Lawrence Berkeley National Laboratory
Recent Work
Title
Pre-metastatic niches: organ-specific homes for metastases.
Permalink
https://escholarship.org/uc/item/5n75t404
Journal
Nature reviews. Cancer, 17(5)
ISSN
1474-175X
Authors
Peinado, Héctor
Zhang, Haiying
Matei, Irina R
et al.
Publication Date
2017-05-01
DOI
10.1038/nrc.2017.6
Peer reviewed
eScholarship.org Powered by the California Digital Library
University of California
A pivotal discovery by Stephen Paget in 1889 (REF.1) pos-
tulated that metastasis is dependent on the interactions
between ‘seeds’ (or the cancer cells) and the ‘soil’ (or the
host microenvironment). Paget’s theory was challenged in
the 1930s by James Ewing, who advocated that metastatic
dissemination could be explained solely by the dynam-
ics of haematogenous flow
2
(FIG.1). Ewing’s perspective
became the prevalent viewpoint until Isaiah Fidler’s
research in the 1970s demonstrated that, although the
mechanical properties of blood flow were important,
successful metastatic colonization could occur only at
certain organ sites
3,4
. In addition to strengthening Paget’s
theory, Fidler’s findings reignited interest in the question
that first captivated Paget
3,5
: why do tumour cells emerge
only as disseminated tumour cells (DTCs) within specific
organs? Is metastatic seeding monoclonal or polyclonal
in nature? Moreover, does metastatic seeding occur only
directly from the primary tumour or is secondary seeding
from one metastatic organ to another also a biologically
relevant event? This organ specificity observed in metas-
tasis is known as organotropism and remains one of the
most intriguing unanswered questions in cancer research.
Additional fundamental discoveries revealed that
tumours induce the formation of microenvironments
in distant organs that are conducive to the survival
and outgrowth of tumour cells before their arrival at
these sites
6–9
. These predetermined microenvironments
are termed ‘pre-metastatic niches’ (PMNs)
7
(FIGS1,2).
Since the existence of the PMN was first demonstrated,
numerous studies have identified various molecules
that regulate its stepwise evolution, highlighting the
complex molecular and cellular changes that occur
in the PMN to support future metastatic tumour
growth
6,10,11
(FIG.2). PMNs are the result of the combined
systemic effects of tumour-secreted factors and tumour-
shed extracellular vesicles (EVs) that promote a tempo-
ral sequence of events during the evolution of PMNs.
Vascular leakiness is the earliest event in this sequence,
followed by the alteration of local resident cells, such as
fibroblasts, and the recruitment of non-resident cells,
such as bone marrow-derived cells (BMDCs), to these
PMNs, subsequently attracting circulating tumour cells
(CTCs)
6,7,12
. Although congruent with both Paget’s and
Ewing’s theories, the concept of the PMN is unique
as it proposes that the primary tumour preconditions
specific organ sites for future metastatic disease (that
is, before CTC arrival) via tumour-derived factors.
Therefore, in contrast to the metastatic niche, which
is initiated and shaped upon CTC arrival, the PMN
represents an abnormal, tumour growth-favouring
microenvironment devoid of cancercells.
In addition to the contribution of tumour-secreted
factors to PMN formation, there are other tumour-
independent pathological and physiological processes
involved, such as the effects of surgery, infection and
ageing (that is, the ageing bone marrow), which alter
the local milieu and help to create a microenvironment
that is sufficiently receptive to colonization by CTCs
13–
15
. It has become clear that extracellular matrix (ECM)
remodelling is crucial for establishing the PMN (FIG.2).
Moreover, PMNs are probable sites of immune dereg-
ulation, owing to the presence of a pro-tumorigenic,
Correspondence to D.L.
Children’s Cancer and Blood
Foundation Laboratories,
Departments of Pediatrics,
and Cell and Developmental
Biology, Drukier Institute for
Children’s Health, Meyer
Cancer Center, Weill Cornell
Medicine, New York,
New York 10021, USA.
dcl2001@med.cornell.edu
doi:10.1038/nrc.2017.6
Published online 17 Mar 2017
Disseminated tumour cells
(DTCs). Thought to originate
from CTCs that reach distant
organs and survive in
these new distant
microenvironments.
Tumour-secreted factors
Also known as the tumour
secretome. The totality of
factors released by tumour
cells into their immediate
environment or into the
systemic circulation. They
include growth factors,
hormones, cytokines,
chemokines and extracellular
matrix components, as well as
extracellular vesicles.
