Evolving Paradigms in Triple-Negative Breast Cancer: Treatment

February 27, 2017
Evolving Paradigms, Triple Negative Breast Cancer, Volume 2, Issue 1

Breast cancer is a leading cause of cancer death in women, with 12% to 20% of cases classified as triple-negative breast cancer.

Treatment of TNBC

Surgical treatment of TNBC

With a lack of ERs, PRs, and HER2 as potential treatment targets, TNBC cannot be treated with conventional breast cancer hormonal-based or trastuzumab-based treatments. Patients are typically treated with a combination of surgery, chemotherapy, and radiation.1In fact, an emphasis is placed on local and regional treatment, such as radiation and surgery, for early-stage disease.

With the exception of advanced and inoperable metastatic breast cancer, breast-conserving surgery (BCS), simple mastectomy, or radical mastectomy have remained initial treatments for breast cancer. However, 75% to 85% women with the BRCA1 mutation have the TNBC subtype that influences the type of surgery selected. Although mastectomy to reduce risk of recurrence is preferred over BCS, it is not clear if surgical recommendations should be altered for patients with TNBC. For example, a retrospective study of patients with TNBC treated with mastectomy compared with BCS assessed overall survival (OS) and DFS. The investigators concluded that BCS in TNBC should be considered in selected patients.15

Research also indicates that BCS is an appropriate option in TNBC because recurrence after BCS is no higher than that for other types of breast cancer. Furthermore, when choosing a surgical approach, the preferences of the patient, as well as clinical and pathologic variables, should be considered.

Radiation Therapy in TNBC

Radiation therapy (RT) after BCS has shown to be beneficial in the long term for patients with breast cancer. Therefore, RT is indicated in TNBC after mastectomy or after BCS despite some controversy.1For example, one study suggested possible radio resistance due to ER-negative status in TNBC. This is thought to be related to more time available for DNA damage repair from radiation in ER-negative cells.15

Conversely, as a result of the aggressive and rapid-growing nature of TNBC, RT after BCS may not be as effective as mastectomy. Because many patients with TNBC have a mutated BRCA1 gene, which lacks double-stranded DNA repair, theoretically they should be very radiosensitive. Some researchers suggest that RT of the surrounding breast tissue after BCS could eliminate remaining mutated BRCA1 foci and lessen the risk of local or regional recurrence.1

Therefore, in light of the conflicting data, either BCS or modified radical mastectomy is considered an acceptable choice in early stages of TNBC. Research has indicated an improvement in 5-year relapse-free survival (RFS) and OS with the addition of RT to modified radical mastectomy. Treatment may also include neo-adjuvant chemotherapy with or without RT. Retrospective data indicate that patients treated with chemotherapy and reclassified as stage I or II may not benefit from RT compared with patients with persistent nodal disease. However, with a lack of controlled, prospective studies, the standard of care continues to involve RT after neoadjuvant chemotherapy.16

Chemotherapy in TNBC

Although TNBC is considered to be quite responsive to chemotherapy, the combination of aggressive behavior and metastatic course, with a short disease-free period, can make the prognosis of TNBC poor. Chemotherapeutic targets include cell proliferation with anthracyclines, p53 with taxanes, and DNA repair complexes with platinum and taxane regimens.1

Platinum-based regimens

Platinum therapies, such as cisplatin, lead to double-strand breaks and cross linking of double-stranded DNA. In cases of the BRCA1 mutation, platinum therapy leads to cell death by inability to repair the damaged DNA. With research indicating that TNBC may be more sensitive to platinum-based regimens, there is a new push for understanding how to incorporate these regimens into breast cancer treatments. For example, 21% of patients with TNBC treated with neoadjuvant cisplatin in a phase II study had complete response (CR) with 50% showing a good response.17

Ezzat and colleagues combined cisplatin with paclitaxel and found a 63% partial response rate and a 28% CR rate in patients with HER2-negative, ER-negative breast cancer. Likewise, patients with TNBC treated with a combination of neoadjuvant cisplatin, paclitaxel, and epirubicin for 8 weeks had a 65% pathologic CR.1

Taxane-based regimens

Taxanes target genetic instability and have been shown to be beneficial in both TNBC and basal-like breast cancer. Meta-analyses show benefit with adjuvant treatment with taxanes, and some research indicates that taxanes may be more effective in receptor-negative cancers. Furthermore, taxanes were associated with a CR of 45% in basal-like breast cancer compared with 6% in luminal subtypes. In addition, patients with TNBC who were treated with 4 cycles of 5-uorouracil, epirubicin, and cyclophosphamide (FEC), followed by 8 weeks of paclitaxel, demonstrated more benefit than those treated with 6 cycles of FEC.17It is unclear if anthracyclines alone are beneficial to patients with TNBC, but evidence supports the combination of anthracyclines with taxanes. Despite evidence of TNBC benefitting from regimens that include taxanes, in vitro studies indicate a possible resistance to taxanes in BRCA1-mutated tumors.1​

Anthracycline-based regimens

Anthracycline-based chemotherapy targets and destabilizes DNA and thus is considered important in the treatment of breast cancer and TNBC. Liedtke et al compared patients with TNBC with patients with breast cancer but not with TNBC and found that treatment with an anthracycline-containing regimen yielded a 3-year disease-free survival that was similar in both groups (CR 94% versus 98%). However, patients with TNBC who did not meet a pathologic CR had worse outcomes for 3-year disease-free survival than did patients who did not have TNBC (68% vs 88%, P = .0001). This was thought to be secondary to an earlier relapse rate, which was much higher in the TNBC group that had incomplete response.13

Likewise, younger patients with TNBC who received high-dose, rapidly cycled treatments containing epirubicin and cyclophosphamide versus a conventional adjuvant regimen demonstrated a 71% 5-year disease-free survival versus 26% in the conventional group.

