The steroid hormone oestrogen is central to normal female physiology, reproduction and behaviour, through its effects on cellular process including cell proliferation and cell survival. The effects are mediated by nuclear oestrogen receptors (ERα and ERβ; BOX 1). ERα is respon-sible for many of the effects of oestrogen on normal and cancerous breast tissue, through ligand-activated trans-criptional regulation (genomic actions) and by acting as a component of membrane and cytoplasmic signalling cascades (non-genomic actions)1 (FIG. 1).
Sustained exposure to endogenous or exogenous oestrogen is a well-established cau of breast cancer 2,3, underpinning the u of anti-oestrogens and aromata inhibitors in breast cancer prevention 4–6. At least 70% of breast cancers are classified as ER‑positive breast cancers 7, and interfering with oestrogen action has been a main-stay of breast cancer treatment for more than a cen-tury. Early therapies included surgical removal of the ovaries, but the synthesis of competitive inhibitors of oestrogen–ER binding during the 1970s led to the first, and to date most successful, targeted cancer therapy: the lective oestrogen receptor modulator (SERM) tamoxifen 8. Adjuvant therapy with tamoxifen almost halves the rate of dia recurrence and reduces the annual breast cancer death rate by one-third, mak-ing a significant contribution to the 25–30% decrea in breast cancer mortality in the past two decades 9. Subquently, other new, effective endocrine therapies have been develope
d that target oestrogen synthesis (such as aromata inhibitors 10) or ER signalling (such as other SERMs and ‘pure’ anti‑oestrogens 11).
One-third of women treated with tamoxifen for 5 years will have recurrent dia within 15 years 9, and so endocrine-resistant dia may reprent up to one-quarter of all breast cancers. Therefore, two major challenges for the successful treatment of breast can-cer are the development of more specific biomarkers that predict therapeutic respon to endocrine therapy and the identification of new therapeutic targets for endocrine-resistant dia. This Review summarizes and evaluates the recent insights into the mechanisms of endocrine resistance that have been made through candidate gene approaches, as well as more global gene expression profiling and functional genetic screens. We necessarily focus on tamoxifen resistance, as the experience with this drug is more extensive and the clinical data more mature than for other drugs. Many of the broad concepts discusd will probably also apply to resistance to aromata inhibitors and other anti-oestrogens, although the lack of clinical cross- resistance 10–12 indicates that some resistance mechanisms are independent.
Molecular mechanisms of resistance
The primary mechanism of de novo or intrinsic resistance to tamoxifen is lack of expression of ERα.
Recently, a cond intrinsic mechanism has been documented in which patients carrying inactive alleles of cytochrome P450 2D6 (CYP2D6) (approximately 8% of Caucasian women) fail to convert tamoxifen to its active meta-bolite, endoxifen, and are conquently less responsive to tamoxifen 13. By contrast, a plethora of mechanisms
*Cancer Rearch Program, Garvan Institute of Medical Rearch, Sydney, New South Wales 2010, Australia. ‡
St Vincent’s Clinical School, Faculty of Medicine, University of New South Wales, New South Wales 2052, Australia. e‑mails:
e.au ; r.au doi:10.1038/nrc2713
Aromata inhibitors
Drugs that function by blocking aromata, the enzyme that converts androgens to
oestrogens in tissues including the breast and adipo tissue. Examples include anastrazole, letrozole and exemestane.
ER-positive breast cancers
In current clinical practice, ER‑positive breast cancers are tho with immunohisto‑chemically detectable ER α levels .
Biological determinants of endocrine resistance in breast cancer
Elizabeth A. Musgrove*‡ and Robert L. Sutherland*‡
Abstract | Endocrine therapies targeting oestrogen action (anti-oestrogens, such as tamoxifen, and aromata inhibitors) decrea mortality from breast cancer, but their efficacy is limited by intrinsic and acquired therapeutic resistance. Candidate molecular biomarkers and gene expression signatures of tamoxifen respon emphasize the
importance of deregulation of proliferation and survival signalling in endocrine resistance. However, definition of the specific genetic lesions and molecular process that determine clinical endocrine resistance is incomplete . The development of large-scale computational and genetic approaches offers the promi of identifying the mediators of endocrine
resistance that may be exploited as potential therapeutic targets and biomarkers of respon
in the clinic.
h e r a p e u T i c r e s i s Ta n c e
Adjuvant therapy
A drug treatment (for example, chemotherapy or endocrine therapy) that is given after the primary therapy (for example, surgery and/or radiotherapy), with the aim of increasing the overall effectiveness of treatment.
