cancer terapies

8/8/2019 Cancer Terapies

http://slidepdf.com/reader/full/cancer-terapies 1/12

The Requirement of MultimodalTherapy (Vaccine, Local Tumor

Radiation, and Reduction of Suppressor Cells) to

Eliminate Established Tumors

Chie Kudo-Saito,

1

Jeffrey Schlom,

1

Kevin Camphausen,

2

C. Norman Coleman,

2

and James W. Hodge

1

Abstract Purpose: Numerous immune-based strategies are currently being evaluatedfor cancer therapy in

preclinicalmodels and clinical trials.Whereas many strategies look promisingin preclinical models,

they are often evaluated before or shortly following tumor implantation. The elimination of well-

established tumors often proves elusive. Here we show that a multimodal immune-based therapy

can be successfully employed to eliminate established tumors.

Experimental Design: This therapy consists of vaccines directed against a self-tumor-

associated antigen, the use of external beam radiation of tumors to up-regulate Fas on tumor cells,

and the use of a monoclonal antibody (mAb) to reduce levels of CD4+

CD25+

suppressor cells.

Results: We show here for the first time that (a) antigen-specific immune responses induced by

vaccines were optimally augmented when anti-CD25 mAb was given at the same time as vacci-

nation; (b) anti-CD25 mAb administration in combination with vaccines equally augmentedT-cellimmune responses specific for a self-antigen as well as those specific for a non ^ self antigen; (c)

whereas the combined use of vaccines and anti-CD25 mAb enhanced antigen-specific immune

responses, it was not sufficient to eliminate established tumors; (d) the addition of external beam

radiation of tumors to the vaccine/anti-CD25 mAb regimen was required for the elimination of

established tumors; and (e) Tcells from mice receiving the combination therapy showed signifi-

cantlyhigher T-cell responses specificnot only for the antigenin thevaccine but alsofor additional

tumor-derived antigens (p53 and gp70).

Conclusions: These studies reported here support the rationale for clinical trials employing

multimodal immune-based therapies.

Numerous cancer vaccines are being investigated targeting

various tumor-associated self-antigens. CD4+CD25+ immuno-

suppressive/immunoregulatory T (Treg) cells have beenimplicated in the maintenance of immunologic tolerance toself-antigens. Clinical studies have shown an increase in levels

of CD4+CD25+ cells in cancer patients (1–9), suggesting that the increment of these cells is correlative to the stage of cancer

progression. Therefore, it has been hypothesized that Treg celldepletion may improve antitumor efficacy with the use of

cancer vaccines. It has been shown in murine tumor modelsthat either (a) depletion of CD4+CD25+ cells by anti-CD25

monoclonal antibody (mAb) administration (10–12), or (b)adoptive cell transfer after CD4+CD25+ cell depletion (13, 14),

could result in significant antitumor activity. In addition,

depletion of Treg cells via anti-CD25 mAb administration hasbeen combined with whole tumor cell vaccine therapy (15)

and peptide vaccine therapy (16). In these studies, however,

anti-CD25 mAb was given several days before or only 1 day

after tumor implantation. No reports exist as to whether anti-CD25 mAb administration could increase antitumor efficacy induced by therapy of more established tumors.

We previously developed recombinant poxviral vectors that contain the transgenes for a carcinoembryonic antigen (CEA),

a triad of T-cell costimulatory molecules (B7-1, intracellular adhesion molecule [ICAM-1], LFA-3; designated TRICOM),

and the combination of the four transgenes (CEA/TRICOM). Two types of poxvirus vectors were developed: replication-

competent recombinant vaccinia (rV) and replication-defectiverecombinant fowlpox (rF; refs. 17–19). In the study reported

here, we first examine whether anti-CD25 mAb administration

can enhance antigen-specific T-cell responses and antitumor immunity induced by CEA/TRICOM vaccines. We sought to

determine the optimal timing for anti-CD25 mAb administra-

tion in relation to vaccine administration to enhance antigen-specific T-cell immune responses. Based on these data, we next

conducted in vivo therapy of well-established CEA-positivetumors using CEA-transgenic mice, which express CEA as aself-antigen; therapy consisted of vaccines, or anti-CD25 mAb,

and the combination of both with and without external beam

radiation of tumors. We show here for the first time that (a) antigen-specific

immune responses induced by vaccines were optimally aug-mented when anti-CD25 mAb was given at the same time as

www.aacrjournals.org Clin Cancer Res 2005;11(12) June 15, 20054533

Cancer Therapy: Preclinical

Authors’ Affiliation: 1Laboratory of Tumor Immunology and Biology, and2Radiation Oncology Branch, Center for Cancer Research, National Cancer

Institute, NIH, Bethesda, Maryland

Received11/2/04; revised 3/17/05; accepted 3/18/05.

The costs of publicationof this article were defrayed inpart by the paymentof page

charges. This article must therefore be hereby marked advertisement in accordance

with18 U.S.C. Section1734 solely to indicatethis fact.

Requests for reprints: Jeffrey Schlom, Laboratory o f Tumor Immunology and

Biology, National Cancer Institute, NIH, 10 Center Drive, Room 8B09, Bethesda,

MD 20892. Phone: 301-496-4343; Fax: 301-496-2756; E-mail: js141c@ nih.gov.



vaccination; (b) anti-CD25 mAb administration in combination with vaccines equally augmented T-cell immune responses

specific for a self-antigen as well as those specific for a non–

self antigen; (c) whereas the combined use of vaccines and anti-CD25 mAb enhanced antigen-specific immune responses, it wasnot sufficient to eliminate established tumors; (d) the addition

of external beam radiation of tumor to the vaccine/anti-CD25mAb regimen was required for the elimination of established

tumors; (e) mice receiving the combination therapy and cured of tumors showed significantly higher T-cell responses specific not only for CEA but also an antigen cascade for additional tumor-derived antigens (p53 and gp70).

Materials and Methods

Animals and tumor cells. Female C57BL/6 mice were obtained fromthe National Cancer Institute, Frederick Cancer Research Facility

(Frederick, MD). Female C57BL/6 mice transgenic for human CEA

were obtained from a breeding pair provided by Dr. John Thompson

(Institute of Immunobiology, University of Freiburg, Germany). The

generation and characterization of the CEA-transgenic mouse has been

previously described (20, 21). Mice were housed and maintained under pathogen-free conditions in microisolator cages until used for experi-

ments at 6 to 8 weeks of age.

The following murine tumor cell lines (H-2b) were used: lymphoma

EL-4 cells, parental colon adenocarcinoma MC38 cells, MC38 express-

ing human CEA (designated MC38-CEA +; ref. 22), and melanoma B16cells (CEA À gp70+). These cells except EL-4 cells were trypsinized, and

all cells were washed in PBS before use.Recombinant poxviruses. The recombinant vaccinia virus designated

rV-CEA/TRICOM contains the murine B7-1, ICAM-1, and LFA-3 genesin combination with the human gene CEA as described elsewhere (23).

The recombinant fowlpox virus encoding the same genes was

designated rF-CEA/TRICOM (23). The recombinant vaccinia virus

designated rV-LacZ/TRICOM was constructed in a similar manner and

contains these three costimulatory molecule genes and LacZ gene

encoding h-galactosidase (called h-gal). The recombinant fowlpox virusencoding the same genes was designated rF-LacZ/TRICOM. Therecombinant fowlpox virus containing murine GM-CSF gene (desig-

nated rF-GM-CSF) was also used for vaccination (24). Therion Biologics

Corp. (Cambridge, MA) kindly provided all orthopox viruses. Antibodies and flow cytometric analysis. Anti-murine CD25 mAb

(PC61) and the isotype rat IgG1 (R187) were purified from culture

supernatant of each hybridoma cell line (American Type Culture

Collection, Manassas, VA) using a protein G column for in vivo

administration.

To analyze the percentage of CD4+CD25+ cells in mice, inguinal

lymph nodes, spleens, and peripheral blood cells were prepared intosingle cell suspensions and RBC were removed for flow cytometric

analysis. The following antibodies were purchased from BD Phar-

Mingen (San Diego, CA) and used for analysis: PE-conjugated anti-

CD25 (3C7, rat IgG2b), CyChrome-conjugated anti-CD4 (rat IgG2a),

FITC-conjugated anti-CD3e (hamster IgG1), PE-conjugated anti-CTLA-4(hamster IgG1), anti-rat IgG2a to detect anti-GITR mAb-binding cells,

each appropriate isotype control, and anti-CD16/CD32 mAb to block

Fc receptors. Anti-GITR mAb was purchased from R&D Systems

(Minneapolis, MN). To evaluate the generation of gp70-specific CTLsin mice, cells were stained with FITC-conjugated anti-CD3e, CyChrome-

conjugated anti-CD8 mAb (rat IgG2a) and PE-conjugated p15E604-611 /

H-2K b-tetramer (called gp70-tetramer) obtained from the NIH Tetramer

Facility (25). Cell samples from blood and spleens of naive mice were

tested as a control for this tetramer assay, and 2% to 3% of these cells

were positive in the CD8+gp70-tetramer + fraction after gating CD3+.

