8
Seminars in Cancer Biology 22 (2012) 33–40 Contents lists available at SciVerse ScienceDirect Seminars in Cancer Biology jou rn al h om epa g e: www.elsevier.com/locate/semcancer Review Cancer-related inflammation: Common themes and therapeutic opportunities Frances R. Balkwill a,, Alberto Mantovani b a Centre for Cancer and Inflammation, Barts Cancer Institute, Charterhouse Square, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, EC1M6BQ, United Kingdom b Istituto Clinico Humanitas IRCCS, Università degli Studi di Milano, Via Manzoni 113, 20089 Rozzano, Milan, Italy a r t i c l e i n f o Keywords: Inflammation Macrophages Cytokines Chemokines Tumor microenvironment a b s t r a c t Inflammatory cells and mediators are an essential component of the tumor microenvironment. Inflam- matory circuits can differ considerably in different tumors in terms of cellular and cytokine networks and molecular drivers. However, macrophages are a common and fundamental component of cancer promot- ing inflammation. Drivers of macrophage functional orientation include tumor cells, cancer-associated fibroblasts, T cells and B cells. Dissection of the diversity of cancer-related inflammation is instrumental to the design of therapeutic approaches that target cancer-related inflammation. © 2011 Elsevier Ltd. All rights reserved. 1. Introduction to cancer-related inflammation Smouldering, non-resolving inflammation is one of the con- sistent features of the tumor microenvironment. The connection between inflammation and cancer, first perceived in the nineteenth century is now accepted as enabling characteristic of cancer [1,2]. Current estimates suggest that about 25% of cancers are associated with chronic inflammation sustained by infections (e.g. hepatitis) or inflammatory conditions of diverse origin (e.g. prostatitis) [1]. Moreover, tumors that are not epidemiologically related to inflam- mation are characterized by the presence of inflammatory cells and mediators [1,3]. Inflammation and cancer are connected by two pathways: extrinsic pathways from conditions that cause non-resolving smouldering inflammatory responses, and intrinsic pathways driven by oncogenes or tumor suppressor genes that activate the expression of inflammation-related programmes [1]. In this review we will first focus on some recent advances in cancer-related inflammation (CRI), emphasizing its diversity as well as the common emerging themes. In the past decade our understanding CRI has increased to the point when we can begin to translate our knowledge into new approaches to cancer prevention and treatment. This is the second focus of our review. Abbreviations: CRI, cancer-related inflammation; EMT, epithelial mes- enchymal transition; TAFs, tumor-associated fibroblasts; TAM, tumor-associated macrophages; TSLP, thymic stromal lymphopoietin. Corresponding author. Tel.: +44 207882 3587; fax: +44 207882 3885. E-mail address: [email protected] (F.R. Balkwill). 2. Polarization and diversity of inflammatory cells in the tumor microenvironment Tumor associated macrophages (TAM) are a common compo- nent of CRI and will be used as a paradigm of its complexity. Cells of the monocyte–macrophage lineage exhibit considerable plastic- ity and diversity [1,4–8]. TAM populations in murine tumors can be quite diverse and hypoxia may be one driver of this [9]. Sub- sets have been identified between mouse and human monocytes [8]. It remains to be determined whether diversity of TAM reflects their origin from different monocyte precursors or microanatom- ical differences (e.g. oxygen tension in different part of the cancer tissue). Molecular pathways driving TAM polarization can differ consid- erably in tumors arising at different sites. For instance CD4 T cells, B cells, antibodies and Fc receptors orchestrate the M2-like phe- notype of tumor promoting TAM in a model of human papilloma virus-driven squamous cancer [10]. In contrast, in a mammary car- cinoma model, Th2-derived IL-4 is responsible for M2 polarization and promotion of metastasis [11]. Complement can also be a mech- anism of myelomonocytic cells recruitment [12]. B cells can skew macrophage function and promote tumor progression, using IL-10 and lymphotoxin (LT) [13–16] and B regulatory cells have been shown to enhance carcinogenesis [14]. Thus, mechanisms of regu- lation of myelomonocytic cells in tumors can be different organ or tumor contexts but M2-like skewing is a recurrent common denominator. Definition of TAM diversity in different human can- cers will be required to translate this recent progress into clinical benefit. Early studies indicated that in situ proliferation of mature mononuclear phagocytes could contribute to TAM accumulation [17]. Recent evidence suggests that proliferation can indeed con- tribute to macrophage accumulation, in particular during type II 1044-579X/$ see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.semcancer.2011.12.005

Cancer-related Inflammation Common Themes and Therapeutic Opportunities

Embed Size (px)

Citation preview

  • Seminars in Cancer Biology 22 (2012) 33 40

    Contents lists available at SciVerse ScienceDirect

    Seminars in Cancer Biology

    jou rn al h om epa g e: www.elsev ier .com/

    Review

    Cancer s a

    Frances Ra Centre for Can on ScQueen Mary Unb Istituto Clinic , Milan

    a r t i c l

    Keywords:InammationMacrophagesCytokinesChemokinesTumor microe

    an e in diages aage ion ofes th

    1. Introdu

    Smouldesistent featbetween incentury is nCurrent estiwith chronor inammMoreover, tumors that are not epidemiologically related to inam-mation are characterized by the presence of inammatory cells andmediators [1,3].

    Inammation and cancer are connected by two pathways:extrinsic pathways from conditions that cause non-resolvingsmoulderindriven by oexpression

    In this rcancer-relawell as theunderstandtranslate ouand treatme

    Abbreviatioenchymal tranmacrophages;

    CorresponE-mail add

    ariza mic

    or af CRIono

    divte dive b

    [8]. It remains to be determined whether diversity of TAM reectstheir origin from different monocyte precursors or microanatom-ical differences (e.g. oxygen tension in different part of the cancertissue).

    Molecular pathways driving TAM polarization can differ consid-

    1044-579X/$ doi:10.1016/j.g inammatory responses, and intrinsic pathwaysncogenes or tumor suppressor genes that activate theof inammation-related programmes [1].eview we will rst focus on some recent advances inted inammation (CRI), emphasizing its diversity as

    common emerging themes. In the past decade ouring CRI has increased to the point when we can begin tor knowledge into new approaches to cancer preventionnt. This is the second focus of our review.

    ns: CRI, cancer-related inammation; EMT, epithelial mes-sition; TAFs, tumor-associated broblasts; TAM, tumor-associated

    TSLP, thymic stromal lymphopoietin.ding author. Tel.: +44 207882 3587; fax: +44 207882 3885.ress: [email protected] (F.R. Balkwill).

    erably in tumors arising at different sites. For instance CD4 T cells,B cells, antibodies and Fc receptors orchestrate the M2-like phe-notype of tumor promoting TAM in a model of human papillomavirus-driven squamous cancer [10]. In contrast, in a mammary car-cinoma model, Th2-derived IL-4 is responsible for M2 polarizationand promotion of metastasis [11]. Complement can also be a mech-anism of myelomonocytic cells recruitment [12]. B cells can skewmacrophage function and promote tumor progression, using IL-10and lymphotoxin (LT) [1316] and B regulatory cells have beenshown to enhance carcinogenesis [14]. Thus, mechanisms of regu-lation of myelomonocytic cells in tumors can be different organor tumor contexts but M2-like skewing is a recurrent commondenominator. Denition of TAM diversity in different human can-cers will be required to translate this recent progress into clinicalbenet.

    Early studies indicated that in situ proliferation of maturemononuclear phagocytes could contribute to TAM accumulation[17]. Recent evidence suggests that proliferation can indeed con-tribute to macrophage accumulation, in particular during type II

    see front matter 2011 Elsevier Ltd. All rights reserved.semcancer.2011.12.005-related inammation: Common theme

    . Balkwill a,, Alberto Mantovanib

    cer and Inammation, Barts Cancer Institute, Charterhouse Square, Barts and The Londiversity of London, EC1M6BQ, United Kingdom

    o Humanitas IRCCS, Universit degli Studi di Milano, Via Manzoni 113, 20089 Rozzano

    e i n f o

    nvironment

    a b s t r a c t

    Inammatory cells and mediators arematory circuits can differ considerablymolecular drivers. However, macrophing inammation. Drivers of macrophbroblasts, T cells and B cells. Dissectto the design of therapeutic approach

    ction to cancer-related inammation

    ring, non-resolving inammation is one of the con-ures of the tumor microenvironment. The connectionammation and cancer, rst perceived in the nineteenthow accepted as enabling characteristic of cancer [1,2].mates suggest that about 25% of cancers are associatedic inammation sustained by infections (e.g. hepatitis)atory conditions of diverse origin (e.g. prostatitis) [1].

    2. Poltumor

    Tumnent oof the mity andbe quisets halocate /semcancer

    nd therapeutic opportunities

    hool of Medicine and Dentistry,

    , Italy

    ssential component of the tumor microenvironment. Inam-fferent tumors in terms of cellular and cytokine networks andre a common and fundamental component of cancer promot-

    functional orientation include tumor cells, cancer-associated the diversity of cancer-related inammation is instrumentalat target cancer-related inammation.