Pre-metastatic niches: organ-specific
homes for metastases
Héctor Peinado
1,2
*, Haiying Zhang
1
*, Irina R.Matei
1
*, Bruno Costa-Silva
1,3
,
Ayuko Hoshino
1
, Goncalo Rodrigues
1,4
, Bethan Psaila
5
, Rosandra N.Kaplan
6
,
Jacqueline F.Bromberg
7
, Yibin Kang
8,9
, Mina J.Bissell
10
, Thomas R.Cox
11
,
Amato J.Giaccia
12
, Janine T.Erler
13
, Sachie Hiratsuka
14
, Cyrus M.Ghajar
15
and David Lyden
1,16
Abstract
|
It is well established that organs of future metastasis are not passive receivers of
circulating tumour cells, but are instead selectively and actively modified by the primary
tumour before metastatic spread has even occurred. Sowing the ‘seeds’ of metastasis requires
the action of tumour-secreted factors and tumour-shed extracellular vesicles that enable the
‘soil’ at distant metastatic sites to encourage the outgrowth of incoming cancer cells. In this
Review, we summarize the main processes and new mechanisms involved in the formation of
the pre-metastatic niche.
NATURE REVIEWS
|
CANCER ADVANCE ONLINE PUBLICATION
|
1
REVIEWS
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Extracellular vesicles
(EVs). A heterogeneous
population of membrane-
surrounded structures released
by cells into the intercellular
space and the circulation. Their
sizes range from 30 nm to 5 μm
in diameter and they include
exosomes (typically
30–150 nm), microvesicles
(150–1,000 nm) and apoptotic
bodies (1–5 μm).
Vascular leakiness
Loss of vascular integrity
resulting in increased
permeability of vessels to
macromolecules and cells that
normally face resistance or do
not cross endothelial barriers.
Circulating tumour cells
(CTCs). Rare cells shed by solid
tumours into the systemic
circulation at an estimated
frequency of 1:500,000–
1:1,000,000 circulating cells.
Metastatic niche
Microenvironment in distant
organs that supports the
survival and outgrowth of
tumour cells.
inflammatory milieu induced by tumour-secreted
factors, which creates immunosuppression and
coagulation disorders
6,10,16
(FIG.2).
In this Review, we summarize the progression of
PMN formation. We discuss the main mechanisms
involved in PMN generation and their impact on
host cells in pre-metastatic organs. We also discuss a
novel concept: the generation of specialized micro-
environments, named ‘sleepy niches’, where tumour cells
could survive in a dormant state (BOX1).
Setting the clinical stage
Most of the work exploring tumour-directed PMN for-
mation has used orthotopic and transgenic mouse models
of metastasis, and most of our understanding of PMN
biology is based on studies of lung metastasis, one of
the most frequent sites of metastasis in both preclinical
models and patients
17,18
. Preclinical models of metastasis
enable researchers to test therapies in laboratory condi-
tions that mimic disease present in patients. Nonetheless,
there are several challenges associated with setting up
spontaneous metastasis experiments such as the limited
availability of syngeneic mouse models and the fact that
metastasis is not confined to one metastatic organ
18
.
Currently, technical limitations, such as the lack of
specific probes to track PMNs in real time by positron
emission tomography (PET) and nuclear magnetic
resonance (NMR) or the difficulty of obtaining pre-
metastatic tissues from patients, remain the biggest
obstacles in clarifying the clinical importance of PMNs
for future metastasis. However, this gap in knowledge is
now being addressed by several groups. These include
a large cohort study assessing PMNs and metastatic
niches in omental tissues from patients undergoing surgi-
cal resection of primary ovarian carcinomas (F. Balkwill,
personal communication), as well as studies in pre-
metastatic livers from patients undergoing surgical
resection of primary pancreatic tumours (D.L. and W. R.
Jarnagin, unpublished data).
Improving the understanding of PMN biology in
patients requires the development of new technolo-
gies for better imaging techniques, such as the devel-
opment of imaging probes to differentiate PMNs from
healthy tissue. Thus, efforts to further uncover the clin-
ical relevance and other ramifications of these niches
are on going. For instance, much research is needed to
explore the possibility of persisting PMNs that coexist
once metastatic niches are created in a specific organ
site and to determine whether established metastatic
niches (or metastases on their own) can modulate
secondary PMNs at additional organ sites of future
metastasis (FIG.1).
