InhibitionofCSF-1ReceptorImprovestheAntitumorEf cacy of ... · Thomas G. Graeber1,2,4,7, Brian L. West 9, Gideon Bollag , and Antoni Ribas1,2,3,4,8 Abstract Colony stimulating factor

Microenvironment and Immunology

Inhibition ofCSF-1Receptor Improves theAntitumor Efficacyof Adoptive Cell Transfer Immunotherapy

Stephen Mok1, Richard C. Koya10, Christopher Tsui8, Jingying Xu1, Lídia Robert8, Lily Wu1,4,5,6,Thomas G. Graeber1,2,4,7, Brian L. West9, Gideon Bollag9, and Antoni Ribas1,2,3,4,8

AbstractColony stimulating factor 1 (CSF-1) recruits tumor-infiltrating myeloid cells (TIM) that suppress tumor

immunity, including M2 macrophages and myeloid-derived suppressor cells (MDSC). The CSF-1 receptor(CSF-1R) is a tyrosine kinase that is targetable by small molecule inhibitors such as PLX3397. In this study, weused a syngeneic mouse model of BRAFV600E-driven melanoma to evaluate the ability of PLX3397 to improvethe efficacy of adoptive cell therapy (ACT). In this model, we found that combined treatment producedsuperior antitumor responses compared with single treatments. In mice receiving the combined treatment, adramatic reduction of TIMs and a skewing of MHCIIlow to MHCIIhi macrophages were observed. Further-more, mice receiving the combined treatment exhibited an increase in tumor-infiltrating lymphocytes (TIL)and T cells, as revealed by real-time imaging in vivo. In support of these observations, TILs from these micereleased higher levels of IFN-g . In conclusion, CSF-1R blockade with PLX3397 improved the efficacy of ACTimmunotherapy by inhibiting the intratumoral accumulation of immunosuppressive macrophages. CancerRes; 74(1); 153–61. �2013 AACR.

IntroductionEstablished solid tumors consist of both transformed neo-

plastic cells and nontransformed host cells such as stromalcells, lymphocytes, dendritic cells, macrophages, and myeloid-derived suppressor cells (MDSC). In order to escape immuneresponses, tumor cells manipulate the surrounding tumormicroenvironment by producing cytokines that suppress cyto-lytic T cells and recruit immunosuppressive cells (1–3). Colony-stimulating factor 1 (CSF-1) is a cytokine frequently producedby several cancers, including melanoma (4, 5). The secretedCSF-1 binds to the tyrosine kinase receptor CSF-1 receptor(CSF-1R) on the myeloid cells, which results in increasedproliferation and differentiation of myeloid cells into typeM2 macrophages and MDSCs, and their recruitment intotumors (6, 7). The M2-polarized macrophages and MDSCs use

several mechanisms to induce an immunosuppressive tumorenvironment, such as the release of arginase I or induciblenitric oxide synthase, leading to T-cell inhibition (1). Therefore,an immunosuppressive tumor milieu mediated by CSF-1 maylimit the antitumor activity of tumor immunotherapy and leadto low response rates (3).

In prior studies, we have established a BRAFV600E mutantmurinemelanoma cell line, SM1, to provide a relevantmodel ofmelanoma in fully syngeneic immunocompetent mice (8).BRAFV600E is the driver oncogene in approximately 50% ofhuman melanomas (9). Besides being driven by the BRAFV600E

oncogene, SM1 has multiple genomic aberrations in a patternsimilar to 108 human melanoma cell lines based on results ofhigh-density single-nucleotide polymorphism/copy numberalteration arrays. It includes amplification of oncogenicBRAFV600E and of themicrophthalmia-associated transcriptionfactor, and a deletion of CDKN2A. In the SM1 model, adoptivecell transfer (ACT) of melanoma-targeted T cells inducesantitumor responses that are enhanced by the addition of theBRAF inhibitor vemurafenib (8). In addition, this mousemodelhas been used to test the antitumor effects of several immunemodulating antibodies, including anti-CTLA4, anti-PD-1, anti-TIM3, and agonistic anti-CD137 (41BB; ref. 10). Only anti-CD137 had significant antitumor activity alone or in combi-nation with BRAF inhibitor therapy, suggesting that there maybe factors in the tumor microenvironment that inhibit theeffector arm of the immune system.

PLX3397 is a potent tyrosine kinase inhibitor that is selectedfor its ability to inhibit CSF-1R. It is currently in clinicaldevelopment as a single agent or in combination therapy forthe treatment of patients with glioblastoma, breast cancer, andother cancers through its inhibition of the CSF-1R, and also

Authors' Affiliations: 1Department of Molecular and Medical Pharmacol-ogy; 2Jonsson Comprehensive Cancer Center; 3Surgery, Division of Sur-gical Oncology; 4Institute forMolecular Medicine; Departments of 5Urologyand 6Pediatrics; 7Crump Institute for Molecular Imaging; 8Department ofMedicine, Division of Hematology-Oncology, University of California, LosAngeles, Los Angeles; 9Plexxikon Inc., Berkeley, California; and 10RoswellPark Cancer Institute, Buffalo, New York

S. Mok and R.C. Koya contributed equally to this article.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

CorrespondingAuthors:Antoni Ribas, Division ofHematology-Oncology,11-934 Factor Building, 10833 Le Conte Avenue, Los Angeles, CA 90095-1782. Phone: 310-206-3928; Fax: 310-825-2493; E-mail:[email protected]; and Richard C. Koya,[email protected]

doi: 10.1158/0008-5472.CAN-13-1816

�2013 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 153

on November 5, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst November 18, 2013; DOI: 10.1158/0008-5472.CAN-13-1816

http://cancerres.aacrjournals.org/

acute myelogenous leukemia through its inhibition of FLT3-ITD. In preclinical models, CSF-1R inhibitors includingPLX3397 have been reported to inhibit the immunosuppressivetumor milieu and facilitate immune responses to cancer(11–14). We hypothesized that a combination of PLX3397 andACT immunotherapy would improve the tumor microenvi-ronment through the inhibition of immunosuppressive mye-loid cells, resulting in better T-cell antitumor functions. Ourresults demonstrate significantly enhanced efficacy of thecombined treatment, mediated by decreasing TIMs andincreasing activated tumor-infiltrating lymphocytes (TIL)compared with either of the single treatment groups.

