Cancer Treatment Reviews 68 (2018) 102–110

Contents lists available at ScienceDirect

Cancer Treatment Reviews

Anti-Tumour Treatment

Androgen receptor in triple negative breast cancer: A potential target for the T
targetless subtype
L. Gerratanaa, D. Basilea, G. Buonob, S. De Placidob, M. Giulianob, S. Minichilloc, A. Coinud,
⁎
F. Martoranae, I. De Santob, L. Del Mastrof, M. De Laurentiisg, F. Puglisia,h, G. Arpinob,
a
Department of Medicine (DAME), University of Udine, 33100 Udine, Italy
b
Clinical Medicine and Surgery Department, University of Naples Federico II, 80131 Naples, Italy
c
Department of Medical Oncology, Bellaria Hospital, Azienda USL Bologna, 40139 Bologna, Italy
d
Oncology Unit, Giovanni Paolo II Hospital, 07026 Olbia, Italy
e
Center of Experimental Oncology and Haematology, University of Catania, 95131 Catania, Italy
f
Department of Medical Oncology, IRCCS Azienda Ospedaliera Universitaria San Martino-IST, Genoa, Italy
g
Breast Unit, ‘Fondazione G. Pascale’ Istituto Nazionale Tumori, 80131 Naples, Italy
h
Department of Clinical Oncology, CRO Aviano National Cancer Institute, 33081 Aviano, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: Triple negative breast cancer (TNBC) represents the 15–20% of all breast cancers (BC) and is characterized by a
Luminal androgen receptor positive very aggressive behavior. Recent data suggest that TNBC is not a single disease, but it is rather an umbrella for
Triple negative breast cancer different ontology-profiles such as basal like 1 and 2, mesenchymal, and the luminal androgen receptor (LAR).
Epithelial-to-Mesenchymal Transition The LAR subtype is characterized by the expression of the Androgen Receptor (AR) and its downstream effects.
Antiandrogen therapy
Notwithstanding the role of the AR in several signaling pathways, its impact on a biological and clinical
standpoint is still controversial. The LAR subtype has been associated with better prognosis, less chemotherapy
responsiveness and lower pathologic complete response after neoadjuvant treatment. Clinical evidence suggests
a role for anti-androgen therapies such as bicalutamide, enzalutamide and abiraterone, offering an interesting
chemo-free alternative for chemo-unresponsive patients, and therefore potentially shifting current treatment
strategies.

Introduction The biological role of AR

Androgen receptor (AR) is a steroid hormonal receptor that belongs The AR gene is located on chromosome Xq11-12 and encodes for a
to the nuclear receptors family together with estrogen (ER), gluco- 110 kDa cytoplasmic polypeptide comprising four distinct functional
corticoid, progesterone (PR) and mineralcorticoid receptor. It links a regions: a N-terminal region involved in transcriptional activation, a
transcription factor that controls specific genes involved in different, regulatory domain at the amino terminal (AF-1 site), a DNA binding
sometimes opposite, cellular processes: it can stimulate or suppress domain composed of two zinc fingers, a hinge region with a nuclear
both cell proliferation and apoptosis, depending on the concurrent localization signal and a C-terminal ligand-binding domain (AF-2 site)
signaling pathways activated [1–6]. Androgen receptor is expressed in (Fig. 1) [12]. The role of AR in breast cancer carcinogenesis is complex.
about 70–90% of breast cancers and its expression varies from 10% to Without its ligand, AR is found in the cytoplasm kept inactive by a
50% in triple negative breast cancer (TNBC) [7–11]. heterocomplex with heat-shock proteins and a chaperone complex
AR seems to play a major role in TNBC carcinogenesis. However, its (HSP-70, HSP-90). Circulating androgens bind to the C-terminal ligand-
impact on patient prognosis and its predictive role in patients with binding domain leading to a conformational change which allows AR
TNBC are still controversial. The present review focuses on AR biology dimerization. After ligand binding, receptor-hormone complex trans-
and covers the current clinical evidences on both predictive and prog- locates into the nucleus where it promotes a co-activator-mediated
nostic implications of AR in TNBC. transcription of target genes (transcriptional/genomic modality of AR
activation), and an inactivation of AR transcription through a negative
feed-back [1,13].

⁎
Corresponding author at: Clinical Medicine and Surgery Department, Division of Medical Oncology, University of Naples Federico II, Pansini 5, 80131 Naples, Italy.
E-mail address: [email protected] (G. Arpino).

https://doi.org/10.1016/j.ctrv.2018.06.005
Received 20 April 2018; Received in revised form 8 June 2018; Accepted 9 June 2018
0305-7372/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

Fig. 1. Androgen receptor gene. Androgen receptor gene is mapped to the long arm of X chromosome (Xq11-12). The androgen receptor protein is encoded by 8
exons (1–8) separated by introns up to 26 Kb in size 8. The protein is composed by distinct functional regions. The exon 1 encoded the N-terminal region (NTD), exons
2 and 3 encoded a DNA binding domain (DBD). The 5′ region of exon 4 encoded for a hinge region, while the 3′ region of exon 4–8 encoded a ligand binding domain
(LBD).