Organ-specific PMNs
Increasing clinical evidence reveals the existence of
PMNs in tissue samples derived from cancer patients, as
observed in sentinel lymph nodes (LNs) resected from
patients with colorectal, prostate, breast, thyroid, blad-
der, gastric and renal cell carcinomas
8
. LN PMNs were
identified in mouse models of experimental metastasis
and in patients with breast cancer, showing that pro-
vasculogenic vascular endothelial growth factor receptor 1
(VEGFR1)
+
myeloid progenitor cells colonize LNs primed
by tumour-secreted factors before the arrival of CTCs
19
.
Subsequently, it was demonstrated that lymphangiogenesis
precedes CTC arrival at future sites of LN metastasis
20,21
.
Tumour cells encounter immune cells within LNs and the
lung and interact with them directly or indirectly through
tumour-secreted factors or EVs
22,23
. These interactions
likely modulate immune responses againstCTCs
24
.
Liver PMN formation, which occurs during the ini-
tial phases of visceral metastasis (that is, meta stasis of
gastrointestinal tract cancers, breast cancer and mela-
noma), also relies on BMDC recruitment
25–27
. Enzymes
involved in ECM remodelling, such as tissue inhibitor
of metallo peptidase 1 (TIMP1), generate liver PMNs
through stromal cell-derived factor 1 (SDF1; also
known as CXCL12)–C-X-C chemokine receptor type 4
(CXCR4)-dependent recruitment of neutrophils
28
. More
recently, exosomes were demonstrated to be biomarkers
and functional contributors to liver PMNs
29
. Specifically,
exosomes derived from human pancreatic cancer cell
lines expressing macrophage migration inhibitory factor
(MIF) ‘educated’ Kupffer cells to produce transforming
growth factor-β (TGFβ), which promoted ECM remodel-
ling by hepatic stellate cells, enabling recruitment of bone
marrow-derived macrophages. Thus, exosomal MIF
levels may serve as an early biomarker for liver PMN
Author addresses
1
Children’s Cancer and Blood Foundation Laboratories, Departments of Pediatrics, and Cell
and Developmental Biology, Drukier Institute for Children’s Health, Meyer Cancer Center,
Weill Cornell Medicine, New York, New York 10021, USA.
2
Microenvironment and Metastasis Group, Department of Molecular Oncology, Spanish
National Cancer Research Center (CNIO), Madrid 28029, Spain.
3
Systems Oncology Group, Champalimaud Research, Champalimaud Centre for the
Unknown, Avenida Brasília, Doca de Pedrouços, 1400-038 Lisbon, Portugal.
4
Graduate Program in Areas of Basic and Applied Biology, Abel Salazar Biomedical Sciences
Institute, University of Porto, 4099-003 Porto, Portugal.
5
Centre for Haematology, Department of Medicine, Hammersmith Hospital, Imperial
College London, London W12 0HS, UK.
6
Center for Cancer Research, Pediatric Oncology Branch, National Cancer Institute,
National Institutes of Health, Building 10-Hatfield CRC, Room 1-3940, Bethesda,
Maryland 20892, USA.
7
Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York
10065, USA.
8
Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544, USA.
9
Rutgers Cancer Institute of New Jersey, New Brunswick, New Jersey 08903, USA.
10
Biological Systems and Engineering Division, Lawrence Berkeley National Laboratory,
Berkeley, California 94720, USA.
11
The Garvan Institute of Medical Research and The Kinghorn Cancer Centre, Cancer
Division, St Vincent’s Clinical School, Faculty of Medicine, University of New South Wales,
Sydney, NSW 2010, Australia.
12
Department of Radiation Oncology, Stanford University, Stanford, California 94305, USA.
13
Biotech Research and Innovation Centre (BRIC), University of Copenhagen (UCPH),
Copenhagen 2200, Denmark.
14
Department of Pharmacology, Tokyo Women’s Medical University School of Medicine,
8-1 Kawada-cho, Tokyo 162-8666, Japan.
15
Public Health Sciences Division/Translational Research Program, Fred Hutchinson Cancer
Research Center, Seattle, Washington 98109, USA.
16
Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York,
New York 10065, USA.