Immunofluorescence staining was done after Fc receptor-blocking with anti-CD16/CD32 mAb (30 minutes on ice). Cells were incubated

with antibodies or gp70-tetramer for 30 minutes on ice and washed

with 1% bovine serum albumin/PBS thrice. The immunofluorescence

was compared with the appropriate isotype-matched controls andanalyzed with Cellquest software using a FACSCalibur cytometer

(Becton Dickinson, Mountain View, CA).Treatments with TRICOM vaccines and anti-CD25 mAb. In

protocols for assays to determine the timing for anti-CD25combination with rV-CEA/TRICOM, C57BL/6 mice were vaccinated

s.c. with rV-CEA/TRICOM [1 Â 108 plaque-forming units (pfu)/

mouse] admixed with rF-GM-CSF (107 pfu/mouse) on day 0. Thesemice were injected i.p. with anti-CD25 mAb (300 Ag/mouse) on day À8, À4, À2, 0, 1, 2, 4, or day 8. In protocols for assays to examine

immune responses to a self-antigen and a foreign antigen at the sametime, CEA-transgenic mice were vaccinated s.c. with the mixture of rV-

CEA/TRICOM (5 Â 107 pfu/mouse) and rV-LacZ/TRICOM (5 Â 107

pfu/mouse) admixed with rF-GM-CSF (1 Â 107 pfu/mouse),

immediately after i.p. injection with anti-CD25 mAb (300 Ag/mouse).

These mice were sacrificed 14 days after vaccination, and splenic T cells were used for in vitro assays.

Lymphocyte proliferation assay. To evaluate CD4+ T-cell responses

specific for antigens, splenic T cells were tested for proliferation in

response to protein or peptide antigens as previously described (26).

Briefly, pooled splenic T cells (1.5 Â 105 cells per well) were cultured

in 96-well flat-bottomed plates with irradiated naive syngeneic

splenocytes as antigen-presenting cells (5 Â 105 cells per well) and

stimulants: human CEA protein (1.56-50 Ag/mL, Aspen Bio, Littleton,

CO), h-gal protein (1.56-50 Ag/mL, Prozyme, San Leandro, CA), or

p53108-122 class II peptide (0.16-5 Ag/mL, LGFLQSGTAKSVMCT;

ref. 27). As a positive control, cells were stimulated with a T-cell

mitogen Concanavalin A (2.5 Ag/mL, Sigma-Aldrich, St. Louis, MO). T

cells and antigen-presenting cells were cultured with medium only as

a negative (background) control. Cells were cultured for 5 days. 3H-

thymidine (1 ACi/well) was added to the wells for the last 18 to

24 hours and harvested using a Tomtec cell harvester (Wallac, Inc.,

Gaithersburg, MD). The incorporated radioactivity was measured using

a liquid scintillation counter (Wallac 1205 Betaplate, Wallac). The

mean proliferation of negative control responses was subtracted

from proliferation in response to antigens: CEA protein, h-gal protein,

or p53 class II peptide. The data were averaged and graphed as DcpmF SD. To evaluate interleukin 10 (IL-10) release from CD4+ T cells in

response to these antigens, the supernatant was collected 48 hours

after culture, and IL-10 concentration was measured using mouse IL-

10 immunoassay kit (Biosource International, Camarillo, CA).

Nonspecific IL-10 production cultured in medium only was subtracted

from that induced by antigens.Cytotoxicity assay. To evaluate CD8+ T-cell responses specific for

antigens, first, spleens were pooled and dispersed into single cell

suspensions and stimulated with the H-2Db-restricted peptideCEA 526-533 (10 Ag/mL, EAQNTTYL; refs. 21, 28), the H-2K b-restricted

peptide h-gal96-103 (DAPIYTNV; ref. 29), the H-2Db-restricted peptide

p53232-240 (2 Ag/mL, KYMCNSSCM; ref. 30), or the H-2K b-restricted

peptide p15E604-611 (1 Ag/mL, KSPWFTTL, called gp70 peptide;

ref. 25). Six days later, bulk lymphocytes were separated by centri-

fugation through a Ficoll-Hypaque gradient.

Using these recovered lymphocytes, tumor-killing activity was testedas described previously (26). Briefly, the recovered lymphocytes, 51Cr-

labeled target tumor cells (EL-4, 5Â 103 cells per well), andeach peptide

were incubated for 5 hours (96-well U-bottomed plates), and radioac-

tivity in supernatants was measured using a g-counter (Corba Auto-gamma, Packard Instruments, Downers Grove, IL). As control peptides,

VSV-N52-59 (RGYVYQGL; ref. 31) was used forH-2Db-restricted peptides,

or ovalbumin257-264 (SIINFEKL; ref. 32) was used for H-2K b-restricted

peptides. In some experiments, MC38-CEA + cells, the parental MC38

cells or B16 cells were used as a target without peptide. The percentage of tumor lysis was calculated as follows: % tumor lysis = [(experimental

cpm À spontaneous cpm) / (maximum cpm À spontaneous cpm)]Â 100. Nonspecific 51Cr release in response to appropriate control


www.aacrjournals.orgClin Cancer Res 2005;11(12) June 15, 2005 4534



peptide was subtracted fromthat induced by eachtumor antigen peptide.

The data were averaged and graphed as D% F SD.Cytokine production assay. The recovered lymphocytes separated as

described above were also tested for cytokine production from CD8+ T

cells in response to antigen peptides. Lymphocytes (5 Â 105 cells per

well) were restimulated with fresh irradiated naive splenocytes (5 Â 106

cells per well) and each peptide: 10 Ag/mL of CEA peptide, 10 Ag/mL of h-gal peptide, 2 Ag/mL of p53 peptide, or 1 Ag/mL of gp70 peptide.

Twenty-four hours later, the supernatant fluid was collected and

analyzed for murine IFN-g using the Cytometric Bead Array kit (BDPharMingen). Nonspecific IFN-g production in response to appropriate

control peptide was subtracted from that induced by the each tumor

antigen peptide.Combination therapy with CEA/TRICOM vaccines, anti-CD25 mAb,

and/or g radiation. MC38-CEA + tumor cells were implanted s.c. into

the right leg of CEA-transgenic mice. Eight days after tumor implanta-

tion, mice were treated s.c. with rV-CEA/TRICOM and i.p. with anti-

CD25 mAb(300 Ag/mouse). In a protocol combining radiation therapy,

the tumor-implanted legs were irradiated at 8 Gy on day 14 according to

the method described elsewhere (33). The dose used (8 Gy) was

predetermined to have a minimal effect on the growth rate of tumors

implanted. Mice were then boosted with rF-CEA/TRICOM on days 15

and 22. As a late-phase therapy, the triple combination therapy was

started on day 13; mice were treated s.c. with rV-CEA/TRICOM and i.p.

with anti-CD25 mAb on day 13, irradiated at 8 Gy on day 19, and

boosted with rF-CEA/TRICOM on day 20. Each virus was injected at 1 Â

108 pfu/mouse admixed with 1Â 107 pfu/mouse of rF-GM-CSF. The sizeof solid tumors was measured using calipers one to two times a week.

The tumor volumes were calculated as follows: tumor volume (mm3) =

0.5 Â length Â width2. Mice were sacrificed when either size (length or

width) of tumors exceeded 20 mm.

In an indicated experiment, CD4+ T cells and/or CD8+ T cells were

depleted from the mice receiving the multimodal therapy with vaccines, tumor radiation and anti-CD25 mAb using anti-CD4

antibody ascitic fluid (GK1.5 hybridoma), and/or anti-CD8 antibody

ascitic fluid (Lyt2.2 hybridoma). The antibody ascitic fluid

(10Â dilution, 100 AL/dose) was injected into mice on days 5, 6,

and 7 after tumor implantation, and the therapy was started. The

antibody ascitic fluid was injected every week for the duration of theexperiment. The condition of T-cell depletion was validated by flow cytometry using CyChrome-conjugated anti-CD4 mAb and CyChrome-

conjugated anti-CD8 (BD PharMingen); >98% of the relevant cell

population was depleted.Statistical analysis. Significant differences were statistically evalu-

ated using ANOVA with repeated measures using Statview 4.1 (Abacus

Concepts, Inc., Berkeley, CA). For graphical representation of data,y -axis error bars indicate the SD of the data for each point on the

graph.