However, AR can also be activated through a non-transcriptional/ positive TNBC presents an up-regulation of the EGFR ligand amphir-
non-genomic mechanism that does not need DNA or RNA interaction egulin, involved in tumor proliferation mediated by the EGFR signaling
and that modulates AR activity by signal transduction in an ERK-de- pathway. Enzalutamide seems to decrease this effect in cell lines ex-
pendent or -independent manner. ERK-mediated AR signaling involves pressing AR [16].
cytoplasmic AR which interacts with phosphoinositide 3-kinase (PI3K),
Src proteins and Ras GTPase. Non ERK-mediated AR signaling may AR and epithelial-to-mesenchymal transition
involve the mammalian target of rapamycin (mTOR) phosphorylation,
the forkhead box protein O1 (FOXO1) inactivation, and the protein Recently, the zinc-finger enhancer binding protein (ZEB1) tran-
kinase A (PKA) activation and results in increased cell proliferation scription factor, has been associated with AR positive TNBC subtypes.
(Fig. 2) [6,14]. ZEB1 activation has been associated to an Epithelial to Mesenchymal
Transition (EMT) phenotype and predicts for poor patient survival due
AR and cell signal transduction pathways to a higher metastatic potential [26]. In TNBC, the ZEB1 - AR cross-talk
is probably due to a direct binding of ZEB1 to the E-box sequence on the
AR enriched TNBC cell lines frequently carry PI3KCA mutations AR promoter [27]. Interestingly, a morphological switch from a me-
which make them very sensitive to PI3K/mTOR inhibition. The cross- senchymal to epithelial phenotype was observed after ZEB1 knockdown
talk between these two pathways have been suggested to promote in TNBC cell lines. ZEB1 suppression in TNBC cell lines was also asso-
cancer cell growth [15,16]. Additionally, AR phosphorylation via ciated with a decrease of AR mRNA and AR downstream targets and a
phosphorylated AKT abolishes AR-induced apoptosis resulting in in- sensitization to bicalutamide. Consistently, the treatment with bicalu-
creased cell survival [17,18]. AR expression may also up-regulate tamide reduced the expression of ZEB1 in treated cells [26,28–30].
PTEN, due to more frequent mutations of AR in the kinase domain AR plays a critical role in cancer metastasis development also by
(exon 20), than in the catalytic domain (exon 9) which leads to an in- promoting migration and invasion, through the extracellular matrix
crease in PTEN levels [19]. It has been observed that AR expression in degradation. Preclinical models have demonstrated that AR induces the
ER-negative MDA-MB-453 cells induces PTEN which represses PI3KCA expression of metalloproteinase (MMP), in particular, MMP2 and
activation and reduces AR activity [20]. At the same time, PTEN acts MMP9 [31]. The decreased anchorage-independent growth and inva-
with the protein killin (KLLN) and induces p53 and p73, resulting in sion, and the increased apoptosis reported in clinical trials with en-
increased apoptosis. These preclinical data explain the anti-pro- zalutamide in AR-positive TNBC subtypes, including mesenchymal
liferative effect of AR that could cause the favorable prognosis in terms stem-like, mesenchymal-like and basal-like, further support this hy-
of DFS and OS seen among AR-positive TNBC patients [21–23]. pothesis [32].
GATA-3, a transcription factor involved in mammary gland devel-
opment and its luminal cell differentiation also seems to interact with AR and cell cycle regulators
AR. It has been demonstrated that GATA-3 may limit the response to
chemotherapy by activating the downstream targets of ER signaling, Noteworthy, AR interacts also with cell cycle regulators and BC
even in ER negative breast cancers, probably under the influence of the susceptibility genes such as Poly-ADP-ribose-polymerase 1 (PARP1) and
AR. Indeed, Naderi at al. found that the activation of AR in ER negative BRCA1. BRCA1 plays a key role in double-strand breaks repair when a
cells induced the expression of FOX1A, which is a downstream target of DNA damage occurs, while PARP1 is important in the base excision
GATA-3 itself [24]. It was also demonstrated that GATA-3 expression repair process for DNA single-strand breaks repair. It has been recently
was strongly correlated with AR-positivity especially in apocrine TNBCs reported that BRCA1 and PARP1 act as coactivators of AR and promote
[24,25]. AR-targeted gene transcription. Preclinical evidences suggest that
Moreover, gene microarray and ChIP-seq analysis showed that AR- PARP1 inhibitors in AR positive TNBC reduce cell migration and

103
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

Fig. 2. Androgen receptor activation. Androgen receptor resides in the cytoplasm in an inactive form through a heterocomplex with heat-shock proteins and a
chaperone complex (HSP-70, HSP-90). Hormone binding induces a conformational change which allows AR activation. Transcriptional/genomic modality of AR
activation: receptor-hormone complex translocates into the nucleus and interacts with co-activators, co-repressor and transcription modulators. As a result, it
promotes the transcription of target genes. Non-transcriptional/non-genomic modality of AR activation: ERK-mediated AR signaling involves phosphoinositide 3-
kinase (PI3K), Src proteins and Ras GTPase. Non ERK-mediated AR signaling involves the mammalian target of rapamycin (mTOR) phosphorylation, the forkhead box
protein O1 (FOXO1) inactivation, and the protein kinase A (PKA) activation.

invasion, thereby preventing DNA repair as well as reducing AR ac- could be used in combination with chemotherapy in the treatment of
tivity. This implies an important role of AR in BRCA1-dependent tumor this subgroup of TNBC. Even though nowadays less investigated, the
suppression. Furthermore, BRCA1-mutated BC show a lower expression crosstalk between AR and the immune system is not less important.
of AR and preclinical studies have highlighted an increased apoptosis Indeed, treatment with AR inhibitors may increase the recruitment of
when PARP1 inhibitors were combined with AR inhibitors. However, cytotoxic T cell, which could enhance susceptibility to immunotherapy
further studies are required to clarify the crosstalk and the role of these [15,31].
pathways in TNBC cells [22,33,34].
Genomic profiling in TNBC and AR expression
AR, angiogenesis and immune system
Molecular characteristics of the main gene ontology-based profiles
Recently, it was observed that AR pathway could be influenced by
the hypoxia-inducible factor 1 alfa (HIF-1a) and the vascular en- Genomic profiling strategies have been explored to shed light on the
dothelial growth factor (VEGF). Preclinical evidence show that the deep heterogeneity which characterize TNBC. The IHC-defined TNBC
treatment with dutasteride, a dual blocker of both the type-1 and type-2 profile is actually composed by a wide range of molecular profiles that
isoform of the steroid-5 alfa-reductase (SRD5A1), was associated with a show profoundly different gene ontologies.
reduction in protein expression of VEGF and HIF-1a, resulting in an In 2011, Lehmann et al. reported six molecular subtypes of TNBC
increased chemosensitivity, and dose- dependent decrease in cell via- each characterized by potentially new therapeutic targets: 2 basal like
bility of about 40% [35]. If confirmed in clinical prospective trials it classes (BL1 and BL2), an immunomodulatory (IM), a mesenchymal

104
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

Fig. 3. A and B. Lehmann classification 2011 and 2016. (A) Triple negative breast cancer (TNBC) was classified into main six subgroups: two basal like classes
(BL1 and BL2), an immunomodulatory (IM), a mesenchymal (M), a mesenchymal stem cell (MSL) and the luminal androgen receptor (LAR) class. (B) Triple negative
breast cancer (TNBC) was classified into main six subgroups: two basal like classes (BL1 and BL2), an immunomodulatory (IM), a mesenchymal (M), a mesenchymal
stem cell (MSL) and the luminal androgen receptor (LAR) class.