Materials and MethodsMice, cell lines, and reagents

C57BL/6 mice, OT-1 transgenic mice (The Jackson Labora-tories), pmel-1 (Thy1.1) transgenic mice (kind gift from Dr.Nicholas Restifo, Surgery Branch, National Cancer Institute,Bethesda, MD), and NOD/SCID/g chainnull (NSG) mice (NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ; The Jackson Laboratory) were bredand kept under defined-flora pathogen-free conditions at theAssociation for Assessment and Accreditation of LaboratoryAnimal Care International-approved animal facility of theDivision of Experimental Radiation Oncology, University ofCalifornia, Los Angeles (UCLA), Los Angeles, CA, and usedunder the UCLA Animal Research Committee protocol #2004-159. The B16murinemelanoma cell line was obtained from theAmerican Type Culture Collection and maintained in Dulbec-co's Modified Eagle Medium (Mediatech, Inc.) with 10% fetalcalf serum (FCS; Omega Scientific) and 1% penicillin, strepto-mycin, and amphotericin (Omega Scientific). The SM1 murinemelanoma was generated from a spontaneously arising tumorin BRAFV600E mutant transgenic mice as previously described(15). SM1 was maintained in RPMI (Mediatech) with 10% FCS(Omega Scientific), 2 mmol/L L-glutamine (Invitrogen) and 1%penicillin, streptomycin, and amphotericin. SM1-OVA wasgenerated by the stable expression of ovalbumin (OVA)through lentiviral transduction as previously described (15).PLX3397 was obtained under a materials transfer agreementwith Plexxikon Inc.. PLX3397 was dissolved in dimethyl sulf-oxide (DMSO; Fisher Scientific). For in vivo studies, PLX3397was dissolved in DMSO, and then in a suspension made bydilution into an aqueousmixture of 0.5%hydroxypropylmethylcellulose and 1% polysorbate (PS80; Sigma-Aldrich). Of note,100 mL of the suspended drug was administered by daily oralgavage into mice at 50 mg/kg when tumors reached 3 mm indiameter. For macrophage depletion studies, 1 mg of clodro-nate (Clodrosome) was injected intraperitoneally (i.p.) every 5days. For antibody-mediated depletion studies, 250 mg of anti-CD8 antibody, 200mg of anti-CSF-1, or isotype control antibody(BioXCell) was injected i.p. every 3 days.

Cell viability assaysMurine melanoma cells (5� 103 cells per well) and activated

C57BL/6 splenocytes (5 � 104 cells per well) were seeded on96-well flat-bottomed plates with 100 mL of 10% FCS mediaand incubated for 24 hours. Graded dilutions of PLX3397 or

DMSO vehicle control, in culture medium, were added to eachwell in triplicate and analyzed byusing a tetrazoliumcompoundMTS-based colorimetric cell proliferation assay (Promega).

Adoptive cell transfer therapy in vivo modelsB16, SM1-OVA, or SM1 cells were implanted subcutane-

ously in C57BL/6 mice, and when tumors reached 5 mm indiameter, mice were conditioned for ACT with a lympho-depleting regimen of 500 cGy of total body irradiation. Thenthey received 2� 105 or 1� 106 OVA257-264 peptide-activatedOT-1 splenocytes or gp10025-33 peptide-activated pmel-1splenocytes intravenously as previously described (15). Inboth cases, the ACT was followed by 3 days of daily intra-peritoneal administration of 50,000 IU of interleukin (IL)-2.Tumors were followed by caliper measurements three timesper week.

Flow cytometry analysisSM1 tumors, lungs, blood, bone marrow, and spleens were

harvested from mice. Tumors and lungs were further digestedwith collagenase (Sigma-Aldrich). TIMs obtained fromdigested SM1 tumors were stained with antibodies to Gr-1,CD11b, F4/80, MHCII (eBiosciences), and Ly6C (BD Bios-ciences). TILs were stained with antibodies with CD3, Thy1.1(BD Biosciences), CD4, and CD8 (eBiosciences) and analyzedwith a LSR-II or FACSCalibur flow cytometers (BD Bio-sciences), followed by Flow-Jo software (Tree-Star) analysisas previously described (16). Intracellular IFN-g staining wasdone as previously described (16).

Immunofluorescence imagingStaining was performed as previously described (15).

Briefly, sections of OCT (Sakura Finetek) cryopreservedtissues were blocked in donkey serum/PBS and incubatedwith primary antibodies to Gr-1 (BD Biosciences) or F4/80(Abcam), followed by secondary donkey anti-rat antibodiesconjugated to DyLight488 (Jackson Immunoresearch Labo-ratories). Negative controls consisted of isotype-matchedrabbit or rat immunoglobulin G in lieu of the primaryantibodies listed earlier. 40,6-Diamidino-2-phenylindole(DAPI) was used for the visualization of nuclei. Immunoflu-orescence images were taken in a fluorescence microscope(Axioplan-2; Carl Zeiss Microimaging).

Bioluminescence imagingOT-1 or pmel-1 splenocytes were retrovirally transduced to

express firefly luciferase as previously described (15), and usedfor ACT. Bioluminescence imaging (BLI) was performed with aXenogen IVIS 200 Imaging System (Xenogen/Caliper LifeSciences) as previously described (15).

Statistical analysisData were analyzed with GraphPad Prism (version 5) soft-

ware (GraphPad Software). A Mann–Whitney test or ANOVAwith Bonferroni posttest was used to analyze experimentaldata. Survival curves were generated by the actuarial Kaplan–Meier method and analyzed with the Jump-In software (SAS)with log-rank test for comparisons from the time of tumor

Mok et al.