    2011 Elsevier Ltd. All rights reserved.

    tion and diversity of inammatory cells in theroenvironment

    ssociated macrophages (TAM) are a common compo- and will be used as a paradigm of its complexity. Cellscytemacrophage lineage exhibit considerable plastic-ersity [1,48]. TAM populations in murine tumors canverse and hypoxia may be one driver of this [9]. Sub-een identied between mouse and human monocytes

  • 34 F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40

    inammation [18]. M-CSF and IL-4 may underlie the proliferativecapacity of macrophages including TAM [19]. Whether and to whatextent TAM proliferate in human tumors remains to be dened.

    Tissue repair is an important component of the inammatoryresponse along been associated important stors, expresand pancrethought to asuggest thamouse modcers mobilizThese cells local broba cancer-princlude a reling factorbreast canctriple-negasurvival [23

    A recenthe tumor m, FAP-, emice, rapidaccompanieity [24]. Thecontribute t

    Other inmodiers oneutrophilsrelease of cproteases [siderably intissues, butand subvers

    3. Cytokin

    TNF-, Itory cytokinfocus on th

    Unlike tstitutively from animathe growthinduced tubowel (revmodels, TNautonomouchemokineinhibitory ftor (VEGF) microenviroand developangiogenic to cause difcells in the cells, othernant cell-d[32] and CD

    It is not microenviroduced by m

    [34]; in a model where chemical damage led to liver cancer, Kupf-fer cell-derived TNF was one of the mitogens driving proliferationof hepatocytes in which DNA damage had already been caused bythe carcinogenic agent diethylnitrosamine [35]. In both a chem-

    inducric c

    and

    etheirectlate Ehan90s a

    st of tFR1. T

    canyte il celation- ped bent,ents velse sig

    [40]eve

    activ[29].tokin

    supp, andperfa

    larvevelssornic rs devogenvae pssorof thsamemunimen

    resus evat thahila de cytinoglatio

    is h otutes-prome of er of

    cancinom

    cananciand sia STs arend broblasts are key to this process. Tumors haveconsidered wounds that do not heal [20]. Tumor-broblasts (TAFs) are a major component of CRI anource of tumor promoting cytokines and growth fac-sing a pro-inammatory gene signature in skin, breastatic cancers that is regulated by NF- [21]. TAFs arerise from mesenchymal stem cells [22] but recent datat other bone marrow cells can regulate their activity. Inels of breast cancer endocrine signals originating in can-e Sca+ cKit granulin-positive bone marrow cells [23].do not act directly to promote tumor growth but act onlasts in the tumor microenvironment to switch them toomoting phenotype. Genes induced in TAFs by granulinange of chemokines, cytokines, and matrix remod-s already implicated in tumor-promotion. In humaner high granulin expression correlated with aggressivetive, basal-like tumor subtype and with reduced patient]. This is another example of cell co-operation in CRI.t paper highlighted the importance of broblasts inicroenvironment. When broblast activation protein-

    xpressing broblasts were depleted in tumor-bearing hypoxic necrosis of tumor and stromal cells occurredd by IFN- and TNF--mediated CD8+ T cell cytotoxic-se data suggest that, at least in some cancers, broblastso the inammatory immunosuppressive milieu.ammatory cells that have recently been identied asf some inammatory tumor microenvironments are

    [2527] and mast cells, important not only for theirytokines and chemokines but also reactive oxygen and2,28]. Thus, the cellular networks involved differ con-

    different tumors and in tumors originating in different common themes emerge and these include skewingion of macrophage function.

    es and cancer-related inammation

    L-6 and IL-1 are among the most studied inamma-es in the tumor microenvironment and this review willese.heir normal counterparts, many malignant cells con-produce small amounts of TNF-. There is evidencel models that malignant cell-derived TNF- enhances

    and spread of syngeneic, xenogeneic, and carcinogen-mors of skin, ovary, pancreas, pleural cavity andiewed in Ref. [29]). For instance, in ovarian cancerF- is an important component of a malignant cell-s network of inammatory cytokines, including thes CXCL12 and CCL2, the cytokines IL-6 and macrophageactor, MIF as well as vascular endothelial growth fac-[30]. This network then acts on the ovarian cancernment, particularly affecting the leukocyte inltratement of blood vessels in peritoneal tumor deposits. Theactions of TNF- may be due, at least in part, to its abilityferentiation of myeloid progenitor cells into endothelialtumor microenvironment [31]. Apart from endothelial

    host cells targeted by the paracrine actions of malig-erived TNF- include tumor-associated macrophages4 cells [33].only malignant cells that can make TNF- in the tumornment. In a genetic model of liver cancer, TNF- pro-yeloid cells promoted inammation-associated tumors

    ically of gastmation[29]).

    Whmay dstimul to enthe 19[37].

    Movia TNhumanleukocin renastimul

    TNFproducvironmof patiHigh le11-gencancer

    Howtumorfactor this cytumorservedTNF sutor. Flygenes dsuppreoncogetumorof oncthe larsuppreptosis if the this imis detr

    Thetumorthe facDrosopto havof carcderegu

    IL-6as witcontribtumora rangeffectohumanlar car

    IL-6malignptosis [43]. Vgenesied model of colorectal cancer, and a genetic modelancer, macrophage TNF- was implicated in inam-

    subsequent tumor development (reviewed in Ref.

    r made by malignant cells or host cells or both TNF-y contribute to oncogene activation, DNA damage andMT, e.g. [36]. This may partly explain the ability of TNF-ce metastatic activity of tumor cells as rst reported innd further elucidated by Michael Karins group in 2009

    hese pro-tumor actions of TNF- appear to be mediatedhis TNF receptor is found on tumor and stromal cells incer biopsies whereas TNFR2 is generally present on thenltrate, although it is also present on malignant cellsl carcinoma [38] where it may allow autocrine growth

    acting via Etk-VEGFR2 cross-talk [39].rotein is found in many different human cancers, bothy the malignant cells and or other cells in the microen-

    and elevated levels are found in plasma (see below)with advanced cancer disease (reviewed in Ref. [29]).

    of expression of TNF- made a major contribution to annature of poor prognostic signicance in Stage 1 lung.r, it is also clear that high doses of TNF- can have anti-ity that justify its original naming as a tumor necrosis

    Recent insight into the apparent paradoxical actions ofe comes from simpler organisms. In Drosophila bothressing and tumor-promoting roles of TNF are con-

    oncogenic Ras is the switch [41]. Flies have just onemily member called dTNF/eiger (egr) and one TNF recep-ae mutant for scribble group polarity tumor suppressorop imaginal disk tumors [42] and y TNF acts as a tumor

    via the JNK signaling pathway [41]. However, whenas (Rasv12) is introduced into the model, large invasiveelop. In ies lacking egr/dTNF, there was no invasionic cells, the clones remained at their site of origin andupated. This mechanism may have evolved as a tumor

    function of the y innate immune system causing apo-e mutant and potentially tumorigenic cells. However,

    cells gain a ras mutation, they are able to re-directe response to provoke invasive growth in a way thattal to the host.lts suggest that acquisition of Ras mutations may helpde host immune responses, and this is supported byt rasV12 is a common mutation in human cancers. Theata may also help us understand why it is advantageousokines such as TNF produced early on in the processenesis. A common feature of tumor progression is then of epithelial cell polarity and adhesion.another major mediator of acute inammation andher major cytokines, dysregulation of IL-6 signaling

    to many inammatory diseases [43]. IL-6 also hasoting actions on both malignant and stromal cells inxperimental cancer models [4345], is a downstreamoncogenic ras [46] and has been implicated in severalcers including multiple myeloma [47] and hepatocellu-a [48].

    be an autocrine or paracrine growth factor in somees, especially those of hematologic origin, it blocks apo-ignals through STAT3 that is activated in many cancersAT3, genes that promote cell proliferation and angio-

    induced in the tumor microenvironment by IL-6 [43].

