Aging Fibroblasts Fuel Ovarian Cancer Growth Through a Single Secreted Protein

Deep inside every solid tumor lies a community of support cells that can be as treacherous as the cancer cells themselves. Among these, cancer-associated fibroblasts have long been known to remodel tissue, suppress immune responses, and feed tumor growth. Now a study published in Genome Medicine has identified a surprising twist in high-grade serous ovarian cancer, the most lethal form of ovarian malignancy: a subset of these fibroblasts enters a state of cellular senescence, and rather than becoming inert, they secrete a matrix protein called Tenascin-C that actively drives tumor progression. The findings, from a team at Southern Medical University in Guangzhou, China, trace the entire signaling chain from the aging fibroblast to the tumor cell membrane and back again, and they point to a specific molecular target that could be exploited therapeutically.

Cellular senescence is a state in which cells stop dividing permanently, usually in response to stress such as DNA damage, telomere shortening, or oxidative injury. Senescent cells are not simply dormant; they often develop what researchers call the senescence-associated secretory phenotype, a program in which they release a cocktail of inflammatory cytokines, growth factors, and enzymes that degrade the extracellular matrix. In many cancers, this secretory output is thought to create a pro-tumorigenic niche. Yet the precise contribution of senescent fibroblasts in ovarian cancer, and the identity of the molecules they use to communicate with tumor cells, had remained poorly defined. The new work set out to close that gap by combining large-scale genomic data with laboratory experiments and clinical samples.

The researchers began by mining publicly available single-cell RNA sequencing datasets, which measure gene expression in thousands of individual cells, alongside spatial transcriptomic data that preserve information about where cells sit within the tumor architecture. They also analyzed clinical specimens obtained from patients at Nanfang Hospital, with ethics approval and informed consent. By applying gene signatures associated with senescence to the fibroblast populations identified in these data, the team showed that cancer-associated fibroblasts in high-grade serous ovarian cancer carry a markedly elevated senescence program compared with other stromal and immune cell types in the same tumors. Spatial analysis placed these senescent fibroblasts in close proximity to malignant epithelial cells, hinting at a functional dialogue between the two compartments.

To translate that observation into something clinically useful, the team built a prognostic model based on the transcriptomic signature of the senescent fibroblasts. They used least absolute shrinkage and selection operator regression, a machine-learning technique that selects a compact set of predictive genes from a much larger candidate list, to derive a risk score for individual patients. When applied to cohort data, a high score on this senescent-fibroblast signature predicted significantly worse overall survival. The model therefore offers a way to stratify patients at diagnosis, potentially identifying those whose tumors are being nurtured by an aged stromal microenvironment and who might benefit from more aggressive or targeted treatment strategies.

The next question was mechanistic: which molecule, or molecules, do senescent fibroblasts use to promote tumor growth? Cell-cell communication analysis, a computational approach that matches ligands expressed by one cell type with receptors expressed by another, flagged Tenascin-C as a leading candidate. Tenascin-C is a large glycoprotein of the extracellular matrix that is abundant in embryonic tissues, largely absent from healthy adult tissue, and re-expressed in wounds and tumors. Among the genes in the prognostic signature, Tenascin-C stood out as a secreted effector produced by senescent fibroblasts and capable of signaling directly to tumor cells, making it an attractive linchpin for the observed pro-tumor effect.

Functional experiments in cell culture and in animal models confirmed the suspicion. When tumor cells were exposed to conditioned medium from senescent cancer-associated fibroblasts, they became more invasive and more migratory, hallmarks of malignant progression. Depleting or blocking Tenascin-C blunted these effects, demonstrating that the protein is a necessary mediator rather than a bystander. In vivo, the presence of senescent fibroblasts accelerated tumor growth in mouse models, and interference with the Tenascin-C axis reduced that acceleration. Together, these assays moved the finding from correlation to causation: the aged stromal cells promote ovarian cancer progression, and they do so substantially through Tenascin-C.

