Liquid biopsy for detection of DDR gene alterations in metastatic prostate cancer (mPC): Diagnostic performance and clinical correlations.

218 Background: DNA-damage-repair (DDR) gene alterations, particularly those involving BRCA1/2 genes, are clinically relevant in the management of mPC. Tumor tissue remains the standard DNA source for molecular profiling, but sample availability and quality of retrieved DNA are limiting factors. Liquid biopsy (LB), based on circulating free DNA (cfDNA), offers a minimally invasive alternative for real-time monitoring of DDR gene alterations, especially in consideration of clonal heterogeneity. Methods: We conducted a retrospective study at the Veneto Institute of Oncology including a random sample of 50 mPC patients (pts) previously tested for DDR alterations on tumor tissue: 25 DDR-mutated (DDRmut) and 25 DDR wild-type (DDRwt). cfDNA was extracted from plasma and analysed with amplicon-based NGS (BRCA1/2) and capture-based NGS (16 DDR genes). Mutations found at LB were classified according to IARC/ACMG-AMP guidelines. Clinical variables, genomic findings, and outcomes were compared between DDRmut and DDRwt pts. The study was approved by local Ethics Committee and all pts signed the informed consent. Results: Median age at diagnosis was 68 years; 64% had high Gleason score (GS, 8 to 10), 56% had high-volume metastatic (HV) disease at diagnosis. DDRmut pts had higher PSA at the time of LB (0.76 vs. 0.08 ng/mL; p=0.004), longer time to castration resistance (29 vs. 7 months; p=0.002), and more frequent family history of cancer (56% vs 30%; p=0.044). Median cfDNA concentration was 5.27 ng/mL, higher in GS 8 to 10 vs 6 to 7 (5.57 vs. 4.51 ng/mL; p<0.001). No correlation was found with disease volume or PSA levels. cfDNA quantity and quality were adequate for NGS testing in all DDRmut and 18 DDRwt pts. LB detected the same DDR variants previously reported in tumor tissue analysis in 20/25 DDRmut pts. LB revealed six previously unknown DDR variants in 18 evaluable DDRwt pts involving BRCA2 gene (3pts, 2 Likely Pathogenic, 1 Not Present), ATM (2 pts, both Not Present) RAD51B (2 pts, both Not Present), and BRIP1 (1 pt, Not Present), with 2 pts bearing 2 concomitant variants. Median VAF ranged from 0.5 to 3.8%. All 6 pts had positive family history for tumors related to DDR mutations and were being treated with ARPi (androgen receptor pathway inhibitor) with undetectable or declining PSA. Overall concordance with archival tissue was 79% (κ=0.41). Conclusions: In this mPC cohort, cfDNA concentration correlated with Gleason score rather than PSA levels. LB demonstrated moderate concordance with tissue-based DDR profiling and revealed previously undetected but likely pathogenetic variants even in patients with biochemical response.. These findings support the potential role of LB in longitudinal monitoring of DDR alterations; however, methodological standardization and prospective validation are required before routine clinical implementation.