When taking into the account the heterogeneity of TNBC, it is not clear if TNBC andBRCA1carriers are as sensitive to anthracycline-containing treatments. For example, when looking at patients with TNBC treated with FEC, those who were carriers ofBRCA1had a CR rate of 17% versus 42% in non-BRCA1carrier TNBC. Finally, pooled data indicate no benefit with anthracycline-containing regimens in HER2-negative breast cancers. Subgroup analyses of studies using anthracyclines in TNBC show mixed results in basal-like breast cancers and TNBC.1

Adjuvant chemotherapy for TNBC

Because there are no approved targeted treatments for TNBC, adjuvant treatment with an anthracycline-based and taxane-based regimen for early-stage TNBC is common, based on recommendations of the National Cancer Comprehensive Network (NCCN). The 5-year risk of recurrence in small, node-negative TNBC is low and therefore has a lower treatment benefit. The algorithm-based NCCN guidelines for node-negative TNBC do not recommend adjuvant chemotherapy for breast tumors 0.5 cm or less. For tumors 0.6 cm to 1.0 cm, chemotherapy can be considered after a balanced discussion with the patient. For patients with lymph node-positive TNBC with any size tumor, adjuvant chemotherapy with an anthracycline/taxane-based regimen is recommended in the guidelines (FIGURE 3).16

Figure 3. General treatment algorithm for adjuvant chemotherapy for early TNBC.

Adapted from www.NCCN.org.

Neoadjuvant chemotherapy in TNBC

Whether or not to consider neoadjuvant chemotherapy in TNBC takes into account several factors. The first involves the resectability of the tumor and lymph nodes, with the goal of having clear margins. The second factor is the likelihood that the tumor could be reduced enough to allow for BCS versus mastectomy.12

Neoadjuvant chemotherapy allows for treatment efficacy more rapidly than conventional adjuvant therapies. Interestingly, studies comparing neoadjuvant treatment in non-TNBC and TNBC have demonstrated higher response rates and better predicted long-term outcomes in the TNBC groups. Factors that predict a favorable response rate to neoadjuvant treatment are Ki-67 expression and ER negativity.1

TNBC and basal-like breast cancers have higher CR rates to neoadjuvant breast cancer treatments than luminal breast cancers, despite overall poorer outcomes.12There is an association between survival outcomes and the rate of CR; therefore, CR is becoming a valuable endpoint when assessing the efficacy of neoadjuvant chemotherapy. Retrospective analysis of neoadjuvant anthracycline/ taxane-based treatment in patients with TNBC demonstrated higher CR rates (38% vs 12%) than in patients with non-TNBC and longer disease-free survival in patients with evidence of CR. Finally, encouraging results and new strategies to manage side effects have led to interest in platinum-based treatments for neoadjuvant treatment of TNBC.

Chemotherapy in metastatic and advanced TNBC

TNBC is characterized as metastatically aggressive with metastases commonly to the brain, liver, and lung. Because there are no standard treatments for metastatic TNBC, treatment planning should consider the patient’s performance status and preference, comorbidities, prior treatments, and disease characteristics.1

Several systemic treatment options are available for advanced and metastatic HER2-negative breast cancers. Treatment of advanced TNBC follows the same algorithm as for other advanced subtypes, with single chemotherapy agents for asymptomatic disease and combination regimens for rapid progression and symptomatic forms.

Based on a randomized phase III study, the first-line treatment is paclitaxel.12However, several trials have indicated no specific benefit of using taxanes in TNBC over other breast cancer types.1

Platinum agents have shown efficacy inBRCA1cancers and also have some response in sporadic TNBC. For instance, in first-line and second-line treatment for advanced TNBC both cisplatin and carboplatin demonstrated a 30% response rate.12

Investigational Targets for TNBC

Because there are no approved targeted treatments for TNBC, chemotherapy continues to be the primary treatment of TNBC.10Identifying potential treatment targets for TNBC may open the door to more personalized treatment.

Glycoprotein nonmetastatic melanoma b (gpNMB)

Glycoprotein nonmetastatic melanoma B overexpression is associated with TNBC, along with other cancers, and is associated with a decrease in OS and disease-free survival and poor prognosis. In breast tissue, gpNMB expression is thought to regulate the ability of cells to allow tumor growth, metastasis, and recruit vasculature.

