GENETIC IMMUNOTHERAPY BY INTRAPLEURAL, INTRAPERITONEALAND SUBCUTANEOUS INJECTION OF IL-2 GENE-MODIFIED LEWIS LUNGCARCINOMA CELLSYuji HEIKE1, Minako TAKAHASHI 1, Tatsuo OHIRA1, Ichiro NARUSE1, Seiji HAMA 1, Yuichiro OHE2, Takashi KASAI2, Hisao FUKUMOTO1,Kristin J. OLSEN3, Eckhard E. PODACK3 and Nagahiro SAIJO1*1Pharmacology Division, National Cancer Center Research Institute, Tokyo, Japan2Department of Internal Medicine, National Cancer Center Hospital, Tokyo, Japan3Department of Microbiology and Immunology, University of Miami School of Medicine, Miami, FL, USA

The induction and augmentation of tumor non-specificimmunity and of tumor-specific immunity by intrapleural,intraperitoneal and subcutaneous injection of interleukin-2(IL-2) gene-modified Lewis lung carcinoma (LLC) cells (LLC-IL2) was tested in C57BL/6 mice. Intrapleural injection of LLCcells induced lung tumors with a malignant effusion, intraperi-toneal injection induced peritoneal tumors with ascites andsubcutaneous injection induced subcutaneous tumors. Intra-pleural injection of irradiated LLC-IL2 cured pre-existing lungLLC tumors and extended the survival of the mice but did notaffect survival of mice with pre-existing peritoneal tumorsnor did it affect the growth of s.c. tumors. Intraperitonealinjection of irradiated LLC-IL2 cured pre-existing LLC perito-neal tumors and extended the survival of the mice but did notaffect survival of mice bearing lung tumors nor did it affect thegrowth of s.c. tumors. Subcutaneous injection of irradiatedLLC-IL2 did not affect the growth of preexisting s.c. tumorsand also did not improve survival of mice bearing the lung orperitoneal tumors. Injection with irradiated LLC-IL2 by allroutes, i.e., intrapleural, intraperitoneal and s.c., protectedagainst subsequent re-challenge with LLC. Eight days afterthe initial immunization (early stage of immunization), non-adherent mononuclear cells in the peritoneal cavity of themice treated with intraperitoneal injection of irradiatedLLC-IL2 displayed enhanced cytotoxicity against LLC, B16-F10 and P815 cells, while the cytotoxic activity of spleen cellsin the same mice did not change. The efficiency of induction oftumor-specific immunity was the strongest after intraperito-neal immunization and weakest after s.c. immunization. Invitro analysis using the spleen cells of mice immunized withirradiated LLC-IL2 suggested that CD81 T cells play a keyrole in tumor-specific immunity. Int. J. Cancer 73:844–849,1997.r 1997 Wiley-Liss, Inc.

The incidence of lung cancer has been increasing in manydeveloped countries, where it represents the most common cause ofcancer death. Despite recent improvements in cancer therapy, theprognosis for lung cancer has not improved.

Since the 1970s, immunotherapy against lung cancers usingcytokines (Clamonet al., 1993; Jettet al., 1994; Schilleret al.,1995; Scudelettiet al.,1993; Yanget al.,1991), biological responsemodifiers (Watanabe and Iwa, 1984; Yamamuraet al.,1979; 1983),adoptive transfer of lymphokine-activated killer (LAK) cells(Bernsteinet al.,1991; Dillmanet al.,1993) and tumor-infiltratinglymphocytes (TIL) (Dillmanet al., 1993) has been attempted.However, therapeutic trials have failed to improve therapeuticefficacy.

Tumor vaccine therapy using cytokine gene-modified cancercells has been studied as a novel approach in anti-cancer immuno-therapy in many laboratories. Basic experiments revealed that thes.c. injection of cancer cells modified by cytokine genes, such asinterleukin 2 (IL-2; Belldegrunet al., 1993; Gansbacheret al.,1990; Oheet al., 1993; Rosenthalet al., 1994), IL-4 (Oheet al.,1993; Ohiraet al.,1994b; Golumbeket al.,1991), IL-6 (Porgadoret al.,1992), IL-7 (McBrideet al.,1992), granulocyte/macrophagecolony stimulating factor (GM-CSF; Dranoffet al., 1993), inter-feron (IFN-a, b, g; Belldegrunet al.,1993; Rosenthalet al.,1994;

Watanabeet al., 1989; Porgadoret al., 1993) and tumor necrosisfactor (TNF; Ohiraet al., 1994a; Asher et al., 1991), inducedtumor-specific immunity. Our group also reported the possibility ofs.c. vaccination with IL-2 gene-modified cancer cells in mice usingLewis lung carcinoma (LLC) models (Oheet al.,1993; Ohiraet al.,1994a and b). In the United States and some other countries,clinical trials of tumor vaccination therapy using cytokine gene-modified cancer cells have already been initiated against metastaticmelanoma, renal cell carcinoma and some other advanced cancers.A recent clinical study revealed that the tumor vaccine therapiesusing IL-2 or GM-CSF gene-modified cancer cells could induceimmunological reactions such as delayed-type hypersensitivity andtissue infiltration with T cells. To date, however, no obvious clinicalresponse has been demonstrated.