Cancer Res; 74(1) January 1, 2014 Cancer Research154




challenge towhenmice were sacrificed due to tumors reaching14 mm in maximum diameter, or the end of the study periodhad been reached.

ResultsPLX3397 does not have direct cytotoxic effects againstSM1 and preserves splenocytesWe tested the effects of single agent PLX3397 against SM1

using an in vitro MTS cell proliferation assay after 72 hoursof exposure to rule out a potential direct antitumor effect ofthis CSF-1R inhibitor in the SM1 murine melanoma cell line.

An IC50 was not reached even at 1 mmol/L (Fig. 1A). In additionto its resistance in MTS assays, exposure of SM1 to PLX3397 atthe range of concentrations between 10 nmol/L and 1 mmol/Lshowed no inhibition of the downstream mitogen–activatedprotein kinase signaling pathway (Fig. 1B). SM1 does notexpress either CSF-1R or c-kit by fluorescence-activatedcell sorting (FACS) analysis and gene expression profiling(data not shown); these are the two main targets of PLX3397.SM1 also produces high levels of CSF-1 protein and its level isnot affected by PLX3397 at 1 mmol/L (Supplementary Fig. S1).

We then ruled out that PLX3397 had a potential detri-mental effect against T cells. Increasing concentrations ofPLX3397 did not negatively alter the viability of murinesplenocytes after 72 hours of treatment (Fig. 1C). In contrast,1 mmol/L of PLX3397 was enough to downregulate pErk leveland to kill cells that were dependent on CSF-1/CSF-1R forgrowth such as myeloid cells and microglia cells (12, 17).Therefore, because single-agent PLX3397 did not affect theviability of either SM1 cells or murine splenocytes, thissupports SM1 as a permissive model for testing the effectsof PLX3397 to improve the antitumor activity of ACTimmunotherapy.

Combined therapy with PLX3397 and ACTimmunotherapy improves antitumor responses againstSM1 tumors

Lymphodepleted C57BL/6 mice with established subcu-taneous SM1-OVA tumors received ACT of splenocytesobtained from OT-1 mice expressing an OVA-specific T-cellreceptor (TCR). We titrated the application of immunother-apy to provide a suboptimal antitumor effect that wassimilar to the antitumor effect of single agent PLX3397 sothat the effect of combination could be revealed (Fig. 2A).The combined therapy for PLX3397 and OT-1 TCR trans-genic ACT demonstrated superior antitumor effects com-pared with either therapy alone in duplicate experimentsand improved overall survival (Fig. 2B; Supplementary Fig.S2a). As the OVA model is based on the recognition of aforeign antigen, we further confirmed the results in thepmel-1 ACT model that is based on transgenic T cells witha TCR recognizing gp100, a murine melanosomal antigenendogenously expressed by SM1 (Fig. 2C; ref. 8). In replicatestudies, the combined therapy with pmel-1 ACT andPLX3397 also had superior antitumor response comparedwith either single agent therapy alone and improved survival(Fig. 2D; Supplementary Fig. S2a). To also test the generalapplicability of combining PLX3397 and pmel-1 ACT, anoth-er murine melanoma model, B16 was used. Combinedtherapy also demonstrated superior antitumor responsecompared with either therapy alone in duplicate experi-ments (Supplementary Fig. S2b).

Decrease in tumor-infiltrating macrophages byinhibition of CSF-1R by PLX3397

To analyze whether PLX3397 changed the magnitude ofTIMs, including macrophages and MDSCs, we analyzed theirpresence in tumors by immunofluorescence. There was a de-crease in the quantity of F4/80(þ) macrophages in both

0 1(10 nmol/L)

2(100 nmol/L)

3(1,000 nmol/L)

Cell v

iab

ilit

y (

%)

Log10 of the PLX3397

concentration in nmol/L

Ce

ll v

iab

ilit

y (

%)

Log10 of the PLX3397

concentration in nmol/L

10 125 1,000 125 1,000

VC PLX3397 (nmol/L)

1 h 24 h

pErk

Erk

Actin

10

VC PLX3397 (nmol/L)

0

20

40

60

80

100

0

20

40

60

80

100

1(10 nmol/L)

2(100 nmol/L)

3(1,000 nmol/L)

0

A

B

C

Figure 1. Effects of PLX3397 on SM1 melanoma and primary T cells. A,murine SM1 melanoma cells were exposed to increasing concentrationsof PLX3397 for 72 hours for IC50 determination using an MTS assay. B,immunoblotting for analysis of signaling molecules after PLX3397exposure of SM1 cells at 10, 125, and 1000 nmol/L for 1 or 24 hours. C,effects of PLX3397 on murine splenocyte viability. Cell viability assay(MTS) of ex vivo activated C57BL/6 splenocytes at 72-hour time pointwith increasing doses of PLX3397.

CSF-1R Blockade Improves Immunotherapy

www.aacrjournals.org Cancer Res; 74(1) January 1, 2014 155




the PLX3397 single agent group and the combined groupcompared with vehicle or ACT single treatment groups(Fig. 3A). Analysis of tumors, spleens, lungs, blood, and bonemarrow frommice treated with PLX3397 demonstrated that theeffects of PLX3397 were specific for intratumoral macrophagesas opposed to a systemic depletion ofmacrophages. Therewas alocal (tumor) decrease in quantity of F4/80(þ) CD11b(þ)macrophages but no significant decrease systemically (Fig.3B and D; Supplementary Fig. S3). To further characterize thephenotype of the remaining macrophages, MHCII expressionwas analyzed by FACS. There was a shift in macrophagephenotype from MHCIIlow to MHCIIhi upon treatment withPLX3397 (Fig. 3C and D; Supplementary Fig. S3).

