Genomic instability score (GIS) and benefit from olaparib (ola) and bevacizumab (bev) maintenance in high-grade ovarian cancer (HGOC): Phase III PAOLA-1 GINECO/ENGOT-ov25 trial exploratory analysis.

Abstract

5576 Background: The PAOLA-1 trial showed that adding ola to bev as maintenance therapy improved overall survival (OS) of HGOC patients with BRCA1/2 mutations (BRCAm) or homologous recombination (HR) deficiency (HRD) defined by the MyChoice HRD Plus assay with a GIS threshold of 42. This post hoc analysis of PAOLA-1 explored if alternative thresholds could better identify patients who benefit most from ola. Methods: New cutoffs were determined through OS analyses using Cox proportional hazards models with an interaction term for GIS. Tumors were categorized into HRP (GIS<42), HRDlow (42–60 for BRCA1/2 wild-type [BRCAwt], 42–67 for BRCAm), and HRDhigh (>60 for BRCAwt, >67 for BRCAm). Genomic analyses included promoter methylation, BRCA loss of heterozygosity (LOH) and HR repair gene mutations. Results: Among 623 patients, 194 (31%) were BRCAm and 429 (69%) BRCAwt. Main clinical prognostic features were well-balanced across BRCAwt/HRDhigh, BRCAwt/HRDlow and BRCAwt/HRP as well as among BRCAm/HRDhigh, BRCAm/HRDlow and BRCAm/HRP. Ola+bev improved progression-free survival (PFS) and OS in BRCAwt/HRDhigh and BRCAm/HRDhigh (Table). Ola+bev improved PFS but not OS in BRCAwt/HRDlow and BRCAm/HRDlow. HRP tumors showed no PFS or OS benefit regardless of BRCA status. BRCA1 / RAD51C promoters were methylated in 75% of BRCAwt/HRDhigh, 47% of HRDlow, and 3% of HRP. HRDlow tumors had fewer HR repair gene mutations than HRDhigh. Among HRP/BRCAm, 37% lacked BRCA LOH, suggesting functional BRCA. Conclusions: Our post-hoc subgroup analyses suggest that refined GIS thresholds identify three distinct populations of HGOC patients with varying survival benefits from ola+bev maintenance. Optimized GIS cutoffs may further improve patient stratification in future PARP inhibitors trials. BRCA WT BRCA Mut HRP HRD low HRD high p HRP HRD low HRD high p N 277 72 80 19 124 51 BRCA mutation 0.36 BRCA1 ‐ ‐ ‐ 11 (57.9%) 82 (66.1%) 38 (74.5%) BRCA2 ‐ ‐ ‐ 8 (42.1%) 42 (33.9%) 13 (25.5%) No BRCA LOH ‐ ‐ ‐ 7 (36.8%) 1 (0.8%) 2 (4.0%) p<0.001 HR gene methylation (NA=134) p<0.001 No 187 (96.9%) 26 (53.1%) 13 (24.5%) ‐ ‐ ‐ BRCA1 1 (0.5%) 15 (30.6%) 34 (64.2%) ‐ ‐ ‐ RAD51C 5 (2.6%) 8 (16.3%) 6 (11.3%) ‐ ‐ ‐ mPFS (95%CI ), months Ola + bev 16.6(15.2-18.2) 28.9(20.3-NR) 38.9(22.1-NR) 21.2(13.9-NR) 51.4(38.9-NR) 75.2(NR-NR) Placebo + bev 16.2(13.9-18.8) 16.4(12.9-27.7) 17.0(12.9-23.4) 20.3(14.7-NR) 19.4(16.6-24.0) 15.5(8.7-NR) HR (95%CI) 1.00(0.76-1.32) 0.51(0.29-0.91) 0.42(0.24-0.72) 0.80(0.28-2.26) 0.39(0.24-0.62) 0.17(0.07-0.41) mOS (95%CI ), months Ola + bev 36.8(30.7-40.9) 54.0(48.3-NR) NR(54.1-NR) 47.0(24.2-NR) NR(NR-NR) 75.2(NR-NR) Placebo + bev 40.4(33.0-53.3) 52.4(45.8-NR) 41.2(34.0-NR) 43.1(29.0-NR) NR(59.8-NR) 55.2(29.8-NR) HR (95%CI) 1.19(0.87-1.62) 1.07(0.55-2.07) 0.49(0.26-0.94) 0.88(0.28-2.79) 0.61(0.32-1.17) 0.15(0.05-0.50)