Glembatumumab vedotin (CDX-011) is an antibody-drug conjugate that binds to gpNMB and releases auristatin E, which leads to cell cycle arrest and subsequent apoptosis. Research has indicated that the efficacy of glembatumumab depends on the amount of gpNMB expressed on the surface of the targeted cell.

Initial studies involved locally advanced or metastatic breast cancer and unresectable melanoma. Patients with breast cancer who received the maximum tolerated dose every 3 weeks demonstrated 62% tumor shrinkage. Furthermore, the progression-free survival (PFS) for patients with gpNMB-positive breast cancer was 17.3 weeks versus 9.1 weeks.

The EMERGE trial aimed to understand the safety and efficacy of glembatumumab in gpNMB-positive metastatic breast cancer with previous intensive treatment.10Patients included in the study had at least 5% of epithelial or stromal cells expressing gpNMB. They were randomly assigned to either investigators’ choice of chemotherapy (n = 41) or to glembatumumab (n = 83). The investigators concluded that glembatumumab vedotin was well tolerated. However, the primary endpoint was not met for all the study subjects, with overall objective response rate of 6% in the glembatumumab group versus 7%. For tumors with overexpression of gpNMB (greater than 25%) the overall objective response was 30%. The glembatumumab group had less hematologic toxicity but more rash, neuropathy, alopecia, and pruritus.14

The METRIC trial is recruiting study participants to demonstrate the effectiveness and safety of glembatumumab vedotin in advanced TNBC (NCT01997333). This randomized, multicenter trial is looking at PFS as the primary outcome measure. Secondary outcomes include objective response rate, OS, adverse events (AEs), and duration of response. Patients will be randomized to either glembatumumab vedotin or to capecitabine.15

mTOR/AKT/PI3K Pathways and Inhibitors

Mutations inP53,PTEN, andPI3Kare associated with TNBC and inhibition studies indicate that targeting mTOR, a serine-threonine protein kinase, may have more antitumor effect than therapies that target tyrosine kinase receptors. The PI3K/AKT pathway regulates the mTOR complexes, and its activation is associated with cellular transformation. Likewise, overexpression of this pathway is related to poor cancer prognosis. Other effects of this pathway include cell migration, transcription, cell cycle progression, and survival.9

Loss of PTEN is associated with mTOR activation, which is common to TNBC tumors. Two mTOR inhibitors, everolimus and temsirolimus, have been studied in HER2-negative and TNBC (FIGURE 4).16

Figure 4. The distribution of TNBC subtypes.

Adapted from The Cancer Genome Atlas. Bar graphs illustrate the percentage of subtype in TNBC.

In a phase II trial of everolimus and carboplatin for patients with metastatic TNBC, Singh et al found the clinical benefit rate was 36% (95% CI, 21.1-57.4) with an OS of 16.6 months (95% CI, 7.3 months to not reached) and median PFS of 3 months (95% CI, 1.6- 4.6 months). The reported AEs were thrombocytopenia, neutropenia, and anemia.17

A study by Yunokawa et al found that everolimus had favorable activity against basal-like TNBC cell lines. Furthermore, there was an association between CK5/6 and EGFR markers and a positive response to everolimus in TNBCs.18

There are reports that activation of mTOR might be associated with resistance to cisplatin, which in theory could be avoided with the use of everolimus.1

A study looking at temsirolimus with neratinib in patients with metastatic TNBC or HER2-amplified breast cancer is ongoing (NCT001111825). The aim of the open-label study is to identify the benefit and safety of this combination in TNBC or HER2-amplified breast cancer that progressed despite treatment.19

Finally, therapies that target both PI3K and mTOR, such as BEZ235, demonstrate targeted inhibition and tolerability. Furthermore, TNBC with LAR or mesenchymal-like expression,PIK-3CAmutation, and loss of PTEN may be more sensitive to these dual inhibitors.

The most common AEs reported with these inhibitors are rash, nausea, diarrhea, mucositis, and hyperglycemia.20

Vascular Endothelial Growth Factor Pathway

The highly proliferative nature of TNBC requires that it has persistent angiogenesis and therefore it is logical that these tumors would have higher expression of VEGF. Bevacizumab is an antiVEGF monoclonal antibody.2

Sunitinib was evaluated in a phase II study for metastatic patients with breast cancer who were previously heavily treated with taxanes and anthracyclines. The ORR was 11%, but 56% of patients required dose adjustment secondary to AEs such as nausea, diarrhea, fatigue, and mucosal in ammation.31The patients with TNBC demonstrated a relative response rate of 15%.2

Another multi-kinase inhibitor is sorafenib, which has activity against tumor angiogenesis and proliferation.2A trial comparing capecitabine and sorafenib with capecitabine and placebo in patients with advanced or metastatic HER2-negative breast cancer found an improvement in PFS in the sorafenib arm. However, the dose used in the study was associated with rash, diarrhea, neutropenia, hypertension, and mucosal in ammation, all AEs that required discontinuation in 20% of patients versus 9% in the placebo arm.32Trials specific to the TNBC population are underway for two other TKIs: apatinib (NCT01176669) and cediranib (NCT01116648).33,34

