DNA-Damage-Repair Gene Alterations in Genitourinary Malignancies

Naturally occurring DNA damage arises continuously during cell life, leading to single-strand DNA breaks (SSBs) or DNA double-strand breaks (DSBs). High-fidelity and timely repair of SSBs and DSBs is therefore necessary for maintaining genomic integrity and cell survival by minimizing mutagenesis or improper apoptosis, which may otherwise lead to carcinogenesis [1]. Thus, DNA damage repair (DDR) alterations are commonly identified in various malignancies, including genitourinary cancers. We aimed to describe herein the known evidence of DDR alterations in prostate cancer (PCa), urothelial, and renal cell carcinoma (RCC) and its therapeutic implications.

DSBs are a highly toxic form of DNA damage, arising from intrinsic and extrinsic sources [2]. Eukaryotes are equipped with several mechanisms to repair DSBs, including nonhomologous end-joining, microhomology-mediated end-joining, or homologous recombination (HR) repair (HRR), among others [3]. HR is a tightly regulated molecular mechanism essential for maintaining genomic stability by repairing DSBs or lesions that stall DNA replication forks. Its main pathway involves serine/threonine kinase ATM which activates several targets, including BRCA1, through CHK2 protein [4]. BRCA2 and RAD51 will form a complex with PALB2 after activation by BRCA1, subsequently leading to G2-M checkpoint arrest [5, 6]. The role of those proteins can be divided between sensors of the DSBs, such as ATM or CHK2, and effectors of the reparation, such as BRCA1/2 or PALB2.

In HR-proficient cells, broken DNA is accurately restored by the use of the respective sequence from second chromosome during the cell cycle phases of late S to G2. On the contrary, DSBs occurring in HR-deficient (HRD) cells lead to cell cycle arrest and cell death.

SSBs repair relies on base excision repair (BER) through poly (ADP-ribose) polymerase (PARP) enzyme, nucleic acid excision repair through xeroderma pigmentosum enzyme and polymerases, and mismatch repair (MMR) through MLH1 and MSH2. PARP also plays a role in transcriptional regulation of various genes, including androgen receptors (ARs) [7].

PARP inhibitors (PARPi) are targeted therapies inducing both a catalytic inhibition of PARP enzyme and a trapping of PARP1 and 2 at the same time, avoiding SSB reparation by preventing BER enzyme to access DNA. The trapping efficiency differs between the different PARPi, talazoparib having the greater stereospecific PARP-DNA trapping ability than other PARPi [8].

PARPi exploit defects in the DNA repair pathway through synthetic lethality. Indeed, if PARylation is inhibited and the BER pathway is impaired, SSBs persist and cause stalled replication forks and eventually DSBs. At this step, DSBs will be repaired with the HRR system in HR-proficient but not HRD cells [9].

Castro et al. [10] and others have described and characterized BRCA2 as an important prognostic factor for aggressive PCa; however, the mutation frequency was low, and most estimates suggested that BRCA2 accounted for a very small fraction of PCa (1–2%), even when early onset family history-positive cases were examined [11, 12]. Of note, familial forms of PCa are observed in 20% of the cases. In a seminal paper, Robinson et al. [13] identified mutations in three DNA repair genes, BRCA1/2 and ATM, at a surprisingly high rate in men unselected for age at diagnosis or family history, but rather for aggressive disease.

More recently, Pritchard et al. [13, 14] demonstrated an elevated rate of mutations (11%) in a number of DNA repair genes in men with metastatic PCa (with 5.3% mutations of BRCA2), with more than 100 genetic pathogenic variants identified. In the most advanced setting of metastatic castration-resistant PCa (mCRPC), mutations in HRR pathway are observed in up to 27% of mCRPC, mainly involving BRCA2 (12–18%), ATM (3–6%), CHEK2 (2–5%), and BRCA1 (<2%). Overall, germinal alterations are present in roughly 30–50% of cases of HRR alterations [14, 15] with reported discrepancies due to the existence of bi-allelic somatic alterations, and the lack of constitutional testing in most studies.