In a previous study, we reported that transfer of the IL-2 geneinto LLC abrogated its tumorigenicity and that s.c. injections ofIL-2 gene-modified LLC (LLC-IL2) induced tumor-specific immu-nity. In this study, we have evaluated therapeutic effects of inducingtumor-specific immunity by intrapleural, intraperitoneal and s.c.injections of irradiated LLC-IL2 on pre-existing tumors. We alsodiscuss the effectiveness of genetic immunotherapy using IL-2gene-modified cancer cells.

MATERIAL AND METHODS

Cell lines and culture

The LLC cell line originates from a spontaneous lung carcinomaof a C57BL/6 mouse (Sugiura and Stock, 1955). LLC-IL2 is themurine IL-2 gene-transduced LLC cell line established in ourlaboratory (Oheet al.,1993). Briefly, the mouse IL-2 gene from theATCC (Rockville, MD) was introduced into the Sal I site of theeukaryotic cDNA expression vector, BMGNeo, using polymerasechain reaction-supported protocols. The transfected cells weretreated with 400 µg/ml G418 for 2 weeks and were furthermaintained in G418-free condition for at least 1 week before theexperiments. The amount of recombinant mouse (rm)-IL-2 secretedwas 895 U/53 106 cells/ml/48 hr. LLC and LLC-IL2 cells weremaintained in Iscove’s medium (IBL, Fujioka, Japan) with 10%fetal cow serum (FCS; IBL). B16-F10, a metastatic subline ofmurine melanoma B16-F10 (Fidleret al., 1975) and the naturalkiller (NK) cell-resistant mastocytoma P815 (Baralet al., 1995)were cultured in RPMI-1640 medium with 10% FCS.

Contract grant sponsor: NIH-NCI; Contract grant numbers: 5P30 CA14395, CA 59351, 39201 and CA 57904; Contract grant sponsors: Ministryof Health and Welfare and Ministry of Education, Science and Culture(Japan).

*Correspondence to: Pharmacology Division, National Cancer CenterResearch Institute, Tsukiji 5-1-1, Chuo-ku, Tokyo 104, Japan. Fax: 81-3-3542-1886. E-mail: nsaijo@gan2.ncc.go.jp

Received 28 February 1997; Revised 25 July 1997

Int. J. Cancer:73,844–849 (1997)

r 1997 Wiley-Liss, Inc.

Publication of the International Union Against CancerPublication de l’Union Internationale Contre le Cancer

AnimalsInbred 5-week-old female C57BL/6 mice were purchased from

Charles River (Atsugi, Japan). The animals were maintained at theNational Cancer Center Research Institute under standard condi-tions, according to institutional guidelines.

Antibodies and recombinant interleukin-2Anti-CD4 (rat IgG2a) and anti-CD8 monoclonal antibodies

(MAbs) (rat IgG2a) were purchased from PharMigen (San Diego,CA). rm IL-2 was purchased from Genzyme (Cambridge, MA).The IL-2 was diluted with saline including 1% of C57BL/6 serumto 10,000 U/ml.

Preparation of tumor cells forin vivo experimentsTumor cells were harvested during exponential growth of the

cell culture forin vivoexperiments. The cells were washed twice inHanks’ balanced solution (IBL) and resuspended at the indicatedconcentration in Hanks’ balanced solution. Cell viability wasgreater than 90% as determined by Trypan blue dye exclusion. Thecells were injected into the pleural cavity, peritoneal cavity and s.c.according to the experimental schedules. Irradiation of the cells forimmunization was performed at 100 Gy, which completely blockedthe proliferation of LLC and LLC-IL2in vitro growth inhibitionassays. In this condition, IL-2 production from irradiated LLC-IL2was maintained for at least 1 weekin vitro and for 5 daysin vivo.

Calculation of the s.c. tumor volumesSubcutaneous tumor volumes were determined using the for-

mula V5 ab2/2, where a is the longest diameter of the tumor and bis the shortest diameter of the tumor.

Prevention of s.c. tumor formation by intrapleural, intraperitonealand s.c. immunization of irradiated LLC-IL2

Irradiated LLC-IL2 cells (53 104/100 µl/mouse) were injectedinto the pleural cavity, the peritoneal cavity or s.c. into the rightflank on days 1, 5 and 9. The control groups were immunized withirradiated LLC-neo on days 1, 5 and 9 and on days 1, 2, 5, 6, 9 and10 with or without 10 U/mouse rmIL-2, respectively. On day 20,2 3 105 cells/100 µl/mouse of LLC were injected s.c. into left flankof the immunized mice. Each group consisted of 10 mice. Thetreated mice were observed every 2 days, and development of thetumor in each group was checked until day 50. Statistical analysiswas performed usingx2 analysis.

Effects of intrapleural, intraperitoneal or s.c. injection ofirradiated LLC-IL2 on the growth of the pre-existing tumors

This experiment was designed to evaluate the therapeuticefficacy of intrapleural, intraperitoneal or s.c. injection of irradiatedLLC-IL2 on pre-existing s.c., pleural and peritoneal LLC tumors.In the first group, 23 105 cells/100 µl/mouse of LLC wereinoculated s.c. in the right flank on day 1. This resulted in thedevelopment of s.c. tumors, and the mice died. In the second group,2 3 106 cells/100 µl/mouse of LLC were inoculated into the pleuralcavity on days 1 and 3. This resulted in the development of lungtumors with a malignant pleural effusion and also in the death of theanimals [mean survival time (MST) 32.86 2.7 days; Fig. 1]. In thethird group, 23 106 cells/100 µl/mouse of LLC were inoculatedinto the peritoneal cavity on days 1 and 3. This resulted inperitoneal tumor formation with the development of tumor ascitesin all of the mice, which all died within 50 days.

