Prostate cancer is a clinically heterogeneous disease with many patients having an indolent course requiring no interventions and others who either present with or progress to metastasis. While underlying dominant driving mutations are not widespread, there have been a number of key genomic mutations that have been consistently identified in prostate cancer patients, across the disease spectrum including gene fusion/chromosomal rearrangements (TMPRSS2-ERG), androgen receptor (AR) amplification, inactivation of tumor suppressor genes (PTEN/PI3-K/AKT/mTOR, TP53, Rb1) and oncogene activation (c-MYC, RAS-RAF).1 More significantly, defects in DNA repair appear to be central in increasing one’s susceptibility to malignant transformation. Because cellular DNA is continually subject to damage, there are coordinated pathways designated for repairing DNA, maintaining genomic integrity, and ensuring cell survival. This process of DNA repair requires excision of the damaged DNA followed by repair via mechanisms including include homologous recombination repair (HRR), base excision repair (BER), nucleotide excision repair (NER) and mismatch repair (MMR). Repair itself may occur via two mechanisms, single (SSBR) or double stranded break repair (DSBR). While the details of these repair pathways are beyond the scope of this manuscript, it is valuable to understand that poly(adenosine diphosphate [ADP]-ribose) polymerase (PARP) enzyme and BRCA 1/2 (BReast CAncer gene 1 and 2) and ATM (Ataxia-Telangiesctasia Mutated) gene products play important roles in this process.2
Examining patients with advanced prostate cancer, Pritchard and colleagues were among the first to demonstrate the value of assessing inherited genetic changes. Among 692 patients with metastatic prostate cancer, they identified mutations in 20 DNA-repair genes in 82 men (11.8%),3 with significant geographic heterogeneity, even among these recognized cancer centers with a prevalence of 8.8% in patients treated at the University of Washington and 18.5% in patients treated at Memorial Sloan Kettering, potentially reflecting referral biases. In a similar study of patients with metastatic castrate resistance prostate cancer, Castro et al. found a prevalence of germline DNA damage repair gene mutations of 16.2%.4 In an analysis which spanned the disease spectrum, Nicolosi and colleagues found germline variants in 620 of 3607 patients (17.2%), of which BRCA1/2 comprised only a small proportion.5
In patients with metastatic castrate-resistant prostate cancer in the PROREPAIR-B cohort, Castro et al. found that germline BRCA2 mutations specifically were associated with significantly worse prostate cancer specific mortality (17 months compared with 33 months, hazard ratio 2.11, p-value = 0.033), though an aggregate assessment of ATM, BRCA1, BRCA2, or PALB2 was not associated with prostate cancer-specific survival (different in survival 10 months, p=0.264).4 However, while these mutations are associated with poor prognosis, they also offer potential therapeutic options.
Over the past decade, treatment of men with advanced prostate cancer has been revolutionized with numerous new treatment options for men with castration-sensitive metastatic prostate cancer, non-metastatic castrate-resistant prostate cancer, and metastatic castrate-resistant prostate cancer. Leading this process has been changes in therapeutic options for men with metastatic castrate-resistant prostate cancer. Coupled with these new therapeutic options is an increasing understanding of the heterogeneity in response between patients, with the potential to tailor therapy for patients most likely to benefit. Perhaps the strongest example of such tailed therapy in patients with metastatic castrate-resistant prostate cancer is the use of poly(adenosine diphosphate [ADP]-ribose) polymerase (PARP) inhibitors.
