Background: Deregulation of the cyclin-dependent kinases 4 and 6 (CDK4/6) – p16 – Rb signaling pathway is commonly found in ovarian cancer (OC). Palbociclib is an inhibitor of CDK4/6 and was recently shown to delay disease progression in postmenopausal women with advanced breast cancer. Here we study the safety and efficacy of palbociclib in patients (pts) with recurrent OC. Methods: Asymptomaticpts with RECIST and/or CA125 measurable disease failing chemotherapy or anti-hormonal therapy were given oral palbociclib 125 mg once daily for 3 weeks followed by 1 week off over 28-day cycles. Response was assessed every 8 wks. The primary endpoint was proportion of patients progression-free at 6 months. Secondary endpoints included ORR (RECIST or GCIG CA125 criteria), PFS and safety. Tumor was collected for targeted next generation sequencing (NGS). NCT01536743 Results: As of January 28, 2016, 37 of 40 pts were accrued from 2 centers. Safety and efficacy data are available for 37 and 30 pts, respectively. Pt characteristics were: Serous (78%), RECIST measurable disease (87%), elevated CA125 (100%) and ECOG PS 0 (67%). Median age was 61 years (range 42-78). Median number of prior chemotherapy regimens was 3 (range 1-12). The proportion of patients who were progression-free at 6 months were 9/30 (30%). Using RECIST median PFS was 3.7 months (95%CI, 1.2-6.2), and 1/26 PR (4%), 17/26 SD (65%) and 8/26 PD (31%) were seen. Using GCIG CA125 criteria median PFS was 4 months (95%CI, 0.3-7.6), 1/30 CR (3%), 3/30 PR (10%), 18/30 SD (60%) and 8/30 PD (27%) were seen. Toxicity was minimal; grade 2 events included anemia (2), nausea (1) abdominal pain (1), grade 3/4 events included neutropenia (5), thrombocytopenia (4), hypokalemia (1) and emesis (1). 1 pt experienced a bowel obstruction and 1 pt died due to disease progression within 30 days of treatment discontinuation. Efficacy is being correlated with targeted NGS data including CDKN2A, RB or CCNE. Conclusions: CDK4/6 inhibition with palbociclib was well tolerated and demonstrated single-agent activity in heavily pretreated unselected OC patients. Ongoing predictive biomarker analyses may facilitate patient selection. Clinical trial information: NCT01536743