Background: The PARPi olaparib (Ola) was recently granted FDA accelerated approval in advanced BMOC. Despite impressive clinical activity in BMOC, the impact of baseline pt factors remains unclear. We hypothesized that the platinum chemotherapy (Plt chemo) to PARPi interval (PTPI) can be used to refine the conventional prediction of response to PARPi based on the clinical categorization of pts into Plt sensitive (Plt-S) and Plt resistant (Plt-R). Methods: Retrospective study of pt with advanced BMOC treated in relapsed setting with ≥ 200 mg bid Ola between 4/2006 – 8/2013 in multiple centers. Pearson Chi², odds ratios (OR) and Fisher’s exact probability tests were used for statistical analyses. Results: 108 advanced BMOC pts were assessed; median age 55y (range 38-79); 83 (77%) pts had high grade serous carcinoma. BRCA1:BRCA2 mutations ratio 71:29. Median prior lines of chemo: 3 (range 1-10). 41 (38%) pts had prior breast cancer (BC). 64% had Plt-S disease and 36% had Plt-R disease prior to Ola. Median PTPI was 53 weeks (w) (range 4-244). Median PTPI was 68.7w for Plt-S pts versus (vs) 25.9w for Plt-R pts (p < 0.0001). RECIST complete or partial responses (CR/PR) were observed in 23/65 (35%) Plt-S pts vs 5/38 (13%) Plt-R pts (p < 0.02). Pts with > 52w PTPI had higher CR/PR rates than those with < 52w PTPI independent of their Plt status (37% vs 18%, respectively, p = 0.053). Pts who had only 1 prior line of chemo had higher CR/PR rates vs pts who had > 1 prior line (p < 0.005). No differences in CR/PR rates were noted between pts with BRCA1 vs pts with BRCA2 mutations, pts with prior BC vs pts without, or use of chemo for BC vs none. Median progression-free survival was 70w for pts with PR/CR vs 28w for pts without (log-rank, p = 0.0004). Median survival was 161w for pts with CR/PR and 64w for pts without (log-rank, p = 0.0005). Conclusions: These data suggest that prediction of response to PARPi in advanced BMOC based on conventional Plt-S/Plt-R categorization may be refined by PTPI. Treatment of Plt-R BMOC pts with a non-Plt agent prior to Ola in order to prolong PTPI may be an appropriate clinical strategy. These findings should be validated prospectively in a larger data set.