REVIEWS

Targeting macrophages:
therapeutic approaches in cancer
Luca Cassetta1 and Jeffrey W. Pollard1,2*
Abstract | Infiltration of macrophages in solid tumours is associated with poor prognosis and
correlates with chemotherapy resistance in most cancers. In mouse models of cancer,
macrophages promote cancer initiation and malignant progression by stimulating angiogenesis,
increasing tumour cell migration, invasion and intravasation and suppressing antitumour
immunity. At metastatic sites, macrophages promote tumour cell extravasation, survival and
subsequent growth. Each of these pro-tumoural activities is promoted by a subpopulation of
macrophages that express canonical markers but have unique transcriptional profiles, which
makes tumour-associated macrophages (TAMs) good targets for anticancer therapy in humans
through either their ablation or their re‑differentiation away from pro-tumoural towards
antitumoural states. In this Review, we evaluate the state of the art of TAM-targeting strategies,
focusing on the limitations and potential side effects of the different therapies such as toxicity,
rebound effects and compensatory mechanisms. We provide an extensive overview of the
different types of therapy used in the clinic and their limitations in light of known macrophage
biology and propose new strategies for targeting TAMs.

1
MRC Centre for Reproductive
Health, College of Medicine
and Veterinary Medicine,
Queen’s Medical Research
Institute, The University of
Edinburgh, Edinburgh, UK.
2
Department of
Developmental and Molecular
Biology, Albert Einstein
College of Medicine, Bronx,
NY, USA.
*e-mail:
jeff.pollard@ed.ac.uk
doi:10.1038/nrd.2018.169
Published online 26 Oct 2018
The concept that has emerged over the past century
suggests that macrophages represent an evolutionarily
ancient, dispersed, homeostatic system, on a par with
the nervous and endocrine systems … Macrophages
are essential for survival and provide an attractive
target to manipulate the host, for both immunologic
and metabolic purposes.
Siamon Gordon, The macrophage, past, present,
future. European Journal of Immunology 37 (Suppl. 1),
S9–S17 (2007).
Metastases cause 90% of cancer deaths worldwide.
In most cases, this is because metastatic tumour cells
become resistant to therapy and develop methods to
evade immune responses. Resistance to cancer treatment
can be intrinsic to the tumour cells, but it is often con-
ferred by non-malignant cells that make up the tumour
microenvironment (TME). The TME is composed of
tissue-resident cells and a large proportion of recruited
immune cells that, in certain solid tumours such as
breast cancer, can constitute up to 50% of the tumour
mass. Original theories assumed that these immune
cells were part of the body’s response to reject tumours.
Indeed, it is still proposed that, at the earliest stage of
tumour onset, the immune system reacts to the presence
of cancer by activating T cells and macrophages, which
clear the tumour and reduce the incidence of cancer 1.
However, once tumours progress past this initial state,
the immune TME is modified to support the tumour and
promote its progression while suppressing any immune
cell-mediated cytotoxicity 2. In this process, substantial
clinical and experimental evidence indicates that macro­
phages — present abundantly in most tumour types
— have a major regulatory role in promoting tumour
progression to malignancy 3.
Tumour-associated macrophages (TAMs), at least
in mouse models, largely originate from bone marrow
monocytes that are recruited through inflammatory sig-
nals released by cancer cells in the primary and meta­
static tumour, where they differentiate into TAMs and
facilitate tumour progression4,5. However, in cancers
such as gliomas and those of the pancreas, TAMs can
also derive from erythro-myeloid progenitors (EMPs)
developed in the yolk sac at the embryonic stage6–8
(BOX 1). Nevertheless, in either case, the TME promotes
differentiation of these progenitors to new tumour-
associated phenotypes that vary depending on their
location in the tumour but are generally pro-tumoural.
Several strategies in immuno-oncology have been
developed in the past few years to reactivate the adap-
tive and innate immune systems to mount a robust anti-
tumoural immune response as an alternative approach
to classic anticancer treatments, to which tumours gen-
erally develop resistance. Clinical trials with immune

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 887

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Box 1 | Macrophage lineage and TAM origin redefined

The understanding of the origins of macrophages has recently undergone a profound shift owing to the use of modern
lineage tracing techniques. These methods use inducible Cre recombinases expressed from cell-specific promoters —
such as the inducible Csf1r‑iCre/Esr1 mouse crossed with floxed, coloured reporter mice — that permanently tag
progeny cells at specific times in development. These methods, together with single-cell and bulk RNA sequencing, have
allowed precise developmental origins of macrophages to be ascertained. A revelation in these studies is that most
tissue-resident macrophages are not derived from bone marrow progenitors, as previously thought, but instead from the
yolk sac or fetal liver270,271. Detailed lineage tracing has indicated that macrophages originate from at least three different
embryonic sources (from erythro-myeloid progenitor (EMP) in the yolk sac and fetal liver, and from macrophage/dendritic
cell progenitor cells (MDPs) in the bone marrow) and differentiate into tissue-specific macrophages according to their
origin, or to other cell types (such as common dendritic cell progenitors (CDP; see the figure) 272,273. Some tissue-specific
macrophages in adults are almost exclusively derived from one source (such as those in the brain, which are derived from
yolk sac, and those in the intestine, which are derived from bone marrow), while in other tissues — such as the skin —
different proportions of macrophages derive from one source or another. In some tissues, such as heart, the yolk sac‑
derived macrophages are replaced from fetal liver monocytes (however, see REFS269,271 for different lineage
interpretations)271,274,275. The presence of persistent embryonic populations throughout life in most tissues suggests that
these tissues harbour pre-macrophages (pMACs) that can proliferate to give rise to mature macrophages275. These data
have also revealed different regulatory mechanisms; for example, bone marrow‑derived cells require transcription factor
cMYB, which is not required for yolk sac‑derived macrophages270. Nevertheless, the dominant transmembrane receptor
controlling the differentiation and survival of almost all macrophages regardless of their origins is the colony-stimulating
factor 1 receptor (CSF1R). In its absence, almost all macrophages are dramatically depleted in mice, with some notable
exceptions such as the lung, in which they are regulated by CSF2 (REF.276). However, it should be recognized that in
addition to this requirement for CSF1R there are tissue-specific growth factors and chemokines as well as
microenvironmental cues that specify macrophage local identity277.
These observations open a number of intriguing questions, for example, about the importance of the phenotype of
macrophages according to their lineage compared with their tissue environment, whether the replacement of yolk
sac-derived or fetal liver-derived macrophages with bone marrow-derived macrophages results in identical phenotypes
and whether macrophages from different origins can be individually targeted. In the context of tumours, these questions
are important because, although in many mouse models most TAMs seem to be of bone marrow origin, there are several
exceptions such as in models of pancreatic cancer and glioma, in which the populations are a mixture of fetal
liver-derived and bone marrow‑derived macrophages 7,8. For example, in pancreatic cancer models, it is the yolk
sac‑derived macrophages but not the bone marrow‑derived macrophages that are pro-tumoural, suggesting that origin
is important57. These different origins might be of clinical relevance, as they could allow independent targeting of
subpopulations6. It also raises questions as to whether inhibition of bone marrow‑derived macrophage recruitment might
result in compensation by yolk sac-derived and/or fetal liver‑derived tissue progenitors or vice versa. Such observations
again emphasize the importance of understanding macrophage origin, heterogeneity and dynamics in the tumour
microenvironment.

Yolk sac Fetal liver Bone marrow

EMP
FLT3L

EMP
MDP CDP Dendritic cell
IL-34 CSF1

CSF1
pMACs FL monocyte
IL-34 CSF1

CSF1
Brain Pancreas Spleen Liver Langerhans • Kidney Intestine Patrolling
microglia cells • Lung
(CSF2) Classical Non-classical
• Heart monocyte monocyte

Tissue macrophages Dendritic cell

F4/80high TAM F4/80low

Figure adapted from REF.183, Springer Nature Limited.

Nature Reviews | Drug Discovery

888 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

checkpoint inhibitors (such as monoclonal antibodies
(mAbs) against cytotoxic T lymphocyte antigen 4
(CTLA4), programmed cell death 1 (PD1) and PD1
ligand (PDL1) that potentiate the activity of cytotoxic
CD8+ T cells) have shown success in the treatment of
cancers such as melanoma and lung cancer. However, in
most cases, only a small fraction of patients fully respond
to immunotherapy for unknown reasons9.
In mouse models, TAMs promote the recovery of
tumours from biologic therapies, chemotherapies and
radiotherapies through a mixture of activities that
include promotion of angiogenesis, maintenance of
stem cells and inhibition of immune responses10. In
some tumours, macrophage infiltration also interferes
with efficacy of immunotherapy, neutralizing efforts to
reactivate CD8+ T cells. For this reason, several thera-
peutic strategies to modulate TAM function, infiltration
or activation are emerging to both block resistance to
conventional therapies and promote T cell-based thera-
pies3,10,11. This article discusses the mechanism of action
by which macrophages exert these activities and also the
manner in which they may be targeted therapeutically. It
describes macrophage diversity and how this promotes
tumour progression in primary and metastatic sites.
Finally, it also discusses ways to exploit this diversity
therapeutically to create an antitumoural microenvi-
ronment and to improve chemotherapy, radiotherapy
or immunotherapy.
Macrophages in the TME
Macrophages are innate immune cells that populate
all tissues12 and have multiple roles in development,
homeostasis and tissue repair 13. They have different
embryological origins (BOX 1) and expand not only in
response to infiltration of monocytes but also through
local proliferation of tissue-resident macrophage pro-
genitors. Although macrophages express canonical
markers such as colony-stimulating factor 1 receptor
(CSF1R) in humans and mice, and adhesion G pro-
tein-coupled receptor (ADGRE; also known as F4/80)14
only in mice, studies indicate considerable transcrip-
tomic diversity between macrophage populations even
within a tissue. This diversity is consistent with their
diverse origin, their adaptation to different tissue niches
and their involvement in different pathologies. CSF1R
is a transmembrane tyrosine kinase class III receptor
that is required for the presence of the vast majority
of macrophages15. It binds to two ligands, CSF1 and
interleukin‑34 (IL‑34), and regulates macrophage dif-
ferentiation, proliferation and survival in humans and
mice16. IL‑34 shows overlapping functions with, but
no sequence similarity to, CSF1 and binds to a differ-
ent binding site on CSF1R with a significantly higher
affinity than CSF1 (REF.17). IL‑34 regulates the develop-
ment of a subset of macrophages, particularly micro-
glia and Langerhans cells18. The role of IL‑34 in cancer
progression is still largely unknown, although a recent
study by Baghdadi et al.19 showed that IL‑34 produced
by chemoresistant cancer cells was able to sustain the
immunosuppressive functions of TAMs and promote
the survival of cancer cells.
Although in the late 19th century the scientific
community accepted that tumours were a mixture of
malignant and normal cells, it was not until the 1970s
that the modern concept of the TME and the idea that
immune cells can actually promote cancer growth were
introduced20. However, even at this time, the dominant
view was that macrophages were tumoricidal, as differ-
ent studies showed that activated macrophages could
kill tumour cells in vitro21–23. This view began to change
in the early 1990s when several key studies identified a
potential role of macrophages in tumour progression
by showing correlations of high intra-tumoural density
and increased expression of CSF1 with clinical markers
of poor prognosis (grade, stage and so on)24. In 2001,
the genetic ablation of Csf1, and thereby macrophages,
in the polyoma middle T (PyMT) model of breast can-
cer resulted in the inhibition of tumour progression and
metastasis, formally proving that macrophages can be
pro-tumoural25. After this discovery, several independent
studies in different cancers confirmed the pro-tumoural
role of TAMs and identified several subtypes of TAM in
mice and their roles in promoting primary tumour pro-
gression and metastasis3,26–28. Consistent with these exper-
imental mouse models, a meta-analysis of 55 studies of
different human cancers indicated that high infiltration of
TAMs correlated with poor overall survival in breast, gas-
tric, oral, ovarian, bladder and thyroid cancers but not in
colorectal cancer (CRC)29 (FIG. 1). The correlation between
TAM infiltration and progression in cancer was further
confirmed by additional recent meta-analyses in breast
cancer, gastric cancer, Hodgkin lymphoma and non-
small-cell lung cancer (NSCLC)30–33. More specifically,
overexpression of CSF1 and the urokinase plasminogen
activator (uPA) in CSF1R‑positive ovarian cancer cell lines
improved their invasive properties34. At the clinical level,
high serum levels of CSF1 are found in several cancers
including metastatic breast cancer 35, and they correlate
with poor prognosis in ovarian and endometrial can-
cers36–38. In endometrial cancer, epithelial CSF1 synthesis
is an independent predictor of survival39. In breast can-
cer, a CSF1 expression signature was described that pre-
dicted poor survival40. CSF1R expression combined with
CD68‑positive macrophage infiltration is considered an
independent predictor of short progression-free survival
in Hodgkin lymphoma41. Normal breast and normal ovar-
ian tissues do not express CSF1R and CSF1 (REFS35,42) but,
sometimes, breast and ovarian tumour cells can express
this receptor–ligand pair 43–45. Nevertheless, expression of
CSF1R on cancer cells is rare and is probably due to acti-
vation as a result of demethylation of a recently acquired
retroviral element in humans; thus, it is not found in
mice46. These data over the past 18 years, in many differ-
ent cancer contexts, have supported the conclusion that,
in general, TAMs promote tumour progression to malig-
nancy through their interactions with cancer cells (FIG. 1).
TAMs in solid tumours
Cancer initiation
The activation of key transcriptional factors such as
nuclear factor-κB (NF-κB), signal transducer and acti-
vator of transcription 3 (STAT3) and hypoxia-inducible

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 889

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

a Percentage of survival b Immune infiltrates in tumour

Bladder
Bladder
Type of cancer

Breast
Bowel
Breast
Stomach

Type of cancer
Lung
Bowel

0 20 40 60 80 100
Percentage (%) Stomach
Localized Invasive Metastasized
Lung

B cells CD8+ γδ T Macrophages Mast

c Macrophage density as a predictor of poor survival T cells cells cells
+
Plasma CD4 NK Dendritic
Macrophage density cells T cells cells cells
Patient survival Polymorphonuclear
cells

Figure 1 | Macrophage infiltration and survival in cancer. a | Five-year survival values are high if cancer is restricted to
the primary site (localized), even if it is invasive. However, survival is dramatically reduced if the cancer has metastasized to
Naturethat
distant sites. Data are expressed as 5‑year survival values (expressed in percentages) in tumours Reviews | Drug Discovery
are localized (blue
dots), invasive (grey dots) or metastasized (red dots). b | CIBERSORT analysis of tissue microarray data sets of solid human
tumours revealed the average immune cell composition of bladder, breast, bowel, stomach and lung cancers. Data are
expressed as estimated fractions of leukocyte RNA. As shown, immune infiltrates vary according to cancer type; however,
macrophages represent the major population infiltrating most human cancers. c | Increased human macrophage density
(red wedge) correlates with markers of poor survival (grey wedge) in most cancer types. NK, natural killer. Part a
republished with permission of The Economist, from Targeting tumours, Loder, N., 2017; permission conveyed through
Copyright Clearance Center, Inc. (REF.278). Part b is adapted from REF.279, Springer Nature Limited.

