Role of the Microenvironment in Liver Metastasis: from Pre- to Prometastatic Niches Pnina Brodt

Published OnlineFirst October 19, 2016; DOI: 10.1158/1078-0432.CCR-16-0460

Review Clinical Cancer Research Role of the Microenvironment in Liver Metastasis: From Pre- to Prometastatic Niches Pnina Brodt

Abstract

Liver metastases remain a major barrier to successful manage- further enhance this intercellular communication network. Both ment of malignant disease, particularly for cancers of the gastro- resident and recruited cells can play opposing roles in the pro- intestinal tract but also for other malignancies, such as breast gression of metastasis, and the balance of these divergent effects carcinoma and melanoma. The ability of metastatic cells to determines whether the tumor cells will die, proliferate, and survive and proliferate in the liver is determined by the outcome colonize the new site or enter a state of dormancy. Moreover, of complex, reciprocal interactions between tumor cells and this delicate balance can be tilted in favor of metastasis, if factors different local resident subpopulations, including the sinusoidal produced by the primary tumor precondition the microenviron- endothelium, stellate, Kupffer, and inﬂammatory cells that are ment to form niches of activated resident cells that promote tumor mediated through cell–cell and cell–extracellular matrix adhesion expansion. This review aims to summarize current knowledge on and the release of soluble factors. Cross-communication between these diverse interactions and the impact they can have on the different hepatic resident cells in response to local tissue damage clinical management of hepatic metastases. Clin Cancer Res; 22(24); and inﬂammation and the recruitment of bone marrow cells 1–12. 2016 AACR.

Introduction entry into the liver (4, 5). Emerging evidence suggests that in addition, a premetastatic phase could set the stage for liver Cancer metastasis remains the major challenge to successful colonization by disseminating tumor cells. management of malignant disease. The liver is the main site of metastatic disease and a major cause of death from gastrointes- tinal malignancies, such as colon, gastric, and pancreatic carci- Evidence for premetastatic niche formation in the liver nomas as well as melanoma, breast cancer, and sarcomas (1). The term "premetastatic niche" was coined to describe a Circulating metastatic cells that enter the liver encounter microenvironment in a secondary organ site that has been unique cellular populations. These include the parenchymal rendered permissive to metastatic outgrowth in advance of hepatocytes and the nonparenchymal hepatocytes, including liver cancer cell entry through the activity of circulating factors sinusoidal endothelial cells (LSEC), hepatic stellate cells (HSC), released by the primary tumor (6–8). Although the dependency Kupffer cells (KC), dendritic cells, liver-associated lymphocytes, of metastasis on premetastatic niches remains controversial and and portal fibroblasts. Circulating and bone marrow–derived difficult to verify in the clinical setting (9, 10), it appears that a immune cells are also recruited to the liver in response to, and consensus is emerging on two scores: (i) that their existence may possibly in preparation for, invading tumor cells and can affect the have important implications for the clinical management of final outcome (reviewed in refs. 2, 3; see Table 1). metastatic disease, and (ii) that much can be learned from their This review summarizes our current understanding of the role interrogation about the cellular/molecular changes required to of the liver microenvironment in the growth of hepatic metastases facilitate tumor cell growth in a distant site, such as the liver. and the reciprocal interactions between metastatic tumor cells and Two recent studies have confirmed their potential relevance to different liver cell populations that regulate the process. liver metastasis. In a mouse model of metastatic pancreatic ductal adenocarcinoma (PDAC), Costa-Silva and colleagues Pre- and Prometastatic Niches Drive the showed that PDAC-derived exosomes taken up by hepatic KCs, Progression of Liver Metastasis upregulate TGFb production, leading to increased fibronectin production by HSCs and recruitment of bone marrow–derived The process of liver metastasis has previously been divided into macrophages. They identified macrophage migration inhibitory four major phases (detailed in Table 1) that follow tumor cell factor (MIF) as essential for premetastatic niche formation and metastasis. Moreover, in exosomes derived from patients with stage I PDAC who later developed liver metastasis, MIF levels Departments of Surgery, Medicine, and Oncology, McGill University and the were found to be higher than in patients whose tumors did not McGill University Health Centre, Montreal, Quebec, Canada. progress, thus identifying MIF as a potential predictor of PDAC Corresponding Author: Pnina Brodt, McGill University Health Centre, 1001 liver metastasis (11). This group also showed that exosomes Boulevard Decarie, Montreal, Quebec H4A 3J1, Canada. Phone: 514-934-1934, from human breast and pancreatic cancer cell lines that metas- ext. 36692; Fax: 514-938-7033; E-mail: [email protected] tasize selectively to the lung (MDA-MB-231) or liver (BxPC-3 doi: 10.1158/1078-0432.CCR-16-0460 and HPAF-II) fused preferentially with fibroblasts and epithelial 2016 American Association for Cancer Research. cells in the lung and KCs in the liver, and this was mediated by

www.aacrjournals.org OF1

Brodt

the exosomal integrin laminin receptors a6b1anda6b4 and the Cancer cells can escape these cytotoxic effects by forming fibronectin receptor avb5, respectively. In exosome-fused KCs, clusters with blood or other cancer cells that shield them from the proinflammatory factors S100P and S100A8 were upregu- the lethal effects of shear stress or NK-mediated cytotoxicity lated (Table 1; Fig. 1). Significantly, in PDAC patients, a corre- (reviewed in ref. 29). The release of inflammatory mediators, lation was documented between the levels of circulating, av- such as IL1b, TNFa, and IL18, can also initiate a cascade that bearing exosomes and disease stage, suggesting that exosomes facilitates their rapid exit from the vasculature to a less "toxic" may play a role clinically and their levels and integrin cargo may microenvironment (see below and Fig. 2). predict metastases in specific organs (12). Kowanetz and colleagues reported that tumor-derived GM-CSF could mobilize The sinusoidal endothelium actively participates in different þ þ Ly6C Ly6G myeloid cells into metastases, enabling niches in phases of metastasis the lung and liver, and also identified S100A8 as a driving factor Although the rapid inflammatory response initiated by cancer (13). Tumor-derived TIMP-1 was identified as another potential cell entry can lead to cell death, it may also have a tumor- inducer of increased liver susceptibility to metastasis, acting via protective effect by upregulating the expression of LSEC cell hepatic SDF-1 and neutrophil recruitment (14). Interestingly, adhesion molecules (CAM) and in this way, enhance cancer S100A8 was identified as a driver of liver premetastatic niche cell adhesion and transendothelial migration into the space of formation in several studies (12–15), suggesting that it may Disse, where they can escape the cytotoxic effects of KCs and have clinical utility as a predictor of liver metastasis. NK cells (30). Cancer cells can adhere to inflammation-induced The stage of primary tumor development at which premeta- E-selectin, VCAM-1, and ICAM-1, either as single cells or in static niches can be formed is difficult to assess in the clinical association with host cells, such as KCs or neutrophils (reviewed setting. Costa-Silva and colleagues found exosomal MIF upre- in ref. 31). Blockade of this inflammatory cascade was shown gulation in mice with pretumoral pancreatic lesions, and high to reduce liver metastasis (32–34), and TNFR1 signaling was plasma exosomal MIF levels were also detected in patients identified as a major driver of this process (35). with stage I PDAC (11), suggesting that they could be formed Ultimately, the outcome of these opposing effects on tumor cell at very early stages of cancer development. The potential survival and progression to the next phase depends on multiple contribution of circulating cancer cells (CTC) that may be factors, including the expression on the cancer cells of the respec- present early in tumor development and even after resection tive counter receptors, namely the E-selectin ligands sLewa and of the primary tumor (16, 17) is also unknown. sLewx and CD44 isoforms (36; reviewed in refs. 29, 37), and VCAM-1 and ICAM-1 counter receptors integrins a4b1 and LFA-1 Resident and Immune Cells Can Play and Mac-1 (38), respectively. E-selectin binding triggers a signal- Diverse and Opposing Roles in the ing cascade in both tumor cells and LSECs, leading to diapedesis and transendothelial migration (39) and altering gene expression Development of Liver Metastases in a tumor type–specific manner (40). Clinical studies have Host innate resistance mechanism can destroy tumor cells documented increased expression of E-selectin in and around prior to extravasation colorectal cancer liver metastases and elevated soluble E-selectin, Blood-borne cancer cells entering the liver first encounter the ICAM-1, and VCAM-1 levels that correlated with disease outcome LSECs, KCs, and hepatic natural killer (NK) cells (pit cells) that in patients with colorectal cancer (reviewed in ref. 23). together mount the first line of defense (18). Cancer cells LSECs may have other metastasis-promoting functions. LSEC- entrapped in the sinusoids may undergo destruction due to secreted fibronectin can induce epithelial–mesenchymal transi- mechanical stress and deformation-associated trauma or may die tion (EMT) in colorectal cancer cells via integrin a9b1, enhancing due to phagocytosis by KCs or cytolysis by NK cell–released ERK signaling and tumor invasion (41). Human LSECs were perforin/granzyme (18; reviewed in ref. 4). NO, ROS, and toxic shown to induce colorectal cancer migration and EMT via MIF radicals released by LSECs in response to local ischemia/reperfu- (42), thereby increasing their metastatic potential. sion can contribute to tumor cell death (19, 20). Release of NO LSECs contribute to tumor-induced angiogenesis. A recent and IFNg by LSECs and NK cells can result in upregulation of study identified Notch1 as a negative regulator of sinusoidal tumor cell Fas and apoptosis (18) that can be augmented by NK endothelial cells sprouting into micrometastases and thereby of and LSEC-derived TNFa (21). TNFa is one of the numerous angiogenesis and liver metastasis (43), raising questions about cytokines and chemokines unleashed by KCs in response to the advisability of Notch targeting for cancer therapy. Moreover, inflammation (Table 1; ref. 22). In addition to causing cell death liver metastasis of different carcinomas, including colorectal directly, some of these factors can mobilize and activate addi- cancer, can co-opt the sinusoidal endothelium at the tumor–liver tional innate immune cells, such as neutrophils, thereby adding to interface, giving rise to a histologic growth pattern termed the the local tumoricidal capacity (reviewed in ref. 23). In several "replacement" or "sinusoidal" growth pattern that seems to result animal tumor models, including colon cancer, loss of NK cells from tumor cell invasion between LSECs and the matrix in the increased cancer cell growth in the liver, whereas enhanced NK space of Disse and is characterized by the formation of intrameta- activity reduced liver metastasis (24–26). Indirect evidence also static vessels that appear continuous with the sinusoidal vessels suggests that susceptibility to NK-mediated immune attack can (reviewed in ref. 23). The factors that drive this co-opting mech- affect the ability of human cancer cells to generate liver metastases anism remain to be identified, but a recent study by Im and (27). Circulating neutrophils can also release various factors colleagues suggests that Ang2 may play a regulatory role (44). abundant in their granules to kill tumor cells (see Table 1), and Collectively, the evidence shows that LSECs are not a passive their cytokines and chemokines can activate the tumoricidal barrier to tumor cell extravasation but participate actively in the potential of resident KCs and recruit host immune T cells with metastatic process. Cancer cell interaction with LSECs can recip- antitumorigenic activities (reviewed in ref. 28). rocally alter the phenotypes of both cell types, and this may lead to

