Integrated bulk and single-cell transcriptomic analysis of ferroptosis-related programs in high-grade serous ovarian...

High-grade serous ovarian cancer (HGSOC) shows marked molecular and microenvironmental heterogeneity. How ferroptosis-related genes (FRGs) relate to immune and metabolic states across bulk tumors and single cells is incompletely defined.

We analyzed TCGA-OV RNA-seq data from 421 patients (419 evaluable for survival) using a curated 25-gene FRG panel, assessing expression, co-expression, survival, and immune/checkpoint associations. Single-cell data from GSE300897 (nine treatment-naive HGSOC donors; 16,939 cells after QC) scored FRG and pathway activity across nine subtypes and compared chemo-refractory versus chemo-sensitive tumors by donor-level pseudobulk analysis.

FTL, FTH1, RPL8, and LDHA were among the most highly expressed FRGs; inter-gene coordination was modest. The aggregate FRG score was not prognostic (log-rank p = 0.80; AUCs 0.52, 0.54, 0.46 at 1, 3, 5 years) but correlated with cytotoxic, CD8 T-cell, macrophage, and Treg signatures. At single-cell resolution, the FRG score was highest in myeloid cells, with enriched iron-metabolism gene expression. Refractory tumors showed higher myeloid and lower fibroblast/plasma-cell fractions, and malignant cells showed directional increases in GLS, CS, and TFRC with decreases in SLC7A11, HMOX1, and GPX4 (none surviving correction). The FRG score showed reproducible positive coupling with hypoxia, glycolysis, and NF-κB programs across platforms. qRT-PCR confirmed higher FTH1, FTL, GPX4, and SLC7A11 in SKOV3 and OVCAR3 versus IOSE80 cells.

Ferroptosis-related transcriptional programs in HGSOC are cell-type dependent and associate with immune infiltration and selected inflammatory/metabolic states, without independent prognostic value. These findings nominate myeloid iron-handling and antioxidant shifts in malignant cells as testable hypotheses for future validation.