Bevacizumab (BEV) with or after chemotherapy (CT) for platinum-resistant recurrent ovarian cancer (PROC): Exploratory analyses of the AURELIA trial.

5551 Background: In the open-label randomized phase III AURELIA trial, adding BEV to CT for PROC significantly improved progression-free survival (hazard ratio [HR] 0.48; p < 0.001) and response rate (27% vs 12%) vs CT alone, but not overall survival (OS). Methods: Eligible patients (pts) had measurable/assessable OC that had progressed < 6 mo after platinum CT. After CT selection, pts were randomized to CT ± BEV until progression (PD), unacceptable toxicity or consent withdrawal. Crossover to BEV at PD was optional in the CT arm but prohibited in the BEV–CT arm. Exploratory post hoc analyses assessed factors potentially affecting the decision to crossover to BEV, and efficacy and safety according to post-PD BEV. Results: 179 pts were randomized to BEV–CT and 182 to CT alone. 72 pts (40%) in the CT-alone arm crossed over to BEV after PD and 110 never had BEV. There were no significant differences in pt characteristics between these subgroups at baseline but at the time of PD, 51% vs 35%, respectively, had ECOG performance status (PS) 0 (p = 0.034). 3-month landmark analyses, excluding 28 pts who died or were lost to follow-up before this time, showed significantly longer OS in pts who received BEV either with CT (HR 0.72, 95% CI 0.54–0.97) or after PD (HR 0.69, 95% CI 0.48–0.99) vs those who never received BEV. These differences remained significant after adjusting for identified prognostic factors for OS (chosen CT, platinum-free interval, ECOG PS, baseline ascites, baseline CA-125). When analyzed from the time of PD, the OS HR vs never BEV was 0.84 (95% CI 0.62–1.14) for upfront (pre-PD) BEV–CT and 0.55 (95% CI 0.38–0.79) for BEV after PD. Tolerability was similar with pre- vs post-PD BEV. Conclusions: There appears to be no clear difference in OS between upfront vs post-PD BEV therapy. Pts who never received BEV had the worst OS. These analyses of a non-randomized phase of the study with no information on other post-PD therapies do not allow definitive conclusions about upfront vs post-PD BEV efficacy. Nevertheless, as 60% of pts randomized to CT alone never received BEV, upfront treatment with BEV–CT is important, as reserving BEV until after PD may deny some pts the opportunity to benefit from BEV. Clinical trial information: NCT00976911.