**1. Introduction**

Epithelial ovarian cancer (EOC) is the second most common and the most lethal gynecologic malignancy responsible worldwide for ≈ 207,000 deaths [1]. EOC affects mainly postmenopausal women and is typically diagnosed at an advanced stage due to the absence of specific symptoms compounded with no effective early screening modalities that allow to detect the disease when it is localized [2, 3].

Several studies have demonstrated that advanced EOC represents a heterogeneous group of malignancies with complex molecular and genetic features associated with specific pathogenic pathways alteration and phenotypic clinical behavior [4]. Highgrade serous ovarian carcinomas (HGSOC) represent the most common histotype and are linked to poor prognosis [4]. The vast majority of HGSOC arises from the fallopian tube (FT) as a precursor known as serous tubal intraepithelial carcinoma [5]. Moreover, several findings suggested that FT and primary peritoneal cancer (PPC) share the same pathobiology and genetic aberrations associated with HGSOC. As such, patients diagnosed with "pelvis serous carcinoma" should be considered as collectively having the same disease and, therefore, they should receive uniform treatment options [5].

Furthermore, the Cancer Genome Atlas (TCGA) project has identified that approximately 50% of HGSOC exhibits defects in homologous recombination repair (HRR) pathway, which is major biochemical machinery for the repair of DNA doublestrand breaks (DSBs) in mammalian cells [6]. DSBs represent the most serious manifestation of DNA damage because, if left unrepaired, they can lead to genomic instability, which is considered one of the main features of carcinogenesis [7]. Given that breast cancer gene (BRCA)1 and (BRCA)2 are the tumor suppressor genes (wildtype BRCA allele is lost during tumorigenesis) involved at different stages of HRR, carriers of deleterious heterozygous germline mutations in the BRCA1 and BRCA2 genes have significantly elevated risks of developing breast, ovarian, and other cancers [7]. At the same time, tumors that exhibit homologous recombination deficiency (HRD) are susceptible to specific systemic treatments, including poly ADP-ribose polymerase (PARP) inhibitors (PARPis) [8].

Poly(ADP-ribose)polymerases (PARPs) are a family of nuclear ubiquitous enzymes, which regulate the biological functions of a variety of proteins by catalyzing their posttranslational modification, named PARylation, using NAD+ as substrate [9]. By post-transcriptionally modifying multiple proteins, PARPs act as signal transducers, contributing to the regulation of various cellular functions including the signaling pathway that leads to the resolution of DNA strand breaks. In this context, PARPs play as promoters of genomic integrity and stability, activating different mechanisms for DNA repair, stabilizing replication forks, and modeling the chromatin structure [10].

PARP1 is the most abundant and studied member of the PARP family, accounting for 80–90% of total PARP activity in the cell, and it is also known as the major PARproducing enzyme in eukaryotes [11–13]. PARP1 is centrally involved in the early response to cellular oxidative and genotoxic stress, to which cells are constantly exposed [10]. In this context, PARP1 acts as a crucial DNA damage sensor, participating in the DNA damage response (DDR), the network of molecular pathways that maintain the genomic integrity by recognizing DNA damages and orchestrating their repair [14]. In DDR, PARP1's first action is to trigger the repair of DNA single-strand breaks (SSBs) to ensure cellular genomic stability. Indeed, if not repaired, SSBs are likely to be converted, during DNA replication, into DSBs, the most harmful form of DNA damage that led to the genomic instability, eventually responsible for the development of many diseases, including cancer [10, 15].

In addition, PARP1 promotes the repair of DNA DSBs through the high-fidelity homologous recombination (HR), by activating and recruiting multiple proteins such as ATM, Mre11, and Nbs11 to DSB lesions, and simultaneously inactivating DNAdependent protein kinases that favor the more error-prone nonhomologous endjoining (NHEJ) [16].

