Systemic Treatment Strategies for Patients with Hereditary Breast Cancer Syndromes

This article is a review of systemic treatment strategies for patients with hereditary breast cancer syndromes involving germline mutations, including the most common hereditary breast cancer syndromes: BRCA‐related breast cancer syndrome, Li‐Fraumeni syndrome, Cowden syndrome, Peutz‐Jeghers syndrome, and hereditary diffuse gastric cancer syndrome.

Abstract

Hereditary breast cancer syndromes are associated with an increased risk of breast cancer and constitute a unique patient population, making up approximately 5%–10% of breast cancer cases in the United States. By virtue of the germline mutations that define these syndromes, invasive breast cancers in these patients have unique mechanisms that can be rationally targeted for therapeutic opportunities distinct from standard of care treatments in nongermline mutation associated breast cancers. This review intends to describe existing data on several of the most common hereditary breast cancer syndromes, including BRCA‐related breast cancer syndrome, Li‐Fraumeni syndrome, Cowden syndrome, Peutz‐Jeghers syndrome, and hereditary diffuse gastric cancer syndrome, specifically focusing on rational therapeutics utilized in these distinct patient subgroups and completed or ongoing clinical trials evaluating their efficacy. By exploiting the distinct biologic features associated with these syndromes, tailored treatment strategies have the potential for improved efficacy and lower toxicity. Knowledge of the emergence of these targeted cancer therapies is critical for appropriate management in these patients, extending beyond treatment to highlight the need for appropriate genetic screening to allow for early recognition of these patients and therefore appropriate treatment.

Implications for Practice.

Molecular testing allows for identification of germline mutations that place individuals at high risk for breast cancer and that are associated with distinct histopathology and molecular characteristics that define the invasive breast cancer cases that these patients develop. These unique characteristics may ultimately provide rational targets for systemic treatments with improvements in both morbidity and efficacy. Identification of patients with these germline mutations is important for not only appropriate screening and prophylaxis, but knowledge of therapies specifically targeting several of the most common hereditary breast cancer syndromes is essential to ensure appropriate treatment of invasive breast cancers in these patients.

Introduction

Patients with invasive breast cancer resulting from highly penetrant cancer susceptibility genes make up approximately 5%‐10% of breast cancer cases in the United States [1]. These patients comprise a unique patient population with a growing repertoire of therapeutic options. This review article aims to review systemic treatment strategies for patients with hereditary breast cancer syndromes involving germline mutations. Specifically, this article will review the following most common hereditary breast cancer syndromes: BRCA‐related breast cancer syndrome, Li‐Fraumeni syndrome, Cowden syndrome, Peutz‐Jeghers syndrome, and hereditary diffuse gastric cancer syndrome.

BRCA‐Related Breast Cancer Syndrome (BRCA1 and BRCA2)

Germline deleterious mutations in BRCA1 (17q11) and BRCA2 (13q12–q13) are associated with a substantially increased risk of breast cancer as compared with the general population, with a cumulative lifetime breast cancer risk of 46%–60% in BRCA1 carriers and 43%–55% in BRCA2 carriers [2], [3], [4]. These mutations have a prevalence of 5%–10% in the general breast cancer population and up to 10%–20% in those patients with triple‐negative breast cancer (TNBC) [5], [6], [7], and they have been suggested to be responsible for approximately half of all hereditary breast and ovarian cancer cases [8]. In addition to increased breast cancer risk, other cancer risks associated with BRCA1 and BRCA2 mutations include melanoma, ovarian, prostate, and pancreatic cancer.

BRCA‐associated breast cancers have a unique biology, most commonly high‐grade, triple‐negative, basal tumor subtype. Triple‐negative biology comprises approximately 60%–80% of breast cancers in BRCA1 carriers and approximately 14%–50% of those in BRCA2 carriers [9], [10], [11], [12]. Additionally, BRCA‐associated breast cancers have been found to be associated with a higher median Oncotype DX (Genomic Health, http://www.oncotypedx.com/) recurrence score in hormone‐receptor‐positive, node‐negative disease as compared with non‐BRCA carrier breast cancer controls [13]. However, studies relating BRCA status to outcome in breast cancer are still conflicting. A meta‐analysis by Zhong et al. in 2015 suggested inferior overall survival (OS), but not worse progression free survival (PFS) in BRCA1 mutation carriers with breast and ovarian cancer as compared with noncarrier patients [14]. Notably, however, this OS difference was not seen in BRCA2 mutation patients and there was significant heterogeneity between these small studies cited [14]. This result is in contrast to a 2011 study showing similar overall prognosis of TNBC in BRCA carriers and noncarriers within the first 5 years following initial diagnosis [15].

Consistent with the role of BRCA1 and BRCA2 in DNA repair, germline mutations in these genes lead to impaired homologous repair of chromosomal double strand breaks, with at least a five‐fold reduction in DNA double strand break repair [16]. This subsequently predisposes patients to chromosomal instability and leads to unique treatment opportunities, including sensitivity to DNA‐damaging agents, ionizing radiation, and poly (ADP‐ribose) polymerase (PARP) inhibitors [17], [18], [19].