*These authors contributed equally to this work
REVIEWS
2
|
ADVANCE ONLINE PUBLICATION www.nature.com/nrc
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Single-organ PMN formation
with monoclonal metastasis
Multiple and organ-specific PMN
formation with monoclonal or
polyclonal metastasis
Multiple-organ PMN
formation with
organ-specific
monoclonal metastasis
b
c
a
Primary tumour
• PMN establishment
• Metastatic cell seeding
Lymph
node
Dormant niche
Metastatic niche
PMN
Lungs
Lymphatic
dissemination
Haematogenous
dissemination
Tumour cell Dormant tumour cell
Distinct tumour cell subclones
Metastatic tumour cell
Tumour-secreted
factors and EVs
Tumour
cell seeding
Extracellular matrix
(ECM). Comprising molecules,
specifically proteoglycans and
fibrous proteins (fibronectin,
collagen, elastin and laminin)
secreted by stromal cells into
the microenvironment, that
generate an intricate network
of macromolecules that fill the
intercellular space.
Orthotopic
Derived from the Greek orthos,
meaning right and topos,
meaning place, this terminology
refers to grafting a tumour into
the place in the body where it
would normally arise and grow.
Transgenic
Relating to or denoting an
organism that contains genetic
material into which DNA from
an unrelated organism has
been artificially introduced.
Omental tissues
A double fold of peritoneum
attached to the stomach and
connecting it with certain
organs of the abdominal
viscera, composed of the
greater and the lesser
omentum, which are the
membranes of the bowels.
Neutrophils
Also known as
polymorphonuclear
leukocytes. Mature granular
white blood cells with a
multilobular nucleus and
cytoplasm containing very fine
granules. They are typically the
first responders to acute
inflammation, such as bacterial
infection, injury or certain
cancers.
Exosomes
Extracellular vesicles (typically
30–150 nm in diameter) of
endocytic origin, released into
the extracellular space by all
cell types through the fusion of
multivesicular bodies with the
plasma membrane.
Kupffer cells
Specialized liver-resident
phagocytic macrophages that
line the walls of the liver
sinusoid blood vessels.
Stellate cells
Pericytes that reside in the
area between liver sinusoid
blood vessels and hepatocytes.
They play a prominent role in
liver fibrosis and may function
as liver-resident
antigen-presenting cells.
formation. Another exosomal biomarker, glypican 1, may
also function in liver PMN formation, as well as being an
indicator of early pancreatic cancer
30
. Thus, the mecha-
nisms of PMN formation at sites of visceral metastasis
share common features, such as BMDC influx and the
horizontal transfer of information via tumour-secreted
factors and exosomes.
Bone is a common site of metastasis for many types of
solid tumour
31
. Primary tumours and the circulating fac-
tors derived from them can condition target cells residing
in the bone to support PMNs and metastatic cell colo-
nization
14,32,33
. The reciprocal crosstalk between tumour
cells and bone cells creates an environment ripe for osteo-
blastic (that is, bone-forming) or osteolytic (that is, bone-
degrading) metastases. In osteolytic metastases (for
example, from breast cancers), hypoxic tumour-secreted
factors, such as the ECM-modifying enzyme lysyl oxi-
dase (LOX), are essential for the formation of PMN
osteolytic lesions
33
(discussed further below). Other
secreted factors such as parathyroid hormone-related
protein (PTHrP)
34
, interleukin-6 (IL-6)
34–37
and matrix
metalloproteinases (MMPs)
38–40
are known to activate
osteoblasts, releasing high levels of receptor activator
of nuclear factor-κB (NF-κB) ligand (RANKL), which
in turn induces osteoclast differentiation. These factors
in turn promote the survival, growth and continued
pro-osteolytic signalling of cancer cells
14,41
. In osteo-
blastic metastases (for example, from prostate cancers)
there is still scarce information regarding PMN forma-
tion. Primary tumour-secreted factors such as WNT,
Figure 1
|
Factors involved in PMN formation. a
|
Tumour-secreted factors and extracellular vesicles (EVs) shed by the
primary tumour (blue arrows) induce pre-metastatic niche (PMN) formation in target organs before metastatic cell
seeding occurs. Tumour cells seeding (pink arrows) the PMN thrive and give rise to micrometastasis. Cells seeding
non-PMN areas lack a supportive tissue microenvironment and fail metastatic colonization. Specific niches — known as
dormant niches — have contrasting effects to PMNs on circulating tumour cells (CTCs), promoting tumour cell dormancy
instead of metastatic outgrowth. Molecules such as thrombospondin 1 (TSP1), deposited around stable microvasculature
in response to bone morphogenetic protein (BMP) and transforming growth factor-β (TGFβ) secreted by stromal cells, are
crucial mediators of cancer cell quiescence and thus dormancy. Dormant niches may arise in specific metastatic organs
that constitute a growth-suppressive microenvironment for disseminated tumour cells (DTCs) (see also BOX1). b
|
In most
types of solid cancer, a primary tumour has the ability to seed more than one organ. Tumour-secreted factors and EVs can
promote PMN formation in different organs, modulating the fate of tumour cell seeding at each metastatic site.
