FormalPara Key Points

This comprehensive literature review provides an update on oral poly(ADP-ribose) polymerase (PARP) inhibitors for the treatment of breast cancer (BC).

The review focuses on olaparib and talazoparib, PARP inhibitor monotherapies approved for patients with deleterious/suspected deleterious germline BRCA-mutated, human epidermal growth factor receptor 2-negative BC. Olaparib is approved in the USA for metastatic BC and in Europe for locally advanced/metastatic BC. Talazoparib is approved for locally advanced/metastatic BC in the USA and Europe.

The review also discusses the investigation of PARP inhibitors for the treatment of early-stage BC, as well as in novel combinations and in other BC populations with high unmet needs, including those with triple-negative BC, somatic BRCA mutations, and mutations in other genes associated with defects in homologous recombination repair of DNA.

1 Introduction

Breast cancer (BC) is the second most common cancer in the world and the most common malignancy in women, with approximately 2.09 million new cases diagnosed in 2018 (accounting for 12% of all cancers) [1]. Men account for fewer than 1% of patients with BC [2]. Although survival rates are improving, BC is still the fourth most common cause of death from cancer (627,000 deaths among women in 2018) [1, 3, 4]. Risk factors for developing BC include family history, age, environmental and lifestyle factors associated with carcinogen exposure, and hormonal changes [5,6,7,8]. The risk of developing BC is about two times higher if there is one first-degree relative affected by the disease and may be five times higher if the relative had BC at a young age [7, 8].

Up to 10% of patients with BC have inherited (germline) DNA mutations, often leading to loss of function in genes implicated in DNA repair and cell-cycle checkpoint activation. The remaining ~ 90% of cases are caused by acquired (somatic) genetic and epigenetic alterations [5, 6]. Loss-of-function mutations in two important BC susceptibility genes that are critical in the DNA damage response (DDR), BRCA1 and BRCA2, are detected in at least 5% of unselected patients with BC and in approximately 30% of patients with a positive family history of breast or ovarian cancer [5, 6, 9, 10]. In carriers of BRCA1 or BRCA2 mutations, the risk of developing BC by 80 years of age is as high as 70%, compared with a 10% risk for women in the general population [9, 11]. Germline BRCA (gBRCA) mutations are particularly common in certain populations. For example, in a study of 732 women of Ashkenazi Jewish heritage who underwent genetic testing, 11% had one of three gBRCA founder mutations [12]. Extensive analyses have revealed that somatic BRCA1 mutations are uncommon in unselected patients, although expression of BRCA1 is often reduced, in non-hereditary (sporadic) BC [10, 12,13,14,15]. BRCA mutation and hormone receptor status are also interlinked. Individuals with a gBRCA1 mutation are more likely to develop triple-negative BC (TNBC) than hormone receptor-positive (HR+) disease, whereas patients with gBRCA2 mutations tend to develop HR+ BC. gBRCA mutations are found in up to 23% of patients with TNBC and in 5% of patients with HR+ disease [16,17,18,19,20,21].

Treatment options are limited at present for patients with gBRCA-mutated BC, and the presence of these mutations is associated with younger age at BC diagnosis, aggressive disease characteristics, and higher risk of disease recurrence [22, 23]. Thus, this patient population has a high unmet need. Chemotherapy has been the mainstay of treatment for patients with gBRCA-mutated TNBC, and endocrine therapy plays an important role in gBRCA-mutated HR+ disease [24]. However, despite aggressive treatment, many patients will relapse and eventually die from their disease, and still others present with metastatic disease at initial diagnosis [25,26,27]. Hence, the goal of producing effective biomarker-targeted oral medications such as poly(ADP-ribose) polymerase (PARP) inhibitors is of major importance.

Two PARP inhibitor monotherapies, olaparib and talazoparib, have been approved by the United States Food and Drug Administration (FDA) and European Medicines Agency (EMA) for deleterious or suspected deleterious gBRCA-mutated, human epidermal growth factor receptor 2 (HER2)-negative BC, based on positive outcomes in phase 3 trials (OlympiAD and EMBRACA) [28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46]. Specifically, olaparib is FDA-approved for metastatic BC and EMA-approved for locally advanced/metastatic BC, and talazoparib is FDA- and EMA-approved for locally advanced/metastatic BC. Of the other three PARP inhibitors (niraparib, rucaparib, and veliparib) currently in global clinical trials for the treatment of BC, veliparib is in phase 3 development for HER2-negative, gBRCA-mutated locally advanced/metastatic BC and has shown promising outcomes when administered with platinum-based chemotherapy (BROCADE3 trial) [47, 48]. The differing activities of PARP inhibitor therapies may explain potential differences in their clinical efficacy and safety profiles [49,50,51,52,53].

PARP inhibitor therapies are now being investigated for the treatment of earlier stages of BC, as well as in novel combinations and in patients without gBRCA mutations, including somatic BRCA mutations and mutations in other DDR genes. This comprehensive literature review provides an overview of the use of PARP inhibitors in the treatment of BC, including background on their mechanism of action, relevant clinical trials, and discussion of the implications for their use in clinical practice and future directions.