(M), a mesenchymal stem cell (MSL) and the luminal androgen receptor respond to antimitotic agents such as platinum salts and PARP in-
(LAR) class, characterized by AR expression (Fig. 3A) [36,37]. hibitors. The BL2 subtype is characterized by the expression of EGFR,
The recently refined version of TNBC molecular classification de- TP63, MET and activation of glycolysis and gluconeogenesis pathways.
fined four main subtypes, BL1 and BL2, M, and LAR, with unique The M group includes more than half of the metaplastic carcinomas and
ontologies and differential response to therapy [38] (Fig. 3B). Biolo- is characterized by EMT and frequent PIK3CA mutations or PTEN de-
gical pathways involving cell cycle control, DNA damage response and ficiency. These tumors respond to tyrosine kinase (TKI) and mTOR in-
high cell proliferation characterize the BL1 profile. These tumors hibitors. Interestingly, eribulin mesylate could be particularly

105
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

beneficial in this subtype since it is capable to suppress EMT. Notably, with a 80% sensitivity and 65% specificity [47].
although AR mainly characterizes the LAR profile, it seems to play an
important role also in non-LAR subtypes, such as M, BL1 and BL2 Prognostic implications of AR in TNBC
[15,32,39]. The LAR subtype is closely linked to histological apocrine
type tumors and was so termed because it can be defined as TNBC by The prognostic impact of AR among TNBC patients is controversial.
IHC but histologically and genetically is similar to ER-positive BC. Gene Several studies have highlighted the favorable prognosis of LAR TNBC
ontologies defining the LAR subtype are enriched in hormonally regu- given the lower Ki-67 and mitotic index and the lower tumor grade and
lated pathways including steroid synthesis and androgen/estrogen clinical stage at diagnoses [20,48,49]. In detail, a recent meta-analysis
metabolism [37]. Because of this, phase II and III clinical trials reported by Wang et al. analyzed data from 2826 women with TNBC from 13
encouraging results in term of clinical benefit after treatment with both trials conducted between 2007 and 2015 and showed that AR, ex-
bicalutamide and enzalutamide among patients with AR-positive TNBC pressed in the 24.4% of the overall TNBC study cohort, was sig-
[40,41]. PI3K inhibitors in addition to an AR antagonist seems to be nificantly associated with post-menopausal status (26.9% of patients
more effective in treating AR-positive TNBC because PIK3CA mutations with AR expressing tumors were postmenopausal and 13.4% were pre-
are frequently activated in these tumors. Further studies testing the menopausal), low tumor grade (40.8% of patients with AR expressing
clinical effect of concurrent treatment of PI3K inhibitors and AR tumors were G1-2 and 23% were G3) and with a high risk of nodal
blockades are ongoing [5,36]. involvement (28.8% of patients with AR expressing tumors were node
LAR TNBC have lower proliferation rates compared to the other positive and the 22.6% were node negative) [50]. Consistently, Maeda
TNBC subtypes, resulting in a partial chemoresistance; consistently, a et al. reported an association between AR and both low clinical stage
retrospective analysis of 130 patients treated with neoadjuvant che- and nuclear grade, among 23 patients with TNBC and Gasparini et al
motherapy has shown a lower pCR rate in LAR (10%) in respect to BL1 showed that high grade TNBC presented lower AR expression
tumors (52%) [42]. (p < 0.01) [48,49]. In a cohort of 203 asian patients, AR positive
Among the different gene ontology-based classification systems, the TNBC showed a lower Ki67 proliferation index [51,52].
LAR definition seems one of the most solid. A newly published classi- Sutton and al. suggested a higher incidence of distant metastases in
fication system based on both RNA and DNA profiling, identified 4 AR-negative tumors [51,52]. A single study exploring the prevalence of
molecularly defined TNBC subtypes: LAR, Mesenchymal (MES), Basal- AR expression in 88 patients with inflammatory breast cancer (IBC)
Like Immune-Suppressed (BLIS), and Basal-Like Immune-Activated showed that only 5 of the 17 TNBC were AR-positive. Interestingly,
(BLIA), characterized by different prognosis and potential therapeutic women with AR-negative TNBC had inferior 5-year survival rates
targets [43]. Interestingly, DNA analysis highlighted profile-specific compared with the AR-positive TNBC and the other histologic sub-
gene amplifications and targetable molecular expression. LAR was groups (p < 0.03) [53].
found to be particularly characterized by both AR and MUC1 markers, Three recent meta-analyses have shown longer disease-free survival
remarking the strategic importance of anti-AR therapy but also the (DFS) in AR-positive versus AR-negative breast cancer patients. In de-
potential role of MUC1 vaccines as an effective treatment for this sub- tail, Qu et al reviewed 12 studies including 5270 patients with breast
type. Interestingly, in contrast with other profiles, LAR subtype iden- cancer. The overall rate of AR expression in these studies was 65.2%.
tified by this new classification system share the same genetic and The combined hazard ratio (HR) of DFS for all 12 eligible studies was
biologic characteristics of those identified by Lehmann/Pietenpol el al 0.52 (95% CI 0.43–0.64), suggesting that AR expression in breast tu-
[37,43]. mors was an indicator of low risk of recurrence. The HR of overall
survival (OS) for all studies was 0.66, but it was not statistically sig-
New approaches and future applications nificant [54]. Similarly, Kim et al. selected 16 articles published be-
tween 1992 and 2013. With DFS data available for 521 TNBC patients,
New, alternative, approaches have been explored to enable trans- AR-positive tumors had a significant lower risk of relapse compared to
ferability of molecular profiling to the clinic. the other TNBC subgroups (OR for DFS 0.44, p 0.002) [55]. Finally, an
A recently published transcriptome analysis identified 4 distinct additional meta-analysis by Wang et al. confirmed that women with
TNBC clusters according to RNA expression. Among these, the LAR AR-positive TNBC display a 20% lower risk of recurrence compared
cluster was characterized by gene ontology enriched in hormone-de- with AR-negative TNBC (HR 0.8, p < 0.05). While DFS findings in all
pendent pathways. Spearman’s correlation analysis highlighted a sig- these analyses were concordant, both Qu and Wang found no associa-
nificant association between the LAR subtype defined according to the tion between AR status and overall survival, whereas the study by Kim
Lehmann/Pietenpol classification and the LAR cluster identified by the et al. showed an overall survival benefit for AR-positive TNBC patients
transcriptome-driven profiles. In detail, an upregulated estrogen de- (OR 0.26, p 0.001) [50,54,55]. Results of a recent prospective study by
pendent signaling pathway was highlighted in the LAR cluster defined Asano et al corroborate these latest findings as 59 of 190 TNBC patients
by RNA analyses, confirming the pivotal role of the anti-androgen (29.5%) who displayed AR-positive status had a significantly favorable
therapy but also suggesting a potential impact of traditional anti-es- Cancer Specific Survival (p = 0.0034) [56]. In summary, despite initial
trogen therapies [44]. studies suggested a potential negative prognostic role for AR in TNBC
Recently, some efforts have been made to integrate gene-expression [57–60], a growing body of evidences indicates that AR expression is
profiles with MicroRNAs (miRNAs) expression levels [45]. miRNAs are associated with a favorable prognosis. Data concerning OS are still
short non-coding RNAs that regulate the function of target genes at the weak and will need further prospective studies.
post-transcriptional phase and are involved in cancer progression and
metastasis. In particular, miR-363 seem to be a promising target, but Predictive value of AR in TNBC
results are still limited [46].
Parker et al. developed a treatment-focused approach based on Despite its clinical aggressive behavior, TNBC is commonly con-
Next-Gen RNA-sequencing analysis using a “from bedside back to sidered more sensitive to chemotherapy compared to others histological
bench” strategy. An initial training set was built by analyzing 80 sam- subtypes given the higher expression of proliferation-related genes in
ples of patients treated with enzalutamide and 42 from untreated pa- this subgroup of BC [59]. However, the molecular tumor features as-
tients and was used as a basis to develop a gene expression model of sociated with LAR TNBC may result in a less responsive phenotype to
biological subtype according to treatment response. This new approach chemotherapy.
was capable to predict a 16-weeks clinical benefit from enzalutamide Several studies, particularly in the neoadjuvant setting [16,42],
and therefore better identify androgen-sensitive tumors among TNBC have investigated whether AR positivity is a chemo-resistance marker