OF2 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

The Microenvironment in Liver Metastasis

Table 1. Cells and mediators involved in the different phases of liver metastasis Phase of the metastatic Soluble factors involved process Cell type involved Cell surface markers Recruiting or activating Produced References Premetastatic niche KCs CD11b, F4/80 M-CSF, GM-CSF, MIF TGFb, S100P, S100A8 (11–14) HSCs a-SMA, desmin, GFAP KC-derived TGFb and TNFa Fibronectin (11, 13) MDSCs CD11b, Ly6C, Ly6G Tumor-derived GM-CSF S100A8 (12, 15) Neutrophils CD11b, Ly6G TIMP-1, SDF-1 S100A8 (7, 14, 67) Post-tumor invasion (1) Microvascular phase: LSECs CD31, TNFR1, TNFR2, avb3, TNFa,IL1b, IL18 ROS, NO, IFNg, TNFa (5, 18–20, 23, 31) Tumor cells trapped in E-selectin, VCAM-1, ICAM-1 the vasculature KCs CD11b, F4/80 M-CSF, GM-CSF, IFNg TNFa, IL1, IL6, IL8, IFNg, (5, 23, 31, 45–47, 51) MIP-1a, MIP-2, IP-10, MCP-1, CCL5, VEGF Neutrophils CD11b, Ly6G S100A8, S100A9, IL8, CCL2 TNFa, ROS, defensins, (28, 31, 64) perforins, elastase Hepatic NK cells NK1.1, NK1.2, FcgRIII CXCL9, CXCL10, IL1, IL12, Grz, Perforin, IFNg, (5, 31, 48) IL15, IL18 TNFa (2) Extravascular, HSCs a-SMA, desmin, GFAP KC-derived TGFb, PDGF, ECM deposition, MCP-1, (74–77, 84) preangiogenic phase: bFGF, SDF-1, IGF-I, CCL2 CCL5, CCL21, TGFb Tumor cells transmigrate Neutrophils CD11b, Ly6G; N1: ICAM-1; S100A8, S100A9, CXCL1, TNFa, ROS, NOS, (28, 60, 64, 71) into the space of Disse, N2: CXCR4 CXCL2, CXCL5, TNFa, MMP-8, MMP-9, activating a local stromal IL8, G-CSF, IFNg, CCL2, elastase, cathepsin G, response CCL5, TGFb VEGF KCs CD11b, F4/80 M-CSF, GM-CSF, IFNg IL6, VEGF, HGF, (5, 50, 53) MMP-9, MMP-14 (3) Angiogenic phase: Blood-derived CD11b, F4/80, CD68, CD163, TNFa, MCP-1 VEGF, MMPs, bFGF (57, 60) Micrometastases are monocytes CD206 vascularized HSCs a-SMA, desmin, GFAP KC-derived TGFb, PDGF, VEGF, angiopoietin-1, (74, 79–82, 88) bFGF, SDF-1, IGF-I, CCL2 MMP-2, -9, and -13 LSECs CD31, TNFR1, ICAM-1 VEGF, Ang1, inhibited by Fibronectin, MIF (23, 41–44) Notch1 and regulated by Ang2 Macrophages (M2) CD11b, F4/80, Ly6C, CD163 TGFb, IL10 NO, TNFa, IFN, VEGF, (57–61) EGF, bFGF, TGFb, IL10 (4) Growth phase: Hepatocytes Albumin, tyrosine Adhesion to tight junction AREG, EGF, HB-EGF, (92–98, 101–105, Metastases expansion aminotransferase integral proteins ErB2, IGF-I, HGFL, 107) ErB3, bFGF Tregs CD4, CD25, Foxp3 TGFb, IL10, CCL5, CCL22 VEGF, TGFb, IL10 (35, 111, 124) NOTE: Shown are the four phases of the metastatic process that follow tumor cell invasion into the liver. Listed are the hepatic cells known to be involved at each phase, the mediators regulating their recruitment and activation, and the factors they produce to block or promote the metastatic process. The process has been simplified and divided into distinct phases in the interest of clarity. However, it should be viewed as dynamic with overlapping cellular and molecular drivers at different phases. Moreover, as the metastatic process advances, the cancer and hepatic cells evolve and their phenotypes changeasa consequence of their interactions. These newly acquired properties help propel the process forward (see also Figs. 1–3). Abbreviations: ECM, extracellular matrix; MDSC, myeloid-derived suppressor cell; MMP, matrix metalloproteinase, NK, natural killer; Treg, regulatory T cell. intravascular tumor cell destruction but can also promote metas- tumor cell phagocytosis or apoptosis, within hours of tumor cell tasis through enhanced tumor cell migration and increased entry (47). This tumoricidal effect may require cooperation with angiogenesis (Table 1; Figs. 2 and 3). local or recruited NK cells (48), and its efficiency in removing cancer cells likely depends on the tumor load (49). An intravital The role of macrophages is context dependent microscopy study (50) recently revealed that 74% of intra-arte- KCs constitute approximately 10% of all liver cells. Recent rially injected rat colon cancer cells adhered to the KCs within studies suggest that they may be established prenatally and 6 hours, and this was associated with increased liver TNFa and maintained into adulthood, independently of replenishment IL1b levels, consistent with our own findings (51, 52) and from blood monocytes but in an M-CSF and GM-CSF–dependent indicative of KC activation. Extensive phagocytosis of tumor manner (45). They reside mainly in the hepatic sinusoids and cells was observed within 2 hours postinjection. Interestingly, are anchored to the endothelium by long cytoplasmic processes although elimination of KCs 2 days prior to tumor injection (46; reviewed in ref. 47). In their interaction with invading can- increased liver metastasis, it had no significant effect up to 1 week cer cells, KCs can play diverse and opposing roles, depending after tumor injection, suggesting that KCs exerted their most on factors such as the stage of the metastatic process, tumor potent antitumor effect within the first 24 hours of tumor cell load, and interactions with other immune cells. As discussed, entry into the liver (50). A bimodal effect of KC depletion was also KCs can promote metastasis by orchestrating the premetastatic documented in another study of murine colorectal cancer, and it niche. However, tumor–KC interactions can have opposing was attributed to decreased VEGF and increased iNOS levels in effects. Several studies have documented a rapid adhesion of livers subjected to late-stage KC depletion (53). Indeed, tumor tumor cells to KCs in the sinusoidal lumen, resulting mainly in cells that survive the initial tumoricidal assault by KCs can benefit

www.aacrjournals.org Clin Cancer Res; 22(24) December 15, 2016 OF3

Brodt

αβ ? ABTumor-derived exosomes

MIF

TGFβ S100A8 aKC G-CSF Exosomes PDAC

LSEC Tumor-derived G-CSF Tumor-derived Vessel lumen exosomes BM-derived myeloid cells