For the treatment of these tumors, 53 106 cells/100 µl/mouse ofirradiated LLC-IL2 were injected into the pleural and peritonealcavity and s.c. into the left flank on days 5, 6, 9 and 13. The controlgroups were treated with irradiated LLC-neo on days 5, 9 and 13with or without 450 U of rmIL-2 given on days 5, 6, 9, 10, 13 and14. Ten mice were tested in each group. In s.c. tumor models, thetumor diameter was measured every 2 days, and tumor volume andsurvival in each group were compared. In pleural tumor models andperitoneal tumor models, the mice were observed every 2 days,

with survival of the tumor-bearing mice noted. The survival curvewas plotted using the Kaplan-Maier method, and statistical analysiswas performed using the generalized Wilcoxon method.

Second challenge of LLC or B16-F10 cells in mice that hadpreviously rejected LLC tumor after immunization and in micetreated with irradiated LLC-IL2

At the end of the first experiment involving LLC challenge, themice were divided into 2 groups. The first group comprised 8tumor-free mice after intrapleural immunization, 10 tumor-freemice after intraperitoneal immunization, and 5 tumor-free miceafter s.c. immunization, while the second group comprised 8 micewhose lung tumor had disappeared after intrapleural treatment and9 mice whose tumor ascites had resolved after intraperitonealtreatment. Sixty days after initial inoculation with LLC, 23 105

cells/100 µl/mouse of LLC or B16-F10 were injected s.c. into the

FIGURE 1 – Effect of intrapleural, intraperitoneal and s.c. injectionsof irradiated LLC-IL2 on the survival of pre-existing lung tumors. Ondays 1 and 3, 23 106 naive LLC cells were injected into the peritonealcavity of mice (large solid arrows). For treatment of the tumor ascites,5 3 106 cells/100 µl of irradiated LLC-IL2 were injected into thepleural cavity(a), into the peritoneal cavity(b) or s.c.(c) on days 5, 9and 13 (open arrows) without or with rm IL-2 (narrow arrows).d (a),Untreated group;s (b), irradiated LLC-IL2-treated group;m (c),irradiated LLC-neo-treated group;n (d), irradiated LLC-neo with rmIL-2-treated group.p values:a/a-b, 0.006;a/a-c, 0.54;a/a-d, 0.18;b/a-b, 0.52;b/a-c, 0.76;b/a-d, 0.22;c/a-b, 0.50;c/a-c, 0.47;c/a-d, 0.34;a/b-B/b,,0.005;a/b-c/b, ,0.007.

845GENETIC IMMUNOTHERAPY WITH LLC-IL-2

left flank. The number of mice that developed s.c. tumors wasevaluated every 2 or 3 days until day 90.

In vitro cytotoxicity of PEC and spleen cells in the early stages ofintraperitoneal immunization

This experiment was designed to determine the presence ofeffector cells in early stages of immunization in peritoneal immuni-zation models; 13 106 cells/100 µl/mouse of irradiated LLC-IL2were injected on days 1 and 5. The control groups were injectedwith 1 3 106 cells/100 µl/mouse of irradiated LLC-neo on days 1and 5. On day 8, the peritoneal exudate cell (PEC) and the spleenswere taken from the immunized mice forin vitro cytotoxicityassays. Mononuclear cells were separated by Lymphoprep (density1.077 g/m; Nycomed, Oslo Norway) centrifugation at 400g for 20min in the TOMY TS-1 rotor. The mononuclear cells from thespleen were mixed with 2,000 of51Cr-labeled target cells atdifferent effector/target ratios in 96-well microtiter plates (Falcon3072, Becton Dickinson, Lincoln Park, NJ) in a total volume of 200µl. Mononuclear cells from PEC were cultured in Falcon 3003culture plates for 1 hr. The non-adherent cells were separated fromthe adherent ones, and viable non-adherent cells were mixed with2,00051Cr-labeled target cells at effector/target ratios of 12.5, 25and 50 in 96-well microtiter plates (total volume 200 µl). Inhibitionof the reaction with MAbs was performed by adding 10 µg/ml ofanti-CD4 or anti-CD8 MAbs. The plates were further incubated for6 hr at 37°C. Cultures were terminated by centrifugation at 300g for10 min at 4°C, and 100 µl of the supernatants were assayed in agamma counter. Percent specific lysis was calculated as follows:

experimental [51Cr] release2 spontaneous [51Cr] release

maximal [51Cr] release2 spontaneous [51Cr] release3 100.

Maximal release was determined by solubilization of target cellsin 0.01% NaOH. Spontaneous release was determined by incubat-ing the target cells in RPMI-1640 medium without effector cells.The spontaneous release was less than 20% of total release.Statistical significance was analyzed by using Student’st test.

In vitro cytotoxicity of LLC-specific killer cells in the late stagesof intraperitoneal immunization

Mice were immunized by intraperitoneal injection of 53 106

irradiated LLC-IL2 on days 1, 5, 9 and 13. On day 20, the spleencells were taken from the mice forin vitro cytotoxicity assays. Thespleen cells were treated with 0.8% of NH4Cl in 17 mM Tris buffer(pH 7.65) to lyse the red cells and were co-cultured with irradiatedLLC-IL2 for 10 days for stimulationex vivo. In this culturemedium, enough IL-2 for maintaining the splenocyte was producedby irradiated LLC-IL2 cells. Viable lymphocytes were separated byLymphoprep centrifugation and admixed with 2,00051Cr-labeledtarget cells at an effector/target ratio of 25 in 96-well microtiterplates (total volume 200 µl). The plates were further incubated for 6hr at 37°C. The cultures were terminated by centrifugation at 300gfor 10 min at 4°C, and 100 µl of the supernatant were assayed in agamma counter. Inhibition of the test with MAbs was performed asdescribed above. Statistical significance was analyzed using Stu-dent’st test.