2 DNA Repair, PARP Inhibition, and Synthetic Lethality

DNA damage and deficiencies of repair are central features of cancer pathology. Healthy cells defend themselves against DNA damage through five major DDR pathways, thus maintaining genomic integrity (Fig. 1). Base excision repair deals with single-strand breaks, nucleotide excision repair addresses helix-distorting damage, while mismatch repair corrects replication errors. Double-strand breaks can be repaired either by the homologous recombination repair (HRR) pathway, using the sister chromatid as a template, or by the more error-prone template-independent mechanism of non-homologous end-joining [51, 54, 55].

Fig. 1
figure 1

modified from O’Connor MJ [54])

DNA damage response pathways (

At least 450 proteins are thought to be involved in DDR pathways, including PARP1 and PARP2 [54]. PARP enzymes are integral to the base excision repair pathway. PARP1 attaches to the damaged DNA strand, allowing nicotinamide adenine dinucleotide (NAD+) to bind to its active site (Fig. 2). ADP-ribose moieties from NAD+ are transferred to target proteins, a process called PARylation, which mediates the recruitment of single-strand DNA repair effectors. PARP1 autoPARylates, leading to its release from DNA and restoration of a catalytically inactive state [51, 53, 56].

Fig. 2
figure 2

The role of PARP in base excision repair of single-strand breaks in DNA. LigIII DNA ligase 3, NAD+ nicotinamide adenine dinucleotide, PARP poly(ADP-ribose) polymerase, pol b, DNA polymerase beta, XRCC1, X-ray repair cross-complementing protein 1

Double-strand breaks form when single-strand breaks are not repaired. Both BRCA1 and BRCA2 proteins play critical roles in the HRR pathway [55]. Initiation of HRR involves recognition of double-strand breaks by the kinases ataxia telangiectasia mutated (ATM) and ataxia telangiectasia and Rad3-related (ATR), and signal transduction by phosphorylated CHK2 (another kinase) and BRCA1 proteins [54, 55, 57]. BRCA1 is a multifunctional protein, with roles beyond direct involvement in HRR, including cell cycle progression, transcription of DDR genes, and apoptosis [51, 58, 59]. In the HRR pathway, BRCA1 forms a multiprotein scaffold that organizes repair proteins at the DNA repair site [57, 60,61,62]. BRCA2 facilitates HRR by recruiting the recombinase RAD51 at the DNA repair site [57]. Along with BRCA1 and BRCA2, multiple HRR genes, including ATM, BARD1, BRIP1, CHEK2 (encodes CHK2), MRE11A, PALB2, RAD50, RAD51C, and RAD51D, are also implicated in hereditary cancer risk [55].

Most late-phase trials of PARP inhibitors have assessed efficacy in patient populations with a vulnerability in their tumor cells, namely HRR deficiency [51, 53, 54, 56, 63, 64]. Tumor cells with HRR gene mutations are targeted by PARP inhibitor therapies through a mechanism known as synthetic lethality (Fig. 3) [51, 54]. PARP inhibitors bind to PARP, inhibiting PARylation, and also trap inactivated PARP on DNA, thereby blocking replication forks, leading to their collapse and the generation of double-strand breaks [51, 52, 54, 56]. If PARP enzymes are inhibited in cells lacking functional HRR proteins (e.g., BRCA1, BRCA2), double-strand breaks can be repaired by the non-homologous end-joining pathway. However, the error-prone nature of this template-independent repair pathway ultimately leads to tumor cell death. By contrast, healthy cells should be spared, thus providing patients with benefits that are not achieved with conventional chemotherapy [54, 56]. In addition to roles in DDR, PARP enzymes are involved in transcription, apoptosis, and immune function; hence, multiple mechanisms of action may contribute to PARP inhibitor efficacy [51].

Fig. 3
figure 3

Synthetic lethality by PARP inhibitors in HRR-deficient cancer cells (modified from O’Connor [54]). HRR homologous recombination repair, PARP poly(ADP-ribose) polymerase

Preclinical data show that the potency of PARP trapping and cytotoxic specificity for HRR-deficient cells differ among the PARP inhibitors, which may explain differences in their clinical efficacy and safety profiles [49,50,51,52,53]. For example, veliparib is a weak PARP1 trapper and may not elicit the same level of synthetic lethality compared with stronger trappers (olaparib, talazoparib, rucaparib, niraparib). Talazoparib is 100-fold more potent at trapping PARP1 than niraparib, which in turn is more potent than rucaparib and olaparib [50,51,52,53, 56]. However, talazoparib has reduced cytotoxic specificity for HRR-deficient cells [50].

3 PARP Inhibitors as Monotherapies for Locally Advanced and/or Metastatic Breast Cancer

Olaparib and talazoparib monotherapies are approved for the treatment of patients with deleterious or suspected deleterious gBRCA-mutated, HER2-negative BC [37,38,39,40,41,42,43,44]. Specifically, olaparib is FDA-approved for metastatic BC and EMA-approved for locally advanced/metastatic BC, and talazoparib is FDA- and EMA-approved for locally advanced/metastatic BC. These approvals were, respectively, gained from the FDA and EMA for olaparib in January 2018 and April 2019 and for talazoparib in October 2018 and June 2019, based on positive outcomes in the OlympiAD and EMBRACA phase 3 trials [29, 33, 38, 40, 41, 44]. Both clinical trials were statistically powered to detect between-treatment differences in the primary endpoint, progression-free survival (PFS), in the overall patient population; subgroup analyses of PFS often included limited numbers of patients [29, 33]. Niraparib, rucaparib, and veliparib are also in clinical development as monotherapies for BRCA-mutated locally advanced/metastatic BC [65,66,67,68,69,70,71,72,73]. Enrollment in the BRAVO phase 3 trial of niraparib was stopped prematurely because of a high rate of discontinuation in the control arm [65,66,67]. A summary of PARP inhibitor monotherapy clinical trials in locally advanced/metastatic BC is shown in Table 1.