106
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

in TNBC. Using samples from patients enrolled in the GeparTrio phase Table 1
III neo-adjuvant trial, Loibl et al evaluated AR expression and its impact Antiandrogen therapy in LAR-TNBC: clinical trials.
on outcome [60]. Overall, 637 core biopsies from primary breast cancer N Treatment CBR CI
patients treated with neoadjuvant docetaxel/doxorubicin/cyclopho-
sphamide (TAC) chemotherapy were analyzed and AR was detected in Gucalp et al. 452 Bicalutamide 150 mg continuous daily 19% 95%
[40] schedule
53.2% of tumors. In AR-positive tumors, pathological complete re-
Traina et al. 118 Enzalutamide 160 mg continuous daily 35% 95%
sponse (pCR) rate was 12.8% compared to 25.4% in AR-negative tu- [41] schedule
mors (P < 0.0001). Among the TNBC subgroup AR expression pre- Bonnefoi 30 Abiraterone 1000 mg + prednisone 20% 95%
dicted a better DFS (AR-positive 85.7% vs. AR-negative 65.5% log-rank et al. 5 mg twice/day continuous daily
P = 0.0544) and OS (95.2% vs. 76.2%; log-rank P = 0.0355). Within [62] schedule
Gucalp et al. 33 Bicalutamide 100 mg continuous daily Ongoing
the non-pCR subgroup, AR positivity selected a group with a significant
[63] schedule + Palbociclib 100 mg daily analysis
better DFS (P = 0.045) and OS (0.021) but not within the pCR group 3 weeks on 1 week off
[60]. Masuda et al. retrospectively classified 146 TNBC tumors ac-
cording to their gene expression profile and found that LAR tumors had
a lower pCR rate as none of the 20 LAR patients achieved pCR after neo- CR plus PR plus SD at 16 weeks (Table 1). CBR at 24 weeks, response
adjuvant chemotherapy [42]. Similarly, pCR was significantly less fre- rate (RR) and safety were evaluated as well. As in the Gucalp study,
quent in AR-positive compared with AR-negative TNBC in a prospective evaluable patients were defined as having AR IHC ≥ 10% and a re-
trial conducted on 117 Japanese women [56]. Recently, a study in- sponse assessment. Outcomes were evaluated according to an an-
vestigating the efficacy of neoadjuvant cisplatin plus paclitaxel with or drogen-driven gene signature (Dx). Out of the 404 samples tested for
without everolimus in 145 TNBC patients, demonstrated that low levels AR IHC, 79% had AR > 0% and 55% had at least AR = 10%, sug-
of AR expression (< 10%) were more likely to be associated with pCR gesting a higher AR prevalence than previously reported. Among the 75
than higher AR levels. These findings were consistent in both control evaluable patients, this trial showed a CBR at 16 and 24 weeks of 35%
and experimental arm. Interestingly, authors observed no significant and 29%, respectively. Median PFS was higher in Dx-positive patients
modifications in AR levels in serial samples, obtained before, during (32 vs 9 weeks), two CR and 5 PR were reported. The most frequent
and after the treatment, suggesting that AR expression is not affected by adverse events were fatigue, decreased appetite and nausea.
chemotherapy. Taken together, these results seem to suggest a negative Another phase II trial tested the potential role of abiraterone, given
predictive role of AR in the neo-adjuvant chemotherapy setting, being the mechanism of action of this drug as selective inhibitor of CYP17
AR positivity correlated with a lower pCR rate. Nonetheless, women [61] (Table 1). Abiraterone was administered orally on a continuous
displaying an AR positive TNBC had a better DFS and OS, indicating daily schedule at the dose of 1000 mg in heavily pretreated women with
that in LAR pCR may not be an appropriate surrogate marker of sur- AR-positive TNBC, adding five milligrams of prednisone twice a day to
vival. Testing AR status at diagnosis could lead to a better selection of avoid adverse effects correlated to increased mineralocorticoid levels.
patients who are likely to benefit from a more aggressive neo-adjuvant Primary endpoint of the study was CBR at 6 months. Objective response
treatment. New therapeutic strategies, such are the combination of rate, duration of response, PFS and safety were secondary endpoints.
chemotherapy with an anti-androgen (i.e. NCT02689427), could lead to Starting from 146 patients screened, 30 patients were considered eli-
a de-esclation of chemotherapy in this subtype. gible. Overall, 6 patients showed a clinical benefit at 6 months (20%),
Notably, AR expression seems to reduce TNBC radiosensitivity too, with CR in 1 patients and SD more than 6 months in 5 patients and an
although preliminary evidence suggests that bicalutamide might restore objective response rate of 6.7%. Median PFS was 2.8 months. Fatigue,
the effect of therapeutically directed ionizing radiation in these patients hypertension, hypokalemia and nausea were the most common adverse
[61]. More studies are required to further confirm these findings. events, predominantly grade 1 and 2. Two patients had a treatment-
In addition, other markers have been explored in order to further related serious adverse event, 1 with grade 3 hypokalemia and the other
refine AR’s predictive potential, such as GATA-3 in the neoadjuvant with grade 3 adrenal insufficiency [62].
setting or CK5/6 and p53 [46,62,63]. An ongoing trial presented at ASCO 2016, is currently evaluating
the combination of bicalutamide and palbociclib: preliminary phar-
Androgen receptor as a therapeutic target in TNBC: Clinical macokinetic results showed a good safety profile, while efficacy data
evidence have yet to be released (Table 1) [63]. Several clinical trials are cur-
rently ongoing in both the (neo)-adjuvant (NCT02689427;
The first clinical trial reporting activity of antiandrogen therapy in NCT01889238) and the metastatic settings (NCT01889238) testing
advanced breast cancer was published by Gucalp et al in 2013 and enzalutamide alone and in combination with other drugs (Table 2).
conducted by the Translational Breast Cancer Research Consortium Moreover, the PI3K/mTOR pathway has been explored in order to
(TBCRC). It was an open-label, single-arm study testing the AR an- enhance anti-AR endocrine therapy, similarly to the ER targeted
tagonist bicalutamide at the dose of 150 mg administered orally on a strategy. Preclinical evidence showed that AR-positive TNBC cell lines
continuous daily schedule, for the treatment of women with metastatic are sensitive to PI3K/mTOR inhibitors in association with bicalutamide.
AR-positive TNBC. Primary endpoint was the clinical benefit rate (CBR) An additional study found that the combination of the mTOR inhibitor
defined as the total number of patients who show a complete response Rapamycin and Enzalutamide has anti-proliferative effects against LAR
(CR), partial response (PR), or stable disease (SD) > 6 months, while xenograft models in mice [64]. Recently a combination of taselisib
secondary endpoints were progression-free survival (PFS), safety and (PI3Kα inhibitor) and Enzalutamide was launched, but recruitment has
toxicity. Twenty-six patients were evaluable for the primary endpoint been suspended (NCT02457910) (Table 1).
out of the 452 screened [40] (Table 1). Clinical benefit rate was 19%
with a median PFS of 12 weeks (range 6.25–57.5 months), giving a
proof of concept of the potential clinical role of targeting AR in TNBC Conclusion
treatment. Overall, the treatment was well tolerated and the most
common treatment-related adverse events (AEs) were fatigue, hot fla- The AR is an emerging and promising therapeutic target in breast
shes, limb edema, and elevation of liver function tests. cancer and, in particular, in the TNBC subtype, both because of the lack
Data from a phase II trial evaluating a new generation AR antago- of a well-established targetable feature and the presence of a solid
nist, enzalutamide, in AR positive TNBC, were presented at the 2015 molecular subtype with different prognosis and clinical behavior.
ASCO meeting [61]. Primary endpoint of the trial was CBR defined as Notwithstanding the effectiveness of an endocrine therapy-based