TGFβ KC 1

HSC S100A8 S100A8 TGFβ 3 Neutrophil 2 Space of Disse

Potential therapeutic strategies: aHSC 1. Enhancing KC tumoricidal Ly6C+Ly6G+ activities Fibronectin myeloid cells 2. Inducing HSC apoptosis 3. Inhibition of inﬂammatory mediators Hepatocyte

Figure 1. The premetastatic niche in the liver. Shown is a diagrammatic representation of the multistep process involved in the formation of premetastatic niches in the liver based on data described in refs. 11, 12. The symbols used for depiction of different hepatic cells are listed in Fig. 3. A and B, A diagrammatic representation of a PDAC is shown in inset A, and an enlarged image of a Kupffer cell activated by PDAC-derived exosomes is shown in inset B. Encircled numbers denote cellular and molecular interactions that can potentially be targeted for liver metastasis prevention, as detailed in the right- hand corner box. BM, bone marrow.

from their protumorigenic functions. Adhesion to KC can facil- muramyl dipeptide may be most effective if achieved before itate tumor cell extravasation, among others, by sequential acti- tumor cells enter the liver in large numbers (47, 55). The contri- vation of endothelial CAMs (29, 54). In addition, KCs can bution of CTCs that may be present before liver micrometastases produce cytokines and growth factors, including IL6, hepatocyte are established or after surgical resection of the primary tumor growth factor (HGF), and VEGF, and matrix metalloproteinases (16, 17) to the tumor load and KC activation is unknown. This (MMP), such as MMP-9 and MMP-14, that can accelerate tumor multifaceted role of KCs in liver metastasis may explain the cell invasion into and within the parenchymal space as well as limited success to date of KC-targeting interventions (47). promote tumor cell proliferation and angiogenesis, thereby Importantly, in addition to KCs, blood-derived monocytes enhancing liver metastasis (see Table 1; Figs. 2 and 3). can also be recruited to the tumor site in response to local Taken together, the evidence suggests that KC targeting could injury and inflammation and differentiate locally into mature become an effective antimetastatic strategy only within a very CD11bloF4/80hi macrophages in a CCR2-dependent manner narrow window. Limiting KC activity may be beneficial if it could (56). These macrophages can also trigger HSC activation and prevent premetastatic niche formation, whereas enhancing the KC fibrogenesis (57), a process important in the early stages of tumoricidal potential using agents such as IFNg, GM-CSF, or extravascular tumor expansion (see below).

OF4 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

The Microenvironment in Liver Metastasis Vessel lumen Fas/ LSEC I. Microvascular phase FasL 1 α 2 Potential therapeutic strategies: Tumor cell TNF sLew×/ 1. TNF inhibition aKC E-Sel. 2. E-selectin blockade

TNFR1 TGFβ VEGF MMP-9 HSC

VEGF-R TNFα aHSC M1 III. Angiogenic phase 7 Potential therapeutic strategies: 7. Blocking M2 polarization 3 TGFβ IL10 8. VEGF inhibition Space of Disse 9. N1 induction M2 TGFβ 8 N1 4 CoIIV Arg MMP-9 β MDSC VEGF TGF 9 II. Preangiogenic phase ROS CD8+ N2 TNFα Potential therapeutic strategies: T cell 3. TGFbRII blockade NO 6 5 4. MDSC depletion TNFα 5. TNFR2 blockade Treg 6. PD-1/PD-L1 blockade MMPs VEGF TGFβ aHSC IGF-I Hepatocyte IGF Fas/ FasL 10 IV. Growth phase Parenchyma 11 Potential therapeutic strategies: 10. IGF-I neutralization 11. Claudin-2 blockade

Figure 2. The multistep process of liver colonization by disseminated cancer cells. Shown is a diagrammatic representation of the major cell types, soluble factors, and extracellular matrix (ECM) components in the liver microenvironment that impact the progression of liver metastasis. The four major phases in the process are delineated by dashed lines. Top, major intercellular interactions that occur upon entry of the tumor cells into the sinusoidal vessels (the microvascular phase) and precede tumor extravasation; middle, events following tumor cell extravasation into the space of Disse, divided into the preangiogenic phase (middle left) that is orchestrated by stellate cell activation, ECM deposition, cytokine production, and the recruitment of innate and adaptive immune cells and culminates in neovascularization (the vascular phase; middle right); bottom, these interactions enable tumor expansion (the growth phase). The symbols used for depiction of different hepatic cells are listed in Fig. 3 or indicated in the diagram. The encircled numbers denote potential targets for therapeutic intervention, as detailed in the boxes inserted within each of the depicted phases. MDSC, myeloid-derived suppressor cell; MMP, matrix metalloproteinase; Treg, regulatory T cell.

Macrophages have inherent plasticity, and their phenotypes and IL10 (Table 1; ref. 62). Evidence regarding the role of can change within a spectrum of activation states between the macrophage polarization in liver metastasis is scant, because polar M1 and M2 phenotypes (58, 59). M1 macrophages are macrophages within or around hepatic metastases have, with few tumoricidal due to high NO and TNFa production levels and exceptions, not been subtyped. A recent clinical study identified produce Th1, Th17, and the NK-attracting chemokines CXCL9 the M2/M1 ratio in hepatic resections as a correlate of colorectal and CXCL10 (60). In contrast, M2 macrophages can promote cancer metastasis (63), implicating M2 macrophages in the clin- tumor growth through production of growth factors, such as ical disease. Furthermore, F4/80, the cell surface marker frequent- VEGF, EGF, bFGF, and TGFb (61). In addition, M1 macrophages ly used to identify KCs, is also expressed on recruited monocytes, activate Th1-type immune responses that can further amplify and strategies used to eliminate macrophages in vivo are not KC M1/killer-type activity through the production of IFNg, whereas specific. The source and identities of macrophages in many M2 macrophages induce regulatory T cells (Treg) and thereby an published studies remain therefore to be verified. As immune immunotolerant microenvironment through release of TGFb editing of the tumor microenvironment and the conversion of

www.aacrjournals.org Clin Cancer Res; 22(24) December 15, 2016 OF5

Brodt

Cell type Antimetastatic effectsPrometastatic effects References

Liver sinusoid Produce ROS, NO, IFNγ, Express CAM, protect tumor (5, 18–20, 31, 33, 41) endothelial cells TNFα cells from apoptosis, facilitate (LSEC) transmigration, participate in angiogenesis, induce EMT

Hepatic stellate cells (HSC) Participate in premetastatic (11, 23, 77, 82–86) niche formation, drive ﬁbrogenic response, ECM production, mediate angiogenesis, recruit inﬂammatory cells, produce MMPs Resting Activated

Hepatocytes Mediate tumor adhesion, produce (93–96, 101, 103, 105) growth factors (e.g., IGF-I, HGFL), alter tumor transcriptome, undergo EMT to provide stromal support, enhance angiogenesis via bFGF

Kupffer cells Phagocytose tumor cells, cause Participate in premetastatic (5, 11, 12, 47–53) (KC) apoptosis (early in the process), niche formation; activate LSEC produce TNFα, mobilize CAM; produce VEGF, IL6, neutrophils and NK cells HGF, and MMP-9; activate HSC

Macrophages M1 macrophages produce NO, M2 macrophages produce (59–63) TNFα, and T-cell– and NK- VEGF, EGF, and bFGF; induce recruiting chemokines Tregs via IL10 and TGFβ

Myeloid-derived Induce Tregs, block T-cell (113, 117, 123, 124) suppressor cells expansion, suppress cytotoxic (MDSC) T-cell function, produce ROS, produce arginase

Natural killer (NK) cells Produce perforin and TNFα, (4) induce Fas/FasL interaction, kill tumor cells

Neutrophils N1 neutrophils produce ROS, Participate in premetastatic (28, 64, 77) NO, TNFα, and defensins; recruit niche formation, enhance CD8+ T cells transendothelial migration, entrap tumor cells; N2 neutrophils produce VEGF, MMP-9, and CCL5

T lymphocytes Cytotoxic T cells kill tumor Tregs inhibit effector T-cell (109, 111) cells, release perforin and expansion, eliminate cytotoxic TNFα, recruit macrophages T cells, subvert immune activation, produce VEGF

Figure 3. Parenchymal, nonparenchymal, and immune cells of the liver and their role in metastasis. Listed are the parenchymal and nonparenchymal cells of the liver and their anti- and prometastatic functions. Symbols shown for each cell type were used to depict intercellular communications in Figs. 1 and 2. EMT, epithelial- mesenchymal transition; Treg, regulatory T cell.