Table 1 Clinical trials of oral PARP inhibitors as monotherapies for locally advanced and/or metastatic breast cancer

3.1 Olaparib in the Phase 3 OlympiAD Trial

OlympiAD was an open-label, randomized, multicenter, international, phase 3 trial comparing the efficacy and safety of olaparib versus single-agent standard therapy of the physician’s choice (TPC; capecitabine, eribulin, or vinorelbine in 21-day cycles) in patients with gBRCA-mutated, HER2-negative metastatic BC. An open-label design was required owing to the different treatment options available for use in the TPC arm; however, the intended regimen had to be specified by the physician prior to randomization. All patients had received no more than two prior lines of chemotherapy for metastatic BC. Based on 2:1 randomization, 205 patients were assigned to oral olaparib (300 mg tablet twice daily) and 97 patients to TPC. The primary endpoint of PFS was assessed by blinded independent central review. Prespecified secondary endpoints included overall survival (OS), objective response rate (ORR), and health-related quality of life (HRQoL) [29].

Median PFS was significantly longer with olaparib (7.0 months) versus TPC (4.2 months; hazard ratio [HR] 0.58, 95% confidence interval [CI] 0.43–0.80; p < 0.001). PFS HRs were consistent across a range of patient subgroups, including those with and those without prior exposure to chemotherapy for metastatic BC and in patients with TNBC, an important consideration given the limited treatment options available for TNBC [29]. Post hoc analyses suggested that patients with visceral metastases benefit from improvements in PFS, when investigated by location (lung/pleura, liver, and brain/central nervous system) [74]. Another post hoc analysis showed that, in the few patients whose tumors did not show loss of heterozygosity (6% of 125 tested patients), there was no evidence for a reduction in the efficacy of olaparib, based on PFS [30].

In the final prespecified analysis of OS, conducted after 192 deaths (64% of patients), no significant difference was detected in median OS with olaparib (19.3 months) versus TPC (17.1 months; HR 0.90, 95% CI 0.66–1.23; p = 0.513) [32]; survival was 18.9% for olaparib versus 14.2% for TPC at 48 months in a post hoc follow-up analysis [28]. In both treatment arms, patients received other medications after discontinuing study treatment (2.0% and 11.3% in the olaparib and TPC arms, respectively, were subsequently treated with a PARP inhibitor), which may have contributed to these OS outcomes [28]. In an exploratory subgroup analysis in the first-line setting for metastatic disease, there appeared to be greater OS benefit for patients treated with olaparib (22.6 months) than TPC (14.7 months; HR 0.51, 95% CI 0.29–0.90; n = 87); this difference was greater than that observed between the treatment arms in the overall trial population [32]. The OS benefit in the second- or third-line setting for metastatic disease was 18.8 months for patients treated with olaparib and 17.2 months with TPC (HR 1.13, 95% CI 0.79–1.64; n = 215) [32]. Possible differences in OS benefit associated with therapeutic line may be related to clinical factors such as development of resistance to medication [75].

ORR in the olaparib arm was more than double the rate observed with TPC when assessed by blinded independent central review (59.9% vs. 28.8%) [29], and also when investigator-assessed (57.6% vs. 22.2%) [32]. Similarly, ORR with olaparib was more than double that with TPC in patients with visceral metastases (lung/pleura, liver, and brain/central nervous system) in post hoc analyses [74].

HRQoL assessments were based on patient-completed European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module (EORTC QLQ-C30) questionnaires. HRQoL consistently improved with olaparib versus TPC, with a higher proportion of olaparib-treated patients rating their best overall response as ‘improvement’ (33.7% vs. 13.4%); median time to deterioration of HRQoL was not reached with olaparib versus 15.3 months with TPC. In post hoc analyses of symptoms and functioning, only nausea/vomiting symptoms were worse during treatment with olaparib than with TPC, and olaparib versus TPC delayed time to deterioration on all functional subscales (physical, role, social, cognitive, and emotional) [31].

In the primary analysis, median treatment duration was 8.2 (range 0.5–28.7) months for olaparib and 3.4 (range 0.7–23.0) months for TPC [29, 32]. Most adverse events (AEs) in the olaparib arm were grade 1/2, and the proportion of patients reporting grade 3 or higher AEs was lower with olaparib (38.0%) than with TPC (49.5%). In the olaparib arm, the most common AEs of any grade were nausea (58.0%), anemia (40.0%), and vomiting (32.2%), and the most common grade 3 or higher AEs were anemia (16.1%), neutropenia (9.3%), fatigue (3.4%), and decreased white blood cell count (3.4%) [29, 32]. Cumulative toxicities were not evident [32]. Regarding management of AEs, olaparib dose interruptions did not significantly affect treatment duration, and few patients discontinued olaparib treatment because of AEs (< 5%). These findings indicate that, although patients should be carefully monitored, toxicities can be effectively managed by supportive treatment, dose interruptions, and dose reductions, enabling patients to gain benefit by remaining on treatment with olaparib [32].