SERMs
Drugs such as tamoxifen that bind the oestrogen receptor and thereby block the effects of oestrogen on tissues such as the breast but that function similarly to oestrogen in other tissues such as bone. Unlike oestrogen, the drugs are not steroidal in structure.
‘Pure’ anti-oestrogens Drugs that bind the oestrogen receptor, thereby blocking the effect of oestrogen, but have no detectable oestrogen‑like effects. Most have a steroidal structure.
Intrinsic resistance
The failure to respond to initial drug therapy.have been postulated to account for acquired resistance
following prolonged exposure to tamoxifen, some of
which may also account for intrinsic resistance in the
clinic. Much of the published information on the
potential molecular mechanisms has been derived
from ERα-positive breast cancer cell lines and from
variants of the cell lines lected for adaptation to
sustained exposure to anti-oestrogens or withdrawal of
oestrogen. Such models identify mechanisms that can
induce tamoxifen resistance in vitro rather than tho
that actually mediate resistance in patients with breast
cancer, and they have veral other potential limita-
tions. The include the degree to which the few ERα-
positive breast cancer cell lines studied reflect the range
of ER-positive phenotypes in situ and the abnce of
epithelial–stromal and tumour–host interactions that
probably modulate nsitivity in vivo. Furthermore, the
mechanisms that are responsible for the clinical obr-
vation that tamoxifen-resistant cancers often respond to
cond-line endocrine therapies10–12 remain unclear.
notwithstanding the potential limitations, stud-
ies of candidate genes involved in oestrogen signalling
(FIG. 1) or anti-oestrogen regulation of cell proliferation
and survival (FIG. 2)and more global unbiad approaches
using cell line models have yielded important concepts
and hypothes that correlate with tamoxifen resist-
ance in the clinic, and they have provided a basis for
new therapeutic approaches. Deregulation of various
aspects of oestrogen signalling is a common mechanism
for resistance, but unrelated mechanisms that provide
tumour cells with alternative proliferative and survival
stimuli also confer resistance.
As detailed information on the biology of the many
molecules implicated in tamoxifen resistance in vitro
and the means by which they cau resistance is sum-
marized in a ries of excellent recent reviews11,12,14–21,
we focus here on recent developments that shed light
on potential mechanisms of prognostic or predictive
importance. As shown by the examples in TABLE 1, most
of the molecules that modulate tamoxifen nsitivity in
experimental models are correlated with dia out-
come not only in women treated with tamoxifen but
also in a wider population of patients with breast can-
cer. Furthermore, becau adjuvant tamoxifen has been
the ‘therapy of choice’ in ER-positive breast cancer for
more than 25 years, most available patient cohorts do
not allow the comparison of therapeutic responsiveness
and outcome in well-matched control populations that
differ only with respect to tamoxifen therapy. Therefore,
it is difficult to distinguish specific differences in tumour
repon to tamoxifen from the broader effects of the
underlying biology of the dia.
ER and co-regulators. Respon to tamoxifen is rare
in ER-negative breast cancer, and so ERα expression is
currently the principal biomarker of respon to endo-crine therapy. Early studies implicated the loss of ERα expression or ERα mutations as potential mechanisms of acquired resistance. However, loss of ERα expres-sion occurs in only a minority (15–20%) of resistant breast cancers22 and <1% of ER-positive tumours have ERα mutations14,20,23. More recently, expression of a new truncated variant of ERα, ERα36, in the prence of full-length ERα has been associated with reduced responsiveness24. The development of antibodies that can distinguish between ERα, ERβ and naturally occur-ring ERβvariants (BOX 1)has led to the identification of respons in ERβ-positive but ERα-negative cancers and a potential role for the carboxy-terminally truncated var-iants of ERβ (ERβ2/cx and ERβ5) in tamoxifen respon-siveness25,26. in addition, the oestrogen-related receptor ERRγ is overexpresd and mediates tamoxifen resistance in lobular invasive breast cancer models27.
