Circulating tumor cells reveal immune escape tricks that may guide immunotherapy

Cancer cells that slip into the bloodstream face a gauntlet of immune attack before they can ever seed a distant metastasis, and a new review argues that how they survive that gauntlet could one day help doctors navigate immunotherapy. The paper, published open access in Biomarker Research on 9 October 2026, organizes the immune phenotype of circulating tumor cells (CTCs) across every stage of dissemination, from the moment a tumor cell sheds into the blood to the moment it attempts to colonize a new organ. Its central message is deliberately cautious: CTC immune profiling is a promising research strategy, but heterogeneous assays, small retrospective cohorts and limited prospective validation mean it cannot yet be used to select treatments.

The stakes are high. Metastasis accounts for roughly 90 percent of cancer deaths globally, and CTCs are the cells that make it possible, having detached from primary or metastatic tumors and entered the circulation with the potential to colonize distant organs. Since CTCs were first described by Thomas Ashworth in 1869, they have evolved from a pathological curiosity into clinically validated biomarkers: enumeration with the FDA-cleared CellSearch system is established as an independent prognostic indicator in metastatic breast, prostate and colorectal cancers. What the new review proposes is a shift in emphasis, from simply counting these rare cells to characterizing the immune molecules they carry.

The urgency comes from the limitations of current immunotherapy biomarkers. Immune checkpoint inhibitors that block interactions such as PD-1/PD-L1 and CTLA-4/B7 can produce durable benefit in non-small cell lung cancer, melanoma and urothelial carcinoma, but response varies widely across cancer types, regimens and patient populations, and reliable predictive biomarkers remain a clinical need. Tissue PD-L1 immunohistochemistry is affected by spatial and temporal heterogeneity, biopsy sampling and differences in antibody clones and scoring systems. Tumor mutational burden and microsatellite instability describe other aspects of tumor biology but say nothing directly about the immune phenotype of intact circulating cells. Serial blood sampling, the review argues, could in principle capture cell-level immune changes during treatment that a one-time tissue biopsy cannot.

What makes CTCs biologically interesting is that they are a selected subpopulation. They have survived immune pressure in the tumor microenvironment and then again in the bloodstream, where the estimated circulating half-life of a CTC is only 1 to 2.4 hours, reflecting intense selective pressure. Their phenotype therefore need not mirror that of the primary tumor. Candidate readouts include checkpoint ligands such as PD-L1, CD47, B7-H3 and HLA-G, antigen-presentation molecules, and innate immune axes such as HLA-E interacting with the CD94/NKG2A receptor on natural killer cells. Single-cell comparisons of primary tumors and CTCs reveal differences in antigen-presentation and interferon-response programs, although such comparisons cannot establish when each phenotype was acquired.

During circulatory transit, CTCs deploy both cell-intrinsic and cell-extrinsic defenses. On the intrinsic side, PD-L1 is the best-studied checkpoint ligand on CTCs and has been reported across cancer types, sometimes differing from matched tissue PD-L1 status. Loss or remodeling of MHC class I can reduce recognition by cytotoxic T cells, but it can also increase susceptibility to NK-cell missing-self recognition, an outcome that depends on the balance of activating and inhibitory signals. On the extrinsic side, platelet cloaking is the best-characterized shield: activated platelets form a physical coat that hides activating ligands from NK-cell receptors, transfer their own MHC class I onto the tumor-cell membrane to create a pseudo-normal surface, and release TGF-β and other soluble mediators that blunt NK-cell effector function. Most of this evidence, the review stresses, comes from experimental models and ex vivo co-culture rather than patient-derived CTCs, and the proportion of platelet-coated CTCs in patients has not been systematically quantified.

Neutrophils provide a second protective interface that the authors carefully split into two distinct arms. Single-cell profiling of CTC-neutrophil clusters in patients with breast cancer and in mouse models showed that neutrophil-associated CTCs carry transcriptional programs favoring cell-cycle progression and that such clusters seed metastases more efficiently than CTCs traveling alone. Separately, neutrophil extracellular traps, or NETs, released during systemic inflammation or after surgery sequester CTCs in the hepatic and pulmonary microvasculature, promote adhesion and extravasation, and can shield tumor cells from contact-dependent cytotoxicity. The distinction matters therapeutically: dissolving NETs and preventing cluster formation predict different effects on immune exposure, and an intervention that removes a NET barrier will not necessarily dissociate a neutrophil-containing cluster.

