Identification of epithelial, mesenchymal, and platelet-associated circulating tumour cells with translational implications in oral squamous cell carcinoma

Abstract

Abstract Metastatic dissemination is driven by rare Circulating Tumour Cells (CTCs) that exhibit pronounced phenotypic plasticity and dynamic interactions with different components in the bloodstream. In Oral Squamous Cell Carcinoma (OSCC), limited resolution of epithelial, mesenchymal, and platelet-associated CTC-states has constrained biological understanding of early tumour spread and hindered translational interpretation. In this study, we identify, count and characterize distinct epithelial, mesenchymal, and platelet-associated CTC-states in a cohort of 33 patients using a strategic workflow involving hematopoietic cell depletion with multiparametric flow-cytometric analysis. The FACS (Fluorescence-assisted cell sorting)-strategy incorporated epithelial tumour-associated markers (EpCAM, EGFR, and Cytokeratin), mesenchymal markers (VIM and N-Cadherin), and platelet/leukocyte markers (CD41 and CD45) to resolve biologically distinct CTC-subpopulations. Analytical robustness was established through spike-in experiments using OSCC cell line i.e., Cal27, demonstrating linear detection across clinically relevant ranges(1-100cells/mL of blood, Spearman’s r  = 0.8738, p  < 0.0001) and sensitive recovery of rare tumour cells with a Limit of Detection of 1 cell/mL, Limit of Quantification of 10 cells/mL, and Limit of Blank of 1.779 cells/mL. Molecular validation of sorted populations using whole transcriptome amplification-quantitative PCR confirmed tumour-associated marker expression and absence of hematopoietic contamination. Application of this approach in OSCC patients revealed detectable CTCs in 70% of the cases (ranging from 14.20 to 340.5 CTCs-related events/mL). Importantly, CTCs were identified across discrete phenotypic states, including platelet-associated CTC-clusters that were not captured by epithelial marker-restricted strategies. Quantitative comparison of CTC subtype proportions revealed majority (nearly 83%) of CTCs present as platelet clusters, followed by Epithelial-Single CTCs (21.21%). In comparison, CTCs with only mesenchymal phenotype constitute only 3.22% of the total CTC-population. Receiver operating characteristic analysis supported the discriminatory capacity of CTC enumeration and enabled the definition of a biologically relevant threshold of > 19.40 events mL for CTC positivity. Taken together, our findings outline a practical workflow for identifying and quantifying different CTC-states in OSCC, with platelet-associated CTCs emerging as the predominant population detected. These findings support the importance of incorporating phenotypic heterogeneity and platelet-associated tumour cell interactions into future liquid biopsy-based translational pathology research and clinical biomarker development in OSCC.