Relative risk (RR) of PCa in germinal BRCA2 alteration carriers is 4.65-fold higher compared to noncarriers, with an RR of 2.5–8.6 [16]. These alterations are associated with higher stage and grade at diagnosis and poorer cancer-specific survival when BRCA2 mutation is located outside the ovarian cancer cluster region [17‒19]. Risk for PCa is also slightly higher in BRCA1 alteration carriers as compared with noncarriers (RR 1.35) but without adverse features.

These results led to recommend digital rectal examination along with PSA testing from the age of 40 years old for patients with BRCA2 mutation in the French Guidelines [20], based on the results of the IMPACT study [21]. Nevertheless, as shown by Shuai Li et al. [22], cumulative risk for PCa in BRCA2 mutation carriers is mainly driven by the overall incidence of PCa in elderly, suggesting that early screening at 40 years old is not supported by robust data. Of note, BRCA2 carriers were at increased risk of other cancers such as male breast (RR 44 vs. noncarriers), stomach (RR 6.39), and pancreatic (RR 3.34) cancers. In the NCCN guidelines updates, genetic counseling is recommended in patients with aggressive disease (≥T3a, intraductal, ≥Gleason 8, positive lymph nodes), Ashkenazi Jewish Ancestry, ≥2 relatives with cancer in the spectrum of hereditary breast-ovarian cancer or Lynch syndrome.

Intensification of curative options, such as extended androgen deprivation therapy, in localized PCa with DDR mutations is not supported by any robust evidence to date, although BRCA mutation has been confirmed as an independent prognostic factor for metastasis-free survival (HR 2.36) and cancer-specific survival (HR 2.17) [23]. To our knowledge, no phase 3 study has assessed efficacy of PARPi in localized PCa in combination with surgery, radiation therapy, and androgen deprivation therapy.

If active surveillance is proposed to BRCA2 alteration carriers with low-risk PCa, the patient should be well informed that no clear guidelines of specific follow-up in this setting are available and that recent publications found a higher risk of grade reclassification in progressors [24]. These recent data suggest that the natural history of BRCA2-mutated PCa is probably accelerated even in the localized setting.

Somatic HRR alterations seem to occur early in PCa development, with genetic alterations being stable over time [25, 26]. Therefore, HRR mutations may be detected in archived primary tumor biopsies or specimen without the need to repeat testing on metastatic tissue samples.

The ESMO provided guidance for precision medicine, and every patient with metastatic PCa should be at least tested for somatic alterations of BRCA1/2 [27]. The NCCN recommends searching for somatic alterations of BRCA1/2, ATM, PALB2, FANCA, RAD51D, CHEK2, and CDK12 for metastatic PCa and microsatellite instability for mCRPC (www.nccn.org). Mutations should be sought using NGS preferentially with large panel, since recent data suggest that not only BRCA1/2 mutations (i.e., PALB2 loss) can confer a PARPi sensitivity.

Determination of DDR gene status in mCRPC patients has been proposed to select first-line treatment options (AR pathway vs. taxane-based chemotherapy) with discrepant results. Thus, in PROREPAIR-B study, germinal BRCA2 mutations had a deleterious impact on mCRPC outcomes that may have been affected by the first line of treatment (with a shorter survival when using docetaxel as first-line therapy before AR pathway inhibitors) [17]. This benefit was also confirmed by Antonarakis et al. [28]. In the opposite, Annala et al. [29] reported shorter progression-free survival in 22 germline DDR carriers versus noncarriers (3.3 vs. 6.2 months, respectively) when treated with first-line AR pathway inhibitors. The multicenter BRCA2MEN study is currently planned to validate the role of BRCA2 as a predictive biomarker to select the first-line therapy in patients with mCRPC.

