Therapeutic Potential of PARP Inhibitors in the Treatment of Metastatic Castration-Resistant Prostate Cancer

Simple Summary In recent years, the development of sequencing techniques to reveal the genomic information of prostate cancer tumors has allowed for the emergence of targeted therapies. Genomic aberrations in tumor cells have become popular due to the successful development of PARP inhibitors, which are particularly active in those tumors harboring DNA repair genomic defects. This review focuses on PARP inhibitors, two of which were approved for use by the US Food and Drug Administration in 2020 in metastatic castration-resistant prostate cancer. The article highlights the development of PARP inhibitors in the preclinical setting, summarizes the impactful clinical trials in the field, and discusses the need for continued research for further success in treating men with advanced prostate cancer. Abstract Metastatic castration-resistant prostate cancer (mCRPC) is an incurable malignancy with a poor prognosis. Up to 30% of patients with mCRPC have mutations in homologous recombination repair (HRR) genes. Poly (ADP-ribose) polymerase (PARP) inhibitors take advantage of HRR deficiency to kill tumor cells based on the concept of synthetic lethality. Several PARP inhibitors (PARPis) have been successful in various malignancies with HRR gene mutations including BRCA1/2, especially in breast cancer and ovarian cancer. More recently, olaparib and rucaparib were approved for mCRPC refractory to novel hormonal therapies, and other PARPis will likely follow. This article highlights the mechanism of action of PARPis at the cellular level, the preclinical data regarding a proposed mechanism of action and the effectiveness of PARPis in cancer cell lines and animal models. The article expands on the clinical development of PARPis in mCRPC, discusses potential biomarkers that may predict successful tumor control, and summarizes present and future clinical research on PARPis in the metastatic disease landscape.


Introduction
Metastatic castration-resistant prostate cancer (mCRPC) patients include men with distant metastases who have evidence of disease progression defined by either prostate-specific antigen (PSA) progression, new metastases, or clinical symptoms, despite a castrate level of testosterone [1]. While the success of androgen deprivation therapy (ADT) in suppressing tumor progression is almost form cytotoxic PARP-DNA complexes [23]. Different PARPis were found to have varying potency of trapping PARP-DNA complexes not related to their catalytic inhibitory properties. PARP inhibition leads to stalling of replication forks, inducing double-strand DNA breaks and the need for HR [24,25]. Tumor cells with two mutated BRCA genes are unable to start HR, which ultimately leads to cell death, while normal cells still possess at least one functioning BRCA gene to begin HR to repair DNA and prevent apoptosis. Cells that have either BER deficiency or HR deficiency may survive; if both pathways are deficient, the cells die. This is the concept for using PARPis to selectively target malignancies with BRCA1/2 germline or somatic mutations, which is termed "synthetic lethality." This model was supported by two landmark studies published in 2005-BRCA1/2 dysfunction sensitized tumor cells to PARPis in vitro, resulting in selective tumor cell death [26,27].

Preclinical Development of PARP Inhibitors
Olaparib (previously known as AZD2281 and KU-0059436) was discovered to be a potent orally bioavailable PARPi against BRCA-1/2 deficient cell lines. It was determined to have both good in vitro cellular potency (IC 50 for PARP-1 inhibition of 6 nM) and in vivo efficacy because it led to 80% tumor inhibition when fed in combination with temozolomide to mice [22]. Olaparib also had in vitro activity against ataxia telangiectasia mutated (ATM)-deficient lymphoid tumors and in vivo activity in mice xenografted with ATM-deficient mantle cell lymphoma tumor [28,29].
Rucaparib (PF01367338 and AG014699) was selected from a group of several compounds tested for a clinical trial based on excellent chemosensitization, radiosensitization, aqueous solubility, and safety profile [30]. Rucaparib was cytotoxic to human cancer cell lines with mutated BRCA1/2 and was then found to be effective in a wide range of ovarian cancer cell lines (including homologous recombination pathway deficiencies not including BRCA1/2 mutations) alone and in combination with other agents [31,32].
Research has shed some light on how PARPs are involved in prostate cancer and how PARPis are beneficial against prostate cancer, although the available literature is scant. The erythroblast transformation specific (ETS) gene fusion family is involved in the progression of a variety of cancers, and TMPRSS2:ERG is a prostate cancer-specific gene fusion; it has been noted that this gene fusion product interacts with PARP-1, and that olaparib was able to suppress ETS-positive (but not ETS-negative) prostate cancer cell invasion [37], yet no clinical data confirming this cellular effect is available. In fact, in a clinical trial with a subset of mCRPC patients who received veliparib and abiraterone, there was no difference in response rate between patients with or without ETS fusions [38]. The combination of rucaparib with radiation therapy was synergistic for prostate cancer cells expressing the TMPRSS2:ERG gene fusion, as these cells showed enhanced sensitivity towards rucaparib, which increased radiation response [39]. Schiewer et al. [40] demonstrated in prostate cancer cells that PARP-1 modulated both androgen-receptor function and response to DNA damage, suggesting its involvement in prostate cancer progression and maintenance of castration resistance; in this study, olaparib and veliparib successfully decreased androgen-receptor target gene expression and tumor growth using murine models and ex vivo prostate tumor cultures.

