5589 Background: Niraparib was approved as maintenance therapy for pts with advanced ovarian cancer (OC) after complete or partial response (CR/PR) to first-line (1L) platinum-based chemotherapy (CT) based on results from the PRIMA study. In this post hoc analysis, we identified factors associated with long-term PFS. Methods: PRIMA is a phase 3, randomized study of pts with newly diagnosed advanced high-grade serous or endometrioid ovarian, primary peritoneal, or fallopian tube cancer at high risk for relapse with CR/PR after 1L platinum-based CT. Pts were randomized 2:1 to niraparib 200 or 300 mg or placebo QD, and stratified by best response to 1L CT (CR/PR), neoadjuvant CT (yes/no), and homologous recombination deficiency (HRD) status (deficient [HRd]/proficient or not determined [HRp/nd]). Investigator-assessed PFS was dichotomized in niraparib-treated pts with progressive disease (PD)/censoring <2 years vs PD/censoring ≥2 years after randomization (data cut, Nov 17, 2021). Logistic regression modeling using backward elimination (significance level=0.15) was used to identify baseline covariates associated with long-term PFS. Results: Of 487 niraparib-treated pts, 152 (31%) had PD/censoring ≥2 years after randomization. Odds ratios (OR) for PFS ≥2 years by baseline subgroup are shown (Table). The logistic regression model showed that BRCA mutation and HRD status, International Federation of Gynecology and Obstetrics (FIGO) stage, fallopian tube as the site of the primary tumor, and absence of baseline nontarget lesions were associated with longer PFS. Eastern Cooperative Oncology Group (ECOG) performance score, body mass index, age, time from CT to randomization, number of target lesions, best response after CT, number of CT cycles, type of debulking surgery, and residual disease were not associated with PFS ≥2 years. Safety has been reported previously ( Ann Oncol. 2022;33[suppl 7]:S789). Conclusions: Results suggest that long-term PFS in pts treated with niraparib may be associated with BRCA/HRd biomarker status, FIGO stage, primary tumor site, and number of baseline non-target lesions. These data are hypothesis generating. Clinical trial information: NCT02655016 . [Table: see text]