Migrasomes Shed by Metastasizing Tumor Cells Carry Antigens That Can Prime CD8 T Cells, Mouse Study Finds

• Circulating tumor cells migrating through blood vessels release migrasomes, vesicle-like structures that carry tumor-specific antigens and can activate CD8-positive T cells, according to a study published in Nature Immunology by teams at Tsinghua University and Zhejiang University.

• In mouse models, blocking migrasome formation by suppressing the protein Tspan4 increased metastatic spread, while increasing Tspan4 reduced metastases and prolonged survival. In T-cell-deficient nude mice the effect reversed, indicating dependence on mature T cells.

• Migrasomes reached the spleen and lymph nodes, were taken up mainly by macrophages and dendritic cells, and were cross-presented to CD8-positive T cells. Proteomic analysis found enrichment of cancer-testis antigens including BRDT and MSH4, with BRDT contributing most to anti-metastatic immunity.

• Injecting migrasomes reduced metastatic burden in mice with established metastases, and combining them with a PD-1 inhibitor increased the effect further. Migrasomes from spontaneous breast tumors also reduced lung nodules.

• All efficacy evidence is preclinical. Human data are limited to detecting migrasome-enriched fractions in patient plasma and migrasome production by patient-derived cancer cells. No migrasome vaccine has been tested in a clinical trial, and the source reports no effect sizes, p values, or safety data.

• Study type: Preclinical mechanistic and therapeutic study combining mouse metastasis models with descriptive human biospecimen analysis. This is not a clinical trial and has no registration.

• Central question: Do migrasomes released by migrating tumor cells carry tumor antigens that the immune system can recognize, and can they be used to limit metastasis?

• Experimental systems: Mouse metastasis models with loss- and gain-of-function manipulation of Tspan4; immunodeficient nude mice; a spontaneous breast cancer model; human plasma from breast and lung cancer patients; primary cancer cells from patients with oral squamous cell carcinoma.

• Intervention tested in mice: Migrasome administration alone and in combination with PD-1 blockade.

• Reported by: Yu Li (Tsinghua University) and Jiang Dong (Zhejiang University) and colleagues, Nature Immunology, 2026, as stated in the source.

• Not reported in the source: sample sizes, effect estimates, confidence intervals, p values, adverse events, funding, conflicts of interest, or registry information.

Metastasis as an immunological exposure event, not only an escape

Metastasis is the step that makes most cancers lethal, and the standard explanation is one of concealment: tumor cells that survive the bloodstream are assumed to slip past immune surveillance on their way to distant organs. A study now described in Nature Immunology proposes that the opposite happens as well. While circulating tumor cells migrate through vessels, they leave behind migrasomes, and those discarded structures carry tumor-specific antigens that can be picked up by antigen-presenting cells and used to train CD8-positive T cells against the tumor.

The work comes from the groups of Yu Li at Tsinghua University and Jiang Dong at Zhejiang University. The claim is not that a new drug has been validated in patients. It is that a behavior long read as pure immune evasion also generates a measurable immunological signal, and that this signal can be exploited therapeutically in mice. The distinction matters for how the finding should be read: the mechanism is interesting, the mouse efficacy data are encouraging, and the clinical question remains entirely open.

What migrasomes are, and why Tspan4 became the test

Migrasomes are vesicle-like structures that form on the retraction fibers trailing behind a migrating cell. They are not shed randomly; they are built at defined positions as the cell pulls its rear edge forward, which makes them a plausible way for a motile cell to deposit material along its path. Because migrating tumor cells in the bloodstream are exactly the cells that must move to seed distant metastases, the researchers asked whether migrasomes act as a vehicle for tumor antigens that the immune system can see.

To test that, they manipulated migrasome formation rather than the tumor's antigen repertoire. Tspan4 is described in the study as a key protein for migrasome formation, giving the team a single lever to reduce or increase migrasome production and observe what happens to metastatic spread.

Loss of migrasomes increased metastasis; the effect vanished without T cells

Suppressing Tspan4, and thereby reducing migrasome formation, led to a clear increase in metastatic lesions in the mice. The reverse experiment pointed the same way: overexpressing Tspan4 reduced metastasis and extended survival. On its own, that pair of results would be compatible with a purely cell-biological explanation, in which migrasomes somehow restrain tumor cell movement or colonization independently of immunity.

The nude-mouse experiment rules that out. In immunodeficient nude mice, which lack mature T cells, the relationship reversed: losing Tspan4 reduced metastasis rather than increasing it. The anti-metastatic effect of migrasomes therefore appears to depend on an intact T-cell compartment. This is the strongest mechanistic argument in the study, because the same genetic manipulation produced opposite outcomes depending on whether T cells were present.

