Prostate cancer is a leading cause of cancer death in men. A recent study suggests that prostate tumors with defects in DNA damage repair (DDR) genes may respond to PARP inhibitor therapy. In breast and ovarian cancer a homologous recombination deficiency (HRD) score optimally defines treatment groups for DNA damaging therapy. This study investigates mutations in DDR genes and molecular signatures for HRD, microsatellite instability (MSI), and high mutation load (HML) in PC.
DNA was extracted from 50 radical prostatectomies and 45 transurethral resections of the prostate, and analyzed using a next generation sequencing assay targeting 43 genes, and genomic regions used to generate HRD, HML, and MSI scores.
DDR genes were considered non-functional if both alleles were mutated and/or deleted. If the second allele is intact these genes were considered defective but functional. Non-functional DDR genes (CDK12, PALB2, RPA1, ATM, and BRCA2) were observed in 7 tumors. 11 tumors were observed with DDR gene defects in 8 additional genes. DDR gene mutation status was significantly associated with high Gleason score (n = 84; p = 0.0028). Mean HRD scores were higher in tumors with non-functional DDR genes (n = 7) compared to non-mutant tumors (n = 49) (31.7 vs. 14.6; p = 0.0039), but not in tumors with defective DDR genes (n = 11) (14.0 vs. 14.6; p = 0.92). HRD scores were associated with high Gleason score (Gleason ≤7 mean = 11.3; Gleason >7 mean = 20.9; p = 0.00064). 3 MSI positive tumors were identified, all had Gleason scores >7.
HRD scores, non-functional DDR genes (but not DDR gene defects), and MSI are all associated with higher Gleason scores in PC. A significant proportion of aggressive prostate tumors carry molecular signatures associated with response to therapies targeting DDR deficiencies or to immunotherapeutics. This study demonstrates the importance of assessing both alleles when identifying prostate tumors with DDR gene mutations. In this study an HRD score of ≥20 captures three times as many potential responders to HRD-dependent therapies compared to non-functional DDR gene mutations. Further studies are required to investigate response to targeted therapies in PC.
Clinical trial identification
Legal entity responsible for the study
K. Timms, C. Neff, J. Reid, C. Solimeno, D. Pruss, S. Stone, Z. Sangale, A. Gutin, J. Lanchbury: Myriad Employee, salary and stock options. J. Cuzick: Consulting/Advisory-Becton Dickinson.