107
 L. Gerratana et al.

Table 2
Ongoing clinical trials.
Agent Study population Study design Patients (n) Primary end point Status ClinicalTrials.gov Identifier

Bicalutamide 150 mg vs AR+ metastatic TNBC Randomized, open-label, Phase III 262 estimated CBR Recruiting NCT03055312
chemotherapy
Bicalutamide 150 mg AR+ metastatic TNBC Single center, Phase II 1 CBR Terminated NCT02348281
Bicalutamide Ribociclib AR+ metastatic or locally advanced TNBC Nonrandomized, open-label, Phase 58 estimated Safety/tolerability, CBR Recruiting NCT03090165
I/II
Bicalutamide AR+ metastatic TNBC Randomized, open-label, Phase II 60 estimated CBR, PFS Not yet recruiting NCT02353988
Bicalutamide AR+/ER−/PR − metastatic BC Nonrandomized, open-label, Phase 26 CBR Results reported NCT00468715
II
Bicalutamide Palbociclib AR+ metastatic breast cancer Nonrandomized, open-label, Phase 51 Safety/tolerability, PFS Recruiting NCT02605486
I/II
Enzalutamide Paclitaxel AR+ TNBC, stage I–III breast cancer Nonrandomized, open label, Phase 37 estimated pCR, RCB-I Recruiting NCT02689427
IIB
Enzalutamide AR + TNBC, stage I–III Nonrandomized, open-label, Phase 200 estimated 1-year dose compliance Recruiting NCT02750358
II rate
Enzalutamide AR+ advanced TNBC Nonrandomized, open-label, Phase 118 CBR Active, not recruiting NCT01889238
II
4-OH-testosterone AR+ TNBC or ER+/PR+/HER2− or ER+/PR−/HER2− Nonrandomized, open-label, Phase 90 estimated CBR Recruiting NCT02067741

108
advanced BC II
AZD5312 Solid carcinomas with AR expression Nonrandomized, open-label, Phase I 32 Safety, tolerability Completed NCT02144051
Enobosarm AR+ advanced TNBC Nonrandomized, open-label, Phase 55 estimated CBR Recruiting NCT02368691
II
Enobosarm Pembrolizumab AR+ metastatic TNBC Nonrandomized, open-label, Phase 29 estimated Safety, tolerability, ORR Recruiting NCT02971761
II
Abiraterone acetate plus prednisone AR+ metastatic or locally advanced TNBC Nonrandomized, open-label, Phase 31 estimated CBR Active, not recruiting NCT01842321
II
Abiraterone acetate AZD8186 Advanced CRPC, SqNSCLC, advanced TNBC Nonrandomized, open-label, Phase I 180 estimated Safety, tolerability Recruiting NCT01884285
AZD2014
Abiraterone acetate Postmenopausal women with ER+ or AR+ metastatic or Nonrandomized, open-label, Phase 77 Safety, toxicity, CBR Completed NCT00755885
locally advanced BC I/II
Orteronel AR+ metastatic BC Nonrandomized, open-label, Phase 86 estimated CBR Recruiting NCT01990209
II
Seviteronel AR+ metastatic TNBC Nonrandomized, open-label, Phase 48 estimated CBR Recruiting NCT02130700
II
Seviteronel Advanced AR+ TNBC or ER+ BC Nonrandomized, open-label, Phase 110 estimated Safety, tolerability, CBR Recruiting NCT02580448
I/II

