Combined drug therapeutic strategies for the effective treatment of Triple Negative Breast Cancer

TNBC (Triple Negative Breast Cancer) is a subtype of breast cancer with an aggressive phenotype which shows high metastatic capability and poor prognosis. Owing to its intrinsic properties like heterogeneity, lack of hormonal receptors and aggressive phenotype leave chemotherapy as a mainstay for the treatment of TNBC. Various studies have demonstrated that chemotherapy alone or therapeutic drugs targeting TNBC pathways, epigenetic mechanisms and immunotherapy alone have not shown significant improvement in TNBC patients. On the other hand, a combination of therapeutic drugs or addition of chemotherapy with therapeutic drugs has shown substantial improvement in results and proven to be an effective strategy for TNBC treatment. This review sheds light on effective combinational drug strategies and current clinical trial status of various combinatorial drugs for the treatment of TNBC.


Introduction
A search of term 'triple-negative breast cancer' in PubMed hits more than 7000 publications; of which 5000 were published in the last 5 years. TNBC (triple negative breast cancer) is an intrinsically heterogeneous disease which accounts for nearly 15-20% cases among 1.7 million new breast cancer cases diagnosed annually across the world [1].
Chemotherapy remains the mainstay for the treatment of TNBC due to lack of targeted therapies. Hormone-targeted drugs like tamoxifen, aromatase inhibitors and Her2-targeted drugs like trastuzumab are ineffective towards the treatment of TNBC due to the absence of receptors. A localized breast cancer can be primarily treated by surgery, while the metastasized breast cancer treatment focuses on improving the quality of life (QOL) by increasing the outcome of pCR (pathological clinical response), PFS (progression-free survival) and prolonging the OS (overall survival) rate of the patient. The rapidly increasing evidence of research and lack of therapeutic options show the significance of investigating effective therapeutic strategies for the treatment of TNBC.

Molecular characteristics of TNBC
TNBC is a breast cancer subtype defined as lack of expression of hormonal receptors (oestrogen (ER) negative (<1%), progesterone (PR) negative (<1%) and HER2/neu) [2,3]. TNBC is a breast cancer subtype with similar characteristics of basal-like with an aggressive phenotype and high metastatic rate. TNBC exhibit properties of high histological grade [4] with distinct pathological and clinical features and associated with poor prognosis [5]. The 5-year survival rate for TNBC is 70% less than other breast cancer subtypes having 80% survival rates [6].
'BRCAness' can be defined as inherited and acquired mutations in DNA repair mechanisms in breast cancer cells [7]. BRCAness enriched phenotype in TNBC can be used as a biomarker for the exploitation of therapeutic options and clinical implications [8,9]. TNBC showed a high prevalence of BRCA mutations when compared with other subtypes are implicated in the pathogenesis of TNBC cells. The developmental pathways like Wnt/β-Catenin, Notch, Hh (Hedgehog) are associated with invasion, migration, metastatic potential and also self-renewal ability of cancer stem cells. PARP inhibitors directly interact and inhibit molecules associated with DNA repair to increase the cellular damage ultimately leading to apoptosis. Most of the potential inhibitors directly induced apoptosis in TNBC by up-regulation of Bad, Caspase 3 and down-regulation of BCL-2, BCL-XL and survivin. While several inhibitors showed therapeutic response through control of tumour growth and antiproliferative effect on TNBC cells, some inhibitors increased chemosensitivity of TNBC cells. So, synergistic targeting of chemotherapy drugs and therapeutic inhibitors may prove to be an effective way of treatment for TNBC.
of breast cancer [10][11][12]. Studies showed that 15-20% of TNBC patients carry BRCA1/2 germline mutations [10]. In recent years, gene expression signatures have been linked with TNBC to unravel distinct molecular subtypes [13]. TNBCs overlap up to 70% with basal-like breast cancer but are clinically and histopathologically distinct [14]. Based on the gene expression profiling and meta-analysis of 21 datasets of breast cancer, TNBCs are categorized into seven subclasses: Basal-like subclass (Basal-like 1 and Basal-like 2), Mesenchymal (M), MSL (mesenchymal stem-like), IM (immunomodulatory), LAR (luminal androgen receptor) and others. Identification of distinct TNBC subtypes may provide biomarkers for selection of patients in designing clinical trials and may help in the prediction of response to the treatment [13].
