The Resounding Impact of DNA Repair Deficiency in Prostate Cancer

Abstract

An estimated one fifth or more of metastatic castration resistant prostate cancer (mCRPC) harbor defects in genes involved in DNA repair pathway (e.g. BRCA2, BRCA1 and others). Early evidence suggests these alterations may be predictive of therapeutic response to PARP inhibitors and platinum chemotherapy, thought to reflect principles of synthetic lethality and are currently being investigated in an increasing number of prospective clinical trials. Other studies have examined these alterations as prognostic biomarkers and in association with response to currently available treatments. A smaller fraction of men (5–10%) with mCRPC have evidence of microsatellite instability and defects in the DNA mismatch repair pathway, which may predict therapeutic response to immune checkpoint inhibitors. Loss of function of these two critical DNA repair pathways serves as new candidate predictive biomarkers for treatment strategies that represent net gains in the treatment toolbox for prostate cancer. Additionally, more than one tenth of men with mCRPC carry genetic alterations of DNA repair in their germline DNA, which may indicate high- to moderate-penetrance heritable cancer risk and have important implications for family members. In some cases, cascade genetic testing of family will direct modified strategies for screening and prevention of multiple cancers. Further study in each of these arenas is ongoing, although the potential for resounding impact is clear.

NEW PRECISION TARGETS

The prostate cancer research field has recently witnessed a series of important discoveries revolving around genes critical to DNA repair and is beginning to harness the potential of these findings in earnest. In 2015, the Cancer Genome Atlas Research Network reported findings from 333 primary prostate cancers and the identification of 19% of primary tumors with mutations in DNA repair genes, including 3% in the homologous recombination repair gene, BRCA2¹. In the same year, the International SU2C/PCF/AACR Prostate Cancer Dream Team applied exome sequencing to 150 metastatic biopsies and found approximately 20% of metastatic prostate cancers with alterations in genes critical to DNA repair, notably involving homologous recombination repair (BRCA2, ATM and BRCA1) as well as mismatch repair (MLH1 and MSH2)². Other studies have validated the high prevalence of DNA repair alterations in metastatic castration resistant prostate cancers (mCRPC)^3,4.

Deficiencies in specific DNA repair pathways have been characterized in other cancers where treatments exploiting these deficiencies using principles of synthetic lethality are used (Figure 1). The basic rationale is that cancers with specific inactivation of one of a number of DNA repair pathways will render the cancer more reliant on the remaining intact pathways. However, drugs such as PARP inhibitors can inactivate one of the remaining DNA repair pathways, which is lethal in the cancer (where there is insufficient DNA repair capacity to compensate), or is relatively non-toxic in non-cancerous cells (where there is sufficient remaining intact DNA repair capacity to compensate). These discoveries represent major strides in the field, as they represent molecular subsets that may benefit from precision therapies and thus are candidate predictive biomarkers.

Figure 1.

Schematic overview of major types of DNA damage, along with the corresponding DNA damage repair mechanisms, key genetic drivers and treatment implications. Boxes denote components with particular new relevance to advanced prostate cancer.

HOMOLOGOUS RECOMBINATION DEFICIENCY

Following on the discovery that a significant proportion of metastatic castration resistant prostate cancer harbors defects in DNA repair was early, biologically plausible evidence of clear therapeutic consequence--that not only PARP inhibitors but also platinum chemotherapy may have particular efficacy in prostate cancer with homologous recombination repair deficiency^5–7: the TOPARP-A study showed that 14/16 (88%) of heavily pre-treated patients with metastatic castration resistant prostate cancer (mCRPC) who had defects in DNA repair genes had a response to the poly ADP ribosylase (PARP) inhibitor olaparib⁵. There are now a substantial number of clinical trials underway investigating PARP inhibitors in metastatic castration resistant prostate cancer with evidence of homologous DNA repair inactivation (Table 1).

Table 1.