Suppression of PARP1 leads to the accumulation of unrepaired DNA SSBs and the stalling of replication forks [17]. The persistence of SSBs culminates in the collapse of stalled replication forks into highly cytotoxic DSBs. HR is considered the highest fidelity machinery to repair DSBs and indispensable to maintain the genomic integrity. HR, as a complex mechanism, involves a large number of proteins that operate from the DSBs detection to the effective DNA repair. In this context, BRCA1 and BRCA2 are crucial players to guarantee HR high efficiency. Indeed, both proteins interact with many HR effectors, participating in the DSB detection and guiding the formation of the complex that effectively repairs the DNA strand. BRCA2 is even

more important since it is responsible for the recruitment and loading onto the DNA strand of RAD51, the recombinase, defined as the catalytic core of HR, that guides homology search and strand invasion.

In case of PARP inhibition, HR acts as a compensatory pathway to maintain the genomic integrity and guarantee cell survival [16]. Normal cells are BRCA-proficient, thus able to efficiently repair DSBs and survive under PARP inhibition. On the contrary, cancers harboring BRCA1 or BRCA2 mutations become HR-defective and highly vulnerable to the effects of PARP inhibition, facing a genomic instability that turns into cell death [18]. This relationship has been defined as synthetic lethality (SL) and it has been exploited as a strategy to selectively target cancers with somatic and germline BRCA1 and BRCA2 mutations [18]. The concept of SL was originally derived from genetic studies on gene-to-gene interactions and their consequent impact on cell viability. According to this genetic principle, two genes are synthetic lethal if their simultaneous mutation causes cell death, while the mutation of either gene alone is compatible with cell viability. (**Figure 1**).

This SL relationship can be explained by the presence of a buffering effect, which links two genes and is lost in case of simultaneous mutation. The SL concept can be extended to proteins encoded by synthetic lethal genes and, in turn, to the cellular pathways. According to this, SL has been exploited for drug discovery to selectively treat cancers, harboring a specific gene mutation, with drugs targeting the synthetic lethal partner. Notably, taking advantage of a mutation present only in cancer cells, SL approach promises to be selective, killing cancer cells while sparing normal ones.

The evidence of a synthetic lethal interaction between PARP inhibition and BRCA1/2 mutation suggested a clinical strategy to treat cancers with loss-of-function mutations in either BRCA1 or BRCA2 genes with PARPis as drugs [14, 16]. PARP1 is the primary target of clinically used PARPis. Initially, PARP1 inhibitors were developed to be used as potentiators of DNA damaging chemo- or radiotherapeutic agents [14, 19]. Later, PARPis revealed their potential as single agents in the treatment of BRCAness tumors with an HR-defective condition. BRCAness refers to tumors with specific genomic signatures (other than BRCA mutation) that cause HR-deficiency and thus susceptibility to SL of PARPis [20]. PARPis block the catalytic activity of PARP1 by directly binding to the NAD+ pocket, responsible for the synthesis of PAR chains. For this reason, the originally proposed mechanism of action (MOA) to explain the SL effect, described PARPis as direct inhibitors of the PARylation, which causes the impairment of DNA repair proteins and the phenocopying effect of deleting PARP1 [19]. Indeed, persistent unrepaired DNA breaks can cause the collapse of replication forks with the formation of DSBs, not repaired by HR-deficient cells. However, the most credited MOA of PARPis is their ability to trap PARP1 on DNA strand, ultimately preventing its release from the DNA strand by the inhibition of autoPARylation and PARP1 conformational change. The trapped PARP1 acts as an obstacle, causing unstable replication forks and consequent accumulation of DNA lesions, which are eventually repaired by error-prone mechanisms in HR-deficient

**Figure** 1**.** *Concept of synthetic lethality.*

cells [19]. This mechanism explains the PARPis cytotoxic effect and most likely accounts for the SL effect in BRCAness tumors.

This chapter examines the clinical development studies, which lead to PARPis approval, the safety and the management of adverse events associated to this new class of drugs, and rational consideration that should guide the use of PARPis in the frontline setting.