O’Shaughnessy and colleagues conducted a meta-analysis of patients with TNBC with first-line treatments and bevacizumab. Patients demonstrated a 35% decrease in risk of death or disease progression, and their response rate improved by 19%.26Furthermore, when bevacizumab and docetaxel were compared with placebo and docetaxel in HER2-negative metastatic breast cancer, patients demonstrated an increase in PFS.27Subgroup analysis of patients with TNBC revealed an 8.2-month median PFS in the bevacizumab plus docetaxel group. Adding bevacizumab to the docetaxel regimen did not worsen the toxicity profile; however, there were more grade 3/4 AEs in the bevacizumab group, including febrile neutropenia, neutropenia, and hypertension.2

In contrast, the RIBBON-1 trial comparing chemotherapy (capecitabine, taxane-based, or anthracycline-based) with bevacizumab showed improvements in PFS when used as a first-line treatment for metastatic breast cancer.28However, these results did not show clear benefit with the addition of bevacizumab in patients with TNBC. The RIBBON-2 trial used chemotherapy plus bevacizumab in patients with metastatic breast cancer who had already received cytotoxic therapy. The TNBC group demonstrated an increase in median PFS of 3.3 months (P = .0006).2The use of bevacizumab with anthracycline-based and taxane-based regimens in the neoadjuvant setting resulted in signicantly increased CR rates in the GeparQunto Trial.29

Recently developed therapies, such as ramucirumab, bind to VEGFR2 and therefore may have more complete angiogenesis inhibition.2A phase III clinical trial of docetaxel plus ramucirumab is underway for patients with HER2-negative metastatic, unresectable, or locally recurrent breast cancer (NCT00703326).30

Two therapies target cell surface-receptor tyrosine kinases, such as VEGFRs, which are important for tumor angiogenesis. Sunitinib and sorafenib are anti-VEGFR tyrosine kinase inhibitors (TKIs) that have been studied in the TNBC subgroup.2

Fibroblast growth factor and receptor

FGFR is being investigated in breast cancer because of its role in cellular brous proliferation in TNBC. It involves the RAS-RAF— mitogen–activated protein kinase cascade.11Molecular profiling of TNBC demonstrated that a subgroup had amplification of FGFR2.35Furthermore, these cell lines were very susceptible to FGFR inhibition.11

Lucitanib (E-3810) is an inhibitor of VEGFR1 and FGFR1, and preclinical models highlight its activity against angiogenesis.36Using xenografts of advanced TNBC, Bello and colleagues demonstrated lasting tumor regression with a combination of E-3810 and paclitaxel.37

Epidermal growth factor receptor

The EGFR is another tyrosine kinase within the HER2 family. More than 50% of TNBC and 65% to 72% of basal-line breast cancer have dysregulation of the EGFR pathway.11Adding cetuximab, a monoclonal antibody that targets EGFR, to cisplatin doubled the ORR in patients with metastatic TNBC.38Likewise, patients demonstrated a longer median PFS and similar OS to cisplatin alone.2The most commonly reported AEs were fatigue, neutropenia, and an acne-like rash.38

In stage IV TNBC, cetuximab plus carboplatin was associated with a response in only 20% of patients. The investigators suggested that although TNBC involves activation of the EGFR pathway, there may be an alternative mechanism for this pathway activation.39

Erlotinib used for neoadjuvant treatment of TNBC with carboplatin and docetaxel had a 40% pathologic CR rate. Furthermore, there was a strong correlation between CR and BRCA mutations, with 100% CR in BRCA-mutated patients versus 27% in non-BRCA—mutated patients (P = .006).2 Neoadjuvant treatment of TNBC with ixabepilone with and without cetuximab is being studied (NCT01097642).40

Combination treatment with MET and EGFR inhibitors MGCD265 and erlotinib or crizotinib and erlotinib demonstrated effective abrogation of tumor growth and resulted in significantly decreased treatment response variability compared with monotherapy.41This study suggests that combined EGFR and MET inhibition may be beneficial for treating patients with TNBC.

Androgen receptors in TNBC

Evidence suggests that 10% to 35% of all patients with TNBC have AR-positive gene expression.2A recent study found that the AR is more likely to be co-expressed in ER-positive breast cancer compared with ER-negative breast cancer (56.0% vs 28.1%).12Furthermore, AR expression was significantly correlated with decreased DFS in TNBC. In ER-positive patients, AR expression is thought to inhibit proliferation. However, in ER-negative patients AR expression may promote tumorigenesis. The efficacy of bicalutamide was tested on TNBC cell lines, and the agent showed that the LAR subtype was more sensitive.11A clinical trial involving bicalutamide in patients with metastatic ER-negative, PR-negative, or AR-positive breast cancer is ongoing (NCT00468715).42

Results of a phase II trial investigating enzalutamide in 118 women with AR-positive TNBC were presented at the 2015 ASCO Annual Meeting.43Of 75 evaluable patients, 35% achieved a clinical benefit at 16 weeks of therapy and 29% achieved clinical benefit at 24 weeks. A total of 7 partial responses and 2 complete responses were observed. Median PFS was 14.7 weeks.