RESULTS

Prevention of s.c. tumor formation by intrapleural, intraperitonealand s.c. immunization with irradiated LLC-IL2

This experiment was conducted to determine the effectiveness ofintrapleural, intraperitoneal and s.c. injection of irradiated LLC-IL2 as a tumor vaccine. Immunization by intrapleural or intraperi-toneal injection of 53 104 irradiated LLC-IL2 on days 1, 5 and 9induced anti-tumor immunityin vivo.As shown in Table I, 8 of 10mice immunized by intrapleural injection of irradiated LLC-IL2rejected subsequent s.c. challenge with parental LLC. Tumorgrowth in 2 of the mice was markedly delayed. All 10 mice

immunized intraperitoneally with LLC-IL2 rejected subsequentchallenge with parental LLC. Three of 10 mice given an intrapleu-ral injection of irradiated LLC-neo and 3 of 10 mice givenintrapleural injection with irradiated LLC-neo plus rmIL-2 rejectedsubsequent LLC challenge. Four of 10 mice given an intraperito-neal injection of irradiated LLC-neo and 3 of 10 mice given anintraperitoneal injection of irradiated LLC with rmIL-2 alsorejected subsequent LLC challenge. The s.c. injection of irradiatedLLC-IL2 also induced tumor-specific immunity, with 5 of 10 micerejecting the following LLC challenge. Two of 10 mice given s.c.injection of irradiated LLC-neo and 1 of 10 mice given irradiatedLLC-neo plus rmIL-2 rejected subsequent LLC challenge.

Moreover, the mice that rejected the first challenge with LLCalso rejected the second challenge of LLC 60 days after the initialinoculation but could not reject s.c. challenge ofB16-F10 on day 60.

Effects of intrapleural, intraperitoneal and s.c. injection ofirradiated LLC-IL2 on the growth of pre-existing LLC s.c. tumors

This experiment was conducted to determine the effectiveness ofintrapleural, intraperitoneal and s.c. injection of irradiated LLC-IL2 on pre-existing s.c. LLC tumors. None of the sites injected withirradiated LLC-IL2 had any therapeutic effect on the growth ofpre-existing LLC tumors. There was no difference in the rate oftumor growth between the different groups (data not shown).

Effects of intrapleural, intraperitoneal and s.c. injection ofirradiated LLC-IL2 on the survival of mice bearing pre-existinglung tumors

Inoculation of 23 106 LLC cells into the pleural cavity on days1 and 3 caused the development of lung tumors and malignantpleural effusions in all mice, with subsequent death (MST 32.86 2.7days). Intrapleural injection of irradiated LLC-IL2 on days 5, 9 and13 prevented the growth of lung tumors and extended the survivalof mice. Eight of 10 mice treated with irradiated LLC-IL2 becametumor free and survived throughout the observation period (Fig.1a). In contrast, intrapleural injection of irradiated LLC-neo notproducing IL-2 had no effect, and all these mice died (MST35.16 2.6 days). A combination of irradiated LLC-neo and rmIL-2resulted in 3 of 10 mice becoming tumor free during the observa-tion period. Treatment by intraperitoneal (Fig. 1b) and s.c. (Fig. 1c)injection with irradiated LLC-IL2 produced no effects on thegrowth of lung tumors or on the survival of the mice. The MST ofthe mice treated with intraperitoneal injection of irradiated LLC-

TABLE I – PREVENTION OF s.c. TUMOR FORMATION BY IMMUNIZATIONWITH IRRADIATED LLC-IL2 1

Immunization

Tumor rejection

Firstchallenge LLC

Second challenge

LLC B16-F10

Intrapleural immunizationIrradiated LLC-IL2 8/102 4/4 0/4Irradiated LLC-neo 3/103 — —Irradiated LLC-neo1 rm IL-2 3/10 — —

Intraperitoneal immunizationIrradiated LLC-IL2 10/104 5/5 0/5Irradiated LLC-neo 4/105 — —Irradiated LLC-neo1 rm IL-2 3/106 — —

Subcutaneous immunizationIrradiated LLC-IL2 5/107 3/3 0/2Irradiated LLC-neo 2/108 — —Irradiated LLC-neo1 rm IL-2 1/109 — —

1On days 1, 5 and 9, 53 104 irradiated LLC-IL2 cells or irradiatedLLC-neo cells with or without rm IL-2 were injected into the pleuralcavity, peritoneal cavity or s.c. On day 20, 23 105 LLC wereinoculated into the right flank. Statistical analysis was performed byx2

analysis;p values: 2–3, 0.03; 4–5, 0.003; 5–6, 0.63; 7–8, 0.16; 8–9,0.53; 2–4, 0.14; 2–7, 0.16; 4–7, 0.01.

846 HEIKE ET AL.

IL2 was 35.86 2.9 days and that of the mice treated with s.c.injection was 35.96 2.5 days.

