Extracellular vesicles, or EVs, are small lipid-enclosed particles released by virtually every cell type in the body, including tumor cells. For decades they were dismissed as cellular debris, but they are now recognized as a fundamental communication system: EVs ferry proteins, lipids and nucleic acids between cells, influencing processes such as immune regulation, angiogenesis and metastatic preparation. Because the molecular cargo of an EV reflects the cell that released it, tumor-derived vesicles are increasingly viewed as a liquid biopsy window into the biology of a cancer without the need for invasive tissue sampling. Cytokines, the small proteins that immune and stromal cells use to signal to one another, are known to shape tumor development and progression, and the question driving this study was whether cytokines bound to or packaged within EVs form a measurable and clinically informative signature in the bloodstream.

To answer it, the researchers isolated EVs from plasma using ultracentrifugation, a technique that spins samples at extremely high speeds to pellet vesicles out of solution. They applied a rigorous characterization pipeline to confirm that the particles they recovered were genuine EVs: transmission electron microscopy provided direct visualization of vesicle morphology, nanoparticle tracking analysis quantified particle size and concentration, flow cytometry probed surface markers, and western blotting verified the presence of canonical EV-associated proteins. The same isolation approach was applied to breast cell lines grown in the laboratory, giving the team a controlled comparison against vesicles produced by healthy breast tissue cells.

The clinical core of the study involved 46 patients with ER+ breast cancer enrolled in the Neoletexe trial, alongside 13 healthy individuals as controls. EV-associated cytokines were profiled using a Luminex multiplex assay, a bead-based platform capable of measuring dozens of proteins simultaneously from a small sample volume. Crucially, the researchers sampled patients at three timepoints: before treatment began (T0), and at two points during neoadjuvant endocrine therapy (T1 and T2). This longitudinal design meant that each patient could serve partly as her own control, revealing how the vesicle-bound cytokine landscape shifted as aromatase inhibitors stripped estrogen from the equation and the tumor responded, or failed to respond, to the hormonal blockade.

The headline finding was that three EV-associated cytokines were significantly elevated in ER+ breast cancer patients compared with healthy individuals: interleukin 10 (IL-10), interleukin 12 (IL-12) and interferon gamma inducible protein 10, better known as IP-10 or CXCL10. Each of these molecules carries immunological weight. IL-10 is generally an anti-inflammatory cytokine that can dampen immune attacks on tumors, IL-12 is a potent activator of T cells and natural killer cells, and IP-10 is a chemokine recruited into action by interferon gamma that guides immune cells to sites of inflammation and, in many cancers, is intertwined with the tumor microenvironment’s immune landscape. Their enrichment on circulating vesicles in patients suggests that EV-associated cytokines capture disease-relevant immune signaling that differs measurably from the healthy state.

The IP-10 result gained independent support from tumor tissue analysis. Using gene expression data, the team found that CXCL10, the gene encoding IP-10, was overexpressed in breast cancer tissue relative to both normal and benign breast tissue, a pattern corroborated through the GEPIA analysis platform drawing on The Cancer Genome Atlas. This convergence between the vesicle protein measurements in blood and the transcriptional activity inside tumors strengthens the argument that EV-associated IP-10 genuinely reflects processes occurring at the disease site rather than incidental background variation. It is exactly this kind of cross-validation that candidate biomarkers need before they can move from exploratory science toward clinical utility.

Perhaps the most technically interesting result concerned how the vesicle-bound cytokines behaved once endocrine therapy started. Levels of EV-associated IL-10 and IL-12 decreased significantly during NET, indicating that the treatment does not merely shrink or slow the tumor but measurably reshapes the immune signaling carried by circulating vesicles. Notably, the treatment-associated changes were more pronounced in the EV-associated cytokines than in the corresponding free-circulating cytokines measured in the same plasma samples, and more pronounced than changes in cytokine gene expression measured in tumor tissue from a subset of the same patients. This suggests that vesicle-packaged cytokines act as a kind of amplifier or concentrator of systemic immune signals, potentially offering greater analytical sensitivity than conventional plasma cytokine assays, which have long been hampered by low concentrations, short half-lives and interference from non-specific binding.

The study was not without its sobering results. When the researchers asked whether any of the measured cytokines predicted treatment response, as assessed by the change in Ki-67, the standard proliferation marker used to gauge how actively tumor cells are dividing under therapy, only eotaxin, a chemokine best known for recruiting eosinophils, showed a significant association. None of the three disease-elevated cytokines tracked treatment response in this way. This dissociation between disease detection and response prediction is an important nuance: a marker can be excellent at distinguishing patients from healthy individuals yet still fail to forecast how an individual tumor will behave under therapy. The authors are careful to frame their findings as evidence that EV-associated cytokines reflect disease- and treatment-related changes, while emphasizing that further studies are required to determine their potential clinical relevance as biomarkers.

Methodologically, the study demonstrates the value of triangulating across molecular compartments. By measuring the same cytokines in three formats, vesicle-bound in plasma, free-circulating in plasma, and as gene expression in tumor tissue, the researchers could show that the vesicle fraction carries complementary rather than redundant information. This matters because the field of liquid biopsy has often been dominated by circulating tumor DNA and extracellular RNA, while the protein cargo of EVs, and cytokines in particular, has remained comparatively underexplored. The finding that EV-associated signals respond more dynamically to endocrine therapy than either free cytokines or tumor transcript levels hints that vesicles may stabilize and enrich labile signaling proteins, protecting them from degradation and protease activity in the bloodstream.

The broader implications reach into how ER-positive breast cancer is monitored. Neoadjuvant endocrine therapy is typically given for months before surgery, and clinicians currently rely on imaging and serial biopsies to judge whether the regimen is working. A blood test that captures treatment-induced shifts in tumor-associated immune signaling could, if validated, allow earlier and less invasive assessment. The Norwegian team’s work, funded in part by Akershus University Hospital and the Norwegian Breast Cancer Research Network and conducted under ethical approval from the Regional Ethics Committee of Southeast Norway, provides a carefully characterized early step in that direction. With 46 patients and 13 controls, the cohort is modest, and ultracentrifugation-based EV isolation, while thorough, is labor-intensive and not yet standardized across laboratories. Larger, multi-center studies with standardized vesicle isolation and longitudinal follow-up will be needed to establish whether IL-10, IL-12 and IP-10 on circulating vesicles can graduate from candidate markers to clinically actionable tools. For now, the study adds a compelling piece to the growing evidence that the smallest messengers in our blood may carry some of the biggest clues about cancer.

Subject of Research: Extracellular vesicle-associated cytokines as circulating biomarkers in ER-positive breast cancer and neoadjuvant endocrine therapy

Article Title: Analysis of circulatory extracellular vesicle–associated cytokines as candidate disease-associated markers in ER-positive breast cancer

Article References: Huynh, H. M., Nawaz, M. S., Jabeen, S., Bjørnetrø, T., Geisler, J., Kristensen, V., & Tahiri, A. (2026). Analysis of circulatory extracellular vesicle–associated cytokines as candidate disease-associated markers in ER-positive breast cancer. BMC Cancer. https://doi.org/10.1186/s12885-026-17052-5

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17052-5

Keywords: extracellular vesicles, cytokines, ER-positive breast cancer, liquid biopsy, biomarkers, neoadjuvant endocrine therapy, IL-10, IL-12, IP-10, CXCL10, aromatase inhibitors, Neoletexe trial