3.2 Talazoparib in the Phase 3 EMBRACA Trial

EMBRACA was an open-label, randomized, multicenter, international, phase 3 trial comparing the efficacy and safety of talazoparib versus single-agent standard TPC (capecitabine, eribulin, gemcitabine, or vinorelbine in 21-day cycles) in patients with gBRCA-mutated, locally advanced/metastatic BC. All patients had received no more than three chemotherapy regimens for advanced BC. Based on 2:1 randomization, 287 patients were assigned to treatment with talazoparib (1 mg once daily) and 144 patients to TPC. In both treatment arms, 94% of patients had metastatic disease. The primary endpoint was PFS, assessed by blinded independent central review. Prespecified secondary endpoints included OS and ORR; HRQoL was assessed as an exploratory endpoint [33].

Median PFS was significantly longer with talazoparib (8.6 months) versus TPC (5.6 months; HR 0.54, 95% CI 0.41–0.71; p < 0.001). PFS HRs were consistent across a range of patient subgroups, including those with and those without prior exposure to chemotherapy, patients with TNBC and patients with visceral disease [33, 36]. PFS HRs with talazoparib and TPC were also consistent in the TNBC and HR+ patient subgroups when analyzed by prior exposure to one line and at least two lines of chemotherapy and no prior exposure [36].

In the final analysis of OS, conducted after 324 deaths (75% of patients), no significant difference was detected in median OS with talazoparib (19.3 months) versus TPC (19.5 months; HR 0.85, 95% CI 0.67–1.07; p = 0.17); survival probability was 0.19 (95% CI 0.14–0.25) for talazoparib versus 0.07 (95% CI 0.02–0.15) for TPC at 48 months. Notably, 4.5% and 32.6% of patients randomized to talazoparib and TPC, respectively, received subsequent therapy with a PARP inhibitor (at the time of the EMBRACA trial, olaparib was an approved treatment for metastatic BC associated with a gBRCA mutation) [45, 46].

Investigator-assessed ORR in the talazoparib arm (62.6%) was more than double that in the TPC arm (27.2%) [33]. As with PFS, ORR was higher with talazoparib than with TPC regardless of exposure or lack of prior exposure to chemotherapy in the TNBC and HR+ patient subgroups [36].

Compared with TPC, patients who received talazoparib had significant overall improvement in HRQoL, and delay in time to deterioration across multiple functions and symptoms, including pain and fatigue [34, 36]. Improvements in HRQoL and delay in time to deterioration for pain and fatigue observed during treatment with talazoparib versus TPC were irrespective of Eastern Cooperative Oncology Group performance status at baseline [76].

Median treatment duration was 7.0 (range 0.8–36.9) months for talazoparib and 4.5 (range 0.5–18.3) months for TPC [35]. The proportion of patients treated with talazoparib who experienced grade 3 or 4 hematologic AEs was higher (55% vs. 38%), and grade 3 non-hematologic AEs (32% vs. 38%) was lower, than with TPC. The most common AEs of any grade with talazoparib were hematologic (67.8% of patients), including anemia (52.8%), neutropenia (34.6%), and thrombocytopenia (26.9%), which were frequently grade 3 (38.5%, 17.8%, and 11.2%, respectively). The majority of non-hematologic toxicities were grade 1 or 2, including fatigue, nausea, headache, alopecia, and vomiting [33, 35]. In general, cumulative risks of common hematologic AEs (anemia, neutropenia, and thrombocytopenia) and selected non-hematologic AEs (nausea, fatigue, vomiting, and alopecia) plateaued after weeks 25 and 50, respectively. In a post hoc analysis, talazoparib was associated with a lower rate of serious AE-associated hospitalizations than with TPC (46.8 vs. 71.9 hospitalizations per 100 patient-years, respectively). Patients with common AEs (anemia, nausea, or vomiting) reported favorable outcomes such as better HRQoL during treatment with talazoparib compared with TPC. Few patients discontinued talazoparib treatment because of AEs (5.9%), indicating that toxicities could be effectively managed by supportive care and dose modifications [33, 35].

3.3 Indirect Comparison of Olaparib and Talazoparib: OlympiAD Versus EMBRACA

In the absence of head-to-head evidence for olaparib and talazoparib, an indirect treatment comparison using a Bayesian fixed-effect approach has been performed using published data from the OlympiAD and EMBRACA trials [77]. This analysis suggests that olaparib and talazoparib are equally efficacious with respect to PFS in the populations tested. There was no difference in AE-related discontinuations, although their safety profiles differed. Olaparib was predicted to have fewer common hematologic AEs of any grade (anemia, thrombocytopenia, and neutropenia; odds ratio (OR) 0.37, 0.23, 0.54, respectively) and alopecia (OR 0.22), but an increased risk of nausea (OR 2.39) and vomiting (OR 2.13), relative to talazoparib [77]. These indirect treatment comparisons are limited by differences in how AEs are reported in the published literature and by differences in study design. For instance, the chemotherapies used in the TPC control arms of the two studies differed; notably, gemcitabine was allowed in the EMBRACA trial but not in the OlympiAD trial [29, 33, 77].