All the mice (n5 4) that were tumor free after intrapleuralinjection of irradiated LLC-IL2 rejected subsequent s.c. re-challenge with LLC on day 60 but could not reject s.c. challengewith B16-F10 on day 60 (data not shown).

Effects of intrapleural, intraperitoneal and s.c. injection ofirradiated LLC-IL2 on the survival of mice bearing pre-existingperitoneal tumors

The injection of 23 106 LLC cells on days 1 and 3 caused thedevelopment of peritoneal tumors with tumor ascites in all mice,resulting in subsequent death (MST 31.66 2.7 days; Fig. 2).Intraperitoneal injection of irradiated LLC-IL2 on days 5, 9 and 13prevented the growth of peritoneal tumors and extended the

survival of mice (Fig. 2b). All the mice (n5 10) treated withintraperitoneal injection of irradiated LLC-IL2 became tumor-freeand survived during the observation period. In contrast, intraperito-neal injection of irradiated LLC-neo not producing IL-2 had noeffect, and all these mice treated died within 32 days after tumorinjection. A combination of irradiated LLC-neo and rmIL-2 had apartial effect on the development of tumor ascites, with 3 of 10mice becoming tumor free and surviving during the observationperiod.

Treatment by intrapleural and s.c. injection with irradiatedLLC-IL2 provided no therapeutic benefits on either the growth ofthe peritoneal tumors or the survival of mice. The survival of micetreated with intrapleural and s.c. injections was almost the same asthat of the control group (MST 32.06 2.4 days and 31.06 2.3days, respectively).

Mice (n 5 5) in which the peritoneal tumor resolved afterintraperitoneal injection of irradiated LLC-IL2 rejected subsequents.c. re-challenge with LLC 60 days after the initial inoculation butcould not reject s.c. challenge with B16-F10 on day 60 (data notshown).

In vitro analysis of effector cells in the late stage of immunizationin intraperitoneal immunization models

Spleen cells from the mice immunized intraperitoneally withirradiated LLC-IL2 showed LLC-specific tumor cytotoxicity afterin vitro re-stimulation (Table II). Cytotoxicity of the cells againstanother syngeneic tumor B16-F10 and P815 was negligible.Additionally, no cytotoxicity of the spleen cells from control miceagainst LLC was detectable. These results indicate that there is aninduction of specific cytotoxicity in immunized mice with irradi-ated LLC-IL2. Anti CD8 MAb blocked the cytotoxicity of spleencells, while anti CD4 MAb had no effect. These data suggest thattumor-specific immunity in immunized mice was mediated byCD81 T lymphocytes.

In vitro cytotoxicity of PEC and spleen cells in the early stageof immunization in intraperitoneal immunization models

To test thein vitro cytotoxicity of the PEC and spleen cells ofirradiated LLC-IL2-treated mice in the early stages of immuniza-tion (until day 8 after initial immunization), cytotoxicity againstLLC, B16-F10 and P815 cells was evaluated using the51Cr-releaseassay. As shown in Figure 3, non-adherent PEC from miceimmunized with intraperitoneal injection of irradiated LLC-IL2showed enhanced cytotoxicity against LLC, B16-F10 and P815cells when compared either with non-immunized mice or with miceimmunized with irradiated LLC-neo.

In contrast, the spleen cells from the mice immunized withintraperitoneal injection of irradiated LLC-neo did not showenhanced cytotoxicity against any of the target cells tested.

DISCUSSION

Tumor vaccine therapy using cytokine gene-modified cancercells has been studied as a novel approach for anti-cancerimmunotherapy in many laboratories. Basic experiments revealedthat the s.c. injection of cancer cells modified by cytokine genes,

FIGURE 2 – Effect of intrapleural, intraperitoneal and s.c. injectionsof LLC-IL2 on the survival of pre-existing peritoneal tumors. On days1 and 3, 23 106 naive LLC cells were injected into the peritonealcavities (large solid arrows). For treatment of the tumor ascites, 53 106

cells/100 µl of irradiated LLC-IL2 were injected into the pleural cavity(a), into the peritoneal cavity(b) and s.c.(c) on days 5, 9 and 13 (openarrows) without or with rm-IL-2 (narrow arrows).d (a), Untreatedgroup;s (b), irradiated LLC-IL2 treated group;m (c), irradiated LLC-neo-treated group;n (d), irradiated LLC-neo with rm IL-2-treatedgroup.p values:a/a-b, 0.94;a/a-c, 0.62;a/a-d, 0.27;b/a-b, ,0.001;b/a-c, 0.22;b/a-d, 0.06;c/a-b, 0.91;c/a-c, 0.97;c/a-d, 0.71;a/b-B/b,,0.0001;b/b-c/b, ,0.0001.

TABLE II – IN VITROEVALUATION OF TUMOR-SPECIFIC KILLER CELLS1

Effector cells% Cytotoxicity in target cells

LLC B16-F10 P815

Re-stimulated spleen cellsWithout MAb 21.56 0.8 8.56 3.0 5.16 2.5With anti-CD4 MAb 17.76 2.8 6.96 2.5 5.06 2.6With anti-CD8 MAb 7.56 3.1 4.96 1.3 4.26 3.8

1Spleen cells from mice immunized with intraperitoneal injection ofirradiated LLC-IL2 were restimulatedin vitro as described in Materialand Methods. The cytotoxicity of the effector cells was evaluated by a6 hr 51Cr-release assay. Effector target ratio: 25.