factor 1α (HIF1α) by chronic inflammation (caused by
persistent infection, exposure to irritants or autoimmune
disease) or by oncogene activation results in the pro-
duction of cytokines and chemokines that engage the
innate immune system and especially macrophages47.
Macrophages, in turn, can contribute to cancer initia-
tion by producing pro-inflammatory mediators such
as IL‑6, tumour necrosis factor (TNF) and interferon-γ
(IFNγ), growth factors including epidermal growth factor
(EGF) and WNTs, proteases and reactive oxygen species
(ROS) and nitrogen species that may create a mutagenic
microenvironment 48,49. Thus, chronic inflammation,
infection or irritation is associated with the initiation of
many cancer types such as those in the colon or stom-
ach50. Genetic ablation of Stat3 in macrophages causes
exuberant inflammation and the formation of invasive
tumours in the colon in a mouse model of inflammatory
bowel disease51. Similarly, the deletion of the anti-inflam-
matory cytokine IL‑10, which signals through STAT3, in
macrophages is also associated with tumour initiation and
promotion52,53. By contrast, removal of ROS production in
myeloid cells, including macrophages, inhibits carcino-
genesis in an intestinal cancer model49. These chronic
inflammatory responses are involved with complicated
interactions with the microbiome, changes in dominance
of bacterial species and breakdown of mucosal barriers
that allow infiltration of pathogenic types54. Thus, in ani-
mal models, in some cases, the initiation of inflamma-
tion-induced cancer can be suppressed by prophylactic
treatment with broad-spectrum antibiotics55.
The role of macrophages in the transition from a
benign to a malignant tumour has been studied in only
a few cancer models (mammary, skin and pancreatic
cancers)6,25,56 and, at least in mammary tumours, pre-
mature recruitment of macrophages by overexpression
of CSF1 promotes the transition to malignancy 25. It is
thought that in the first stages of tumour formation mac-
rophages are mainly tumoricidal, as they reflect an acti-
vated state, although the evidence for this conclusion is
limited. What is certain, however, is that, as the tumour
grows, T helper 2 (TH2) cells in the microenvironment
that ‘educate’ macrophages to become pro-tumoural and
bias the immune response from a cytotoxic to a support-
ive role begin to dominate. In mouse models of breast
cancer, this state is driven by IL‑4 secreted from CD4+
T cells57,58 recruited to the tumour through unknown
mechanisms.
The transition to malignancy is also exacerbated by
macrophages through their production of angiogenic
factors, secretion of growth factors and production of
proteases.
Tumour progression requires the creation of a vascu-
lar network for oxygenation, nutrition and waste disposal.
This process, known as the ‘angiogenic switch’, is mainly
characterized by a dramatic increase in new vessel forma-
tion that often involves vessel dilatation and recruitment
of perivascular cells59–61. TAMs support these processes by
the production of pro-angiogenic factors such as vascular
endothelial growth factor A (VEGFA)62 and angiogenic
CXC chemokines (CXCL8 and CXCL12)63. Additional
factors such as WNT7B, transforming growth factor-β
(TGFβ), TNF and thymidine phosphorylase contribute
to the angiogenic process by recruitment and activation
of endothelial cells or other cells such as fibroblasts or
pericytes that further support the generation of vascular
networks in the microenvironment61,64. The identification

890 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

of a subpopulation of TAMs that are characterized by
the expression of the angiopoietin 1 receptor (TIE2; also
known as TEK) and that lie in close association with ves-
sels confirmed the central role of TAMs in tumour angio­
genesis65, as their depletion inhibited tumour growth and
metastasis66.
Highly invasive tumour cells are able to move from
the primary tumour and intravasate into the blood or
lymphatic vessels, where they migrate to distant sites
to establish micro-metastases 28. TAMs are in part
responsible for the creation of this invasive TME. TAMs
directly help tumour cells to migrate via a paracrine loop
between macrophages and tumour cells that involves
secretion of EGF family ligands from macrophages
and CSF1 from tumour cells, which improves tumour
cell invasive properties. TAMs also produce cathepsins
and matrix-remodelling enzymes that stimulate this pro-
cess and increase intravasation by the upregulation of
metalloproteinases such as matrix metalloproteinase 9
(MMP9)67–69. Tumour cells are attracted to vessels, where
they engage with perivascular TIE2+ macrophages that
act as a conduit for the escape of tumour cells in part
through expression of VEGFA, which allows bursts of
vessel permeability 70. This tripartite structure consist-
ing of perivascular TAMs, cancer cells that have height-
ened motility owing to the expression of the invasive
isoform of the actin-binding protein mammalian ena-
bled (MENA) and endothelial cells has been called the
tumour microenvironment for metastasis (TMEM).
Similar structures identified in clinical samples led to the
identification of a prognostic clinical score that predicts
the chance of metastasis in breast cancer regardless of
tumour stage or clinical subtype71.
Transcriptional profiling of invasive TAMs also
revealed upregulation of key components of the WNT
and Hedgehog gene families, which are mainly involved
in tissue patterning and development in homeostatic
conditions72,73. The importance of WNT signalling was
confirmed by the identification of WNT7B as one of
the key factors secreted by TAMs to promote tumour
progression and metastasis, in part through effects on
angiogenesis but also through promotion of tumour cell
invasion61.
TAM-mediated immune suppression
Macrophages can potentially mount a robust anti­
tumoural response, as they are able to directly eliminate
cancer cells if properly activated by IFNγ. They also
can support the adaptive immune response through
presentation of tumour antigens and the production of
chemokines and cytokines to recruit and activate cyto-
toxic CD8+ T cells and natural killer (NK) cells. However,
these macrophage activities are restricted by the TH2 cell
dominance in the TME, which profoundly affects mac-
rophage functions such that their phenotypes resemble
those involved in tissue development and repair, with
a consequent suppression of antitumoural response3,13.
TAM-induced immune suppression is mediated
by the expression of inhibitory receptors, including
non-classical major histocompatibility complex (MHC)
class I (MHC‑I) molecules such as HLA‑E and HLA‑G,
which are able to negatively modulate the activation
of NK cells and T cells by interaction with CD94 and
leukocyte immunoglobulin-like receptor subfamily B
member 1 (LIR1, also know as ILT2), respectively 74.
TAMs express T cell immune checkpoint ligands, such
as PDL1, PDL2, B7‑1 (also known as CD80) and B7‑2
(also known as CD86), which directly inhibit T cell
functions75,76, while also secreting several cytokines,
such as IL‑10 and TGFβ, that contribute to the main-
tenance of a strong immunosuppressive microenvi-
ronment by inhibiting CD4+ (TH1 cells and TH2 cells)
and CD8+ T cells and by inducing regulatory T (Treg)
cell expansion. TAM-mediated release of chemokines
such as CCL2, CCL3, CCL4, CCL5 and CCL20 fur-
ther contributes to the recruitment of Treg cells in the
TME3. TAMs also directly inhibit T cell cytotoxicity
through depletion of l‑arginine (which is essential for
the re‑expression of the T cell receptor (TCR) after anti-
gen engagement on T cells) via the release of arginase 1,
which metabolizes l‑arginine to urea and l‑ornithine3.
Similarly, depletion of tryptophan or production of tryp-
tophan metabolites by indoleamine 2,3‑dioxygenase
(IDO) expressed by macrophages can inhibit cytotoxic
T cells77,78.
TAMs in metastasis
To establish metastasis, invasive cancer cells need to
avoid eradication by the immune system, survive in
the blood or lymphatic circulation, arrest at a distant
site and survive and grow in these often hostile envi-
ronments. In mouse models of breast cancer and CRC,
metastasis-associated macrophages (MAMs) promote
these latter steps by increasing extravasation, sending
survival and growth signals to tumour cells and inhib-
iting anti-cytotoxic T cells5,79,80,81. Importantly, ablation
of these MAMs or blockade of their recruitment results
in inhibition of metastasis and in some cases prolonged
survival of mice5,79,82, suggesting that they represent
thera­peutic targets.
Mechanistically, metastatic cells attract bone marrow-
derived classical monocytes (that express lymphocyte
antigen 6C2 (LY6C)) in mice and CD14highCD16negative
in humans) through a CCL2–CCR2 mechanism5. Once
extravasated, these monocytes (called MAM precursor
cells (MAMPCs)) differentiate into MAMs79 through
engagement of a chemokine cascade that involves
CCR1–CCL3 autocrine signalling 80. MAMs found
in the lungs of mice with metastatic disease are tran-
scriptionally different from resident macrophages and
MAMPCs83 and are characterized by the expression of
the markers CD11b, VEGF receptor 1 (VEGFR1; also
known as FLT1), CXCR3 and CCR2 (REFS79,80). Moreover,
lung intravital and ex vivo experiments showed that
macrophages in the lung interact physically with meta­
static cells84 and support their survival through a vas-
cular cell adhesion protein 1 (VCAM1)-dependent
and AKT-dependent mechanism as well as by induc-
ing epithelial to mesenchymal transition by producing
TGFβ85,86. Moreover, MAMs and metastatic cancer cells
engage in a crosstalk that improves MAM retention
in the metastatic foci, which further supports tumour

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 891

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

• VEGFR1+
• CCR2+
Metastasis • CXCR4–
• TIE2–
• Arginase+
• MARCO+
TAM • IL-10+
• Wnt signaling • CCL22+ Immune
Invasion • CTSS+ • MRC1+ regulation
and/or • CTSB+ • MHC-IIhigh
intravasation • EGF+ • CD11b+
• CSF1R+
• F4/80+ • IL-12
• MHC-IIhigh Inflammation
• iNOS+
• CXCR4 + • TNF+
Resistance • B7.1 and/or
to therapy • MRC1+
• TIE2+ B7.2

• VEGFR1+
• VEGF+
• TIE2+ Angiogenesis
• CXCR4+
Stem cell • CTSB+
maintenance WNT proteins • CTSS+

Figure 2 | Macrophage diversity drives tumour progression to metastasis and resistance to therapy.
Tumour-associated macrophages (TAMs) that express canonical macrophage markers in the Nature Reviews
mouse (CD11b,| Drug Discovery
colony-stimulating factor 1 receptor (CSF1R) and adhesion G protein-coupled receptor (ADGRE; also known as F4/80))
can be polarized to adopt different pro-tumoural functions depending on the environment, as indicated. These activities
promote tumour initiation through inflammation, tumour progression to malignancy via increasing angiogenesis,
immunosuppression, invasion, intravasation and at distant sites tumour cell extravasation and persistent growth. Each of
these functions is performed by a subpopulation of macrophages with a different transcriptome and cell surface markers,
as indicated. EGF, epidermal growth factor; MARCO, macrophage receptor with collagenous structure; MHC-II, major
histocompatibility complex class II; TIE2, angiopoietin 1 receptor; TNF, tumour necrosis factor; VEGF, vascular endothelial
growth factor; VEGFR1, vascular endothelial growth factor receptor 1.

growth through expression of unidentified survival and TAMs in conventional therapies

growth factors80. MAMs and their precursors inhibit
cytotoxic T cells, suggesting that they are also involved in
the maintenance of an immunosuppressive environment
that further promotes metastasis83. Similar mechanisms
have been observed in bone metastasis, in which another
class of macrophages — the bone-degrading osteoclasts
— are activated by metastatic cells to engage a vicious
cycle because the bone resorption liberates entrapped
growth factors that further promote metastatic growth87.
There are currently limited data available on tumoural
infiltration in human metastasis, although MAMs that
express high levels of VEGFR1 have been identified in
lymph node metastases82.
Targeting TAMs in cancer therapy
The subpopulations of macrophages with identifia-
ble markers, transcriptomes and phenotypes (FIG. 2)
described above are therefore attractive therapeutic
targets for combination therapies including standard of
care and immunotherapy. In mouse models and clin-
ical contexts, TAMs can synergize with the anticancer
therapy or, alternatively, induce pro-tumoural functions
by the activation of tissue repair mechanisms. Here, we
describe the latest studies reporting dichotomous behav-
iours of TAMs after standard therapy, and strategies to
improve anti-macrophage therapeutics (FIGS. 3,4).
Radiotherapy. Ionizing radiation is designed to specifically
target cancer cells by inducing DNA damage in cells that
usually have impaired DNA repair mechanisms. However,
the effect of radiotherapy on the TME, and especially on
TAMs, is only partially understood88. After ablative radio-
therapy (single dose of 10 Gy), the innate immune system
is activated by inflammatory cytokines, such as IL‑1 and
TNF, and pro-fibrotic factors such as TGFβ that recruit
macrophages with a tissue-reparative phenotype and con-
tribute to tumour recurrence and progression89–93. A recent
study by Pinto et al. showed that fractionated cumulative
radiation dose regimens similar to those used during can-
cer treatment induced a pro-inflammatory phenotype
in macrophages in vitro but did not alter their ability to
promote cancer invasion and cancer angiogenesis94. It is
becoming a priority to tailor dose and fractionation of
radiotherapy in different cancers, as the response elicited
by macrophages and the stroma in general can influence
the outcome95. Consequently, macrophage targeting in
combination with radiotherapy is a potential therapeutic
strategy to modulate the stroma and allow better tumour
killing but, as yet, there are no clinical trials reported.
Chemotherapy. The TME and, in particular, macrophages
play an important part in chemotherapy response and
resistance96–98. The first observation that suggested a

