Identification of parthanatos-related molecular subtypes and development of prognostic risk models in ovarian cancer based on multi-omics analysis.

Abstract

Ovarian cancer (OC) remains the most lethal gynecologic malignancy due to late-stage diagnosis, high recurrence rates, and chemoresistance. Parthanatos, a distinct caspase-independent form of programmed cell death mediated by poly (ADP-ribose) (PAR), plays a critical role in DNA damage response and tumor progression. In this study, we identified parthanatos-related genes (PRG) associated with OC prognosis based on integrated analyses of TCGA and GEO datasets, and developed a PRG-based risk model with significant prognostic value. Immune infiltration analysis and single-cell RNA sequencing revealed that high-risk patients exhibited a more immunosuppressive tumor microenvironment. Drug sensitivity profiling and clinical subgroup analyses further supported the model's clinical applicability. Among the screened PRG, TGFBI was selected for experimental validation, and in vitro assays including qRT-PCR, western blotting, colony formation, and Transwell migration demonstrated its role in promoting OC cell proliferation, invasion, and clonogenicity. These findings underscore the prognostic and biological significance of parthanatos in OC and suggest that targeting key PRG such as TGFBI may offer novel therapeutic strategies to improve patient outcomes.