A study in 2006 showed that TNBC is linked to ethnic and menopausal differences which are not observed in ER+/Her2− and ER+/Her2+ breast cancer. The study also reports that prevalence of TNBC in African American women is 47%, twice when compared with white women which accounts only 22%, and this rate further increases to three-fold when considering factors like age and stage of diagnosis. African American premenopausal women diagnosed with breast cancer showed 39% of TNBC [15].

Pathways and therapeutic targets in TNBC
Cancer is a network of complex signalling pathways controlled by a cascade of events. Some pathways are highly regulated and are indispensable for the growth, survival, invasion and progression of TNBC. Various pathways are targeted and only a few pathways are found to be sensitive and effective targets for the treatment of TNBC ( Figure 1).
NF-κB is a key regulator of inflammatory response, apoptosis and angiogenesis in TNBC and shows four-fold differential expression when compared with normal breast cells [16]. Resistance in cancer cells is developed by abnormal activation of the NF-κB pathway [17]. More than 750 natural and synthetic inhibitors like small molecules, antioxidants, small RNA/DNA, peptides, viral and microbial proteins have been identified as inhibitors of the NF-κB pathway [18]. These inhibitors are used to treat various types of diseases and cancers, but there are no therapeutic drugs for TNBC which may directly interact with NF-κB pathway and thereby treat TNBC. Studies have shown that apoptosis in TNBC is also regulated by the NF-κB pathway. Genistein, a relatively nontoxic and one of the major soy isoflavones, induce apoptosis in TNBC cells by down-regulating the expression of BCL-2, BCL-xL and Cyclin B1 possibly mediated by activation of NF-κB through Notch-1 signalling pathway [19]. Plumbagin inactivates DNA-binding activity of NF-κB and BCL-2 and induces apoptosis in TNBC cells with no effect on normal breast cells [20]. Fenofibrate has antiproliferative effects and induces apoptosis by activation of the NF-κB pathway in TNBC by up-regulation of Bad and activation of Caspase-3, down-regulation of BCL-xL, survivin [21].
JAK/STAT pathway is a key regulator of cellular functions like cell differentiation, proliferation, migration, survival and apoptosis [22]. STAT3 is overexpressed in more than 50% of TNBCs associated with poor prognosis and invasive phenotype [23,24]. Metformin selectively inhibits STAT3 and restricts the growth of the tumour and induces apoptosis in TNBC cells [25]. Ruxolitinib, an inhibitor of JAK1/2 is approved for myelofibrosis treatment [26]. This drug in combination with paclitaxel, doxorubicin and cyclophosphamide is being tested in Phase II clinical trials for triple negative inflammatory breast cancers (Trial Ref.: NCT02876302). In a study, the results showed that JAK2 gene is amplified in TNBC cells treated with chemotherapy when compared with the tumours before the treatment indicating the JAK2 role in chemoresistance of TNBC. Ruxolitinib failed to inhibit tumour progression in JAK2 amplified TNBC cells. BSK805, a JAK2-specific inhibitor when combined with chemotherapy reduced the tumour growth in mice [27].
mTOR is a downstream constituent of PI3K/AKT pathway and regulates cellular functions like cell growth, survival, protein turnover and translocation. It exists in two different complexes, mTORC1 and mTORC2. mTORC1 is involved in activation of protein translation and mTORC2 is responsible for AKT phosphorylation. Clinical efficiency of numerous drugs targeting mTOR in TNBC patients is under investigation. Everolimus exhibited antitumour activity in basal-like breast cancer cells in preclinical studies [37]. BEZ235 has shown resistance to the TORC1/2 activity which further activates NOTCH1 that increases population of cancer stem cells. NOTCH activation depends upon FGFR (fibroblast growth factor receptor) 1 (FGFR1)-mitochondrial metabolism. Thus, a combined approach of TORC1/2 inhibitor and FGFR1-mitochondrial metabolism antagonists is required [38]. Some clinical trials have shown that addition of everolimus to paclitaxel in Phase II/III TNBC patients did not show any significant improvement in response ration (RR) and pCR [39][40][41].