OF6 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

The Microenvironment in Liver Metastasis

tumor-associated (TAM) M2 macrophages to M1 killer macro- HSCs orchestrate a prometastatic microenvironment that is phages are emerging as potentially relevant anticancer strategies required for transition from the avascular to the vascular (62), further characterization of both the source and the status stage of metastasis of macrophage populations involved in liver metastasis will HSCs orchestrate the characteristic fibrogenic response of the become critical. liver to injury. Normally quiescent in the space of Disse (74), they are activated (aHSC) in response to liver damage and The neutrophils: A double-edged sword inflammatory stimuli, acquire a myofibroblast-like phenotype þ Neutrophils are part of the innate immune response to patho- (a-SMA ), and produce ECM rich in collagen I and IV (74, 75). gens and are also rapidly activated in response to invading cancer Chemokines and cytokines released by aHSCs (Table 1) also cells (28). Bone marrow–derived neutrophils are mobilized to recruit inflammatory/immune cells (76, 77), thereby shaping sites of inflammation or cancer via their cell-surface receptor the immune microenvironment. CXCR2 and in response to chemokines and cytokines (Table 1) In addition to their role in premetastatic niche formation secreted by activated macrophages, endothelial cells, or the tumor (above), HSCs can orchestrate a prometastatic niche following cells (reviewed in ref. 28). At the tumor site, the neutrophils can tumor extravasation. In response to growth factors and inflam- exert opposing effects that depend on the inflammatory/immune matory mediators (detailed in Table 1), HSCs are activated (74) context (reviewed in ref. 64). They can release cytolytic factors and can trigger a process akin to the early events in liver repair. (detailed in Table 1) or inhibit tumor growth indirectly via This was observed in animal models (4) and is also evidenced þ recruitment of CD8 cytotoxic T cells and macrophages. This by the increased production of collagen IV in and around may require direct contact with the tumor cells (64), such as can hepatic metastases in clinical specimens (78). Macrophages, occur within the liver sinusoidal lumen (65, 66). hepatocytes, and LSECs contribute to this process by releasing On the other hand, neutrophils can be mobilized into pre- TGFb and/or TNFa (57). In turn, aHSC-derived proangiogenic metastatic niches in the liver in response to S100A8 and S100A9 factors, such as VEGF and angiopoietin-1 (79, 80), initiate (7) and were shown to contribute to niche formation in an angiogenesis (3), and this is enhanced by HSC-derived MMPs orthotropic colon carcinoma model (67). Moreover, in the vas- that facilitate endothelial cell migration and tumor invasion cular lumen, the physical interaction with tumor cells that (23, 74; reviewed in refs. 5, 81; Table 1; Fig. 2). could lead to tumor cell kill may, alternatively, anchor circulating Several lines of evidence confirm the essential role of HSCs in tumor cells to the vascular endothelium and enhance their migra- liver metastasis. Olaso and colleagues showed that myofibro- tion into the extravascular space. Tumor-derived cytokines, such blasts that infiltrated avascular micrometastases of B16 mela- as IL8, induce expression of integrins, such as CD11b/CD18, on noma cells induced a stromal response favorable to angio- the neutrophils, increasing their adhesion to tumor cells via genesis, which preceded endothelial cell recruitment and was the counter receptor ICAM-1 and facilitating transendothelial followed by colocalization of myofibroblasts and endothelial migration, as shown in murine melanoma and lung carcinoma cells within angiogenic structures (82, 83). HSC and macro- models (66, 68). Tumor cell entrapment in the sinusoids can phage recruitment into metastatic sites and subsequent angio- also be mediated by neutrophil-produced extracellular (DNA) genesis were shown to be CCR2/CCL2 dependent (84). More traps, resulting in increased tumor retention in the sinusoids, recently, Eveno and colleagues (85), using immunofluores- enhanced tumor cell adhesion, proliferation, migration and inva- cence, showed that 9 days postinjection of colorectal cancer sion, and increased liver metastasis (69, 70). Neutrophils can also LS174 cells into SCID mice, the hepatic micrometastases con- release extracellular matrix (ECM)–degrading proteinases, such sisted of proliferating cancer cells, a well-organized network of as MMP-8, MMP-9, elastase, and cathepsin G, thereby increasing aHSC and laminin deposits, but no vascular network. As the tumor invasion (64), and can promote angiogenesis through the liver metastases grew, an organized vascular network appeared þ release of VEGF (Table 1; Figs. 2 and 3). and laminin colocalized with CD31 endothelial cells. Co- Moreover, a TGFb-mediated neutrophil polarization that injection of tumor cells and aHSCs enhanced metastasis. More- alters the phenotype of tumor-associated neutrophils from over, analysis of liver metastasis from patients with colorectal tumor inhibitory (N1) to tumor promoting (N2) was recently cancer revealed a surface-marker expression pattern similar to described (71). TGFb produced by M2 macrophages may con- that observed in the coinjection studies. tribute to this process (reviewed in ref. 59). However, the contri- Recently, suppression of TGFb receptor II signaling by bution of polarized neutrophils to liver metastasis remains to QGAP1 was shown to prevent HSC activation, and IQGAP1 be confirmed. downregulation was observed in myofibroblasts associated Clinical studies implicate neutrophils in tumor promotion. with human colorectal cancer liver metastases, consistent with þ With few exceptions, elevated circulating neutrophil counts or their activation in this context (86). Furthermore, a-SMA cells, neutrophil-to-lymphocyte ratios were associated with poorer whose presence correlated with the degree of fibrous stroma, outcomes and distant metastases in patients with various were identified in liver metastases of colon, gastric, and pan- epithelial malignancies (72), including carcinomas of the gas- creatic adenocarcinomas (87). trointestinal tract. High neutrophil counts may also be associ- The importance of HSCs to metastatic niche formation and ated with increased resistance to therapy (64, 72). The 5-year their role in primary liver cancer (88) have raised interest in survival for patients with colorectal cancer undergoing hepatic HSC targeting as an anticancer/antimetastatic strategy. There is resections that had neutrophil-to-lymphocyte ratios greater little direct evidence that such an approach can succeed, due than 5 was worse than for those with normal ratios (73). partially to the present lack of agents that specifically target However, high neutrophil infiltration into the tumor site may HSCs without toxicity. Indirect evidence from a rat model of be a consequence of advanced tumor stage (and, therefore, cholangiocarcinoma (89) and pancreatic stellate cell targeting poorer prognosis) rather than its cause. in a PDAC model (81) suggests a potential therapeutic benefit,