Poly (adenosine diphosphate-ribose) polymerase inhibitors

Poly (adenosine diphosphate-ribose) polymerase (PARP) enzymes, especially PARP1, are involved in single-strand DNA break repair, and loss of this function leads to accumulation of these breaks. Furthermore, pathways that are related to BRCA1 and BRCA2 also repair the breaks. Preclinical results of inhibition of PARP1 demonstrated that TNBC cells are more sensitive than non-TNBC cells. Several PARP inhibitors are being evaluated as possible therapeutic options in TNBC. An oral PARP inhibitor, olaparib, has shown activity against BRCA-mutated tumors and as a single agent for advanced breast cancer. The reported AEs were mild and included nausea, vomiting, fatigue, and anemia.

In a phase I trial of olaparib plus paclitaxel as first-line or second-line treatment of metastatic TNBC, Dent and colleagues demonstrated an objective response rate in 3 of 9 patients in the first cohort and 4 of 10 patients in the second cohort.2However, this combination had significant interactions and high rates of neutropenia. Other AEs reported were diarrhea and nausea. The investigators suggested considering alternative dosing and scheduling.44

Olaparib is also being investigated with paclitaxel or carboplatin in patients with TNBC (NCT00707707 and NCT00516724).45,46Likewise, olaparib plus cisplatin is being investigated for use in advanced solid tumors (NCT00782574),47including locally advanced TNBC.2

Veliparib is another oral PARP inhibitor that targets both PARP1 and PARP2. A single-arm, phase II trial of the combination of veliparib and temozolomide in patients with metastatic breast cancer, including 15 patients with TNBC,2concluded that the combination was active in metastatic breast cancer with 1 CR and 2 partial responses. Progression of disease was noted in 14 patients. The reported AEs included grade 3/4 thrombocytopenia and neutropenia.48

Although its mechanism is not well understood, iniparib was developed initially as a PARP inhibitor. However, iniparib has been shown to result in cell-cycle arrest and increase the effects of DNA damaging pathways, and may have antiproliferative activity in TNBC cell lines.2

Combining gemcitabine and carboplatin, with and without iniparib, for metastatic TNBC in an open-label, phase II study demonstrated improved rate of clinical benefit (34%-56%; P = .01) and ORR (32% to 52%; P = .02). Likewise, the OS increased from 7.7 months to 12.3 months (HR for death, 0.57; P = .01). The most common AEs in both groups were neutropenia, anemia, fatigue, leukopenia, and thrombocytopenia. There was little difference in the rate of AEs between the two groups; thus, toxicity was not significantly increased with the addition of iniparib.49However, a phase III trial of the gemcitabine, carboplatin, and iniparib combination in advanced TNBC did not meet the study’s primary endpoints.2Studies of neoadjuvant treatment of TNBC with paclitaxel alone or with iniparib (NCT01204125) and treatment of TNBC with the PARP inhibitor rucaparib (NCT0104970) are underway.50,51

Src tyrosine kinase inhibitors

Overexpression of the Src tyrosine kinase is associated with more metastatic progression and invasion in breast cancer.52Dasatinib is an oral inhibitor of the Src kinases and was originally approved in leukemia with the Philadelphia chromosome.2

Pichot and colleagues demonstrated synergistic activity of doxorubicin and dasatinib in breast cancer cells and interference with migration and invasion in the triple-negative cell line.53Likewise, synergy was demonstrated with dasatinib, cetuximab, and cisplatin in TNBC cell lines.54Cell viability was reduced and apoptosis was induced more often in the 3-drug combination versus cisplatin and cetuximab.2Dasatinib is being studied as adjuvant treatment in women with ER-negative breast cancer, with the aim of preventing breast cancer in the unaffected breast (NCT01471106).55Side effects associated with dasatinib include cytopenias, bleeding, pul- monary arterial hypertension, and swelling.56

Histone deacetylase inhibitors

Histone deacetylase inhibitors, such as panobinostat, target pathways involved in apoptosis, inhibition of angiogenesis, tumor-suppressor genes, cell-cycle arrest, and invasion. In cancers other than TNBC, panobinostat has demonstrated clinical response with limited side effects.2

In TNBC cell lines, panobinostat was shown to decrease cell survival and proliferation, and block cell cycle progression. Panobinostat induced apoptosis in all TNBC cell lines except one. Mice models revealed that panobinostat reduced tumor formation.57Panobinostat is being studied for metastatic breast cancer and TNBC when combined with letrozole (NCT01105312).58

Mitogen-activated protein-kinase kinase

In TNBC mouse models, inhibition of the mitogen-activated protein-extracellular signal-regulate kinase (MEK) with PI3K/mTOR has shown activity.16Trametinib is an oral MEK inhibitor under investigation for multiple malignancies.59

A study to understand how MEK functions in TNBC is being conducted in which GSK1120212 is given for a short time (NCT01467310).60Likewise, a combination of trametinib and GSK2141795, an Akt inhibitor, is being studied in metastatic TNBC in a phase II trial (NCT01964924).61

7-Hydroxystauroporine and checkpoint kinase inhibitors

Mutations in TP53 are common in TNBC and are associated with checkpoint regulation; 7-hydroxystauroporine (UCN-01) is a checkpoint kinase 1 inhibitor.2

A phase II trial of the combination of irinotecan and UCN-01 in metastatic TNBC demonstrated limited activity and found that patients with metastatic TNBC and with TP53 mutations had a poorer prognosis.62The OS in patients with the TP53 mutation was 5.5 months versus 20.3 months without the TP53 mutation.