Olaparib
The first clinical trial of olaparib (NCT00516373) included three patients with advanced prostate cancer, including one patient with a BRCA2 mutation who had a greater than 50% reduction in PSA level and resolution of bone metastases [41]. The open-label phase II trial (NCT01078662) by Kaufman et al. [42] included eight patients with mCRPC (one with BRCA1 mutation, seven with BRCA2 mutation) and a median of two prior therapies. Median total duration of olaparib treatment was 223.5 days, given at 400 mg twice daily. The one BRCA1 mutation patient and three of the seven BRCA2 mutation patients responded to the treatment. Two of the eight prostate cancer patients had stable disease that persisted for at least eight weeks. Median duration of response was 327 days, and median time to onset of response was 54.5 days. Progression-free survival (PFS) at 6 months was 62.5%, median overall survival was 18.4 months, and 50.0% of the patients were alive at 12 months.
Further encouraging efficacy data was reported in the open-label Phase II Trial of Olaparib in Patients with Advanced Castration Resistant Prostate Cancer (TOPARP-A trial, NCT01682772), which focused solely on mCRPC [43]. The primary endpoint was based on objective response rate (ORR) according to Response Evaluation Criteria in Solid Tumors (RECIST), PSA decline, or reduction in circulating tumor-cell count. The trial enrolled 50 patients, although one was lost to follow-up early and was not included in the later analysis. All 49 patients had previously received treatment with other agents, including docetaxel, abiraterone, enzalutamide, and cabazitaxel. Next-generation sequencing data revealed that 16 patients had tumor aberrations in DNA-repair genes, including BRCA1, BRCA2, ATM, FANCA, PALB2, HDAC2, and CHEK2. Median duration of olaparib was 12 weeks. Patients with these DNA-repair gene aberrations had a significantly better response to olaparib of 400 mg twice daily (14/16 compared to 2/33 biomarker-negative patients). All seven patients with BRCA2 loss had PSA levels decrease by over 50% from baseline, and the five who had measurable disease had radiologic partial response. Four of five patients with ATM mutations had response as well. The results in the TOPARP-A trial led the FDA in January 2016 to give olaparib a breakthrough therapy designation for mCRPC with BRCA1/2 or ATM mutations previously treated with a taxane-based chemotherapy and either enzalutamide or abiraterone.
On a continuum, TOPARB-B was an open-label phase II trial that included 98 mCRPC patients with known aberrations to 18 DNA-repair genes who received olaparib: 49 received 300 mg twice daily and 49 received 400 mg twice daily [44]. Overall, 43 of the 98 patients achieved a confirmed composite response. Patients with the BRCA1/2 mutation had the best response and longest median radiographic progression-free survival (rPFS) compared to patients with the 16 other gene mutations, but because olaparib did have an effect on these other mutations as well, this study supported the genomic stratification of mCRPC and olaparib's potential in mCRPC patients with tumor gene aberrations besides the BRCA1/2 mutation.
These data leveraged conducting a confirmatory, phase III trial, PROfound (NCT02987543). This prospective, randomized, open-label study evaluated the efficacy and safety of 300 mg of olaparib twice daily versus 160 mg of enzalutamide daily or 1000 mg of abiraterone daily in 387 patients with mCRPC and 15 HRR gene alterations (BRCA1, BRCA2, ATM, BRIP1, BARD1, CDK12, CHEK1, CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D or RAD54L) [45]. The patients had all been treated previously with enzalutamide or abiraterone, and some patients had also been treated previously with taxane chemotherapy. Cohort A included 245 patients (162 received olaparib and 83 received the control treatment) with at least one alteration in BRCA1, BRCA2, or ATM, while cohort B included 142 patients (94 received olaparib and 48 received the control treatment) with the 12 other gene alterations. The primary endpoint was imaging-based PFS in cohort A. In cohort A, the olaparib group was better than the control group in terms of significantly increased median imaging-based PFS (7.4 vs. 3 In cohort A, more adverse events (AEs), including grade 3 or higher, were noted in the olaparib group compared to the control group in the PROfound trial, and in line with the known safety profile of PARP inhibitors [46,47]. While common AEs included anemia, nausea and decreased appetite, serious side effects associated with olaparib include the development of myelodysplastic syndrome, acute myeloid leukemia, and pneumonitis [48]. Thus, it is often helpful to obtain a complete blood count at baseline and then monthly to monitor for clinically significant changes.
This first biomarker-selected mCRPC study led the FDA in May 2020 to approve the use of olaparib for patients with mCRPC and HRR gene mutations who progressed despite previous treatment on enzalutamide or abiraterone [49]. Foundation-One was the approved companion diagnostic test, but other tissue and circulating tumor DNA assays are commercially available and future validation studies will clarify their role in identifying these biomarkers.
The efficacy of olaparib monotherapy in mCRPC patients without HRR mutations is under investigation. In a double-blinded, randomized, placebo-controlled phase II trial of mCRPC not required to have an HRR mutation (NCT0197221), 71 patients who received 300 mg of olaparib twice daily combined with 1000 mg of abiraterone daily were compared to 71 patients who received abiraterone and placebo [50]. Median rPFS was 13.8 months for olaparib and abiraterone compared to 8.2 months for abiraterone alone (HR 0.65, 95% CI, 0.44 to 0.97, p = 0.034), suggesting possible benefits for olaparib in mCRPC patients without HRR mutations. The ongoing phase III trial PROPEL (NCT03732820) builds on the success of the phase II trial that used the combination of olaparib and abiraterone in genomic unselected mCRPC patients. The primary outcome measure is rPFS, with expected completion in 2021.