Clinical benefit rate (CBR), Minimal Residual Disease (RCB-I), Progression-free survival (PFS), overall response rate (ORR).
Cancer Treatment Reviews 68 (2018) 102–110
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

strategy, which could enhance both outcome and quality of life, ex- 2016.04.013.
ploiting the interlink between AR and EMT could be a potentially new [25] Naderi A, Hughes-Davies L. A functionally significant cross-talk between androgen
receptor and ErbB2 pathways in estrogen receptor negative breast cancer. Neoplasia
approach in the treatment of this subtype. 2008;10:542–8. http://dx.doi.org/10.1593/neo.08274.
[26] Eger A, Aigner K, Sonderegger S, Dampier B, Oehler S, Schreiber M, et al. DeltaEF1
Conflict of interest is a transcriptional repressor of E-cadherin and regulates epithelial plasticity in
breast cancer cells. Oncogene 2005;24:2375–85. http://dx.doi.org/10.1038/sj.onc.
1208429.
The authors declare that there are no conflicts of interest. [27] Graham TR, Yacoub R, Taliaferro-Smith L, Osunkoya AO, Odero-Marah VA, Liu T,
et al. Reciprocal regulation of ZEB1 and AR in triple negative breast cancer cells.
Breast Cancer Res Treat 2010;123:139–47. http://dx.doi.org/10.1007/s10549-009-
References 0623-7.
[28] Ohira T, Gemmill RM, Ferguson K, Kusy S, Roche J, Brambilla E, et al. WNT7a
[1] Shah PD, Gucalp A, Traina TA. The role of the androgen receptor in triple-negative induces E-cadherin in lung cancer cells. Proc Natl Acad Sci USA
breast cancer. Women’s Heal 2013;9:351–60. http://dx.doi.org/10.2217/WHE. 2003;100:10429–34. http://dx.doi.org/10.1073/pnas.1734137100.
13.33. [29] Graham TR, Zhau HE, Odero-Marah VA, Osunkoya AO, Kimbro KS, Tighiouart M,
[2] Gerratana L, Fanotto V, Bonotto M, Bolzonello S, Minisini AM, Fasola G, et al. et al. Insulin-like growth factor-I - Dependent up-regulation of ZEB1 drives epi-
Pattern of metastasis and outcome in patients with breast cancer. Clin Exp thelial-to-mesenchymal transition in human prostate cancer cells. Cancer Res
Metastasis 2015;32:125–33. http://dx.doi.org/10.1007/s10585-015-9697-2. 2008;68:2479–88. http://dx.doi.org/10.1158/0008-5472.CAN-07-2559.
[3] Bonotto M, Gerratana L, Poletto E, Driol P, Giangreco M, Russo S, et al. Measures of [30] Singh M, Spoelstra NS, Jean A, Howe E, Torkko KC, Clark HR, et al. ZEB1 expression
outcome in metastatic breast cancer: insights from a real-world scenario. Oncologist in type I vs type II endometrial cancers: a marker of aggressive disease. Mod Pathol
2014;19:608–15. http://dx.doi.org/10.1634/theoncologist.2014-0002. 2008;21:912–23. http://dx.doi.org/10.1038/modpathol.2008.82.
[4] Dai X, Li T, Bai Z, Yang Y, Liu X, Zhan J, et al. Breast cancer intrinsic subtype [31] Christenson JL, Butterfield KT, Spoelstra NS, Norris JD, Josan JS, Pollock JA, et al.
classification, clinical use and future trends. Am J Cancer Res 2015;5:2929–43. MMTV-PyMT and derived met-1 mouse mammary tumor cells as models for
[5] Ahn SG, Kim SJ, Kim C, Jeong J. Molecular classification of triple-negative breast studying the role of the androgen receptor in triple-negative breast cancer pro-
cancer. J Breast Cancer 2016;19:223. http://dx.doi.org/10.4048/jbc.2016.19.3. gression. Horm Cancer 2017;8:69–77. http://dx.doi.org/10.1007/s12672-017-
223. 0285-6.
[6] Pietri E, Conteduca V, Andreis D, Massa I, Melegari E, Sarti S, et al. Androgen re- [32] Barton VN, D’Amato NC, Gordon MA, Lind HT, Spoelstra NS, Babbs BL, et al.
ceptor signaling pathways as a target for breast cancer treatment. Endocr Relat Multiple molecular subtypes of triple-negative breast cancer critically rely on an-
Cancer 2016;23:R485–98. http://dx.doi.org/10.1530/ERC-16-0190. drogen receptor and respond to enzalutamide in vivo. Mol Cancer Ther
[7] Gucalp A, Traina TA. Targeting the androgen receptor in triple-negative breast 2015;14:769–78. http://dx.doi.org/10.1158/1535-7163.MCT-14-0926.
cancer. Curr Probl Cancer 2016;40:141–50. http://dx.doi.org/10.1016/j. [33] Gibson BA, Kraus WL. New insights into the molecular and cellular functions of poly
currproblcancer.2016.09.004. (ADP-ribose) and PARPs. Nat Rev Mol Cell Biol 2012;13:411–24. http://dx.doi.org/
[8] Luo X, Shi Y-X, Li Z-M, Jiang W-Q. Expression and clinical significance of androgen 10.1038/nrm3376.
receptor in triple negative breast cancer. Chin J Cancer 2010;29:585–90. [34] Luo J, Jin J, Yang F, Sun Z, Zhang W, Shi Y, et al. The correlation between PARP1
[9] Niemeier LA, Dabbs DJ, Beriwal S, Striebel JM, Bhargava R. Androgen receptor in and BRCA1 in AR positive triple-negative breast cancer. Int J Biol Sci
breast cancer: Expression in estrogen receptor-positive tumors and in estrogen re- 2016;12:1500–10. http://dx.doi.org/10.7150/ijbs.16176.
ceptor-negative tumors with apocrine differentiation. Mod Pathol 2010;23:205–12. [35] Von Wahlde MK, Hülsewig C, Ruckert C, Götte M, Kiesel L, Bernemann C. The anti-
http://dx.doi.org/10.1038/modpathol.2009.159. androgen drug dutasteride renders triple negative breast cancer cells more sensitive
[10] He J, Peng R, Yuan Z, Wang S, Peng J, Lin G, et al. Prognostic value of androgen to chemotherapy via inhibition of HIF-1α-/VEGF-signaling. Gynecol Endocrinol
receptor expression in operable triple-negative breast cancer: a retrospective ana- 2015;31:160–4. http://dx.doi.org/10.3109/09513590.2014.971235.
lysis based on a tissue microarray. Med Oncol 2012;29:406–10. http://dx.doi.org/ [36] Uscanga-Perales GI, Santuario-Facio SK, Ortiz-López R. Triple negative breast
10.1007/s12032-011-9832-0. cancer: deciphering the biology and heterogeneity. Med Univ 2016;18:105–14.
[11] Park S, Koo J, Park HS, Kim J-H, Choi S-Y, Lee JH, et al. Expression of androgen http://dx.doi.org/10.1016/j.rmu.2016.05.007.