  • F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40 35

    IL-6 is a critical tumor promoter in colitis-associated cancer. Pro-duced by lamina propria myeloid cells, IL-6 enhances proliferationof tumor-initiating cells and protects normal and malignant intesti-nal epithelial cells from apoptosis in a STAT3-dependent manner[44,45]. Cotumor size

    IL-6/gp1paracrine aras-transfortributes to tfavors progroblastoma

    In ovar6 enhancechemothera6 receptor In additionregulating tumor-promwith advanpoor prognmalignant asecrete IL-6co-culturedment [30,58cell autocrinnetwork invCCL2, CXCLin the tumo

    There isinstance, IL-cells and thcinogenesishas been obcytokine prcredited to cells inducethis cytokin

    4. Chemok

    Chemokcancer-relaprogressionsenescencemetastases

    CCL2 (Mciated to recUnequivocamatory CC [63,66]. M2by CC cheminvestigated

    Many capromoting broblast ois complex cells that corole for the cphocytes arof human ptumor-residable to favoing lymph nthymic stro

    conducted entirely with human cells and tissues the authors founda signicant association between tumor Th2 cell inltrates and poorprognosis with the ratio between Th2/Th1 cells in tumor biopsiesbeing independently predictive of patient survival. Using biopsies,

    d cematownstSLP.igrat

    cells brinpromnal e

    Hyp arenvirriton

    28 w canted ssivealigntory or foune.oxiaon or thiT regkiner wh

    rape

    undto trationnicalors o

    rgeti

    re issibleells, cal sg wi

    memnvir

    the fn widvanses (and ese raabin

    histo m

    patphoc

    these CDt hadntinuous treatment with recombinant IL-6 increasedin colitis-induced cancer models.30/STAT3 signaling also provides autocrine andmplication loop in lung adenocarcinoma [49] andmed cancer cells [46]. Bone marrow-derived IL-6 con-he formation of a bone marrow microenvironment thatression of neuroblastoma and increases survival of neu-

    cells [50].ian cancer, there is pre-clinical evidence that IL-s tumor cell survival and increases resistance topy via JAK/STAT signaling in tumor cells [51] and IL-alpha transignaling on tumor endothelial cells [52]., IL-6 has pro-angiogenic properties [53], as well asimmune cell inltration, stromal reaction and theoting actions of Th17 lymphocytes [54]. In patients

    ced disease, high plasma levels of IL-6 correlate withosis [55,56], and elevated levels are also present inscites [57]. Some ovarian cancer cell lines constitutively, and its production is enhanced when these cells are

    with other cells from the ovarian cancer microenviron-,59]. We have found that this IL-6 is part of a malignante cytokine network in ovarian cancer cells [30,60]. Thisolves co-regulation of the cytokines TNF- and IL-1,

    12 and VEGF and has paracrine actions on angiogenesisr microenvironment.

    strong renewed interest in IL-1 in relation to CRI. For1 is involved in the generation of IL-17 producing CD+ Te IL-23/IL-17 axis has been shown to promote skin car-

    [61]. Direct evidence for a role of IL-1 in human cancertained in multiple myeloma. Proteasome inhibitors andoduction inhibitors thalidomide and lenalidomide areact by disrupting this axis. IL-1 released by myelomas IL-6 production by bone marrow stromal cells, ande is in turn a growth factor for myeloma cells.

    ines and cancer-related inammation

    ines and their receptors are a key component ofted inammation affecting several pathways of tumor

    including leukocyte recruitment and function, cell, tumor cell proliferation and survival, invasion and

    [62].CP-1) and inammatory CC chemokines have long asso-ruitment of TAM in tumors, in particular TAM [6365].l genetic evidence for a non-redundant role of inam-chemokines in carcinogenesis has now been obtained

    polarization and survival of TAM are also promotedokines. CCL2 and its cognate receptor CCR2 have been

    in mouse tumors in particular in prostate cancer [67].ncer microenvironment papers focus on the tumor-or tumor-inhibiting roles of a particular immune,r endothelial cell but, as discussed above, the realityand dynamic interactions between the many differentmprise a malignant tumor. One recent study showed ahemokines CCL17 and 22 in such interactions. Th2 lym-e abundant in the brotic stroma that is characteristicancreatic cancer. De Monte et al. [68] hypothesized thatent dendritic cells, conditioned by local factors, may ber differentiation of tumor-specic Th2 cells in the drain-odes. They also had an idea that an IL-17-like cytokine,mal lymphopoietin (TSLP) might be involved. In work

    isolateinambly domake Tthat mCD4+ Tand 22major

    A CCL28.vesselsmicroeintrapeangle.

    CCLovarianassociasuppreThis mregulareceptto immVEGFA

    Hyprejecticountegenic chemobody o

    5. The

    Ourbegin prevenpre-climediat

    5.1. Ta

    Thebe posnant ca cliniStartinfamilymicroetrial ofbinatiowith aresponmass) respongemcitfrom ahints ttreatedof lymeled ina mouagonislls and short-term primary cultures, they found thatry cytokines produced by the malignant cells (proba-ream of oncogenic mutations) stimulate broblasts to

    TSLP activates/matures resident tumor dendritic cellse to draining lymph nodes where they in turn activate

    to a Th2 phenotype. Th2-attracting chemokines CCL17g the CD4+ cells back to the tumor where they have aoting inuence.xample of cell co-operation and chemokines involvesoxia, immune evasion and the formation of new blood

    key enabling characteristics of a progressing tumoronmentand recent work [69] shows that, at least ineal ovarian cancer, CCL28 might link this vicious tri-

    as frequently and strongly upregulated when humancer cells were exposed to hypoxia. CCL28 is normallywith mucosal immunity but it also recruits immuno-

    T regulatory, Treg cells, during liver inammation.ant cell-produced CCL28 recruited FoxP3 positive T

    cells that also expressed chemokine receptor CCR10ar CCL28. And the Treg cells did not only contribute

    tolerancethey also produced the angiogenic factor

    can induce a type of cell death that can trigger immunef tumorsthe induction of CCL28 is a mechanism tos via recruitment of immune suppressive and angio-ulatory cells. It will be interesting to see if differents have similar actions in tumors at other sites in theether CCL28 has a more universal role.

    utic opportunities

    erstanding CRI has increased to the point when we cannslate our knowledge into new approaches to cancer

    and treatment. This section will review some recent and clinical studies relating to the cells and solublef CRI.

    ng TAM

    evidence from experimental cancer models that it may to re-educate tumor-promoting TAM to reject malig-e.g. [70]. There is one promising recent example oftudy that involved modulating the function of TAM.th the hypothesis that activation of the TNF receptorber CD40 may reverse the immunosuppressive tumor

    onment, Beatty et al. conducted a small Phase II clinicalully human CD40 agonist antibody CP-870,893 in com-th gemcitabine chemotherapy, in twenty-one patientsced pancreatic cancer [71]. Four patients had partiali.e. a greater than 50% reduction in the size of a tumorleven patients had a period of disease stabilization. Thiste was greater than the 5% expected response rate toe alone and the progression-free survival was increasedoric rate of 2.3 months to 5.6 months [71]. Intriguingechanism of action were found in two biopsies fromientsa prominent cellular inltrate that was devoidytes. To investigate this further, clinical trial was mod-

    KPC mouse genetic model of pancreatic cancer using40 agonist. Within 18 h of administration, the CD40

    bound to tumor-associated macrophages, TAM that

  • 36 F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40

    had up-regulated MHC Class II and the co-stimulatory moleculeCD86. These cells were than able to lyse the pancreatic tumor cellsin vitro and in vivo destruction of the tumor stroma was observed.Macrophage depletion prevented these actions of the CD40 ago-nist, but thCD8+ T cellsnecessarilythe abundabe just as eterms of clinwarrant fur

    5.2. Targeti

    If TNF-anti-TNF-apeutic actas reportedplantable, xcancers (remight be beThis has beeas single ageRef. [29]). Fbody inixipatients wiII study in ostable diseacept (a soluTNF-) anddisease wittreatment. in the pre-m

    There is in cancer with chroninhibit cytomatory cell[72]all acttion, bindinhave directof cytotoxicon the immare of partition of the inammatopromotion.

    The therbeen evalucastration-rwhich IL-6 was 52%: by

    We comPhase II clresistant ovon tissue mthat intensiciated withwith siltuxproductiontumor-assoproducing itrial, one pwhile seventreated for levels of IL-

    levels of factors that were reduced by siltuximab treatment inthe patients signicantly correlated with high IL-6 pathway geneexpression and macrophage markers in microarray analyses ofovarian cancer biopsies. Hence we concluded that IL-6 stimulates

    mato mac

    inhibnical

    rgeti

    ibodgatey habodised td VEthers [75s in exte

    that at s2-po

    cellsonocl of toinganeoia eefucooma-ted wedial triaR4 or inl, in

    ytokien tyclam-term908asis ostat

    bicyUnexom thatop

    ells fvity cytot

    ula

    ammiseaseutic

    the cmmdent

    the o oneve imdy [7e therapy was still active in mice depleted of CD4+ or. This means that cancer immune surveillance does not

    depend on stimulating T cells, but that re-educatingnt macrophages in the tumor microenvironment mayffectiveor more. These results, while not dramatic inical response, can be seen as encouraging and certainlyther clinical investigation of CD40 agonists.