DNA‐Damaging Agents

Platinum agents exert their cytotoxic effects by crosslinking and damaging DNA strands that can only be repaired by homologous recombination, which is typically absent in BRCA‐mutated cells. In this setting, platinum agents have been well‐studied in breast cancer patients with germline BRCA mutations, with studies evaluating their use in both the neoadjuvant and metastatic settings.

Multiple studies have evaluated neoadjuvant chemotherapy in BRCA‐mutated breast cancers, specifically focusing on pathologic complete response (pCR) rates, a known surrogate for clinical efficacy and outcome. In a 2011 study by Arun et al., BRCA mutation carriers had up to a 45% pCR when treated with standard neoadjuvant anthracycline plus taxane based chemotherapy [20]. Neoadjuvant cisplatin monotherapy has also been shown to have improved pCR rates in this patient population. A 2010 study by Silver et al. evaluating neoajudvant cisplatin monotherapy in TNBC patients notably showed a pCR in the two patients with BRCA1 germline mutations [21]. A sub‐analysis of all 28 TNBC patients included in the study subsequently showed improved response rates to cisplatin in those patients with lower BRCA1 mRNA expression and BRCA1 promoter methylation [21]. A larger observational study evaluating pCR rates in BRCA1‐positive breast cancer patients found high rates of pCR in patients receiving neoadjuvant cisplatin (pCR in 10 of 12 patients, 83%), substantially higher than pCR rates in BRCA1‐mutated breast cancer patients treated with cyclophosphamide, methotrexate, and fluorouracil (CMF, 7%); doxorubicin and docetaxel (AT, 8%); or doxorubicin and cyclophosphamide with and without fluorouracil ([F]AC, 22%) [22]. A subsequent study in 2014 treated 107 women with stage I–III breast cancer with known BRCA1 mutation with 4 cycles of neoadjuvant cisplatin and found a high pCR rate of 61% (65 of 107 patients) [23].

Similar high pCR rates have been found with neoadjuvant carboplatin combinations in patients with homologous recombination deficiency (HRD). The multicenter placebo‐controlled trial TBCRC008 reported at the 2015 San Antonio Breast Cancer Symposium (SABCS) investigated pCR following 12 weeks of neoadjuvant carboplatin and albumin‐bound paclitaxel with or without vorinostat in patients with estrogen receptor (ER)‐positive or TNBC. The pCR rate was similar in both arms (vorinostat 25.8%, placebo 29%) but notably a higher pCR rate was found in patients with HRD, defined as an HRD score ≥ 42 using the Myriad myChoice HRD test (Myriad, https://myriad.com/products-services/companion-diagnostics/mychoice-hrd/) and/or tumor BRCA mutation (50% vs. 7.7% in those without HRD). Ultimately, the study showed more than a six‐fold increase in pCR in tumors with HRD as compared with those without HRD, suggesting a predictive role of HRD status in response to platinum therapy [24].

When used in the metastatic setting, cisplatin has been associated with a high overall response rate in breast cancer patients with germline BRCA1 mutations. In a phase II study including 20 BRCA1‐mutated metastatic breast cancer (MBC) patients, treatment with 6 cycles of cisplatin was associated with an overall response rate (ORR) of 80%, with 9 out of 20 patients (45%) experiencing a complete clinical response and 7 out of 20 patients (35%) experiencing a partial response (PR) [25]. In a separate multicenter phase II trial, TBCRC009, patients with metastatic TNBC received first or second line cisplatin or carboplatin as monotherapy. In an exploratory analysis of the 11 patients with germline BRCA 1 or 2 mutations, an ORR of 54.5% was found, notably higher than the ORR seen in the general TNBC population (25.6%). In the 66 patients without germline BRCA 1 or 2 mutations, a separate exploratory analysis showed that a BRCA‐like genomic instability signature, found in 32 patients, distinguished responding versus nonresponding tumors [26].

Use of other, nonplatinum DNA‐damaging agents in BRCA‐mutated breast cancers is currently undergoing investigation. Methotrexate, which functions as a DNA‐damaging agent by preventing DNA synthesis through inhibition of dihydrofolate reductase, is currently being studied in a multicenter phase II clinical trial in patients with BRCA‐associated breast or ovarian cancer in combination with six mercaptopurine [27].

Antimicrotubule Agents

Taxane chemotherapy agents, including docetaxel and paclitaxel, work by binding reversibly to microtubules, thereby stabilizing the microtubule and interfering with the normal process of microtubule reorganization, ultimately leading to disruption of mitosis.

Several studies have evaluated the use of taxane agents in breast cancer patients with germline BRCA mutations, overall finding poorer outcomes in patients with BRCA mutations. In a 2008 study evaluating patients with BRCA1‐mutated breast cancer, patients who received neoadjuvant docetaxel with doxorubicin had a lower ORR (6 of 15, 40%) as compared with those receiving other DNA‐damaging therapies (29 of 29, 100%) [28]. Similarly, in a phase III randomized control trial comparing carboplatin with docetaxel in patients with metastatic or recurrent locally advanced BRCA‐mutated breast cancer, ORR was 68% in those patients receiving carboplatin versus 33.3% in those who received docetaxel with PFS of 6.8 months versus 3.1 months, respectively [29].