How migrasomes reach the immune system: spleen, lymph nodes, and cross-presentation

The study traced where the shed material goes. Migrasomes entered the spleen and lymph nodes, where they were taken up mainly by macrophages and dendritic cells. These antigen-presenting cells then cross-presented the tumor antigens carried by the migrasomes to CD8-positive T cells, the cytotoxic population that kills target cells. In other words, the migrating tumor cell deposits a sample of its own antigenic identity in exactly the tissues where immune responses are organized.

Cancer-testis antigens, with BRDT in the lead

Proteomic analysis of the migrasomes showed enrichment of cancer-testis antigens, including BRDT and MSH4, alongside a range of mutated proteins. Cancer-testis antigens are appealing immunotherapy targets because they are normally restricted to germline tissue but are expressed by some tumors, which gives the immune system a way to discriminate. Among the candidates identified, BRDT contributed most to the anti-metastatic immunity observed in the models, according to the study. That ranking is important for future vaccine design, although the source does not report the magnitude of the contribution or how it was quantified.

Turning the mechanism into treatment: migrasome administration in mice

The researchers then moved from observation to intervention. In mice with already established metastatic disease, injecting migrasomes significantly reduced metastatic burden. Combining migrasome administration with a PD-1 inhibitor strengthened the anti-metastatic effect further, which fits a model in which migrasomes supply the antigenic stimulus and checkpoint blockade removes a brake on the responding T cells.

Two additional results broaden the concept. First, migrasomes derived from the primary tumors of a spontaneous breast cancer model also reduced lung metastatic nodules, suggesting the approach does not depend on an artificial, engineered tumor line. Second, the combination with PD-1 blockade implies that migrasome-based vaccination could be positioned alongside existing immunotherapy rather than as a replacement. These are proof-of-concept findings in animals; the source does not report effect sizes, confidence intervals, or p values for any of them, so the size of the benefit cannot be judged from the available description.

Human samples: signals, not proof

The human component of the study is descriptive. Plasma from patients with breast cancer and lung cancer contained fractions enriched for migrasomes, indicating that the structures can be detected in circulation in people with these tumors. Separately, primary cancer cells taken from patients with oral squamous cell carcinoma were able to generate migrasomes rich in mutated proteins and testis-related proteins. Together these observations suggest the biology is not confined to mice and that patient-derived material can produce antigen-bearing migrasomes.

What these data do not show is that migrasome-based vaccination works in humans, that the plasma signal tracks with disease stage or outcome, or that migrasomes detected in patient blood are immunologically active in a human immune system. Detection is not efficacy, and the sample is described only as small.

What the findings do not yet show

• No clinical data. Every efficacy result described comes from mouse models. No migrasome vaccine has entered a clinical trial, and no human dose, schedule, or response rate has been established.

• No quantitative detail. The source reports directions of effect rather than sizes. Without effect estimates, confidence intervals, or p values, the magnitude and precision of the anti-metastatic benefit cannot be assessed.

• No safety data. The source does not describe adverse events, immune-related toxicity, or tolerability for migrasome administration, alone or with PD-1 blockade. An agent designed to intensify antigen presentation carries the theoretical risk of autoimmunity, which would need systematic evaluation.

• Unresolved specificity. Whether migrasome release differs meaningfully across tumor types, and whether the antigens enriched in migrasomes are reliably tumor-restricted, are open questions.

• Durability and manufacturing unknown. How long any immune memory persists, and how a patient-specific migrasome product would be produced at scale, are not addressed.

Why the idea is biologically interesting

The conceptual shift is that metastasis is not purely a story of invisibility. A tumor cell that migrates leaves a trail of antigen-bearing material in the circulation and in lymphoid organs, and in the models studied that trail can be turned against the tumor. If the mechanism holds in people, the therapeutic logic follows: use a patient's own tumor-derived migrasomes as a personalized vaccine, possibly with checkpoint blockade, to teach the immune system to recognize the cells capable of seeding new lesions.

That remains a hypothesis. The study establishes a mechanism and a proof of concept in animals, and adds limited human descriptive support. Whether migrasome vaccination can reproduce these effects in clinical trials requires substantial further validation, and the field will need quantitative preclinical reporting and formal safety assessment before that question can be answered.

Funding, registration, and conflicts

The source does not report funding sources, sponsor roles, conflict-of-interest disclosures, or registry information. No registration applies, as the work is preclinical and no clinical trial is described.

References

1. Jiang D, He J, Xie R, Shi M, Liu S, Jia H, Yang B, Ruan X, Tao X, Xiang Y, Chen Y, Jiao L, Feng X, Yu L. Metastasis enables immunogenicity through migrasome-mediated antigen release. Nat Immunol. 2026 Oct;27(10):2088-2101. doi: 10.1038/s41590-026-02641-0. Epub 2026 Sep 22. PMID: 42773266; PMCID: PMC13630567.