847GENETIC IMMUNOTHERAPY WITH LLC-IL-2

such as IL-2 (Belldegrunet al.,1993; Gansbacheret al.,1990; Oheet al.,1993; Rosenthalet al.,1994), IL-4 (Oheet al.,1993; Ohiraetal., 1994b; Golumbeket al., 1991), IL-6 (Porgadoret al., 1992),IL-7 (McBride et al., 1992), GM-CSF (Dranoffet al., 1993),IFN-a, b, g (Belldegrun et al., 1993; Rosenthalet al., 1994;Watanabeet al.,1989; Porgadoret al.,1993) and TNF (Ohiraet al.,1994b; Asheret al.,1991), induced tumor-specific immunity.

Based on these results, clinical trials of s.c. cancer vaccinesusing cytokine gene-modified cancer cell were planned in somelaboratories. To date, although some of the reports have suggestedthat these therapies induce immunological responses such asdelayed type hypersensitivity and tissue infiltration with T cells inpatients, there is no evidence that they are clinically useful.

In this study, we have evaluated the effectiveness of intrapleural,intraperitoneal and s.c. injections of irradiated LLC-IL2 in prevent-ing the growth of parental LLC cells and the therapeutic efficacyagainst pre-existing tumors in mice.

Injection with irradiated LLC-IL2 by all routes,i.e., intrapleural,intraperitoneal and s.c., protected mice against subsequent re-challenge with LLC. Although the immunity was also partiallyinduced by irradiated LLC-neo, IL-2 gene transfer enhancedefficiency markedly. We checked that IL-2 gene transfection intoLLC and B16-F10 cells did not enhance expression of IL-2R orHLA-class I molecules. The specificity of immunity was shown bythe fact that animals immunized or treated with irradiated LLC-IL2did not reject the s.c. challenge with another syngeneic tumor cellline, B16-F10.In vitro analysis of effector cells in intraperitonealimmunization models suggested that CD81 T lymphocytes wereresponsible for this immunity. CD41 T cells did not appear tocontribute to immunity in this model.

Our results suggest that immunization into the peritoneal orpleural cavity was more efficient in inducing tumor-specificimmunity than s.c. immunization. As shown Table I, all miceimmunized intraperitoneally with 13 105 irradiated LLC-IL2 cellscompletely rejected s.c. challenge with LLC. After intrapleuralimmunization, 8 of 10 mice rejected s.c. challenge with LLC, with

only 5 of the 10 mice immunized s.c. rejecting re-challenge. Inother experiments, we evaluated the correlation between the cellnumbers used for vaccination and the efficacy of induction oftumor-specific immunity using several immunization schedules.These showed that intraperitoneal immunization was the mostefficient in inducing tumor-specific immunity. (Immunization on 3occasions with 13 104 irradiated LLC-IL2 cells induced completeimmunity against the LLC tumors.) Higher cell numbers (at least1 3 106 irradiated LLC-IL2 cells/mouse) were necessary for theinduction of complete tumor-specific immunity using s.c. immuni-zation. These data suggest that the route of immunization withgene-modified cells is important in obtaining effective anti-tumorimmunity. Similar results have been obtained by other investigatorstesting for immunization in viral antigen models (Kundiget al.,1995).

Intrapleural (Fig. 1) and intraperitoneal (Fig. 2) injection ofirradiated LLC-IL2 showed therapeutic efficacy against pre-existing LLC lung tumors and peritoneal tumors, respectively, withan improvement in survival. However, injection with irradiatedLLC-IL2 did not exhibit any effect if the tumors were in sitesdifferent from the site of injection. Thus, intraperitoneal injectiondid not affect pre-existing s.c. or lung tumors, andvice versa.Theonly exception was the s.c. injection of irradiated LLC-IL2, whichshowed no effect not only on the growth of pre-existing lung andperitoneal tumors, but also on s.c. tumors. It did also not affectsurvival. These data therefore suggest that therapeutic efficacy oftumor vaccine therapy using IL-2 gene-modified cancer cellsagainst pre-existing tumors is limited. It has previously beenreported (Oheet al., 1993) that the s.c. injection of LLC-IL2suppresses the growth of pre-existing s.c. tumors. In these experi-ments, treatment was started 5 days after tumor injection. This mayaccount for the contradictory findings.

In vitro assays demonstrated that the non-specific killer activityof nonadherent PEC cells were activated in an early stage ofimmunization. The candidate killer cells are IL-2-activated NKcells, T cells, monocyte/macrophages, LAK cells and neutrophils.In preliminary experiments, we tested therapeutic efficacy byintraperitoneal injection of irradiated LLC-IL2 on pre-existingperitoneal tumors in SCID mice and SCID/nude/beige mice. Thetherapeutic efficacy in the SCID mice was not inferior to thatobtained in normal mice, but that in SCID/nude/beige mice wasinferior (data not shown). These results suggest that IL-2-activatedNK cells and/or LAK cells play a major role in mediating thetherapeutic effect against pre-existing tumors. We also tried tocollect the effector cells in the pleural cavity after LLC-IL2injection but failed because of the difficulty in getting enough pureeffector cells to characterize them.