4 Treatment Pathways: Germline BRCA Mutation Testing and PARP Inhibitor Therapy

Choice of treatment for BC is based on the clinical characteristics of the individual patient, their disease history, and patient preference [85]. Treatment options are influenced by tumor hormone receptor status (presence or absence of estrogen and progesterone receptors) and HER2 gene amplification [85,86,87]. gBRCA testing, which already has an established predictive role in BC risk assessment, can now be used to inform therapeutic choice. PARP inhibitors are recommended over nonplatinum single-agent chemotherapy for the treatment of patients with advanced BC associated with a gBRCA mutation [87], and platinum compounds also show efficacy [88]. Early provision of genetic counseling and testing, possibly at the time of BC diagnosis, may be beneficial with regard to making informed decisions about primary surgical and other medical interventions [89].

Proposed positions of gBRCA testing and PARP inhibitor therapy in possible treatment pathways for patients with HER2-negative BC are shown in Fig. 4. As indicated in the FDA and EMA labels, patients with BC should be tested for gBRCA mutations before treatment with olaparib or talazoparib [37, 39, 42, 43]. The treatment pathways in Fig. 4 are aligned with the FDA- and EMA-licensed indications for these PARP inhibitors [37, 39, 42, 43] and also with evidence-based US and European treatment guidelines [24, 85, 90]. In particular, olaparib or talazoparib should be used in the treatment of patients with deleterious/suspected deleterious gBRCA-mutated, HER2-negative, locally advanced/metastatic BC after receiving chemotherapy in the (neo)adjuvant or metastatic settings and, if considered appropriate, after patients with HR+ tumors have received endocrine therapy [37, 39, 42, 43]. Olaparib is approved in the USA for metastatic BC and in Europe for locally advanced/metastatic BC; talazoparib is approved for locally advanced/metastatic BC in the USA and Europe.

Fig. 4
figure 4

Possible treatment pathways for germline BRCA-mutated, HER2-negative breast cancer and proposed positions of germline BRCA mutation testing (author opinion, based on treatment guidelines and licensed indications [24, 37, 39, 42, 43, 85, 90,91,92]). aRed star denotes potential positions of gBRCA mutation testing in the treatment pathways. bThe PD-L1 inhibitor atezolizumab plus albumin-bound paclitaxel. For patients with visceral crisis (organ dysfunction) and PD-L1+, first-line treatment could be CT or PARPi. For patients with visceral crisis (organ dysfunction) and PD-L1-, first-line CT may be appropriate. cDouble-headed arrows show that therapies can be provided in either sequence. dOlaparib and talazoparib are PARPi monotherapies approved for deleterious/suspected deleterious gBRCA-mutated, HER2-negative BC. Olaparib is approved in the USA for gBRCA-mutated metastatic BC and in Europe for gBRCA-mutated locally advanced/metastatic BC; talazoparib is approved for gBRCA-mutated locally advanced/metastatic BC in the USA and Europe. eIn Europe, the PI3K inhibitor alpelisib plus fulvestrant is approved for use after disease progression following ET as monotherapy. In the USA, alpelisib plus fulvestrant is approved for use after disease progression on or after an ET-based regimen. fAlt. Tx includes everolimus plus ET. Return arrows show that patients can receive more than one line of Alt. Tx. Alt. Tx alternative treatment to PARPi or CT, BC breast cancer, CDK4/6i cyclin-dependent kinase 4 and 6 inhibitor, CT chemotherapy, ET endocrine therapy, gBRCAm germline BRCA mutation, HER2 human epidermal growth factor receptor 2, HR+ hormone receptor-positive, IOT immuno-oncology therapy, L line, PARPi PARP poly(ADP-ribose) polymerase inhibitor, PD-L1 programmed cell death ligand 1, PI3Ki phosphoinositide 3-kinase inhibitor, TNBC triple-negative breast cancer

5 Identification of Patients Who Could Potentially Benefit from PARP Inhibition

The advent of PARP inhibitor therapies provides the prospect of biomarker-targeted treatment for BC; however, there is a need to efficiently identify who may benefit from treatment and to ensure accessibility to genetic testing [93]. Patients with BC who may be eligible for PARP inhibitor therapy are being missed, even when using established diagnostic guidelines and techniques [94,95,96,97]. In the OlympiAD trial of olaparib for metastatic BC, the majority of gBRCA mutations were detected during screening for the trial [29]. Potential reasons for the lack of uptake of BRCA testing are discussed in Sect. 5.1.

5.1 Issues with Uptake of BRCA Mutation Testing

Identification of BRCA mutations through early genetic screening allows increased monitoring and surveillance for breast (and other) cancers, and may provide the patient and their family with the opportunity for counseling, earlier stage BC diagnosis, and risk-reducing interventions [98,99,100]. However, some patients with BRCA mutations may be missed owing to undertesting; in the USA, only 5.1% and 2.7% of eligible women (based on family history of BRCA mutation-associated cancers) reported uptake of genetic counseling and testing, respectively [101, 102]. Eligibility for and uptake of BRCA testing varies among countries [103,104,105], and use of international testing criteria is not feasible for all countries owing to disparities in resources [106]. There are racial disparities in BRCA testing uptake [101, 107,108,109,110,111,112]. Testing rates also vary widely according to BC receptor subtype [104, 113,114,115,116,117,118].