Role of developmental pathways in TNBC
Wnt/β-catenin signalling plays a major role in embryonic development and tumorigenesis by regulating cell proliferation, differentiation and survival [42][43][44]. Previous studies reported that aberrant activation of Wnt/β-catenin signalling in TNBC results in poor prognosis [44,45]. Knockdown of β-catenin in TNBC cells significantly decreased cell migration and made TNBC cells more sensitive to chemotherapeutic drugs like cisplatin and doxorubicin [46]. Highly conserved developmental transcription factor SOX4 (sex-determining region Y-box 4) plays a key role in Wnt signalling [47]. SOX4 knockdown has shown to decrease the migration and proliferation in TNBC. Wnt/β-catenin pathway inhibitor ICRT-3 has been reported to inhibit proliferation of TNBC cells [48]. LRP5 and LRP6 of the LDLR (low-density lipoprotein receptor) family are the essential co-receptors for Wnt/β-catenin signalling [43]. LRP6 is overexpressed in TNBC and its knockdown suppresses Wnt/β-catenin signalling in vivo. Thus, LRP6 can act as a potential therapeutic target in the treatment of TNBC [49]. To activate Wnt/β-catenin signalling, Wnt binds to both FZD (Frizzled) proteins and LRP5/6. It has been demonstrated that FZD 7 was overexpressed in TNBC and its suppression inactivates Wnt/β-catenin pathway [50]. Secreted glycoproteins like WIF1 and FZD are reported to act as Wnt antagonists. Both the proteins inhibit the interaction of Wnt with FZD receptor hindering the transcription of activated genes by β-catenin/TCF/LEF transcriptional complex [43]. Recently, it has been reported that salinomycin induces degradation of Wnt co-receptor LRP6 [51,52] and also has potential to inhibit the breast cancer cell proliferation [43].
Hh (Hedgehog) signalling dysregulation confers aggressive TNBC phenotype and enhances the invasion, migration and metastatic potential of TNBC cells [53,54]. Previous clinical studies highlighted the key role of Hh signalling in cancer stem cell reprogramming and EMT (epithelial-to-mesenchymal) in TNBC [55,56]. The Hh pathway is associated with embryonic patterning and mediates stem cell renewal by activating the expression of BMI-1, a potent regulator of self-renewal in cancer stem cells [57]. It involves three ligands -IHH (Indian Hedhehog), SHH (Sonic Hedgehog) and DHH (Desert Hedgehog); Transmembrane receptor, PTCH ( Patched) and co-receptor, SMO (Smoothened) [58]. There are three glioma-associated oncogenes (GLI) transcription factors, GLI1, GLI2 and GLI3. However, GLI1 and GLI2 are the most studied ones and responsible for cell proliferation and survival [59]. SMO is the most pharmacologically targeted pathway in TNBC. Various SMO inhibitors were clinically tested and few gave the positive response as Hh antagonists (NCT01071564, NCT02027376 and NCT01757327) [60]. However, in preclinical studies, resistance to these Hh antagonists was observed in TNBC. Thus, a rationale for the GLI-targeted approach was suggested [61]. So far, numerous direct and indirect GLI inhibitors have been clinically tried like GANT61, GANT58 and Glabrescione B (GLaB). These drugs interfere with GLI DNA binding by inhibiting the output of transcription in Hh signalling pathway [62].