Organ-specific seeding has also been observed and extensively reported in the literature. Most often, organ-specific
metastasis is the result of metastatic colonization by a specific tumour subclone. Metastatic dissemination can be
achieved via the haematogenous or lymphatic system. Interestingly, some cancer types, such as sarcomas, preferentially
spread through blood vessels
200,201
, whereas others, such as epithelium-derived cancers, spread via the lymphatic
vessels
200,201
. c
|
Distinct PMNs form in different organs as a result of seeding by specific subclones derived from the primary
tumour or owing to distinct microenvironments present at secondary sites of metastasis. Organ-specific PMNs may
promote the formation of polyclonal metastatic lesions, composed of different tumour cell subclones, as opposed to the
‘traditional’ view that metastatic lesions are monoclonal in origin.
REVIEWS
NATURE REVIEWS
|
CANCER ADVANCE ONLINE PUBLICATION
|
3
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Early PMN
Metastatic niche
Evolving PMN
b
c
a
Vascular disruption
Tumour cell
Tumour
initiating cell
Immune cell
Epithelial cell
Endothelial cell
Platelet
CTCs
Fibroblast
Haematopoietic
progenitor cell
Clot
formation
BMDC
activation
BMDC mobilization
ANGPTL4
↑CCL2
PLGF
TF
Lung
Lung
Lung
Bone marrow
Tumour-derived
EV
Tumour-secreted
factor
Stroma-secreted
factor
Fibrin
ECM
↑FAK
VEGFA
Adhesion
↑E-selectin
ECM
remodelling
Inflammation
S100A4, SDF1,
CXCR4 and BV8
TGFβ
LOX
IL-6
TLR3
TNF
TNC
CCL2
HIF1α
TIMP1
MMPs
↑Periostin
↑S100s
↑SAA
Primary tumour
Proliferation
Adhesion
Invasion
Recruitment
Survival
Extravasation
↑WNT
signalling
↑Notch
signalling
↑Chemotacic
signals
• CXCL5
• CXCL7
↑FN
Versican
ANGPT2
RAMP2 loss
E-selectin
MMP3MMP10
MET
Tumour-secreted
factors and EVs
Immune cell recruitment
Tumour cell
seeding
Blood–brain barrier
(BBB). A complex structure
formed by the tight
interactions between the brain
endothelium, surrounded by
the basal lamina and stabilized
by pericytes, glial cells and
neurons.
bone morphogenetic proteins (BMPs), fibroblast growth
factor (FGF), insulin-like growth factor I (IGF1) and
IGF2, endothelin 1, prostate-specific antigen (PSA; also
known as KLK3) and vascular endothelial growth factor
A (VEGFA) influence osteoblast activity directly or indi-
rectly
42–45
and therefore may be candidates for driving
osteoblastic PMNs. Whether these factors could execute
these changes distally is still a matter ofstudy.
Last but not least, the brain, as an immuno-
privileged site protected by the unique properties of
the blood–brain barrier (BBB), remains a challenge, as
brain metastasis is one of the deadliest complications
of treatment failure
46
. Several recent studies have pro-
vided considerable insight into the evolution of BBB
vascular permeability during cancer metastasis and its
consequences for therapeutic efficacy
46–49
. For exam-
ple, reduction in expression of the astrocytic basement
membrane component laminin-α2 and the CD13
+
pericyte subpopulation are observed when BBB per-
meability is compromised
47
. Although the brain PMN
remains mostly uncharted territory, glucose metabo-
lism is one of the most relevant mechanisms involved
in breast cancer metastasis to the brain
50
as will be
discussed furtherbelow.
REVIEWS
4
|
ADVANCE ONLINE PUBLICATION www.nature.com/nrc
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.