Dinaciclib (MK-7965) is a small molecule that inhibits CDK1, CDK2, CDK5, and CDK9.2A phase II trial investigating the use of dinaciclib versus capecitabine in patients with advanced breast cancer did not show superiority to capecitabine; however, the investigators noted that it did show some antitumor activity and was generally tolerated. The most common AEs were neutropenia, leukopenia, febrile neutropenia, and elevated aspartate aminotransferase.66This trial did not evaluate patients with breast cancer based on subpopulations, histologic markers, or molecular markers.2

A phase I study evaluating dinaciclib plus epirubicin was conducted to determine the maximum tolerated dose in patients with metastatic TNBC. The dose-limiting toxicities included febrile neutropenia, syncope, and vomiting. The investigators did not escalate the dose past the second cohort because of the toxicities. No treatment responses were noted and the accrual was stopped. The investigators concluded that the combination had substantial toxcities without evidence of efficacy for TNBC.67

Using TNBC patient—derived xenografts and three established TNBC cell lines, Rajput and colleagues looked at the efficacy of dinaciclib. They concluded that dinaciclib led to cell cycle arrest at the G2/M phase and to apoptosis.68Finally, in BRCA1 wild-type TNBC cell lines, investigators demonstrated that the use of CDK inhibition with dinaciclib led to the cancer cells being more sensitive to PARP inhibition with veliparib. The investigators noted that the combination may be an option for TNBC treatment.69

Proteasome inhibitors

Initially approved for hematologic malignancies, bortezomib is a proteasome inhibitor, with preclinical studies indicating activity against breast cancer.2Studies show that bortezomib induces apoptosis in TNBC cell lines63and suggest that bortezomib downregulates CIP2A-dependent p-Akt, which may be associated with more aggressive breast cancers.2A recent study by Song and colleagues demonstrated that delivery of bortezomib with nanoparticles increases the circulation half-life and attenuates tumor growth in the basal-like TNBC mouse model.64

Cycline-dependent kinase (CDK) inhibitors

Small interfering RNAs that target and disable survival gene expression are a potential new area for targeting TNBC cells. For example, Kren and colleagues completed a preclinical study on the use of cyclin dependent kinase 11 (CDK11) and casein kinase II for RNA interference in TNBC cells. CDK11 is a survival protein kinase involved in the regulation of RNA, mitosis, splicing, and transcription. Casein kinase II is another survival protein kinase that is involved in preventing cancer cell death.

Kren et al found that 100% of TNBC cells stained positive for CDK11. When they attempted to downregulate CDK11 and casein kinase II in the breast cancer cells, they noted a signicant decrease in messenger RNA (mRNA) and protein expression, clone survival, and cell viability. Doing so also led to induction of cell death changes.65

Dinaciclib (MK-7965) is a small molecule that inhibits CDK1, CDK2, CDK5, and CDK9.2 A phase II trial investigating the use of dinaciclib versus capecitabine in patients with advanced breast cancer did not show superiority to capecitabine; however, the investigators noted that it did show some antitumor activity and was generally tolerated. The most common AEs were neutropenia, leukopenia, febrile neutropenia, and elevated aspartate aminotransferase.66This trial did not evaluate patients with breast cancer based on subpopulations, histologic markers, or molecular markers.2

A phase I study evaluating dinaciclib plus epirubicin was conducted to determine the maximum tolerated dose in patients with metastatic TNBC. The dose-limiting toxicities included febrile neutropenia, syncope, and vomiting. The investigators did not escalate the dose past the second cohort because of the toxicities. No treatment responses were noted and the accrual was stopped. The investigators concluded that the combination had substantial toxcities without evidence of efficacy for TNBC.67

Using TNBC patient—derived xenografts and three established TNBC cell lines, Rajput and colleagues looked at the efficacy of dinaciclib. They concluded that dinaciclib led to cell cycle arrest at the G2/M phase and to apoptosis.68Finally, in BRCA1 wild-type TNBC cell lines, investigators demonstrated that the use of CDK inhibition with dinaciclib led to the cancer cells being more sensitive to PARP inhibition with veliparib. The investigators noted that the combination may be an option for TNBC treatment.69

Wnt/frizzled pathway targets

Wnt signaling, which is involved in cell proliferation, differentiation, and migration, is implicated in several cancers.2Activation of the beta-catenin/Wnt pathways is associated with poor clinical outcomes and TNBC.70Transcriptome analysis of TNBC cell lines resulted in 72 Wnt targets that were expressed more often in TNBC, and these may indicate chronic activation of the Wnt pathway in TNBC.71

