Abstract 966: Exploration of a novel HRD signature (HRDsig) as a biomarker for rucaparib benefit in ARIEL2

Abstract

Abstract Introduction: The ARIEL2 (Parts 1 and 2) all-comers study tested the effectiveness of the PARP inhibitor rucaparib in patients (pts) with platinum-sensitive or resistant/refractory relapsed high-grade ovarian cancer. Pre-specified analyses identified associations of BRCA1/2 mutation status and genomic LOH (gLOH) with prolonged PFS. Recently, a novel scar-based measure of HRD was described [HRDsig; AACR #1249], and we retrospectively examined its predictive value in the ARIEL2 study. Methods: ARIEL2 (CO-338-017; NCT01891344) was an international multicenter, two-part, phase 2 open-label study conducted across 64 sites. Tumor tissues were profiled with comprehensive genomic profiling for all classes of alterations in at least 287 genes (FoundationOne®). HRDsig was called using a machine learning based algorithm with a broad set of genome-wide copy number and short variant features, independent of gLOH (AACR 2022 #1249). Survival analysis was limited to samples where both gLOH and HRDsig could be evaluated (n=394). Hazard ratios were estimated using a univariate Cox proportional hazards model and objective response rates (ORR) were compared using Fisher’s exact test. gLOH high was defined using a cutoff of 16%, based on ARIEL2 and subsequently FDA approved as a complementary diagnostic. BRCA1 promoter methylation was quantified by digital droplet PCR. Results: HRDsig(+) was identified in 56% (251/449) of cases, including 92% (108/117) of those with deleterious BRCA1/2 alterations and 43% (143/332) of BRCAwt. In the intention to treat (ITT) and in pts with platinum sensitive (plat-sen) disease, HRDsig(+) was predictive of PFS benefit on rucaparib (ITT HR = 0.63 [0.50-0.80], p<0.001; plat-sen HR = 0.44 [0.32-0.60]; p<0.001), similar to gLOH-high (ITT HR = 0.70 [0.56-0.87], p=0.0016; plat-sen HR 0.55 [0.41-0.74], p<0.001). In BRCAwt pts with plat-sen disease (n=179), HRDsig was predictive of objective response and PFS on rucaparib, (ORR 28% in HRDsig(+) vs 10% in HRDsig(-), p=0.002; PFS HR = 0.66 [0.48-0.91]; p=0.012). Tumors with RAD51C/D alterations (5/5; 100%) were identified as HRDsig(+). Most other HRR alterations showed little association with HRDsig, including ATM (0/5 HRDsig(+)), and CHEK2 (0/4 HRDsig(+)). Additionally, 33 BRCAwt pts were identified as BRCA1 methylation positive in the cohort, with 32/33 (97%) identified as HRDsig(+), similar to gLOH-high (30/33; 91%). Conclusions: HRDsig(+) was associated with rucaparib benefit overall and in BRCAwt pts with platinum-sensitive ovarian cancer in this study. HRDsig(+) status exhibited strong association with deficiency caused by both epigenetic (BRCA1 methylation) and genetic (HRR mutation) mechanisms. Additional studies should further explore the utility of this biomarker for pt selection in ovarian cancer and other relevant cancer types to inform the use of PARP inhibitors or other DNA damaging agents. Citation Format: Ethan S. Sokol, Russell W. Madison, Dexter X. Jin, Kuei Ting Chen, Zoe Fleischmann, Justin Newberg, Alexa Shrock, David Fabrizio, Jie He, Neeru Bhardwaj, Kevin K. Lin, Iain A. McNeish, Elizabeth M. Swisher. Exploration of a novel HRD signature (HRDsig) as a biomarker for rucaparib benefit in ARIEL2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 966.