Potential barriers to BRCA testing uptake and genetic counseling for eligible women with or without a diagnosis of BC include: lack of understanding and knowledge about genetic counseling and testing by physicians and patients; lack of perceived benefits of counseling; lack of perceived risk of having a mutation; cost of testing; and fear of insurance discrimination [94, 109, 119,120,121]. Patients’ attitudes to BRCA testing (the predisposing factor), income (the enabling factor), and risk of carrying a BRCA mutation (the need factor) predict uptake of BRCA testing [122]. Uptake of BRCA testing may be increased in the following ways: provision of free genetic counseling; greater dissemination of information to at-risk individuals; genetic counseling that covers strategies for individuals to discuss their diagnosis with family members; and awareness and implementation of population-based testing as a preventive measure [93, 109, 123,124,125].

5.2 Future Directions to Identify Eligible Patients

Future avenues to identify patients who may benefit from treatment with PARP inhibitors include early detection of somatic BRCA mutations and other gene mutations that result in HRR deficiency in primary tumors and metastases. PARP inhibitor therapies are now being investigated in patients with non-gBRCA HRR gene mutations (see Sect. 6.3) and in neoadjuvant and adjuvant settings (see Sect. 6.1). Use of PARP inhibitor therapies at early stages of BC and in patients without gBRCA mutations are both subject to confirmation of PARP inhibitor efficacy in clinical trials and have yet to gain approval from licensing authorities, including the FDA and EMA. Increased detection of actionable genetic mutations, at earlier stages of disease, would require wider access to BRCA-specific and multiple-gene panel testing, and validation of predictive models to establish probabilities of having gene mutations [126, 127]. Evaluation of mutations in various HRR genes could be fundamental to identify patients suitable for PARP inhibitor therapy, as has been suggested by studies of prostate cancer [56]. Accordingly, a suite of biomarkers correlating with PARP activity has recently been identified in human cancer cell lines, and this could be used as patient selection criteria for expanding the clinical development of PARP inhibitors [128]. In addition, given that immune checkpoint inhibitors that target the programmed cell death ligand 1 (PD-L1) and the programmed cell death 1 (PD-1) receptor are now being investigated as combination therapies with PARP inhibitors in patients with BC [129,130,131,132,133,134,135] (see Sect. 6.2), there may be merit in determining PD-L1 levels in patients who could be eligible for this treatment option [136].

6 Overview of New Directions for PARP Inhibitors

Advances in our knowledge are resulting in potential commencement of PARP inhibitor therapies in patients with earlier stage BC and in combination with other therapies. As with other cancer therapies, resistance to PARP inhibitor therapy occurs in patients with advanced cancer [51]. Resistance to PARP inhibitor therapy may result from multiple mechanisms. For example, HRR could be reactivated by secondary mutations that restore the open reading frames of HRR genes such as BRCA1, BRCA2, PALB2, and RAD51C/D, by mutations leading to mitigation of replication stress, or by mutations in genes for PARP1 or drug effluxion pumps. Early-stage tumors should harbor fewer acquired resistance mechanisms that adversely affect duration of response, in comparison to advanced disease [56]. Thus, treatment of earlier stage disease and use of PARP inhibitor combination therapies may enhance their antitumor effects.

6.1 PARP Inhibitors for Early-Stage Breast Cancer

Treatment of early-stage BC with PARP inhibitors is the subject of several clinical studies, including a phase 3 trial of neoadjuvant veliparib, phase 1/2 trials of neoadjuvant niraparib and talazoparib, and phase 2/3 trials of olaparib as a neoadjuvant and adjuvant treatment (Table 2).

Table 2 Clinical trials of oral PARP inhibitors for early breast cancer in the neoadjuvant and adjuvant settings

At present, there are no specific targeted therapies available for TNBC, which shares some phenotypic and molecular similarities with gBRCA-mutated BC. There is increasing evidence that PARP inhibitor therapies may be effective in the treatment of patients with non-gBRCA HRR gene mutations (see Sect. 6.3). TNBCs often harbor somatic BRCA or other HRR mutations, or BRCA genes may be silenced through promoter hypermethylation, which may result in susceptibility to PARP inhibitor therapy [137]. PARTNER is a three-stage phase 2/3 trial, designed to assess the safety, schedule selection, and efficacy of neoadjuvant olaparib in combination with platinum-based chemotherapy for patients with TNBC and/or gBRCA-mutated BC [138, 139]. Based on 159 patients (target N = 527), preliminary safety data support the combination. A large phase 2 study, the PETREMAC trial, is also ongoing (N = 200); olaparib is one of several treatment options being investigated in this trial for patients with TP53-mutated or TP53-wild-type BC [137, 140]. The primary outcome measure of PETREMAC is the predictive and prognostic value of mutations in 300 cancer-related genes, assessed in BC tissue by next-generation sequencing before starting neoadjuvant therapy. Olaparib monotherapy in 32 treatment-naïve patients with TNBC yielded a high ORR (56%). Of the 18 responders, 16 had HRR defects (gene mutations or BRCA1 promotor hypermethylation), which were found in only four of the 14 non-responders. After excluding patients with gBRCA (n = 4) or gPALB2 mutations (n = 1), ORR was 52% (n = 14/27), thus indicating potential efficacy in patients without gBRCA mutations. In the phase 2 GeparOLA trial (N = 107), in patients with HR+ or TNBC and HRR deficiency (deleterious BRCA mutations and/or high HRR deficiency scores), pathological complete response rates were 55.1% with the combination of olaparib and paclitaxel, relative to 48.6% with carboplatin and paclitaxel; both combinations were followed by treatment with epirubicin and cyclophosphamide [141]. Pathological complete response rates were higher with olaparib combination therapy than with carboplatin and paclitaxel in patients under 40 years of age (76.2% vs. 45.5%) and in those with HR+ tumors (52.6% vs. 20.0%).