892 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

role of TAMs in mediating resistance to chemotherapy
was the demonstration that CSF1 inhibition was able
to reverse chemoresistance of breast cancer cell lines in
xenograft mouse models99. DeNardo100 et al. confirmed
and extended this initial observation and showed that
tumour biopsy samples from cancer patients who
received neoadjuvant therapy had a much larger infil-
trate of CD45+CD11b+CD14+ macrophages than those
from patients who received only surgery. They also
demonstrated that treatment of PyMT mice with pacl-
itaxel in combination with anti‑CSF1R antibodies signif-
icantly reduced tumour burden and vessel density and
increased cytotoxic T cell infiltration compared with
treatment with paclitaxel alone.
TAMs are responsible for the increased produc-
tion of cathepsins that leads to increased lymphangio-
genesis and metastasis after paclitaxel treatment 101,102.
Upon 5‑fluorouracil treatment of CRC, TAMs release
the diamine putrescine, which confers on cancer cells
resistance to apoptosis in a JNK-dependent and caspase
3‑dependent manner 103. Doxorubicin, another chemo-
therapeutic agent, can also induce the selective accumu-
lation of perivascular MRC1+ TAMs around the blood
vessels. These perivascular TAMs promoted recurrence
after therapy partly through the expression of VEGF and
subsequent increased angiogenesis. Selective targeting of
the recruitment of this perivascular TAM population by
CXCR4 blockade showed reduced tumour relapse after
chemotherapy, suggesting that this targeting strategy is
of clinical utility 63.
Immunotherapy. Mutated tumour cells can express
tumour antigens as a product of oncogenic viruses,
differentiation antigens or tumour-specific mutations
(tumour neo-antigens). The production of neo-antigens
is not equal across tumour types, with tumours such
as melanoma and lung cancers expressing the highest
rate of neo-antigens, and haematological malignancies
(acute myeloid leukaemia (AML), acute lymphoblas-
tic leukaemia and chronic myelogenous leukaemia)
expressing the lowest 104. T cells are able to recognize
tumour antigens loaded to the MHC on the cancer
cell through binding to the TCR; however, to get com-
pletely activated, they require interaction of CD28
with co‑stimulatory B7 molecules (CD80 and CD86)
expressed on the antigen-presenting cell (APC). Cancer
cells do not express B7 molecules, and so, without a sec-
ond stimulatory signal provided by other APCs (such as
dendritic cells and macrophages) recruited by inflam-
matory signals, the antitumoural T cell response will
not start.
The identification of tumour antigens led to the
development of several tumour vaccination strategies in
the 1980s, in which tumour-derived antigens (DNA or
peptides) were injected together with cytokines in order
to enhance the immunological response. Results from
these trials, however, were not as striking as expected105,
suggesting that the regulation of T cell activation in the
tumour is complex. Both preclinical and clinical stud-
ies indicated that tumours are infiltrated by immuno-
suppressive cells (Treg cells, TAMs, cancer fibroblasts
and myeloid-derived suppressor cells) and that they
are exposed to an immunosuppressive cytokine milieu
(IL‑10 and TGFβ).
With this increasing understanding, the immuno-
therapy field has had a rapid expansion particularly with
the discovery of immune checkpoint inhibitors, such as
mAbs against CTLA4 and PD1, whose main function is
to remove the ‘brakes’ on T cells, allowing an effective
antitumoural immune response106. One anti‑CTLA4
(ipilimumab) and two anti‑PD1 (pembrolizumab and
nivolumab) neutralizing antibodies are currently US
Food and Drug Administration-approved for the treat-
ment of several cancers including melanoma, advanced
renal carcinoma, gastric cancer, NSCLC and CRC;
hundreds of clinical trials on multiple solid cancers are
ongoing to evaluate their efficacy. Although efficacy in
melanoma and some other cancers is high, especially
when anti‑CTLA4 and anti‑PD1 therapies are combined,
in many other cancers, only a small fraction of treated
patients respond. One hypothesis is that the efficacy of
checkpoint inhibitors could be improved by the modu-
lation of immunosuppressive cells such as TAMs. TAMs
express high levels of PDL1 and PDL2, as well as PD1.
PD1 expression in TAMs also negatively correlates with
the ability of these cells to phagocytose cancer cells, and
therefore, TAM-specific PD1 inhibition reduces tumour
growth107.
Moreover, CSF1 expression correlates with accumula-
tion of CD8+ T cells and CD163+ TAMs in melanoma, and
anti‑PD1 and anti‑CSF1R combination therapy induced
regression of melanoma in preclinical studies108. Clinical
trials combining checkpoint inhibitors and anti-TAM
agents (such anti‑CSF1R antibodies; see below) are cur-
rently ongoing in different solid tumour contexts109–119.
Methods of TAM targeting
TAM depletion. The dependence of macrophages on
CSF1R signalling makes this an attractive target to selec-
tively deplete macrophages. Consequently, different anti-
bodies and small molecules mainly targeting CSF1R are
being studied in different clinical trials both as mono­
therapies or in combination with standard therapy or
immunotherapies.
The small molecules under clinical development and
study are PLX3397, JNJ‑40346527, PLX7486, ARRY‑382
and BLZ945. Preclinical studies showed that PLX3397, or
pexidartinib, reduced the number of tumour-associated
microglia and glioblastoma invasion120,121; in glioma
mouse models, the tyrosine kinase inhibitors dovitinib
and vatalanib increased sensitivity to PLX3397. PLX3397
was tested in a phase I dose-escalation clinical trial fol-
lowed by a phase II extension study in patients with
advanced tenosynovial giant cell tumours122, a tumour
type characterized by the high expression of CSF1 and
CSF1R123,124. Results from the phase I and phase II stud-
ies indicated that PLX3397 was tolerated at a dose of
1,000 mg, and in the extension study, 12 out of 23 patients
(52%) showed antitumour responses after treatment122.
In a phase II study in patients with recurrent glioblas-
toma, PLX3397 treatment was tolerated, and PLX3397 was
able to pass the blood–tumour barrier; however, PLX3397

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 893

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

1 • Metastasis 2
Trabectedin Phagocytosis
• Dissemination
Blood 6 • Angiogenesis
vessel Monocyte
recruitment Anti-VEGFR SIRP1α inhibitors

• Anti-CCR2 TRAIL • Anti-MUC1 Anti-CD47

Monocyte VEGF
• Anti-CSF1R • Anti-EGFR
SIRPα
Bisphosphonates
CD47
CCR2
Caspase 8
CSF1R
DICER
Survival Apoptosis DICER
CCL2 inhibitor
Tumour 5ʹ
cells Anti-CCL2 3ʹ
Pre-miRNA

CSF1 TLR8 TLR7

Anti-CSF1 TLR9

3 TLR agonists
IL-10 Checkpoint and
B7-1 arginase inhibitors
and/or
B7-2 PDL1
and/or FcγRIIB MARCO
Dendritic cell TGFβ PDL2
PI3Kγ CD40

CD28 HDAC
PD1 inhibitors
CD40
T cell or agonists
NK cell Arginase PI3Kγ
inhibitors Anti-MARCO

Treg cell
4
5
Immune suppression Reprogramming

Pro-tumour macrophage Anti-tumour macrophage

Figure 3 | Targeting and reprogramming TAM pro-tumoural activities.
The figure shows pro-tumoural tumour-associated macrophage (TAM)
activities (angiogenesis, immune escape, dissemination, and so on)
that have been targeted in either preclinical models or therapeutic trials
in humans. The strategies fall into six main groups: (1) blocking
pro-tumoural functions such as those that promote intravasation,
angiogenesis or metastatic cell extravasation and/or persistent growth;
(2) promoting phagocytosis of tumour cells by macrophages; (3) using
checkpoint or anti-immunosuppressive cytokine and/or protein
inhibitors to allow cytotoxic T cell activity; (4) reprogramming TAMs to
become antitumour macrophages either directly through tumour cell
killing or by T cell activation; (5) inhibiting the tumour suppressive
microenvironment; and (6) inhibitingNature Reviews
recruitment Drugmonocytic
of| the Discovery
progenitors of TAMs; CSF1, colony-stimulating factor 1; CSF1R, CSF1
receptor; EGFR, epidermal growth factor receptor; HDAC, histone
deacetylase; MARCO, macrophage receptor with collagenous structure;
miRNA, microRNA; NK, natural killer; PD1, programmed cell death 1;
PDL1, PD1 ligand 1; PI3Kγ, phosphoinositide 3‑kinase-γ; SIRPα, signal
regulatory protein-α; TGFβ, transforming growth factor-β; TLR, Toll-like
receptor; TRAIL, tumour necrosis factor-related apoptosis-inducing
ligand; Treg cell, regulatory T cell; VEGF, vascular endothelial growth
factor; VEGFR, VEGF receptor.

treatment did not show improvement in 6‑month pro-
gression-free survival values when compared with those
of the group receiving radiotherapy and temozolomide125.
JNJ‑40346527, another kinase inhibitor, was tested in a
phase I/II study on 21 patients with relapsed or refrac-
tory Hodgkin lymphoma. One patient showed complete
remission, and eleven patients showed stable disease126.
Several phase I clinical trials are ongoing with additional
CSF1R inhibitors: PLX7486 is being tested as a single
agent in patients with advanced-stage solid tumours (ten-
osynovial giant cell tumour and tumours of any histology
with activating mutations in brain-derived neurotrophic
factor (BDNF))127, and ARRY‑382 is currently involved in
two phase I clinical trials128,111 in patients with metastatic
disease and advanced-stage solid tumours. BLZ945 was
reported to alter macrophage polarization and to block
glioma progression129 with promising results when used
in combination with inhibitors of insulin-like growth
factor 1 receptor (IGF1R) and phosphoinositide 3‑kinase
(PI3K)130 and is currently under assessment in a trial in
advanced-stage solid tumours as a single agent or in com-
bination with the anti‑PD1 antibody PDR001 (REF.112).

894 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

PEGylated nanoparticles
A type of therapeutic delivery
system that results from
PEGylation (a chemical
modification that involves the
conjugation of polyethylene
glycol (PEG) to different
molecules) of nanoparticles.
There are three anti‑CSF1R mAbs under clinical eval-
uation, RG7155, IMC‑CS4 and FPA008. RG7155 (emac-
tuzumab) is a humanized mAb that binds to CSF1R and
blocks its dimerization. Preclinical studies showed that
RG7155 depletes CSF1R+CD163+ macrophages in vitro
and in vivo. In mouse models of CRC and fibrosarcoma,
treatment with a chimeric version of RG7155 reduced
the number of infiltrating TAMs and increased the
CD8+:CD4+ T cell ratio124; moreover, RG7155 showed sig-
nificant reduction of CSF1R+CD163+ macrophages and
T cell TME composition in patients with advanced-stage
diffuse-type giant cell tumours treated with RG7155 as
monotherapy or in combination with paclitaxel124. These
promising results led to a dose-escalation phase I clinical
trial in 12 patients with tenosynovial giant cell tumours to
investigate the clinical benefit of RG7155; the compound
showed no dose toxicity, and common adverse effects
reported were facial oedema, asthenia and pruritus. In
the dose-expansion phase, out of the 28 patients tested,
24 patients (86%) showed an objective response, and
2 (7%) achieved a complete response131. There are two
other antibodies currently under phase I clinical trials:
FPA008 — currently under clinical investigation in three
ongoing clinical trials116,132,133 in diffuse-type tenosynovial
giant cell tumour and advanced-stage solid tumours —
and IMC‑CS4, which is also under clinical investigation
in three clinical trials134–136 on solid tumours including
pancreatic, prostate and breast cancers.
Overall, these preliminary results suggest that target-
ing the CSF1–CSF1R axis could be a promising strategy.
Indeed, in tumours overexpressing CSF1, such as the
synovial giant cell tumours, it is likely that this approach
will be advanced as an effective treatment. Nevertheless,
reported toxicity has limited dose escalation, as it is
problematic to deplete all macrophages from the body
for a long period of time.
Another therapeutic strategy is to selectively deplete
TAMs and not the other components of the stroma. An
example of this strategy is the use of bisphosphonates.
These inorganic compounds are stable, and their struc-
ture is identical to pyrophosphatases of the bone matrix,
so they can be metabolized rapidly by osteoclasts and
inhibit their resorption. Moreover, they are already in
use as anticancer agents for the treatment of haemato-
logical and solid malignancies137,138. Bisphosphonates
are mainly subdivided into two classes on the basis of
their structure and mechanism of action: clodronate,
etidronate and tiludronate belong to the first group, and
alendronate, ibandronate, pamidronate, risedronate and
zolenodrate belong to the second group. At the preclin-
ical level, bisphosphonates exhibited direct and indirect
antitumour properties. They are reported to inhibit can-
cer cell proliferation, induce tumour cell apoptosis, block
angiogenesis, inhibit cell adhesion and invasion and
interfere with immune surveillance through activation
of γδ T cells139. Different studies showed that bisphos-
phonates are also able to inhibit proliferation, migration
and invasion of macrophages, causing apoptosis140,141.
They, however, mainly affect osteoclasts that share the
same lineage with macrophages and thus have been used
in preclinical bone metastasis models.
A more directed approach when using bisphospho-
nates is to encapsulate clodronate, for example, in lipo­
somes (clodrolip), which are preferentially taken up by
macrophages owing to their phagocytic activities. This
treatment reduced macrophage tumour infiltration
in experimental models of bone metastasis from lung
cancer 142 and lung metastasis from breast cancer 79 and
thereby limited metastatic outgrowth. Similar results
were obtained in mice injected with human melanoma
cells, in which clodrolip reduced tumour mass and angio­
genesis143. Clodrolip in combination with anti-VEGF
mAbs showed antitumour properties in mice injected
with teratocarcinoma and rhabdomyosarcoma cells, with
a significant reduction of TAM infiltration144. Moreover,
in mouse models of metastatic hepatocellular carcinoma,
the combination treatment with clodrolip and sorafenib
caused decreased tumour burden, angiogenesis and
metastasis145. Clodrolip was also tested in dogs with
spontaneous soft tissue sarcomas, in which it depleted
CD11b+ macrophages in tumours and decreased IL‑8
serum levels, although the antitumour properties of
the treatment were not significant 146. Zoledronic acid,
another bisphosphonate, reduced tumour burden in a
mouse model of bone metastases from breast cancer 147
and to modulate the TME by reducing the number of
TAMs and their polarization status148. Recently, Comito
et al.149 demonstrated that treatment with zoledronic
acid impairs macrophage polarization, reduces macro­
phage-induced angiogenesis and decreases tumour
invasion in prostate cancer. Current studies using
nanotechnology are trying to optimize the delivery of
bisphosphonates by their encapsulation in stealth lipo­
somes or in PEGylated nanoparticles; preclinical tests were
promising, showing better antitumoural activity and
lower TAM numbers than those seen with treatment
with free bisphosphonates150–152.
At the clinical level, clodronate and zoledronic acid
were tested in several clinical trials on a variety of dif-
ferent cancers with inconsistent results that suggest the
need to optimize better combination treatments and
for longer clinical trials, as discussed in several meta-
analyses153,139,154. There are currently two ongoing clin-
ical trials evaluating the effect of zoledronic acid in
triple-negative breast cancer 155 and stage IIIb and IV
lung cancer 156. Clodronate is also being tested in differ-
ent clinical trials as a neoadjuvant agent in patients with
breast cancer 157 and in combination with chemother-
apy and hormonal therapy 158; additional trials include
a clodronate–chemotherapy combination treatment in
patients with metastatic refractory prostate cancer 159.
Trabectedin is a tetrahydroisoquinoline alkaloid
that was initially isolated from the Caribbean tuni-
cate Ecteinascidia turbinata160. It is an anti-neoplastic
drug approved in Europe, Russia and South Korea for
the treatment of advanced-stage tissue sarcoma and plat-
inum-sensitive relapsed ovarian cancer in combination
with PEGylated liposomal doxorubicin161–163. Trabectedin,
in addition to targeting tumour cells, specifically induces
apoptosis of monocytes and macrophages in the tumour
by the activation of caspase 8 through a TNF-related
apoptosis-inducing ligand (TRAIL; also known as

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 895

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Recruitment Survival Reprogramming

Monocyte Cancer cell TAM TAM Apoptosis Pro-tumoural Tumoricidal

macrophage macrophage
Drug name Target Inhibitor type Drug name Target Inhibitor type Drug name Target Inhibitor type
Carlumab CCL2 mAb Clodronate NA Small molecule Hu5F9-G4 CD47 mAb
PF04136309 CCR2 Small molecule Zoledronic acid NA Small molecule CC-90002 CD47 mAb
PLX3397 CSF1R Small molecule Trabectedin Caspase 8 Small molecule TTI-621 SIRPα Fusion protein
PLX7486 CSF1R Small molecule PLX7486 CSF1R Small molecule CP-870,893 CD40 Agonistic antibody
JNJ-40346527 CSF1R Small molecule JNJ-40346527 CSF1R Small molecule RO7009789 CD40 Agonistic antibody
ARRY-382 CSF1R Small molecule ARRY-382 CSF1R Small molecule Imiquimod TLR7 Small molecule
BLZ945 CSF1R Small molecule BLZ945 CSF1R Small molecule 852A TLR7 Small molecule
IMC-CS4 CSF1R mAb IMC-CS4 CSF1R mAb IMO-2055 TLR9 Small molecule
R05509554 CSF1R mAb R05509554 CSF1R mAb BLZ945 CSF1R Small molecule
RG7155 CSF1R mAb RG7155 CSF1R mAb
FPA008 CSF1R mAb FPA008 CSF1R mAb

Figure 4 | Selective examples of anti-TAM drugs currently under clinical trial investigation. For each of the strategies
targeting tumour-associated macrophages (TAMs) (recruitment, survival and reprogramming),
Nature there are drugs
Reviews currently
| Drug Discovery
being tested in clinical trials as monotherapies or in combination with chemotherapy and immunotherapy. In many cases,
one anti-TAM drug can affect more than one process, such as colony-stimulating factor 1 receptor (CSF1R) inhibitors that
inhibit recruitment, survival and function. mAb, monoclonal antibody; NA, not applicable; TLR, Toll-like receptor.