One mechanism by which ERα regulates gene expres-sion is through protein–protein interactions with other transcription factors — for example, activator protein 1 (Ap1), specificity protein 1 (Sp1) and nuclear factor-κB (nF-κB) (FIG. 1). incread Ap1 and nF-κB transcriptional activity are also associated with endocrine resistance28–30. ERα function is regulated by post-translational modifica-tions (phosphorylation, methylation and sumoylation) that influence interactions with other proteins, i
ncluding transcriptional co-regulators19 and cytoplasmic signalling molecules (FIG. 1). There is significant evidence to show that effects on the end points contribute to endocrine resistance18,20,31. Overexpression and incread phosphor-ylation of ERα co-activators, particularly nuclear receptor co-activator 3 (nCOA3; also known as AiB1 or SRC3), leads to constitutive ERα-mediated transcription, which confers resistance in vitro and in xenograft models12,16 and is associated with reduced responsiveness to tamoxifen in patients32. Transient methylation of ERα at R260 by protein arginine N-methyltransfera 1 (pRMT1) results
Cytochrome P450 2D6 (CYP2D6)
A member of the large and diver superfamily of cytochrome P450 enzymes. CYP2D6 catalys the conversion of tamoxifen into its active metabolites, endoxifen and 4‑hydroxytamoxifen. It is highly polymorphic, so its activity is variable between individuals.
Acquired resistance
In contrast to intrinsic resistance, an initial respon to drug therapy followed by subquent dia progression. in the formation of cytoplasmic complexes that contain
ERα, pi3K, the tyrosine kina SRC and focal adhesion
kina (FAK; also known as pTK2) and that activate Akt
(FIG. 1). However, it is not known whether this methyla-
tion event, which is frequent in breast cancer33, is associ-
ated with the endocrine respon. Another example is
the ER co-activator pElp1, which in many breast cancers
localizes to the cytoplasm, where it can confer tamoxifen
resistance31. pElp1 functions as a scaffold that modu-
lates ER interaction with SRC, leading to activation of
SRC and the Erk family kinas and also promotes
oestrogen activation of pi3K31(FIG. 1).
Receptor tyrosine kina signalling. The bidirectional
crosstalk between ER and receptor tyrosine kina sig-
nalling is evidenced by the early obrvations of recip-
rocal expression of ER and members of the epidermal
growth factor receptor (Egfr) family such as EGFR and
ERBB2 (also known as HER2)34. Growth factors of the
Egf and insulin-like growth factor (igf) families can
modulate tamoxifen nsitivity in vitro35; although breast
cancer cells are quiescent and innsitive to growth fac-
tor stimulation following treatment with the pure anti-
oestrogen iCi 182780 (fulvestrant), tamoxifen treatment
does not lead to growth factor innsitivity36. This has
focud attention on receptor tyrosine kina expression
and function as potential mediators of endocrine resist-
ance. incread expression of EGFR, ERBB2 and iGF1
receptor (iGF1R) can elicit tamoxifen resistance37–40, as
can activation of the components of their downstream
signalling pathways, particularly the Erk and pi3K path-
ways41–43. in some cas, deregulation of the signalling
pathways occurs as a result of genetic or epigenetic
modifications, such as amplification of ERBB2, activat-
ing mutations in PIK3CA, which encodes a catalytic
subunit of type i pi3Ks, and loss of heterozygosity or
methylation of PTEN, a tumour suppressor that inhibits
the pi3K pathway20,21. in other cas, however, deregula-
tion of the pathways reflects aberrations in upstream
regulators, such as the activation of Akt in association
with the loss of PTEN expression or overexpression of
ERBB2 (REFS 20,21) and activation of iGF1R and ERBB3
following the loss of PTEN40. How the events mediate
tamoxifen resistance has not been fully elucidated, but
veral potential contributing factors have been sug-
gested (FIGS 1,3):decread ERα expression mediated by
ERK activation; loss of ER-mediated repression of EGFR
and ERBB2 and conquent activation of mitogenic sig-
nalling cascades; ligand-independent activation of ER or
its co-activators through phosphorylation; upregulation
of key cell cycle regulators, for example MYC and the
D-type and E-type cyclins, through constitutive activa-
tion of mitogenic signalling pathways; and the inhibition
of apoptosis through constitutive activation of survival
signalling.
Overexpression of ERBB2 is one of the best-
characterized mechanisms of endocrine resistance21.