www.aacrjournals.org Clin Cancer Res; 22(24) December 15, 2016 OF7

Brodt

but this remains to be verified. Of note, portal fibroblasts and The role of Tregs. Treg cells may be thymically derived (natural, þ bone marrow–derived fibrocytes (90) can also contribute to the nTregs) or induced from na€ve CD4 T-cell precursors (iTreg) stromal response. Liver-invading cancer cells located in portal in response to IL10 and TGFb. They exert an immunosup- tracts and unable to activate HSCs may engage portal tract pressive effect by different mechanisms that result in the fibroblasts that produce IL8, a chemokine involved in invasion inhibition of an effective, antitumorigenic T-cell response and angiogenesis (91). Specific targeting of HSCs may, there- (reviewed in ref. 111). Although high Treg levels were docu- fore, not be sufficient to deprive metastatic cells of a growth- mented in the blood and local tumor sites of patients with promoting stroma. various epithelial cancers (e.g., ref. 112), relatively little is known about their role in liver metastasis. Connolly and Hepatocytes can promote metastasis directly and indirectly colleagues, using models of early, preinvasive pancreatic The role of the parenchymal hepatocytes in liver metastasis is neoplasia and advanced colorectal cancer showed that Tregs not well understood. Tumor cell adhesion to hepatocytes was accelerated the development of liver metastases (113). We identified as one of the earliest events in liver metastasis for- recently documented the accumulation of Tregs around colon mation and a correlate of the metastatic potential (92, 93). carcinoma MC38 liver micrometastases and showed that this Desmosomes (94), integrins av, a6, and b1thatbindto was TNFR2 dependent (35). hepatocyte ECM (93), osteopontin binding via counter recep- Intriguingly, in 57 colon cancer specimens from patients under- tors CD44 and integrin av (95), and claudins (96, 97) were all going chemotherapy or chemoimmunotherapy, higher Treg infil- implicated. Adhesion of human colorectal cancer cells to hepa- tration scores were associated with a better prognosis and a better tocyte-derived ECM was shown to upregulate the expression of outcome (114). However, the relationship between Treg accu- genes involved in tumor cell survival, motility, and prolifera- mulation in the primary tumors and in the corresponding liver tion, in particular EGF family members implicated in liver metastases has not been examined. Although Treg-targeting metastasis (98) such as AREG, EGFR, HBEGF,anderbB2 and strategies are in development (115, 116; reviewed in ref. 110), stem cell markers CD133 (PROM1)andLGR5 (99), suggesting their potential therapeutic benefit for liver metastases has not that colorectal cancer adhesion to hepatocyte ECM induces yet been examined. autocrine growth-promoting mechanisms for tumor expansion. Tumor cell adhesion to hepatocytes can have other conse- The role of MDSCs. MDSCs are a heterogeneous population of quences. The FasL/FLICE-like inhibitory protein on murine myeloid cells that can be of the monocytic (Mo-MDSC, þ þ þ þ MC38 cells was shown to induce apoptosis in hepatocytes by CD11b Ly6C ) or granulocytic (G-MDSC, CD11b Ly6G ) þ þ activating Fas signaling, and this destructive process created a lineages (117). In humans, MDSCs are CD33 and/or CD11b niche for tumor expansion (100). Furthermore, hepatocytes and HLA-DR (118). Under normal physiologic conditions, bone produce several growth factors, including IGF-I (101). As we marrow–derived immature myeloid cells differentiate into have shown, blockade of IGF-I signaling in metastatic tumor mature granulocytes or monocytes, able to mediate host innate cells could inhibit liver metastasis (102–104). Other factors immune responses in the target tissue. However, in the tumor include the HGF-like protein/macrophage-stimulating protein microenvironment, these precursors do not mature and exert (HGFL) that can enhance tumor growth, motility, and inva- immunosuppressive and tumor-promoting effects instead sion via the Ron receptor (105); heregulin, a ligand of ErbB3 (119). Although they express granulocyte and monocyte surface shown to enhance integrin avb5-mediated cell migration and markers, these cells have been characterized on the basis of erbB3/erbB2 signaling in human colorectal cancer cells (106); functional criteria, namely their immunosuppressive capabilities and the proangiogenic factor bFGF (107). (117, 120). An increase in circulating MDSCs in cancer patients The clinical relevance of tumor–hepatocyte interactions is and their recruitment into tumor sites have been documented not clear. Three different histologic growth patterns have been (117, 121, 122). identified in liver metastases specimens, namely the desmo- In the liver, MDSCs can be recruited to the metastases plastic, pushing, and replacement growth patterns. The latter through chemokines, such as CXCL1 and CXCL2, produced two are characterized by close tumor–hepatocyte proximity by LSECs and KCs or by activated HSCs (77). There, they can (23). Whether or not direct adhesion between metastatic promote tumor growth by inducing immunosuppression cancer cells and hepatocytes influences the eventual growth (123), producing arginase (117), and increasing Treg numbers pattern of colorectal cancer liver metastases is not known. (Table 1; Fig. 2; ref. 124). Clinical and experimental evidence Blocking tumor–hepatocyte interactions could have beneficial confirms their role in metastasis (125). Recruitment of tumor- antimetastatic effects, as recently demonstrated by Tabaries promoting myeloid cells into colorectal cancer liver metastases and colleagues (108). has recently been documented, and their depletion was shown to result in a marked reduction in liver metastasis (126). We An immunosuppressive microenvironment facilitates recently reported that MDSCs accumulate in hepatic microme- metastatic expansion tastases of MC38 cells within days of tumor cell injection high Tumor cells can activate specific T-cell–mediated immune and documented a relative preponderance of Ly6C cells responses that curtail tumor expansion through various mechan- (the more suppressive, arginase-high subtype; ref. 127). We þ þ isms (reviewed in ref. 109). However, tumor cells can evade also identified CD33 HLA-DR TNFR2 myeloid cells in the T-cell–mediated killing, among others, by giving rise to and/or periphery of hepatic metastases from patients with colorectal recruiting immunosuppressive cells, such as myeloid-derived cancer, implicating MDSCs in the clinical disease (35). suppressor cells (MDSC) and regulatory T cells (Treg) that alter Several strategies are being developed to selectively elim- the immune landscape, inducing a state of immune tolerance that inate MDSCs, including pharmacologic induction of MDSC dif- is permissive to tumor expansion (reviewed in ref. 110). ferentiation, inhibition of MDSC expansion from bone marrow

OF8 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

The Microenvironment in Liver Metastasis

precursors, or blockade of their function (128). However, some of interrogation, our understanding of the relationship between the drugs developed are not MDSC specific, or their long-term the genetic background and clinical history of the patient, the administration is not possible due to toxicity (128). Their trans- genomic/transcriptomic profileofthecancercells,andthetype lational potential remains therefore to be verified. of response mounted by the microenvironment may improve, opening new avenues for prevention and/or treatment of liver Conclusions metastatic disease. Cancer cells entering the liver encounter a unique and complex fl microenvironment. Parenchymal and nonparenchymal liver Disclosure of Potential Con icts of Interest fl cells, as well as recruited inflammatory and immune cells, par- No potential con icts of interest were disclosed. ticipate in the response to invading tumor cells and may inhibit or favor the progression of metastasis. The factors that determine the Acknowledgments outcome of these opposing effects and the pattern of growth of the The author is indebted to Dr. Janusz Rak (McGill University Health Centre) for insightful editorial comments and to Simon Milette for the superb artistic metastases are not yet well understood. This duality of function rendering of the metastatic niches of the liver and for his help with the table. and the shifts in the types of cells and mediators involved at different stages of the metastatic process render the attempts to Grant Support target specific cellular interactions in the microenvironment This work was made possible by support from the Canadian Institute for extremely challenging and may explain the slow progress to date Health Research and by a PSR-SIIRI-843grant from the Quebec Ministere de fi in identifying agents that can successfully and speci cally l'Economie, de l'Innovation et des Exportations. inhibit metastatic disease. With the advent of genomic profiling for personalized cancer treatment and the increased availability Received February 21, 2016; revised September 19, 2016; accepted September of surgically resected liver metastases for cellular/molecular 22, 2016; published OnlineFirst October 19, 2016.

References 1. Hess KR, Varadhachary GR, Taylor SH, Wei W, Raber MN, Lenzi R, the liver through SDF-1/CXCR4-dependent neutrophil recruitment in et al. Metastatic patterns in adenocarcinoma. Cancer 2006;106: mice. Hepatology 2015;61:238–48. 1624–33. 15. Zhang Y, Davis C, Ryan J, Janney C, Pena~ MMO. Development and 2. Kmiec Z. Cooperation of liver cells in health and disease. Adv Anat characterization of a reliable mouse model of colorectal cancer metastasis Embryol Cell Biol 2001;161:III–XIII. to the liver. Clin Exp Metastasis 2013;30:903–18. 3. Smedsrod B, Le Couteur D, Ikejima K, Jaeschke H, Kawada N, Naito M, 16. Bork U, Rahbari NN, Scholch S, Reissfelder C, Kahlert C, Buchler et al. Hepatic sinusoidal cells in health and disease: update from the 14th MW, et al. Circulating tumour cells and outcome in non-met- International Symposium. Liver Int 2009;29:490–501. astatic colorectal cancer: a prospective study. Br J Cancer 2015; 4. Vidal-Vanaclocha F. The prometastatic microenvironment of the liver. 112:1306–13. Cancer Microenviron 2008;1:113–29. 17. Tsai WS, Chen JS, Shao HJ, Wu JC, Lai JM, Lu SH, et al. Circulating tumor 5. Vidal-Vanaclocha F. The tumor microenvironment at different stages of cell count correlates with colorectal neoplasm progression and is a hepatic metastasis. In: Brodt P, editor. Liver metastasis: biology and prognostic marker for distant metastasis in non-metastatic patients. clinical management. Dordrecht, the Netherlands: Springer Scien- Sci Rep 2016;6:24517. ceþBusiness Media B.V.; 2011. p. 1. 18. Braet F, Nagatsuma K, Saito M, Soon L, Wisse E, Matsuura T. The hepatic 6. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, et al. sinusoidal endothelial lining and colorectal liver metastases. World VEGFR1-positive haematopoietic bone marrow progenitors initiate the J Gastroenterol 2007;13:821–5. pre-metastatic niche. Nature 2005;438:820–7. 19. Wang HH, McIntosh AR, Hasinoff BB, Rector ES, Ahmed N, Nance DM, 7. Hiratsuka S, Watanabe A, Aburatani H, Maru Y. Tumour-medi- et al. B16 melanoma cell arrest in the mouse liver induces nitric oxide ated upregulation of chemoattractants and recruitment of mye- release and sinusoidal cytotoxicity: a natural hepatic defense against loid cells predetermines lung metastasis. Nat Cell Biol 2006;8: metastasis. Cancer Res 2000;60:5862–9. 1369–75. 20. Yanagida H, Kaibori M, Yoshida H, Habara K, Yamada M, Kamiyama Y, 8. Hiratsuka S, Watanabe A, Sakurai Y, Akashi-Takamura S, Ishibashi S, et al. Hepatic ischemia/reperfusion upregulates the susceptibility of Miyake K, et al. The S100A8-serum amyloid A3-TLR4 paracrine hepatocytes to confer the induction of inducible nitric oxide synthase cascade establishes a pre-metastatic phase. Nat Cell Biol 2008;10: gene expression. Shock 2006;26:162–8. 1349–55. 21. Hehlgans T, Pfeffer K. The intriguing biology of the tumour necrosis 9. Psaila B, Lyden D. The metastatic niche: adapting the foreign soil. Nat Rev factor/tumour necrosis factor receptor superfamily: players, rules and the Cancer 2009;9:285–93. games. Immunology 2005;115:1–20. 10. Sceneay J, Smyth MJ, Moller A. The pre-metastatic niche: finding common 22. Ramadori G, Moriconi F, Malik I, Dudas J. Physiology and pathophys- ground. Cancer Metastasis Rev 2013;32:449–64. iology of liver inflammation, damage and repair. J Physiol Pharmacol 11. Costa-Silva B, Aiello NM, Ocean AJ, Singh S, Zhang H, Thakur BK, et al. 2008;59 Suppl 1:107–17. Pancreatic cancer exosomes initiate pre-metastatic niche formation in 23. Van den Eynden GG, Majeed AW, Illemann M, Vermeulen PB, Bird NC, the liver. Nat Cell Biol 2015;17:816–26. Hoyer-Hansen G, et al. The multifaceted role of the microenvironment in 12. Hoshino A, Costa-Silva B, Shen TL, Rodrigues G, Hashimoto A, Tesic liver metastasis: biology and clinical implications. Cancer Res 2013;73: Mark M, et al. Tumour exosome integrins determine organotropic 2031–43. metastasis. Nature 2015;527:329–35. 24. Bertin S, Neves S, Gavelli A, Baque P, Brossette N, Simoes S, et al. 13. Kowanetz M, Wu X, Lee J, Tan M, Hagenbeek T, Qu X, et al. Granu- Cellular and molecular events associated with the antitumor response locyte-colony stimulating factor promotes lung metastasis through induced by the cytosine deaminase/5-fluorocytosine suicide gene mobilization of Ly6GþLy6Cþ granulocytes. Proc Natl Acad Sci therapy system in a rat liver metastasis model. Cancer Gene Ther U S A 2010;107:21248–55. 2007;14:858–66. 14. Seubert B, Grunwald B, Kobuch J, Cui H, Schelter F, Schaten S, et al. Tissue 25. Takehara T, Uemura A, Tatsumi T, Suzuki T, Kimura R, Shiotani A, et al. inhibitor of metalloproteinases (TIMP)-1 creates a premetastatic niche in Natural killer cell-mediated ablation of metastatic liver tumors by