No difference in MDSC number and ratio of PMN/MO-MDSC in combined treatment of ACT with PLX3397

The level of MDSCs in SM1 tumors was first analyzed byimmunofluorescence. The MDSCs were present at a low levelin tumors and their level seemed to be unaltered by ACT orPLX3397 treatment (Fig. 4A). To better enumerate themagnitude and distribution of MDSCs in vivo, we analyzedtheir presence in tumors, spleens, lungs, blood, and bonemarrow by flow cytometry. Confirming the immunofluores-cence data, there was no change in the already low (�4%–6%) baseline quantity of Gr-1(þ) CD11b(þ) MDSCs follow-ing treatment with PLX3397 (Fig. 4B and D; SupplementaryFig. S3). We then examined whether PLX3397 altered theratio between the two recognized subsets of MDSCs, thepolymorphonuclear MDSCs (PMN-MDSC, Gr-1hi Ly6Clow),and the monocytic MDSCs (MO-MDSC, Gr-1low Ly6Chi).There was no significant difference between PLX3397 trea-ted and nontreated groups in MDSC subsets, just as therewas no change in total MDSCs (Fig. 4C and D; Supplemen-tary Fig. S3).

Effect of CSF-1 or macrophage depletion overlappedwith the effects of PLX3397

In order to confirm the role of CSF-1 in the activity ofPLX3397, PLX3397 treatment was combined with anti-CSF-1antibody in mice that received ACT with pmel-1 splenocytes.There was no improvement in antitumor activity in thePLX3397, pmel-1 ACT, and anti-CSF-1 antibody treatmentgroups, compared with mice receiving PLX3397 and pmel-1ACT therapy with an isotype control antibody (Fig. 5A),suggesting that the effects of PLX3397 and anti-CSF-1 anti-body-mediated depletion might be overlapping. To deter-mine if the target of PLX3397 was macrophages, mice withestablished SM1 tumors were treated with PLX3397 incombination with clodronate, an agent that depletes macro-phages. There was no enhanced antitumor response in thecombined treatment group of PLX3397 and clodronate com-pared with either single treatment group (Fig. 5B). Thesestudies suggest that PLX3397 mediated its effects throughinhibition of immunosuppressive CSF-1–responding intra-tumoral macrophages.

PLX3397 increases the expansion, distribution, andfunctional activation of intratumoral lymphocytes

To analyze whole animal T-cell distribution in the pres-ence or absence of PLX3397, we genetically labeled theadoptively transferred OT-1 T cells with firefly luciferasetransgene for in vivo BLI. There was an increased expansion,in vivo distribution to tumor, and tumor targeting by adop-tively transferred OT-1 T cells when mice were treated withPLX3397 (Fig. 6A). The quantitative analysis of T-cell-asso-ciated luciferase activity in mice treated with OT-1 ACTcombined with PLX3397 showed increased accumulationwithin OVA antigen-matched tumors over time (Fig. 6B;Supplementary Fig. S4a). Furthermore, we repeated BLI

(tumor ~3 mm

day 3)

0 5 10 15 20

Tu

mo

r v

olu

me

(m

m3)

Tu

mo

r v

olu

me

(m

m3)

Day

VehiclePLX3397OT-1 ACTPLX3397+OT-1 ACT

P = 0.00002

P = 0.1

P = 0.000004

C

0 6 12 18 Day

VehiclePLX3397pmel-1 ACTPLX3397+pmel-1 ACT

P = 0.002

P = 0.03

P = 0.00000004

500cGy IL-2PLX3397

p.o. DailySM1 Tumor

pmel-1 Mouse

T Lymphocytes

(tumor ~5 mm day 7)

(tumor ~3 mm

day 3)

500cGy IL-2PLX3397

p.o. Daily

T Lymphocytes(tumor ~5 mm day 7)

SM1-OVA

Tumor

OT-1 Mouse

0

400

800

1,200

1,600

0

400

800

1,200

D

A BFigure 2. Combined antitumoractivity of ACT immunotherapy andPLX3397 in the OVA and pmel-1models. A, schematic of the OT-1ACT model based on adoptivelytransferring OT-1 splenocytes intolymphodepleted mice withpreviously established SM1 tumorsstably expressing OVA antigen(SM1-OVA). B, tumor growthcurves of established SM1-OVAtumors in C57BL/6 mice throughday 20 posttumor implantation. C,schematic of pmel-1 ACT model,with lymphodepleted miceharboring SM1 tumors thatadoptively received pmel-1splenocytes and PLX3397. D,tumor growth curves of establishedSM1 tumors in C57BL/6 micethrough day 18. p.o., orally.

Mok et al.





using pmel-1 T cells also labeled with the firefly luciferasetransgene and used for ACT. Again, PLX3397 increased theexpansion of pmel-1 T cells distributed to gp100 positiveantigen-matched SM1 tumors (Fig. 6C). Mice that receivedpmel-1 ACT combined with PLX3397 also demonstrated anincreased intratumoral accumulation of luciferase activity totumors over time (Fig. 6D; Supplementary Fig. S4b). We thenanalyzed the activation state of TILs by detecting cytokineproduction. In two replicate experiments, TILs collectedfrom mice treated with the combination of pmel-1 ACT andPLX3397 showed a higher ability to secrete IFN-g and torespond to short term ex vivo restimulation with the gp100antigen (Fig. 6E). Therefore, treatment with PLX3397increased both the number and functionality of adoptivelytransferred antitumor antigen-specific T cells.

The antitumor activity of PLX3397 is T-cell-dependentSingle agent PLX3397 treatment demonstrated a weak but

reproducible antitumor response compared with vehicle con-trol (Figs. 2B, C, and 5B). We tested if this antitumor activitywas mediated by endogenous cytotoxic CD8 T cells. SM1tumors were implanted in C57BL/6 mice without irradiation

and received PLX3397 in combination with anti-CD8 antibody.The depletion of CD8þ cells abrogated the antitumor activityof single agent PLX3397 (Fig. 7A). To further test the role ofimmune cells in the antitumor activity of PLX3397, immuno-deficient NSG mice were implanted with SM1 tumors anddosed with PLX3397. In these immunodeficientmice there wasno antitumor activity of PLX3397 compared with mice receiv-ing vehicle control (Fig. 7B). Finally, we tested whether theantitumor effects of PLX3397 was also dependent on cytotoxicCD8 T cells in the pmel-1 ACT model. C57BL/6 mice withestablished SM1 tumors were treated with a combination ofPLX3397, anti-CD8 antibody, and pmel-1 ACT. There was noantitumor activity of pmel-1 ACT combined with PLX3397 inmice that received CD8-depleting antibody (Fig. 7C). Collec-tively, these studies highlight the role of CD8þ T cells aseffectors of the antitumor activity of PLX3397 in the SM1murine melanoma model.

