Genomic instability is one of the hallmarks of cancer. The incidence of genetic alterations in homologous recombination repair genes increases during cancer progression, and 20% of prostate cancers (PCas) have defects in DNA repair genes. Several somatic and germline gene alterations drive prostate cancer tumorigenesis, and the most important of these are BRCA2, BRCA1, ATM and CHEK2. There is a group of BRCAness tumours that share phenotypic and genotypic properties with classical BRCA-mutated tumours. Poly(ADP-ribose) polymerase inhibitors (PARPis) show synthetic lethality in cancer cells with impaired homologous recombination genes, and patients with these alterations are candidates for PARPi therapy. Androgen deprivation therapy is the mainstay of PCa therapy. PARPis decrease androgen signalling by interaction with molecular mechanisms of the androgen nuclear complex. The PROFOUND phase III trial, comparing olaparib with enzalutamide/abiraterone therapy, revealed increased radiological progression-free survival (rPFS) and overall survival (OS) among patients with metastatic castration-resistant prostate cancer (mCRPC) with BRCA1, BRCA2 or ATM mutations. The clinical efficacy of PARPis has been confirmed in ovarian, breast, pancreatic and recently also in a subset of PCa. There is growing evidence that molecular tumour boards are the future of the oncological therapeutic approach in prostate cancer. In this review, we summarise the data concerning the molecular mechanisms and preclinical and clinical data of PARPis in PCa.