Results

Depletion of CD4+CD25+ Treg cells from mice treated with

anti-CD25 mAb. In previous studies, anti-CD25 mAb hasbeen given over a wide range (250-1,000 Ag/mouse) for tumor

prevention or the elimination of tumors when given several

days before tumor transplant (11, 15, 34). Studies were first conducted here to determine the optimal dose of anti-CD25

mAb to deplete CD4+CD25+ Treg cells from C57BL/6 micefrom lymph nodes, peripheral blood, and spleens. Anti-CD25mAb was given at doses of 75, 150, 300, or 600 Ag/mouse

(Fig. 1A-C). Five days after administration, inguinal lymph

node cells, spleen cells, and peripheral blood cells wereanalyzed for CD4+CD25+ cells by flow cytometry. The results

shown in Fig. 1A-C show that a dose of 300 Ag/mouse wasoptimal for the reduction of CD4+CD25+ cells in lymph nodes,

peripheral blood, and spleens. This dose was thus used for thefollowing studies.

Next, we examined the time course of CD4+CD25+ cell

depletion after anti-CD25 mAb administration. As seen inFig. 1D, the percent of CD4+CD25+ cells in lymph nodes greatly decreased 1 day after administration of anti-CD25 mAb and

remained decreased until day 28. Virtually identical results wereseen in peripheral blood (Fig. 1E). Reduction of CD4+CD25+

cells in spleens followed a similar pattern (Fig. 1F) but was not as pronounced as that seen in lymph nodes or peripheralblood. Mice treated with isotype control antibody showed nodepression of CD4+CD25+ cells through the 35-day observation

period. Determination of optimal timing for anti-CD25 mAb admin-

istration in combination with a viral vaccine. Previous studies

have shown that CEA-specific cellular immune responses wereenhanced by vaccination with CEA/TRICOM vectors (17, 19).

Here, we evaluated whether anti-CD25 mAb administrationcould enhance these T-cell responses. C57BL/6 mice were s.c.

vaccinated on day 0 with rV-CEA/TRICOM, and anti-CD25

mAb was i.p. injected on day À8, À4, À2, 0, 1, 2, 4, or 8, and

splenic lymphocytes were evaluated on day 14.Figure 2A shows CD4+ T-cell proliferation specific for CEA

protein. T-cell proliferation in response to CEA protein was not noted in mice treated with anti-CD25 mAb alone (4). Whereas

the addition of anti-CD25 mAb on days À8, À4, À2, 0, 1, and2 enhanced CEA-specific CD4+ responses compared with

vaccine alone, optimal responses were clearly observed whenthe antibody was given at the same time as vaccination

(P = 0.0004 versus vaccine alone).Figure 2B depicts IFN-g production from CD8+ T cells in

response to CEA peptide when mice received vaccines and anti-CD25 mAb at different times. These results also show that

administration of the antibody with vaccines at the same timegives optimal CEA-specific responses (50-fold higher than

vaccine alone).

Enhancement of T-cell immune responses specific for self andnon– self antigens by vaccines in combination with anti-CD25

mAb. It has been described that a role for regulatory T cells is

to suppress self-reactive immune responses (35). However, acontrolled comparison of the effect of elimination of regulatory

T cells on the generation of self versus non– self immune

responses in the identical host has yet to be investigated. To

investigate the relevancy of Treg depletion in a self-antigen

system, experiments were conducted using CEA-transgenic mice

(20, 21). CEA-transgenic mice were vaccinated with an

admixture of rV-CEA/TRICOM (self) and rV-LacZ/TRICOM(nonself) in combination with anti-CD25 mAb. Mice were

sacrificed 14 days later, and splenic lymphocytes were evaluatedusing in vitro assays.

Figure 3A and B shows CD4+ T-cell proliferation specificfor CEA or h-gal. CD4+ responses to h-gal (nonself) were,as expected, greater than those to CEA (self). The addition

of anti-CD25 mAb enhanced both CEA-specific immune

responses (P = 0.002) and h-gal immune responses (P = 0.001at 6.25 Ag/mL protein). These results indicate that anti-CD25

mAb enhanced vaccine-mediated CD4+ responses to a greater extent to the self-antigen than the non– self antigen. Wenext examined vaccine-induced CD8+ antigen-specific T-cell

responses with and without the addition of anti-CD25 mAb

(Fig. 3C and D). These results showed, however, a similar


Multimodal Therapy to Eliminate Established Tumors



increase in CD8+ responses for self (P = 0.0001) and non– self (P = 0.008 at 1.56 Ag/mL protein) antigens with the additionof anti-CD25 mAb to vaccines.

Phenotypic analysis of T cells as a consequence of vaccines, anti-

CD25 mAb, and the combination. Studies were then conductedto determine the phenotype of T cells as a consequence of

vaccines, anti-CD25 mAb, and the combination of vaccines

and anti-CD25 mAb. As seen in Fig. 4A, CD4+CD25+ cellsconstituted f4% to 5% of CD4 cells in spleens. This number

was not altered with the administration of vaccines. The

addition of anti-CD25 mAb, with or without vaccines, reducedthe percent of CD4+CD25+ cells f10-fold (Fig. 4A). When one

examined the CD25+CTLA4+ phenotype, it was seen that vaccines greatly enhanced this phenotype and that the additionof anti-CD25 mAb to the vaccines greatly reduced the presence

of these cells (Fig. 4B). An even more striking change in

phenotype was seen when one examined the CD25+GITR +

phenotype (Fig. 4C). This phenotype was reduced by the anti-CD25 mAb and enhanced by the administration of vaccines.

The addition of anti-CD25 mAb to vaccines, however, greatly

reduced this phenotype. We next examined the cytokineproduction from T cells as a consequence of vaccines, anti-CD25 mAb, or the combination of both. As can be seen in

Fig. 4D, the addition of anti-CD25 mAb to vaccines reduced the

IL-10 production from CD4+ T cells in response to the CEA antigen by 2.4-fold. This result was even more dramatic when

one looks at the reduction in h-gal-specific IL-10 production

from CD4+ T cells. The addition of anti-CD25 mAb to vaccines totally eliminated the production of IL-10 in

response to h-gal protein in mice vaccinated with a h-gal

vaccine. It was interesting to note that the inverse of this wasseen when one evaluated IFN production in response to CEA-

or h-gal-specific peptides (Fig. 4E). These results clearly show a shift to the Th1 phenotype for both the self and non–self immune responses with the combination of vaccines and anti-

CD25 mAb.

Multimodal therapy of established tumors. Previous studieshave shown that when antiCD25 mAb is given before or 1 day after tumor implantation, antitumor effects can be observed.

We confirmed and extended these results in this model system.

Fig.1. Depletion of CD4+

CD25+

Treg cells from C57BL/6micetreated withanti-CD25 mAb. A-C, dose responses.C57BL/6 mice were injected i.p. with 0 (PBS), 75,150,300, or600 Ag/mouse of anti-CD25 mAb (n = 3). Five dayslater, lymphnode cells ( A), peripheral blood cells (B), andspleen cells (C) were pooledand analyzed for CD4+CD25+

cells by flow cytometry. D-F, time course after anti-CD25mAb administration. C57BL/6 micewere injectedi.p. with300 Ag/mouse of anti-CD25 mAb (n = 3). Lymph node cells(D), peripheral blood cells (E ), and spleencells (F ) were

pooledand analyzed for CD4

+

CD25

+

cells by flow cytometryon the indicated days after anti-CD25 mAb administration.Open columns, percentage of CD4+CD25+ cells in total cells.Closed columns, percentage of CD25+ cells in CD4+ cells.





When anti-CD25 mAb was given on day 0, 3, or 6 after tumor implantation, tumor growth was inhibited compared with that

of the control group (P < 0.002; data not shown). However, no

antitumor effect of anti-CD25 mAb was seen when mAb wasgiven to mice with established tumors 8 days post-tumor implantation (P = 0.4). Based on these studies, further

combination therapy studies were conducted 8 days after tumor implantation.