www.aacrjournals.org Clin Cancer Res; 22(24) December 15, 2016 OF9

Brodt

hydrodynamic injection of IFNalpha gene to mice. Int J Cancer 2007; 46. Bouwens L, Baekeland M, De Zanger R, Wisse E. Quantitation, tissue 120:1252–60. distribution and proliferation kinetics of Kupffer cells in normal rat liver. 26. Tatsumi T, Takehara T, Yamaguchi S, Sasakawa A, Miyagi T, Jinushi M, Hepatology 1986;6:718–22. et al. Injection of IL-12 gene-transduced dendritic cells into mouse liver 47. van der Bij GJ, Oosterling SJ, Meijer S, Beelen RH, van Egmond M. tumor lesions activates both innate and acquired immunity. Gene Ther Therapeutic potential of Kupffer cells in prevention of liver metastases 2007;14:863–71. outgrowth. Immunobiology 2005;210:259–65. 27. Pfister K, Radons J, Busch R, Tidball JG, Pfeifer M, Freitag L, et al. Patient 48. Timmers M, Vekemans K, Vermijlen D, Asosingh K, Kuppen P, Bouwens L, survival by Hsp70 membrane phenotype: association with different et al. Interactions between rat colon carcinoma cells and Kupffer cells routes of metastasis. Cancer 2007;110:926–35. during the onset of hepatic metastasis. Int J Cancer 2004;112:793–802. 28. Spicer J, Brodt P, Ferri LE. Role of inflammation in the early stages of liver 49. Bayon LG, Izquierdo MA, Sirovich I, van Rooijen N, Beelen RH, Meijer S. metastasis. In: Brodt P, editor. Liver metastasis: biology and clinical Role of Kupffer cells in arresting circulating tumor cells and controlling management. New York: Springer; 2011. p. 155–85. metastatic growth in the liver. Hepatology 1996;23:1224–31. 29. Paschos KA, Majeed AW, Bird NC. Natural history of hepatic metastases 50. Matsumura H, Kondo T, Ogawa K, Tamura T, Fukunaga K, Murata S, et al. from colorectal cancer - pathobiological pathways with clinical signifi- Kupffer cells decrease metastasis of colon cancer cells to the liver in the cance. World J Gastroenterol 2014;20:3719–37. early stage. Int J Oncol 2014;45:2303–10. 30. Glinskii OV, Huxley VH, Glinsky GV, Pienta KJ, Raz A, Glinsky VV. 51. Khatib AM, Auguste P, Fallavollita L, Wang N, Samani A, Kontogiannea M, Mechanical entrapment is insufficient and intercellular adhesion is essen- et al. Characterization of the host proinflammatory response to tumor tial for metastatic cell arrest in distant organs. Neoplasia 2005;7:522–7. cells during the initial stages of liver metastasis. Am J Pathol 2005;167: 31. Brodt P. Role of the host inflammatory response in colon carcinoma 749–59. initiation, progression and liver metastasis. In: Beauchemin N, Huot J, 52. Khatib AM, Kontogiannea M, Fallavollita L, Jamison B, Meterissian S, editors. Metastasis of colorectal cancer. New York, NY: Springer; 2010. Brodt P. Rapid induction of cytokine and E-selectin expression in the liver p. 289–319. in response to metastatic tumor cells. Cancer Res 1999;59:1356–61. 32. Jiao S-F, Sun K, Chen X-J, Zhao X, Cai N, Liu Y-J, et al. Inhibition of tumor 53. Wen SW, Ager EI, Christophi C. Bimodal role of Kupffer cells during necrosis factor alpha reduces the outgrowth of hepatic micrometastasis colorectal cancer liver metastasis. Cancer Biol Ther 2013;14:606–13. of colorectal tumors in a mouse model of liver ischemia-reperfusion 54. Jessup JM, Samara R, Battle P, Laguinge LM. Carcinoembryonic antigen injury. J Biomed Sci 2014;21:1. promotes tumor cell survival in liver through an IL-10-dependent path- 33. Khatib AM, Fallavollita L, Wancewicz EV, Monia BP, Brodt P. Inhibition of way. Clin Exp Metastasis 2004;21:709–17. hepatic endothelial E-selectin expression by C-raf antisense oligonucleo- 55. Asao T, Shibata HR, Batist G, Brodt P. Eradication of hepatic metastases of tides blocks colorectal carcinoma liver metastasis. Cancer Res 2002;62: carcinoma H-59 by combination chemoimmunotherapy with liposomal 5393–8. muramyl tripeptide, 5-fluorouracil, and leucovorin. Cancer Res 1992;52: 34. Yoshimoto K, Tajima H, Ohta T, Okamoto K, Sakai S, Kinoshita J, et al. 6254–7. Increased E-selectin in hepatic ischemia-reperfusion injury mediates liver 56. Tosello-Trampont A-C, Landes SG, Nguyen V, Novobrantseva TI, Hahn metastasis of pancreatic cancer. Oncol Rep 2012;28:791–6. YS. Kuppfer cells trigger nonalcoholic steatohepatitis development in 35. Ham B, Wang N, D'Costa Z, Fernandez MC, Bourdeau F, Auguste P, et al. diet-induced mouse model through tumor necrosis factor-a production. TNF receptor-2 facilitates an immunosuppressive microenvironment in J Biol Chem 2012;287:40161–72. the liver to promote the colonization and growth of hepatic metastases. 57. Dey A, Allen J, Hankey-Giblin PA. Ontogeny and polarization of macro- Cancer Res 2015;75:5235–47. phages in inflammation: blood monocytes versus tissue macrophages. 36. Elliott VA, Rychahou P, Zaytseva YY, Evers BM. Activation of c-Met and Front Immunol 2014;5:683. upregulation of CD44 expression are associated with the metastatic 58. Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, et al. phenotype in the colorectal cancer liver metastasis model. PLoS One Macrophage activation and polarization: nomenclature and experimental 2014;9:e97432. guidelines. Immunity 2014;41:14–20. 37. Witz IP. The selectin-selectin ligand axis in tumor progression. Cancer 59. Schouppe E, De Baetselier P, Van Ginderachter JA, Sarukhan A. Instruction Metastasis Rev 2008;27:19–30. of myeloid cells by the tumor microenvironment: open questions on the 38. Greenwald E, Yuki K. A translational consideration of intercellular adhe- dynamics and plasticity of different tumor-associated myeloid cell popu- sion molecule-1 biology in the perioperative setting. Transl Perioper Pain lations. Oncoimmunology 2012;1:1135–45. Med 2016;1:17–23. 60. Galdiero MR, Bonavita E, Barajon I, Garlanda C, Mantovani A, Jaillon S. 39. Tremblay PL, Huot J, Auger FA. Mechanisms by which E-selectin regulates Tumor associated macrophages and neutrophils in cancer. Immunobiol- diapedesis of colon cancer cells under flow conditions. Cancer Res ogy 2013;218:1402–10. 2008;68:5167–76. 61. Mills CD. Anatomy of a discovery: m1 and m2 macrophages. Front 40. Aychek T, Miller K, Sagi-Assif O, Levy-Nissenbaum O, Israeli-Amit M, Immunol 2015;6:212. Pasmanik-Chor M, et al. E-selectin regulates gene expression in metastatic 62. Mills CD, Lenz LL, Harris RA. A breakthrough: macrophage-directed colorectal carcinoma cells and enhances HMGB1 release. Int J Cancer cancer immunotherapy. Cancer Res 2016;76:513–6. 2008;123:1741–50. 63. Cui YL, Li HK, Zhou HY, Zhang T, Li Q. Correlations of tumor-associated 41. Ou J, Peng Y, Deng J, Miao H, Zhou J, Zha L, et al. Endothelial cell-derived macrophage subtypes with liver metastases of colorectal cancer. Asian Pac fibronectin extra domain A promotes colorectal cancer metastasis via J Cancer Prev 2013;14:1003–7. inducing epithelial-mesenchymal transition. Carcinogenesis 2014;35: 64. Sionov RV, Fridlender ZG, Granot Z. The multifaceted roles neutro- 1661–70. phils play in the tumor microenvironment. Cancer Microenviron 42. Hu CT, Guo LL, Feng N, Zhang L, Zhou N, Ma LL, et al. MIF, secreted by 2015;8:125–58. human hepatic sinusoidal endothelial cells, promotes chemotaxis and 65. McDonald B, Spicer J, Giannais B, Fallavollita L, Brodt P, Ferri LE. Systemic outgrowth of colorectal cancer in liver prometastasis. Oncotarget inflammation increases cancer cell adhesion to hepatic sinusoids by 2015;6:22410–23. neutrophil mediated mechanisms. Int J Cancer 2009;125:1298–305. 43. Banerjee D, Hernandez SL, Garcia A, Kangsamaksin T, Sbiroli E, 66. Spicer JD, McDonald B, Cools-Lartigue JJ, Chow SC, Giannias B, Kubes P, Andrews J, et al. Notch suppresses angiogenesis and progression of et al. Neutrophils promote liver metastasis via Mac-1-mediated interac- hepatic metastases. Cancer Res 2015;75:1592–602. tions with circulating tumor cells. Cancer Res 2012;72:3919–27. 44. Im JH, Tapmeier T, Balathasan L, Gal A, Yameen S, Hill S, et al. G-CSF 67. Yamamoto M, Kikuchi H, Ohta M, Kawabata T, Hiramatsu Y, Kondo K, rescues tumor growth and neo-angiogenesis during liver metastasis under et al. TSU68 prevents liver metastasis of colon cancer xenografts by host angiopoietin-2 deficiency. Int J Cancer 2013;132:315–26. modulating the premetastatic niche. Cancer Res 2008;68:9754–62. 45. Yona S, Kim KW, Wolf Y, Mildner A, Varol D, Breker M, et al. Fate mapping 68. Huh SJ, Liang S, Sharma A, Dong C, Robertson GP. Transiently entrapped reveals origins and dynamics of monocytes and tissue macrophages under circulating tumor cells interact with neutrophils to facilitate lung metas- homeostasis. Immunity 2013;38:79–91. tasis development. Cancer Res 2010;70:6071–82.

