Transcriptomic Analysis of EpCAM/CD45 Dual-Positive Ovarian Cancer Stem-Like Cells Reveals Inflammatory Signatures and Association with SYK
DOI:
https://doi.org/10.18502/ajmb.v18i4.22927Keywords:
Gene expression profiling, Interferons, Neoplastic stem cells, Ovarian neoplasms, Protein interaction mapping, Transcription factorsAbstract
Background: Dual‑positive EpCAM⁺/CD45⁺ ovarian cancer cells in malignant ascites are proposed as ovarian cancer stem-like cells, yet their cellular origin and trans-criptomic programs remain debated.
Methods: In this work expression data from ascites‑derived EpCAM⁺/CD45⁺ dual‑ positive ovarian cancer stem‑like cells (DpOCSCs; GSE75036) and reference non‑ dual‑positive ovarian cancer stem cells (OCSCs; GSE90125, GSE216627) was in-tegrated. The workflow included cross‑platform harmonization, differential expression analysis, pathway enrichment, protein–protein interaction mapping, and evaluation in independent cohorts (TCGA, CPTAC, survival databases).
Results: DpOCSCs showed 347 up‑regulated and 23 down‑regulated genes (|log₂FC|>1, FDR<0.1). Enrichment analyses revealed overrepresentation of interferon‑α/γ re-sponses, TNF, NF‑κB, IL‑6, JAK‑STAT3, and CXCR4 signaling, and pathways linked to natural killer cell‑mediated cytotoxicity, indicating a distinct inflammatory signature. Network analysis identified 17 up‑regulated hub genes; 10 had higher mRNA in ovarian tumors versus normal tissues (TCGA), and 3 (SYK, CD74, MX1) showed elevated pro-tein in CPTAC. In bulk ovarian cancer cohorts, these hubs were exploratorily associated with copy‑number amplifications, immune‑checkpoint expression, immunosuppressive infiltration, and clinicopathological features, including lymphatic/venous invasion and therapy outcomes. Survival and ROC analyses indicated context‑dependent associations of SYK, CD74, and MX1 with overall and progression‑free survival; SYK achieved an AUC ≈0.7 for predicting 6‑month relapse, particularly in paclitaxel‑treated patients. Transcription‑factor enrichment implicated Wnt/β‑catenin, Hippo, and NF‑κB as upstream regulators.
Conclusion: These in silico findings are hypothesis‑generating and nominate a SYK‑associated inflammatory network for functional assessment in DpOCSCs and as a potential source of exploratory biomarkers for recurrent ovarian cancer.
