AUTHOR=Wu Yifan , Chen Cunte , Cui Yingwen , Zou Ruoyao , Yang Yaoxiang , Sun Fengjie , Du Yanzi , Wang Peipei TITLE=PD-1 and LAG-3 were optimal combination of immune checkpoints for predicting poor clinical outcomes of patients with ovarian cancer JOURNAL=Frontiers in Immunology VOLUME=Volume 16 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2025.1656242 DOI=10.3389/fimmu.2025.1656242 ISSN=1664-3224 ABSTRACT=BackgroundAlthough immune checkpoint blockade (ICB) therapy has transformed the therapeutic landscape for ovarian cancer (OC), the predictive utility of immune checkpoint (IC) expression signatures in stratifying clinical outcomes requires further systematic interrogation.MethodsTranscriptomic profiles from 147 OC patients within The Cancer Genome Atlas (TCGA) cohort were interrogated to assess the prognostic significance of ICs. These genomic findings were subsequently validated through immunohistochemical analysis of an independent institutional cohort comprising 74 OC tissue specimens.ResultsBoth TCGA and validation cohorts demonstrated that elevated expression of PD-1 and LAG-3 correlated with inferior overall survival (OS) in patients with OC. Importantly, among the ICs, PD-1/LAG-3 co-expression emerged as the optimal combinatorial biomarker, independently predicting adverse outcomes [hazard ratio (HR) = 1.74, 95% confidence interval (CI): 1.12-2.70, P < 0.001]. The derived nomogram model incorporating PD-1/LAG3 status, TNM stage, histologic grade, and age generated patient-tailored 1–5 year OS rate estimates. Notably, risk stratification using this model significantly enhanced prognostic precision versus conventional parameters (TNM stage or histologic grade) alone, especially in patients with serous cystadenocarcinoma.ConclusionElevated IC expression correlated with poor OS in OC patients. Specifically, PD-1/LAG-3 co-expression emerged as the optimal prognostic biomarker pair, representing a promising therapeutic target for dual checkpoint blockade strategies in OC.