Poly-ADP-ribose polymerase inhibitors (PARPis) have revolutionized the management of BRCA-mutated (BRCA ^mut ) and homologous recombination deficiency (HRD)-positive ovarian cancer (OC). While long-term analyses clearly support the use of PARPi as maintenance therapy after first-line chemotherapy, recent data have raised concerns on detrimental overall survival (OS) in non-BRCA ^mut OC, a greater incidence of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), and unfavorable outcomes following subsequent platinum-based chemotherapy in pretreated OC patients. In this report we discuss the long-term follow-up results from phase III trials in pretreated OC patients, which led to the Food and Drug Administration's withdrawal of PARPi indications in this setting. We summarize the newly available evidence concerning the risk of MDS/AML and the post-progression efficacy results after PARPi. We emphasize the importance of long-term follow-up and real-world data coming from international registries to define the efficacy and safety of stopping PARPi at relapse at a pre-specified time. To this point, biomarkers able to identify the patients who will experience long-term remission with PARPi maintenance or develop early resistance are urgently needed to guide treatment decision and duration.