Molecular and spatial differences in the tumor microenvironment of high-grade serous ovarian cancers with short...

Survival of patients diagnosed with advanced-stage high-grade serous ovarian cancer (HGSOC) varies widely. Understanding the biological and molecular differences between tumors associated with short-term and long-term survival may inform prognosis and treatment.

Omental tumor samples were collected from 15 short-term survivors (STS; <12 months) and 15 long-term survivors (LTS; ≥ 120 months) with stage III-IV HGSOC who underwent primary debulking surgery followed by platinum-based chemotherapy. Bulk tumor sections were analyzed using whole-genome RNA sequencing. Corresponding H&E-stained whole-slide images were processed using machine learning-based cell segmentation to identify cell phenotypes. Computational image analysis was used to extract morphometric features. The integrated transcriptomic and image-derived data were used to quantify proportions of cancer cells, fibroblasts, and immune cells, and to identify differences in nuclear texture, cancer cell spatial organization, and collagen architecture.

STS was associated with higher fibroblast content (48.6% vs 33.5%, p < 0.05), lower cancer cell content (48.0% vs 64.3%, p < 0.05), and a lower tumor-fibroblast proportion (0.9 vs 1.8, p < 0.05) than LTS. Computational image analyses identified significant differences in nuclear texture, cancer cell spatial organization, and collagen architecture. STS correlated with upregulation of fibroblast-associated genes (ITGAV, p < 0.05; INHBA, p < 0.0001) and epithelial-mesenchymal transition markers (ZEB1, p < 0.001; CALD1, p < 0.001), while LTS was enriched for immune signaling pathways.

Omental metastases from STS and LTS patients have distinct molecular and spatial profiles. The enrichment of fibroblast-associated gene signatures in STS and immunoreactive pathways in LTS underscores a divergent tumor microenvironment with potential clinical relevance.