As seen in Fig. 5A and B, administration of vaccines 8 dayspost-tumor implantation also had no antitumor effects(P = 0.52). As seen in Fig. 5C, the addition of anti-CD25 to

vaccines also showed no additive antitumor effects (P = 0.6

versus control; P = 0.9 versus vaccine alone). This wasdisappointing in light of the demonstration of enhanced

CD4+ and CD8+ T-cell responses specific for CEA using thiscombination therapy.

We have previously reported in two separate studies that theadministration of 8 Gy external beam radiation to this tumor

shows no antitumor effects (33, 36). We have also previously shown that the use of this dose of radiation up-regulates Fas on

tumor cells and facilitates vaccine-mediated killing of tumors

(33). These results are confirmed and extended in Fig. 5D(P = 0.02 versus vaccine alone). However, it should be notedthat whereas significant antitumor effects were seen with

vaccines plus tumor radiation, there was no elimination of the 14 established tumors in these experiments. The addition of

anti-CD25 mAb to this combination therapy (vaccines pluslocal radiation of tumors), however, showed significantly

greater antitumor effects (P = 0.01 versus vaccine plus radiationand P = 0.0001 versus vaccine plus anti-CD25 mAb). It should

also be emphasized that employing this multimodal therapy of vaccines, local tumor radiation, and anti-CD25 mAb complete-

ly eliminated established tumors from 7 of 14 mice (Fig. 5E).Next, we attempted to treat larger tumors with this multimodal

regimen by withholding vaccine therapy until day 13 after tumor implantation. When mice were vaccinated s.c. with rV-

CEA/TRICOM in combination with anti-CD25 mAb on day 13,irradiated at the site of tumors on day 19, and boosted s.c. with

rF-CEA/TRICOM on day 20, tumor growth was again signifi-cantly inhibited on day 28 (P = 0.0001 compared with the

untreated control group). Although no tumors were eliminatedin those mice, tumor growth was substantially inhibited in 6 of

10 mice. Tumor growth was strongly inhibited after radiation in both

groups receiving rV-CEA/TRICOM with/without anti-CD25

mAb at priming (Fig. 5D and E). However, tumors werecompletely eliminated only from the group of mice that received anti-CD25 mAb at the prime vaccination (Fig. 5E). To


Fig. 2. Enhancement of CEA-specificT-cell immuneresponses induced by rV-CEA/TRICOMin combination withanti-CD25 mAb. C57BL/6 mice were usedin these assays(n = 3). Control group was treated with PBS (o). AsamAb monotherapy control, micewere injected i.p. withanti-CD25 mAb (4) onday 0. As a vaccination control, micewere vaccinated s.c. withrV-CEA/TRICOM on day 0 (.).For combination treatment groups, micewere injectedi.p.with anti-CD25 mAbon dayÀ8,À4,À2,0,1, 2,4,or8 incombination with rV-CEA/TRICOM on day 0 (E).rV-CEA/TRICOM was admixed with rF-GM-CSF.Mice were sacrificedon day14; spleniclymphocytes werepooledand usedfor assays. A, CEA-specific CD4+ T-cellproliferation. Cell proliferation was measured by3H-thymidine incorporation. Data is depictedas Dcpm; cellproliferationin response to medium alone was subtractedfrom that inresponse to CEA protein (50Ag/mL). SDs arebased on the meanof triplicate wells. B, CEA-specific IFN-g

production from CD8+

Tcells. Data is depicted asDpg/mL;IFN-g production in response to the control peptide wassubtracted from thatinduced by CEA peptide (10 Ag/mL).Dottedline, levelof response (.) seen inmice vaccinatedwith rV-CEA/TRICOM.




examine which effector cells were infiltrating into tumor sites in

0 to 3 days after radiation (days 14-17 after tumor implanta-tion) when tumor mass was drastically reduced, cells were

harvested from tumors and analyzed using flow cytometry. Thepercentage of CD4 T cells (CD3+CD4+) , CD 8 T c ell s

(CD3+CD8+), natural killer (NK) cells (NK1.1+), dendritic cells(CD11c+MHCII+), B cells (CD19+MHC II+), macrophages

(Mac-3+), and neutrophils (Ly-6G+) was compared betweentwo groups (CEA/TRICOM with or without anti-CD25 mAb).

We found that CD8 T cells, dendritic cells, and NK cells wereincreased in tumors in both groups on day 14 (before

radiation) compared with those of nonvaccinated mice; there were less of these cells in tumors of mice receiving only vaccine/

radiation regimen compared with mice treated with theaddition of anti-CD25 mAb to the vaccine/radiation regimen.Of particular note, large differences were seen in the percentage

of NK cells on days 14 to 17; 5% to 8% in tumors of mice not

receiving anti-CD25 mAb versus 20% to 30% in tumors of micereceiving anti-CD25 mAb. An increase of NK cells was not seen

in peripheral blood and spleens, and the NK increase seen at tumor sites on days 14 to 17 was not evident after day 22. Nodifferences were seen in terms of the other cell populations.

These results indicated that CD8 T cells and NK cells could be

associated with antitumor responses induced by the multi-

modal therapy at local tumor sites in a short time period after

tumor radiation. To determine if immune memory was initiated as a

consequence of tumor elimination, mice cured of tumors wererechallenged with MC38-CEA + tumor cells 30 days after the last

vaccination. All mice completely rejected this rechallenge. Inaddition, we then challenged these mice with CEA-negative

parental MC38 tumors; six of seven mice (86%) rejected theparental MC38 tumors. To determine if the antitumor

immunity could cross-react to another tumor type, mice werechallenged with syngeneic B16 melanoma cells (s.c.) 30 days

after the challenge of parental MC38 tumors. None of thesemice rejected the B16 tumors; however, tumor growth was

significantly suppressed compared with that in the control mice(P = 0.0001). Tumor volume of control mice at day 18 was1,927 F 351 mm3, and tumor volume of mice cured after the

multimodal therapy was 344 F 115 mm3. In addition, mice

cured after the multimodal therapy in another experiment werechallenged with EL-4 lymphoma cells (i.v.) 85 days after the

last vaccination. All of those mice remained tumor free for theduration of the experiment (60 days after the EL-4 challenge),

whereas all control mice died of tumor on days 20 to 22 after

EL-4 implantation. The antitumor effects seen with these CEA-

negative tumors indicated that immune responses to other

Fig. 3. Enhancement of self andnon ^ selfantigen-specific

T-cell immune responses in CEA-transgenic mice vaccinatedwithTRICOM vectors in combination with anti-CD25 mAb.CEA-transgenic mice were used inthese assays (n = 3).Control group was treated with PBS (o). As a mAb control,mice were injectedi.p. anti-CD25 mAb alone (4). As avaccination control (.), mice were vaccinated s.c.withrV-CEA/TRICOM (self-antigen) and rV-LacZ/TRICOM(non ^ self antigen). For combination treatment group,mice were vaccinated with rV-CEA/TRICOM and rV-LacZ/TRICOMimmediately after i.p. injection with anti-CD25 mAb(E).Vaccines were admixed withrF-GM-CSF. Mice weresacrificed14 days after treatment; spleniclymphocytes werepooledand used for assays. A, CEA-specific CD4+

T-cell proliferation. The inset is depictedby a different scale.B, h-gal-specific CD4+ T-cell proliferation. Asterisks, meancell proliferation in response to medium alone (negativecontrol, range 444-569 cpm, P > 0.05 among groups). Datais depicted asDcpm; negative control responses were

subtracted from that in response to CEA proteinor h-galprotein. SDs are based on the meanof triplicate wells.C, CEA-specific CTL activity. D, h-gal-specific CTL activity.Tumor lysis was measured by 51Cr release in supernatants(E/T ratio = 30:1). Asterisks, mean

51Cr releasein response to

the control peptide (control response, range 7.1-10.0%,P > 0.05 among groups). Data is depictedas D%; controlresponses were subtracted fromthat inducedby CEApeptide or h-gal peptide. SDs are based on the mean oftriplicate wells.





antigens commonly expressed on these tumors were induced by the multimodal therapy regimen.

