Somatic BRCA2 bi-allelic loss in the primary prostate cancer was associated to objective response to PARPi in a sporadic CRPC patient

Response prediction biomarkers and drug combinations of PARP inhibitors in prostate cancer

PARP inhibitors are a group of inhibitors targeting poly(ADP-ribose) polymerases (PARP1 or PARP2) involved in DNA repair and transcriptional regulation, which may induce synthetic lethality in BRCAness tumors. Systematic analyzes of genomic sequencing in prostate cancer show that ~10%-19% of patients with primary prostate cancer have inactivated DNA repair genes, with a notably higher proportion of 23%-27% in patients with metastatic castration-resistant prostate cancer (mCRPC). These characteristic genomic alterations confer possible vulnerability to PARP inhibitors in patients with mCRPC who benefit only modestly from other therapies. However, only a small proportion of patients with mCRPC shows sensitivity to PARP inhibitors, and these sensitive patients cannot be fully identified by existing response prediction biomarkers. In this review, we provide an overview of the potential response prediction biomarkers and synergistic combinations studied in the preclinical and clinical stages, which may expand the population of patients with prostate cancer who may benefit from PARP inhibitors.

Genomics of lethal prostate cancer at diagnosis and castration resistance

The genomics of primary prostate cancer differ from those of metastatic castration-resistant prostate cancer (mCRPC). We studied genomic aberrations in primary prostate cancer biopsies from patients who developed mCRPC, also studying matching, same-patient, diagnostic, and mCRPC biopsies following treatment. We profiled 470 treatment-naive prostate cancer diagnostic biopsies and, for 61 cases, mCRPC biopsies, using targeted and low-pass whole-genome sequencing (n = 52). Descriptive statistics were used to summarize mutation and copy number profile. Prevalence was compared using Fisher's exact test. Survival correlations were studied using log-rank test. TP53 (27%) and PTEN (12%) and DDR gene defects (BRCA2 7%; CDK12 5%; ATM 4%) were commonly detected. TP53, BRCA2, and CDK12 mutations were markedly more common than described in the TCGA cohort. Patients with RB1 loss in the primary tumor had a worse prognosis. Among 61 men with matched hormone-naive and mCRPC biopsies, differences were identified in AR, TP53, RB1, and PI3K/AKT mutational status between same-patient samples. In conclusion, the genomics of diagnostic prostatic biopsies acquired from men who develop mCRPC differ from those of the nonlethal primary prostatic cancers. RB1/TP53/AR aberrations are enriched in later stages, but the prevalence of DDR defects in diagnostic samples is similar to mCRPC.