OF10 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

The Microenvironment in Liver Metastasis

69. Cools-Lartigue J, Spicer J, McDonald B, Gowing S, Chow S, Giannias B, 91. Mueller L, Goumas FA, Affeldt M, Sandtner S, Gehling UM, Brilloff S, et al. et al. Neutrophil extracellular traps sequester circulating tumor cells and Stromal fibroblasts in colorectal liver metastases originate from resident promote metastasis. J Clin Invest 2013;123:3446–58. fibroblasts and generate an inflammatory microenvironment. Am J Pathol 70. Tohme S, Yazdani HO, Al-Khafaji AB, Chidi AP, Loughran P, Mowen K, 2007;171:1608–18. et al. Neutrophil extracellular traps promote the development and pro- 92. Wang J, Fallavollita L, Brodt P. Inhibition of experimental hepatic metas- gression of liver metastases after surgical stress. Cancer Res 2016;76: tasis by a monoclonal antibody that blocks tumor-hepatocyte interaction. 1367–80. J Immunother Emphasis Tumor Immunol 1994;16:294–302. 71. Fridlender ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, et al. Polarization 93. Mook ORF, van Marie J, Jonges R, Vreeling-Sindelarova H, Frederiks WM, of tumor-associated neutrophil phenotype by TGF-beta: "N1" versus Van Noorden CJF. Interactions between colon cancer cells and hepato- "N2" TAN. Cancer Cell 2009;16:183–94. cytes in rats in relation to metastasis. J Cell Mol Med 2008;12:2052–61. 72. Donskov F. Immunomonitoring and prognostic relevance of neutrophils 94. Shimizu S, Yamada N, Sawada T, Ikeda K, Nakatani K, Seki S, et al. in clinical trials. Semin Cancer Biol 2013;23:200–7. Ultrastructure of early phase hepatic metastasis of human colon carci- 73. Halazun KJ, Aldoori A, Malik HZ, Al-Mukhtar A, Prasad KR, Toogood GJ, noma cells with special reference to desmosomal junctions with hepa- et al. Elevated preoperative neutrophil to lymphocyte ratio predicts tocytes. Pathol Int 2000;50:953–9. survival following hepatic resection for colorectal liver metastases. Eur 95. Huang J, Pan C, Hu H, Zheng S, Ding L. Osteopontin-enhanced hepatic J Surg Oncol 2008;34:55–60. metastasis of colorectal cancer cells. PLoS One 2012;7:e47901. 74. Friedman SL. Hepatic stellate cells: protean, multifunctional, and enig- 96. Tabaries S, Dupuy F, Dong Z, Monast A, Annis MG, Spicer J, et al. Claudin- matic cells of the liver. Physiol Rev 2008;88:125–72. 2 promotes breast cancer liver metastasis by facilitating tumor cell inter- 75. Gressner AM, Bachem MG. Molecular mechanisms of liver fibrogen- actions with hepatocytes. Mol Cell Biol 2012;32:2979–91. esis–a homage to the role of activated fat-storing cells. Digestion 97. Georges R, Bergmann F, Hamdi H, Zepp M, Eyol E, Hielscher T, et al. 1995;56:335–46. Sequential biphasic changes in claudin1 and claudin4 expression are 76. Muhanna N, Doron S, Wald O, Horani A, Eid A, Pappo O, et al. Activation correlated to colorectal cancer progression and liver metastasis. J Cell Mol of hepatic stellate cells after phagocytosis of lymphocytes: a novel path- Med 2012;16:260–72. way of fibrogenesis. Hepatology 2008;48:963–77. 98. Radinsky R, Risin S, Fan D, Dong Z, Bielenberg D, Bucana CD, et al. Level 77. Zhao W, Zhang L, Xu Y, Zhang Z, Ren G, Tang K, et al. Hepatic stellate and function of epidermal growth factor receptor predict the metastatic cells promote tumor progression by enhancement of immunosuppres- potential of human colon carcinoma cells. Clin Cancer Res 1995;1:19–31. sive cells in an orthotopic liver tumor mouse model. Lab Invest 99. Zvibel I, Wagner A, Pasmanik-Chor M, Varol C, Oron-Karni V, Santo EM, 2014;94:182–91. et al. Transcriptional profiling identifies genes induced by hepatocyte- 78. Burnier JV, Wang N, Michel RP, Hassanain M, Li S, Lu Y, et al. Type IV derived extracellular matrix in metastatic human colorectal cancer cell collagen-initiated signals provide survival and growth cues required for lines. Clin Exp Metastasis 2013;30:189–200. liver metastasis. Oncogene 2011;30:3766–83. 100. Li H, Fan X, Stoicov C, Liu JH, Zubair S, Tsai E, et al. Human and mouse 79. Copple BL, Bai S, Burgoon LD, Moon J-OK. Hypoxia-inducible factor- colon cancer utilizes CD95 signaling for local growth and metastatic 1a regulates expression of genes in hypoxic hepatic stellate cells spread to liver. Gastroenterology 2009;137:934–44. important for collagen deposition and angiogenesis. Liver Int 2011; 101. Long L, Nip J, Brodt P. Paracrine growth stimulation by hepatocyte- 31:230–44. derived insulin-like growth factor-1: a regulatory mechanism for carci- 80. Taura K, De Minicis S, Seki E, Hatano E, Iwaisako K, Osterreicher CH, et al. noma cells metastatic to the liver. Cancer Res 1994;54:3732–7. Hepatic stellate cells secrete angiopoietin 1 that induces angiogenesis in 102. Long L, Rubin R, Baserga R, Brodt P. Loss of the metastatic phenotype in liver fibrosis. Gastroenterology 2008;135:1729–38. murine carcinoma cells expressing an antisense RNA to the insulin-like 81. Kang N, Shah VH, Urrutia R. Membrane-to-nucleus signals and epigenetic growth factor receptor. Cancer Res 1995;55:1006–9. mechanisms for myofibroblastic activation and desmoplastic stroma: 103. Samani AA, Chevet E, Fallavollita L, Galipeau J, Brodt P. Loss of tumor- potential therapeutic targets for liver metastasis? Mol Cancer Res igenicity and metastatic potential in carcinoma cells expressing the 2015;13:604–12. extracellular domain of the type 1 insulin-like growth factor receptor. 82. Olaso E, Salado C, Egilegor E, Gutierrez V, Santisteban A, Sancho-Bru P, Cancer Res 2004;64:3380–5. et al. Proangiogenic role of tumor-activated hepatic stellate cells in 104. Wang N, Rayes RF, Elahi SM, Lu Y, Hancock MA, Massie B, et al. The IGF- experimental melanoma metastasis. Hepatology 2003;37:674–85. Trap: novel inhibitor of carcinoma growth and metastasis. Mol Cancer 83. Badiola I, Olaso E, Crende O, Friedman SL, Vidal-Vanaclocha F. Discoidin Ther 2015;14:982–93. domain receptor 2 deficiency predisposes hepatic tissue to colon carci- 105. Wagh P, Peace BE, Waltz SE. The met-related receptor tyrosine kinase Ron noma metastasis. Gut 2012;61:1465–72. in tumor growth and metastasis. Adv Cancer Res 2008;100:1–33. 84. Yang X, Lu P, Ishida Y, Kuziel WA, Fujii C, Mukaida N. Attenuated liver 106. Yoshioka T, Nishikawa Y, Ito R, Kawamata M, Doi Y, Yamamoto Y, et al. tumor formation in the absence of CCR2 with a concomitant reduction in Significance of integrin alphavbeta5 and erbB3 in enhanced cell migra- the accumulation of hepatic stellate cells, macrophages and neovascular- tion and liver metastasis of colon carcinomas stimulated by hepatocyte- ization. Int J Cancer 2006;118:335–45. derived heregulin. Cancer Sci 2010;101:2011–8. 85. Eveno C, Hainaud P, Rampanou A, Bonnin P, Bakhouche S, Dupuy E, 107. Dome B, Hendrix MJ, Paku S, Tovari J, Timar J. Alternative vascularization et al. Proof of prometastatic niche induction by hepatic stellate cells. J Surg mechanisms in cancer: Pathology and therapeutic implications. Am J Res 2015;194:496–504. Pathol 2007;170:1–15. 86. Liu C, Billadeau DD, Abdelhakim H, Leof E, Kaibuchi K, Bernabeu C, et al. 108. Tabaries S, Annis MG, Hsu BE, Tam CE, Savage P, Park M, et al. Lyn IQGAP1 suppresses TbetaRII-mediated myofibroblastic activation and modulates Claudin-2 expression and is a therapeutic target for breast metastatic growth in liver. J Clin Invest 2013;123:1138–56. cancer liver metastasis. Oncotarget 2015;6:9476–87. 87. Terada T, Makimoto K, Terayama N, Suzuki Y, Nakanuma Y. Alpha- 109. Hadrup S, Donia M, Thor Straten P. Effector CD4 and CD8 T cells and smooth muscle actin-positive stromal cells in cholangiocarcinomas, their role in the tumor microenvironment. Cancer Microenviron hepatocellular carcinomas and metastatic liver carcinomas. J Hepatol 2013;6:123–33. 1996;24:706–12. 110. Butt AQ, Mills KH. Immunosuppressive networks and checkpoints con- 88. Ju MJ, Qiu SJ, Fan J, Xiao YS, Gao Q, Zhou J, et al. Peritumoral activated trolling antitumor immunity and their blockade in the development of hepatic stellate cells predict poor clinical outcome in hepatocellular cancer immunotherapeutics and vaccines. Oncogene 2014;33:4623–31. carcinoma after curative resection. Am J Clin Pathol 2009;131:498–510. 111. Facciabene A, Motz GT, Coukos G. T-regulatory cells: key players in tumor 89. Sirica AE, Gores GJ. Desmoplastic stroma and cholangiocarcinoma: immune escape and angiogenesis. Cancer Res 2012;72:2162–71. clinical implications and therapeutic targeting. Hepatology 2014;59: 112. Wolf AM, Wolf D, Steurer M, Gastl G, Gunsilius E, Grubeck-Loebenstein 2397–402. B. Increase of regulatory T cells in the peripheral blood of cancer patients. 90. Brenner DA, Kisseleva T, Scholten D, Paik YH, Iwaisako K, Inokuchi S, Clin Cancer Res 2003;9:606–12. et al. Origin of myofibroblasts in liver fibrosis. Fibrogenesis Tissue Repair 113. Connolly MK, Mallen-St Clair J, Bedrosian AS, Malhotra A, Vera V, 2012;5(Suppl 1):S17. Ibrahim J, et al. Distinct populations of metastases-enabling myeloid