TNFSF10)-dependent mechanism164. These results sug-
gested that the apoptotic receptor family TRAIL could
be a therapeutic target to selectively kill immune cells
and, especially, macrophages. A recent report by Liguori
et al.165 explored this hypothesis and demonstrated that
monocytes and macrophages express the functional
TRAIL receptors TRAILR1 (also known as TNFSF10A)
and TRAILR2 (also known as TNFRSF10B), whereas
neutrophils and lymphocytes express the non-functional
decoy receptor TRAILR3 (also known as TNFRSF10C).
Interestingly, human TAMs in mammary, hepatic and
colon carcinoma, but not resident tissue macrophages,
express functional TRAILRs165, making these receptors
interesting targets for therapy 166.
Inhibition of TAM recruitment. TAM expansion in
the tumour is often mediated by monocytic recruit-
ment through the CCL2–CCR2 axis. CCL2 is a potent
chemo­attractant for monocytes, T cells and NK cells167,
and several mouse studies have demonstrated a role for
it and other chemokines in macrophage accumulation
in the tumours168–171. CCL2 released by tumour cells
recruits classical monocytes that express the receptor
CCR2 to the tumour sites, and inhibition of CCL2 cor-
relates with reduced tumour burden and metastasis in
different experimental models of prostate, breast, lung
and liver cancers and melanoma172. However, withdrawal
of anti‑CCL2 treatment accelerated lung metastasis in
mouse models of breast cancer and resulted in death
of mice owing to a rebound in monocyte recruitment,
which raised important concerns about the long-term
efficacy of this approach173,174. However, high levels of
CCL2 in serum, as well in the tumour, are associated
with poor prognosis in different types of tumours,
such as breast cancer 175,176. For these reasons, several
CCL2‑neutralizing antibodies are now being tested
in clinical trials. The two main drugs currently tested
are carlumab (CNTO 888), an anti‑CCL2 mAb, and
PF‑04136309, a small molecule inhibitor that targets
CCR2.
Carlumab is a human immunoglobin G1κ antibody
that binds to CCL2. In prostate cancer mouse models,
systemic injection of the antibody reduced tumour
growth, infiltration of CD68+ macrophages and vascular
density 177,178. CCL2 inhibition was also able to improve
the effect of paclitaxel and carboplatin therapies in
mouse models of ovarian cancer 179. A phase I clinical
trial was carried out in 2013 to assess tolerance to differ-
ent doses of carlumab in 44 patients with different solid
tumours180. The results indicated that CCL2 levels were
only partially suppressed by carlumab, with an increase of
free CCL2 after treatment over the pretreatment baseline
of more than 1,000‑fold173. A phase II clinical trial was
then carried out in 46 patients with castration-resistant
metastatic prostate cancer, with no reported therapeutic
efficacy of carlumab181.
The CCR2 small molecule inhibitor PF‑04136309
was recently assessed in a phase Ib non-randomized
trial in patients with locally advanced pancreatic can-
cer in combination with FOLFIRINOX chemotherapy
(oxaliplatin and irinotecan plus leucovorin and fluoro-
uracil); 47 patients were treated and 8 patients received
only FOLFIRINOX, while the remaining patients
received FOLFIRINOX and PF‑04136309. Results

896 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

showed that PF‑04136309 treatment in combination
with FOLFIRINOX was safe and well tolerated com-
pared with chemotherapy alone. Patients treated only
with FOLFIRINOX did not show an objective response
to the treatment. By contrast, in the PF‑04136309 plus
FOLFIRINOX group, 16 of the 33 patients, who under-
went repeated imaging assessments, had an objective
tumour response, and 32 of those patients achieved local
tumour control182.
These trials, however, were largely disappointing
and indicated that much more biological knowledge
is needed to achieve effective inhibition of monocyte
recruitment. For example, CCL2–CCR2 signalling is
required for monocyte egress from the bone into the
blood, and CCL2 inhibition results in a severe deple-
tion of monocytes. This depletion is recognized, and
the organism attempts to overcome this deficiency with
a dramatic elevation in CCL2 concentration, which in
turn prevents efficacy of the reagents. In our own stud-
ies using genetic ablation, we have demonstrated the
requirement for CCR2 to recruit monocytes to the pri-
mary tumour in the PyMT model but that in the absence
of CCR2 in monocytes this recruitment is completely
overcome by unknown redundancy mechanisms4. In
addition, there may be compensatory proliferation of
tissue-resident macrophages if recruitment is blocked6.
Therefore, much more biological understanding is
required before this approach to selectively inhibit
monocyte recruitment to the tumour or its metastatic
derivatives is likely to be effective. Indeed, it may be
better to explore inhibition of molecules required for
retention of monocytes, such as CCL3, or molecules that
induce their differentiation4,80.
Reprogramming of TAMs
Despite generally being pro-tumoural, TAMs can,
depending on context, be tumoricidal and also suppress
tumour growth by activating immune responses183.
This suggests that macrophage plasticity can be thera­
peutically exploited to restore antitumour properties
to TAMs184. Indeed, it might indicate the paucity of the
approach of targeting all macrophages, as both pro-
tumoural and antitumoural macrophages will be
depleted. Instead, macrophage reprogramming is a tar-
geting strategy that provides an opportunity to rebalance
the microenvironment immune infiltrate therapeutically
from a pro-tumoural one to one that actively rejects the
tumour in synergy with T cell-enhancing drugs such as
checkpoint inhibitors. It also eliminates the drawbacks
and long-term toxicity of ablation of all macrophages,
such as that seen in the case of anti‑CSF1R therapeu-
tics discussed above. Different methods are currently
being tested at the preclinical and clinical levels, as
discussed below.
Anti‑CD47 antibodies. CD47 is a ubiquitous protein
that regulates cell migration, axon extension, cytokine
production and T cell activation185–188. CD47 interacts
with thrombospondin 1 and signal regulatory protein-α
(SIRPα; also known as SHPS1), which are mainly
expressed by myeloid cells, including dendritic cells and
macrophages189. In the latter case, CD47 inhibits phago-
cytosis by the prevention of myosin IIA accumulation at
the phagocytic synapse190. The result of this interaction
is a ‘do not eat me’ signal that prevents phagocytosis of
autologous cells in homeostatic conditions. This mecha­
nism is tightly regulated, and it is mainly activated in
pro-inflammatory conditions, as CD47‑null mutant
mice show a phenotype against self only during inflam-
matory conditions191. CD47 is overexpressed in a variety
of tumours192–195 through the activation of CD47‑specific
super-enhancers196. It is involved in tumour invasion,
metastasis and, more importantly, inhibition of phago-
cytosis by the innate immune system by interacting with
SIRPα195 expressed on phagocytes.
Several preclinical studies in xenograft mouse models
demonstrated that CD47 inhibition is an effective strat-
egy for tumour therapy 197–199, as it enables killing and
phagocytosis of tumour cells by macrophages. The role
of CD47 in the inhibition of phagocyte-mediated killing
was confirmed recently in human small-cell lung and
ovarian cancer cell lines200,201. Highly phagocytic bone
marrow-derived macrophages, in which SIRPα had been
inhibited by injection of antitumour antibodies against
MUC1 and EGF receptor (EGFR) (cetuximab), could
effectively reach the tumour and engulf human lung car-
cinoma A549 cancer cells, causing tumour regression.
However, the antitumoural effect observed was then lost
owing to their differentiation to TAMs202.
At the moment, there are two anti‑CD47 mAbs
(Hu5F9‑G4 and CC‑90002) and one soluble recombi-
nant SIRPα–crystallizable fragment (Fc) fusion protein
(TTI‑621) currently being tested in phase I clinical trials.
Hu5F9‑G4 showed promising results at the preclinical
level in human AML203 and in paediatric brain tumours204.
There are currently four ongoing clinical trials on different
solid and haematological malignancies205–208 to study the
safety of Hu5F9‑G4 in patients. Preliminary results from
the NCT02216409 trial indicated that Hu5F9‑G4 was tol-
erated with reversible side effects observed such as anae-
mia, headache, nausea and retinal toxicity 209. Two trials
assessing CC‑90002210,211 have been initiated in patients
with haematological malignancies, but results have not
been published yet.
TTI‑621 is a fully human recombinant protein that
blocks the CD47–SIRPα axis and improves killing of
cancer cells. A recent study investigated the efficacy of
TTI‑621 in aggressive AML and B cell lymphoma xeno­
grafts; TTI‑621 successfully improved macrophage-
mediated phagocytosis of cancer cells but not normal
cells. Moreover, in vivo data suggested that TTI‑621
treatment was able to control the growth of haemato-
logical and solid tumours in mouse xenografts models212.
TTI‑621 is currently being tested at the clinical level in
two ongoing clinical trials on haematological and mul-
tiple solid tumours213,214.
Toll-like receptor agonist. Toll-like receptors (TLRs)
are innate immunity pattern recognition receptors that
have fundamental roles in the activation of the innate
immune response215. Activation of TLR by bacterial
particles (such as lipopolysaccharide) and viral nucleic

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 897

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

acids (RNA or DNA) polarizes macrophages towards a
pro-inflammatory phenotype. For this reason, different
TLR synthetic ligands have been tested in cancer models
in order to assess their efficacy in the phenotypic switch
of TAMs to tumoricidal macrophages in the TME216.
In an orthotropic mammary tumour mouse model,
intra-tumoural delivery of TLR7 and TLR9 agonists
caused increased monocyte infiltration in the tumour
and macrophage repolarization217; similar results were
obtained with an agonist of both TLR7 and TLR8
(3M‑052) that induced macrophage repolarization and
improved tumoricidal activity in melanoma218. In pre-
clinical models, the TLR7 ligand imiquimod, the only
TLR agonist approved for clinical use, showed anti­
tumoural activity in basal cell carcinoma, melanoma
and breast cancer skin metastases219–222.
Two TLR7 ligands (imiquimod and 852A) and one
TLR9 ligand (IMO‑2055) are being tested for their
antitumoural properties in clinical trials. Imiquimod
has been tested on several cancers: in a prospective
clinical trial, topical treatment of skin metastases from
breast cancers with imiquimod was well tolerated, and
responders showed histological tumour regression and
increased lymphoid immune infiltration221. 852A has
been tested in five clinical trials in melanoma, leukae-
mia and gynaecological cancers223–227. A phase I clinical
trial on patients with advanced-stage cancer showed
that treatment with 852A three times per week for
2 weeks was well tolerated, with reversible side effects228.
IMO‑2055 has been tested in CRCs and head and neck,
lung and renal cancers229–233. Results from a clinical trial
in patients with advanced metastatic NSCLC showed
good tolerability and potential antitumoural activity of
IMO‑2055 when used in combination with erlotinib and
bevacizumab234.
Anti‑CD40 antibodies. CD40 is a receptor that belongs
to the TNF receptor superfamily, and it is expressed by
APCs such as monocytes, macrophages, dendritic cells
and B cells235, but it can be expressed by endothelial
and epithelial cells as well. The natural ligand of CD40
is CD40L, which is mainly expressed by CD4+ T cells,
basophils and mast cells236. The CD40–CD40L inter­
action upregulates the expression of MHC molecules
and the production of pro-inflammatory cytokines, such
as IL‑12, which primes naive CD4+ and CD8+ T cells
into T helper and cytotoxic cells, respectively. Agonistic
anti‑CD40 antibodies exert tumour inhibitory effects in
several tumour mouse models, an observation that has
opened the way for the development of clinically relevant
anti‑CD40 antibodies.
Interestingly, TAM treatment with CD40 agonists
in combination with anti‑CSF1R antibodies results in a
profound TAM reprogramming before their depletion;
these reprogrammed TAMs create a pro-inflammatory
environment that elicits effective T cell responses even
in tumours that were non-responsive to immune check-
points inhibitors237,238.
Two agonistic anti‑CD40 antibodies are being
tested in clinical trials: CP‑870,893 and RO7009789. In
a phase I dose-escalation study, 29 patients with solid
tumours were treated with a single intravenous dose of
CP‑870,893; common adverse effects included cytokine
release syndrome and alterations of immune cell num-
ber but, in general, the treatment was well tolerated, and
CP‑870,893 led to an objective response and antitumour
activity 239. CP‑870,893 treatment in combination with
carboplatin and paclitaxel was tested in 32 patients with
advanced-stage solid tumours in another phase I trial; 6
out of 30 evaluable patients showed partial response to
treatment and depletion of B cells together with upreg-
ulation of immune co‑stimulatory molecules240. Similar
results were obtained in patients with malignant pleural
mesothelioma treated with CP‑870,893 in combination
with cisplatin and pemetrexed241 and in patients with
advanced-stage pancreatic ductal adenocarcinoma242.
RO7009789 is being studied in four ongoing clinical
trials on advanced-stage solid tumours114,243–245.
Histone deacetylase inhibitors. Histone deacetylases
(HDACs), of which there are 18 in mammals, are
divided into four classes246,247. HDACs are responsible
for removing the acetyl groups on histones, a crucial
process in epigenetic regulation of gene expression.
TMP195, a specific inhibitor of class IIA HDACs248,
can modify the epigenomic profile of monocytes and
macrophages, resulting in, for example, altered CCL1
and CCL2 expression in monocytes and a pro-inflam-
matory phenotype. In a model of luminal B‑type breast
cancer, intraperitoneal injection of TMP195 increased
infiltration of CD11b+ myeloid cells from blood into the
tumour, where they differentiated into antitumoural
macrophages249. This resulted in reduced vessel permea­
bility as well as in reduced vasculature and tumour cell
proliferation. Moreover, the antitumour macrophage
phenotype induced by TMP195 treatment increased the
efficacy and durability of both standard chemotherapeu-
tic regimens (carboplatin and paclitaxel) and immuno-
therapy (anti‑PD1 antibodies). These findings suggest
that class IIA HDAC inhibitors can selectively repro-
gramme monocytes and macrophages in the tumour and
open interesting therapeutic opportunities, although it
remains to be seen whether targeting this HDAC sub-
class will be sufficiently specific if given systemically
to patients.
Anti-MARCO antibody therapy. The macrophage
receptor with collagenous structure (MARCO) is a
pattern recognition receptor that belongs to the class A
scavenger receptor family. MARCO is mainly expressed
by macrophages250, and its expression was linked to
poor prognosis in breast cancer 251. A recent report by
Georgoudaki et al.252 showed that MARCO is expressed
in TAMs of patients with breast cancer and metastatic
melanoma. MARCO neutralization with antibodies
inhibits tumour growth and metastasis in the 4T1 mam-
mary carcinoma model. Similarly, using a B16 melanoma
mouse model, treatment with an anti-MARCO antibody
inhibited tumour growth and improved the effects of
anti‑CTLA4 immunotherapy 252. The antitumour activity
of anti-MARCO therapy was dependent on the ability of
the Fc portion of the anti-MARCO antibody to engage