Interestingly, several studies have suggested that the lack of efficacy with taxane administration is limited to BRCA1 mutation carriers. A trial presented at the 2010 American Society of Clinical Oncology (ASCO) Annual Meeting showed improved ORR in BRCA2 patients as compared with BRCA wild type patients treated with taxane (ORR 75% vs. 36%); however, notably, the groups had similar PFS (4.6 vs. 4.7 months) [30]. A more recent study presented at the 2015 SABCS meeting evaluated the response of patients with stage I–III breast cancer, including 12 with known germline BRCA mutation, to neoadjuvant weekly taxane followed by adriamycin/cyclosphosphamide (AC) or fluorouracil/epirubicin/cyclophosphamide (FEC). Response to taxane administration alone (prior to AC or FEC administration) was performed by magnetic resonance imaging assessment. None of the four patients with BRCA1 mutations had a radiographic complete response after taxanes (0%), while two out of eight BRCA2‐mutated patients had a radiographic complete response after taxanes (25%). This was in comparison to a 16.1% radiographic complete response rate in BRCA wild type patients (18 of 112) after taxane treatment alone. Notably, following taxane plus AC/FEC treatment, pCR rate was 50% of the BRCA mutated patients and 31.3% of the BRCA wild type patients [31]. Overall, these findings suggest that BRCA mutation status predicts taxane resistance, with differential efficacy between BRCA1 and BRCA2 mutation carriers, and suggests that normal BRCA1 may be required for clinical response to antimicrotubule agents.

While of limited use as monotherapy in BRCA patients, taxane administration in combination with DNA damaging agents has been shown to have increased treatment efficacy. The phase II GeparSixto study showed an improved pCR rate of 57.9% in germline BRCA mutated patients treated with neoadjuvant weekly paclitaxel/nonpegylated liposomal doxorubicin, as compared with a 40.2% pCR rate in patients without identified risk for hereditable breast cancer. pCR rate increased by 25% with the addition of weekly carboplatin for patients with germline BRCA mutations (as compared with a 14% increase in pCR rate in BRCA wild type patients) [32]. Another study presented at the 2014 ASCO Annual Meeting evaluated the efficacy of neoadjuvant combination therapy with carboplatin and docetaxel in sporadic and BRCA‐associated TNBC with 86% of patients with deleterious mutations achieving a pCR (12 of 14). This was compared to pCR rates of 50% in the 28 patients with sporadic TNBC [33].

In addition to taxanes, the antimicrotubule agent eribulin has also been evaluated in BRCA‐associated breast cancer patients. The neoadjuvant GeparQuinto study evaluated 74 patients with germline BRCA mutations with TNBC treated with epirubicin/cyclophosphamide, followed by 4 cycles of docetaxel, with or without bevacizumab, finding a higher pCR rate of 50% (37 of 74) in BRCA‐mutated patients as compared with 31.1% Pathologic complete response rate in BRCA wild type patients [34].

Trabectidin

Trabectidin has been shown to block DNA binding of the oncogenic transcription factor FUS‐CHOP. In addition, preclinical and clinical data have suggested that trabectedin may have specific activity against nucleotide excision repair intact or homologous recombination repair deficient MBC, prompting its evaluation in BRCA‐mutated breast cancer patients. When evaluated in heavily pretreated MBC patients with germline BRCA1 and BRCA2 mutations, trabectedin treatment every 3 weeks was found to lead to PR in 6 out of 35 patients (17%), with median PFS of 3.9 months [35]. Similarly, a phase II trial of trabectedin showed activity in BRCA1/2 mutation carrier patients with pretreated MBC, with 4 out of 29 patients (14%) achieving PR and median PFS of 3.3 months [36].

Lurbinectedin

Lurbinectedin (PM01183) binds covalently to DNA and induces the formation of double strand breaks in a wide range of cancer cell lines, with particular activity against platinum resistant tumors and homologous recombination deficient cell lines. In this setting, it was investigated in previously treated MBC patients with germline BRCA1/2 mutations with an ORR of 41% (1 complete response [CR], 6 PR, 6 stable disease [SD], 4 progressive disease [PD] in 17 evaluable patients) with a median duration of response of 5 months. This was as compared with an ORR of 9% and median duration of response of 3.3 months in an unselected cohort. In an exploratory analysis, the ORR in the BRCA mutated cohort was higher in PARP inhibitor naïve patients (64%, 7 of 11 patients) [37]. At the European Society of Medical Oncology (ESMO) Annual Meeting in October 2016, Balmana and colleagues presented a phase II trial of 54 MBC patients with BRCA mutations at initially 7 mg fixed dose and then amended to 3.5 mg/m2 IV every 3 weeks with an overall response rate of 39% with the fixed 7 mg dose and 44% with the 3.5 mg/m2 dosage. More than half the patients had received a prior platinum therapy. There was a differential response in BRCA2 versus BRCA1 with improved ORR (61% vs. 26%), PFS (5.9 vs. 2.1 months) and OS (31.8 vs. 11.8 months) in BRCA2 mutation carriers compared with BRCA1 mutation carriers. An expansion cohort evaluating this agent in patients who previously have been exposed to PARP inhibitors is underway [38].

