PARP Inhibitors in Ovarian Cancer: A Trailblazing and Transformative Journey

In this issue of Clinical Cancer Research, Ison and colleagues (1) describe the FDA approval of the PARP inhibitor niraparib, which is the first PARP inhibitor to be approved by the FDA as maintenance for recurrent ovarian cancer following response to platinum-based chemotherapy. This approval is part of a remarkable and transformative continuum of establishment of the anticancer activity of PARP inhibitors by targeting DNA repair deficiencies. The initial observation that PARP inhibitors have profound in vitro activity against BRCA-mutated (BRCAm) cancer cells was made in 2005 (2), activity of these agents in patients with heavily pretreated BRCA-related cancers was shown in 2009 (3), the first FDA approval of olaparib for the treatment of patients with germline BRCAm (gBRCAm) ovarian cancer who have received at least three lines of prior chemotherapy occurred in 2014, and now this current milestone of the first PARP inhibitor approved as maintenance in patients with recurrent ovarian cancer following response to platinum-based chemotherapy (1). Ovarian cancer has been the model cancer to test these drugs because of the enrichment, especially in high-grade serous carcinoma (HGSC), of underlying DNA repair deficiency that PARP inhibitors exploit through the mechanism of synthetic lethality (4).

Niraparib is a potent inhibitor of PARP1 and PARP2. PARP1, the main member of the PARP family, is a nuclear enzyme that transfers poly-ADP-ribosyl moieties from nicotinamide-adenine-dinucleotide (NAD) to many target proteins in a posttranslational modification termed “parylation.“ PARP1 contributes 85% to 90% of PARP activity (and NAD⁺ utilization), whereas 5% to 10% is facilitated by PARP2. PARP2 is the closest relative of PARP1 and exhibits almost overlapping functions as PARP1 (5); both enzymes are involved in base excision repair. In addition, PARP2 is essential for the viability of PARP1 knockout mice, as PARP1/PARP2 double knockout mice are not viable. Parylation plays an important role in promoting DNA repair by allowing for recruitment of proteins involved in repair of single-strand breaks (SSB) and double-strand breaks (DSB) as well as in stalled replication fork protection and restart (Fig. 1). Inhibition of SSB and trapping of PARP–DNA complexes at the replication fork are the most prevalent mechanisms of action of PARP inhibitors against homologous recombination (HR)–deficient cells, although other mechanisms such as replication fork collapse and enhancement of toxic classic nonhomologous end joining (C-NHEJ) have also been proposed. Beyond DNA repair, PARP1 has numerous alternative functions (Fig. 1), including regulation of transcription, angiogenesis, adaptation to hypoxia, epithelial-to-mesenchymal transition (EMT), metastasis, inflammation, and immune regulation. In this regard, the anticancer activity of PARP inhibitors extends beyond DNA repair, and PARP inhibitors have been considered for management of a number of nononcologic conditions (Fig. 1; refs. 6–9).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

PARP1 functions, results of PARP inhibition, and clinical/translational applications of PARP inhibitors (PARPi). Top, PARP1 functions in repair of DNA damage. PARP1 participates in SSB and DSB repair as well as in stalled replication fork protection and restart. Inhibition of SSB and trapping of PARP–DNA complexes at the replication fork are the most prevalent mechanisms of action of PARPi against HR-deficient cells although other mechanisms such as replication fork collapse, enhancement of toxic classic nonhomologous end joining (C-NHEJ) in PARP1-deficient cells, and inhibition of alternative end joining (Alt-EJ) have also been proposed. Besides anticancer activity, unresolved DSBs may induce immune “priming” of the tumor for response to PD-1/PD-L1 inhibitors via activation of innate immunity (via activation of the STING pathway) and by upregulation of PD-L1 expression in cancer cells. Bottom, PARP1 functions beyond DNA damage repair. 1. PARP1 serves as a potent modulator of gene transcription through activities that include transcription factor (TF) regulation, chromatin regulation, and the ability of PARP1 to serve as transcriptional coregulator and chromatin modifier. PARP1 also interacts with RNA polymerase (pol) II complexes, and can thus both up- and downregulate gene expression. PARP inhibition leads to context-specific stimulation or inhibition of gene expression, which can be exploited in both oncologic and benign indications. 2. PARP1 promotes angiogenesis via upregulation of transcription of proangiogenic factors such as syndecan-4 and Id-1 as well as activation of HIF1α. PARP1 inhibition impairs angiogenesis and this can also explain part of the anticancer activity of PARPi as well as provide the rationale for combinations with antiangiogenic agents. 3. PARP1 also induces epithelial-to-mesenchymal transition (EMT) by upregulating vimentin and Snail1 and can induce metastasis by activation of ETS and NF-κB TFs; PARP inhibition can therefore inhibit EMT and metastasis. 4. Activation of PARP1 occurs by reactive oxygen species (ROS) and contributes to stimulation of inflammation via upregulation of NF-κB, NFAT, and AP-1 TFs. 5. Furthermore, PARP1 has been closely implicated with modulation of the immune response; beyond activation of NF-κB (in macrophages and dendritic cells) and of NFAT (T cells), PARP1 increases the expression of proinflammatory cytokines, IL2 and T helper type 2 cytokines, while it decreases Foxp3⁺ regulatory T cells (Treg). Due to these inflammatory and immunologic properties, PARPi have been proposed for the management of several acute and chronic inflammatory conditions, as well as to limit tissue damage during reperfusion in acute events, such as myocardial infarction (MI), circulatory shock, or brain stroke. AP-1, activator protein-1; ETS, E26 transformation-specific; Foxp3, forkhead box P3; H1, histone H1; HIF1a, hypoxia-inducible factor 1-alpha; IL2, interleukin 2; NF-κB, nuclear factor kappa-light-chain-enhancer of activated B cells; NFAT, nuclear factor of activated T cells; Th2, type 2 helper T cells.