898 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

with the inhibitory Fc receptor FcγRIIB, as previously
demonstrated with other mAb-mediated reprogram-
ming strategies253. This study highlights the feasibility of
antibody-mediated reprogramming of macrophages by
using TAM-derived targets and stresses the importance
of the correct design of the antibodies, especially the Fc
region, for future clinical interventions.
PI3Kγ inhibitors. PI3Ks are involved in almost all types
of signalling in cells254. There are several subclasses of
PI3K, of which class 1B PI3Kγ is mainly expressed in
haematopoietic cells. Mice that lack PI3Kγ expression
show impaired recruitment of inflammatory cells, mainly
macrophages and neutrophils255. Kaneda et al.256 showed
that PI3Kγ is a key regulator of the tumour immune sup-
pression exerted by TAMs; genetic and pharmacological
inhibition of this target induced the expression of MHC
class II (MHC‑II) molecules together with upregulation
of IL‑12 and decreased secretion of IL‑10. As a result,
the inhibition of PI3Kγ in TAMs caused recruitment of
cells associated with antitumour adaptive immunity and
tumour growth inhibition. At the clinical level, patients
with head and neck and lung cancers with low PI3Kγ
activity had a better prognosis and longer overall sur-
vival, suggesting that PI3Kγ could be a potential future
therapeutic target.
Inhibition of microRNA activity. MicroRNAs (mi­RNAs)
are small non-coding RNAs that regulate transcription
and translation in a sequence-specific manner, and
their maturation is regulated by the RNase-III enzyme
DICER257. A recent study showed that inhibition of
DICER in macrophages affects TAM programming
and is associated with tumour regression and altered
infiltration of immune cells258. DICER inhibition repro-
grammed TAMs to express an IFNγ–STAT1 signature
and to become antitumoural. DICER inhibition in TAMs
was also associated with a better response to immunos-
timulatory antibodies258. These data raise the possibili-
ties of identification and targeting mi­RNAs to repolarize
TAMs.
Conclusions and perspectives
TAMs represent a heterogeneous population with
defined functions that differ between primary and
metastatic tumours. Immune cell infiltrates also vary
according to tumour type and, thus, dynamic inter­
actions between TAMs and other immune cells are
likely to vary according to cancer type and stage of pro-
gression. In addition to pro-tumoural effects, TAMs can
also have antitumour effects that in some cases might be
dominant. Consequently, it is necessary to understand
tumour heterogeneity and how it evolves during the pro-
gression to malignancy and also following therapy. It is
also judicious to realize that much of these data derive
from mouse models, and that there is scant knowledge
except at the most descriptive level about TAMs in
human cancers.
Despite these caveats, given the consensus that
TAMs are overall pro-tumoural, several anti-TAM strat-
egies aimed at depletion (such as CSF1R inhibitors),
reprogramming (such as PI3Kγ and HDAC inhibitors)
and targeting of functional molecules (such as argin-
ase 1 and Fc receptors) are in preclinical and clinical
trials (FIG. 4). Despite these advances, each strategy
needs further investigation, as they all have limitations.
For example, the general depletion of monocytes and
macrophages exerted by CSF1R inhibitors is not TAM-
specific and thus has substantial toxicity over time259.
Furthermore, given the complexity of TAM popula-
tions, there is growing awareness that the functioning of
macro­phages and dendritic cells is required for anti‑PD1
(REF.260 ) and anti‑CTLA4 therapies261, respectively. Thus,
potential alternative strategies would be to ablate TAMs
transiently, followed by recovery periods during which
monocytes could be recruited into the tissue to promote
antitumour immune reactions before differentiating into
pro-tumoural TAMs. This strategy is attractive, but it will
require a careful timing and a better knowledge of the
immune interactions ongoing in all phases of tumour
formation262.
Therefore, going forward, a better strategy would be
to specifically target pro-tumoural macrophages and
enhance the activity of antitumoural ones or to repo-
larize existing ones to have antitumoural activities.
In this context, TAM therapy aimed at the functional
modulation of TAM subpopulations through the use of
mAbs showed promising preclinical results. This strat-
egy, combined with the use of Fc receptor inhibitors,
seems to be the most promising one for the modulation
of the TME, as macrophages are able to rapidly take up
anti‑PD1 antibodies through their Fc receptors, limiting
the efficacy of the checkpoint inhibitors. Thus, it will be
fundamental to identify TAM-specific targets (such as
MARCO) to improve therapy specificity. Other meth-
ods of macrophage manipulation have been introduced,
including macrophages that express chimeric antigen
receptors (CAR-macrophages) and designer supramo-
lecules — for example, against CD47 and CSF1R — that
home to SIRPα-expressing macrophages and amplify
immune responses, limiting tumour growth263,264. These
exciting strategies will need considerable refinement to
reach human applications but may well be the wave of
the future.
Another interesting approach might also be to under-
stand the interactions between the microbiome and
macrophages. Such a concept is based on recent studies
suggesting that manipulation of the microbiome alters
cancer incidence as well as responses to therapy 265,266.
As macrophages are sentinel cells for changes in the
microbiota, usually through TLR engagement, under-
standing the exact nature of these interactions and their
consequences could potentially help tailor an anticancer
microbial cocktail267.
Another strategy is to target cancer-associated mye-
loid immature progenitors268, which are cells that are
mainly found in the blood of cancer patients and that
are thought to be the progenitors of tumour-infiltrating
myeloid cells. Several studies indicated that these imma-
ture cells have intrinsic immunosuppressive functions
in vitro and therefore could represent a very promising
target in combination with immunotherapy 269. However,

NATURE REVIEWS | DRUG DISCOVERY VOLUME 17 | DECEMBER 2018 | 899

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

again, the challenge is to identify specific markers that
will allow the selective targeting of these abnormally
expanded immature populations, as there are currently
not enough markers available to distinguish them from
mature myeloid cells.
To improve all therapeutic modalities, a thorough
understanding of the TME is required in humans and
in different cancer types, including the identification of
specific populations of cells, compensatory mechanisms
between cells of the innate system and their mechanisms
of immunosuppression and tissue repair. In this context,
the extensive use of single-cell RNA sequencing, multi­
plex immunohistochemistry and mass cytometry will
considerably increase our knowledge, and it will allow
the selection of novel TAM targets for the modulation
of the TME to improve therapy.

1. Dunn, G. P., Old, L. J. & Schreiber, R. D. The three Es
of cancer immunoediting. Annu. Rev. Immunol. 22,
329–360 (2004).
2. Gajewski, T. F., Schreiber, H. & Fu, Y. X. Innate and
adaptive immune cells in the tumor microenvironment.
Nat. Immunol. 14, 1014–1022 (2013).
3. Noy, R. & Pollard, J. W. Tumor-associated
macrophages: from mechanisms to therapy. Immunity
41, 49–61 (2014).
4. Arwert, E. N. et al. A unidirectional transition from
migratory to perivascular macrophage is required for
tumor cell intravasation. Cell Rep. 23, 1239–1248
(2018).
This paper shows both the recruitment of
monocytes and their ultimate differentiation to
perivascular macrophages that promote tumour
cell intravasation.
5. Qian, B. Z. et al. CCL2 recruits inflammatory
monocytes to facilitate breast-tumour metastasis.
Nature 475, 222–225 (2011).
This is the first paper to show that classical
monocytes are recruited to lung metastases from
breast tumours through a CCL2–CCR2‑dependent
mechanism and that inhibition of this recruitment
reduces metastasis.
6. Zhu, Y. et al. Tissue-resident macrophages in
pancreatic ductal adenocarcinoma originate from
embryonic hematopoiesis and promote tumor
progression. Immunity 47, 597 (2017).
This paper shows that TAMs in pancreatic ductal
adenocarcinoma originate from both bone
marrow-derived monocytes and embryonic-derived
tissue-resident macrophages and that the latter
support tumour progression.
7. Chen, Z. et al. Cellular and molecular identity of
tumor-associated macrophages in glioblastoma.
Cancer Res. 77, 2266–2278 (2017).
8. Bowman, R. L. et al. Macrophage ontogeny underlies
differences in tumor-specific education in brain
malignancies. Cell Rep. 17, 2445–2459 (2016).
9. Topalian, S. L., Drake, C. G. & Pardoll, D. M. Immune
checkpoint blockade: a common denominator
approach to cancer therapy. Cancer Cell 27, 450–461
(2015).
10. Ruffell, B. & Coussens, L. M. Macrophages and
therapeutic resistance in cancer. Cancer Cell 27,
462–472 (2015).
11. Mantovani, A., Marchesi, F., Malesci, A., Laghi, L. &
Allavena, P. Tumour-associated macrophages as
treatment targets in oncology. Nat. Rev. Clin. Oncol.
14, 399–416 (2017).
12. Gordon, S., Plüddemann, A. & Martinez Estrada, F.
Macrophage heterogeneity in tissues: phenotypic
diversity and functions. Immunol. Rev. 262, 36–55
(2014).
13. Pollard, J. W. Trophic macrophages in development
and disease. Nat. Rev. Immunol. 9, 259–270
(2009).
14. Gautier, E. L. et al. Gene-expression profiles and
transcriptional regulatory pathways that underlie the
identity and diversity of mouse tissue macrophages.
Nat. Immunol. 13, 1118–1128 (2012).
15. Chitu, V. & Stanley, E. R. Colony-stimulating factor‑1 in
immunity and inflammation. Curr. Opin. Immunol. 18,
39–48 (2006).
16. Stanley, E. R. & Chitu, V. CSF‑1 receptor signaling in
myeloid cells. Cold Spring Harb. Perspect. Biol. 6,
a021857 (2014).
17. Lin, H. et al. Discovery of a cytokine and its receptor
by functional screening of the extracellular proteome.
Science 320, 807–811 (2008).
18. Wang, Y. et al. IL‑34 is a tissue-restricted ligand of
CSF1R required for the development of Langerhans
cells and microglia. Nat. Immunol. 13, 753–760
(2012).
19. Baghdadi, M. et al. Chemotherapy-induced IL34
enhances immunosuppression by tumor-associated
macrophages and mediates survival of chemoresistant
lung cancer cells. Cancer Res. 76, 6030–6042 (2016).
20. Prehn, R. T. The immune reaction as a stimulator of
tumor growth. Science 176, 170–171 (1972).
This is a study that introduces the idea that
immune cells promote tumour growth.
21. Fidler, I. J. & Schroit, A. J. Recognition and destruction
of neoplastic cells by activated macrophages:
discrimination of altered self. Biochim. Biophys. Acta
948, 151–173 (1988).
22. Mantovani, A., Ming, W. J., Balotta, C., Abdeljalil, B. &
Bottazzi, B. Origin and regulation of tumor-associated
macrophages: the role of tumor-derived chemotactic
factor. Biochim. Biophys. Acta 865, 59–67 (1986).
23. Hibbs, J. B., Vavrin, Z. & Taintor, R. R. l‑Arginine is
required for expression of the activated macrophage
effector mechanism causing selective metabolic
inhibition in target cells. J. Immunol. 138, 550–565
(1987).
24. Lewis, C. E. & Pollard, J. W. Distinct role of
macrophages in different tumor microenvironments.
Cancer Res. 66, 605–612 (2006).
25. Lin, E. Y., Nguyen, A. V., Russell, R. G. & Pollard, J. W.
Colony-stimulating factor 1 promotes progression of
mammary tumors to malignancy. J. Exp. Med. 193,
727–740 (2001).
This is the first study to show that CSF1 and
macrophages promote tumour progression and
metastasis of mammary tumours.
26. Hanahan, D. & Coussens, L. M. Accessories to the
crime: functions of cells recruited to the tumor
microenvironment. Cancer Cell 21, 309–322 (2012).
27. Quail, D. F. & Joyce, J. A. Microenvironmental
regulation of tumor progression and metastasis. Nat.
Med. 19, 1423–1437 (2013).
28. Kitamura, T., Qian, B. Z. & Pollard, J. W. Immune cell
promotion of metastasis. Nat. Rev. Immunol. 15,
73–86 (2015).
29. Zhang, Q. W. et al. Prognostic significance of tumor-
associated macrophages in solid tumor: a meta-analysis
of the literature. PLOS ONE 7, e50946 (2012).
30. Yin, S. et al. The prognostic and clinicopathological
significance of tumor-associated macrophages in
patients with gastric cancer: a meta-analysis. PLOS
ONE 12, e0170042 (2017).
31. Zhao, X. et al. Prognostic significance of tumor-
associated macrophages in breast cancer: a meta-
analysis of the literature. Oncotarget 8, 30576–30586
(2017).
32. Guo, B., Cen, H., Tan, X. & Ke, Q. Meta-analysis of the
prognostic and clinical value of tumor-associated
macrophages in adult classical Hodgkin lymphoma.
BMC Med. 14, 159 (2016).
33. Mei, J. et al. Prognostic impact of tumor-associated
macrophage infiltration in non-small cell lung cancer:
A systemic review and meta-analysis. Oncotarget 7,
34217–34228 (2016).
34. Chambers, S. K., Wang, Y., Gertz, R. E. &
Kacinski, B. M. Macrophage colony-stimulating factor
mediates invasion of ovarian cancer cells through
urokinase. Cancer Res. 55, 1578–1585 (1995).
35. Kacinski, B. M. et al. FMS (CSF‑1 receptor) and CSF‑1
transcripts and protein are expressed by human
breast carcinomas in vivo and in vitro. Oncogene 6,
941–952 (1991).
36. Scholl, S. M. et al. Circulating levels of colony-
stimulating factor 1 as a prognostic indicator in 82
patients with epithelial ovarian cancer. Br. J. Cancer
69, 342–346 (1994).
This is the first study to show that elevated levels
of CSF1 are an independent prognostic factor
associated with poor survival in patients with
ovarian cancer.
37. Price, F. V. et al. Colony-stimulating factor‑1 in primary
ascites of ovarian cancer is a significant predictor of
survival. Am. J. Obstet. Gynecol. 168, 520–527
(1993).
38. Smith, H. O. et al. The role of colony-stimulating factor
1 and its receptor in the etiopathogenesis of
endometrial adenocarcinoma. Clin. Cancer Res. 1,
313–325 (1995).
This is the first prospective study to show that
patients with endometrial cancer have elevated
levels of serum CSF1 and high expression of CSF1
and its receptor at the tissue level compared with
control groups.
39. Smith, H. O. et al. The clinical significance of
inflammatory cytokines in primary cell culture in
endometrial carcinoma. Mol. Oncol. 7, 41–54
(2013).
40. West, R. B. et al. Determination of stromal signatures
in breast carcinoma. PLOS Biol. 3, e187 (2005).
41. Steidl, C. et al. Tumor-associated macrophages and
survival in classic Hodgkin’s lymphoma. N. Engl.
J. Med. 362, 875–885 (2010).
42. Sapi, E. The role of CSF‑1 in normal physiology of
mammary gland and breast cancer: an update. Exp.
Biol. Med. 229, 1–11 (2004).
43. Tang, R. et al. M‑CSF (monocyte colony stimulating
factor) and M‑CSF receptor expression by breast
tumour cells: M‑CSF mediated recruitment of
tumour infiltrating monocytes? J. Cell. Biochem. 50,
350–356 (1992).
44. Scholl, S. M. et al. Expression of colony-stimulating
factor‑1 and its receptor (the protein product of c‑fms)
in invasive breast tumor cells: induction of urokinase
production via this pathway? Ann. NY Acad. Sci. 698,
131–135 (1993).
45. Chambers, S. K., Kacinski, B. M., Ivins, C. M. &
Carcangiu, M. L. Overexpression of epithelial
macrophage colony-stimulating factor (CSF‑1) and
CSF‑1 receptor: a poor prognostic factor in epithelial
ovarian cancer, contrasted with a protective effect of
stromal CSF‑1. Clin. Cancer Res. 3, 999–1007
(1997).
46. Lamprecht, B. et al. Derepression of an endogenous
long terminal repeat activates the CSF1R proto-
oncogene in human lymphoma. Nat. Med. 16, 571–
579 (2010).
47. Crusz, S. M. & Balkwill, F. R. Inflammation and cancer:
advances and new agents. Nat. Rev. Clin. Oncol. 12,
584–596 (2015).
48. Moore, R. J. et al. Mice deficient in tumor necrosis
factor-alpha are resistant to skin carcinogenesis. Nat.
Med. 5, 828–831 (1999).
49. Canli, O. et al. Myeloid cell-derived reactive oxygen
species induce epithelial mutagenesis. Cancer Cell 32,
869–883 (2017).
50. Balkwill, F. R. & Mantovani, A. Cancer-related
inflammation: common themes and therapeutic
opportunities. Semin. Cancer Biol. 22, 33–40
(2012).
51. Deng, L. et al. A novel mouse model of inflammatory
bowel disease links mammalian target of rapamycin-
dependent hyperproliferation of colonic epithelium to
inflammation-associated tumorigenesis. Am. J. Pathol.
176, 952–967 (2010).
52. Kong, L. et al. Deletion of interleukin‑6 in monocytes/
macrophages suppresses the initiation of
hepatocellular carcinoma in mice. J. Exp. Clin. Cancer
Res. 35, 131 (2016).
53. Oft, M. IL‑10: master switch from tumor-promoting
inflammation to antitumor immunity. Cancer Immunol.
Res. 2, 194–199 (2014).
54. Jobin, C. Precision medicine using microbiota. Science
359, 32–34 (2018).
55. Jobin, C. Colorectal cancer: looking for answers in the
microbiota. Cancer Discov. 3, 384–387 (2013).