Another utility of HRR mutation testing is to assess eligibility for PARPi in mCRPC settings. Thus, PARPi recently improved overall survival (OS) in mCRPC HRD patients, after new hormonal therapy (NHT) [30], and four PARPi are currently in late development in PCa: olaparib, rucaparib, niraparib, and talazoparib (Tables 1, 2).

Olaparib was the first one to show an OS improvement [30]. The first positive signals arose from the phase II TOPARP-A study which enrolled 50 pretreated mCRPC patients to receive olaparib [31]. Eighty-eight percent of the HR-altered patients (14/16 patients) had a composite response (objective radiological response, or a biochemical response, or a decrease in the circulating tumor cell), compared to only 6% for the HR-efficient patients. Based on these results, 98 HRD patients with mCRPC were enrolled in TOPARP-B [32]. A preplanned analysis subgroup showed an important objective radiological response for BRCA1/2 patients (11/21, 52.4%) and PALB2 patients (2/6, 33.3%), but not ATM patients (5.3%) [32].

In parallel, the study PROfound compared, in a phase 3 trial, the efficacy of olaparib to enzalutamide or abiraterone in patients with mCRPC previously treated with at least one NHT [30, 33] +/− taxanes (65% of the patients), with patients in the control arm receiving another NHT [30]. The study was positive in the primary endpoint (radiographic progression-free survival [rPFS]) and in OS, while 67% of the patients in the control arm crossed over (Table 1). Given these results, olaparib was approved by the Food and Drug Administration (FDA) for patients with an alteration in 14 of the 15 prespecified HRR genes. Indeed, preclinical data, generated after the PROfound design, suggest that PPP2R2A loss of function does not confer sensitivity to PARPi and was then withdrawn from the FDA list [33]. The European Medicines Agency was more conservative and gave its approval only for BRCA1/2, given concern for a lower benefit regarding ATM. Other studies using rucaparib, niraparib, and talazoparib are presented in Table 1 [34‒36].

Preclinical data suggest synergic action between olaparib and androgen pathway inhibitors: AR signaling regulates DNA repair, thus inducing HRR deficiency and conferring a BRCAness state, and PARPi may increase activity of NHT via AR-dependent transcription [37]. Thus, use of PARPi has also been proposed in combination with NHT. A phase 2 trial showed encouraging results in terms of radiographic progression-free survival (rPFS) (HR 0.65) with olaparib in combination with abiraterone versus abiraterone alone in mCRPC pretreated with docetaxel, irrespective of HRR mutation status [38]. Hence, this clinical benefit led to 2 phase 3 trials comparing abiraterone with either olaparib (PROpel) or niraparib (MAGNITUDE), with both studies having met their primary endpoint with an rPFS improvement in the combination arm (irrespective of HRR mutations in PROpel but without benefit in HRR-negative population in MAGNITUDE), and OS data still immature [39] (Table 2). Other combinations with checkpoint inhibitors are ongoing (NCT02861573, NCT03834519 KEYLINK-010 study, etc.) (Table 3), since PARPi could act as immunomodulatory agents, inducing PD-L1 expression through interferon after activation of the cGAS/STING pathway via accumulation of unrepaired DNA fragments [40].

Urothelial carcinoma (UC) displays a complex genomic landscape, including defective DNA damage response and repair at the somatic genomic level. A recent targeted exome sequencing (n = 237 genes) identified a DDR gene alteration in 54% of patients analyzed (not all being considered deleterious), with a higher frequency in muscle-invasive bladder cancer (BC) (MIBC) compared to low-grade non-MIBC and high-grade upper tract urothelial cancer (UTUC), the latter displaying a distinct mutation profile than BC [41]. In a recent study, compared to urothelial BC, UTUC had higher clonal and subclonal mutation numbers, TP53, PIK3CA, and FGFR3 mutations being the putative driver genes in UTUC, whereas for BC, the driver gene may be BRCA1. Conversely, there was no significant difference in the number of DDR mutations between UTUC and urothelial BC [42].