Interestingly, the Wnt5a oncogene is thought to be a tumor suppressor gene and signals via the frizzled receptors. When breast cancer cells were analyzed for transcription of Wnt5a, investigators noted lower levels compared with normal cells. Likewise, in patients with ER-negative status, the lower levels of Wnt5a was associated with a poor outcome.72

A study currently recruiting patients (NCT02020291) is investigating a hexapeptide known as Foxy-5 that mimics the activity of the Wnt5a molecule, which is thought to be involved in cancer cell migration. Invasive breast cancers that have lower levels or lack the Wnt5a protein are associated with shorter recurrence-free survival. Prospective study participants include those with metastatic breast, colorectal, or prostate cancer and are screened to insure lower or absent levels of Wnt5a.73

A membrane-bound O-acyltransferase known as PORCN (Porcupine) is involved in a necessary step in the Wnt pathway.74In knock-out mice models, loss of PORCN is associated with Wnt signaling inhibition.2Research into this pathway led to the development of a PORCN inhibitor known as LGK974. In vitro and in vivo use of LGK974 has demonstrated inhibition of Wnt signaling. LGK974 has been shown to be effective in breast cancer tumor models, as well.74

Solzak and colleagues found that RNA sequencing data of seven TNBC cells demonstrated hyperactivation and overexpression of both Wnt pathways and PI3K/AKT/mTOR pathways. Likewise, they tested cell-based models against BKM120 and LGK974 combined and alone. They found that the combination produced synergy in its ability to reduce cell viability.75

A phase I, open-label, dose-escalation study of LGK974 is currently recruiting (NCT01351103). Patients included in the trial are those with pancreatic adenocarcinoma, BRAF-mutant colorectal cancer, and other cancers with documented genetic abnormalities in the Wnt pathway.76

A mechanistic study of the active chalconoid cardamonin revealed the involvement of Wnt/beta-catenin signaling in the cardamonin-induced reversal of epithelial—mesenchymal transi- tion (EMT) in a TNBC cell line.77Further data demonstrated EMT blockage and inhibition of TNBC cell invasion and tumor growth.

Prognosis, Recurrence, and Metastasis of TNBC

Despite all the current research into TNBC treatments, TNBC has a poor prognosis, is diagnosed earlier in life, is aggressive, and has a short RFS. Furthermore, the risk of dying secondary to relapse is higher in patients with TNBC.4

After an average follow-up of 96 months for a group of patients with breast cancer, Tian et al found that the TNBC phenotype was inversely associated with OS (P < .05) and that this diagnosis carries a poor prognosis.78Prognosis may also be affected by how the patient initially responds to treatment. For example, patients with TNBC treated with neoadjuvant chemotherapy had higher rates of CR compared with non-TNBC patients. However, their 3-year OS and 3-year PFS were lower. These patients had higher rates of visceral metastases and shorter survival after recurrence whereas patients with TNBC who achieved CR had a similar survival rate to those with non-TNBC.

Residual disease in TNBC after neoadjuvant chemotherapy was found to carry a worse prognosis than that in non-TNBC patients with residual disease.79Likewise, TNBC has been found to correlate with younger age, higher nuclear grade, and biological aggressiveness. Rhee et al found a relapse rate of 14.7% in TNBC versus 6.6% in non-TN breast cancer, with a 4-year RFS rate of 85.5% versus 94.2%. A shorter RFS was associated with triple-negative status, younger age, and close resection margin on multivariate analysis.80

Patients with ipsilateral breast recurrence of TNBC were found to have 5-year survival of 72.7% and disease metastasis—free survival of 48.6% despite standard chemotherapy.81Disease-specific survival at 5 years has been found to be 88% in TNBC, 98% in HR- positive/HER2-negative, and, interestingly, 86% in HR-/HER2+ patients. The study found a 5-year RFS of 84% in patients with TNBC, 95% in patients with HR+/HER2-, and 76% in patients with HR-/HER2+. When Kaplan and Malmgren adjusted the model for race, age, TNM (tumor, node, metastasis) status, and treatment, they found a recurrence risk of 2.32 for TNBC (95%CI, 1.32-4.08), with patients with HR-positive/HER2-negative as the reference group.82Finally, compared with non-TNBCs, OS after recurrence has been found to be signicantly lower in TNBC. Likewise, TNBC is associated with a significantly lower DFS compared with non-TNBC.83

Retrospective analyses of distant metastatic patterns in TNBC demonstrated that pulmonary metastasis is associated with the longest metastatic OS, 16.6 months. This is followed by bone metastasis with 16.3 months and liver metastasis with 8.9 months. Pleural and brain metastases are associated with an OS of 7.5 and 4.3 months, respectively. Location of first metastasis significantly correlated with survival in the TNBC subgroup, with the lowest survival in brain metastasis.84Metastasis to the brain has a higher incidence within the first 5 years of a TNBC diagnosis. Therefore, Chikarmane and colleagues suggest that this should help guide metastatic surveillance and imaging.85