Rucaparib
Rucaparib was granted accelerated approval by the FDA in May 2020 for treatment of mCRPC with a deleterious germline or somatic BRCA mutation previously treated with androgen-receptor-directed therapy and a taxane-based therapy, based on data from TRITON-2 [51]. The Trial of Rucaparib in Prostate Indications (TRITON)-2 is an open-label phase II trial (NCT02952534) evaluating 600 mg of rucaparib twice daily (with gonadotropin-releasing hormone [GnRH] analogue or prior bilateral orchiectomy) in mCRPC patients who progressed on androgen-deprivation therapy and one prior taxane-based chemotherapy with a deleterious germline or somatic alteration in BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK2, FANCA, NBN, PALB2, RAD51, RAD51B, RAD51C, RAD51D, or RAD54L. Of 62 BRCA-mutated patients, 27 had a confirmed ORR, and 15 of these 27 had a response duration of at least six months. TRITON-3 is an actively recruiting, randomized, open-label phase III trial (NCT02975934) studying rucaparib 600 mg twice daily versus either abiraterone, enzalutamide, or docetaxel in patients with mCRPC and a deleterious germline or somatic mutation in BRCA1, BRCA2, or ATM that progressed on androgen-receptor signaling-directed therapy, building on the success of TRITON-2.

Other PARP Inhibitors in Prostate Cancer
Niraparib is being evaluated in the open-label phase II trial GALAHAD (NCT02854436) in mCRPC patients with gene alterations in BRCA1, BRCA2, ATM, FANCA, PALB2, CHEK2, BRIP1, or HDAC2 who progressed despite androgen-receptor-targeted therapy and taxane-based chemotherapy [52]. At the 2019 European Society for Medical Oncology (ESMO) Congress, data presented on 81 patients (46 BRCA and 35 non-BRCA) showed better performance for BRCA patients, who had a 41% objective response rate, 63% complete response rate, median rPFS of 8.2 months, and overall survival of 12.6 months; for non-BRCA patients, the numbers were 9%, 16%, 5.3 months, and 14.0 months, respectively. This led the FDA in October 2019 to give niraparib a breakthrough designation as therapy for BRCA1/2-mutant positive mCRPC [53].

