Background: Options for first-line (1L) maintenance therapy in ovarian cancer (OC) have evolved in the US in recent years, particularly with FDA approvals of poly(adenosine diphosphate [ADP]-ribose) polymerase inhibitors (PARPi). Olaparib was approved in 2018 for 1L maintenance treatment of patients (pts) with advanced OC with BRCA mutation and in 2020 as combination therapy with bevacizumab for 1L maintenance treatment of pts with homologous recombination deficient (HRd)–positive tumors. Additionally, the FDA approved niraparib in 2020 for maintenance treatment of pts with advanced OC regardless of tumor biomarker status. This study aimed to describe use and outcomes of 1L maintenance vs. active surveillance (AS) among PARPi-eligible pts with OC in a real-world setting prior to the most recent 2020 FDA approvals. Methods: This retrospective cohort study included pts with newly diagnosed stage III/IV OC who received 6–9 cycles of 1L platinum-based chemotherapy (PBC) and either primary debulking surgery or interval debulking surgery following neoadjuvant chemotherapy between January 1, 2016, and February 29, 2020, regardless of biomarker status, from the Flatiron Health database, a longitudinal electronic health record-derived database consisting of de-identified patient-level data that are curated via technology-enabled abstraction. The end of the last cycle of 1L PBC was defined as the index date. Pts who started second-line (2L) treatment within 2 months of the index date were excluded. Primary endpoint was time to initiation of 2L systemic therapy (as a surrogate for progression) or death. Inverse probability of treatment weighting and Cox proportional hazard model were used to adjust for baseline differences among pts on maintenance therapy and pts on AS. Results: A total of 463 pts were included in the study, 87.7% from community practices and 12.3% from academic institutions. Of the pts included, 21.0% received maintenance therapy, while 79.0% did not. Of those who received maintenance therapy, 48.5% received bevacizumab, 40.2% received PARPi (olaparib, rucaparib), and 11.3% received paclitaxel. Median progression-free survival (PFS) for pts who received 1L maintenance therapy was 16.1 months, compared with 12.2 months in pts who did not receive 1L maintenance therapy. After adjusting for baseline differences in characteristics and demographics, including age, race, stage of cancer, and BRCA status, pts on maintenance therapy had a statistically significant, 29% lower risk of progression or death than those receiving AS (hazard ratio: 0.71; 95% CI, 0.52–0.99; P= 0.04). Conclusions: In this real-world analysis, the majority of pts did not receive maintenance therapy; however, a PFS benefit was found in those receiving maintenance therapy. Further studies are needed to understand how biomarker status drives practice patterns.