Our data indicate that mice recovering from lung or peritonealtumors after treatment with irradiated LLC-IL2 exhibit tumor-specific immunity and reject subsequent re-challenge with parentalcells. These results suggest that treatment of lung or peritonealtumors with IL-2 gene-modified cancer cells may cure them byenhancing tumor non-specific immunity and inducing tumor-specific immunity and thereby preventing the recurrence or distantspread of the tumor. Since the combination of irradiated LLC andrmIL-2 showed inferior effects on the induction of tumor non-specific and specific immunity, treatment with IL-2 gene-modifiedcancers appears to be more interesting for effective immuno-therapy.

ACKNOWLEDGEMENTS

This work was supported in part by Grants-in-Aid for CancerResearch from the Ministry of Health and Welfare, Second TermComprehensive Ten-Year Strategy for Cancer Control from theMinistry of Education, Science and Culture, Japan and from theBristol-Myers Squibb Foundation. These studies were also sup-ported in part by NIH-NCI5P30 CA 14395, NIH-NCI CA 59351,NIH-NCI 39201 and NIH-NCI CA 57904 to ERP.

FIGURE 3 – In vitro cytotoxicity of PEC and spleen cells fromimmunized mice in the early stage of immunization. Cytotoxicity ofnon-adherent PEC cells and spleen cells of mice immunized againstLLC, B16-F10 and P815 cells were evaluated by51Cr-release assaysinvitro. Effector target ratios were 12.5, 25 and 50.(a) PEC cells frommice immunized with irradiated LLC-neo.(b) Spleen cells from miceimmunized with irradiated LLC-neo.(c) PEC cells from mice immu-nized with irradiated LLC-IL2.(d) Spleen cells from mice immunizedwith irradiated LLC-IL2. Target cells:s, LLC; d, B16-F10;n, P815.Bars represent SDs from triplicate measurements.

848 HEIKE ET AL.

REFERENCES

ASHER, A.L., MULE, J.J., KASID, A., RESTIFO, N.P., SALO, J.C., REICHERT,C.M., JAFFE, G., FENDLY, B., KRIEGLER, M. and ROSENBERG, S.A., Murinetumor cells transduced with the gene for tumor necrosis factor-alpha.Evidence for paracrine immune effects of tumor necrosis factor againsttumors.J. Immunol.,146,3227–3234 (1991).BARAL, E., NAGY, E. and BERCZI, I., Enhancement of natural killer cellmediated cytotoxicity by tamoxifen.Cancer,75,591–599 (1995).BELLDEGRUN, A., TSO, C.L., SAKATA , T., DUCKETT, T., BRUNDA, M.J.,BARSKY, S.H., CHAI, J., KABOO, R., LAVEY, R.S., MCBRIDE, W.H. andDEKERNION, J.B., Human renal carcinoma line transfected with interleu-kin-2 and/or interferon alpha gene(s): implications for live cancer vaccines.J. nat. Cancer Inst.,85,207–216 (1993).BERNSTEIN, Z.P. and 13OTHERS, Interleukin-2 with ex vivo activated killercells: therapy of advanced non-small-cell lung cancer.J. Immunother.,10,383–387 (1991).CLAMON, G., HERNDON, J., PERRY, M.C., OZER, H., KREISMAN, H., MAHER,T., ELLERTON, J. and GREEN, M.R., Interleukin-2 activity in patients withextensive small-cell lung cancer: a phase II trial of cancer and leukemiagroup B.J. Nat. Cancer Inst.,85,316–320 (1993).DILLMAN , R.O., CHURCH, C., OLDHAM , R.K., WEST, W.H., SCHWARTZBERG,L. and BIRCH, R., Inpatient continuous-infusion interleukin-2 in 788patients with cancer. The National Biotherapy Study Group experience.Cancer,71,2358–2370 (1993).DRANOFF, G., JAFFEE, E., LAZENBY, A., GOLUMBEK, P., LEVITSKY, H., BROSE,K., JACKSON, V., HAMADA , H., PARDOLL, D. and MULLIGAN , R.C., Vaccina-tion with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, andlong-lasting anti-tumor immunity.Proc. nat. Acad. Sci. (Wash.),90,3539–3543 (1993).GANSBACHER, B., ZIER, K., DANIELS, B., CRONIN, K., BANNERJI, R. andGILBOA, E., Interleukin 2 gene transfer into tumor cells abrogates tumorige-nicity and induces protective immunity.J. exp. Med.,172, 1217–1224(1990).GOLUMBEK, P.T., LAZENBY, A.J., LEVITSKY, H.I., KARASUYAMA , H., BAKER,M. and PARDOLL, D.M., Treatment of established renal cancer by tumorcells engineered to secrete interleukin-4.Science,254,713–716 (1991).FIDLER, I.J., Biological behavior of malignant melanoma cells correlated totheir survivalin vivo. Cancer Res.,35,218–224 (1975).JETT, J.R., MAKSYMIUK , A.W., SU, J.Q., MAILLIARD , J.A., KROOK, J.E.,TSCHIETTER, L.K., KARDINAL , C.G., TWITO, D.I., LEVITT, R. and GERSTNER,J.B., Phase III trial of recombinant interferon gamma in complete respond-ers with small-cell lung cancer.J. clin. Oncol.,12,2321–2326 (1994).KUNDIG, T.M., BACHMANN, M.F., DIPAOLO, C., SIMARD, J.J.L., BATTEGAY,M., LOTHER, H., GESSNER, A., KUHLCKLE, K., OHASHI, P.S., HENGARTNER, H.and ZINKERNAGEL, R.M., Fibroblasts as efficient antigen-presenting cells inlymphoid organs.Science,268,1343–1347 (1995).MCBRIDE, W.H., THACKER, J.D., COMORA, S., ECONOMOU, J.S., KELLEY, D.,HOGGE, D., DUBINETT, S.M. and DOUGHERTY, G.J., Genetic modification ofa murine fibrosarcoma to produce interleukin 7 stimulates host cellinfiltration and tumor immunity.Cancer Res.,52,3931–3937 (1992).OHE, Y., PODACK, E.R., OLSEN, K.J., MIYAHARA , Y., OHIRA, T., MIURA, K.,NISHIO, K. and SAIJO, J., Combination effect of vaccination with IL2 andIL4 cDNA transfected cells on the induction of a therapeutic immuneresponse against Lewis lung carcinoma cells.Int. J. Cancer,53, 432–437(1993).