PARP Inhibitors

Upon DNA damage, PARP activation leads to ADP‐ribosylation of histones and recruitment of chromatin remodeling enzymes that ultimately creates a relaxed chromatin state favorable for DNA repair. This is an essential component of base excision repair and single strand break repair, both of which homologous recombination deficient cells, such as BRCA‐mutated cells, rely heavily upon. Suppression of PARP catalytic activity blocks the formation of ADP‐ribose polymers at the site of single strand breaks, preventing the recruitment of DNA‐damage repair complexes. Unrepaired single strand breaks eventually lead to stalled replication forks, which collapse into double strand breaks. When unable to be repaired by homologous recombination, such as in BRCA mutated cells, these double strand breaks are highly cytotoxic [39], [40], [41], [42], [43]. Utilizing the rationale of synthetic lethality, PARP inhibitors have been studied extensively in BRCA‐mutated breast cancer. These include veliparib, which functions primarily through suppression of the catalytic activity of PARP, as well as olaparib, talazoparib, rucaparib, and niraparib, which act via trapping PARPs to DNA.

Olaparib

Olaparib is an orally available PARP inhibitor that was FDA approved in 2014 for patients with germline BRCA‐mutated advanced ovarian cancer pretreated with at least three prior lines of chemotherapy. Promising results of olaparib monotherapy have also been seen in BRCA‐mutated breast cancer patients.

In a proof‐of‐concept phase I trial, olaparib was used as monotherapy in 60 patients with advanced solid tumors, including 9 patients with breast cancer and 22 patients with BRCA1/2 mutations (3 out of 9 breast cancer patients had BRCA2 mutations). Objective antitumor activity was found only in BRCA mutation carriers. Of the 19 BRCA patients that could be evaluated, 9 had partial or complete radiologic response (8 with ovarian cancer, 1 with breast cancer) and 2 had radiologically stable disease (1 with ovarian cancer and 1 with breast cancer) [44]. In a subsequent phase II trial, 54 patients with recurrent, advanced BRCA1/2 mutated breast cancer were treated with continuous oral olaparib. Evaluating the 27 patients who received the maximum tolerated dose of 400 mg twice daily, a 41% ORR was noted [45]. A subsequent multicenter phase II basket trial evaluated olaparib monotherapy in 298 patients with heavily pretreated recurrent cancers including ovarian, breast, pancreatic, and prostate cancers with BRCA1/2 mutations. Of the 62 patients with BRCA mutated breast cancer, the ORR was 12.9% (8 of 62), and 47% of patients had disease stabilization for at least 8 weeks. This response rate was better for those patients without prior platinum exposure (20% vs. 9.5%). The lower ORR compared with previous studies was suggested to be due to the heavily pretreated nature of these patients, with a mean of 4.6 prior chemotherapy regimens in the metastatic setting [46]. Heavy pretreatment was also suspected to be a reason for low response to olaparib in BRCA‐mutated breast cancers in a 2011 study by Gelmon et al. in which olaparib monotherapy showed no confirmed objective responses in 10 patients with BRCA‐mutated TNBC [47].

Taken together, these studies suggest efficacy of olaparib monotherapy in BRCA‐mutated breast cancer patients. Notably, predictors of poorer response suggested by these studies includes heavy pretreatment, particularly prior platinum exposure. Ongoing studies include OlympiA, a phase III randomized trial of olaparib as adjuvant monotherapy in germline BRCA‐mutated TNBC patients, which began enrolling in April 2014 with 1,320 patients targeted across 550 sites and 25 countries worldwide [48].

Veliparib

Veliparib (ABT 888) is a potent, orally administered small molecule inhibitor of PARP1 and PARP2. Similar to olaparib, monotherapy with veliparib has been shown to be efficacious in BRCA‐associated breast cancer patients.

In a phase I study of single‐agent veliparib, ORR was 29% (4 of 14 patients) in BRCA‐mutated breast cancer patients as compared with ORR 5% (1 of 21 patients) in BRCA‐wild‐type patients [49]. In a subsequent phase II trial of BRCA‐mutated MBC, single agent veliparib was administered orally at 400 mg twice daily until progression, at which time therapy was transitioned to combination veliparib dosed at 150 mg orally twice daily and carboplatin (area under the curve [AUC] of 5). Forty‐one out of 44 patients enrolled were treated, all with BRCA1/2 mutations. As of the time of presentation at the ASCO Annual Meeting 2014 meeting, the rate of PR was 17% (2 of 12 patients) in BRCA1 patients and 23% (3 of 13 patients) in BRCA2 mutations who had at least 4 cycles of follow‐up. The authors noted time to failure on veliparib of 2.0 months for BRCA1 and 5.1 months for BRCA2 patients. Of the 10 patients who had proceeded to combination treatment with veliparib and carboplatin, the authors noted one PR in a BRCA1 patient [50].