    ng key cytokines in the tumor microenvironment

    was involved in growth of experimental tumors, thenantibodies or other TNF antagonists should have ther-ivity in similar mouse models. This is indeed the case

    in experiments involving carcinogen-induced, trans-enograft and genetic models of common epithelial

    viewed in Ref. [29]). This raised the possibility that itnecial to neutralize TNF- activity in cancer patients.n tested in early Phase clinical trials of TNF antagonistsnts, with some evidence of clinical activity (reviewed inor instance, in a Phase I study using the anti-TNF anti-mab, stabilization of disease was observed in 7 of 41th previously progressing advanced cancer; in a Phasevarian cancer, 6 of 30 progressing patients also showedse after treatment with the TNF- antagonist etaner-ble TNFR2 fusion protein that binds and neutralizes

    in renal cell cancer 14 of 39 patients achieved stableh 3 of 39 obtaining partial responses after iniximabClinical benet of TNF- antagonists has also been seenalignant condition of myelodysplasia.

    as yet no clear idea of mechanisms of action of anti-TNF-patients but nearly twenty years experience in patientsic inammatory disease shows that TNF- antagonistskine and chemokine production, recruitment of inam-s, angiogenesis and extra-cellular matrix degradationions that could be useful in a cancer treatment. In addi-g of TNF antagonists to transmembrane TNF- may

    effects on TNF-producing cells, stimulating a number pathways. Two specic actions of TNF- antagonistsune system in patients with inammatory disease

    cular interest in terms of cancer treatment; modula-function of T regulatory cells and a reduction in Th17ry responses, both of which are implicated in tumor

    apeutic anti-IL-6 antibody siltuximab (CNTO328) hasated in Phase II trials of Castlemans disease [73] andesistant prostate cancer [74]. In Castlemans disease, inis a key pathogenic driver, the objective response rate

    contrast, in prostate cancer, the response rate was 3.2%.bined pre-clinical and in silico experiments with ainical trial of siltuximab in patients with platinum-arian cancer [60]. Automated immunohistochemistryicroarrays from 221 ovarian cancer cases demonstratedty of IL-6 staining in malignant cells signicantly asso-

    poor prognosis. Treatment of ovarian cancer cellsimab reduced constitutive cytokine and chemokine

    and also inhibited IL-6 signaling, tumor growth, theciated macrophage inltrate and angiogenesis in IL-6-ntraperitoneal ovarian cancer xenografts. In the clinicalatient of eighteen evaluable had a partial response,

    others had periods of disease stabilization. In patientssix months, there was a signicant decline in plasma6-regulated CCL2, CXCL12 and VEGF. Gene expression

    inamtumorcan beand cli

    5.3. Ta

    Antinvestitherapof antidecreainducechemosettinghuman

    An shownstromaof CCRtumorited msurvivaunderg

    Cutleukemized dlymphassociacells mClinica

    CXCreceptsurvivatory chas bethe bicaminoCTCE-9metastand pr

    Theagent. cells frfor hemnant csensitito the

    6. Stim

    Innant dtherapbers ofof inadepenswitchment tadaptiantibory cytokine production, tumor angiogenesis and therophage inltrate in ovarian cancer and these actionsited by a neutralizing anti-IL-6 antibody in pre-clinical

    studies [60].

    ng key chemokines and their receptors

    ies against CCL2 or its cognate receptor CCR2 have beend in preclinical models and a strong case for anti CCL2s been made for prostate cancer [67]. Administrationes to CCL2 in mice bearing prostate cancer resulted inumor burden and bone resorption, with lower CCL2-GF release. Combination studies with anti-CCL2 andapy have also yielded improved survival in pre-clinical]. Anti-CCL2 antibodies are currently being evaluated inprostate and ovarian cancer [76].nsive study of CCL2 in breast cancer metastases has

    CCL2 synthesized by metastatic tumor cells and byites of metastases, is critical for continual recruitmentsitive monocytes that enhance the extravasation of. Blockade of CCL2 with neutralizing antibodies inhib-yte recruitment, reduced metastases and prolongedumor-bearing mice [64]. Such anti-CCL2 antibodies are

    clinical evaluation in prostate and ovarian cancer [76].us T cell lymphoma and T-cell acute lymphoblasticxpress the chemokine receptor CCR4. A human-sylated antibody to CCR4 has anti-tumor activity inbearing mice [77]. Anti-CCR4 antibody therapy wasith increased numbers of tumor-inltrating CD56+ NK

    ting ADCC, and reduced the number of FOXP3+ Treg cells.ls are now underway with this antibody.is the most commonly over-expressed chemokine

    human cancer and affects tumor cell proliferation,vasion and metastasis. It is upregulated by inamma-nes such as TNF- and induced by hypoxia. CXCR4argeted by a number of small antagonists, including

    AMD3100 and analogs and peptides designed to theinal region of the chemokine such as T22, TN14003, and

    . CXCR4 antagonists inhibited the primary tumor andin animal models of melanoma, osteosarcoma, breast,e tumors, e.g. [78,79].clam AMD3100 was rst developed as an anti-HIVpectedly, AMD3100 was found to mobilize CD34+ steme bone marrow [80]. At present AMD3100 is clinical useoietic stem cell mobilization [81]. By mobilizing malig-rom the bone marrow niche, AMD3100 enhances theof multiple myeloma or acute myeloid leukemia blastsoxic effect of chemotherapy in preclinical models [81].

    ting good inammation

    ation may not always be bad in the context of malig-e; the cancer cytokine network can also contribute to

    response. It is a question of balance of individual mem-ytokine network. The actions of the cells and cytokinesation in the tumor microenvironment are very context-and one approach to cancer therapy is to attempt totumor-promoting immune suppressive microenviron-

    that kills tumor cells, is anti-angiogenic and promotesmune responses as described above with the anti-CD401].

  • F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40 37

    At the turn of the 19th century a New York surgeon appearedto achieve impressive clinical results by using inammation-stimulating bacterial extracts to treat patients with intractablecancers reviewed in Ref. [29]. We now realize that Coleys mixedtoxins musTLRs, inducrecent apprlocal treatmBCG. Currena good inmacrophagsustained a

    Stressedof danger or secretedfrom non-iinammato17-produciCD8+ T celloutcomes omouse modgen to dencancer patieof metastascan also unchemotheraclinical outways [85].

    Anotherclues aboureceptor asin TLR and iMyD88/sis, but theythat are versearching fomotion, Maactually monation coulchemotheracumstancesTNF- actu

    The exaccan be reliaeven beforesion after sbest stimulimicroenvirothe signalin

    7. Combin

    Over thecancer treatve years rEven theseof cancer cpathway bltumor micrare underwstrategies tgeting CRI mtherapies acombinatiowith CTLA4

    immune escape. As described above, some chemotherapies maycreate a microenvironment that can stimulate good inammationand promote useful adaptive immune responses.

    TAM may also have a profound inuence on a tumors responsemoth

    tumnvirxel cors Cor-exde oatioed m

    this led b

    the mRNA me Alls inion [8

    exterogn

    negant m

    in thindiv

    to brgetireatmn thotatioe ret to r thauted

    vent

    rythi leadotenal an

    sug rece

    cancebo iase [antling wn deg ca

    te caeatestory,].

    potIDs

    signip theancrC? Gis in arbormato

    inlt have been powerful stimulants of Toll-like receptors,ing a range of inammatory mediators. The closestoximation to Coleys work is probably the successfulent of bladder cancer with bacillus CalmetteGuerin,t thinking is that both BCG and Coleys toxins triggerammatory response, via TLRs, that not only stimulateses to kill tumor cells but also promotes development ofnd effective adaptive immunity to the tumor [70].

    and dying tumor cells may emit a particular patternsignals that are either expressed on the cell surface

    into the microenvironment inuencing the switchnammatory silent removal of dead cells to a goodry reaction that can stimulate rapid invasion of IL-ng T cells and generate tumor-specic IFN--producing

    adaptive immune responses that are important to thef some chemotherapies and radiotherapy [82,83]. Inels, dying tumor cells were able to cross-present anti-dritic cells in a TLR4-dependent manner and breastnts with a mutation in TLR4 had an increased frequencyis [84]. Inammasome activation and IL-1 productionderlie activation of protective immunity after cytotoxicpy. Associations of TLR4 and PrX7 polymorphisms withcome are consistent with the relevance of these path-

    paper on TLR signaling and cancer may also give somet good immune responses and TNF [86]. The TLR-sociated signaling adaptor MyD88 plays a critical rolenammatory cytokine signaling. As might be expected

    mice were resistant to DMBA/TPA skin carcinogene- were also resistant to MCA-induced sarcomastumorsy susceptible to immune-surveillance. However, whiler downstream effectors of MyD88-induced tumor pro-rk Smyths group found that TNF knockout mice werere susceptible to MCA-induced sarcoma. One expla-d be that these sarcomas, like tumor cells dying afterpy, are inherently immunogenic and under such cir-

    the inammatory microenvironment stimulated byally protects against tumor development [86].t mechanisms whereby a good inammatory responsebly triggered during cancer therapy are not clear, but

    Coleys time there was evidence for cancer regres-ome bacterial infections. The priority is to nd the

    to change the cytokine network of a tumor-promotingnment to a tumor-inhibiting state and to understandg mechanisms involved.