OHIRA, T., OHE, Y., HEIKE, Y., PODACK, E.R., OLSEN, K.J., NISHIO, K.,NISHIO, M., MIYAHARA , Y., OGASAWARA, H., ARIOKA, H., KATO, H. andSAIJO, N., Gene therapy for for Lewis lung carcinoma with tumor necrosisfactor and interleukin 2 cDNAs co-transfected subline.Gene Ther.,1,269–275 (1994a).

OHIRA, T., OHE, Y., HEIKE, Y., PODACK, E.R., OLSEN, K.J., NISHIO, K.,NISHIO, M., MIYAHARA , Y., OGASAWARA, H., ARIOKA, H., KUNIKANE, H.,FUKUDA, M., KATO, H. and SAIJO, N., In vitro andin vivogrowth of B16F10melanoma cells transfected with interleukin-4 cDNA and gene therapy withthe transfectant.J. Cancer Res. clin. Oncol.,120,631–635 (1994b).

PORGADOR, A., BANNERJI, R., WATANABE, Y., FELDMAN, M., GILBOA, E. andEISENBACH, L., Antimetastatic vaccination of tumor-bearing mice with twotypes of IFN-gamma gene-inserted tumor cells.J. Immunol.,150, 1458–1470 (1993).

PORGADOR, A., TZEHOVAL, E., KATZ, A., VADAI , E., REVEL, M., FELDMAN,M. and EISENBACH, M., Interleukin 6 gene transfection into Lewis lungcarcinoma tumor cells suppresses the malignant phenotype and confersimmunotherapeutic competence against parental metastatic cells.CancerRes.,52,3679–3686 (1992).

ROSENTHAL, F.M., CRONIN, K., BANNERJI, R., GOLDE, D.W. and GANS-BACHER, B., Augmentation of antitumor immunity by tumor cells trans-duced with a retroviral vector carrying the interleukin-2 and interferon-gamma cDNAs.Blood,83,1289–1298 (1994).

SCHILLER, J.H., MORGAN-IHRIG, C. and LEVITT, M.L., Concomitant adminis-tration of interleukin-2 plus tumor necrosis factor in advanced non-smallcell lung cancer.Amer. J. clin. Oncol.,18,1847–1851 (1995).

SCUDELETTI, M., FILACI , G., IMRO, M.A., MOTTA, G., DI GAETANO, M.,PIERRI, I., TONGIANI, S., INDIVERI, F. and PUPPO, F., Immunotherapy withintralesional and systemic interleukin-2 of patients with non-small-cell lungcancer.Cancer Immunol. Immunother.,37,119–124 (1993).

SUGIURA, S. and STOCK, C.C., Studies in a tumor spectrum. III. The effect ofphosphoramides on the growth of a variety of mouse and rat tumors.CancerRes.,15,38–51 (1955).

WATANABE, Y. and IWA, T., Clinical value of immunotherapy for lungcancer by the streptococcal preparation OK-432.Cancer, 53, 248–253(1984).

WATANABE, Y., KURIBAYASHI, K., MIYATAKE , S., NISHIHARA, K., NAKAYAMA ,E., TANIYAMA , T. and SAKATA , T., Exogenous expression of mouseinterferon gamma cDNA in mouse neuroblastoma C1300 cells results inreduced tumorigenicity by augmented anti-tumor immunity.Proc. nat.Acad. Sci. (Wash.),86,9456–9460 (1989).

YAMAMURA , Y., OGURA, T., SAKATANI , M., HIRAO, F., KISHIMOTO, S.,FUKUOKA, M., TAKADA , M., KAWAHARA , M., FURUSE, K., KUWAHARA, O.,IKEGAMI, H. and OGAWA, N., Randomized controlled study of adjuvantimmunotherapy with Nocardia rubra cell wall skeleton for inoperable lungcancer.Cancer Res.,43,5575–5579 (1983).

YAMAMURA , Y., SAKATANI , M., OGURA, T. and AZUMA, T., Adjuvantimmunotherapy of lung cancer with BCG cell wall skeleton (BCG-CWS).Cancer,43,1314–1319 (1979).

YANG, S.C., GRIMM, E.A., PARKINSON, D.R., CARINHAS, J., FRY, K.D.,MENDIGUREN-RODRIGUEZ, A., LICCIARDELLO, J., OWEN-SCHAUB, L.B., HONG,W.K. and ROTH, J.A., Clinical and immunomodulatory effects of combina-tion immunotherapy with low-dose interleukin 2 and tumor necrosis factoralpha in patients with advanced non-small cell lung cancer: a phase I trial.Cancer Res.,51,3669–3676 (1991).

849GENETIC IMMUNOTHERAPY WITH LLC-IL-2