Rucaparib

Rucaparib (AG‐014699, PF‐01367338) is a potent, selective PARP1 and PARP2 inhibitor. When evaluated in a multicenter, single‐arm phase II trial reported at the 2011 ASCO Annual Meeting, 41 patients with BRCA mutated breast (17 patients) and ovarian (24 patients) cancer were treated with intravenous rucaparib on days 1–5 of a 21 day cycle. Of these 41 patients, 38 had RECIST assessments with 5% ORR (2 of 38) and 26% of patients achieving stable disease for at least 4 months (10 of 38). The intermittent dosing schedule was suspected to be the cause of the lower ORR found in this study [51]. To further evaluate the impact of dosing schedules on PARP activity, the same group evaluated rucaparib treatment in BRCA‐mutated advanced breast and ovarian cancer, specifically testing intravenous and oral rucaparib at a range of dosing schedules. The study found that PARP inhibition following rucaparib dosing was transient, recovering within 7 days. Utilizing intermittent dosing with intravenous rucaparib, an ORR of 2% was found, which was much lower than the ORR of 15% seen with continuous oral rucaparib dosing across all six dose levels [52]. Notably, this response rate with continuous oral rucaparib is still lower than that reported in a study presented at the 2014 ASCO Annual Meeting using the recommended phase II dose of rucaparib at 600 mg orally twice daily. With limitations of a small sample size, this study showed a RECIST or CA125 response in 3 out of 4 patients with BRCA‐associated ovarian cancer (80%) and 1 out of 1 patient with BRCA‐associated breast cancer (100%). All responders in the study had a BRCA1/2 mutation [53].

Niraparib

Niraparib (MK4827) is a potent, selective, orally available PARP1 and PARP2 inhibitor shown in a phase I dose escalation study in patients with advanced solid tumors enriched for BRCA mutations to have a 50% ORR (2 of 4 patients with PR) in those patients with BRCA‐associated breast cancer. Of those patients with BRCA‐associated ovarian cancer, 8 out of 20 (40%) had PR [54]. Further studies of niraparib are pending, including BRAVO, a phase III multicenter trial currently enrolling patients with germline BRCA mutations with HER2 negative breast cancer to receive niraparib versus physician's choice (eribulin or capecitabine or gemcitabine or vinorelbine), notably excluding patients who had progressed on prior platinum therapy [55].

Talazoparib

Talazoparib (BMN673) is the most potent and specific inhibitor of PARP1 and PARP2 in clinical development with an IC₅₀ of less than 1 nM that also functions by trapping PARP on DNA. In the first in‐human dose escalation study of talazoparib, eight patients were included with breast cancer (6 of 8 with deleterious BRCA mutations). Objective responses occurred in 2 out of 6 breast cancer patients with BRCA mutations (33%) [56]. In a subsequent study presented at the 2013 SABCS, patients with solid tumors, including 18 with BRCA‐associated breast cancer, were treated with talazoparib from 900–1100 μg/day. Of these 18 patients, 1 had complete response, 6 had PR, and 5 had SD for at least 12 weeks. Notably, four of the BRCA‐associated breast cancer patients enrolled in the trial had not responded to prior platinum, none of which subsequently responded to talazoparib [57].

In a recently presented trial at the 2016 ESMO Annual Meeting, Litton et al. presented a 13‐patient neoadjuvant pilot trial. Patients who were to receive neoadjuvant chemotherapy and had a known BRCA mutation were enrolled to receive 2 months of talazoparib prior to physician's choice systemic chemotherapy. Given the concerns over delaying systemic therapy, the primary endpoints were evaluation of toxicity and ability to enroll this pilot trial in 2 years. Within 8 months, 13 patients were enrolled and able to tolerate the single agent with no reported grade IV toxicities. Median tumor shrinkage after 2 months of talazoparib was 88%. The study was discontinued and expanded to a phase II 6‐month‐of‐talazoparib‐alone neoadjuvant trial to estimate pCR and toxicity [58].

Multiple upcoming studies will evaluate talazoparib in patients with BRCA‐associated breast cancer including the following: the phase II ABRAZO study randomizing patients with BRCA‐associated locally advanced or MBC to one of two cohorts, those previously responding to a platinum‐containing regimen for MBC or those without prior platinum therapy but having previously received more than two prior chemotherapy regimens for MBC [59]; the phase III EMBRACA trial also in BRCA mutated patients with locally advanced and MBC comparing talazoparib given 1 mg/day in 21 day cycles versus physician's choice (capecitabine, eribulin, gemcitabine, or vinorelbine) [60]; the phase II single‐center, nonrandomized, multi‐cohort trial evaluating the use of talazoparib in patients with advanced solid tumors without curative therapeutic options, specifically looking at cohorts including BRCA somatic mutations, BRCA somatic deletions, mutations or homozygous deletions in other BRCA pathway genes including ATM, PALB2, NBS1, and Fanconi Anemia genes [61]. Finally, a phase 2 clinical trial evaluating use of neoadjuvant talazoparib in patient with BRCA associated invasive breast cancer is ongoing [62].

