Objective To investigate the expression of endothelial lipase (LIPG) in ovarian cancer (OC) and its association with platinum resistance and patient prognosis.

Methods Transcriptomic and clinical data were obtained from TCGA database and validated using clinical OC specimens. LIPG expression was assessed by quantitative real-time PCR and immunohistochemistry. Protein-protein interaction analysis, functional enrichment, and regression modeling were performed to assess the impact of the relevant genes on prognosis and platinum-based chemotherapy response, and a predictive model was subsequently constructed by incorporating immune infiltration characteristics.

Results Fifteen LIPG-differential genes were identified, mainly enriched in lipid metabolism pathways. High LIPG expression was significantly associated with shortened overall survival and progression-free survival. APOA5, LIPG, and PNLIPRP3 were confirmed as independent predictors via multivariate regression analysis, and their combined model showed good performance in predicting platinum response. Immune analysis revealed that high-risk patients exhibited downregulated antigen presentation, increased CD8⁺ T cell and dendritic cell infiltration, and reduced macrophage and regulatory T cell proportions. Clinical validation further confirmed that platinum-resistant patients had significantly increased LIPG mRNA and protein expression levels.

Conclusion High LIPG expression is closely associated with poor prognosis and platinum resistance in OC. A multi-gene risk model incorporating LIPG improves predictive accuracy, and LIPG may serve as a potential biomarker and therapeutic target of ovarian cancer.