To determine the possible therapeutic mechanisms associated

with the multimodal therapy regimen, immunologic analysis was conducted using CEA-transgenic mice implanted s.c. withMC38-CEA + tumors (Fig. 5E). First, we examined which effector

T cells were required for the antitumor activity by depleting CD4cells and/or CD8 cells. Mice were treated with vaccines, local

radiation of tumors, and anti-CD25 mAb by the same scheduledescribed in Fig. 5E. Anti-CD4 antibody and/or anti-CD8antibody were injected during the therapy as described in theMaterials and Methods. In this experiment, tumor growth was

significantly inhibited compared with that of the nontreatedcontrol group (P = 0.0001), and the tumors were eradicated in 5

of 10 mice (Fig. 5G). When CD4 cells were depleted from miceduring the multimodal therapy, tumor reduction after radiation

was similar to that of the PBS-treated control group (Fig. 5H,P = 0.67 on day 17). In contrast, when CD8 cells were depleted

from mice during the multimodal therapy, tumor reduction wasnot observed in 8 of 10 mice, and antitumor effects were

significantly decreased compared with that seen in the PBS-

control group (Fig. 5I, P = 0.02 versus Fig. 5G). When both CD4and CD8 cells were depleted, antitumor effects were abrogated

in all mice, and the antitumor effect was more significantly decreased compared with CD8 cell depletion (Fig. 5J, P = 0.002

versus Fig. 5I). There were no tumor-free mice in all cases whenany T-cell subset was depleted (Fig. 5H-J). These results show

that CD8 cells were required to induce strong antitumor effectsby the multimodal therapy, but that CD4 cells are also needed

for complete elimination of tumors.Next, we examined tumor antigen-specific immune responses

of CD4 or CD8 T cells. Mice were treated with vaccines, localradiation of tumors, with or without the addition of anti-CD25

mAb according to the same schedule done in Fig. 5D and E. These mice were sacrificed, and splenocytes were assayed 29 days

after tumor implantation. As seen in Fig. 6A, there was a clear increase in the induction of CEA-specific CD4+ T cells with the

addition of anti-CD25 mAb to the vaccine/radiation regimen;no immune responses were seen in any of the groups to the

human serum albumin control protein. p53 is an antigenknown to be overexpressed in the MC38 tumor cell line (30).

We thus evaluated immune responses to p53 in light of the fact

Fig. 4. Analysis of immunosuppressivefactors. CEA-transgenic mice were used inthese assays (n = 3). Control group wastreated with PBS. As a mAb monotherapycontrol, mice were injected i.p. withanti-CD25 mAb. As a vaccination control,

mice were vaccinated s.c. with rV-CEA/TRICOM and rV-LacZ/ TRICOM. Forcombination treatment group, mice werevaccinated with rV-CEA/TRICOM andrV-LacZ/TRICOM immediately afteri.p. injection with anti-CD25 mAb. Vaccineswere admixed with rF-GM-CSF. Mice weresacrificed 14 days after treatment; spleniclymphocytes were pooled and used forassays. A, % CD4+CD25+ cells in total cells.B, % CD25

+CTLA-4

+cells in CD4

+cells.

The panels were depicted after gatingCD4

+fraction. C, % CD25

+GITR

+cells in

CD4+ cells. The panels were depicted aftergating CD4

+fraction. Inset, CD25

histogram after gating CD4+GITR+ fraction.D, IL-10 production from CD4+ Tcells inresponse to CEA or h-gal protein(50 Ag/mL). Data is depicted as Dpg/mL;

IL-10 production cultured in medium wassubtracted from that induced by CEA orh-gal protein. E , IFN-g production fromCD8+ Tcells in response to CEA or h-galpeptide (10 Ag/mL). Data is depicted asDpg/mL; IFN-g production in response tothe control peptide was subtracted fromthat induced by CEA or h-gal peptide.





that CEA negative tumors were also eliminated upon rechal-lenge of mice cured of tumors. As seen in Fig. 6B, T-cell

responses to the 15-mer p53 peptide were seen in mice that

received vaccines plus radiation. These responses, however, weresubstantially increased when anti-CD25 mAb was added to the

vaccine/radiation regimen (P = 0.0017 at 0.16 Ag/mL protein).

This is a clear demonstration of an antigen cascade enhanced by the multimodal therapy regimen. These results were further

extended when analyzing CTL activity. As seen in Fig. 6C, CEA-specific CTL activity was greatly enhanced when anti-CD25 mAb

was added to the vaccine/radiation regimen. No CTL activity

was observed to the VSV control peptide. Only a slight and

nonstatistical increase was seen in p53-specific CD8+ T-cellresponses with the addition of anti-CD25 mAb (Fig. 6D).

Another endogenous tumor-associated antigen, gp70, haspreviously been shown to be overexpressed in MC38 tumors

(25, 33). As seen in Fig. 6E, there was a strong gp70-specific T-cell response seen in mice receiving vaccines plus local tumor

radiation; moreover, there was a statistically significant increase

in this gp70 T-cell response (P = 0.004 at an E/T = 5:1) in micetreated with the addition of anti-CD25 mAb to the vaccine/radiation regimen. This CTL activity in response to gp70

antigen, not encoded in vaccines but expressed on tumors, wasmuch greater than that seen in response to the other tumor

antigens (Fig. 6E versus C and D).Further studies were conducted to examine the potential role

of specific CD8+ T cells directed against antigens not presented

by the CEA vaccine, in the antitumor response. Studies were

conducted to examine the direct CTL activity against MC38-CEA + tumor cells (CEA +, p53+, and gp70+), the parental MC38

tumor cells (CEA À, p53+, and gp70+), or B16 tumor cells(CEA À, p53+, and gp70+). We first tested CEA-specific CD8+ T

Fig. 5. Enhancement of antitumor efficacyinducedby vaccines andlocal radiation oftumors in combinationwith anti-CD25m Ab.CEA-transgenic mice wereimplanted s.c.with MC38-CEA+ tumors on day0. A-E ,tumor therapy with vaccines, tumorradiation, and/or anti-CD25 mAb. A, control

mice were injectedi.p. with PBS onday 0and s.c. withPBS ondays 15 and 22.B, micewere vaccinated withrV-CEA/TRICOMin combination withisotypeimmunoglobulin on day 8 andboosted s.c. withrF-CEA/TRICOM on days15 and 22. C, mice were vaccinated withrV-CEA/TRICOMin combination withanti-CD25 mAb onday 8 and boosted s.c.withrF-CEA/TRICOM on days15 and22. D, mice werevaccinated withrV-CEA/TRICOMin combination withisotypeimmunoglobulin on day 8. Six dayslater, mice were irradiated at the site oftumors.Then, mice were boosteds.c. withrF-CEA/TRICOM on days15 and 22. E, micewere vaccinated with rV-CEA/TRICOMincombination withanti-CD25 mAb on day 8.Sixdays later, micewere irradiatedat the site

of tumors.Then, micewere boosted s.c.withrF-CEA/TRICOM on days15 and 22.P s are depicted on day 28 compared withthe PBS-control group. Compilation of twoseparate experiments. F-J, T-cell depletionfrom mice receiving the triple combinationtherapy. F, control micewere injected i.p.withPBS onday0 and s.c. withPBS ondays15 and 22. G-J, micewere treatedwith anti-CD4 antibody and/or anti-CD8antibodyand vaccinated withrV-CEA/TRICOMin combination withanti-CD25 mAb onday 8, irradiated at thesite of tumors on day14, and boosted s.c.withrF-CEA/TRICOM on days15 and 22.Eachdepleting antibodywas injected for thedurationof the experiment as described inMaterials and Methods. G, micewere

treated i.p. with PBS as a control. H, micewere treated i.p. withanti-CD4 antibody.I , mice were treated i.p. with anti-CD8antibody. J, mice were treated i.p. withanti-CD4 andanti-CD8 antibody. Eachviruswas admixed with rF-GM-CSF. Tumorvolume was monitored one to twotimesa week.





cells obtained from mice treated with vaccine/radiationregimen with/without anti-CD25 mAb. These cells were

stimulated with CEA peptide for 6 days. CEA-specific CD8+ T

cells from mice receiving only the vaccine/radiation regimen

did not show marked CTL activity against any tumor lines usedfor assays (Fig. 6F). In contrast, CEA-specific CD8+ T cells from

mice receiving the addition of anti-CD25 mAb to the vaccine/

radiation regimen showed marked killing activity directed

Fig. 6. Analysis of therapeutic mechanism induced bythemultimodal therapy withvaccines, local radiationoftumors and anti-CD25 mAb. CEA-transgenic micewereimplanted s.c. with MC38-CEA+ tumors on day0.These micewere treated with CEA/TRICOMvaccines admixed with rF-GM-CSF, tumor radiation,anti-CD25 mAb, and/or the isotype immunoglobulinaccording to thesame protocol described in Fig. 5.The mice were sacrificed; spleniclymphocytes werepooledand usedfor in vitro assays 29 daysafter tumor implantation (n = 4). A and B, tumorantigen ^ specific CD4

+T-cellproliferation.

