Single-cell RNA sequencing of circulating tumour cells in colorectal cancer
摘要
Circulating tumour cells (CTCs) are key mediators of metastasis and exhibit marked phenotypic plasticity driven by epithelial-to-mesenchymal transition (EMT). Traditional marker-based CTC isolation approaches rely on epithelial marker expression, which may fail to capture mesenchymal and hybrid CTC subpopulations. Hybrid CTCs remain poorly characterised in colorectal cancer (CRC). This study explored transcriptionally defined CTC subpopulations in CRC to provide insight into CTC heterogeneity.
MethodsSingle-cell RNA sequencing (scRNA-seq) was performed on peripheral blood mononuclear cell (PBMC) fractions from four treatment-naive CRC patients (AJCC stages I–IV). Integrated analysis with healthy PBMC controls enabled immune cell exclusion and cell-type annotation. CTCs were identified using epithelial and mesenchymal transcriptional scores together with CD45 negativity. Differential expression, pathway enrichment, and pseudotime analyses were used to characterise epithelial, mesenchymal, and hybrid CTC states.
ResultsSubpopulations of epithelial, mesenchymal, and hybrid cells were identified in one CRC patient. Hybrid CTCs exhibited distinct transcriptional features and enrichment of pathways related to RNA metabolism, protein trafficking, mitochondrial energy production, DNA repair, and cytoskeletal organisation. Trajectory inference suggested a continuous EMT spectrum, with hybrid CTCs occupying intermediate pseudotime states characterised by progressive loss of epithelial markers and acquisition of mesenchymal-associated features.
ConclusionThis study explored CTC heterogeneity in CRC using single-cell transcriptomics and identified epithelial, hybrid, and mesenchymal CTC states within the analysed sample. Hybrid CTCs exhibited distinct transcriptional features, providing preliminary insight into the transcriptional diversity of CRC CTCs. Further studies in larger cohorts are required to validate these findings and determine their clinical relevance.