DiscussionTumor immunotherapy with cytokines such as IFN or

IL-2, immune modulating antibodies blocking the cytotoxic

0

100

200

300

400

Tum

or

Splee

nLu

ng

Blood

Bone

mar

row

Me

an

flu

ore

sc

en

ce

inte

ns

ity

of

MH

CII

Vehicle

PLX3397

OT-1 ACT

PLX3397+OT-1 ACT

A

0

5

10

15

Tum

or

Splee

nLu

ng

Blood

Bone

mar

row

Perc

en

tag

eo

fF

4/8

0(+

)an

d

CD

11b

(+)

macro

ph

ag

e Vehicle

PLX3397

OT-1 ACT

PLX3397+OT-1 ACT

D

Vehicle

PLX3397

OT-1 ACT

PLX3397+

OT-1 ACTF

4/8

0

13.9%

0.729%

13.1%

CD11b MHCII

1.63%

38.3

MΦM1/M2–MΦ

(mean fluorescence)

147

58.9

180

B

C

OT-1 ACT PLX3397 + OT-1 ACT

PLX3397Vehicle

Figure 3. Changes in intratumoralmacrophages in responses toPLX3397. C57BL/6 mice withSM1-OVA tumors were gavagedorally daily with PLX3397 andreceived OT-1 ACT for 18 days toassess prolonged effects of thedrug on macrophages. A, tissueimmunofluorescence microscopyto detect macrophages.Representative hematoxylinand eosin (left) andimmunofluorescence formacrophages stained with anti-F4/80-FITC (green, right), andnuclei stained with DAPI (blue,right). B, cells stained for thesurface expression markers ofmacrophages (F4/80þ CD11bþ),M1-macrophage (MHCIIhi), andM2-macrophage (MHCIIlow) wereused for FACS analysis. Bar graphrepresentation of percentage ofF4/80(þ) and CD11b(þ)macrophages. C, bar graphrepresentation of meanfluorescence intensity of MHCIIexpression on macrophages. D,representative FACS plotsdemonstrating percentage ofF4/80(þ) CD11b(þ) macrophagesand mean fluorescence intensityof M1-marophage (MHCIIhi) andM2-macrophage (MHCIIlow) intumor tissue.






T-leukocyte-associated antigen-4 (CTLA4) or the program-med cell death receptor 1 (PD-1) or its ligand PD-L1, andthe adoptive transfer of tumor antigen-specific T cells, resultin tumor responses in patients with advanced cancers, andmelanoma in particular. These immunotherapy-inducedtumor responses are frequently extremely durable and canlast years. However, their low response rate has been one ofthe obstacles remaining to overcome (3, 18). A potentialexplanation is that intratumoral immunosuppressive mye-loid cells may inhibit immunotherapy responses via differentmechanisms (1). These myeloid lineage cells represent var-ious distinct and heterogeneous populations of immuno-suppressive cells, including MDSCs and macrophages. Bysuppressing these myeloid cells with CSF-1R blockade usingPLX3397, we hypothesized that T-cell function withintumors may improve. Hence, the ideal treatment would beto suppress the myeloid cells and increase the number ofeffector T cells at the same time. SM1 tumors grow veryrapidly when implanted in C57BL/6 mice, such that miceoften need to be sacrificed within 2 to 3 weeks of tumor

implantation, limiting the ability to induce an immuneresponse with immune modulating antibodies (8, 10). Toovercome this, and provide the opportunity for a fullyeffective immune response in combination with CSF-1Rblockade, we have used the ACT of T cells expressing TCRsrecognizing a specific tumor antigen on the tumor cells. Thescientific rationale posits that inhibition of CSF-1R signalingin immunosuppressive TIMs will improve the intratumoralmilieu by taking away immunosuppressive factors that limitthe antitumor activity of cytotoxic T lymphocytes.

We explored the potential mechanisms by which PLX3397improves the antitumor effect of ACT. The myeloid cellsinfiltrating SM1 tumors are mainly macrophages. Our datademonstrate that PLX3397 decreases the quantity of macro-phages locally in tumor. Although the classification betweenM1andM2polarizedmacrophages is still controversial,MHCIIis used as a marker to distinguish them (19). With PLX3397,there is a skewing of the population of MHCIIlow to MHCIIhi

macrophages, consistent with a previous report (13). Bydecreasing the presence of immunosuppressive MHCIIlow