    ing therapies

    past decade a number of exciting and more targetedments have entered the clinic, a direct result of twenty-esearch into the genetic basis of malignant disease.

    treatments, however, are liable to generate clonesells that overcome this specic receptor or signalingockade. Now that the importance of the inammatoryoenvironment is recognized and early clinical trialsay, the next steps are to devise and test therapeutichat combine targeting the malignant cells with tar-aybe alongside, or in sequence with anti-angiogenic

    nd immune checkpoint blockade. It is also likely thatn of chemotherapy with immune checkpoint blockade

    will provide useful treatments for cancers and block

    to cherecruitmicroepaclitatic factreceptBlockacombininhibitduringparalletion ofand mgranzytor T cedeplet

    Thegood pshow aimportprolefor an seemsthat taother tcially ipresen

    Sominteresmannecontrib

    8. Pre

    Evemationhave psteroidstudies[89]. Adue toor placlar disesignicincreaseffect oand lunprostawas gring hisage [90

    Oneof NSAhighlydeveloacute pof PDAcreatitcells hinamsistenterapy [87] because common cytotoxic drugs mayor-promoting monocytes/macrophages into the tumoronment. In a genetic mouse model of breast cancer,hemotherapy up-regulated the macrophage chemotac-SF1, CCL8 and IL-34 which led to an increase in CSF1pressing macrophages in the tumor microenvironment.f macrophage recruitment with inhibitors of CSFR1 inn with chemotherapy, enhanced therapeutic activity,etastases. The late-stage carcinomas that did developcombination therapy contained large areas of necrosis,y reduced vessel density in the treated tumors. Inhibi-acrophage inltrate also increased T cells in the tumorsfor a number of cytotoxic effector molecules such as

    and B and perforin-1. Depleting CD8+ cytotoxic effec- the mice abrogated the positive effects of macrophage7].nt of the CD8+ T cell inltrate positively correlates withosis in many cancers and macrophage inltrates usuallyative correlation in the same cancers. This work gives usechanistic insights into the role of the tumors immunee response to chemotherapy and hence the outcomeidual patient. The association found by DeNardo et al.e true for different breast cancer subtypes suggestingng leukocytes in the tumor microenvironment duringents is an option that will be widely applicable, espe-

    se patients who have an unfavorable immune prole atn.cent pre-clinical studies suggest that it may also be oftarget TAM during radiotherapy, as macrophages, in at required intact TNF- signaling and VEGF production,

    to radioresistance [88].

    ion strategies

    ng we have learnt so far about cancer-related inam-s to the hypothesis that anti-inammatory agents maytial as cancer preventative agents. In terms of non-ti-inammatory agents, there are several observationalgesting that aspirin reduces risk of certain cancers, e.g.nt study of particular interest was an analysis of deathser in a cohort of patients randomized to receive aspirinn trials originally planned to study prevention of vascu-90]. In eight trials with over 25,000 individuals, aspiriny reduced deaths from solid cancers with the benetith duration of treatment. The latent period before an

    aths was about 5 years for esophageal, pancreatic, brainncer but was more delayed for stomach, colorectal andncer. The overall effect on 20-year risk of cancer deatht for adenocarcinomas and was independent of smok-

    sex, or a dose of aspirin above 75 mg, but increased with

    ential mechanism for the cancer-preventative actionshas recently been reported. Chronic pancreatitis is acant risk factor for PDAC but a majority of patients who

    disease have no history of this condition. Maybe sub-eatic inammation drives the majority of human casesuerra et al. [91] induced sub-acute asymptomatic pan-the PDAC mouse model. Providing the pancreatic ductaled a K-Ras mutation typical of the human disease, thisry stimulus, which induced atrophy, brosis and a per-trate of macrophages and T lymphocytes, was sufcient

  • 38 F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40

    to induce pre-malignant and malignant lesions. The developmentof the malignant lesions was dependent on the extent of tissuedamage and the inammatory response: inammation abrogatedthe senescence barrier characteristic of low-grade pre-malignantlesions. Remmutation wbut treatmeattenuated to PDACs. Rsies, premahad receivesenescencetion; the opwith anti-ininammatomay reduce

    The effein part, dumore specirandomizedpre-malignmonths aftecell carcinocelecoxib [9

    The cancof cytokinemouse modof early cannols inhibitregulating iclude widethousands inammatoduring TNFunderlying treatment wsmall numbof nine doubodies in pcancer wasauthors conmatory disoverall incrmay be neceor concurrean increaseCrohns disin overall ctherapies as

    The admpatients wito inhibit pinhibitors htory disease

    Finally, in long-termtone deaceteffects that

    9. Biomark

    Plasma chemokinein patients ally a poor p

    generally associated with poor prognosis (reviewed in Ref. [99]). Totake the example of prostate cancer, blood TNF- concentrationsare elevated in those patients with advanced, cachectic disease andTNF- levels correlate positively with extent of disease, e.g. [100].

    or svate

    rian -6 led sulasmlinke-6 le

    disearostted y wecuteof cyrease

    progssocies, ptionile inrs of e of ma ork o

    (e.g. h ass

    A ree lev

    te cacreasstina

    inductal cma n cannths2 andufcis thas iny of at th

    respthe mts.

    t of

    e de

    nces

    antovature dir Nyc marivenn010;14iswas te subantovleranian BZetastaarkably, malignant changes happened even if the K-Rasas switched on after inammation-induced damage,nt of mice with a non-steroidal anti-inammatory drugdevelopment of premalignant lesions and progressionelating the mouse results to human pancreas biop-

    lignant lesions in patients with chronic pancreatitis whod anti-inammatory drugs had high expression of the

    marker P16INK4A and no evidence of cell prolifera-posite was seen in biopsies from patients not treatedammatory drugs. These data also suggest that anti-ry treatments in people diagnosed with pancreatitis

    their risk of developing PDAC.cts of NSAIDs such as aspirin are likely to be, at leaste to inhibition of the inammatory enzyme COX2. Ac inhibitor of COX-2, celecoxib, was evaluated in a

    trial of patients with extensive actinic keratoses, theant precursor of nonmelanoma skin cancers. Elevenr start of the study, the incidence of basal and squamousmas was signicantly decreased in the group receiving2].er-preventative potential of more specic antagonistss such as TNF- or IL-6 is more theoretical. Certainlyel experiments suggest a role for TNF- in promotioncers (e.g. [93]). Both herbal medicines and tea polyphe-

    TNF release, e.g. [94] but given the role of TNF- innnate immunity, increased risk of infection would pre-r use of current TNF- antagonists. However, tens ofof people with rheumatoid arthritis and other chronicry disease are being monitored for cancer incidence- antagonist treatment. Analyses are complicated byimmune system dysfunction in these patients, priorith immunosuppressive and mutagenic drugs, and theer of malignancies so far recorded. In one meta-analysisble blind placebo controlled trials of anti-TNF- anti-atients with rheumatoid arthritis an increased risk of

    recorded [95]. However, in a later review the samecluded that with over fty trials of anti-TNF- in inam-ease now published there was no clear evidence forease in cancer risk [96]. The current view is that cautionssary when considering treatment of patients with pastnt cancer or premalignant lesions and there seems to be

    in rare T cell lymphomas in patients with juvenileease [95,96]. There is also no evidence of an increaseancer incidence in patients receiving with anti-TNF-

    compared to a matched cohort of the general public.inistration of the IL-1 receptor antagonist (anakinra) toth indolent, smouldering myeloma has been reportedrogression to aggressive disease [97]. More potent IL-1ave now entered clinical evaluation for autoinamma-s and may provide better tools to interfere with CRI.on a more general note, many drugs currently used

    treatment of common diseases such as statins, his-ylase inhibitors and PPAR agonists, have anti-cytokine

    may give them a cancer-preventative role [98].

    ers of CRI

    or serum levels of inammatory cytokines ands, and in some cases their soluble receptors, are elevatedwith a range of advanced cancers and this is gener-rognostic sign. For example, raised blood TNF- levels

    Plasmaalso elein ovahigh ILreducehigh pare all ever, ILin this

    In passociahealth

    In arange or dechavingtures asubtypproduc

    Whmarkea rangof plasthe womentsby sucaction.cytokinprostaand introintediseasecolorein plasovariasix moCXCL1

    If ssamplerhythmstabilitsible thand/orto be patien

    Conic

    Non

    Refere

    [1] MN

    [2] SoM

    [3] G2

    [4] Bcy

    [5] Mto

    [6] Qmerum levels of IL-6, which increase with age anyway, ared in many solid and hematological cancers. For instancecancer patients, a number of studies have shown thatvels correlate with stage of disease, poor prognosis, andrvival (e.g. [56]). More recent studies have shown thata IL-6, dysregulated cortisol and vegetative depressiond in patients at diagnosis of ovarian cancer [55]. How-vels do not seem to be as useful a biomarker as CA125se.ate cancer high pre-diagnostic IL-6 was signicantlywith time to prostate cancer progression/death inight prostate cancer cases [101].