The PARPs are a family of enzymes (most abundantly PARP1) that excise bases, playing a key role in repair of DNA single strand breaks.6 PARP inhibition leads to an accumulation of DNA single strand breaks, leading to DNA double-strand breaks at replication forks that are normally repaired by the homologous recombination double stranded DNA repair pathway.7 Key components of this specific pathway are the tumor-suppressor proteins BRCA1 and BRCA 2.8 When cells carrying heterozygous loss-of-function BRCA mutations lose the remaining wild-type allele, there is deficient homologous-recombination DNA repair leading to carcinogenesis. These tumors carrying specific DNA-repair defects can be exploited using PARP inhibitors to induce selective tumor cytotoxicity and spare normal cells. PAPR inhibition in these homologous-recombination repair deficient cells leads to unrepaired DNA single-strand breaks, causing accumulation of DNA double-strand breaks.9,10 This has lead researchers to suggest the term “synthetic lethality” when there is a lethal synergy between two otherwise nonlethal events: a PARP inhibitor inducing a DNA lesion and a genetic loss of function for the homologous recombination DNA repair pathway required to fix it.10 In vitro studies have shown that BRCA1/2 deficient cells were 1000-fold more sensitive to PARP inhibition than wild-type cells.9
Early data for Olaparib
The first phase I trial utilizing the PARP inhibitor olaparib was performed more than a decade ago among a population of cancer patients enriched with carriers of a BRCA1 or BRCA2 mutation.11 Among 60 patients, 22 were BRCA1 or BRCA2 mutation carriers and one patient had a strong family history of BRCA-associated cancers. The olaparib dose and schedule were increased from 10 mg daily for two of every three weeks to 600 mg twice daily continuously. Reversible dose-limiting toxicity was seen in one of eight patients receiving 400 mg twice daily and two of five patients receiving 600 mg twice daily. Subsequently, another cohort was enrolled consisting only of carriers of a BRCA1 or BRCA2 mutation to receive olaparib at a dose of 200 mg twice daily. This phase I trial found objective antitumor activity was reported only in mutation carriers, all of whom had ovarian, breast, or prostate cancer and had received multiple treatment regimens. Subsequently, the TOPARP-A Trial demonstrated that treatment with the PARP inhibitor olaparib was associated with improvements in radiographic progression-free survival and overall survival, specifically among patients with extensively pre-treated mCRPC who had DNA-repair defects:12 16 of 49 patients (33%) had a response to olaparib, with 12 patients remaining on treatment for >6 months. TOPARP-B, was a phase II trial for patients with mCRPC preselected for putatively pathogenic DNA damage repair alterations.13 Because TOPARP-A used a 400 mg olaparib dose and patients with breast cancer typically use a 300 mg dose, patients in TOPARP-B were randomized 1:1 to 400mg or 300mg of olaparib BID, aiming to exclude ≤30% response rate using radiographic (RECIST 1.1), PSA (50% decrease) or CTC conversion criteria. A response was seen in 54% (95%CI 39-69%) of patients in the 400 mg cohort and 39% (95%CI 24-54%) in the 300 mg cohort. Over a median follow-up of 17.6 months, the overall median PFS (mPFS) was 5.4 months. Notably, response rates and mPFS were higher in patients with BRCA1/2 of 83% (mPFS 8.1 months) and PALB2 57% (mPFS 5.3 months).
PROfound: phase III data for Olaparib in mCRPC
On the basis of the above data, the PROfound study recruited men with metastatic castrate-resistant prostate cancer who had progressed on previous abiraterone acetate or enzalutamide administered at the time of non-metastatic castrate-resistant prostate cancer or at the time of metastatic castrate-sensitive prostate cancer14. Patients with prior taxane exposure were allowed. The investigators then used an investigational assay based on the FoundationOne CDx to identify alterations in one of 15 pre-specified genes involved in homologous recombination repair (BRCA 1/2, ATM, BRIP1, BARD1, CDK12, CHEK 1/2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D, RAD54L).
The authors then used biomarker driven stratification to derive two study cohorts: Cohort A had alterations in BRCA1, BRCA2, or ATM while Cohort B had alterations in any of the other 15 included genes. In both cohorts, patients were randomized 2:1 to olaparib vs. abiraterone or enzalutamide. Within each biomarker strata, randomization was stratified based on prior taxane use and measurable disease burden (according to RESIST 1.1 criteria).