In high-grade non-MIBC and MIBC cohorts, there was a strong association between DDR gene mutation and higher tumor mutation burden. Using a more stringent definition of mutation in DDR genes, DDR gene alterations were observed in 23% of patients with ATM, ERCC2, PRKDC, and ATRX mutations being most common (in 4.3, 4.1, 2.4, and 2.2% of the cases, respectively). ERCC2 alterations seemed to be linked to smoking habits in patients in this study [41], and recent data suggested that PRKDC alterations may serve as potential biomarkers of response to immune checkpoint blockade [43].

In addition, recent studies have suggested that there is an association between DDR gene alterations and sensitivity to cisplatin-based regimens and to immunotherapy treatment for UC (PD-1/PD-L1 blockade) [44, 45]. Teo et al. [45] were the first to report the association between immunotherapy and defective DDR mechanisms beyond MMR deficiency (pembrolizumab already approved for MMR-deficient tumors, regardless of primary tumor site). Regarding platinum sensitivity, cell line studies reported that DDR-related mechanism could be of particular relevance in acquired cisplatin resistance mechanisms of individual UC cells and that targeting of CHK1 (encoded by the CHEK1 gene) might be useful to overcome acquired cisplatin resistance of certain subtypes of UC [46].

Recently, the JAVELIN Bladder 100 trial showed there was a survival advantage for the use of avelumab immunotherapy in maintenance setting in patients who had not progressed with 1st line chemotherapy for advanced/metastatic UC [47], with PD-L1 expression by tumor cells being a putative biomarker of response as well as tumor mutational burden [48].

Regarding germline mutations, recent work using targeted sequencing (i.e., targeted clinical germline testing) of known cancer susceptibility genes revealed that 14–24% of UC patients harbored pathogenic germline variants and 18.6% had clinically actionable variants. ATM and BRCA2 were highlighted in these studies as potential UC predisposition genes but account for only a small fraction of the genetic predisposition for the disease since other genes (i.e., MLH1, MSH2, etc.) are also validated as UC risk genes [49]. Individually rare but collectively common germline variants can explain a substantial fraction of the missing genetic predisposition to UC. A recent study on putative deleterious germline variants in DDR genes in UC opened the door to trials of PARPi in properly selected UC patients [50]. Data presented during the ASCO-GU 2022, supporting the use of HRR status to select patients for PARPi treatment in advanced UC, are summarized in Table 4.

In other genitourinary malignancies, such as RCCs, the analysis of pathogenic/likely pathogenic germline variant frequency and spectrum has been determined on a cohort of patients with early onset of the disease [51]. These variants were identified in 17.7% of patients, of which 8.6% were in non-RCC-associated genes, including 7.3% in DDR genes such as BRCA1/2, ATM, BARD1, PALB2, and CHEK2. CHK2 kinase (encoded by the CHK2 gene) has a role in the mitotic phosphorylation of the tumor suppressor BRCA1 [4]. On the basis of its role during DDR, CHK2 loss in human tumors may be exploited to develop novel therapies based on synthetic lethal interactions, but the clinical utility of the strategy of identification of germline variants requires validation and further validation in RCC.

At the somatic level, deleterious DDR gene alterations have been described as recurrent genomic events in patients with advanced RCC (most frequently altered genes being CHEK2 and ATM) and were mostly clonal in a recent cohort [52]. Loss-of-function events in these genes may affect outcomes with immunotherapy in metastatic RCC (with independent correlation with OS), and these hypothesis-generating results deserve further study.

A growing body of evidence supports the need for identification of germinal and somatic DDR alterations in patients with genitourinary malignancies. PARPi have already shown significant survival benefits in patients harboring HHR mutations in advanced settings, paving the way for precision medicine. Further studies will address the benefit of testing at earlier stages of disease to improve outcome.

The authors have no conflicts of interest to declare.

The authors declare no funding source for this manuscript.

Charles Dariane: data collection and manuscript writing. Marc-Olivier Timsit: project development, manuscript writing, manuscript editing, critical revision, and supervision.