    myelogenous leukemia, AML, and myelodysplasia atokines and chemokines were consistently elevatedd compared with normal controls with the patternsnostic signicance [102]. There also seemed to be signa-ated with the distinct cytogenetic abnormalities of AMLossibly more evidence that cytokine and chemokine

    in cancer may be driven by oncogenic changes.ammatory cytokines are not as yet seen as useful

    prognosis and response to treatment, we can now studydifferent cytokines and chemokines in small volumesr serum using multiplexing techniques as shown byf [60,102]. It is possible that response to some treat-cytokine antagonists described above) may be assessedays and some insight may be gained on mechanisms ofcent study, for example, used longitudinal proling ofels during intensity-modulated radiotherapy, IMRT, forncer. Both IFN- and IL-6 were raised during treatmenting levels of IL-2 and IL-1 were associated with gas-l and genitourinary toxicity respectively [103]. Stableced by dendritic cell-based vaccination in patients withancer, was associated with increased Th1 cell cytokines[104]. In a Phase II trial of an anti-IL-6 antibody incer patients, those patients who had stable disease for

    or longer showed signicant declines in plasma CCL2, VEGFall factors that can be induced by IL-6 [60].ent care is taken in preparation and storage of bloodt are taken at a standard time of day (because of diurnal

    cytokine levels) and we have better knowledge of thecytokines in plasma and serum during storage, it is pos-ese mediators may be useful biomarkers of prognosisonse to treatment, especially if we consider the tumorajor source of elevated cytokines in blood of cancer

    interest

    clared.

    ani A, Allavena P, Sica A, Balkwill F. Cancer-related inammation.2008;454:43644.M, Swigart LB, Karnezis AN, Hanahan D, Evan GI, Soucek L. Endogenousintains the tumor microenvironment. Genes Dev 2011;25:90716.ikov SI, Greten FR, Karin M. Immunity, inammation, and cancer. Cell0:88399.SK, Mantovani A. Macrophage plasticity and interaction with lympho-sets: cancer as a paradigm. Nat Immunol 2010;11:88996.ani A, Sica A. Macrophages, innate immunity and cancer: balance,ce, and diversity. Curr Opin Immunol 2010;22:2317., Pollard JW. Macrophage diversity enhances tumor progression andsis. Cell 2010;141:3951.

  • F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40 39

    [7] Mosser DM, Edwards JP. Exploring the full spectrum of macrophage activa-tion. Nat Rev Immunol 2008;8:95869.

    [8] Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K. Developmentof monocytes, macrophages, and dendritic cells. Science 2010;327:65661.

    [9] Movahedi K, Laoui D, Gysemans C, Baeten M, Stange G, Van den BosscheJ, et al.subsets2010;70

    [10] Andreuactivaticer Cell

    [11] DeNardet al. Cnomas 2009;16

    [12] MarkiewKoutoulby comp

    [13] Ammiraderived2010;46

    [14] SchioppB regulasquamo

    [15] OlkhanuTumor-verting

    [16] Wong SMacropImmuno

    [17] Bottazzcuit in tsarcoma

    [18] Jenkins Local msignatu

    [19] HamiltoNat Rev

    [20] Dvorak 1986;31

    [21] Erez N, vated inan NF-k

    [22] Spaeth Mesenctributes2009;4:

    [23] ElkabetHumanmote m2011;12

    [24] KramanSuppresactivati

    [25] Mantovvation a2011;11

    [26] Fridlendof tumoCancer

    [27] Nozawaangioge2006;10

    [28] Soucek requiredcreatic i

    [29] Balkwil[30] Kulbe H

    The innetwork

    [31] Li B, Vinels of tuendothe2009;69

    [32] Hagemaincreasewith ma

    [33] Charleset al. Thovarian

    [34] Pikarskyfunction2004;43

    [35] Maeda death tohepatoc

    [36] Komori J, Marusawa H, Machimoto T, Endo Y, Kinoshita K, Kou T, et al.Activation-induced cytidine deaminase links bile duct inammation tohuman cholangiocarcinoma. Hepatology 2008;47:88896.

    [37] Kim S, Takahashi H, Lin W-W, Descargues P, Grivennikov S, Kim Y, et al.Carcinoma-produced factors activate myeloid cells through TLR2 to stimulateetastaarrisomor n

    al pha07;25

    l-Lamecrosiatholike Mse of arrounorderonic Rev Celidal Mev 200augler

    immrivenn-6 andment

    ollrath130-mll-cyc09;15

    ncrile quireain Ktegrinplicaion. Cauglerer disproductao D, rogeniiencera T, terleud suruan Z,ucers aarian

    CW,gnalinancer ilssonn car05;65iyahad reg08;10tgend

    al. Inatientsambignicaancerante Mlevel iarianccani50 nuchibits06;66

    offelt Se proroughcad Scnglesiat3-Hnitinngowromotantovstem 10;21

    etranoe lymsociat10;59 Different tumor microenvironments contain functionally distinct of macrophages derived from Ly6C(high) monocytes. Cancer Res:572839.

    P, Johansson M, Affara NI, Pucci F, Tan T, Junankar S, et al. FcRgammaon regulates inammation-associated squamous carcinogenesis. Can-

    2010;17:12134.o DG, Barreto JB, Andreu P, Vasquez L, Tawk D, Kolhatkar N,D4(+) T cells regulate pulmonary metastasis of mammary carci-by enhancing protumor properties of macrophages. Cancer Cell:91102.ski MM, DeAngelis RA, Benencia F, Ricklin-Lichtsteiner SK,aki A, Gerard C, et al. Modulation of the antitumor immune responselement. Nat Immunol 2008;9:122535.nte M, Luo JL, Grivennikov S, Nedospasov S, Karin M. B-cell-

    lymphotoxin promotes castration-resistant prostate cancer. Nature4:3025.a T, Moore R, Thompson RG, Rosser EC, Kulbe H, Nedospasov S, et al.tory cells and the tumor-promoting actions of TNF-{alpha} duringus carcinogenesis. Proc Natl Acad Sci U S A 2011;108:106627.d PB, Damdinsuren B, Bodogai M, Gress RE, Sen R, Wejksza K, et al.evoked regulatory B cells promote breast cancer metastasis by con-resting CD4T cells to T-regulatory cells. Cancer Res 2011;71:350515.C, Puaux AL, Chittezhath M, Shalova I, Kajiji TS, Wang X, et al.hage polarization to a unique phenotype driven by B cells. Eur Jl 2010;40:2296307.

    i B, Erba E, Nobili N, Fazioli F, Rambaldi A, Mantovani A. A paracrine cir-he regulation of the proliferation of macrophages inltrating murines. J Immunol 1990;144:240912.SJ, Ruckerl D, Cook PC, Jones LH, Finkelman FD, van Rooijen N, et al.acrophage proliferation, rather than recruitment from the blood, is are of TH2 inammation. Science 2011;332:12848.n JA. Colony-stimulating factors in inammation and autoimmunity.

    Immunol 2008;8:53344.HF. Tumors: wounds that do not heal. New Engl J Med5:16506.Truitt M, Olson P, Hanahan D. Cancer-associated broblasts are acti-

    incipient neoplasia to orchestrate tumor-promoting inammation inappaB-dependent manner. Cancer Cell 2010;17:13547.EL, Dembinski JL, Sasser AK, Watson K, Klopp A, Hall B, et al.hymal stem cell transition to tumor-associated broblasts con-

    to brovascular network expansion and tumor progression. PLoS Onee4992.s M, Gifford AM, Scheel C, Nilsson B, Reinhardt F, Bray MA, et al.

    tumors instigate granulin-expressing hematopoietic cells that pro-alignancy by activating stromal broblasts in mice. J Clin Invest1:78499.

    M, Bambrough PJ, Arnold JN, Roberts EW, Magiera L, Jones JO, et al.sion of antitumor immunity by stromal cells expressing broblaston protein-alpha. Science 2010;330:82730.ani A, Cassatella MA, Costantini C, Jaillon S. Neutrophils in the acti-nd regulation of innate and adaptive immunity. Nat Rev Immunol:51931.er ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, et al. Polarizationr-associated neutrophil phenotype by TGF-beta: N1 versus N2 TAN.Cell 2009;16:18394.

    H, Chiu C, Hanahan D. Inltrating neutrophils mediate the initialnic switch in a mouse model of multistage carcinogenesis. PNAS3:124938.L, Lawlor ER, Soto D, Shchors K, Swigart LB, Evan GI. Mast cells are

    for angiogenesis and macropscopic expansion of Myc-induced pan-slet tumors. Nat Med 2007;13:12118.l F. Tumour necrosis factor and cancer. Nat Rev Cancer 2009;9:36171., Thompson RT, Wilson J, Robinson SC, Hagemann T, Fatah R, et al.ammatory cytokine TNF-a generates an autocrine tumour-promoting

    in epithelial ovarian cancer cells. Cancer Res 2007;67:58592.cent A, Cates J, Brantley-Sieders DM, Polk DB, Young PP. Low lev-mor necrosis factor alpha increase tumor growth by inducing anlial phenotype of monocytes recruited to the tumor site. Cancer Res:33848.nn T, Wilson J, Kulbe H, Charles K, Li F, Pukrop T, et al. TNF-a dependentd c-Jun and NF-k B activity in tumour cell lines upon co-cultivationcrophages. J Immunol 2005;175:1197205.