Recent evidence implicates the loss of transcriptional
repressors and amplification of ERBB2 as mechanisms
that are responsible for incread expression of this recep-
tor. The X-linked tumour suppressor forkhead box p3
(FOXp3) and the zinc finger transcription factor GATA4
can repress ERBB2 expression,even in a cell line with an
approximately tenfold amplification of ERBB2, and their
expression is negatively correlated with ERBB2 expres-
sion in breast cancer44,45. in addition, a recent pivotal
study showed that ERα-mediated repression of ERBB2 is
dependent on competition between the paired-domain
transcription factor pAX2 and the ERα co-activator
nCOA3 for binding and regulation of ERBB2 transcrip-
tion and, in turn, tamoxifen responsiveness46. A direct
relationship between FOXp3 or GATA4 expression
and tamoxifen responsiveness has not been established.
However, incread pAX2 expression and conquent
repression of ERBB2 was associated with incread sur-
vival following tamoxifen treatment, and loss of pAX2
expression in the prence of incread nCOA3 expres-
sion predicted a poor outcome46, indicating that this
mechanism is of direct clinical relevance.
Members of the Src family of tyrosine kinas, par-
ticularly SRC itlf, and their downstream targets are
also commonly overexpresd in breast cancer and
have been implicated in resistance. The Src substrate
BCAR1 (also known as CAS) is a focal adhesion adap-
tor protein that activates proliferative, survival and
invasion pathways. it can induce tamoxifen resist-
ance when overexpresd in vitro47, and BCAR1-
overexpressing breast cancers are less responsive to
tamoxifen48. BCAR1 binds and activates SRC with con-
quent phosphorylation of the Src substrates EGFR
and signal transducer and activator of transcription 5B
(STAT5B) and effects on downstream signalling path-
ways20. However, recent data suggest that the ability of
BCAR1 to confer anti-oestrogen resistance may not
require interaction with SRC49. The putative guanine
nucleotide-exchange factor BCAR3, which synergizes
with BCAR1 to activate SRC50, also caus tamoxifen
resistance in vitro51. in addition, BCAR3 activates
Rac and p21-activated kina 1 (pAK1)52; the latter is
itlf implicated in tamoxifen resistance through ERα
phosphorylation53
.
size of tumours before surgery. Cyclin E1
Cyclin E1 and cyclin E2 are regulatory subunits of kina complexes that contain CDK2 as their catalytic subunit and regulate the G1 to S pha cell cycle transition.
Data from experimental model
Neoadjuvant endocrine therapy leads to decread pro-
liferation54, and in cell culture anti-oestrogen treat-
ment leads to a G1 pha-specific cell cycle arrest and
a conquent reduction in growth rate55. not unexpect-
edly, the molecules pivotal to the anti-oestrogen effects
on cell cycle progression (FIG. 2a) have central roles in
the control of G1 pha progression downstream of
polypeptide growth factor mitogens, as well as oestro-
anti-oestrogen targets confers resistance in vitro and is
associated with reduced tamoxifen responsiveness in
patients. Overexpression of MYC, cyclin E1, cyclin D1
or the cyclin D1 splice variant cyclin D1b, or the inac-
tivation of the RB tumour suppressor — an important
substrate for cyclin-dependent kinas (CDKs) that are
active in G1 pha — and the decread expression of
the CDK inhibitors p21 or p27, results in decread anti-
oestrogen nsitivity in vitro56–64. MYC overexpression
and conquent tamoxifen resistance is accompanied by
transcriptional repression of CDKN1A (which encodes
Nature Reviews |Cancer
Cyclin D1
The regulatory subunit of a kina complex that functions as a growth factor nsor to regulate G1 pha cell cycle progression. The catalytic subunits of cyclin
D1‑dependent kinas are CDK4 and CDK6.Cell survival signalling and apoptosis. Treatment with
high (micromolar) concentrations of anti-oestrogen,
oestrogen withdrawal (mimicking the effects of aroma-
ta inhibitors) or aromata inhibitor treatment of cells
transfected with aromata leads to the activation of the
cellular stress respon and apoptosis in breast cancer
cells17,76. The molecular mechanisms are not well defined,
but veral molecular conquences that promote apop-
tosis have been documented, including the regulation of
Bcl‑2 family members and increas in the apoptotic cond
mesnger ceramide17,76(FIG. 2b). Crosstalk between the
apoptotic effects of anti-oestrogens and the tumour necro-
sis factor (TnF) pathway, as well as anti-oestrogen effects
on survival signalling through the pi3K–Akt, nF-κB
and interferon pathways, is also likely to contribute to
anti-oestrogen-mediated apoptosis17(FIG. 2b). Finally,
intriguing recent obrvations indicate that autophagy is
a mechanism of cell survival in breast cancer cells that are
resistant to apoptotic concentrations of tamoxifen77.