The results of the phase 3 BrighTNess trial (N = 634) generally do not support the addition of veliparib to carboplatin and paclitaxel, followed by doxorubicin and cyclophosphamide, for the neoadjuvant treatment of stage II–III, high-risk TNBC [142]. The addition of veliparib and carboplatin to paclitaxel increased the proportion of patients who achieved a pathological complete response (53%) versus paclitaxel alone (31%), but not relative to carboplatin and paclitaxel (58%). In the subgroup of 70 patients with BRCA mutations, pathological complete response rates were 57% with the veliparib combination and 50% with the combination of carboplatin and paclitaxel.

Positive efficacy data have been reported from two phase 1 studies of neoadjuvant niraparib and talazoparib monotherapy [143,144,145]. Niraparib was administered to 21 patients with somatic or gBRCA-mutated BC, mainly TNBC. Based on 18 patients with magnetic resonance imaging (MRI) and ultrasound results after 2 months of treatment, tumor response rate was 89% by MRI, and all patients had responded according to at least one imaging technique [143, 144]. The pilot study of neoadjuvant talazoparib, which had a planned recruitment of 20 patients, was stopped after recruitment of 13 patients owing to favorable efficacy and safety findings. In the 13 patients, who had gBRCA-mutated BC (n = 9 with TNBC), tumor volumes decreased by a median of 88% (range 30–98%) after 2 months of treatment with neoadjuvant talazoparib [145]. The pilot study was modified into a phase 2 trial (N = 20, n = 15 with TNBC), in which 53% of patients experienced a pathological complete response after 6 months of treatment [146]. A phase 2 study of neoadjuvant talazoparib, with a planned enrollment of 112 evaluable patients with gBRCA-mutated, stage I-III TNBC, was terminated in September 2020 (following recruitment of 61 patients) owing to a change in the sponsor’s clinical development strategy, a decision not related to safety and efficacy [147, 148].

In the adjuvant setting, the phase 3 OlympiA trial is ongoing, investigating olaparib monotherapy in patients with gBRCA-mutated, high-risk, HER2-negative primary BC (N = 1836) [149, 150]. Eligible patients had completed neoadjuvant chemotherapy and surgery or adjuvant chemotherapy. The primary objective is invasive disease-free survival.

6.2 PARP Inhibitors in Combination Therapies, Including with Immunotherapies

The combination of PARP inhibitors and immune checkpoint inhibitors is based on evidence for an interaction between the abnormal presence of unrepaired DNA in the cytoplasm and the stimulator of interferon genes (STING) pathway. STING activation leads to the release of interferons and induction of tumor infiltration by T-cells [151]. PARP inhibitor monotherapies have been shown to trigger antitumor immunity in BRCA1-deficient mice, an effect that was augmented when the PARP inhibitor was combined with an immune checkpoint inhibitor [151,152,153]. Monoclonal antibodies that inhibit the interaction of PD-L1 with the PD-1 receptor, allowing the immune system to target tumor cells, include pembrolizumab, durvalumab, atezolizumab, and avelumab.

Promising efficacy and safety findings have been reported for niraparib combined with pembrolizumab and for olaparib plus durvalumab in two single-armed phase 2 studies, TOPACIO and MEDIOLA (Table 3). In TOPACIO (N = 47 for efficacy, N = 55 for safety), the combination of niraparib and pembrolizumab conferred antitumor activity, regardless of BRCA mutation status, in patients with somatic or gBRCA-mutated and wild-type BRCA advanced/metastatic TNBC [129]. ORR was 21% in the overall population (n = 10/47) and 47% in patients with tumor BRCA mutations (n = 7/15). Disease control rate (DCR) was 49% (80% in patients with tumor BRCA mutations). For the five patients harboring non-BRCA HRR pathway mutations, ORR was 20% (n = 1/5) and DCR was 80% (n = 4/5). In the overall population, ORR was numerically higher in patients with PD-L1-positive TNBC (32%; n = 9/28) than in those with PD-L1-negative TNBC (8%; n = 1/13). In MEDIOLA (N = 30 for efficacy, N = 34 for safety), the combination of olaparib and durvalumab was associated with DCRs of 80% and 50% after 12 and 28 weeks, respectively, and favorable tolerability in patients with gBRCA-mutated metastatic BC [130, 131]. Other ongoing trials of PARP inhibitors combined with immune checkpoint inhibitors include DORA, a phase 2 study of olaparib and durvalumab in platinum-responsive locally advanced (inoperable) or metastatic TNBC, and KEYLYNK-009, a phase 2/3 trial of olaparib and pembrolizumab in locally recurrent inoperable or metastatic TNBC [132,133,134,135, 154,155,156].