A, CEA-specific CD4+ T-cell proliferation. Opensquares, control response to human serum albumin(HSA, 50 Ag/mL) in the group vaccinated withCEA/TRICOM vaccines in combination with radiation

and isotype immunoglobulin. Diamonds, controlresponse to HSA inthe group vaccinatedwithCEA/TRICOM vaccines in combination with radiationand anti-CD25 mAb. B, p53-specific CD4+ T-cellproliferation. Asterisks, mean cell proliferationinresponse to medium alone (negative control, range495-551cpm, P > 0.05 among groups). Data isdepictedas Dcpm; negative control responses weresubtracted from thatin response to CEA proteinor p53class II peptide.SDs are based on the mean oftriplicate wells. C-E , CTL activity inresponse to tumorantigen peptide. C, CEA-specific CTL activity(E/T ratio = 50:1). D, p53-specific CTL activity(2Ag/mL p53 peptide). E , gp70-specific CTL activity(1Ag/mL gp70 peptide). Asterisks, mean51Cr releasein response to the control peptide (control response,range 3.5-4.4%, P > 0.05 among groups). Data isdepictedas D%; control responses were subtracted

fromthat induced by eachantigen peptide. SDs arebased on the meanof triplicate wells. Closed squares,micetreated withvaccines in combination with tumorradiation and isotype immunoglobulin. Crossedsquares, micetreated with vaccines in combinationwith tumor radiation andanti-CD25 mAb. F-H,

direct CTL activity against tumor cells(E/T ratio = 50:1). F , CTL activity of CEA-specificCTLs(CEA-CTLs). Splenic lymphocytes were stimulatedwithCEApeptide for 6 days and used for 51Cr releaseassay. G, CTL activity of p53-specific CTLs(p53-CTLs). Splenic lymphocytes were stimulatedwithp53 peptide for 6 days and used for 51Cr releaseassay. H, CTL activity of gp70-specific CTLs(gp70-CTLs). Splenic lymphocytes were stimulatedwith gp70 peptide for6 days and used for

51Cr release

assay. I, generationof gp70-specific CD8+ Tcells bythe multimodal therapy with vaccines, tumor radiation,and anti-CD25 mAb. Cells fromtumors, blood,and spleens were stained with anti-CD3 mAb,anti-CD8 mAb, and gp70-tetramer. a-c, micewerevaccinated with vaccines in combination withtumor radiation and isotype immunoglobulin.a, tumor-infiltrating Tcells. b, peripheral bloodT cells.c, splenicTcells. d-f, mice werevaccinated withvaccines in combination with tumor radiation andanti-CD25 mAb. d, tumor-infiltratingT cells.e, peripheral bloodTcells. f, splenicTcells.The numbersindicate the percentage of gp70-tetramer-bindingCD8+ Tcells intotal cells.These panels were depictedafter gating theCD3+ fraction.





against MC38-CEA + tumor cells (Fig. 6F). The CTL activity wassignificantly higher than that seen in mice treated with only

vaccine/radiation (Fig. 6F, P = 0.0001). When CEA-negative

MC38 cells or B16 cells were used as a tumor targets, however,CTL activity was not observed, and there were no significant differences between these groups (Fig. 6F). Next, p53-specific

CD8+ T cells, which were stimulated with p53 peptide for 6 days, were tested for direct CTL activity against tumors. P53-

specific CD8+

T cells from mice treated with only vaccine/radiation did not exhibit CTL activity against any tumor linesused for assays (Fig. 6G). However, p53-specific CD8+ T cellsfrom mice receiving the addition of anti-CD25 mAb to the

vaccine/radiation showed marked killing activity directedagainst all tumor lines, and there were significant differences

compared with those seen in mice receiving only vaccine/radiation (Fig. 6G, P < 0.0004). gp70-specific CD8+ T cells from

mice treated with the vaccine/radiation regimen showed slight CTL activity against MC38-CEA + and MC38 tumor cells but not

against B16 (Fig. 6H). In contrast, gp70-specific CD8+ T cellsfrom mice receiving the addition of anti-CD25 mAb to the

vaccine/radiation regimen showed killing activities directed

against all tumor lines, and significant differences wereobserved compared with those of mice receiving only vac-cine/radiation (Fig. 6H, P < 0.0004). Interestingly, the CTL

activity shown by gp70-specific CTLs was the greatest in thisgroup compared with those shown by the other CTLs (Fig. 6H

versus F and G).Because gp70-specific tetramer was available, we conducted

gp70-tetramer binding assays to detect the gp70-specific CD8+

T cells in these mice. It should be pointed out that CEA- and

p53-specific tetramer was not available, but it is clearly shown(Fig. 6A-H) that CEA- and p53-specific T cells are also present

in the periphery after vaccination. Cells were harvested fromtumors, peripheral blood, and spleens of these mice andanalyzed for gp70-tetramer binding using flow cytometry. As

seen in Fig. 6Ia, 15% of CD8+ T cells seen in tumors were

gp70 specific following treatment of mice with vaccines andlocal tumor radiation. This number was increased to 35%

when mice were treated with the multimodal therapy using

vaccines, radiation, and anti-CD25 mAb (Fig. 6Id). Similar increases in gp70-specific CD8+ T cells were also seen in blood

and spleens from mice receiving the multimodal therapy

compared with those from mice receiving only vaccines andradiation (Fig. 6Ie and f).

These studies thus showed that a multimodal therapy was

required for the elimination of established tumors and that anincrease in CD4+ and CD8+ cells specific for a vaccine-directed

antigen (CEA) as well as an antigen cascade of T-cell responses

to other tumor-associated antigens (p53 and gp70) wasassociated with antitumor effects.

Discussion

Previous studies have shown that in vitro depletion of CD4+CD25+ immunosuppressive Treg cells can enhance T-cell

immune responses (3, 37–53). In addition, it has been also

shown that in vivo antitumor activity could be augmented by depletion of CD4+CD25+ cells via anti-CD25 mAb administra-tion (10, 11, 13, 14, 34). Based on the emerging characteristics

of the suppressive actions of CD4+CD25+ cells, many research-

ers hypothesized that a strategy of Treg cell depletion couldaugment the efficacy of cancer immunotherapy. Vaccine therapy

in combination with anti-CD25 mAb has also been conductedusing animal tumor models, with some studies showing

antitumor effects (15, 16). However, in these studies, anti-CD25 mAb was given several days before vaccination, or one

day after tumor implantation. To our knowledge, there are noreports investigating the optimal timing for anti-CD25 mAb in

combination with antitumor vaccination. Here, we show that vaccine-induced T-cell immune responses could be optimally

augmented when anti-CD25 mAb was combined at the sametime as vaccination (Fig. 2).

CD25 (IL-2 receptor) is expressed on Treg cells as well asactivated effector T cells. It is interesting to note that in the

studies reported here, the administration of anti-CD25 mAb at the time of vaccination reduced the number of CD4+CD25+

cells (Fig. 1), but at the same time enhanced vaccine-inducedCD4+ and CD8+ T-cell responses, thus apparently not having an

inhibitory effect on activated T-cell responses (Fig. 2). Additionally, it was shown here that anti-CD25 mAb combi-

nation with vaccines could enhance T-cell immune responses

specific for a self-antigen as well as those specific for a non – self antigen (Fig. 3). To our knowledge, this is the first study that examines vaccine-induced T-cell immune responses to self-

antigens and non – self antigens simultaneously.

It has been reported that Treg cells constitutively express

CTLA-4 and GITR, produce immunosuppressive cytokines, anddown-regulate maturation of dendritic cells (3, 44, 45, 53).

When we examined these immunosuppressive factors, thepercentage of GITR +CD4+CD25+ cells was decreased, and

IL-10 production from CD4+ T cells was strongly inhibitedas a consequence of vaccines plus anti-CD25 mAb (Fig. 4).