Selected trials in prostate cancers related to DNA repair

Phase	Title	Disease	State	Clinicaltrials.gov
III	Study of Olaparib (Lynparza™) Versus Enzalutamide or Abiraterone Acetate in Men With Metastatic Castration-Resistant Prostate Cancer (PROfound Study)	mCRPC	PROFOUND	NCT02987543
III	A Study of Rucaparib Verses Physician's Choice of Therapy in Patients With Metastatic Castration-resistant Prostate Cancer and Homologous Recombination Gene Deficiency (TRITON3)	mCRPC	TRITON3	NCT02975934
II	A Phase 2 Efficacy and Safety Study of Niraparib in Men with Metastatic Castration-Resistant Prostate Cancer and DNA-Repair Anomalies	mCRPC	GALAHAD	NCT02854436
II	A Multicenter, Open-label Phase 2 Study of Rucaparib in Patients With Metastatic Castration-resistant Prostate Cancer Associated With Homologous Recombination Deficiency	mCRPC	TRITON2	NCT02952534
II	Response Rate Study of Talazoparib in Men With DNA Repair Defects and Metastatic Castration-Resistant Prostate Cancer Who Previously Received Taxane-Based Chemotherapy and Progressed on at Least 1 Novel Hormonal Agent (Enzalutamide and/or Abiraterone Acetate/Prednisone)	mCRPC		NCT03148795
II	Olaparib in Men With High-Risk Biochemically-Recurrent Prostate Cancer Following Radical Prostatectomy, With Integrated Biomarker Analysis	BCR		NCT03047135
II	Abiraterone/Prednisone, Olaparib, or Abiraterone/Prednisone + Olaparib in Patients With Metastatic Castration-Resistant Prostate Cancer With DNA Repair Defects	mCRPC	BRCAaway	NCT03012321
I	A Safety and Pharmacokinetics Study of Niraparib Plus an Androgen Receptor-Targeted Therapy in Men With Metastatic Castration-Resistant Prostate Cancer (BEDIVERE)	mCRPC	BEDIVERE	NCT02924766
pilot	Docetaxel and Carboplatin in Treating Patients With Metastatic, Hormone Resistant Prostate Cancer Containing Inactivated Genes in the BRCA 1/2 Pathway	mCRPC	ABCD	NCT02598895
II	The BARCODE 2 Study - The Use of Genetic Profiling to Guide Prostate Cancer Treatment (BARCODE2)	mCRPC	BARCODE-2	NCT02955082
II	Docetaxel and Carboplatin for Patients With mCRPC and DNA-Repair Deficiencies	mCRPC	V-ABCD	NCT02985021
II	Pembrolizumab in Treating Patients With Metastatic Castration Resistant Prostate Cancer Previously Treated With Enzalutamide	mCRPC		NCT02312557
II	Study of Pembrolizumab (MK-3475) in Participants With Metastatic Castration-Resistant Prostate Cancer	mCRPC	KEYNOTE-199	NCT02787005
Ib/II	Study of Pembrolizumab (MK-3475) Combination Therapies in Metastatic Castration-Resistant Prostate Cancer	mCRPC	KEYNOTE-365	NCT02861573

In addition, several reports show improved response to platinum chemotherapy in patients with metastatic prostate cancer containing inactivating mutations of BRCA2, germline and somatic-only. In one case series, 3 heavily pre-treated patients observed to have exceptional responses to addition of carboplatin chemotherapy (up to 30 months of PSA-progression-free survival) underwent tumor next generation sequencing and all were found to have evidence of biallelic inactivation of BRCA2. Notably, 2 patients had underlying germline BRCA2 pathogenic mutations⁶. In another single-institutional study of 141 men treated with at least 2 cycles of carboplatin and docetaxel for mCRPC, pathogenic germline BRCA2 variants were observed in 8/141 men (5.7%; 95% confidence interval, 2.5%-10.9%). Six of 8 BRCA2 mutation carriers (75%) experienced PSA declines >50% within 12 weeks, compared with 23 of 133 non-carriers (17%; absolute difference, 58%; 95% confidence interval, 27%-88%; P < .001)⁸. There are also prospective clinical trials underway examining treatment with platinum chemotherapy for patients with mCRPC (Table 1).

MISMATCH REPAIR DEFICIENCY and MICROSATELLITE INSTABILITY

In May of 2017, the U.S. Food and Drug Administration granted approval to pembrolizumab for patients with unresectable or metastatic solid tumors with progression or no alternative treatments that have microsatellite-instability (MSI) or mismatch repair deficient (MMR) agnostic of tissue/site indication—in this regard, the first FDA-approval of its kind. A small subset of advanced prostate cancers have been reported to have complex MSH2 or MSH6 structural rearrangements resulting in hypermutation^2,9. Preliminary findings from a combination study adding pembrolizumab at time of resistance to enzalutamide revealed one exceptional responder with evidence of tumor microsatellite instability (www.clinicaltrials.gov; NCT02312557)¹⁰. Thus, MSI/MMR suggests that immune checkpoint inhibition will be another precision-guided treatment avenue for a subset of men with advanced prostate cancer. It remains to be determined whether the combination of enzalutamide and pembrolizumab exerts a different effect on prostate cancer compared to monotherapy with pembrolizumab alone, and whether there are therapeutic differences between different timing and sequences of combinations. Other novel approaches include combination of immune checkpoint inhibitors and PARPi, with the goal of forcing errors in tumor replication and potentially rendering greater tumor visibility to the immune system.