    KA, Kulbe H, Soper R, Escorcio-Correia M, Lawrence T, Schultheis A,e tumor-promoting actions of TNF-alpha involve TNFR1 and IL-17 in

    cancer in mice and humans. J Clin Invest 2009;119:301123. E, Porat RM, Stein I, Abramovitch R, Amit S, Kasem S, et al. NF-kBs as a tumour promoter in inammation-associated cancer. Nature1:44616.S, Kamata H, Luo JL, Leffert H, Karin M. IKKbeta couples hepatocyte

    cytokine-driven compensatory proliferation that promotes chemicalarcinogenesis. Cell 2005;121:97790.

    m[38] H

    Tuti20

    [39] AnJ P

    [40] SeUsu

    [41] CgeD

    [42] VD

    [43] Nin

    [44] GILop

    [45] Bgpce20

    [46] Are

    [47] Shinimat

    [48] Ndp

    [49] GpSc

    [50] AInan

    [51] Ddov

    [52] LosiC

    [53] Nia20

    [54] Man20

    [55] Luetp

    [56] ScsiJ C

    [57] Pl6 ov

    [58] Sapin20

    [59] CThthA

    [60] AStSu

    [61] Lap

    [62] Msy20

    [63] VThas20sis. Nature 2009;457:1026.n ML, Obermueller E, Maisey NR, Hoare S, Edmonds K, Li NF, et al.ecrosis factor a as a new target for renal cell carcinoma: two sequen-se II trials of iniximab at standard and high dose. J Clin Oncol:45429.

    ki RS, Sadler TJ, Wang J, Reid MJ, Warren AY, Movassagh M, et al. Tumors factor receptor expression and signaling in renal cell carcinoma. Am

    2010;177:94354., Yanaihara N, Bowman ED, Zanetti KA, Budhu A, Kumamoto K, et al.

    cytokine gene expression signature in lung adenocarcinoma and theding tissue as a prognostic classier. JNCI 2007;99:125769.

    JB, Macagno JP, Stefanatos RK, Strathdee KE, Cagan RL, Vidal M. Onco-as diverts a host TNF tumor suppressor activity into tumor promoter.l 2010;18:9991011., Cagan RL. Drosophila models for cancer research. Curr Opin Genet6;16:106.

    WE, Karin M. The wolf in sheeps clothing: the role of interleukin-6unity, inammation and cancer. Trends Mol Med 2008;14:10919.ikov S, Karin E, Terzic J, Mucida D, Yu G-Y, Vallabhapurapu S, et al.

    Stat3 are required for survival of intestinal epithelial cells and devel- of colitis-associated cancer. Cancer Cell 2009;15:10313.

    J, Phesse TJ, von Burstin VA, Putoczki T, Bennecke M, Bateman T, et al.ediated Stat3 activation in enterocytes regulates cell survival and

    le progression during colitis-associated tumorigenesis. Cancer Cell:91102.B, Lim K-H, Counter CM. Oncogenic Ras-induced secretion of IL-6 isd for tumorigenesis. Genes Dev 2007;21:17149.H, Yarde DN, Meads MB, Huang M, Jove R, Hazlehurst LA, et al. 1

    adhesion enhances IL-6-mediated STAT3 signaling in myeloma cells:tions for microenvironment inuence on tumor survival and prolifer-ancer Res 2009;69:100915.

    WE, Sakurai T, Kim S, Maeda S, Kim KH, Elsharkawy AM, et al. Gen-arity in liver cancer due to sex differences in MyD88-dependent IL-6ion. Science 2007;317:1214.Nolan DJ, Mellick AS, Bambino K, McDonnell K, Mittal V. Endothelialtor cells control the angiogen in switch in mouse lung metastasis.

    2008;319:1958.Song L, Shimada H, Keshelava N, Russell HV, Metelitsa LS, et al.kin-6 in the bone marrow microenvironment promotes the growthvival of neuroblastoma cells. Cancer Res 2009;69:32937.

    Foster R, Bell DA, Mahoney J, Wolak K, Vaidya A, et al. Signal trans-nd activators of transcription 3 pathway activation in drug-resistant

    cancer. Clin Cancer Res 2006;12:505563. Chen MW, Hsiao M, Wang S, Chen CA, Hsiao SM, et al. IL-6 trans-g in formation and progression of malignant ascites in ovarian cancer.Res 2011;71:42434.

    MB, Langley RR, Fidler IJ. Interleukin-6, secreted by human ovar-cinoma cells, is a potent proangiogenic cytokine. Cancer Res:10794800.ra Y, Odunsi K, Chen W, Peng G, Matsuzaki J, Wang R-F. Generationulation of human CD4+ IL-17-producing T cells in ovarian cancer. PNAS5:1550510.orf SK, Weinrib AZ, Penedo F, Russell DW, DeGeest K, Costanzo ES,terleukin-6, cortisol, and depressive symptoms in ovarian cancer. J Clin Oncol 2008;26:48207.a G, Testa U, Panici PB, Foti E, Martucci R, Gadducci A, et al. Prognosticnce of interleukin 6 serum levels in patients with ovarian cancer. Br

    1995;71:3546., Rubin SC, Wong GY, Federici MG, Finstad CL, Gastl GA. Interleukin-n serum and ascites as a prognostic factor in patients with epithelial

    cancer. Cancer 1994;73:18828. A, Schioppa T, Porta C, Biswas SK, Nebuloni M, Vago L, et al.lear factor-kappaB overexpression in tumor-associated macrophages

    M1 inammatory responses and antitumor resistance. Cancer Res:1143240.B, Marini FC, Watson K, Zwezdaryk KJ, Dembinski JL, LaMarca HL, et al.-inammatory peptide LL-37 promotes ovarian tumor progression

    recruitment of multipotent mesenchymal stromal cells. Proc Natli U S A 2009;106:380611.o MS, George J, Kulbe H, Friedlander ML, Rischin D, Lemech C, et al. Il6-if signalling and therapeutic response to the angiogenesis inhibitor,ib, in ovarian clear cell cancer. Clin Cancer Res 2011;17:253848.ski JL, Zhang X, Lingling W, Mattson JD, Chen T, Smith K, et al. IL-23es tumour incidence and growth. Nature 2006;442:4615.ani A, Savino B, Locati M, Zammataro PA, Bonecchi R. The chemokinein cancer biology and therapy. Cytokine Growth Factor Rev:2739.

    S, Borroni EM, Sarukhan A, Savino B, Bonecchi R, Correale C, et al.phatic system controls intestinal inammation and inammation-ed colon cancer through the chemokine decoy receptor D6. Gut:197206.

  • 40 F.R. Balkwill, A. Mantovani / Seminars in Cancer Biology 22 (2012) 33 40

    [64] Qian BZ, Li J, Zhang H, Kitamura T, Zhang J, Campion LR, et al. CCL2recruits inammatory monocytes to facilitate breast-tumour metastasis.Nature 2011;475:2225.

    [65] Bottazzi B, Polentarutti N, Acero R, Balsari A, Boraschi D, Ghezzi P, et al. Regu-lation of the macrophage content of neoplasms by chemoattractants. Science1983;220:2102.

    [66] Nibbs RJ, Gilchrist DS, King V, Ferra A, Forrow S, Hunter KD, et al. The atypicalchemokine receptor D6 suppresses the development of chemically inducedskin tumors. J Clin Invest 2007;117:188492.

    [67] Zhang J, Patel L, Pienta KJ. CC chemokine ligand 2 (CCL2) promotesprostate cancer tumorigenesis and metastasis. Cytokine Growth Factor Rev2010;21:418.

    [68] De Monte L, Reni M, Tassi E, Clavenna D, Papa I, Recalde H, et al. Intratu-mor T helper type 2 cell inltrate correlates with cancer-associated broblastthymic stromal lymphopoietin production and reduced survival in pancreaticcancer. J Exp Med 2011;208:46978.

    [69] Facciabene X, Peng IS, Hagemann K, Balint A, Barchetti LP, Wang PA, et al.Tumour hypoxia promotes tolerance and angiogenesis via CCL28 and Tregcells. Nature 2011;475:22630.

    [70] Hagemann T, Lawrence T, McNeish I, Charles KA, Kulbe H, Thompson RC, et al.Re-educating tumor-associated macrophages by targeting NF-kB. J Exp Med2008;205:12618.

