Assessment of cancer-associated fibroblast signature genes in ovarian cancer patients: impact on immunity, drug resistance, and prognosis

Abstract

Ovarian cancer (OC) is women's third most common gynecologic tumor and is highly lethal. Cancer-associated fibroblasts (CAFs) are associated with cancer at all stages of disease progression and are involved in biological processes, including inflammatory processes, tumor development occurrence, and immune rejection. This study aimed to construct prognosis-related CAFs regulatory factors to predict the survival of OC patients. Datasets of OC patients with complete clinical information were collected from the Gene Expression Omnibus (GEO) and the Cancer Genome Atlas (TCGA) databases. First, we identified potential regulator factors of CAFs in OC based on the xCell algorithm and weighted gene co-expression analysis (WGCNA). Further screening using one-way cox regression analysis and LASSO regression models yielded 22 prognosis-related CAFs regulatory factors, using which a model was constructed. Subsequently, the diagnostic effectiveness of the model was assessed using receiver operating characteristic (ROC) curves, and the validity of the CAFs regulatory factors survival model was verified in three additional independent datasets and single cell data. Meanwhile, experimental validation was conducted using immunohistochemistry and Western blot. The results showed that GAS1 (Growth arrest specific 1) exhibited a higher expression pattern in fibroblasts from ovarian cancer patients.

The assessment of resistance and immune checkpoint differences across various risk score groups indicates that the CAFs regulatory factor survival model is practical for guiding systemic treatment. In summary, this study establishes a prognostic model composed of 22 CAFs regulatory factors to predict the prognosis of ovarian cancer (OC), providing new perspectives for the clinical treatment of OC.