Primary analysis was based on imaging-based progression free survival (soft-tissue according toe RESIST 1.1 and bony according to PCCTWG3 criteria) among patients in Cohort A. Secondary outcomes including PFS in the combined Cohort A+B, objective response rates, time to pain progression, overall survival, reduction in PSA >50%, CTC conversion rate, and safety/toxicity outcomes.
The authors utilized a hierarchical analysis strategy with sequential analysis of each endpoint on the basis of preceding outcomes.
Among 4425 patients who were screened, 4047 had sufficient tissue available for genetic testing of which 2972 had appropriate biomarker testing. 778 patients were identified as having at least one mutation in one of the 15 eligible genes. 387 of these patients met the remainder of eligibility criteria and were randomized.
In assessment of the primary outcome, the authors found significantly improved progression-free survival in patients with mutations of BRCA1, BRCA2, or ATM (hazard ratio 0.34, 95% confidence interval 0.25 to 0.47). Similar results were seen in the combined cohort (hazard ratio 0.49, 95% confidence interval 0.38 to 0.63). Subgroup analysis showed similar results when stratified according to prior taxane use, measurable disease at baseline, location of metastasis (bone only, visceral, or other), performance status, age at randomization, region, and baseline PSA (dichotomized around the median).
While data is immature (data maturity 38%), there is suggestion of improved overall survival in Cohort A (hazard ratio 0.64, 95% confidence interval 0.43 to 0.97, p=0.02 with an alpha significance threshold of 0.01 based on alpha-spending function).
Adverse events were common in both patients on olaparib (any = 95%, grade ≥ 3 = 51%) and in the control group (any = 88%, grade ≥ 3 = 38%).
While these results support the phase I/II data demonstrating the activity of olaparib in patients with mCRPC, particularly among patients with qualifying gene mutations, some have questioned these results, particularly as patients to be eligible must have progressed on abiraterone or enzalutamide (with approximately 20% having received both prior to randomization) and those who were randomized to the control arm received abiraterone or enzalutamide. Given the results of the CARD trial, patients progression following docetaxel and one androgen-axis inhibitor (abiraterone or enzalutamide) are likely to derive significantly greater benefit from cabazitaxel than a switch to a different androgen-axis inhibitor15.
On May 19, 2020, as a result of the data from the PROfound trial, the U.S. Food and Drug Administration approved olaparib for the treatment of patients with germline or somatic homologous recombination repair gene-mutated metastatic castrate resistant prostate cancer (mCRPC) who progressed following treatment with enzalutamide or abiraterone acetate.
Combination therapy with Olaparib:
Based on the initial success with PARP inhibitor monotherapy, there has been recent interest in combination therapy trials. The androgen receptor promotes DNA damage repair, whereas ADT upregulates PARP-mediated repair pathways with synthetic lethality between ADT and PARP inhibition. Furthermore, PARP1 also regulates AR-mediated transcriptional activation. As a result, there is a biologic rationale for combinations of androgen-axis targeting agents and PARP inhibitors. In 2018, results of phase 2 trial assessing olaparib with abiraterone were published in Lancet Oncology.16 There were 142 patients randomly assigned to receive olaparib and abiraterone (n=71) or placebo and abiraterone (n=71). The median rPFS was 13.8 months (95% CI 10.8-20.4) with olaparib and abiraterone and 8.2 months (5.5-9.7) with placebo and abiraterone (HR 0.65, 95% CI 0.44-0.97, p = 0.034). One treatment-related death (pneumonitis) occurred in the olaparib and abiraterone group. Based on the results of this phase 2 study, the ongoing PROpel phase 3 trial will evaluate olaparib + abiraterone in the first line mCRPC setting. Goal enrolment for PROpel is 720 patients, with an estimated trial completion of 2022.
