PFS by blinded independent central review (BICR) in the VELIA trial of veliparib (V) plus carboplatin/paclitaxel (CP) and as monotherapy in newly diagnosed patients (pts) with high-grade serous ovarian cancer (HGSC).

Abstract

6077 Background: The phase III VELIA trial (NCT02470585) demonstrated statistically significant improvement in PFS per investigator (INV) for V added to CP and continued as maintenance (CPV-V) vs. CP alone in pts with newly diagnosed HGSC in the BRCA mutated ( BRCAm), homologous recombination deficient (HRD), and whole populations. Here we present pre-specified analyses of PFS per BICR. Methods: Pts with Stage III-IV HGSC received V or Placebo (PL) with CP (6 cycles) and as maintenance (30 additional cycles). Primary analysis of PFS by INV compared CPV-V to CP alone in the BRCAm, HRD, and whole populations. Exploratory analyses of PFS in BRCA wildtype (wt) and non-HRD HGSC were performed. Radiologic tumor assessments were also prospectively submitted to an independent central reviewer for blinded assessment per RECIST v 1.1. PFS per BICR and rates of concordance between INV and BICR for determination of disease progression were analyzed. Safety data from the primary analysis were previously reported. Results: 1140 total pts were enrolled (CPV-V 382; CP 375). In the whole population, 26% of HGSCs were BRCAm and 55% were HRD. Concordance rates between INV and BICR were 68-85% by arm for each population. Analyses of PFS per BICR and per INV were consistent (Table). PFS was prolonged in the CPV-V vs. CP arm in all primary and exploratory populations assessed. Conclusions: Analyses of PFS per BICR supported the primary analysis of PFS per INV in the BRCAm, HRD, and whole populations, as well as exploratory BRCAwt and non-HRD populations. Median PFS per BICR was longer compared to PFS per INV assessments in all populations and in both arms. These findings support the reliability of PFS by INV in ovarian cancer trials. Alternate strategies like audits may be appropriate to support PFS by INV with less time and expense than full BICR. Clinical trial information: NCT02470585. [Table: see text]