The percentage of CD4+CD25+CTLA-4+ cells was increased with vaccines but reduced with the addition of anti-CD25

mAb. In addition, the percentage of activated dendritic cells(CD11c+MHC II+ cells) was increased compared with those

seen in mice receiving vaccines alone (data not shown). These

data would suggest that anti-CD25 mAb/vaccine combination

could augment T-cell immune responses induced by vaccines

via inhibition of immunosuppressive factors. The studies reported here (see Fig. 5) show that the combined

use of vaccines, external beam radiation of tumors, and anti-

CD25 mAb resulted in optimal antitumor effects. We analyzedthe potential mechanisms induced by the triple combination

therapy using CEA + tumor–bearing mice. When mice received

anti-CD25 mAb in combination with vaccines and radiation,CEA-specific T-cell immune responses (particularly, CEA-specif-ic CTL activity) were significantly enhanced compared with

those in mice not receiving anti-CD25 mAb (Fig. 6A and C).Moreover, T-cell immune responses specific for the other tumor

antigens (p53 and gp70), not encoded in vaccines, were strongly increased in mice receiving the triple combination therapy. p53-

specific CD4+ T-cell responses were greatly augmented by theaddition of anti-CD25 mAb to the vaccine/radiation combina-

tion therapy (Fig. 6B). In addition, the CTL activity wasgreatly induced in response to gp70 (Fig. 6E), and a high

level of gp70-specific CTLs was generated in mice receiving the triple combination therapy (Fig. 6I). At tumor challenge

of mice cured after the multimodal therapy, tumor growth of CEA-negative tumor cells was eliminated or strongly sup-

pressed. This suggests that the tumor therapy was accompanied





by the induction of an antigen cascade of immune responsesto other antigens commonly expressed on these tumors.

Many studies have shown antitumor efficacy of anti-CD25

mAb itself (10, 11, 13, 14, 34). In these studies, however,anti-CD25 mAb was given several days before or immediately after tumor implantation. In the tumor model employed here,

the administration of anti-CD25 mAb to mice with well-established tumors 8 days after tumor implantation was

ineffective (Fig. 5C). The administration of vaccines 8 daysafter tumor implantation was also ineffective (Fig. 5B). Wehave also previously shown that the single treatment withexternal beam radiation of established tumors is also

insufficient to cause antitumor effects (33, 36). These studiesdid show, however, that external beam radiation of tumors at

doses below that which cause antitumor effects up-regulatedthe expression of Fas on tumor cells and thus rendered them

more susceptible to vaccine-induced T-cell mediated lysis. It was further shown that the up-regulation of Fas was necessary

for tumor killing by the fact that a tumor expressing adominant-negative Fas was insensitive to killing. It was also

shown in these studies that both CEA expression in the tumor

and the use of a vaccine expressing the CEA transgene wereboth necessary for antitumor effects (33).

The studies reported here show the complexity of achieving acure of a well-established tumor. Here we have employed three

different modalities to achieve this goal: (i) vaccines to induceantigen-specific T-cell responses, (ii) external beam radiation of

tumors to up-regulate Fas and thus make tumor cells moresusceptible to T-cell killing, and (iii) the use of anti-CD25 mAb

to eliminate CD4+

CD25+

suppressor cells. Recent in vitrostudies have shown that external beam radiation of a range of tumor types at doses below cytotoxic doses will up-regulate Fasand render human tumor cells more susceptible to antigen-

specific T-cell killing (54). Several vaccine trials employing TRICOM vaccines are also in progress (55), and external beam

radiation of tumors is a well-established modality for thetreatment and/or palliation of a range of human tumors. The

use of several drugs, antibodies, or fusion proteins to reduce or eliminate human suppressor cells is also currently being

evaluated in the clinic. The studies reported here thus formthe rationale for potential clinical trials employing multimodal

immune mediated therapies.


References1. Woo EY, Chu CS, Goletz TJ, et al. Regulatory

CD4(+)CD25(+) T cells in tumors from patientswith early-stage non-small cell lung cancer andlate-stage ovarian cancer. Cancer Res 2001;61:4766^72.

2. Sakata H, Murakami S, Hirayama R. Serum solubleinterleukin-2 receptor (IL-2R) and immunohisto-chemical staining of IL-2R/Tac antigen in colorectalcancer. Int J Clin Oncol 2002;7:312 ^ 7.

3. Woo EY,Yeh H, Chu CS, et al. Cutting edge: Regula-tory T cells from lung cancer patients directly inhibitautologous T cell proliferation. J Immunol 2002;168:4272^6.

4. Wolf AM,Wolf D, Steurer M, et al. Increase of regula-

tory T cells in the peripheral blood of cancer patients.Clin Cancer Res 2003;9:606^ 12.

5. Sa sa d a T , Kim u r a M, Yo sh id a Y, Ka n a i M,Takabayashi A. CD4+CD25+ regulatory T cells inpatients with gastrointestinal malignancies: possibleinvolvement of regulatoryT cells in disease progres-sion. Cancer 2003;98:1089 ^ 99.

6. Pavlidis N, Nicolaides C, Bairaktari E, et al. Solubleinterleukin-2 receptors in patients with advancedcolorectal carcinoma. Int J Biol Markers 1996;11:6 ^ 11.

7. Zaloudik J, Lauerova L, Janakova L, et al. Signifi-cance of pre-treatment immunological parameters incolorectal cancer patients with unresectable metasta-ses to the liver. Hepatogastroenterology 1999;46:220^7.

8. Saito H,Tsujitani S, Ikeguchi M, Maeta M, Kaibara N.Serum level of a s oluble receptor for interleukin-2 as aprognostic factor in patients with gastric cancer. On-cology1999;56:253 ^ 8.

9. Pavone L, Andrulli S, Santi R, Majori M, Buzio C.Long-term treatment with low doses of interleukin-2and interferon-a: immunological effects in advancedrenal cell cancer. Cancer Immunol Immunother 2001;50:82^6.

10.Tawara I,TakeY, Uenaka A, Noguchi Y, Nakayama E.Sequential involvement of two distinct CD4(+) regu-latoryTcells during the course of transplantable tumorgrowth and protection from 3-methylcholanthrene-induced tumorigenesis by CD25-depletion. Jpn JCancer Res 2002;93:911 ^ 6.

11. Jones E, Dahm-VickerM, Simon AK, et al. Depletionof CD25+ regulatory cells results in suppression ofmelanoma growth and induction of autoreactivity inmice. Cancer Immun 2002;2:1.

12. Li J, Hu P, Khawli LA, Epstein AL. Completeregression of experimental solid tumors by com-bination LEC/chTNT-3 immunotherapy andCD25(+) T-cell depletion. Cancer Res 2003;63:8384^92.

13. Shimizu J,Yamazaki S, Sakaguchi S. Induction oftumor immunity by removing CD25

+CD4

+T cells:

a common basis between tumor immunity and auto-immunity. J Immunol 1999;163:5211 ^ 8.

14. Tanaka H, Tanaka J, Kjaergaard J, Shu S. Deple-tion of CD4+ CD25+ regulatory cells augments thegeneration of specific immune T cells in tumor-draining lymph nodes. J Immunother 2002;25:207 ^ 17.

15. Sutmuller RP, van Duivenvoorde LM, van Elsas A,et al. Synergism of cytotoxicT lymphocyte-associatedantigen 4 blockade and depletion of CD25(+) regula-tory T cells in antitumor therapy reveals alternativepathways for suppression of autoreactive cytotoxicT lymphocyte responses. J Exp Med 2001;194:823^32.

16. Casares N, Arribillaga L, Sarobe P, et al. CD4+/CD25+ regulatory cells inhibit activation of tumor-primed CD4+ Tcells with IFN-g-dependent antiangio-genic activity, as well as long-lasting tumor immunityelicited by peptide vaccination. J Immunol 2003;171:5931 ^ 9.

17. Grosenbach DW, Barrientos JC, Schlom J,Hodge JW. Synergy of vaccine strategies to amplifyantigen-specific immune responses and antitumoreffects. Cancer Res 2001;61:4497 ^ 505.

18. Greiner JW, Zeytin H, Anver MR, Schlom J.

Vaccine-based therapy directed against carci-noembryonic antigen demonstrates antitumor ac-tivity on spontaneous intestinal tumors in theabsence of autoimmunity. Cancer Res 2002;62:6944 ^ 51.

19. Aarts WM, Schlom J, Hodge JW. Vector-basedvaccine/cytokine combination therapy to enhanceinduction of immune responses to a self-antigenand antitumor activity. Cancer Res 2002;62:5770 ^ 7.

20. Eades-Perner AM, van der Putten H, Hirth A, et al.Mice transgenic for thehumancarcinoembryonic anti-gen gene maintain its spatiotemporal expressionpattern. Cancer Res1994;54:4169 ^ 76.

21. Kass E, Schlom J,Thompson J, et al. Induction ofprotectivehost immunity to carcinoembryonicantigen(CEA), a self-antigen in CEA transgenic mice, by

immunizing with a recombinant vaccinia-CEA virus.Cancer Res1999;59:676^ 83.