Other PARP inhibitors in mCRPC:
In addition to olaparib, there are a number of other PARP inhibitors being assessed in patients with advanced prostate cancer. The TRITON2 trial assessed rucaparib 600 mg BID in patients with mCRPC associated homologous recombination repair gene alterations, initially presented at ESMO 2018.17 For the patients with BRCA1/2 alteration, there was a 44% confirmed overall response rate and 51% confirmed PSA response rate while patients harboring an ATM and CDK12 alteration did not receive significant benefit. On the basis of the data from TRITON2, on May 15, 2020, the US Food and Drug Administration (FDA) approved rucaparib for patients with mCRPC and BRCA mutations (germline or somatic) who had progressed following treatment with androgen-axis targeted treatment and taxane-based chemotherapy. The TRITON3 trial, discussed below, is set to serve as the confirmatory trial for this indication.
The GALAHAD trial was a phase II study of niraparib in patients with mCRPC and biallelic DNA-repair gene defects, initially presented at ASCO 2019.18 For this study, a patient’s plasma sample was evaluated to identify DNA repair defects, including mutations in BRCA1/2, ATM, FANCA, PALB2, CHEK2, BRIP1 or HDAC2. The composite response rate was 18/29 (62%) in the BRCA1/2 patients, and non-BRCA 5/21 (23.8%). Finally, veliparib has also been assessed among patients with previously treated mCRPC in a biomarker-stratified phase II trial: patients were randomized to abiraterone + veliparib versus abiraterone alone. The authors found no difference in terms of PSA response rate, objective response rate, or progression-free survival.19 Veliparib was also assessed in combination with temozolomide in a small pilot study among men with mCRPC.20
Ongoing trials of PARP inhibitors in mCRPC
As may be expected given the recent promising results of PARP inhibitor treatment in both monotherapy and combination therapy in patients with advanced prostate cancer.
- Olaparib: The PROpel trial is randomizing patients with mCRPC to first line treatment with Olaparib + abiraterone vs abiraterone alone, without biomarker selection. Further, KEYNOTE-365 Cohort A is testing pembrolizumab + Olaparib,21 while KEYLYNK-010 is randomizing 780 patients to pembrolizumab + olaparib versus abiraterone or enzalutamide in patients with mCRPC.
- Talazoparib: The TALAPRO-1 trial will assess talazoprabib in patients with progressive disease following both taxane and androgen axis inhibitor treatment (and biomarker selection for DDR mutations likely to sensitize to PARP inhibitor) while TALAPRO-2 will assess talazopraib + enzalutamide versus enzalutamide alone in first-line treatment of mCRPC without biomarker selection.
- Rucarparib: the TRITON3 trial is randomizing patients with mCRPC and mutations in BRCA1, BRCA2, or ATM to rucaparib or abiraterone/enzalutamide/docetaxel following progression on an androgen axis inhibitor.
- Niraparib: The MAGNITUDE trial is recruiting patients with mCRPC to first-line therapy with niraparib + abiraterone or abiraterone alone. Analysis is stratified by the presence (cohort A) or absence (cohort B) of DDR mutations.
Germline mutations in DNA-repair genes are relatively uncommon in patients with prostate cancer, though the prevalence increases among patients with advanced disease and somatic changes are also common in this disease. While such mutations are associated with a worse prognosis, they open the possibility of targeted therapy using PARP inhibitors. Phase I, II, and III data have now demonstrated benefit to the use of PARP inhibitors in patients with alterations in DNA repair genes, especially BRCA2. In the past week, two PARP inhibitors (olaparib and rucarparib) have received FDA approval for patients with advanced prostate cancer. Numerous ongoing studies will continue to refine the role of these agents, both alone and in combination, for patients with advanced prostate cancer.
Published June 29, 2020
Written by: Christopher J.D. Wallis, MD, PhD, Vanderbilt University Medical Center, Nashville, TN, and Zachary Klaassen, MD, MSc, Medical College of George, Augusta, GA