A

0

20

40

60

80

Tum

or

Splee

n

Lung

Blood

Bone

mar

row

Pe

rce

nta

ge

of

Gr-

1(+

) a

nd

CD

11

b(+

) M

DS

C Vehicle

PLX3397

OT-1 ACT

PLX3397+OT-1 ACT

0%

20%

40%

60%

80%

100%

Vehic

le

PLX

3397

OT

-1 A

CT

PLX

3397 +

OT

-1 A

CT

Vehic

le

PLX

3397

OT

-1 A

CT

PLX

3397 +

OT

-1 A

CT

Vehic

le

PLX

3397

OT

-1 A

CT

PLX

3397 +

OT

-1 A

CT

Vehic

le

PLX

3397

OT

-1 A

CT

PLX

3397 +

OT

-1 A

CT

Vehic

le

PLX

3397

OT

-1 A

CT

PLX

3397 +

OT

-1 A

CT

Tumor Spleen Lung Blood Bone marrow

Rela

tive p

erc

en

tag

e o

f M

O/P

MN

-MD

SC

PMN

MO

B

C

D

Vehicle

PLX3397

OT-1 ACT

PLX3397+

OT-1 ACT

Gr-

1

Ly6

G

6.26%

6.50%

5.34%

5.69%

78.3%

18.5%

65.7%

31.5%

55.9%

40.2%

59.5%

37.1%

MDSC

CD11b

PMN/MO-MDSC

Ly6C

PLX3397Vehicle


Figure 4. Analysis of MDSCs withPLX3397 exposure. C57BL/6 micewith SM1-OVA tumors weregavaged orally daily with PLX3397and received OT-1 ACT for 18 daysto assess prolonged effects of thedrug on MDSCs. A, tissueimmunofluorescence microscopyto detect MDSCs on day 14 post-ACT. Representative hematoxylinand eosin (left) andimmunofluorescence for MDSCsstained with anti-Gr-1-FITC (green,right), and nuclei stained with DAPI(blue, right). B, cells stained for thesurface expression markers ofMDSCs (Gr-1þ CD11bþ), MO-MDSC (Gr-1low Ly6Chi), and PMN-MDSC (Gr-1hi Ly6Clow) were usedfor FACS analysis. Bar graphrepresentation of percentage ofGr-1(þ) CD11b(þ) MDSCs. C, bargraph representation of ratiobetween MO-MDSC and PMN-MDSC. D, representative FACSplots demonstrating percentagesof Gr-1(þ) CD11b(þ) MDSC, MO-MDSC (Gr-1low Ly6Chi), and PMN-MDSC (Gr-1hi Ly6Clow) in tumortissue.

Mok et al.





macrophages, PLX3397 likely facilitates the intratumoral traf-ficking of adoptively transferred lymphocytes and their anti-tumor functions. Our studies also suggest that macrophageinhibition alone by PLX3397 is not sufficient for an antitumorresponse. An immune response is indeed needed for theantitumor effect of PLX3397 in the SM1 model because thiseffect requires CD8 T cells. Therefore, the main beneficialeffects of PLX3397 in this model are derived from the abilityto improve T-cell effector functions indirectly through theinhibition of intratumoral immunosuppressive macrophages.

Reports have shown that different tumor models such asprostate and lung carcinoma models are MDSC-dominantand these MDSCs are heavily infiltrated in tumors andsystemic organs such as spleen, blood, and bone marrow(14, 20). By contrast, our SM1 model has few MDSCs infil-trating the tumor or accumulating in the spleen over time.This is consistent with a recent report that shows patientswith melanoma have far less MDSCs than other cancers andalso compared with many nonmelanoma murine tumormodels (21). In addition, prior reports have shown thatCSF-1R blockade therapy reduced the number of myeloidcells including MDSCs and macrophages in both tumors andsystemic organs (4, 14). In contrast, in our model, CSF-1Rblockade with PLX3397 mainly targeted the more abundantmacrophages instead of MDSCs, suggesting the role of CSF-1may be tumor model-dependent. Different tumor models,genetic backgrounds, or treatments may induce differentgrowth factors or cytokines in the tumor microenvironment.As myeloid cell tumor infiltration is a complex processregulated by different pathways, it is possible that differenttumor models and tumors from different patients producingdifferent cytokines like CSF-1, IL-34, or CCL2 result in

Vehicle PLX3397


0

4,000

8,000

12,000

0 2 4 6 8 10 12 14

Co

un

t/p

ixel

Number of days after ACT

Vehicle PLX3397 OT-1 ACT PLX3397 + OT-1 ACT

0

1,000

2,000

3,000

0 2 4 6 8 10

Co

un

t/p

ixe

l

Number of days after ACT

Vehicle

PLX3397

pmel-1 ACT

PLX3397 + pmel-1 ACT

Vehicle PLX3397

pmel-1 ACT PLX3397 + pmel-1 ACT

A B

C D

IFNγ

CD3

52.6%

56.7% 81.5%

70.4%

pmel-1 ACT

PLX3397 +

pmel-1 ACT

Thy1.1

E

Figure 6. Effects of PLX3397 on the distribution and cytokine-producing functions of adoptively transferred lymphocytes. A, in vivo BLI of TCR transgenic T-celldistribution.OT-1 transgenic T cellswere transducedwith a retrovirus-firefly luciferase andused for ACT. Representative figure at day 5depicting four replicatemice pergroup. B, quantitation of BLI of serial images with region of interest analysis at the site of tumors obtained through day 13 post-ACT of OT-1 transgenic Tcells expressing firefly luciferase, with four mice per group. C, pmel-1 transgenic T cells transduced with retrovirus-firefly luciferase were used for ACT. Day 5representative figure showing four replicatemice per group. D, quantitativemeasure of BLI signal with region of interest at the tumor site obtained through day 10 post-ACTof luciferase expressing pmel-1 T-cells. C, effects on cytokine production upon antigen restimulation. SM1 tumor-bearingC57BL/6mice receivedpmel-1ACTwithor without PLX3397. At day 5 post-ACT, TILs were isolated for intracellular IFN-g staining analyzed by FACS after 5-hour ex vivo exposure to the gp10025-33 peptide.