On the clinical side, PD-L1-positive CTCs have received the most attention. In non-small cell lung cancer, early studies linked PD-L1-positive CTCs with poorer outcomes during PD-1 blockade, though platform and threshold heterogeneity limit their use as stand-alone treatment selectors, and limited longitudinal data suggest that dynamic changes during treatment may be more informative than a single baseline measurement. In locally advanced head and neck squamous cell carcinoma treated with induction chemotherapy and concurrent chemoradiotherapy, PD-L1 overexpression in the EpCAM-enriched CTC fraction at the end of treatment was associated with poorer outcomes, although that study did not evaluate response to PD-1 blockade. Across cancer types, reported positivity and outcome associations differ by disease, treatment setting, enrichment method and scoring definition, and discordance with tissue PD-L1 may reflect biological differences between compartments as well as sampling and analytical variation rather than superior predictive accuracy.

Technology both enables and constrains the field. The CellSearch platform, which uses EpCAM-mediated enrichment followed by cytokeratin, DAPI and CD45 staining, offers the most standardized clinical framework, and its fourth fluorescent channel has been adapted for PD-L1 and other markers, but EpCAM-based capture under-represents mesenchymal or hybrid CTCs and a single extra channel cannot resolve coordinated changes in PD-L1, MHC-I, CD47 and epithelial-mesenchymal state. Multiparametric immunofluorescence, flow cytometry, mass cytometry and single-cell multiomics extend the analysis to coordinated immune phenotypes, while live-cell microfluidic capture enables functional assays such as co-culture with immune effector cells. A persistent statistical problem is rarity: at low CTC concentrations a standard 7.5 milliliter blood draw may contain no tumor cells even when the circulating burden is substantial, which is why in vivo approaches such as photoacoustic flow cytometry and intravascular capture devices, though currently limited to detection and enumeration rather than immune phenotyping, represent a long-term direction toward dynamic surveillance.

The review also maps therapeutic strategies that target CTC immune vulnerabilities, while flagging their immaturity. Antiplatelet approaches have epidemiological and experimental support for reducing metastatic risk but must balance immune exposure of CTCs against bleeding risk. NET-targeting strategies such as DNase-based degradation may expose CTCs to immune attack. CD47-SIRPα blockade is supported by solid-tumor models and non-CTC clinical studies, but CTC CD47 expression alone should not be taken to predict benefit, particularly because CD47 is also expressed on normal cells. Preclinical work linked cardiac glycosides to CTC-cluster disruption, and a recent proof-of-concept trial showed that digoxin can reduce CTC cluster size in metastatic breast cancer, though the authors caution that the next step is to test whether cluster reduction actually changes immune susceptibility and clinical outcome rather than to claim cluster dissociation as established anti-metastatic therapy.

To move the field forward, the authors propose a four-stage validation roadmap: discovery and analytical validation, in which an assay must reproducibly measure the intended immune feature with defined thresholds, denominators and reference materials; clinical validation, which must separate prognostic associations from predictive ones requiring a comparator arm; clinical utility demonstration, in which randomized biomarker-strategy or enrichment designs must show that acting on the CTC readout improves decisions beyond tissue biomarkers, ctDNA, imaging and clinical factors; and finally implementation, with regulatory review, quality systems, reimbursement and turnaround times compatible with treatment decisions. Until those conditions are met, the authors conclude, CTC immune phenotyping should guide research questions rather than individual treatment decisions, and serial measurements should be investigated as potential pharmacodynamic or risk markers whose added value over total CTC burden and established biomarkers remains to be proven prospectively.

Subject of Research: Immune phenotyping of circulating tumor cells as a candidate biomarker for immunotherapy

Article Title: The immune phenotype of circulating tumor cells: from escape mechanisms to immunotherapy navigation

Article References: Zhang, P., Li, M., Li, L., Fu, J., Li, Q., Liu, T., Zeng, R., Zhou, Y., Mao, R., Wang, X., Li, S., Zhang, X., & Yang, Z. (2026). The immune phenotype of circulating tumor cells: from escape mechanisms to immunotherapy navigation. Biomarker Research, 14(1), Article 139. https://doi.org/10.1186/s40364-026-01015-w

Image Credits: AI Generated

DOI: 10.1186/s40364-026-01015-w

Keywords: circulating tumor cells, immune evasion, immunotherapy, PD-L1, liquid biopsy, biomarkers, metastasis, platelets, neutrophil extracellular traps, CD47, checkpoint inhibitors, CellSearch