900 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.

56. de Visser, K. E., Eichten, A. & Coussens, L. M.
Paradoxical roles of the immune system during
cancer development. Nat. Rev. Cancer 6, 24–37
(2006).
57. DeNardo, D. G. et al. CD4(+) T cells regulate
pulmonary metastasis of mammary carcinomas by
enhancing protumor properties of macrophages.
Cancer Cell 16, 91–102 (2009).
This study elucidates a pro-tumoural role of
IL‑4‑expressing CD4+ TH2 cells, which regulate the
phenotype of macrophages in mammary tumours
and promote tumour progression.
58. Ruffell, B., Affara, N. I. & Coussens, L. M. Differential
macrophage programming in the tumor
microenvironment. Trends Immunol. 33, 119–126
(2012).
59. Bruno, A. et al. Orchestration of angiogenesis by
immune cells. Front. Oncol. 4, 131 (2014).
60. Bergers, G. & Benjamin, L. E. Tumorigenesis and the
angiogenic switch. Nat. Rev. Cancer 3, 401–410
(2003).
61. Yeo, E. J. et al. Myeloid WNT7b mediates the
angiogenic switch and metastasis in breast cancer.
Cancer Res. 74, 2962–2973 (2014).
62. Lin, E. Y. & Pollard, J. W. Tumor-associated
macrophages press the angiogenic switch in breast
cancer. Cancer Res. 67, 5064–5066 (2007).
63. Hughes, R. et al. Perivascular M2 macrophages
stimulate tumor relapse after chemotherapy. Cancer
Res. 75, 3479–3491 (2015).
64. Hirano, H., Tanioka, K., Yokoyama, S., Akiyama, S. &
Kuratsu, J. Angiogenic effect of thymidine
phosphorylase on macrophages in glioblastoma
multiforme. J. Neurosurg. 95, 89–95 (2001).
65. De Palma, M. et al. Tie2 identifies a hematopoietic
lineage of proangiogenic monocytes required for
tumor vessel formation and a mesenchymal
population of pericyte progenitors. Cancer Cell 8,
211–226 (2005).
This is the first study to identify a novel population
of pro-angiogenic monocytes expressing TIE2,
which are recruited to the tumour and account for
most of the pro-angiogenic activity among myeloid
cells in the tumour.
66. Mazzieri, R. et al. Targeting the ANG2/TIE2 axis
inhibits tumor growth and metastasis by impairing
angiogenesis and disabling rebounds of
proangiogenic myeloid cells. Cancer Cell 19,
512–526 (2011).
67. Wyckoff, J. et al. A paracrine loop between tumor cells
and macrophages is required for tumor cell migration
in mammary tumors. Cancer Res. 64, 7022–7029
(2004).
This study identifies a direct crosstalk between
tumour cells and macrophages that involves EGF
and CSF1 and is required for tumour cell migration.
68. Goswami, S. et al. Macrophages promote the invasion
of breast carcinoma cells via a colony-stimulating
factor‑1/epidermal growth factor paracrine loop.
Cancer Res. 65, 5278–5283 (2005).
69. Hernandez, L. et al. The EGF/CSF‑1 paracrine invasion
loop can be triggered by heregulin beta1 and CXCL12.
Cancer Res. 69, 3221–3227 (2009).
70. Harney, A. S. et al. Real-time imaging reveals local,
transient vascular permeability, and tumor cell
intravasation stimulated by TIE2hi macrophage-
derived VEGFA. Cancer Discov. 5, 932–943 (2015).
71. Rohan, T. E. et al. Tumor microenvironment of
metastasis and risk of distant metastasis of breast
cancer. J. Natl Cancer Inst. 106, dju136 (2014).
72. Ojalvo, L. S., King, W., Cox, D. & Pollard, J. W. High-
density gene expression analysis of tumor-associated
macrophages from mouse mammary tumors. Am.
J. Pathol. 174, 1048–1064 (2009).
73. Ojalvo, L. S., Whittaker, C. A., Condeelis, J. S. &
Pollard, J. W. Gene expression analysis of
macrophages that facilitate tumor invasion
supports a role for Wnt-signaling in mediating their
activity in primary mammary tumors. J. Immunol.
184, 702–712 (2010).
74. Morandi, F. & Pistoia, V. Interactions between HLA‑G
and HLA‑E in physiological and pathological
conditions. Front. Immunol. 5, 394 (2014).
75. Santarpia, M. & Karachaliou, N. Tumor immune
microenvironment characterization and response to
anti‑PD‑1 therapy. Cancer Biol. Med. 12, 74–78
(2015).
76. Buchbinder, E. I. & Desai, A. CTLA‑4 and PD‑1
pathways: similarities, differences, and implications of
their inhibition. Am. J. Clin. Oncol. 39, 98–106
(2016).
NATURE REVIEWS | DRUG DISCOVERY
77. Mellor, A. L., Keskin, D. B., Johnson, T., Chandler, P.
& Munn, D. H. Cells expressing indoleamine
2,3‑dioxygenase inhibit T cell responses. J. Immunol.
168, 3771–3776 (2002).
78. Mbongue, J. C. et al. The role of indoleamine 2,
3‑dioxygenase in immune suppression and
autoimmunity. Vaccines 3, 703–729 (2015).
79. Qian, B. et al. A distinct macrophage population
mediates metastatic breast cancer cell extravasation,
establishment and growth. PLOS ONE 4, e6562
(2009).
This is the first study to identify a novel
subpopulation of TAMs that accumulate in metastatic
lesions and promote tumour dissemination.
80. Kitamura, T. et al. CCL2‑induced chemokine cascade
promotes breast cancer metastasis by enhancing
retention of metastasis-associated macrophages.
J. Exp. Med. 212, 1043–1059 (2015).
81. Zhao, L. et al. Recruitment of a myeloid cell subset
(CD11b/Gr1 mid) via CCL2/CCR2 promotes the
development of colorectal cancer liver metastasis.
Hepatology 57, 829–839 (2013).
82. Qian, B. Z. et al. FLT1 signaling in metastasis-
associated macrophages activates an inflammatory
signature that promotes breast cancer metastasis.
J. Exp. Med. 212, 1433–1448 (2015).
83. Kitamura, T. et al. Monocytes differentiate to immune
suppressive precursors of metastasis-associated
macrophages in mouse models of breast cancer.
Front. Immunol. 8, 2004 (2018).
84. Entenberg, D. et al. In vivo subcellular resolution
optical imaging in the lung reveals early metastatic
proliferation and motility. Intravital 4, e1086613
(2015).
85. Chen, Q., Zhang, X. H. & Massague, J. Macrophage
binding to receptor VCAM‑1 transmits survival
signals in breast cancer cells that invade the lungs.
Cancer Cell 20, 538–549 (2011).
86. Gao, D. et al. Myeloid progenitor cells in the
premetastatic lung promote metastases by inducing
mesenchymal to epithelial transition. Cancer Res.
72, 1384–1394 (2012).
87. Camacho, D. F. & Pienta, K. J. A multi-targeted
approach to treating bone metastases. Cancer
Metastasis Rev. 33, 545–553 (2014).
88. Madani, I., De Neve, W. & Mareel, M. Does ionizing
radiation stimulate cancer invasion and metastasis?
Bull. Cancer 95, 292–300 (2008).
89. Moding, E. J., Kastan, M. B. & Kirsch, D. G.
Strategies for optimizing the response of cancer and
normal tissues to radiation. Nat. Rev. Drug Discov.
12, 526–542 (2013).
90. Russell, J. S. & Brown, J. M. The irradiated tumor
microenvironment: role of tumor-associated
macrophages in vascular recovery. Front. Physiol. 4,
157 (2013).
91. Kaur, P. & Asea, A. Radiation-induced effects and the
immune system in cancer. Front. Oncol. 2, 191 (2012).
92. Shiao, S. L. & Coussens, L. M. The tumor-immune
microenvironment and response to radiation therapy.
J. Mammary Gland Biol. Neoplasia 15, 411–421
(2010).
93. Barker, H. E., Paget, J. T., Khan, A. A. &
Harrington, K. J. The tumour microenvironment after
radiotherapy: mechanisms of resistance and
recurrence. Nat. Rev. Cancer 15, 409–425 (2015).
94. Teresa Pinto, A. et al. Ionizing radiation modulates
human macrophages towards a pro-inflammatory
phenotype preserving their pro-invasive and pro-
angiogenic capacities. Sci. Rep. 6, 18765 (2016).
95. Hellevik, T. & Martinez-Zubiaurre, I. Radiotherapy
and the tumor stroma: the importance of dose and
fractionation. Front. Oncol. 4, 1 (2014).
96. Correia, A. L. & Bissell, M. J. The tumor
microenvironment is a dominant force in multidrug
resistance. Drug Resist. Updat. 15, 39–49 (2012).
97. Nakasone, E. S. et al. Imaging tumor-stroma
interactions during chemotherapy reveals
contributions of the microenvironment to resistance.
Cancer Cell 21, 488–503 (2012).
98. Shaked, Y. Balancing efficacy of and host immune
responses to cancer therapy: the yin and yang
effects. Nat. Rev. Clin. Oncol. 13, 611–626 (2016).
99. Paulus, P., Stanley, E. R., Schafer, R., Abraham, D. &
Aharinejad, S. Colony-stimulating factor‑1 antibody
reverses chemoresistance in human MCF‑7 breast
cancer xenografts. Cancer Res. 66, 4349–4356
(2006).
This study shows the antitumoural effects of
anti‑CSF1 therapy in MCF‑7 breast cancer
xenografts.
©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS
100. DeNardo, D. G. et al. Leukocyte complexity predicts
breast cancer survival and functionally regulates
response to chemotherapy. Cancer Discov. 1, 54–67
(2011).
This study demonstrates that inhibition of
macrophage recruitment in the tumour can
influence response to chemotherapy, suggesting
that macrophage targeting is a therapeutic strategy
for cancer treatment.
101. Alishekevitz, D. et al. Macrophage-induced
lymphangiogenesis and metastasis following paclitaxel
chemotherapy is regulated by VEGFR3. Cell Rep. 17,
1344–1356 (2016).
102. Shree, T. et al. Macrophages and cathepsin proteases
blunt chemotherapeutic response in breast cancer.
Genes Dev. 25, 2465–2479 (2011).
103. Zhang, X. et al. Macrophages induce resistance to
5‑fluorouracil chemotherapy in colorectal cancer
through the release of putrescine. Cancer Lett. 381,
305–313 (2016).
104. Schumacher, T. N. & Schreiber, R. D. Neoantigens
in cancer immunotherapy. Science 348, 69–74
(2015).
105. Rosenberg, S. A., Yang, J. C. & Restifo, N. P. Cancer
immunotherapy: moving beyond current vaccines. Nat.
Med. 10, 909–915 (2004).
106. Sharma, P. & Allison, J. P. Immune checkpoint
targeting in cancer therapy: toward combination
strategies with curative potential. Cell 161, 205–214
(2015).
107. Gordon, S. R. et al. PD‑1 expression by tumour-
associated macrophages inhibits phagocytosis
and tumour immunity. Nature 545, 495–499
(2017).
This study shows that human and mouse TAMs
express PD1, which inhibits tumour phagocytosis.
Inhibition of PD1 on TAMs reduced tumour growth,
suggesting that checkpoint inhibitor therapy is also
effective on TAMs.
108. Neubert, N. J. et al. T cell-induced CSF1 promotes
melanoma resistance to PD1 blockade. Sci. Transl
Med. 10, eaan3311 (2018).
109. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02452424
(2018).
110. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02777710
(2017).
111. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02880371
(2018).
112. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02829723
(2018).
113. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02323191
(2018).
114. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02760797
(2018).
115. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02713529
(2018).
116. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02526017
(2018).
117. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02718911
(2018).
118. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02807844
(2018).
119. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02554812
(2018).
120. Coniglio, S. J. et al. Microglial stimulation of
glioblastoma invasion involves epidermal growth factor
receptor (EGFR) and colony stimulating factor 1
receptor (CSF‑1R) signaling. Mol. Med. 18, 519–527
(2012).
121. Yan, D. et al. Inhibition of colony stimulating factor‑1
receptor abrogates microenvironment-mediated
therapeutic resistance in gliomas. Oncogene 36,
6049–6058 (2017).
122. Tap, W. D. et al. Structure-guided blockade of CSF1R
kinase in tenosynovial giant-cell tumor. N. Engl.
J. Med. 373, 428–437 (2015).
123. Molena, B. et al. Synovial colony-stimulating factor‑1
mRNA expression in diffuse pigmented villonodular
synovitis. Clin. Exp. Rheumatol 29, 547–550
(2011).
VOLUME 17 | DECEMBER 2018 | 901
REVIEWS