    [71] Beatty GL, Chiorean EG, Fishman MP, Saboury B, Teitelbaum UR, Sun W, et al.CD40 agonists alter tumor stroma and show efcacy against pancreatic car-cinoma in mice and humans. Science 2011;331:16126.

    [72] Tracey D, Klareskog L, Sasso EH, Salfeld JG, Tak PP. Tumor necrosis factorantagonist mechanisms of action: a comprehensive review. Pharmacol Ther2008;11

    [73] van Rhetuximabdisease

    [74] Dorff TBClinical(siltuximpretrea2010;16

    [75] Rozel S,anti-CCcancer m

    [76] Garber 2009;10

    [77] Ito A, IsCCR4 min the mouse m

    [78] Kim SY,the CXCpulmon

    [79] PorvasnEffects 2009;69

    [80] Liles WCtion of hCXCR4 a

    [81] De Clermobiliz

    [82] Zitvogeand imm

    [83] Ma Y, Atributiochemot

    [84] Apetoh like recchemot

    [85] Ghiringhelli F, Apetoh L, Tesniere A, Aymeric L, Ma Y-Y, Ortiz C, et al. Activationof the NLRP3 inammasome in dendritic cells induces IL-1beta-dependentadaptive immunity against tumors. Nat Med 2009;15:11708.

    [86] Swann JB, Vesely MD, Silva A, Sharkey J, Akira S, Schreiber RD, et al. Demon-stration of inammation-induced cancer and cancer immunoediting duringprimary tumorigenesis. PNAS 2008;105:6526.

    [87] DeNardo DG, Brennan DJ, Rexhepaj E, Ruffell B, Shiao SL, Madden SF, et al.Leukocyte complexity predicts breast cancer survival and functionally regu-lates response to chemotherapy. American Association for Cancer Research.Cancer Discov 2011:216.

    [88] Meng Y, Beckett MA, Liang H, Mauceri HJ, van Rooijen N, Cohen KS,et al. Blockade of tumor necrosis factor alpha signaling in tumor-associatedmacrophages as a radiosensitizing strategy. Cancer Res 2010;70:153443.

    [89] Cuzick J, Otto F, Baron JA, Brown PH, Burn J, Greenwald P, et al. Aspirin andnon-steroidal anti-inammatory drugs for cancer prevention: an interna-tional consensus statement. Lancet Oncol 2009;10:5017.

    [90] Rothwell PM, Fowkes FG, Belch JF, Ogawa H, Warlow CP, Meade TW. Effect ofdaily aspirin on long-term risk of death due to cancer: analysis of individualpatient data from randomised trials. Lancet 2011;377:3141.

    [91] Guerra C, Collado M, Navas C, Schuhmacher AJ, Hernandez-Porras I, CanameroM, et al. Pancreatitis-induced inammation contributes to pancreatic can-cer by inhibiting oncogene-induced senescence. Cancer Cell 2011;19:72839.

    [92] Elmets CA, Viner JL, Pentland AP, Cantrell W, Lin HY, Bailey H, et al.Chemoprevention of nonmelanoma skin cancer with celecoxib: a random-ized, double-blind, placebo-controlled trial. J Natl Cancer Inst 2010;102:183544.oore ecrosi999;5:jiki HF-a redicinr tamoongartF antd masklingurr Opst JA,ductiont mu

    produ009;84inarel010;14ndersonesisichalapha catientark JRg precidenornblapress

    endenlood 2hristenngitueatmeurgdoiomarll-bas7:24479.e F, Fayad L, Voorhees P, Furman R, Lonial S, Borghaei H, et al. Sil-, a novel anti-interleukin-6 monoclonal antibody, for Castlemans

    . J Clin Oncol 2010;28:37018., Goldman B, Pinski JK, Mack PC, Lara Jr PN, Van Veldhuizen Jr PJ, et al.

    and correlative results of SWOG S0354: a phase II trial of CNTO328ab), a monoclonal antibody against interleukin-6, in chemotherapy-

    ted patients with castration-resistant prostate cancer. Clin Cancer Res:302834.

    Galban CJ, Nicolay K, Lee KC, Sud S, Neeley C, et al. Synergy betweenL2 and docetaxel as determined by DW-MRI in a metastatic boneodel. J Cell Biochem 2009;107:5864.

    K. First results for agents targeting cancer-related inammation. JNCI1:11102.hida T, Yano H, Inagaki A, Suzuki S, Sato F, et al. Defucosylated anti-onoclonal antibody exercises potent ADCC-mediated antitumor effectnovel tumor-bearing humanized NOD/Shi-scid, IL-2Rgamma(null)odel. Cancer Immunol Immunother 2009;58:1195206.

    Lee CH, Midura BV, Yeung C, Mendoza A, Hong SH, et al. Inhibition ofR4/CXCL12 chemokine pathway reduces the development of murineary metastases. Clin Exp Metastasis 2008;25:20111.ik S, Sakamoto N, Kusmartsev S, Eruslanov E, Kim WJ, Cao W, et al.of CXCR4 antagonist CTCE-9908 on prostate tumor growth. Prostate:14609., Broxmeyer HE, Rodger E, Wood B, Hubel K, Cooper S, et al. Mobiliza-ematopoietic progenitor cells in healthy volunteers by AMD3100, antagonist. Blood 2003;102:272830.cq E. The AMD3100 story: the path to the discovery of a stem celler (Mozobil). Biochem Pharmacol 2009;77:165564.l L, Kepp O, Kroemer G. Decoding cell death signals in inammationunity. Cell 2010;140:798804.

    ymeric L, Locher C, Mattarollo SR, Delahaye NF, Pereira P, et al. Con-n of IL-17-producing gamma delta T cells to the efcacy of anticancerherapy. J Exp Med 2011;208:491503.L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A, et al. Toll-eptor 4-dependent contribution of the immune system to anticancerherapy and radiotherapy. Nat Med 2007;13:10509.

    [93] Mn1

    [94] FuTNmo

    [95] BTNan

    [96] AC

    [97] LuInle62

    [98] D2

    [99] Age

    [100] Malp

    [101] Stinin

    [102] KexpB

    [103] CLotr

    [104] BbceR, Owens D, Stamp G, East N, Holdworth H, Arnott C, et al. Tumours factor-a decient mice are resistant to skin carcinogenesis. Nat Med82831., Suganuma M, Kurusu M, Okabe S, Imayoshi Y, Taniguchi S, et al. Neweleasing inhibitors as cancer preventive agents from traditional herbale and combination cancer prevention study with EGCG and sulindacxifen. Mutat Res 2003;523:11925.z T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. Anti-ibody therapy in rheumatoid arthritis and the risk of serious infectionslignancies. JAMA 2006;295:227585.

    J, Bongartz T. Malignancy and biologic therapy in rheumatoid arthritis.in Rheumatol 2008;20:3349.

    Lacy MQ, Zeldenrust SR, Dispenzieri A, Gertz MA, Witzig TE, et al.n of a chronic disease state in patients with smoldering or indo-ltiple myeloma by targeting interleukin 1{beta}-induced interleukinction and the myeloma proliferative component. Mayo Clin Proc:11422.lo CA. Anti-inammatory agents: present and future. Cell0:93550.n GM, Nakada MT, DeWitte M. Tumor necrosis factor-a in the patho-

    and treatment of cancer. Curr Opin Pharmacol 2004;4:31420.ki V, Syrigos K, Charles P, Waxman J. Serum levels of IL-6 and TNF-orrelate with clinicopathological features and patient survival ins with prostate cancer. Br J Cancer 2004;91:1227., Li H, Kraft P, Kurth T, Giovannucci EL, Stampfer MJ, et al. Circulat-diagnostic interleukin-6 and C-reactive protein and prostate cancerce and mortality. Int J Cancer 2009;124:26839.u SM, McCue D, Singh N, Chen W, Estrov Z, Coombes KR. Recurrention signatures of cytokines and chemokines are present and are inde-tly prognostic in acute myelogenous leukemia and myelodysplasia.010;116:425161.sen E, Pintilie M, Evans KR, Lenarduzzi M, Menard C, Catton CN, et al.dinal cytokine expression during IMRT for prostate cancer and acutent toxicity. Clin Cancer Res 2009;15:557683.rf SK, Claesson MH, Nielsen HJ, Rosenberg J. Changes in cytokine andker blood levels in patients with colorectal cancer during dendriticed vaccination. Acta Oncol 2009:18.

    Cancer-related inflammation: Common themes and therapeutic opportunities1 Introduction to cancer-related inflammation2 Polarization and diversity of inflammatory cells in the tumor microenvironment3 Cytokines and cancer-related inflammation4 Chemokines and cancer-related inflammation5 Therapeutic opportunities5.1 Targeting TAM5.2 Targeting key cytokines in the tumor microenvironment5.3 Targeting key chemokines and their receptors

    6 Stimulating good inflammation7 Combining therapies8 Prevention strategies9 Biomarkers of CRIConflict of interestReferences