22. Robbins PF, KantorJA, Salgaller M, et al. Transduc-tion and expression of the human carcinoembryonicantigen gene in a murine colon carcinoma cell line.Cancer Res 1991;51:3657 ^ 62.

23. Hodge JW, Sabzevari H, Yafal AG, et al. A triad ofcostimulatory molecules synergize to amplify T-cellactivation. Cancer Res1999;59:5800 ^ 7.

24. Kass E, Panicali DL, Mazzara G, Schlom J, GreinerJW. Granulocyte/macrophage-colony stimulatingfac-tor produced by recombinant avian poxvirusesenriches the regional lymph nodes with antigen-pre-senting cells and acts as an immunoadjuvant. CancerRes 20 01;61:20 6 ^ 14.

25. Yang JC, Perr y-Lalley D. The envelope proteinof an endogenous murine retrovirus is a tumor-associated T-cell antigen for multiple murine tumors.JImmunother 2000;23:177^ 83.

26. KantorJ,Irvine K, Abrams S, et al.Antitumoractivityand immune responses induced by a recombinantcarcinoembryonic antigen-vaccinia virus vaccine.J Natl Cancer Inst 1992;84:1084 ^ 91.

27. Zwaveling S, Vierboom MP, Ferreira Mota SC,et al. Antitumor efficacy of wild-type p53-specificCD4(+) T-helpercells. CancerRes 2002;62:6187^ 93.

28. Kalus RM, Kantor JA, Gritz L, et al. The use ofcombination vaccinia vaccines and dual-gene vac-cinia vaccines to enhance antigen-specific T-cell im-munity via T-cell costimulation. Vaccine 1999;17:893^903.

29. Overwijk WW, Surman DR,Tsung K, Restifo NP.Identification of a Kb-restricted CTL epitope of h-galactosidase: potential use in development of immu-nization protocols for ‘‘self’’ antigens. Methods 1997;12:117 ^ 23.

30. Hilburger Ryan M, Abrams SI. Characterization ofCD8+ cytotoxicT lymphocyte/tumor cell interactionsreflecting recognition of an endogenously expressedmurine wild-type p53 determinant. Cancer ImmunolImmunother 2001;49:603 ^ 12.

31. Esquivel F,Yewdell J, Bennink J. RMA/S cells pres-ent endogenously synthesized cytosolic proteins toclass I-restrictedcytotoxicT lymphocytes. J Exp Med1992;175:163 ^ 8.

32. el-Shami K, Tirosh B, Bar-Haim E, et al. MHC classI-restricted epitope spreading in the context of tu-mor rejection following vaccination with a single




immunodominant CTL epitope. Eur J Immunol 1999;29:3295 ^ 301.

33. Chakraborty M, Abrams SI, Coleman CN, et al.External beam radiation of tumors alters phenotypeof tumor cells to render them susceptible to vac-cine-mediated T-cell killing. Cancer Res 2004;64:4328^37.

34. Onizuka S,Tawara I, Shimizu J, et al.Tumor rejectionby in vivo administration of anti-CD25 (interleukin-2receptor a) monoclonal antibody. Cancer Res 1999;

59:3128 ^ 33.35. Hsieh CS, Liang Y, Tyznik AJ, et al. Recognition of

the peripheral self by naturally arising CD25+ CD4+ Tcell receptors. Immunity 2004;21:267 ^ 77.

36. Chakraborty M, Abrams SI, Camphausen K, et al.Irradiation of tumor cells up-regulates Fas and enhan-ces CTL lytic activity and CTL adoptive immunother-apy. J Immunol 2003;170:6338 ^ 47.

37. Read S, Malmstrom V, Powrie F. Cytotoxic Tlymphocyte-associated antigen 4 plays an ess entialrole in the function of CD25(+)CD4(+) regulatorycells that control intestinal inflammation. J Exp Med2000;192:295 ^ 302.

38. TakahashiT,TagamiT,Yamazaki S, et al. Immunologicself-tolerance maintained by CD25(+)CD4(+) regula-toryT cells constitutively expressing cytotoxicT lym-phocyte-associated antigen 4. J Exp Med 20 00;192:303^10.

39. Piccirillo CA, Shevach EM. Cutting edge: control ofCD8+ Tcell activation by CD4+CD25+ immunoregula-tory cells. J Immunol 2001;167:1137 ^ 40.

40. Bystry RS, Aluvihare V,Welch KA, Kallikourdis M,Betz AG. B cells andprofessional APCs recruitregula-tory T cells via CCL4. Nat Immunol 2001;2:1126 ^ 32.

41. Nakamura K, Kitani A, Strober W. Cell contact-dependent immunosuppression by CD4(+)CD25(+)regulatory T cells is mediated by cell surface-boundtransforming growth factor h. J Exp Med 2001;194:629^44.

42. Sakaguchi S, Sakaguchi N, Shimizu J, et al. Immu-nologic tolerance maintained by CD25

+CD4

+regula-

tory T cells: their common role in controllingautoimmunity, tumor immunity, and transplantationtolerance. Immunol Rev 2001;182:18^ 32.

43. Chai JG, Tsang JY, Lechler R, et al. CD4+CD25+ Tcells as immunoregulatoryT cells in vitro. Eur J Immu-nol 2002;32:2365^ 75.

44. Morse MA, ClayTM, Mosca P, Lyerly HK. Immuno-regulatory T cells in cancer immunotherapy. ExpertOpin BiolTher 2002;2:827^ 34.

45 . McHugh RS, Whitters MJ, Piccirillo CA, et al.CD4(+)CD25(+) immunoregulatory T cells: geneexpression analysis reveals a functional role forthe glucocorticoid-induced TNF receptor. Immunity2002;16:311 ^ 23.

4 6. Chen W, Wahl SM. TGF-h: the missing link inCD4(+)CD25(+) regulatoryT cell-mediated immuno-suppression. Cytokine Growth Factor Rev 2003;14:85^9.

47. Jonuleit H, Schmitt E, Schuler G, Knop J, EnkAH. Induction of interleukin 10-producing, nonpro-liferating CD4(+) T cells with regulatory properties

by repetitive stimulation with allogeneic immaturehuman dendritic cells. J Exp Med 2000;192:1213 ^ 22.

48. Jonuleit H, Schmitt E, Steinbrink K, Enk AH. Den-dritic cells as a tool to induce anergic and regulatoryTcells.Trends Immunol 2001;22:394 ^ 400.

49. Jonuleit H, Schmitt E, Kakirman H, et al. Infectioustolerance: human CD25(+) regulatory T cells conveysuppressor activity to conventional CD4(+) T helpercells. JExp Med 2002;196:255^ 60.

50. Kursar M, Bonhagen K, FensterleJ,et al. RegulatoryCD4+CD25+ T cells restrict memory CD8+ T cellresponses. J Exp Med 2002;196:1585 ^ 92.

51. Somasundaram R, Jacob L, Swoboda R, et al.Inhibition of cytolytic T lymphocyte proliferationby autologous CD4

+/CD25

+regulatory T cells in

a colorectal carcinoma patient is mediated bytransforming growth factor-h. Cancer Res 2002;62:5267^72.

52. Dieckmann D, Bruett CH, Ploettner H, Lutz MB,Schuler G. Human CD4(+)CD25(+) regulatory, con-tact-dependent T cells induce interleukin 10-produc-ing, contact-independent t ype 1-like regulatory T cells[corrected]. J Exp Med 2002;196:247^ 53.

53. Manzotti CN,Tipping H, Perry LC, et al. Inhibitionof human T cell proliferation by CTLA-4 utilizesCD80 and requires CD25+ regulatory T cells. Eur JImmunol 2002;32:2888 ^ 96.

54. Garnett CT, Palena C, Chakarborty M, Tsang KY,Schlom J, Hodge JW. Sublethal irradiation of humantumor cells modulates phenotype resulting in en-hanced killing by cytotoxicT lymphocytes. CancerRes 2004;64:7985^ 94.

55. Marshall JL, Gulley JL, Arlen PM, et al. Phase I

study of sequential vaccinations with fowlpox-CEA(6D)-TRICOM alone and sequentially withvaccinia-CEA(6D)-TRICOM, with and withoutgranulocyte-macrophage colony-stimulating factor,in patients with carcinoembryonic antigen-expressingcarcinomas. J Clin Oncol 2004;23:720 ^ 31.


www aacrjournals orgClin Cancer Res 2005;11(12) June 15 2005 4544

cancer terapies

Documents