Niraparib has an MTD of 300 mg once daily, and following repeated doses of niraparib, its mean half-life is 36.5 hours. Niraparib is metabolized by carboxylesterases to an inactive metabolite, which then undergoes glucuronidation differs from the other FDA-approved PARP inhibitors, olaparib and rucaparib, which instead are metabolized through CYP enzymes. Niraparib has an off-target effect of norepinephrine, epinephrine, and dopamine transporter interactions, which likely gives rise to the toxicities of hypertension, headaches, and palpitations.

In the NOVA study that led to the FDA maintenance approval, 553 patients with recurrent platinum-sensitive HGSC, sensitive to their penultimate platinum as well as their most recent platinum regimen, were randomized 2:1 to either niraparib or placebo at the completion of chemotherapy and were prospectively placed into one of two groups—gBRCAm or non-gBRCAm—based on upfront germline BRCA testing. Within the non-gBRCAm group, patients were further stratified by retrospectively performed HR deficiency (HRD) testing (10). Efficacy results were performed simultaneously for the gBRCA and non-gBRCA groups, and the primary endpoint was progression-free survival (PFS) measured from the time of treatment randomization to cancer progression or death from any cause. Independent radiologic review was used to define cancer progression. Niraparib significantly improved median PFS in all three primary efficacy groups, which included gBRCAm, overall non-gBRCAm and non-gBRCAm/HRD positive, as well as in the exploratory groups including somatic BRCAm/HRD positive, BRCA wild-type/HRD positive, and even in the HRD negative/BRCA wild-type group. Because of the observed myelosuppression of niraparib and especially the 33% risk of grade 3 and 4 thrombocytopenia observed in the NOVA study, the FDA package insert contains detailed recommendations for weekly blood count assessment for the first month after starting niraparib and then monthly thereafter, as well as specific dose modifications for myelosuppression management.

The FDA approval of niraparib has been transformative for many reasons; although the NOVA study eligibility excluded any cancers but HGSC, the FDA approval was histology nonspecific, thus allowing any patient with ovarian cancer irrespective of their cancer's histology to receive niraparib as long as there is demonstration of benefit from platinum. Platinum and PARP inhibitors have overlapping mechanisms of response and resistance, thus the use of a PARP inhibitor in the maintenance setting serves to improve remission length postresponse to platinum. In addition, the approval was based on PFS improvement, and for several reasons, survival benefit may not be observed in maintenance studies because of the availability of PARP inhibitors and eventual crossover of placebo patients to PARP inhibitor use (11).

Two other phase III PARP inhibitor maintenance studies have been published, SOLO2 and ARIEL3, both of which have led to FDA approvals of olaparib and rucaparib, respectively, as maintenance therapies, and clinicians in the United States now have three PARP inhibitors to select from in the maintenance setting (12, 13). The PFS benefits in the BRCAm and the HRD-positive groups in these phase III studies are remarkably similar, leading to the conclusion that these drugs have comparable efficacy. Besides PARP inhibitor toxicities that include gastrointestinal toxicity, bone marrow suppression, and some fatigue, an initial safety signal of early PARP inhibitor studies was an increased risk of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), but in these phase III maintenance studies, the MDS and AML risk seems quite comparable for the PARP inhibitor– and placebo-treated patients.

In addition, PARP inhibitor combinatorial strategies are actively being explored. Several rationales exist behind these combinations including overcoming PARP inhibitor resistance by addition of agents that inhibit HR in tumors with de novo or acquired HR proficiency or priming the immune system by PARP inhibitors to facilitate response to immune checkpoint blockade in both HR-deficient and HR-proficient tumors. Combinations that have reached phase III testing include olaparib and the antivascular agent cediranib; noteworthy phase II results (14) of efficacy in BRCA wild-type as well as BRCAm ovarian cancer have led to randomized studies in both BRCAm and BRCA wild-type recurrent cancer (NCT02446600 and NCT02502266). In addition, phase II studies of combined PARP inhibitor and immune checkpoint blockade, including niraparib and pembrolizumab (NCT02657889), in platinum-resistant ovarian cancer have also yielded notable early results, especially so because of the relatively low overall response rate of single-agent checkpoint blockade for recurrent ovarian cancer.

Finally, PARP inhibitors have activity beyond ovarian cancer; olaparib has shown impressive activity in both BRCAm breast cancer and prostate cancer. With the increased use of next-generation sequencing and identification of mutated genes involved in DNA repair, PARP inhibitors are poised for use as single agents in cancers with documented abnormalities in DNA repair. In addition, because of PARP enzymes' multiple functions in cancer cells, combination strategies with non-myelosuppressive biologic agents may offer patients with DNA repair–proficient cancers an opportunity to benefit from a PARP inhibitor.

Disclosure of Potential Conflicts of Interest

P.A. Konstantinopoulos is a consultant/advisory board member for AstraZeneca, Merck, and Pfizer. U.A. Matulonis is a consultant/advisory board member for 2X Oncology, Fujifilm, Geneos, Immunogen, Merck, Mersana, and Myriad Genetics.

Authors' Contributions

Conception and design: P.A. Konstantinopoulos, U.A. Matulonis

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.A. Konstantinopoulos, U.A. Matulonis

Writing, review, and/or revision of the manuscript: P.A. Konstantinopoulos, U.A. Matulonis

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): U.A. Matulonis