Objective To develop a methylation-based predictive model for PARP inhibitor (PARPi) efficacy in advanced high-grade serous ovarian cancer (HGSOC) patients. Design Retrospective analysis of tumor tissues from 96 advanced HGSOC patients receiving PARPi maintenance therapy (78 sensitive, 18 insensitive). Setting A tertiary gynecologic oncology center at Women’s Hospital, Zhejiang University School of Medicine, China. Population or Sample 96 advanced HGSOC patients receiving PARPi maintenance therapy. Methods Pyrosequencing assessed methylation levels across promoter regions of 10 HRR genes. A nomogram prediction model was constructed using LASSO regression to identify key methylation sites associated with PARPi sensitivity in a training cohort (n=68) and validated in an independent cohort (n=28). Performance was compared to conventional BRCA/HRD testing. Main Outcome Measures PARPi sensitivity prediction accuracy measured by AUC of ROC curves and comparison with BRCA/HRD testing. Results 93 of 96 methylation sites showed significantly lower median methylation levels in PARPi-sensitive patients (pConclusions HRR gene promoter methylation, particularly PALB2, RAD51D, and CHEK2, is a potent predictor of PARPi efficacy in HGSOC. The nomogram provides a robust clinical tool outperforming standard BRCA/HRD testing.