124. Ries, C. H. et al. Targeting tumor-associated
macrophages with anti-CSF‑1R antibody reveals a
strategy for cancer therapy. Cancer Cell 25, 846–859
(2014).
This is the first clinical report showing activity of an
anti‑CSF1R antibody in diffuse-type giant cell
tumours.
125. Butowski, N. et al. Orally administered colony
stimulating factor 1 receptor inhibitor PLX3397 in
recurrent glioblastoma: an Ivy Foundation Early Phase
Clinical Trials Consortium phase II study. Neuro Oncol.
18, 557–564 (2016).
126. von Tresckow, B. et al. An open-label, multicenter,
phase I/II Study of JNJ‑40346527, a CSF‑1R
inhibitor, in patients with relapsed or refractory
Hodgkin lymphoma. Clin. Cancer Res. 21,
1843–1850 (2015).
127. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT01804530
(2018).
128. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT01316822
(2018).
129. Pyonteck, S. M. et al. CSF‑1R inhibition alters
macrophage polarization and blocks glioma
progression. Nat. Med. 19, 1264–1272 (2013).
This is an important paper in mouse models of
glioblastoma that shows that macrophages can be
repolarized to become antitumoural.
130. Quail, D. F. et al. The tumor microenvironment
underlies acquired resistance to CSF‑1R inhibition in
gliomas. Science 352, aad3018 (2016).
131. Cassier, P. A. et al. CSF1R inhibition with
emactuzumab in locally advanced diffuse-type
tenosynovial giant cell tumours of the soft tissue: a
dose-escalation and dose-expansion phase 1 study.
Lancet Oncol. 16, 949–956 (2015).
132. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02471716
(2018).
133. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT03158272
(2018).
134. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT01346358
(2018).
135. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02265536
(2018).
136. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT03153410
(2018).
137. Rodan, G. A. & Fleisch, H. A. Bisphosphonates:
mechanisms of action. J. Clin. Invest. 97, 2692–2696
(1996).
138. Stresing, V., Daubiné, F., Benzaid, I., Mönkkönen, H. &
Clézardin, P. Bisphosphonates in cancer therapy.
Cancer Lett. 257, 16–35 (2007).
139. Van Acker, H. H., Anguille, S., Willemen, Y.,
Smits, E. L. & Van Tendeloo, V. F. Bisphosphonates for
cancer treatment: mechanisms of action and lessons
from clinical trials. Pharmacol. Ther. 158, 24–40
(2016).
140. Moreau, M. F. et al. Comparative effects of five
bisphosphonates on apoptosis of macrophage cells
in vitro. Biochem. Pharmacol. 73, 718–723 (2007).
141. Rogers, T. L. & Holen, I. Tumour macrophages as
potential targets of bisphosphonates. J. Transl Med.
9, 177 (2011).
142. Hiraoka, K. et al. Inhibition of bone and muscle
metastases of lung cancer cells by a decrease in the
number of monocytes/macrophages. Cancer Sci. 99,
1595–1602 (2008).
143. Gazzaniga, S. et al. Targeting tumor-associated
macrophages and inhibition of MCP‑1 reduce
angiogenesis and tumor growth in a human melanoma
xenograft. J. Invest. Dermatol. 127, 2031–2041
(2007).
144. Zeisberger, S. M. et al. Clodronate-liposome-mediated
depletion of tumour-associated macrophages: a new
and highly effective antiangiogenic therapy approach.
Br. J. Cancer 95, 272–281 (2006).
145. Zhang, W. et al. Depletion of tumor-associated
macrophages enhances the effect of sorafenib in
metastatic liver cancer models by antimetastatic and
antiangiogenic effects. Clin. Cancer Res. 16,
3420–3430 (2010).
146. Guth, A. M., Hafeman, S. D., Elmslie, R. E. &
Dow, S. W. Liposomal clodronate treatment for
tumour macrophage depletion in dogs with soft-tissue
sarcoma. Vet. Comp. Oncol. 11, 296–305 (2013).
902 | DECEMBER 2018 | VOLUME 17
147. Daubiné, F., Le Gall, C., Gasser, J., Green, J. &
Clézardin, P. Antitumor effects of clinical dosing
regimens of bisphosphonates in experimental breast
cancer bone metastasis. J. Natl Cancer Inst. 99,
322–330 (2007).
148. Coscia, M. et al. Zoledronic acid repolarizes tumour-
associated macrophages and inhibits mammary
carcinogenesis by targeting the mevalonate pathway.
J. Cell. Mol. Med. 14, 2803–2815 (2010).
149. Comito, G. et al. Zoledronic acid impairs stromal
reactivity by inhibiting M2‑macrophages polarization
and prostate cancer-associated fibroblasts.
Oncotarget 8, 118–132 (2017).
150. Marra, M. et al. Cutting the limits of
aminobisphosphonates: new strategies for the
potentiation of their anti-tumour effects. Curr. Cancer
Drug Targets 9, 791–800 (2009).
151. Marra, M. et al. Nanotechnologies to use
bisphosphonates as potent anticancer agents: the
effects of zoledronic acid encapsulated into liposomes.
Nanomedicine 7, 955–964 (2011).
152. Marra, M. et al. New self-assembly nanoparticles and
stealth liposomes for the delivery of zoledronic acid: a
comparative study. Biotechnol. Adv. 30, 302–309
(2012).
153. Ben-Aharon, I. et al. Bisphosphonates in the adjuvant
setting of breast cancer therapy—effect on survival: a
systematic review and meta-analysis. PLOS ONE 8,
e70044 (2013).
154. Gnant, M. et al. Endocrine therapy plus zoledronic
acid in premenopausal breast cancer. N. Engl. J. Med.
360, 679–691 (2009).
155. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02347163
(2017).
156. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02622607
(2017).
157. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00127205
(2017).
158. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00009945
(2017).
159. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00003232
(2010).
160. Cuevas, C. & Francesch, A. Development of Yondelis
(trabectedin, ET‑743). A semisynthetic process solves
the supply problem. Nat. Prod. Rep. 26, 322–337
(2009).
161. Le Cesne, A. et al. A retrospective analysis of antitumour
activity with trabectedin in translocation-related
sarcomas. Eur. J. Cancer 48, 3036–3044 (2012).
162. Grosso, F. et al. Efficacy of trabectedin
(ecteinascidin‑743) in advanced pretreated myxoid
liposarcomas: a retrospective study. Lancet Oncol. 8,
595–602 (2007).
163. Monk, B. J. et al. Trabectedin plus pegylated
liposomal doxorubicin (PLD) versus PLD in recurrent
ovarian cancer: overall survival analysis. Eur. J. Cancer
48, 2361–2368 (2012).
164. Germano, G. et al. Role of macrophage targeting in
the antitumor activity of trabectedin. Cancer Cell 23,
249–262 (2013).
This study shows that the antitumoural drug
trabectedin induces monocyte and macrophage
apoptosis through the activation of caspase 8 in a
TRAIL-dependent mechanism.
165. Liguori, M. et al. Functional TRAIL receptors in
monocytes and tumor-associated macrophages: a
possible targeting pathway in the tumor
microenvironment. Oncotarget 7, 41662–41676
(2016).
166. Amarante-Mendes, G. P. & Griffith, T. S. Therapeutic
applications of TRAIL receptor agonists in cancer and
beyond. Pharmacol. Ther. 155, 117–131 (2015).
167. Deshmane, S. L., Kremlev, S., Amini, S. &
Sawaya, B. E. Monocyte chemoattractant protein‑1
(MCP‑1): an overview. J. Interferon Cytokine Res. 29,
313–326 (2009).
168. Chun, E. et al. CCL2 promotes colorectal
carcinogenesis by enhancing polymorphonuclear
myeloid-derived suppressor cell population and
function. Cell Rep. 12, 244–257 (2015).
169. Fang, W. B. et al. CCL2/CCR2 chemokine signaling
coordinates survival and motility of breast cancer cells
through Smad3 protein- and p42/44 mitogen-
activated protein kinase (MAPK)-dependent
mechanisms. J. Biol. Chem. 287, 36593–36608
(2012).
©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
170. Fang, W. B. et al. Targeted gene silencing of CCL2
inhibits triple negative breast cancer progression
by blocking cancer stem cell renewal and M2
macrophage recruitment. Oncotarget 7,
49349–49367 (2016).
171. Peña, C. G. et al. LKB1 loss promotes endometrial
cancer progression via CCL2‑dependent macrophage
recruitment. J. Clin. Invest. 125, 4063–4076 (2015).
This is an important paper showing how loss of a
tumour suppressor gene can induce CCL2
production that results in TAM recruitment.
172. Li, M., Knight, D. A., A. Snyder, L., Smyth, M. J. &
Stewart, T. J. A role for CCL2 in both tumor
progression and immunosurveillance.
Oncoimmunology 2, e25474 (2013).
173. Bonapace, L. et al. Cessation of CCL2 inhibition
accelerates breast cancer metastasis by promoting
angiogenesis. Nature 515, 130–133 (2014).
This study reports the paradoxical pro-metastatic
effect of the cessation CCL2 inhibition in different
tumour models.
174. Hitchcock, J. R. & Watson, C. J. Anti‑CCL2: building a
reservoir or opening the floodgates to metastasis?
Breast Cancer Res. 17, 68 (2015).
175. Ueno, T. et al. Significance of macrophage
chemoattractant protein‑1 in macrophage
recruitment, angiogenesis, and survival in human
breast cancer. Clin. Cancer Res. 6, 3282–3289
(2000).
176. Lebrecht, A. et al. Monocyte chemoattractant
protein‑1 serum levels in patients with breast cancer.
Tumour Biol. 25, 14–17 (2004).
177. Loberg, R. D. et al. CCL2 as an important mediator of
prostate cancer growth in vivo through the regulation
of macrophage infiltration. Neoplasia 9, 556–562
(2007).
178. Loberg, R. D. et al. Targeting CCL2 with systemic
delivery of neutralizing antibodies induces prostate
cancer tumor regression in vivo. Cancer Res. 67,
9417–9424 (2007).
179. Moisan, F. et al. Enhancement of paclitaxel and
carboplatin therapies by CCL2 blockade in ovarian
cancers. Mol. Oncol. 8, 1231–1239 (2014).
180. Sandhu, S. K. et al. A first‑in‑human, first‑in‑class,
phase I study of carlumab (CNTO 888), a human
monoclonal antibody against CC‑chemokine ligand 2
in patients with solid tumors. Cancer Chemother.
Pharmacol. 71, 1041–1050 (2013).
181. Pienta, K. J. et al. Phase 2 study of carlumab (CNTO
888), a human monoclonal antibody against
CC‑chemokine ligand 2 (CCL2), in metastatic
castration-resistant prostate cancer. Invest. New
Drugs 31, 760–768 (2013).
182. Nywening, T. M. et al. Targeting tumour-associated
macrophages with CCR2 inhibition in combination
with FOLFIRINOX in patients with borderline
resectable and locally advanced pancreatic cancer: a
single-centre, open-label, dose-finding, non-
randomised, phase 1b trial. Lancet Oncol. 17,
651–662 (2016).
183. Wynn, T. A., Chawla, A. & Pollard, J. W. Macrophage
biology in development, homeostasis and disease.
Nature 496, 445 (2013).
184. Williams, C. B., Yeh, E. S. & Soloff, A. C. Tumor-
associated macrophages: unwitting accomplices in
breast cancer malignancy. NPJ Breast Cancer 2,
15025 (2016).
185. Brown, E. J. & Frazier, W. A. Integrin-associated
protein (CD47) and its ligands. Trends Cell Biol. 11,
130–135 (2001).
186. Gao, A. G. et al. Integrin-associated protein is a
receptor for the C‑terminal domain of
thrombospondin. J. Biol. Chem. 271, 21–24 (1996).
187. Liu, Y. et al. The role of CD47 in neutrophil
transmigration. Increased rate of migration correlates
with increased cell surface expression of CD47. J. Biol.
Chem. 276, 40156–40166 (2001).
188. Lindberg, F. P. et al. Decreased resistance to bacterial
infection and granulocyte defects in IAP-deficient mice.
Science 274, 795–798 (1996).
189. Okazawa, H. et al. Negative regulation of phagocytosis
in macrophages by the CD47‑SHPS‑1 system.
J. Immunol. 174, 2004–2011 (2005).
190. Barclay, A. N. & Van den Berg, T. K. The interaction
between signal regulatory protein alpha (SIRPα) and
CD47: structure, function, and therapeutic target.
Annu. Rev. Immunol. 32, 25–50 (2014).
191. Bian, Z. et al. Cd47‑Sirpα interaction and IL‑10
constrain inflammation-induced macrophage
phagocytosis of healthy self-cells. Proc. Natl Acad. Sci.
USA 113, E5434–E5443 (2016).
www.nature.com/nrd