A

B

0

400

800

1,200

0 6 12 18 24

Tu

mo

r vo

lum

e (

mm

3)

Tu

mo

r vo

lum

e (

mm

3)

Day

Vehicle+pmel-1

PLX3397+pmel-1

Anti-CSF-1+pmel-1

PLX3397+Anti-CSF-1+pmel-1

0 5 10 15

Day

Vehicle

PLX3397

Clodronate

PLX3397 + Clodronate P = 0.00001

P = 0.06

0

500

1,000

1,500

Figure 5. Effects of PLX3397 mediated by CSF-1R and macrophages. A,tumor growth curves of established SM1 tumors in C57BL/6 mice thatreceived pmel-1 ACT with PLX3397 and anti-CSF-1 antibody. Treatment ofanti-CSF-1 or isotype antibody control was started at the same time withPLX3397 when the tumor diameter reached 3 mm. On day 15, betweenvehicleþpmel-1 and anti-CSF-1þpmel-1, P ¼ 0.000008; betweenanti-CSF-1þpmel-1 and PLX3397þpmel, P ¼ 0.00003; betweenPLX3397þpmel-1 and PLX3397þanti-CSF-1þpmel-1, P ¼ 0.1. B, tumorgrowthcurvesof establishedSM1 tumors inC57BL/6mice treatedPLX3397in combination with clodronate. Clodronate treatment was started at thesame time with PLX3397 when the tumor diameter reached 3 mm.






attraction of different myeloid lineages leading to differen-tial responses to CSF-1R blockade (22–25). For example, ithas been reported that under hypoxic conditions, induced

expression of hypoxia-inducible factor-1a results in theexpression of CXCR4 and SDF-1. The SDF-1/CXCR4 axis isanother pathway that mediates the recruitment of TIMs totumors (26, 27). Future immunotherapy trials may thusbenefit from molecular profiling-based stratification ofpatients in regards to their tumor microenvironmentcytokines.

In conclusion, combined therapy with the CSF-1R inhibitorPLX3397 and TCR-based ACT immunotherapy results insuperior antitumor effects than single agent treatment in amurine model of melanoma. The antitumor activity is medi-ated by inhibition of the myeloid cell-mediated immunosup-pressive tumor microenvironment so that more lymphocytesinfiltrate into the tumor with enhanced functionality.

Disclosure of Potential Conflicts of InterestG. Bollag is the CEO of Plexxicon. B. L. West is an employee of Plexxikon.

No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: S. Mok, R.C. Koya, J. Xu, L. Wu, A. RibasDevelopment of methodology: S. Mok, R.C. Koya, L. Robert, A. RibasAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S. Mok, R.C. Koya, C. Tsui, J. Xu, L. Robert, B.L. West,G. Bollag, A. RibasAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Mok, R.C. Koya, J. Xu, T.G. Graeber, A. RibasWriting, review, and/or revision of the manuscript: S. Mok, R.C. Koya,L. Wu, B.L. West, G. Bollag, A. RibasAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): S. Mok, C. TsuiStudy supervision: R.C. Koya, A. Ribas

Grant SupportThis work was funded by the NIH (grants P01 CA168585, P50 CA086306, and

R21 CA169993), the Seaver Institute, the Louise Belley and Richard Schnarr Fund,the Wesley Coyle Memorial Fund, the Garcia-Corsini Family Fund, the Bila AlonHacker Memorial Fund, the Fred L. Hartley Family Foundation, the Ruby FamilyFoundation, the Jonsson Cancer Center Foundation, the Eli & Edythe BroadCenter of Regenerative Medicine and Stem Cell Research at UCLA, and theCaltech–UCLA Joint Center for Translational Medicine (A. Ribas andT.G. Graebar).

T.G. Graebar is supported by an American Cancer Society Research ScholarAward (RSG-12-257-01-TBE), the Caltech–UCLA Joint Center for TranslationalMedicine, the National Center for Advancing Translational Sciences UCLA CTSIGrant UL1TR000124, and a Concern Foundation's CONquer CanCER NowAward.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received June 26, 2013; revised October 10, 2013; accepted October 23, 2013;published OnlineFirst November 18, 2013.

References1. Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as reg-

ulators of the immune system. Nat Rev Immunol 2009;9:162–74.2. Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating

immunity's roles in cancer suppression and promotion. Science2011;331:1565–70.

3. Kerkar SP, Restifo NP. Cellular constituents of immune escape withinthe tumor microenvironment. Cancer Res 2012;72:3125–30.

4. Priceman SJ, Sung JL, Shaposhnik Z, Burton JB, Torres-Collado AX,MoughonDL, et al. Targeting distinct tumor-infiltratingmyeloid cells byinhibiting CSF-1 receptor: combating tumor evasion of antiangiogenictherapy. Blood 2010;115:1461–71.

5. Tarhini AA, Butterfield LH, Shuai Y, Gooding WE, Kalinski P, Kirk-wood JM. Differing patterns of circulating regulatory T cells andmyeloid-derived suppressor cells in metastatic melanoma patientsreceiving anti-CTLA4 antibody and interferon-alpha or TLR-9 ago-nist and GM-CSF with peptide vaccination. J Immunother 2012;35:702–10.

6. Dai XM, RyanGR,Hapel AJ, DominguezMG,Russell RG, KappS, et al.Targeted disruption of the mouse colony-stimulating factor 1 receptorgene results in osteopetrosis, mononuclear phagocyte deficiency,increased primitive progenitor cell frequencies, and reproductivedefects. Blood 2002;99:111–20.

B

A

0

200

400

600

800

0 7 14

Tu

mo

r vo

lum

e (

mm

3)

Tu

mo

r vo

lum

e (

mm

3)

Tu

mo

r vo

lum

e (

mm

3)

Day

Vehicle

PLX3397P = 0.1

0 7 14

Day

Vehicle+pmel-1

PLX3397+pmel-1

Anti-CD8+pmel-1

PLX3397+Anti-

CD8+pmel-1

P = 0.1

P = 0.01

P = 0.01

C

0

200

400

600

800

1,000

0 5 10 15

Day

VehiclePLX3397Anti-CD8PLX3397 + Anti-CD8

P = 0.06

P = 0.00007

0

100

200

300

Figure 7. Lack of antitumor activity of PLX3397 in immunodeficient miceor with CD8þ T-cell depletion. A, tumor growth curves of establishedSM1 tumors in nonirradiated C57BL/6 mice treated with PLX3397 incombination with anti-CD8 antibody. B, tumor growth curves ofestablishedSM1 tumors inNSGmice treatedwith orwithout PLX3397.C,SM1 tumor-bearing mice received pmel-1 ACT and were treated withPLX3397 and anti-CD8 depleting antibody.

Mok et al.





