776P - Liquid biopsy for the determination of homologous recombination deficiency in patients with high-grade serous ovarian cancer: Results of the BOVARY-pilot study

Background

Ovarian cancer ranks 15th globally in cancer-related deaths. Deleterious variants in BRCA1/BRCA2 genes and alterations in HRR pathway genes or BRCA1/RAD51C promoter methylation cause HRD, impacting response to PARP inhibitors. Tissue biopsy with HRR-gene sequencing and genomic scar evaluation is standard for HRD determination, but some cases lack viable tissue. In the BOVARY-pilot trial (NCT03881683), we explored whether liquid biopsy could substitute FFPE tissue for HRD assessment.

Methods

24 patients with histologically proven stage III or IV high-grade serous ovarian carcinoma (HSOG) were included. All patients gave their informed consent. Matched FFPE tissue/plasma samples were collected at baseline for all patients. DNA was extracted from FFPE samples and cfDNA was extracted from plasma. Next-generation Sequencing (NextSeq 550, Illumina) was utilized to analyze the HRR-gene panel, including BRCA1/2, and to evaluate genomic integrity index (GII) using low-passWGS (Sophia Genetics). Additionally, all FFPE samples were analyzed using Myriad MyChoice CDx Plus for HRR-panel and genomic instability score (GIS). The methylation of BRCA1/RAD51C promoters in all FFPE and cfDNA samples was analyzed using ddPCR (BioRad). Results were finally compared between FFPE/plasma pairs.

Results

20 samples were available for analysis. All SNVs and indels identified in FFPE tissue were also detected in the blood samples (Se=100%, Sp=100%). HRD status was determined for all 20 samples with high agreement between GII and GIS scores (90.0%, 95% CI [0.69; 0.97]). Sensitivity was 92.8% (95% CI [0.68; 0.99]) and Specificity 83.3% (95% CI [0.44; 0.97]). Among the plasma samples, only those with the highest cfDNA concentrations (3/17) yielded a valid HRD status using the low-passWGS approach, and HRD status was concordant for all 3 samples. Additionally, methylation of the BRCA1 promoter was detected in only one HRD sample, with concordant methylation found in the paired cfDNA sample.

Conclusions

Liquid biopsy shows promise as a surrogate for tissue biopsy in determining HRR-gene alterations, genomic instability, and BRCA1 and RAD51C methylation statuses, albeit with some limitations.

Clinical trial identification

NCT03881683.

Editorial acknowledgement

Legal entity responsible for the study

Institut de Cancérologie de Lorraine.

Funding

Tesaro, Sophia Genetics.

Disclosure

A. Harlé: Financial Interests, Personal, Invited Speaker: Biocartis, MSD, Nonacus, Pierre Fabre, Amgen, Roche; Financial Interests, Personal, Advisory Board: AstraZeneca, GSK, HederaDx, Janssen, Sophia Genetics. P. Gilson: Financial Interests, Personal, Invited Speaker: AstraZeneca, BMS. J.L. Merlin: Financial Interests, Personal, Advisory Board: Novartis, Pfizer, Pierre Fabre, MSD, Merck, Roche; Financial Interests, Personal, Invited Speaker: GSK, AstraZeneca. C. Gavoille: Financial Interests, Personal, Advisory Board: Kephren, GSK, Mundipharma, AstraZeneca. All other authors have declared no conflicts of interest.