Looking into the Future: Potential Biomarkers of Response to PARP Inhibitors, Cautious Optimism, and Ongoing Clinical Trials
BRCA1 and BRCA2 are logical candidates to be biomarkers of response to PARPis, based on the current knowledge of DNA damage repair with wild-type BRCA1 and BRCA2 part of the complex in homologous recombination to fix double-strand breaks [55]. However, recent data have sparked debate over just how predictive BRCA1/2-mutated cancers can be to PARPis, especially in non-BRCA-associated cancer types (cancers not including breast, ovary, prostate, or pancreatic cancer) [56]. Response to PARPis in different BRCA1/2-associated cancers varies widely, and BRCA1/2-mutations are not synonymous with HRR deficiency, as other secondary somatic mutations may restore or bypass BRCA function [56,57].
Cyclin-dependent kinase 12 (CDK12) phosphorylates the C-terminal domain of RNA polymerase II, which ensures several functions, including optimal transcription elongation, translation of a subset of human protein-coding genes, and maintenance of genomic stability [58]. A genome-wide synthetic lethal screen involving ovarian cancer cell lines and olaparib determined that CDK12 deficiency may confer sensitivity to PARPis [59]. However, as more clinical data become available, CDK12 mutations in prostate cancer appear to minimally respond to PARPis. In the TRITON2 trial, CDK12 mutated mCRPC patients had dismal response rates to rucaparib (0/10 with a radiographic response, 1/15 with a PSA response), and this cohort was discontinued [60]. Antonarakis et al. [61] revealed early results of a multi-institution retrospective study of 60 men with CDK12-altered prostate cancers that showed 0 of the 11 who received PARPis (10 olaparib, 1 rucaparib) had a PSA response.
ATM (ataxia-telangiectasia mutated) is a phosphatidylinositol-3 related kinase involved in DNA double-strand break repair that generates signaling networks for DNA repair proteins [62]. Response rates to PARPis in ATM-deficient tumor cell lines were seen in chronic lymphocytic leukemia [28], gastric cancer [63], and mantle cell lymphoma [64]. However, other studies revealed ATM deficiency may not be enough to fully sensitize these cells to PARP inhibition; experiments showed olaparib given alone to ATM-deficient cancer cells induced only a cytostatic state, while there are emerging data suggesting that olaparib combined with an ATR (ATM-and RAD3-related) inhibitor provides an additional cytotoxic effect [65,66]. A study using prostate cancer cell lines agreed that ATM loss may not respond to PARPis, but they did respond well to an ATR inhibitor [67].
In the PROFOUND trial, the hazard ratio for progression or death of mCRPC patients with ATM mutation was 1.04 (95% CI of 0.61-1.87) [45]. In the TRITON2 trial, only 2 of 19 mCRPC patients with ATM mutation receiving rucaparib had a radiographic response and 2 of 49 patients had a PSA response [60]. Taken together, cancers with ATM aberrations may rarely respond to PARPis, and the response is, in general, more limited compared with the activity in BRCA1/2 tumors.
Other DNA repair genes such as PALB2, FANCA and the RAD51 family are currently being evaluated in several of the above-mentioned trials, although the number of patients with these mutations enrolled is relatively low. Preclinical data using cell lines have suggested synergy of PARPis with these impaired DNA repair proteins, such as FANCA [68], RAD51C [69], or MRE11 [70]. Limited clinical data exist to make definitive conclusions about the effectiveness of PARPis in patients with these mutations [71]. Notably, prostate cancer patients with PALB2 mutations receiving different PARPis have demonstrated antitumor activity. More robust prospective studies must be done to better determine the reliability of these biomarkers for PARPis in mCRPC.
Overall success of PARPis in mCRPC must still be interpreted cautiously, as the response to patients harboring various mutations is still variable, and eventually patients experience disease progression after prolonged administration of PARPis. Multiple proposed mechanisms of PARPi resistance include restoration of HRR, DNA replication fork protection, reversion mutations, epigenetic modifications, and restoration of PARylation [72]. Resistance to PARPis was already proposed and demonstrated very early in the preclinical setting using a PARPi-resistant pancreatic cancer cell line with the intragenic deletion of c.6174delT of BRCA2 [73]. Because of these possibilities, identifying patients with mCRPC who may develop resistance to PARPis would be helpful. For example, Quigley et al. [74] detected BRCA2 reversion mutations associated with olaparib and talazoparib resistance in mCRPC patients through analysis of circulating cell-free DNA. There continue to be several active clinical trials for mCRPC patients at various stages involving different PARPis, either as monotherapy (Table 1) or combined with other therapies to enhance success ( Table 2). These trials aim to further improve upon the progress so far with PARPis against this incurable malignancy.

Conclusions
The success of PARPis in treating cancer points to the importance of understanding the molecular phenotype of mCRPC, the therapeutic implications of genomic information, and the potential of precision oncology. Olaparib and rucaparib are now available for mCRPC, and other PARPis are likely to be approved soon, based on several ongoing studies. More studies are required to determine the full benefit of these agents, including use in earlier stages of the disease, identification of further predictive biomarkers and evaluation of synergism when combined with other agents.