192. Jaiswal, S. et al. CD47 is upregulated on circulating
hematopoietic stem cells and leukemia cells to avoid
phagocytosis. Cell 138, 271–285 (2009).
This study shows that CD47 expression is
upregulated by leukaemic cells to escape
phagocytosis by macrophages.
193. Chao, M. P. et al. Anti‑CD47 antibody synergizes with
rituximab to promote phagocytosis and eradicate
non-Hodgkin lymphoma. Cell 142, 699–713 (2010).
194. Zhao, H. et al. CD47 promotes tumor invasion and
metastasis in non-small cell lung cancer. Sci. Rep. 6,
29719 (2016).
195. Zhang, H. et al. HIF‑1 regulates CD47 expression in
breast cancer cells to promote evasion of phagocytosis
and maintenance of cancer stem cells. Proc. Natl
Acad. Sci. USA 112, E6215–E6223 (2015).
196. Betancur, P. A. et al. A CD47‑associated super-
enhancer links pro-inflammatory signalling to CD47
upregulation in breast cancer. Nat. Commun. 8,
14802 (2017).
197. Willingham, S. B. et al. The CD47‑signal regulatory
protein alpha (SIRPa) interaction is a therapeutic
target for human solid tumors. Proc. Natl Acad. Sci.
USA 109, 6662–6667 (2012).
198. Edris, B. et al. Antibody therapy targeting the CD47
protein is effective in a model of aggressive metastatic
leiomyosarcoma. Proc. Natl Acad. Sci. USA 109,
6656–6661 (2012).
199. Xiao, Z. et al. Antibody mediated therapy targeting
CD47 inhibits tumor progression of hepatocellular
carcinoma. Cancer Lett. 360, 302–309 (2015).
200. Liu, R. et al. CD47 promotes ovarian cancer
progression by inhibiting macrophage phagocytosis.
Oncotarget 8, 39021–39032 (2017).
201. Weiskopf, K. et al. CD47‑blocking immunotherapies
stimulate macrophage-mediated destruction of small-
cell lung cancer. J. Clin. Invest. 126, 2610–2620
(2016).
202. Alvey, C. M. et al. SIRPA-inhibited, marrow-derived
macrophages engorge, accumulate, and differentiate
in antibody-targeted regression of solid tumors. Curr.
Biol. 27, 2065–2077 (2017).
203. Liu, J. et al. Pre-clinical development of a humanized
anti‑CD47 antibody with anti-cancer therapeutic
potential. PLOS ONE 10, e0137345 (2015).
204. Gholamin, S. et al. Disrupting the CD47‑SIRPα anti-
phagocytic axis by a humanized anti‑CD47 antibody is
an efficacious treatment for malignant pediatric brain
tumors. Sci. Transl Med. 9, eaaf2968 (2017).
205. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02216409
(2018).
206. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02678338
(2018).
207. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02953782
(2018).
208. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02953509
(2018).
209. Sikic, B. I. et al. A first‑in‑human, first‑in‑class phase I
trial of the anti‑CD47 antibody Hu5F9‑G4 in patients
with advanced cancers. J. Clin. Oncol. 34 (Suppl. 15),
3019 (2016).
210. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02367196
(2018).
211. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02641002
(2018).
212. Petrova, P. S. et al. TTI‑621 (SIRPαFc): a
CD47‑blocking innate immune checkpoint inhibitor
with broad antitumor activity and minimal erythrocyte
binding. Clin. Cancer Res. 23, 1068–1079 (2017).
213. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02663518
(2018).
214. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02890368
(2018).
215. Medzhitov, R. & Janeway, C. The Toll receptor family
and microbial recognition. Trends Microbiol. 8,
452–456 (2000).
216. Kaczanowska, S., Joseph, A. M. & Davila, E. TLR
agonists: our best frenemy in cancer immunotherapy.
J. Leukoc. Biol. 93, 847–863 (2013).
217. Le Mercier, I. et al. Tumor promotion by intratumoral
plasmacytoid dendritic cells is reversed by TLR7
ligand treatment. Cancer Res. 73, 4629–4640
(2013).
NATURE REVIEWS | DRUG DISCOVERY
218. Singh, M. et al. Effective innate and adaptive
antimelanoma immunity through localized TLR7/8
activation. J. Immunol. 193, 4722–4731 (2014).
219. Kobold, S., Wiedemann, G., Rothenfußer, S. &
Endres, S. Modes of action of TLR7 agonists in cancer
therapy. Immunotherapy 6, 1085–1095 (2014).
220. Vacchelli, E. et al. Trial watch: immunotherapy plus
radiation therapy for oncological indications.
Oncoimmunology 5, e1214790 (2016).
221. Adams, S. et al. Topical TLR7 agonist imiquimod can
induce immune-mediated rejection of skin metastases
in patients with breast cancer. Clin. Cancer Res. 18,
6748–6757 (2012).
222. Menzies, S., Mc Menamin, M. & Barry, R. Lentigo
maligna successfully treated with combination therapy
of topical tazarotene and imiquimod. Clin. Exp.
Dermatol. 42, 468–470 (2017).
223. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00319748
(2017).
224. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00095160
(2017).
225. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00276159
(2017).
226. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00189332
(2008).
227. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00091689
(2008).
228. Dudek, A. Z. et al. First in human phase I trial of
852A, a novel systemic toll-like receptor 7 agonist, to
activate innate immune responses in patients with
advanced cancer. Clin. Cancer Res. 13, 7119–7125
(2007).
229. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00719199
(2013).
230. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00633529
(2013).
231. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT01040832
(2017).
232. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT00729053
(2018).
233. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT01360827
(2014).
234. Smith, D. A. et al. Antitumor activity and safety of
combination therapy with the Toll-like receptor 9
agonist IMO‑2055, erlotinib, and bevacizumab in
advanced or metastatic non-small cell lung cancer
patients who have progressed following
chemotherapy. Cancer Immunol. Immunother. 63,
787–796 (2014).
235. Khalil, M. & Vonderheide, R. H. Anti‑CD40 agonist
antibodies: preclinical and clinical experience. Update
Cancer Ther. 2, 61–65 (2007).
236. van Kooten, C. & Banchereau, J. CD40‑CD40 ligand.
J. Leukoc. Biol. 67, 2–17 (2000).
237. Hoves, S. et al. Rapid activation of tumor-associated
macrophages boosts preexisting tumor immunity.
J. Exp. Med. 215, 859–876 (2018).
238. Perry, C. J. et al. Myeloid-targeted immunotherapies
act in synergy to induce inflammation and antitumor
immunity. J. Exp. Med. 215, 877–893 (2018).
239. Vonderheide, R. H. et al. Clinical activity and immune
modulation in cancer patients treated with
CP‑870,893, a novel CD40 agonist monoclonal
antibody. J. Clin. Oncol. 25, 876–883 (2007).
240. Vonderheide, R. H. et al. Phase I study of the CD40
agonist antibody CP‑870,893 combined with
carboplatin and paclitaxel in patients with advanced
solid tumors. Oncoimmunology 2, e23033 (2013).
241. Nowak, A. K. et al. A phase 1b clinical trial of the
CD40‑activating antibody CP‑870,893 in
combination with cisplatin and pemetrexed in
malignant pleural mesothelioma. Ann. Oncol. 26,
2483–2490 (2015).
242. Beatty, G. L. et al. A phase I study of an agonist CD40
monoclonal antibody (CP‑870,893) in combination
with gemcitabine in patients with advanced
pancreatic ductal adenocarcinoma. Clin. Cancer Res.
19, 6286–6295 (2013).
243. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02304393
(2018).
©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS
244. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02665416
(2018).
245. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02588443
(2017).
246. Arrowsmith, C. H., Bountra, C., Fish, P. V., Lee, K. &
Schapira, M. Epigenetic protein families: a new
frontier for drug discovery. Nat. Rev. Drug Discov. 11,
384–400 (2012).
247. Choudhary, C., Weinert, B. T., Nishida, Y., Verdin, E. &
Mann, M. The growing landscape of lysine acetylation
links metabolism and cell signalling. Nat. Rev. Mol.
Cell Biol. 15, 536–550 (2014).
248. Lobera, M. et al. Selective class IIa histone
deacetylase inhibition via a nonchelating zinc-binding
group. Nat. Chem. Biol. 9, 319–325 (2013).
249. Guerriero, J. L. et al. Class IIa HDAC inhibition reduces
breast tumours and metastases through anti-tumour
macrophages. Nature 543, 428–432 (2017).
This paper shows that class IIa HDAC inhibitors can
reprogramme TAMs to antitumoural macrophages
and increase the efficacy of chemotherapy and
immunotherapy in preclinical models of cancer.
250. Elomaa, O. et al. Cloning of a novel bacteria-binding
receptor structurally related to scavenger receptors
and expressed in a subset of macrophages. Cell 80,
603–609 (1995).
251. Bergamaschi, A. et al. Extracellular matrix signature
identifies breast cancer subgroups with different
clinical outcome. J. Pathol. 214, 357–367 (2008).
252. Georgoudaki, A. M. et al. Reprogramming tumor-
associated macrophages by antibody targeting
inhibits cancer progression and metastasis. Cell Rep.
15, 2000–2011 (2016).
253. Li, F. & Ravetch, J. V. Inhibitory Fcγ receptor
engagement drives adjuvant and anti-tumor activities of
agonistic CD40 antibodies. Science 333, 1030–1034
(2011).
254. Sotsios, Y. & Ward, S. G. Phosphoinositide 3‑kinase: a
key biochemical signal for cell migration in response to
chemokines. Immunol. Rev. 177, 217–235 (2000).
255. Hirsch, E. et al. Central role for G protein-coupled
phosphoinositide 3‑kinase gamma in inflammation.
Science 287, 1049–1053 (2000).
256. Kaneda, M. M. et al. PI3Kgamma is a molecular
switch that controls immune suppression. Nature
539, 437–442 (2016).
This study shows that PI3Kγ is a key regulator of
TAM pro-tumoural activities and that its inhibition
contributes to TAM reprogramming and tumour
inhibition.
257. Ha, M. & Kim, V. N. Regulation of microRNA bio-
genesis. Nat. Rev. Mol. Cell Biol. 15, 509–524 (2014).
258. Baer, C. et al. Suppression of microRNA activity
amplifies IFN‑γ‑induced macrophage activation and
promotes anti-tumour immunity. Nat. Cell Biol. 18,
790–802 (2016).
This is the first study showing reprogramming of
TAMs to tumoricidal macrophages by the
modulation of miRNA activity.
259. Cannarile, M. A. et al. Colony-stimulating factor 1
receptor (CSF1R) inhibitors in cancer therapy.
J. Immunother. Cancer 5, 53 (2017).
260. Krieg, C. et al. High-dimensional single-cell analysis
predicts response to anti‑PD‑1 immunotherapy. Nat.
Med. 24, 144–153 (2018).
This study shows that the frequency of
CD14+CD16–HLA-DRhigh monocytes is a strong
predictor of response to PD1 immunotherapy in
patients with melanoma.
261. Romano, E. et al. Ipilimumab-dependent cell-mediated
cytotoxicity of regulatory T cells ex vivo by
nonclassical monocytes in melanoma patients. Proc.
Natl Acad. Sci. USA 112, 6140–6145 (2015).
262. Gil Del Alcazar, C. R. et al. Immune escape in breast
cancer during in situ to invasive carcinoma transition.
Cancer Discov. 7, 1098–1115 (2017).
263. Harrer, D. C., Dörrie, J. & Schaft, N. Chimeric antigen
receptors in different cell types: new vehicles join the
race. Hum. Gene Ther. 29, 547–558 (2018).
264. Kulkarni, A. et al. A designer self-assembled
supramolecule amplifies macrophage immune
responses against aggressive cancer. Nat. Biomed.
Eng. 2, 589–599 (2018).
In this paper, the authors design a supramolecule
that inhibits CSF1R signalling and blocks the
CD47–SIRPα pathway in macrophages; this
compound can reprogramme TAMs and exert
potent antitumour effects in aggressive models of
breast cancer and melanoma.
VOLUME 17 | DECEMBER 2018 | 903
REVIEWS

265. Gopalakrishnan, V. et al. Gut microbiome modulates
response to anti‑PD‑1 immunotherapy in melanoma
patients. Science 359, 97–103 (2018).
266. Iida, N. et al. Commensal bacteria control cancer
response to therapy by modulating the tumor
microenvironment. Science 342, 967–970 (2013).
267. Roy, S. & Trinchieri, G. Microbiota: a key orchestrator
of cancer therapy. Nat. Rev. Cancer 17, 271–285
(2017).
268. Veglia, F., Perego, M. & Gabrilovich, D. Myeloid-
derived suppressor cells coming of age. Nat. Immunol.
19, 108–119 (2018).
269. de Haas, N., de Koning, C., Spilgies, L., de Vries, I. J. &
Hato, S. V. Improving cancer immunotherapy by
targeting the STATe of MDSCs. Oncoimmunology 5,
e1196312 (2016).
270. Schulz, C. et al. A lineage of myeloid cells independent
of Myb and hematopoietic stem cells. Science 336,
86–90 (2012).
271. Hoeffel, G. et al. C‑Myb(+) erythro-myeloid
progenitor-derived fetal monocytes give rise to adult
tissue-resident macrophages. Immunity 42, 665–678
(2015).
272. Perdiguero, E. G. & Geissmann, F. The development
and maintenance of resident macrophages. Nat.
Immunol. 17, 2–8 (2016).
273. Lavin, Y., Mortha, A., Rahman, A. & Merad, M.
Regulation of macrophage development and function
in peripheral tissues. Nat. Rev. Immunol. 15, 731–744
(2015).
274. Ginhoux, F. & Jung, S. Monocytes and
macrophages: developmental pathways and tissue
homeostasis. Nat. Rev. Immunol. 14, 392–404
(2014).
275. Mass, E. et al. Specification of tissue-resident
macrophages during organogenesis. Science 353,
aaf4238 (2016).
276. Cecchini, M. G. et al. Role of colony stimulating
factor‑1 in the establishment and regulation of
tissue macrophages during postnatal development
of the mouse. Development 120, 1357–1372
(1994).
277. Okabe, Y. & Medzhitov, R. Tissue-specific signals
control reversible program of localization and
functional polarization of macrophages. Cell 157,
832–844 (2014).
278. [No authors listed]. Technology quarterly: targeting
tumours. Treating cancer: progress on many fronts.
The Economist (London) https://www.economist.com/
technology-quarterly/2017-09-16/treating-cancer (2017).
279. Gentles, A. J. et al. The prognostic landscape of genes
and infiltrating immune cells across human cancers.
Nat. Med. 21, 938–945 (2015).
Acknowledgements
The authors apologize to the many authors whose work they
could not cite owing to space constraints. Research in the
authors’ laboratories is supported by the Wellcome Trust
(101067/Z/13/Z) and Medical Research Council (MRC) Centre
grant MR/N022556/1 to J.W.P.
Author contributions
The authors contributed equally to all aspects of the article.
Competing interests
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

904 | DECEMBER 2018 | VOLUME 17 www.nature.com/nrd

©
2
0
1
8
S
p
r
i
n
g
e
r
N
a
t
u
r
e
L
i
m
i
t
e
d
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.