receptors in primary breast cancer. Ann Oncol 2010;21:488–92. http://dx.doi.org/ [37] Lehmann BD, Bauer JA, Chen X, Sanders ME, Chakravarthy AB, Shyr Y, et al.
10.1093/annonc/mdp510. Identification of human triple-negative breast cancer subtypes and preclinical
[12] Caiazza F, Murray A, Madden SF, Synnott NC, Ryan EJ, O’Donovan N, et al. models for selection of targeted therapies. J Clin Invest 2011;121:2750–67. http://
Preclinical evaluation of the AR inhibitor enzalutamide in triple-negative breast dx.doi.org/10.1172/JCI45014.
cancer cells. Endocr Relat Cancer 2016;23:323–34. http://dx.doi.org/10.1530/ [38] Lehmann BD, Jovanović B, Chen X, Estrada MV, Johnson KN, Shyr Y, et al.
ERC-16-0068. Refinement of triple-negative breast cancer molecular subtypes: implications for
[13] Narayanan R, Dalton JT. Androgen receptor: a complex therapeutic target for breast neoadjuvant chemotherapy selection. PLoS One 2016;11:e0157368. http://dx.doi.
cancer. Cancers (Basel) 2016;8:108. http://dx.doi.org/10.3390/cancers8120108. org/10.1371/journal.pone.0157368.
[14] Basile D, Cinausero M, Iacono D, Pelizzari G, Bonotto M, Vitale MG, et al. Androgen [39] Cuenca-López MD, Montero JC, Morales JC, Prat A, Pandiella A, Ocana A. Phospho-
receptor in estrogen receptor positive breast cancer: Beyond expression. Cancer kinase profile of triple negative breast cancer and androgen receptor signaling. BMC
Treat Rev 2017;61:15–22. http://dx.doi.org/10.1016/j.ctrv.2017.09.006. Cancer 2014;14:302. http://dx.doi.org/10.1186/1471-2407-14-302.
[15] Barton VN, D’Amato NC, Gordon MA, Christenson JL, Elias A, Richer JK. Androgen [40] Gucalp A, Tolaney S, Isakoff SJ, Ingle JN, Liu MC, Carey LA, et al. Phase II trial of
receptor biology in triple negative breast cancer: a case for classification as AR+ or bicalutamide in patients with androgen receptor-positive, estrogen receptor-nega-
quadruple negative disease. Horm Cancer 2015;6:206–13. http://dx.doi.org/10. tive metastatic breast cancer. Clin Cancer Res 2013;19:5505–12. http://dx.doi.org/
1007/s12672-015-0232-3. 10.1158/1078-0432.CCR-12-3327.
[16] Rahim B, O’Regan R. AR signaling in breast cancer. Cancers (Basel) 2017;9:21. [41] Traina TA, Miller K, Yardley DA, Eakle J, Schwartzberg LS, O’Shaughnessy J, et al.
http://dx.doi.org/10.3390/cancers9030021. Enzalutamide for the treatment of androgen receptor-expressing triple-negative
[17] Lin H-K, Yeh S, Kang H-Y, Chang C. Akt suppresses androgen-induced apoptosis by breast cancer. J Clin Oncol 2018;36:884–90. http://dx.doi.org/10.1200/JCO.2016.
phosphorylating and inhibiting androgen receptor. Proc Natl Acad Sci 71.3495.
2001;98:7200–5. http://dx.doi.org/10.1073/pnas.121173298. [42] Masuda H, Baggerly KA, Wang Y, Zhang Y, Gonzalez-Angulo AM, Meric-Bernstam F,
[18] Manning BD, Toker A. AKT/PKB Signaling: navigating the network. Cell et al. Differential response to neoadjuvant chemotherapy among 7 triple-negative
2017;169:381–405. http://dx.doi.org/10.1016/j.cell.2017.04.001. breast cancer molecular subtypes. Clin Cancer Res 2013;19:5533–40. http://dx.doi.
[19] McNamara KM, Sasano H. Androgen and breast cancer: an update. Curr Opin org/10.1158/1078-0432.CCR-13-0799.
Endocrinol Diabetes Obes 2016;23:249–56. http://dx.doi.org/10.1097/MED. [43] Burstein MD, Tsimelzon A, Poage GM, Covington KR, Contreras A, Fuqua SAW,
0000000000000251. et al. Comprehensive genomic analysis identifies novel subtypes and targets of
[20] Wang Y, Romigh T, He X, Tan MH, Orloff MS, Silverman RH, et al. Differential triple-negative breast cancer. Clin Cancer Res 2015;21:1688–98. http://dx.doi.org/
regulation of PTEN expression by androgen receptor in prostate and breast cancers. 10.1158/1078-0432.CCR-14-0432.
Oncogene 2011;30:4327–38. http://dx.doi.org/10.1038/onc.2011.144. [44] Liu Y-R, Jiang Y-Z, Xu X-E, Yu K-D, Jin X, Hu X, et al. Comprehensive transcriptome
[21] Wang Y, He X, Yu Q, Eng C. Androgen receptor-induced tumor suppressor, KLLN, analysis identifies novel molecular subtypes and subtype-specific RNAs of triple-
inhibits breast cancer growth and transcriptionally activates p53/p73-mediated: negative breast cancer. Breast Cancer Res 2016;18:33. http://dx.doi.org/10.1186/
apoptosis in breast carcinomas. Hum Mol Genet 2013;22:2263–72. http://dx.doi. s13058-016-0690-8.
org/10.1093/hmg/ddt077. [45] Shi Y, Yang F, Sun Z, Zhang W, Gu J, Guan X. Differential microRNA expression is
[22] McNamara KM, Moore NL, Hickey TE, Sasano H, Tilley WD. Complexities of an- associated with androgen receptor expression in breast cancer. Mol Med Rep
drogen receptor signalling in breast cancer. Endocr Relat Cancer 2014;21:T161–81. 2017;15:29–36. http://dx.doi.org/10.3892/mmr.2016.6019.
http://dx.doi.org/10.1530/ERC-14-0243. [46] Nakano K, Miki Y, Hata S, Ebata A, Takagi K, McNamara KM, et al. Identification of
[23] Darb-Esfahani S, Denkert C, Stenzinger A, Salat C, Sinn B, Schem C, et al. Role of androgen-responsive microRNAs and androgen-related genes in breast cancer.
TP53 mutations in triple negative and HER2-positive breast cancer treated with Anticancer Res 2013;33:4811–9.
neoadjuvant anthracycline/taxane-based chemotherapy. Oncotarget [47] Parker J, Peterson A, Tudor I, Al E. A novel biomarker to predict sensitivity to
2016;7:67686–98. http://dx.doi.org/10.18632/oncotarget.11891. enzalutamide (ENZA) in TNBC. J Clin Oncol 2015;29:2015–6. http://dx.doi.org/10.
[24] Kim S, Moon BI, Lim W, Park S, Cho MS, Sung SH. Expression patterns of GATA3 1200/jco.2015.33.15_suppl.1083.
and the androgen receptor are strongly correlated in patients with triple-negative [48] Maeda T, Nakanishi Y, Hirotani Y, Fuchinoue F, Enomoto K, Sakurai K, et al.
breast cancer. Hum Pathol 2016;55:190–5. http://dx.doi.org/10.1016/j.humpath. Immunohistochemical co-expression status of cytokeratin 5/6, androgen receptor,