The Notch signalling pathway is a much conserved signalling pathway that is mediated by four receptors (NOTCH 1-4) and five ligands ( -like 1,3,4 and JAGGED-1,2) [63][64][65][66]. Cell-cell contact is a key factor to activate the NOTCH signalling pathway [67]. The signalling cascade is activated by the release of Notch receptor intracellular domain (NICD) with a series of proteolytic cleavage facilitated by γ-secretase [68]. Irregular activation of Notch signalling cascade could initiate malignancies and promote angiogenesis [69]. Previous studies reported that GSI (γ-secretase inhibitors) play a significant role in blocking the Notch signalling pathway [70]. Therefore, numerous preclinical studies have been done on GSI-directed therapy. Researchers confirmed that NOTCH-1 exert a strong influence on tumour proliferation and metastasis. The increased expression of NOTCH-1 has been observed in TNBC that lead to malignancies and poor prognosis [71]. However, it has been recently discovered that NOTCH-4 also plays a pivotal role in the initiation of TNBC and induction of proliferation and tumorigenesis [72]. Targeting NOTCH signalling cascade with GSIs and other drugs should be meticulously explored to increase the survival rate of TNBC patients.
EGFR, also known as HER1 is overexpressed in basal-like cells [80]. EGFR-TKI (tyrosine kinase inhibitor) erlotinib, showed a change in mesenchymal phenotype to epithelial phenotype by up-regulating E-cadherin and down-regulating Vimentin in TNBC cells [82]. Several other EGFR inhibiting agents like panitumumab, cetuximab, gefitinib have shown initial success but failed to produce significant results in clinical studies [83]. Sunitinib is a small-molecule kinase inhibitor, which inhibits both PDGF family and VEGF have shown to reduce tumour volume in xenograft models of TNBC [84]. Bevacizumab reduced progression of metastatic TNBC in 35% of patients in a meta-analysis of Phase III clinical trials [85].

Epigenetic therapies
It is widely believed that aberrant epigenetic changes in histone deacetylation and DNA hypermethylation may lead to silencing of tumour suppressor genes and drive tumorigenesis in cancer cells [86]. A detailed study of DNA methylation signatures using TCGA (The Cancer Genome Atlas) data helped in the separation of TNBC cells from non-TNBC cells. These data helped in the prognosis of patients by categorizing into poor, medium and good outcomes [87]. The first study showed methylation of a BRCA1 promoter in TNBC and few other studies investigated the role of BRCA1 methylation in TNBC. They also found that BRCA1 methylation increases the sensitivity of TNBC cells towards PARP inhibitors [88]. Another study has found that decreased expression of pRb and increased expression of p76 is associated with BRCA1 [89].
DNA hypermethylation decreases expression of tumour suppressor genes. A study revealed that inhibition of STAT3-DNMT1 (DNA methyltransferase 1) at K685 residue by novel inhibitor SH-I-14 has shown to demethylate the promoter regions of tumour suppressor genes and re-expressed PDLIM4 and VHL genes [90]. A study performed on whole-genome methyl CpG binding domain based capture sequencing (MBDCap-Seq) on TNBC tumours and found 36 differentially methylated regions (DMRs) which showed increased hypermethylation specifically in TNBC cells when compared with non-TNBC samples [91]. BRD4 is a BET (bromodomain and extra terminal) protein family member, regulates mitosis and cell cycle progression [92,93]. BRD4 inhibition has shown to suppress important oncogenic drivers [94]. BETi (BET inhibitor) showed direct inhibition of mitotic regulating proteins AURKA/B in TNBC cells and thereby suppressing tumour growth [95]. BETi JQ1 targeted hypoxic inducing genes and angiogenesis dually in TNBC cells [96]. ID4 (inhibitor of differentiation) protein is highly expressed in TNBC cells and down-regulates BRCA1 pathways [97] and exhibits anchorage-independent growth of breast cancer cells [98]. ID4 promoter hypermethylation is known to increase lymph node metastasis [99]. A study also revealed that ID4 and BRCA1 expression are inversely related and unmethylation of ID4 is associated with BRCAness of breast cancer cells [100]. PKD1 (protein kinase D1) encoded by PRKD1 gene is abnormally methylated and silenced in invasive breast cancer cells. DNMT inhibitor decitabine reverses PRKD1 promoter methylation and restores PKD1 expression and suppresses lung metastasis in animal models [101].