PARP Combinations

There have been a large number of studies evaluating the use of combination therapy involving PARP inhibitors. Interestingly, the different PARP inhibitors have shown divergent safety profiles in combination studies, with veliparib combinations showing improved tolerability overall as compared with olaparib combinations. Additionally, the diverse mechanisms of PARP inhibition have been theorized to lead to synergy with different cytotoxic agents. Specifically, veliparib, functioning primarily through suppression of the catalytic activity of PARP, has been shown to enhance the activity of topoisomerase I targeted inhibitors and DNA‐damaging agents, while PARP‐trapping agents, such as olaparib, talazoparib, rucaparib, and niraparib, have been suggested to synergize with alkylating agents. Ongoing clinical trials of combination therapy with PARP inhibitors are summarized in Table 1.

Table 1. Ongoing clinical trials of combination therapy with PARP inhibitors.

Abbreviations: AC, adriamycin/cyclosphosphamide; ASCO; American Society of Clinical Oncology; CR, complete response; IV, intravenous; MBC, metastatic breast cancer; ORR, overall response rate; PARP, poly (ADP‐ribose) polymerase; PR, partial response; TNBC, triple‐negative breast cancer.

Autologous Stem Cell Transplantation

As noted previously, multiple studies have shown high sensitivity of BRCA‐mutated breast cancer cells to DNA‐damaging agents. In this setting, a study was conducted evaluating high‐dose chemotherapy followed by autologous hematopoietic stem cell transplantation in BRCA‐mutated versus BRCA wild type or untested patients with MBC. Evaluating a total of 235 patients, 15 (6.4%) with known BRCA mutations, with a median follow‐up of 53.3 months, the authors noted that at 5 years, 69% of BRCA‐mutated patients remained alive versus 41% for patients without known BRCA mutations. The authors then adjusted for other prognostic factors using a Cox regression multivariate analysis, finding those patients without known BRCA mutations to have a significantly worse OS as compared with BRCA‐mutated patients with a hazard ratio of 3.08 [63].

Cowden Syndrome (PTEN)

Patients with germline mutations in the tumor suppressor gene PTEN, located on 10q23.3, have an increased lifetime risk of not only breast cancer, but also uterine, renal cell, colorectal, and thyroid (follicular or papillary) cancer, as well as melanoma. The estimated lifetime risk of female breast cancer has been found to be as high as 87% in these patients [64]. Few specific treatment options exist for breast cancer patients with Cowden syndrome, but by virtue of the role of PTEN in the PI3 kinase (PI3K) signal transduction cascade, inhibitors targeting the PI3K/AKT/mTOR pathway are attractive candidates in this patient subset.

While clinical trials of PI3K/AKT/mTOR inhibitors in patients with Cowden Syndrome are limited, there have been several early phase studies evaluating safety and efficacy of these targeted inhibitors in patients with Cowden Syndrome. A pilot study of the mTOR inhibitor sirolimus in patients with Cowden syndrome with germline mutations in PTEN included 18 total patients from 16 families, with 8 patients with breast involvement and 5 patients with breast, skin, thyroid, gastrointestinal (GI) polyps, and central nervous system involvement. Additionally, 4 of the 18 patients had a history of breast cancer. Treatment with sirolimus in this pilot population was found to be well‐tolerated and associated with decreased mTOR signaling in addition to improvement in symptoms, skin and GI lesions, and cerebellar function [65]. Notably, modest activity of rapamycin as a single agent in many common cancers has been theorized to be due in part to feedback activation of AKT through insulin receptor substrate 1 (IRS1) or through direct phosphorylation at Ser473 by TOR2. A preclinical study looking at breast cancer cells treated with rapamycin found that in vitro pretreatment with the natural phytoalexin, resveratrol, prevented mTOR inhibitor‐related rebound activation of AKT, suggesting combination treatment may be appropriate in treatment of breast cancers with mTOR activation [66]. Use of inhibitors targeting upstream components of the PI3K/AKT/mTOR pathway, such as dual PI3K‐mTOR inhibitors, PI3K inhibitors, AKT inhibitors, and mTOR complex catalytic site inhibitors, can also abrogate the feedback AKT activation seen with mTOR inhibition. A phase I study of the selective PI3K inhibitor BEZ235 in patient with advanced, unresectable solid tumors presented at the 2010 ASCO Annual Meeting included 13 patients with breast cancer. In addition to being well‐tolerated, of the two patients with PR noted with BEZ235 treatment, one patient had Cowden syndrome and one patient had breast cancer [67].

Li‐Fraumeni Syndrome (TP53)

Li‐Fraumeni is an autosomal dominant syndrome with near complete penetrance associated with germline mutations in the TP53 gene, which codes for a transcription factor involved in proliferation, apoptosis, and genomic stability. Patients with Li‐Fraumeni syndrome have an increased lifetime risk of developing cancer by age 50 of 68% in men and 93% in women. The most common cancer found in these patients in women is breast cancer, with studies suggesting that approximately 1% of all breast cancers are secondary to germline TP53 mutations [68]. While detailed characteristics of breast cancers associated with TP53 mutations are not fully know, it is suggested that these breast cancers are usually early onset with an increased rate of HER‐2/neu positivity [69]. Other cancer risks associated with TP53 mutations include sarcoma, brain cancer, leukemia, and adrenocortical carcinoma.