109
L. Gerratana et al. Cancer Treatment Reviews 68 (2018) 102–110

and p53 as prognostic factors of adjuvant chemotherapy for triple negative breast [57] Choi JE, Kang SH, Lee SJ, Bae YK. Androgen receptor expression predicts decreased
cancer. Med Mol Morphol 2016;49:11–21. http://dx.doi.org/10.1007/s00795-015- survival in early stage triple-negative breast cancer. Ann Surg Oncol 2015;22:82–9.
0109-0. http://dx.doi.org/10.1245/s10434-014-3984-z.
[49] Gasparini P, Fassan M, Cascione L, Guler G, Balci S, Irkkan C, et al. Androgen re- [58] Asano Y, Kashiwagi S, Onoda N, Kurata K, Morisaki T, Noda S, et al. Clinical ver-
ceptor status is a prognostic marker in non-basal triple negative breast cancers and ification of sensitivity to preoperative chemotherapy in cases of androgen receptor-
determines novel therapeutic options. PLoS One 2014;9:e88525. http://dx.doi.org/ expressing positive breast cancer. Br J Cancer 2016;114:14–20. http://dx.doi.org/
10.1371/journal.pone.0088525. 10.1038/bjc.2015.434.
[50] Wang C, Pan B, Zhu H, Zhou Y, Mao F, Lin Y, et al. Prognostic value of androgen [59] Aguilera TA, Rafat M, Castellini L, Shehade H, Kariolis MS, Hui ABY, et al.
receptor in triple negative breast cancer: a meta-analysis. Oncotarget Reprogramming the immunological microenvironment through radiation and tar-
2016;7:46482–91. http://dx.doi.org/10.18632/oncotarget.10208. geting Axl. Nat Commun 2016;7:13898. http://dx.doi.org/10.1038/ncomms13898.
[51] McNamara KM, Yoda T, Miki Y, Chanplakorn N, Wongwaisayawan S, Incharoen P, [60] Hilborn E, Gacic J, Fornander T, Nordenskjöld B, Stal O, Jansson A. Androgen re-
et al. Androgenic pathway in triple negative invasive ductal tumors: its correlation ceptor expression predicts beneficial tamoxifen response in oestrogen receptor-α-
with tumor cell proliferation. Cancer Sci 2013;104:639–46. http://dx.doi.org/10. negative breast cancer. Br J Cancer 2016;114:248–55. http://dx.doi.org/10.1038/
1111/cas.12121. bjc.2015.464.
[52] Sutton LM, Cao D, Sarode V, Molberg KH, Torgbe K, Haley B, et al. Decreased [61] Traina TA, Miller K, Yardley DA, Shaughnessy JO, Cortes J, Awada A, et al.
androgen receptor expression is associated with distant metastases in patients with Advanced AR + triple-negative breast cancer (TNBC). Evaluable Dx + Dx -. J Clin
androgen receptor-expressing triple-negative breast carcinoma. Am J Clin Pathol Oncol 2015;33:15–6. http://dx.doi.org/10.1200/jco.2015.33.15_suppl.1003.
2012;138:511–6. http://dx.doi.org/10.1309/AJCP8AVF8FDPTZLH. [62] Bonnefoi H, Grellety T, Tredan O, Saghatchian M, Dalenc F, Mailliez A, et al. A
[53] Gong Y, Wei W, Wu Y, Ueno NT, Huo L. Expression of androgen receptor in in- phase II trial of abiraterone acetate plus prednisone in patients with triple-negative
flammatory breast cancer and its clinical relevance. Cancer 2014;120:1775–9. androgen receptor positive locally advanced or metastatic breast cancer (UCBG
http://dx.doi.org/10.1002/cncr.28667. 12–1). Ann Oncol 2016;27:812–8. http://dx.doi.org/10.1093/annonc/mdw067.
[54] Qu Q, Mao Y, Fei X, Shen K. The impact of androgen receptor expression on breast [63] Gucalp A. TPS Poster #204b - Phase I/II trial of palbociclib in combination with
cancer survival: a retrospective study and meta-analysis. PLoS One 2013;8:e82650. bicalutamide for the treatment of androgen receptor (AR)+ metastatic breast
http://dx.doi.org/10.1371/journal.pone.0082650. cancer (MBC). Asco 2016;34:Abstract no.TPS1103. doi:10.1200/JCO.2016.34.15_
[55] Kim Y, Jae E, Yoon M. Influence of androgen receptor expression on the survival suppl.TPS1103.
outcomes in breast cancer: a meta-analysis. J Breast Cancer 2015;18:134. http://dx. [64] Robles AJ, Cai S, Cichewicz RH, Mooberry SL. Selective activity of deguelin iden-
doi.org/10.4048/jbc.2015.18.2.134. tifies therapeutic targets for androgen receptor-positive breast cancer. Breast
[56] Asano Y, Kashiwagi S, Goto W, Tanaka S, Morisaki T, Takashima T, et al. Expression Cancer Res Treat 2016;157:475–88. http://dx.doi.org/10.1007/s10549-016-
and clinical significance of androgen receptor in triple-negative breast cancer. 3841-9.
Cancers (Basel) 2017;9:4. http://dx.doi.org/10.3390/cancers9010004.

110