it has been difficult to establish the role of apop-
tosis in the clinical tting. neoadjuvant studies have
yielded conflicting data and have been limited by small
patient numbers and the methodological challenges of
measuring apoptosis in vivo78. nevertheless, many sig-
natures of respon to endocrine therapy include genes
with roles in apoptosis, as discusd below. As tumour
growth reflects the balance between cell proliferation
and cell death, disruption of this balance by effects on
survival signalling and apoptosis are expected to affect
clinical respon. There is accumulating evidence for
the incread expression of anti-apoptotic molecules,
for example BCl-2 and BCl-X
l
, and decread expres-
sion of pro-apoptotic molecules, for example BAK, BiK
and caspa 9, in attenuated respons to tamoxifen17.
Although many of the respons are probably con-
quences of the activation of survival signalling through
the pi3K–Akt pathway, as a conquence of overex-
pression of receptor tyrosine kinas and incread
‘non-genomic’ signalling from cytoplasmic ER,
other pathways have been documented. For example,
incread DnA-binding and transcriptional activity
of nF-κB are features of tamoxifen-resistant cells30, and
tamoxifen nsitivity can be restored by parthenolide,
a specific nF-κB inhibitor30,79. Tamoxifen innsitivity
in vitro is also associated with the downregulation of
iRF1, an interferon-responsive putative tumour sup-
pressor that binds nF-κB and is esntial for apoptosis.
Furthermore, overexpression of a splice variant of human
X-box-binding protein 1 (XBp1), a transcription factor
that controls the unfolded protein respon, is also associ-
ated with tamoxifen resistance in vitro and poor survival
in patients with breast cancer treated with tamoxifen80–82.
nF-κB, XBp1 and iRF1 expression are correlated in
patients with breast cancer83, which may indicate that
the molecules function in a common pathway14.
signatures of tamoxifen responsiveness
The advent of genome-wide gene expression analy-
sis allowed clinical material from patients of known
responsiveness to tamoxifen to be ud as a means of
gaining broad insights into the potential mechanisms
of endocrine resistance. it also helped in the develop-
ment of clinically relevant markers of respon and
potential mechanisms of resistance. This is not without
its own limitations, for example the difficulty of obtaining
tumour tissue at the time when resistance has developed,
rather than before therapy.
Selection on the basis of dia outcome. The identifica-
tion of women who are unlikely to respond to endocrine
therapy, but who may benefit from chemotherapy, is a
a | Anti-oestrogen (AE) treatment of cultured breast cancer cells leads to oestrogen receptor (ER) binding and subquent rapid decreas in the expression of MYC, followed by decread expression of cyclin D1. Downregulation of MYC leads to de-repression of CDKN1A (which encodes p21) transcription. In addition, becau cyclin D1–cyclin-dependent kina 4 (CDK4) complexes function as a cellular ‘sink’ for the CDK inhibitors p21 and p27, the reduction in cyclin D1–CDK4 abundance makes p21 and p27available for cyclin E1–CDK2 binding, and so indirectly contributes to the inhibition of cyclin
E1–CDK2 activity. The decrea in activity of both CDK2 and CDK4 prevents RB phosphorylation (P) and therefore impedes transition into S pha. Treatment with the pure anti-oestrogen ICI 182780, but not tamoxifen, leads to an increa in the expression of p27 and molecular markers that are characteristic of quiescence (G0), that is, the formation of p130–E2F4 complexes and the accumulation of hyperphosphorylated E2F4. The effects are reviewed in REF. 55. b | Proteins and process that are upregulated during anti-oestrogen-induced apoptosis are indicated in green (reviewed in REF. 76); red cross indicate proteins that are downregulated.Apoptotic concentrations of tamoxifen elicit caspa activation downstream of respons such as activation of the stress kinas Jun N-terminal kina (JNK) and p38 MAPK, activation of the intracellular cond mesnger ceramide, transcriptional downregulation of anti-apoptotic molecules including BCL-2, and upregulation of pro-apoptotic molecules such as IRF1, BIK and possibly BAK. In addition, anti-oestrogens have effects on the interferon (IFN) and nuclear factor-κB (NF-κB) pathways, and on survival signalling through Akt downstream of receptor tyrosine kinas (RTKs), as well as synergistic effects on tumour necrosis factor (TNF)-mediated apoptosis. The pro-apoptotic effects of tamoxifen are oppod by autophagy77. JAK, janus kina; Stat, signal transducer and transcription activator.