While no specific systemic treatment recommendations are standard of care in patients with cancer with Li‐Fraumeni syndrome with germline TP53 mutations beyond minimizing radiation exposure, somatic mutations in the TP53 gene have been linked with decreased response to both radiotherapy as well as DNA‐damaging cytotoxic agents, which are considered to induce p53‐dependent apoptosis [70], [71]. To this end, TP53 has been suggested for use as a biomarker in multiple cancer types with prognostic value in assessing outcome to cytotoxic agents. One 2014 study evaluated 36 patients with primary operable esophageal cancer treated neoadjuvantly with cisplatin and fluorouracil, noting that those patients with mutant TP53 had decreased OS (16.7%) with neoadjuvant treatment as compared with patients with normal TP53 marker status (55.6%), suggesting TP53 status predicts response to chemotherapy. In a study of 67 breast cancer patients treated neoadjuvantly with either a DNA‐damaging therapy (FEC) or a microtubule stability therapy (paclitaxel), treatment failure in the FEC group was found to be related to the presence of TP53 gene mutations. Consistent with the proposed induction of p53‐dependent apoptosis with DNA‐damaging cytotoxic agents, apoptosis in those patients treated with FEC was found almost exclusively in tumors with normal p53. Alternatively, paclitaxel treatment failure was more commonly seen in patients with normal p53 status, with the authors theorizing that lack of G1 arrest due to p53 deficiency may be beneficial for those patients with breast cancer with TP53 gene mutations treated with paclitaxel. Overall, this study suggested use of a microtubule stability cytotoxic agent such as paclitaxel in patients with p53 deficient tumors [72].

Several novel agents have been evaluated preclinically and found to radiosensitize p53‐defective human tumor cells, including a Wee1 kinase inhibitor and a chk1 kinase inhibitor. When evaluated in vitro in multiple tumor cell lines, including breast cancer cells, the Wee1 kinase inhibitor, MK‐1775, was found to radiosensitize p53‐defective human tumor cells, supporting a potential role of MK‐1775 in combination with radiation [73]. The increased cytotoxicity in p53‐defective tumor cell lines has been theorized to be due to inhibition of the G2 checkpoint in cells with a defective G1 checkpoint [74]. Based on further observations, defective DNA damage response has also been theorized to be a mechanism of cytotoxicity and a phase I study of AZD1775 (MK‐1775) in patients with refractory solid tumors found a PR in two patients with BRCA mutations, one with head and neck cancer and one with ovarian cancer [75]. Similarly, MK‐8776, a chk1 kinase inhibitor, was found to radiosensitize p53‐defective tumor cells in vitro using cells derived from non‐small cell lung cancer and head and neck squamous cell carcinoma. The mechanisms for radiosensitization were noted to include inhibition of double strand break repair, as well as termination of the G2 block [76].

When evaluated in vitro in multiple tumor cell lines, including breast cancer cells, the Wee1 kinase inhibitor, MK‐1775, was found to radiosensitize p53‐defective human tumor cells, supporting a potential role of MK‐1775 in combination with radiation

Peutz‐Jeghers Syndrome (STK11)

Peutz‐Jeghers syndrome (PJS) is an autosomal dominant syndrome with nearly complete penetrance associated with germline mutations in LKB1 (STK11), located on 19p13.3, which encodes a multifunctional serine/threonine kinase. Women with PJS have a 9.9‐fold increased relative risk for cancer, most notable for gastrointestinal cancer (response rate [RR] = 151) and breast cancer (RR = 20.3) [77]. In another study by Hearle et al in 2006, breast cancer risk was 15% by age 50, 33% by age 60, and 57% by age 70 [78]. Other cancer risks associated with STK11 mutations include those of the pancreas, lung, gonads, and adenoma malignum of the cervix.

While patients with malignancies in the setting of PJS currently receive standard systemic treatments, the mechanism of LKB1 mutation leading to aberrant mTOR activity suggests a potential targeted therapy in these patients. Despite this knowledge, there is limited data to suggest the safety and efficacy of mTOR‐targeted treatments in PJS patients. In fact, two pilot studies of mTOR inhibitors in PJS patients were unsuccessful, with a study of everolimus in patients with PJS with advanced malignancies withdrawn prior to enrollment [79] and a phase II study of everolimus in PJS patients to evaluate safety and efficacy in diminishing large gastrointestinal polyps ending early due to low enrollment of only three patients [80]. While there is limited data in PJS patients, mTOR inhibitors have been used in patients with tuberous sclerosis, an autosomal dominant disorder leading to growths in multiple organs but particularly sub‐ependymal giant cell astrocytomas (SEGA) of the brain and angiomyolipomas (AML) of the renal tract. These growths arise in the setting of germline mutations in tuberous sclerosis 1 or 2 genes that encode proteins functioning downstream of STK11, ultimately leading to constitutive mTOR signaling similar to that seen in PJS patients. In the EXIST‐1 trial, treatment of patients with SEGA associated with tuberous sclerosis complex with the mTOR inhibitor everolimus was associated with at least a 50% reduction in SEGA volume in 35% of patients, as compared with none in the placebo group [81]. The EXIST‐2 trial subsequently showed everolimus to be associated with a 42% response rate for AML associated with tuberous sclerosis complex with an acceptable safety profile [82]. Similarly, rapamycin has been found to be effective at reducing the size of AML associated with tuberous sclerosis complex, however notably showing growth in lesions after discontinuation, suggesting the persistent need for mTOR inhibition in these patients [83], [84].

