewing progenitor states toward dysfunctional, terminally exhausted states under chronic antigen exposure. The exhaustion spectrum runs from progenitor exhausted cells, which express TCF1 and SLAMF6 and retain the capacity to self-renew and respond to immune checkpoint therapy, through intermediate stages, to terminally exhausted cells defined by loss of TCF1, persistent expression of inhibitory receptors such as PD-1, TIM-3 and LAG-3, and rewritten TOX- and EOMES-associated transcriptional profiles. Human immune-aging studies have identified altered chromatin accessibility in CD8+ T cells, providing a direct link between aging and epigenetic remodeling. In chronic viral infection models, exhausted T cells retain a distinct epigenetic state even when PD-1 blockade transiently reinvigorates their function, a finding that raises the possibility that age-associated remodeling further restricts T-cell plasticity in tumors, though the authors note that direct testing is still required.

Regulatory T cells, or Tregs, present a subtler picture. Age-associated DNA methylation changes have been documented in non-malignant lung Tregs, and separate cancer studies have shown EZH2/H3K27me3-dependent regulation of tumor-infiltrating Tregs, but no cited study establishes an age-dependent epigenetic mechanism within intratumoral Tregs. In mice, aging impairs the capacity of Tregs to support central nervous system remyelination, and aspects of this defect can be reversed in a young environment; in humans, reduced tissue-repairing capacity has been observed in aged Tregs after respiratory infection. The review insists these findings should not be read as evidence that aged tumor-infiltrating Tregs uniformly lose suppressive activity. In the TME, Treg abundance, suppressive function and tissue-repair capacity must be evaluated separately, and a Treg-CD8+ T-cell imbalance cannot be inferred from cell counts alone, since suppressive activity, antigen specificity and spatial proximity are equally critical.

Myeloid cells, natural killer cells and the stromal compartment complete the picture. Age-associated DNA methylation differences have been measured in purified circulating human NK cells and in circulating monocytes and granulocytes, but their functional consequences and their applicability to individual tumor-associated subsets remain uncertain. In aged tumor-bearing mice, impaired cDC1-mediated T-cell priming has been demonstrated and rescued therapeutically, yet the underlying myeloid epigenetic mechanism has not been established. On the stromal side, age-related chromatin-accessibility changes have been described in fibroblasts from healthy mammary tissue, and separate melanoma studies show age-dependent extracellular matrix remodeling that alters immune-cell motility. Senescent fibroblasts release the senescence-associated secretory phenotype, a cocktail of cytokines, chemokines, growth factors and matrix-remodeling enzymes that can recruit suppressive myeloid cells in experimental models, though senescent cells also produce damage-associated molecular patterns that may promote antitumor effects. Endothelial cells show age-associated chromatin changes and increased inflammatory responsiveness in senescence models, including increased VCAM1-associated H3K4me3, but the review emphasizes that none of these studies establishes endothelial epigenetic aging as the cause of vascular dysfunction in aged tumors.

These compartments do not age in isolation. The review proposes a conceptual framework of immune-stromal circuits in which senescent fibroblasts may recruit or retain suppressive myeloid cells, myeloid-derived mediators reinforce stromal activation and impair T-cell function, and Tregs further constrain local CD8+ T-cell activity. The authors are explicit that this is a framework for investigation rather than a universally established sequence of events in aged tumors. Transforming growth factor-beta (TGF-beta) offers the clearest supporting example: in metastatic urothelial cancer, a fibroblast TGF-beta response signature was linked to resistance against PD-L1 blockade in immune-excluded tumors, and combined pathway inhibition improved T-cell penetration and antitumor activity in a mouse model. That evidence supports targeting immune-stromal interactions, but it does not prove that TGF-beta-mediated exclusion is intrinsically age-dependent. Environmental factors add further layers: hypoxia can inhibit oxygen-dependent enzymes such as TET, an effect associated with tumor DNA hypermethylation, while metabolic pathways govern the availability of substrates and cofactors for chromatin-modifying enzymes, including acetyl-CoA, S-adenosylmethionine and alpha-ketoglutarate. Whether these stresses accelerate aging-specific epigenetic programs or merely induce generic stress responses remains unresolved, and the authors note that such interactions are difficult to validate experimentally in humans.

The therapeutic implications are substantial but hedged. Epigenetic therapies, including DNA methyltransferase inhibitors, histone deacetylase inhibitors and bromodomain and extraterminal protein inhibitors, are under investigation in cancer and may alter tumor immunogenicity and immune-cell programs. Yet the review is categorical that an antitumor effect of an epigenetic drug does not by itself demonstrate reversal of epigenetic aging; genuine rejuvenation would require restoration of relevant cellular functions accompanied by appropriately validated molecular changes. Senolysis, the selective elimination of senescent cells, and senomorphic interventions that modify senescence-associated phenotypes without killing the cells face a similar dilemma, because senescence can restrain malignant proliferation even as persistently senescent cells degrade the microenvironment. CTLA-4-directed antibodies can influence Treg biology, but human tumor analyses show that intratumoral Treg depletion is not consistently observed, and systemic suppression of regulatory populations risks inflammatory toxicity. Chronological age, meanwhile, has been associated with differences in certain immune checkpoint therapy biomarkers, but these relationships depend on tumor type and biological context, and older age should not be treated as a uniform indicator of immunotherapy resistance.

Methodologically, the authors call for studies that treat age as a continuous variable, document tumor type, stage, anatomical site and treatment history, and use paired tumor and non-tumor sampling with longitudinal pre- and post-treatment analyses. They advocate integrating single-cell RNA sequencing, single-cell ATAC sequencing, methylation profiling, CITE-seq and spatial methods, while stressing that these approaches measure different features and should not be treated as equivalent evidence for epigenetic aging. Association-based signatures require functional validation through perturbation and rescue experiments comparing young and aged cells or hosts. The bottom line of the review is a disciplined form of optimism: tumors may indeed harbor epigenetically altered immune-stromal niches whose functional states are not captured by patient age alone, and precision immunotherapy may ultimately benefit from targeting those dysfunctions while preserving protective immune regulation. But the field must first identify which cell-specific epigenetic changes causally sustain these niches, and determine whether modifying them genuinely improves treatment response, before an epigenetic-age estimate can be trusted as either a biomarker or a therapeutic target.

Subject of Research: Epigenetic aging of the tumor microenvironment and its implications for cancer immunotherapy

Article Title: Epigenetic aging in the tumor microenvironment: mechanisms and implications for immunotherapy

Article References: Takezawa, F., & Sato, Y. (2026). Epigenetic aging in the tumor microenvironment: mechanisms and implications for immunotherapy. Frontiers in Aging, 7, 2008892. https://doi.org/10.3389/fragi.2026.2008892

Image Credits: AI Generated

DOI: 10.3389/fragi.2026.2008892

Keywords: epigenetic aging, tumor microenvironment, immunotherapy, DNA methylation, T-cell exhaustion, senescence, cancer-associated fibroblasts, Tregs, myeloid cells, epigenetic clocks, checkpoint inhibitors, chromatin