Cancer cells disseminate to distant sites by transforming EMT phenotype, which is characterized by loss of E-cadherin expression. TICs which are found in tumour tissues exhibit self-renewing stem cell properties and they also have the ability to grow into a tumour in mice when inoculated at very low numbers [107]. Studies have shown that cancer cells activating EMT acquire TIC's properties expressing CD44 high /CD24 low markers [108][109][110].

Immunotherapies
In 2013, cancer immunotherapy was named as 'Breakthrough of the year' by science magazine [111]. TILs (tumour-infiltrating lymphocytes) are long known to be associated with breast cancer prognosis. The prognostic and predictive values vary between subtypes of breast cancer. Studies showed that TILs highly prevailed in TNBC and were less abundant in other types of breast cancer [112]. TILs are prognostic markers for high OS, increased metastasis-free survival and decreased distant recurrence [113,114]. Stromal TILs are correlated with immunological markers like indoleamine 2,3-dioxygenase (IDO1), CD8α, CCL5 (chemokine (C-C motif) ligand 5) and PD-L1 (programmed cell death ligand-1) to significantly increase pCR rates in chemotherapy [115]. Trop-2 (trophoblast cell-surface antigen) is expressed on multiple solid cancers and found to be a novel target for antibody-mediated drug conjugate (ADC) therapy [116]. IMMU-132 is an ADC, delivers topoisomerase-I inhibitor (SN-38) in its most active (non-glucuronidated) form targeting Trop-2 in TNBC [117].
Immune checkpoints are the molecules of inhibitory pathways in the immune system which play a major role in preventing autoimmunity [118]. Activated CD8 + T cells express inhibitory cytotoxic receptor T-lymphocyte associated antigen 4 (CTLA-4), counteracts the activity of co-stimulatory receptor CD28 and attenuates immune response [119]. Ipilimumab is a monoclonal antibody that targets CTLA-4 to activate T cells and thereby increasing proliferation of T cells and potentiates antitumour immune response [120]. Another 'immune checkpoint' blockade is PD-1 (programmed cell death 1), a T-cell transmembrane receptor expressed on CD8 + T cells. Up-regulation of PD-1 ligands (PD-L1 or PD-L2) blocks T-cell immune response in the tumour microenvironment [121]. Pembrolizumab, a potent inhibitor of PD-1 showed antitumour activity and overall response rate (ORR) of 18.5% in TNBC patients [122] (Figure 3). Other antibodies to take the 'brakes off' T cells to increase the antitumour immune response are under investigation and the current immunotherapy clinical trials are listed in Table 1 (Figure 3).

Combined drug therapy strategies
Although the single-agent therapy has shown positive results in cell lines and preclinical models but failed to get promising results in clinical trials to counter aggressive TNBC, owing to its heterogeneity and acquired drug resistance. Combined drug therapy (CDT) is rapidly gaining popularity and proving to be effective in current clinical trials towards improving pCR, PFS and OS in various cancers. At present, almost 80% of the clinical trials are using  combinatorial drugs to investigate new therapeutic strategies for TNBC treatment. CDT strategies in current clinical trials data are provided (Table 1).
Recently, CDT strategy has been widely used for immunotherapy checkpoint inhibitors to target TNBC effectively. Tremelimumab (CTLA-4i) in combination with duralumin (PD-L1i) is under investigation in Phase II clinical trials (NCT02527434). The effective way of planning combinational strategy is through prediction of effective targets connected to signalling networks that drive cancer progression. Systems biology provided attractive tools to strategize network-based therapies for cancer. Using these tools, a study group identified five most effective and connected targets (VIM, YWHAB, TK1, CSNK2B and HSP90AB1) in TNBC cells. Initially, the targets were validated using cell-based assays. Based on initial results, using animal models they knocked out five targets in vivo and successfully inhibited colony formation, proliferation, migration, anchorage independence and invasion [123].
