Determination of high-grade serous ovarian cancer stem cell-based subtypes and prognostic model and identification of highly expressed VSIG4 and STAB1 in macrophages.
Cancer stem cells are associated with tumorigenesis, aggression, and drug resistance. We aimed to identify stem cell-related subtypes and a prognostic tool, and to investigate potential stem cell-related genes contributing to high-grade serous ovarian cancer (HGSOC).
Stem cell pathways were used to determine tumor subtypes and the least absolute shrinkage and selection operator regression was conducted to construct a prognostic risk model, with robustness validation in external datasets. We assessed immune characteristics and therapeutic responses of risk score. Macrophage subpopulations were identified using single cell data, and pseudo-time analysis revealed the changes of macrophages during cell state transition.
HGSOC patients were stratified into stem cell pathway-related clusters (C1, C2) and stem cell-related clusters (GC1, GC2). Patients in C1 and GC1 exhibited better prognosis, increased ImmuneScore, decreased TumorPurity and low immune escape. Patients in C1 were sensitive to gemcitabine while patients in GC1 were sensitive to cisplatin, cyclophosphamide, gemcitabine and niraparib. Risk score was constructed based on 15 genes (IL2RG, STAB1, C2, CD163, FBXO17, VSIG4, CXCL11, CXCL13, GJB1, GPC3, NPY, KRT16, GRIK5, PI3, and RARRES1) with robustness in prediction. Low-risk patients showed favorable outcomes, high immune infiltration and high immunotherapy response. Novel ligand-receptor pairs LGALS9-HAVCR2 and CD86-CTLA4 were specifically interacted between Macro_1 and T/NK cells. VSIG4 and STAB1 were highly expressed in macrophages and were associated with poor prognosis, high tumor purity and high immune checkpoints.
The results provide novel insights into prognosis prediction and therapeutic responses, and identify VSIG4 and STAB1 as potential biomarkers affecting macrophages in HGSOC.
Stem cell pathways were used to determine tumor subtypes and the least absolute shrinkage and selection operator regression was conducted to construct a prognostic risk model, with robustness validation in external datasets. We assessed immune characteristics and therapeutic responses of risk score. Macrophage subpopulations were identified using single cell data, and pseudo-time analysis revealed the changes of macrophages during cell state transition.
HGSOC patients were stratified into stem cell pathway-related clusters (C1, C2) and stem cell-related clusters (GC1, GC2). Patients in C1 and GC1 exhibited better prognosis, increased ImmuneScore, decreased TumorPurity and low immune escape. Patients in C1 were sensitive to gemcitabine while patients in GC1 were sensitive to cisplatin, cyclophosphamide, gemcitabine and niraparib. Risk score was constructed based on 15 genes (IL2RG, STAB1, C2, CD163, FBXO17, VSIG4, CXCL11, CXCL13, GJB1, GPC3, NPY, KRT16, GRIK5, PI3, and RARRES1) with robustness in prediction. Low-risk patients showed favorable outcomes, high immune infiltration and high immunotherapy response. Novel ligand-receptor pairs LGALS9-HAVCR2 and CD86-CTLA4 were specifically interacted between Macro_1 and T/NK cells. VSIG4 and STAB1 were highly expressed in macrophages and were associated with poor prognosis, high tumor purity and high immune checkpoints.
The results provide novel insights into prognosis prediction and therapeutic responses, and identify VSIG4 and STAB1 as potential biomarkers affecting macrophages in HGSOC.