The study also resolved the signaling logic on both sides of the communication. On the receiving end, Tenascin-C engages syndecan-4, a transmembrane proteoglycan on the surface of tumor cells, which triggers the ERK mitogen-activated protein kinase pathway. ERK activation, in turn, drives epithelial-mesenchymal transition, a developmental program that cancer cells co-opt to lose their adhesive, epithelial character and acquire the motile, invasive traits of mesenchymal cells. Epithelial-mesenchymal transition is a well-established engine of metastasis and treatment resistance, so its induction explains how Tenascin-C exposure translates into poorer clinical outcomes. On the sending end, the researchers found that the nuclear factor kappa B pathway, a central inflammatory signaling cascade, is activated in the senescent fibroblasts and transcriptionally upregulates Tenascin-C expression, linking the senescence program itself to the production of the tumor-promoting factor.

To test how general the phenomenon might be, the team performed a pan-cancer analysis, examining the senescent-fibroblast risk score and Tenascin-C expression across multiple tumor types. The pattern held: high expression of Tenascin-C and a high senescent-fibroblast risk score were associated with worse overall survival across a range of solid tumors, not just ovarian cancer. This breadth suggests that the senescent stromal niche is not a peculiarity of high-grade serous ovarian cancer but a recurring feature of the tumor microenvironment, and that Tenascin-C may represent a broadly relevant node at which aged stroma communicates malignancy to neighboring epithelial cells.

The therapeutic implications are significant. High-grade serous ovarian cancer is typically diagnosed at an advanced stage and, despite initial responses to surgery and platinum-based chemotherapy, frequently recurs and eventually becomes resistant to treatment. Agents that target the tumor microenvironment rather than the tumor cells alone have shown promise in other contexts, and the identification of a single, druggable secreted mediator offers a concrete strategy. Blocking Tenascin-C, interrupting its binding to syndecan-4, or dampening the NF-kappaB-driven secretory program in senescent fibroblasts could, in principle, starve tumors of a key progression signal. Approaches that eliminate senescent cells altogether, so-called senolytics, represent another avenue suggested by this work, though the authors’ data point most directly at the Tenascin-C pathway itself.

As with any study, caveats apply. The prognostic model was developed and evaluated on existing datasets and clinical specimens, and prospective validation in independent patient cohorts will be needed before it can inform clinical decision-making. The mechanistic experiments, while thorough, were conducted in cell culture and animal models, and the complexity of the human tumor microenvironment, with its immune, vascular, and matrix components, may modulate the Tenascin-C pathway in ways not fully captured in the laboratory. Nevertheless, the study delivers an unusually complete chain of evidence, from single-cell atlas to survival statistics to signaling biochemistry, and it reframes cellular senescence in cancer not merely as a marker of an aged microenvironment but as an active, targetable driver of tumor progression. For patients with high-grade serous ovarian cancer, whose options narrow sharply at recurrence, a new molecular handle on the stromal side of the disease is a development worth watching closely.

Subject of Research: Senescent cancer-associated fibroblasts secreting Tenascin-C to promote tumor progression in high-grade serous ovarian cancer

Article Title: Senescent cancer-associated fibroblasts secrete Tenascin-C to promote tumor progression in high-grade serous ovarian cancer

Article References: Li, R., Chen, Y., Huang, X., Cai, J., Yang, Y., Fan, H., Xiang, Y., Meng, F., Huang, L., Li, Y., & Ning, Y. (2026). Senescent cancer-associated fibroblasts secrete Tenascin-C to promote tumor progression in high-grade serous ovarian cancer. Genome Medicine. https://doi.org/10.1186/s13073-026-01795-8

Image Credits: AI Generated

DOI: 10.1186/s13073-026-01795-8

Keywords: ovarian cancer, cancer-associated fibroblasts, cellular senescence, Tenascin-C, tumor microenvironment, epithelial-mesenchymal transition, NF-kappaB, ERK signaling, single-cell RNA sequencing, prognostic model, high-grade serous ovarian cancer, senescence-associated secretory phenotype