A study showed that combination of mTOR inhibitor rapamycin and doxorubicin-loaded cyclic octapeptide liposomes inhibited the expression of HIF-1α in TNBC cells [124]. Combined inhibition of PI3K/AKT/mTOR with chemotherapy showed substantial improvement in PFS of TNBC patients [125]. Other study showed that combined inhibition of CDK4/6 and PI3Kα has greatly increased tumour infiltrating T-cell activation in TNBC cells [126]. TNBC cells which expressed PTEN responded to PARP and HDACis. Combined inhibition of olaparib and SAHA in TNBC cells showed increased DNA damage, decreased proliferation, increased autophagy and apoptosis [127]. HDACi mocetinostat combinedly treated with BETi JQ1 showed synergistic suppression of cell cycle progression genes and induced apoptosis in TNBC cells [128].
Few randomized clinical trials showed that addition of HDACi to DNMTi did not improve the outcomes in the patients [129][130][131]. There is no conclusive evidence that epigenetic inhibitors function by epigenetic mechanisms. These results clearly indicate to reinvestigate how epigenetic drugs work and their mechanism of action [132].

Future directions
The recent study shows that knockdown of PRL-3 (phosphatase of regenerating liver 3) leads cancer cells to senescence. The experimental drug AMPI-109 inactivates PRL-3, making senescent cancer cells sensitive for immunotherapy treatment [133].
There is an increasing evidence indicating the role of PTEN in acquiring chemoresistance in MDR (multidrug resistant) breast cancer cells. Inhibition of miR-19 down-regulates multidrug resistance genes (MDR-1, MRP-1 and BCRP) and restores PTEN expression in MDR breast cancer cells, sensitizing cells to chemotherapeutic agents [134]. Up-regulation of PTEN activity increases the effectiveness of chemotherapy and in combination with ID4 (DNA binding protein inhibitor) can be studied for the effective treatment of TNBC. One of the studies suggested that combination therapy of lapatinib (NF-κB inhibitor) with a proteasome inhibitor may prove to be an effective treatment for TNBC [135].
A study published in 2011, shows that anti-oestrogens or aromatase inhibitors increase the population of ER-negative cells in luminal breast cancer cells thereby increasing resistance to the treatment [136]. This study led to the findings that inhibiting Notch-1 in luminal breast cancers maintains the ER positive state for the effective targeting of ER-based therapies. It is also found that inhibiting Notch-1 can transform ER−/PR−/CK5+ cells to ER+ cells [137]. Therefore, Notch-1 inhibitors like GSI in combination with endocrine therapies can be used as CDT strategy for TNBC treatment.
Several other chemotherapy drugs, epigenetic inhibitors, immunotherapies and combinational therapies showing positive results in vitro should be immediately carried over to clinical trials to determine the effectiveness of the drugs in vivo. As there is an urgent need to find out therapeutic targets for TNBC, we need to explore the new biomarkers and signalling pathways which help in early diagnosis of cancer and finding new therapeutic targets for effective treatment of TNBC.

Conclusion
Despite the fact that combined therapeutic strategies are proven to be effective in various cancers including TNBC, there are few exemptions where some of the valid hypotheses and in vitro results are shown to be ineffective when translated into clinical trials. TNBC is a heterogeneous cancer with varying physiological and pathological characteristics and associated with the aggressive phenotype. So, despite the emergence of various therapeutic strategies for the treatment of TNBC, the effective treatment can be provided by selecting suitable combinational therapy by considering patient-specific molecular characteristics, biomarkers, clinical and pathological features through proper diagnosis.