The application of circulating tumor cells in endometrial cancer: A narrative review

Endometrial cancer (EC) is increasingly common worldwide, and recurrence remains difficult to anticipate using clinicopathologic factors, static tumor molecular classification, imaging, or conventional serum markers alone. This narrative review evaluates circulating tumor cells (CTCs) as a cell-based liquid biopsy for EC, with emphasis on their biological rationale, detection technologies, and potential roles in risk assessment, surveillance, and minimal residual disease monitoring. CTCs preserve intact cellular morphology and phenotypic information that are not captured by circulating tumor DNA (ctDNA), including protein expression, epithelial-mesenchymal transition (EMT)-related features, cell clustering, and intercellular heterogeneity. Current EC studies remain small and methodologically diverse, but collectively suggest that CTC detection is feasible in selected early-stage, high-risk, advanced, and recurrent settings. Detection performance varies substantially according to platform design: CellSearch ® provides standardized EpCAM-dependent enumeration but may underestimate EpCAM-low or EMT-shifted CTCs, whereas size-based, microfluidic, multimarker, and molecular approaches may improve sensitivity at the cost of requiring further specificity validation. With the continued development of CTC profiling, single-cell biopsy, and single-cell sequencing technologies, CTC analysis may move beyond enumeration toward phenotype- and subtype-informed characterization. This review highlights the need to clarify how The Cancer Genome Atlas molecular subtypes relate to CTC count, epithelial/mesenchymal phenotype, cluster status, and protein-marker expression. It also emphasizes that benign gynecologic conditions with invasive or inflammatory features, including endometriosis, should be incorporated into control cohorts to better define false-positive CTC or CTC-like events. Although CTC testing in EC remains in the clinical research stage, emerging automation, artificial intelligence-assisted image interpretation, and scalable microfluidic workflows may improve throughput, reproducibility, and cost-effectiveness. Clinical implementation will require analytical standardization, multicenter prospective validation, adequate benign and healthy controls, predefined clinical endpoints, direct comparison with ctDNA and serum biomarkers, and formal cost-effectiveness assessment.