7. Li J, Chen K, Zhu L, Pollard JW. Conditional deletion of the colonystimulating factor-1 receptor (c-fms proto-oncogene) inmice. Genesis2006;44:328–35.

8. Koya RC, Mok S, Otte N, Blacketor KJ, Comin-Anduix B, Tumeh PC,et al. BRAF inhibitor vemurafenib improves the antitumor activity ofadoptive cell immunotherapy. Cancer Res 2012;72:3928–37.

9. Gray-Schopfer V, Wellbrock C, Marais R. Melanoma biology and newtargeted therapy. Nature 2007;445:851–7.

10. Knight DA, Ngiow SF, Li M, Parmenter T, Mok S, Cass A, et al. Hostimmunity contributes to the anti-melanoma activity of BRAF inhibitors.J Clin Invest 2013;123:1371–81.

11. Chitu V, Nacu V, Charles JF, Henne WM, McMahon HT, Nandi S, et al.PSTPIP2 deficiency in mice causes osteopenia and increased differ-entiation of multipotent myeloid precursors into osteoclasts. Blood2012;120:3126–35.

12. ConiglioSJ, Eugenin E, Dobrenis K, Stanley ER,WestBL, SymonsMH,et al.Microglial stimulationof glioblastoma invasion involves epidermalgrowth factor receptor (EGFR) and colony stimulating factor 1 receptor(CSF-1R) signaling. Mol Med 2012;18:519–27.

13. DeNardo DG, Brennan DJ, Rexhepaj E, Ruffell B, Shiao SL, MaddenSF, et al. Leukocyte complexity predicts breast cancer survival andfunctionally regulates response to chemotherapy. Cancer Discov2011;1:54–67.

14. Xu J, Escamilla J, Mok S, David J, Priceman SJ, West BL, et al.Abrogating the protumorigenic influences of tumor-infiltrating myeloidcells by CSF1R signaling blockade improves the efficacy of radiother-apy in prostate cancer. Cancer Res 2013;73:2782–94.

15. KoyaRC,MokS, Comin-Anduix B, Chodon T, RaduCG, NishimuraMI,et al. Kinetic phases of distribution and tumor targeting by T cellreceptor engineered lymphocytes inducing robust antitumorresponses. Proc Natl Acad Sci U S A 2010;107:14286–91.

16. Vo DD, Prins RM, Begley JL, Donahue TR, Morris LF, Bruhn KW, et al.Enhanced antitumor activity induced by adoptive T-cell transfer andadjunctive use of the histone deacetylase inhibitor LAQ824. CancerRes 2009;69:8693–9.

17. HeY, RhodesSD, Chen S,WuX, Yuan J, Yang X, et al. c-Fms signalingmediates neurofibromatosis Type-1 osteoclast gain-in-functions.PLoS One 2012;7:e46900.

18. McArthur GA, Ribas A. Targeting oncogenic drivers and the immunesystem in melanoma. J Clin Oncol 2013;31:499–506.

19. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regula-tion of myeloid cells by tumours. Nat Rev Immunol 2012;12:253–68.

20. Srivastava MK, Zhu L, Harris-White M, Kar U, Huang M, Johnson MF,et al. Myeloid suppressor cell depletion augments antitumor activity inlung cancer. PLoS One 2012;7:e40677.

21. Gros A, Turcotte S, Wunderlich JR, Ahmadzadeh M, Dudley ME,Rosenberg SA. Myeloid cells obtained from the blood but not fromthe tumor can suppress T-cell proliferation in patients with melanoma.Clin Cancer Res 2012;18:5212–23.

22. Lin H, Lee E, Hestir K, Leo C, Huang M, Bosch E, et al. Discovery of acytokine and its receptor by functional screening of the extracellularproteome. Science 2008;320:807–11.

23. Wei S, Nandi S, Chitu V, Yeung YG, Yu W, Huang M, et al. Functionaloverlap but differential expression of CSF-1 and IL-34 in their CSF-1receptor-mediated regulation of myeloid cells. J Leukoc Biol 2010;88:495–505.

24. Sawanobori Y, Ueha S, Kurachi M, Shimaoka T, Talmadge JE, Abe J,et al. Chemokine-mediated rapid turnover of myeloid-derived sup-pressor cells in tumor-bearing mice. Blood 2008;111:5457–66.

25. Li X, Loberg R, Liao J, Ying C, Snyder LA, Pienta KJ, et al. A destructivecascademediated byCCL2 facilitates prostate cancer growth in bone.Cancer Res 2009;69:1685–92.

26. Kozin SV, Kamoun WS, Huang Y, Dawson MR, Jain RK, Duda DG.Recruitment of myeloid but not endothelial precursor cells facil-itates tumor regrowth after local irradiation. Cancer Res 2010;70:5679–85.

27. Kioi M, Vogel H, Schultz G, Hoffman RM, Harsh GR, Brown JM.Inhibition of vasculogenesis, but not angiogenesis, prevents the recur-rence of glioblastoma after irradiation in mice. J Clin Invest 2010;120:694–705.






2014;74:153-161. Published OnlineFirst November 18, 2013.Cancer Res Stephen Mok, Richard C. Koya, Christopher Tsui, et al. Adoptive Cell Transfer ImmunotherapyInhibition of CSF-1 Receptor Improves the Antitumor Efficacy of

Updated version

10.1158/0008-5472.CAN-13-1816doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/11/18/0008-5472.CAN-13-1816.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/1/153.full#ref-list-1

This article cites 27 articles, 14 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/1/153.full#related-urls

This article has been cited by 33 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/74/1/153To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-1816

http://cancerres.aacrjournals.org/content/suppl/2013/11/18/0008-5472.CAN-13-1816.DC1

http://cancerres.aacrjournals.org/content/74/1/153.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/1/153.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/1/153


Documents

InhibitionofCSF-1ReceptorImprovestheAntitumorEf cacy of ... · Thomas G. Graeber1,2,4,7, Brian L. West 9, Gideon Bollag , and Antoni Ribas1,2,3,4,8 Abstract Colony stimulating factor