Multi-omics study of prognostic models and molecular networks related to ovarian cancer

Ovarian cancer (OV) is considered the most lethal gynecological cancer in women. Despite significant advancements in treatment and risk-stratification methods, these approaches remain far from ideal. This study aims to leverage large-scale public cohorts to identify differentially expressed prognostic genes between OV and normal ovarian tissue and evaluate their impact on patient survival.

Utilizing data from extensive public cohorts and machine learning methods, we conducted a comprehensive screening to identify genes differentially expressed in ovarian cancer compared to normal ovarian tissue. We also developed a risk score for each patient based on these genes. Subsequent analyses explored the immunological profiles and genomic alterations associated with different risk scores.

Our analysis revealed that a high risk score is positively correlated with poor survival in OV patients. The risk score is associated with key oncological pathways, immune-related processes, and genomic changes. Notably, patients with higher risk scores exhibited increased levels of immune cell infiltration and significant remodeling of the immune microenvironment. Furthermore, there is a strong correlation between the risk score and immune checkpoint molecules, suggesting potential benefits from immune checkpoint blockade strategies. The risk score also proved to be a stable and sensitive indicator for predicting sensitivity to various chemotherapeutic drugs.

Through an integrative approach, our study deciphers the prognostic, immune, and therapeutic value of the risk score in OV. This analysis highlights the importance of the risk score in predicting survival, modulating immune response, and guiding chemotherapy sensitivity, thus supporting its utility in improving personalized treatment strategies for ovarian cancer patients.