Hereditary Diffuse Gastric Cancer Syndrome (CDH1)

Hereditary diffuse gastric cancer syndrome (HDGCS) is an autosomal dominant syndrome due in approximately 30% of patients meeting clinical criteria to pathogenic E‐cadherin protein (CDH1) gene germline mutations. In addition to diffuse gastric cancer, patients with HDGCS are at an increased risk of lobular breast cancer. A 2001 study by Pharoah et al. found an estimated cumulative risk of gastric cancer by age 80 of 67% in men and 83% in women. Women additionally had a cumulative risk of 39% for breast cancer [85].

Although no specialized treatments exist for breast cancer patients with HDGCS, in vitro data has shown increased resistance to taxol in cells with HDGCS‐related germline mutations in E‐cadherin [86]. While this suggests limitations in current standard of care therapies for patients with breast cancer in the setting of HDGCS, other in vitro data suggests potential targeted therapies that could be employed in this patient population. Specifically, E‐cadherin mutations associated with HDGCS have been shown to lead to abnormal activation of the epidermal growth factor receptor (EGFR) signaling pathway with accompanying increased cell motility. When treated in vitro with EGFR inhibitors, decreased migratory behavior is noted, suggesting a potential role for EGFR inhibitors in these patients [87], [88]. Further in vitro studies have shown E‐cadherin inactivation to be associated with aberrant Notch‐1 activation and Bcl‐2 overexpression, suggesting further potential targeted therapeutic strategies [86].

Conclusion

Patients with invasive breast cancer due to hereditary breast cancer syndromes constitute a unique patient population with opportunities for individualized, rationally targeted systemic therapies. The most common syndrome leading to increased risk of breast cancer, BRCA‐related breast cancer syndrome, involves germline deleterious mutations in BRCA1 and BRCA2. These patients display particular sensitivity to DNA‐damaging agents such as platinum agents, as well as PARP inhibitors, which by virtue of their blocking the formation of ADP‐ribose polymers at the site of single strand breaks, prevents the recruitment of DNA‐damage repair complexes. Notably, PARP inhibitor activity has been seen both in monotherapy and combination, with varying levels of toxicity and efficacy of the multiple different PARP inhibitors, in part due to differing mechanisms of PARP inhibition, including suppression of the PARP catalytic activity and PARP trapping. While promising data has been shown with PARP inhibitors in BRCA mutated breast cancer, multiple mechanisms of resistance, including secondary mutations leading to restoration of BRCA functionality, mutations in p53 binding protein 1 leading to DNA damage repair rewiring, overexpression of P‐glycoprotein leading to increased drug efflux, and epigenetic silencing leading to decreased PARP expression, are currently under investigation [39]. Other unique therapies trialed in BRCA‐mutated invasive breast cancer patients include trabectedin, given preclinical data suggesting specific activity against nucleotide excision repair intact or homologous recombination repair deficient metastatic breast cancer, and lurbinectedin, which induces the formation of double strand breaks, thus leading to activity against platinum resistant tumors and homologous recombination deficient cell lines. Notably, patients with germline BRCA mutations have been found to have poorer outcome with taxane therapy as compared with control, non‐BRCA‐mutated patients with invasive breast cancer.

Other hereditary breast cancer syndromes leading to invasive breast cancer include Cowden syndrome (germline PTEN mutation) and Peutz‐Jeghers syndrome (germline LKB1 mutation), both of which lead to aberrant signaling in the PI3K/AKT/mTOR pathway, suggesting potential use of targeted therapies specific to this pathway. Hereditary diffuse gastric cancer syndrome, associated with CDH1 gene germline mutations, also leads to an increased risk of lobular breast cancer, and preclinical data suggests abnormal activation of the EGFR‐signaling pathway—as well as aberrant Notch‐1 activation and Bcl‐2 overexpression—in these patients, suggesting possible rationale therapeutic strategies in this patient cohort. Finally, Li‐Fraumeni syndrome, associated with germline TP53 mutations, is associated with increased risk of invasive breast cancer, but to date no specific treatment recommendations are standard of care in patients with cancer with Li‐Fraumeni syndrome with germline TP53 mutations beyond minimizing radiation exposure.

Author Contributions

Manuscript writing: Amanda Parkes, Banu K. Arun, Jennifer K. Litton

Final approval of manuscript: Amanda Parkes, Banu K. Arun, Jennifer K. Litton

Disclosures

Jennifer K Litton: Novartis, Genentech, Medivation, GlaxoSmithKline (RF); Banu Arun: Abbvie, PharmaMar (RF). The other author indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board