Table 3 Clinical trials of oral PARP inhibitors in combination with immunotherapies and other combinations for the treatment of breast cancer

PARP inhibitors are also being evaluated in combination therapies with other agents to treat locally advanced or metastatic BC [47, 48, 157]. In the phase 3 BROCADE3 trial (N = 509), addition of veliparib to carboplatin and paclitaxel resulted in significant improvement in median PFS compared with placebo added to carboplatin and paclitaxel (14.5 vs. 12.6 months; HR 0.71, 95% CI 0.57–0.88; p = 0.002) in patients with gBRCA-mutated, HER2-negative, locally advanced or metastatic BC. The PFS benefit was durable and no additional toxicities were seen, although there was a high degree of toxicity in both treatment arms [47, 48]. A subset of patients (n = 194) were transferred from the combination therapies to veliparib or placebo monotherapy for reasons other than disease progression. Patients treated with veliparib appeared to derive PFS benefit from both monotherapy (HR 0.49, 95% CI 0.33–0.73) and combination therapy (HR 0.81, 95% CI 0.62–1.06). Similar benefit was gained with veliparib monotherapy in patients who transferred from ≤ 6 cycles versus patients who transferred from 7–12 cycles of combination therapy, indicating that the number of prior cycles of combination therapy may not influence the efficacy of subsequent veliparib monotherapy. Overall, these results suggest that veliparib monotherapy may be beneficial following a discontinuation of combination therapy with veliparib plus carboplatin and paclitaxel [158]. Looking further ahead, ongoing trials are investigating PARP inhibitors in novel combinations, including olaparib plus inhibitors of DDR molecules (ATR or Wee1) for metastatic TNBC (VIOLETTE trial), olaparib plus trastuzumab for HER2-positive BC (OPHELIA trial), and talazoparib plus a bromodomain inhibitor (ZEN003694) or a dual mTOR/PI3K inhibitor (gedatolisib) for metastatic or recurrent/unresectable TNBC [159,160,161,162,163].

6.3 PARP Inhibition in Broader Populations of HRR-Deficient Breast Cancer

PARP inhibitors are being investigated for the treatment of BC in patients with non-gBRCA HRR gene mutations or without documented gBRCA mutations (Tables 1, 2, 3) [73, 78,79,80, 84, 133, 134, 137, 154, 157, 160,161,162].

Clinical studies that have positive findings for PARP inhibitors in settings other than gBRCA-mutated BC include single-arm phase 2 studies of rucaparib, olaparib, and talazoparib monotherapy (Table 1). In the RUBY trial, rucaparib monotherapy was investigated in 41 patients with homologous recombination deficiency, including four patients harboring somatic BRCA mutations. Five patients (13.5%) demonstrated clinical benefit, comprising three patients with high loss of heterozygosity (complete response, n = 1; partial response, n = 2), one with a somatic BRCA1 mutation (stable disease) and one patient with a somatic BRCA2 mutation (partial response) [73]. In the Olaparib Expanded study, in 54 patients with metastatic BC and germline mutations in various non-BRCA DDR genes (cohort 1) or somatic mutations in DDR genes including BRCA (cohort 2), ORR was 33% and 31%, respectively. Antitumor activity was reported in patients with somatic BRCA or gPALB2 mutations but not in those with ATM or CHEK2 mutations [78]. The phase 2 study of single-agent talazoparib enrolled patients with BRCA wild-type, HER2-negative, advanced BC and non-BRCA HRR pathway mutations. Based on 12 evaluable patients, ORR was 25% after 6 months (two of the three responders had gPALB2 mutations, the other had gCHEK2, gFANCA and somatic PTEN mutations) and the clinical benefit rate was 50% (the three additional patients harbored gPALB2, somatic ATR, or somatic PTEN mutations) [84].

7 Conclusions

PARP inhibitor therapies are a welcome addition to the treatment arsenal for patients with locally advanced or metastatic gBRCA-mutated, HER2-negative BC. Given that this additional option provides targeted therapy for patients presenting with a gBRCA mutation, patients and healthcare professionals require clear guidance on testing for these mutations. The oral formulation of PARP inhibitors, together with their safety and HRQoL profiles, which are more favorable than for chemotherapy agents, have the potential to improve patient experience and adherence [168]. The most common AEs observed during treatment with PARP inhibitors are generally manageable, but patients should be monitored regularly. New directions for evaluation of PARP inhibitors include earlier stages of BC and in combination with agents that target other HRR-related pathways, with a view to potentially avoiding resistance to PARP inhibitor therapy and expanding indications beyond the gBRCA-mutated population. The advent of PARP inhibitor therapies is likely to have significant implications for the treatment of patients with BC beyond the locally advanced/metastatic setting.