THERAPEUTIC OPPORTUNITIES

These findings have proven exhilarating to the prostate cancer field that has eagerly awaited delivery on the promise of precision oncology: we now have new candidate predictive biomarkers (homologous recombination deficiency, mismatch repair deficiency/microsatellite instability) for new treatments and a clear view of net gains in the therapeutic toolbox. Currently, clinical trials incorporating these targeted agents alone and in combination with other agents for prostate cancer are ongoing and in development (Table 1, www.clinicaltrials.gov).

MECHANISMS OF RESISTANCE

Synthetic lethality in homologous recombination deficient cancers has been extensively explored in ovarian and breast cancers, and some mechanisms of resistance to PARPi and platinum have been described, including reversion mutations (i.e. secondary mutations restoring open reading frames)^11–14. Indeed, several reports in prostate cancer patients have already described reversion mutations of mutations in BRCA2 and PALB2 in the plasma ctDNA as a mechanism of resistance to PARPi and platinum ^15,16 (Cheng, et al. JCO PO, in press; Carneiro, et al, JCO PO, in press). These reports are sobering in that, as with AR-targeting agents, the drive to resistance is a persistent force though the ability to detect resistant clones early may offer opportunities to triage to clinical trials to prevent development of resistance, such as through combination approaches.

GERMLINE IMPLICATIONS

There is little question that the new relevance of DNA repair deficiency in advanced prostate cancer has led to an abundance of new opportunities and therapeutic directions. In 2016, a dedicated germline study of nearly 700 men with metastatic prostate cancer found 11.8% harboring pathogenic germline mutations associated with high- to moderate-penetrance cancer predisposition, including BRCA2, BRCA1 along with a number of other less common DNA repair genes newly associated with prostate cancer¹⁷. This discovery that approximately half of these treatment-actionable genetic alterations lie in the germline DNA (and are therefore heritable) may be an equally ripe opportunity for the field.

We have long recognized that family history of prostate cancer is a major risk factor for developing prostate cancer. The genetic risk is comprised of a combination of common risk and modifying alleles as well as, in some men, relatively rare mutations in moderate-high penetrance cancer risk genes such as BRCA2. Both are important, but knowledge of moderate-high penetrance genes carries management recommendations for carriers who may be at increased risk of multiple cancers (NCCN Clinical Practice Guidelines for Genetic/Familial High-Risk Assessment: Breast and Ovarian, Genetic/Familial High-Risk Assessment: Colorectal.)

A number of groups have now demonstrated the clear limitations of prior criteria (largely family history of cancer) for identification of men with prostate cancer and single-gene, high- to moderate-penetrance cancer predisposition (e.g. BRCA2, BRCA1, etc)^17,18. Moreover, germline carriers of BRCA2 pathogenic mutations have worse prostate-cancer specific outcomes^19–21. Germline mutation carriers who are at risk for prostate cancer may be candidates for modified cancer screening^22,23, and those with localized prostate cancer might be considered for clinical trials of treatment intensification to improve outcomes.

The identification of carriers of rare germline pathogenic mutation carriers among those diagnosed with metastatic prostate cancer can also facilitate cascade genetic testing (genetic testing of at-risk family members where each first degree relative has a 50% chance of inheriting the same risk variant), and with it more informed and tailored cancer risk management in these relatives. Alternate approaches are being explored in universal screening of colon and endometrial cancer patients for Lynch Syndrome ²⁴, and similarly in Hereditary Breast and Ovarian Cancer syndrome (HBOC)^25,26.

Concerted efforts to determine best approaches to implementation of cascade genetic testing of family members are underway, and more work is needed to understand newer, less characterized variants and genes with respect to clarification of prostate cancer risk, consequent measures for early detection and prevention of not only prostate, but potentially also breast, ovarian, colon and endometrial cancers, among others. New dedicated clinics and clinical trials are opening to address delivery, characterize risk and improve management approaches.

SUMMARY

Many new investigative areas and opportunity have arisen from recent discoveries around DNA repair in advanced prostate cancer: precision therapy opportunities (PARPi, platinum, immune-checkpoint inhibitors and new combination approaches), liquid tumor biopsies in the form of circulating tumor cells and cell-free circulating tumor DNA, research and clinical considerations around pathogenic variants in DNA repair genes that could translate to improved management of the next generation (cancer risk management in family members). The full extent of impact will not be realized for a long time to come, but will undoubtedly lead to better outcomes for prostate cancer patients and their families.