www.aacrjournals.org Clin Cancer Res; 22(24) December 15, 2016 OF11

Brodt

cells expand in the liver of mice harboring invasive and preinvasive intra- 122. Meyer C, Cagnon L, Costa-Nunes CM, Baumgaertner P, Montandon N, abdominal tumor. J Leuk Biol 2010;87:713–25. Leyvraz L, et al. Frequencies of circulating MDSC correlate with clinical 114. Correale P, Rotundo MS, Del Vecchio MT, Remondo C, Migali C, outcome of melanoma patients treated with ipilimumab. Cancer Immu- Ginanneschi C, et al. Regulatory (FoxP3þ) T-cell tumor inﬁltration is a nol Immunother 2014;63:247–57. favorable prognostic factor in advanced colon cancer patients undergoing 123. Kusmartsev S, Nefedova Y, Yoder D, Gabrilovich DI. Antigen-speciﬁc chemo or chemoimmunotherapy. J Immunother 2010;33:435–41. inhibition of CD8þ T cell response by immature myeloid cells in cancer is 115. Rech AJ, Vonderheide RH. Clinical use of anti-CD25 antibody daclizu- mediated by reactive oxygen species. J Immunol 2004;172:989–99. mab to enhance immune responses to tumor antigen vaccination by 124. Huang B, Pan PY, Li Q, Sato AI, Levy DE, Bromberg J, et al. Gr-1þCD115þ targeting regulatory T cells. Ann N Y Acad Sci 2009;1174:99–106. immature myeloid suppressor cells mediate the development of tumor- 116. Rech AJ, Mick R, Martin S, Recio A, Aqui NA, Powell DJ Jr, et al. CD25 induced T regulatory cells and T-cell anergy in tumor-bearing host. blockade depletes and selectively reprograms regulatory T cells in concert Cancer Res 2006;66:1123–31. with immunotherapy in cancer patients. Sci Transl Med 2012;4:134ra62. 125. Diaz-Montero CM, Salem ML, Nishimura MI, Garrett-Mayer E, Cole 117. Keskinov AA, Shurin MR. Myeloid regulatory cells in tumor spreading and DJ, Montero AJ. Increased circulating myeloid-derived suppressor cells metastasis. Immunobiology 2015;220:236–42. correlate with clinical cancer stage, metastatic tumor burden, and 118. Almand B, Clark JI, Nikitina E, van Beynen J, English NR, Knight SC, doxorubicin-cyclophosphamide chemotherapy. Cancer Immunol et al. Increased production of immature myeloid cells in cancer Immunother 2009;58:49–59. patients: a mechanism of immunosuppression in cancer. J Immunol 126. Zhao L, Lim SY, Gordon-Weeks AN, Tapmeier TT, Im JH, Cao Y, et al. 2001;166:678–89. Recruitment of a myeloid cell subset (CD11b/Gr1 mid) via CCL2/ 119. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regulation CCR2 promotes the development of colorectal cancer liver metastasis. of myeloid cells by tumours. Nat Rev Immunol 2012;12:253–68. Hepatology 2013;57:829–39. 120. Haverkamp JM, Smith AM, Weinlich R, Dillon CP, Qualls JE, Neale G, 127. Rodriguez PC, Ernstoff MS, Hernandez C, Atkins M, Zabaleta J, Sierra R, et al. Myeloid-derived suppressor activity is mediated by monocytic et al. Arginase I-producing myeloid-derived suppressor cells in renal cell lineages maintained by continuous inhibition of extrinsic and intrinsic carcinoma are a subpopulation of activated granulocytes. Cancer Res death pathways. Immunity 2014;41:947–59. 2009;69:1553–60. 121. Khaled YS, Ammori BJ, Elkord E. Increased levels of granulocytic myeloid- 128. Ugel S, Delpozzo F, Desantis G, Papalini F, Simonato F, Sonda N, et al. derived suppressor cells in peripheral blood and tumour tissue of pan- Therapeutic targeting of myeloid-derived suppressor cells. Curr Opin creatic cancer patients. J Immunol Res 2014;2014:879897. Pharmacol 2009;9:470–81.

OF12 Clin Cancer Res; 22(24) December 15, 2016 Clinical Cancer Research

Role of the Microenvironment in Liver Metastasis: From Pre- to Prometastatic Niches

Pnina Brodt

Clin Cancer Res Published OnlineFirst October 19, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-16-0460

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2016/11/29/1078-0432.CCR-16-0460. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.