In a cohort of women followed for approximately 8 years, in the first 5 years after diagnosis, women with TNBC had a higher likelihood of distant recurrence (HR, 2.6; P <.0001) and death (HR, 3.2; P <.001). This risk of recurrence seemed to peak at the third year after diagnosis and rapidly declined following the peak. Dent noted that although TNBC is considered an aggressive breast cancer, this may actually be temporary in the course of the disease. In this cohort, the survival time after recurrence in the TNBC group was significantly shorter at 9 months versus 20 months (P = .02).86

Finally, survival in TNBC is affected by age, with distant DFS and OS being shorter in younger patients, even with more aggressive treatment strategies. In patients older than 60 years , the average DFS was 8 years versus 4 years in patients 31 to 40 years of age.87

Movement Toward Personalized Medicine

Although TNBC is a heterogeneous group of breast cancers, there is increasing effort to subdivide it into more homogenous groups that may help lead to a more accurate prognosis and response to therapy. Le Du and colleagues propose that 5 molecular subtypes of TNBC have a high potential to drive clinical trials. These include basal-like, mesenchymal-like, immune-associated, AR overexpression in the luminal/apocrine group, and HER2-enriched.88

Because of the genetic instability of the tumors, the cancer can have molecular changes throughout the course of the disease. Therefore, it is not known if molecular classification at the time of diagnosis will continue to be predictive of response after treatment as the tumor evolves. Less invasive methods, such as analysis of circulating tumor DNA or circulating tumor cells, may be an accurate method of monitoring clinical response to treatment and personalizing the treatment as the disease progresses. As more research continues into the molecular subtypes of TNBC, there is the potential to uncover additional biomarkers (FIGURE 5).88

An additional target under research for the treatment of TNBC is the notch signaling pathway, which is involved in self-renewal, angiogenesis, proliferation, and apoptosis. TNBC xenograft models with NOTCH1 rearrangements were found to be sensitive to gamma-secretase inhibitors.89PF-03084014 is a selective gamma-secretase inhibitor that is thought to have antitumor activity by blocking activation of NOTCH receptors.90A phase II, open-label biomarker study of PF-030804014 is underway for nonmetastatic chemoresistant TNBC (NCT02338531).91

There is evidence that the JAK2/STAT3 pathway disruption may be a potential target in TNBC. Janus kinases (JAKs) are another group of tyrosine kinases involved in cell survival and growth. In basal-like breast cancer, JAK/STAT3 was found to be active, and inhibition led to decreased growth in the xenograft models.89Ruxolitinib, an IL-6/JAK/STAT inhibitor, is being studied in a phase I trial for recurrent breast cancer, with a phase II portion specifically for preoperative inflammatory TNBC (NCT02041429).92

Trop-2 is overexpressed in some epithelial cancers and involved in cell-to-cell adhesion. The FDA gave a fast track designation to IMMU-132 (isactuzumab govitecan) for progressive metastatic TNBC.89An open-label, phase II study that is not yet recruiting will look at IMMU-132 and carboplatin in patients with TNBC (NCT02161679).93

In addition to these strategies, other studies are exploring innate immunosurveillance, which has the ability to identify cancerous cells and destroy them. Cancer cells characteristically have the ability to evade the immune system. New evidence points to cytotoxic T-lymphocyte antigen inhibitors and programed cell death-1 inhibitors (PD-1) as potential targeted therapies. Furthermore, the PD-1 ligand, PD-L1, is expressed in about 20% of TNBC and confers a poorer prognosis.94Nanda and colleagues published findings in theJournal of Clinical Oncologyfor pembrolizumab (Keytruda), a selective humanized monoclonal antibody that blocks PD-1, in the treatment of advanced TNBC. Among 27 evaluable patients, the ORR was 18.5% and the median duration of response was not yet reached (range, 15.0 to ≥47.3 weeks).95This study opened the door for immune checkpoint inhibition in TNBC.

Findings from a phase Ib study for the PD-L1 inhibitor atezolizumab (Tecentriq) in combination with nab-paclitaxel (Abraxane) were presented in an abstract by Sylvia Adams, MD, at the 2016 ASCO Annual Meeting.96The confirmed ORR with the combination in the frontline setting was 67%. In the third-line and beyond, the ORR was 29%. An ongoing phase III trial is evaluating the combination of atezolizumab with either chemotherapy or bevacizumab for previously untreated patients with metastatic TNBC (NCT01633970).97

New research into the association of CD44+/CD24- and TNBC found this relationship may play a role in the aggressive nature of TNBC. Therefore, CD44+/CD24- may be another future targeted therapy in the TNBC population.98

Finally, new research into insulin-like growth factor may provide even more options for targeted therapy. For example, insulin-like growth factor-2 mRNA binding protein 3 (IMP3) is expressed in TNBC and its function is not completely understood. Evidence now points to IMP3 being involved in breast tumor initiation and more aggressive diease.99Likewise, insulin-like growth factor binding protein-3 (IGFBP-3) is highly expressed in TNBC and may function as a tumor promotor. The mechanism involves interaction with EGFR and growth-stimulatory signaling. Furthermore, the interaction of IGFBP-3 and EGFR may be asscoiated with TNBC chemoresistance. In-depth understanding of this relationship is another potential target for TNBC treatment.100