Combination Suicide and Cytokine Gene Therapy for Hepatic Metastases of Colon Carcinoma: Sustained Antitumor Immunity Prolongs

Animal Survival
Shu-Hsia Chen, Ken-ichiro Kosai, Bisong Xu, et al. Cancer Res 1996;56:3758-3762.

Updated Version

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/16/3758

Citing Articles

This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/56/16/3758#related-urls

E-mail alerts Reprints and Subscriptions Permissions

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at pubs@aacr.org. To request permission to re-use all or part of this article, contact the AACR Publications Department at permissions@aacr.org.

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research

(CANCER RESF.ARCH 56. 3758-3762,

Augusl 15. 1996]

Combination Suicide and Cytokine Gene Therapy for Hepatic Mtastasesf Colon o Carcinoma: Sustained Antitumor Immunity Prolongs Animal Survival
Shu-Hsia Chen, Ken-ichiro Savio L. C. Woo1'2 Rosai. Bisong Xu, Khiem Pham-Nguyen, Charles Contant, Milton J. Finegold, and
C.C..

Howard Hughes Medical limitale K.P-N.. S. L. C. W.and Departments fii\ur College of Medicine. Houston. Texas 77030

tif Cell Biology S-H.C.. B. X.. S. L C. W.]. Pathology K-i.K.. M. J. F.I and Neurosurgery

ABSTRACT The effectiveness of combination therapy using a suicide gene and cytokine genes for the treatment of metastatic colon carcinoma in the mouse liver was investigated. Pre-established hepatic tumors treated with a recombinant adenoviral vector containing the herpes simplex virus thymidine kinase gene (tk) exhibited substantial regression, although all treated animals suffered from subsequent relapses. Although cotreatment with a mouse interleukin 2 (mIL-2)-containing adenoviral vector induced an effective antitumor immune response, the immunity waned with time, and the treated animals eventually succumbed to hepatic tumor relapse or distant mtastases.In this study, mouse granulocyte macrophage colonystimulating factor (mGM-CSF) gene was tested for its ability to further enhance and prolong the antitumoral cellular immunity. A fraction of the animals treated with tk + mIL-2 + mGM-CSF developed long-term antitumor immunity and survived for more than 4 months without recur rence. This long-term antitumor immunity could be enhanced further by subsequent "vaccination" with mIL-2-expressing parental tumor cells. The results indicate that local expression of GM-CSF in the hepatic tumors and prolonged mil -2 expression are necessary to generate per sistent antitumor immunity that is essential for the prevention of tumor recurrence and long-term animal survival.

cell-derived peptides by APCs, which then present the tumor antigens to the CD4+ T lymphocytes that in turn activate tumor-specific CDS1 cytolytic T cells. In the presence of elevated concentrations of IL-2 secreted locally by tumor cells, tumor rejection is achieved through enhanced antigen-specific T-cell response (4). Improvement of anti gen processing and presentation to tumor-specific T-cell precursors complements IL-2 action and may further enhance the induction of antitumor immune response. GM-CSF is the best-characterized cyto kine that promotes the activation and maturation of specialized APCs (5). Dranoff et al. (6) had demonstrated previously that GM-CSF was the most potent cytokine tested in "cancer vaccine" models utilizing irradiated tumor cells engineered to secrete cytokines in the induction of protective antitumor immunity in vivo. Using appropriate adenoviral vector systems, various doses of cytokine genes can be delivered directly to the tumor cells in vivo for in situ expression and secretion. This permitted us to explore the therapeutic effect of localized expression of GM-CSF, in combination with the tk suicide gene therapy in the treatment of hepatic mtastases of colon cancer. The results indicated that GM-CSF stimulated a long-lasting antitumor response and, in conjunction with subsequent vaccination with irradiated tumor cells transduced with cytokines, achieved long-term survival of tumor-bearing animals.

INTRODUCTION Metastatic colon carcinoma is the second leading cause of death from malignancy in the United States. Eighty c/cof the patients who die of colon cancer have mtastasesin the liver (1). Once hepatic mtastasesoccur, surgery and chemotherapy are the only currently available treatment modalities (2). Thus, the development of new modalities for treatment of mtastasesis important for patients with metastatic colon cancer. We have demonstrated previously in a murine model of hepatic mtastasesof colon cancer that the cytokine gene L-2} acted synergistically with the suicide gene herpes simplex thymidine kinase (tk) to induce a systemic antitumor immunity that resulted in regression of local tumor and protection against distant site challenge of parental tumor cells. The antitumor immunity was attributed to IL-2-mediated activation and proliferation of CDS ^ CTLs (3). However, the antitu mor immunity waned with time, resulting in relapse of tumors in the liver and distant sites. To achieve long-term protection against recur rences and mtastases,it is imperative for this protective immunity to be maintained. Therefore, identification of other cytokines that can enhance and prolong antitumor immunity in combination with IL-2 is critically important for this therapeutic approach. Tumor cell killing by tk/GCV results in the uptake of tumor
Received 1/12/96; accepted ft/18/96. The costs of publication of this article were defrayed in part hy the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 S. L. C. W. is an investigator of the Howard Hughes Medical Institute. 2 To whom requests for reprints should be addressed, at Department of Cell Biology. Baylor College of Medicine. One Baylor Plaza. Houslon. TX 77030. ' The abbreviations used are: IL. interleukin: mIL. mouse IL; APC. antigen-presenting cell; GM-CSF. granulocyie-macrophage colony-stimulating factor; mGM-CSF. mouse GM-CSF; RSV. Rous sarcoma virus; pfu. plaque-forming unit; tk. thymidine kinase: GCV. ganciclovir: ADV. adenovirus.

MATERIALS

AND METHODS

Construction of Recombinant Adenoviral Vectors. Construction of a replication-defective adenoviral vector containing the tk gene under the transcriptional control of the RSV long-terminal repeat (ADV/tk) has been re ported (7). A replication-defective adenoviral vector containing the mIL-2 cDNA under the transcriptional control of the RSV long-terminal repeat promoter (ADV/mIL2) was constructed (3) and plaque purified. A partiallength mGM-CSF cDNA encoding the mature peptide was obtained from R & D Ltd. A I05-bp leader sequence was synthesized with overlapping oligonucleotide fragments that covered the leader sequence, the initiation codon. and Kozak consensus sequence for ribosome binding and was utilized to construct a full-length mGM-CSF cDNA. To ensure correct nucleotide sequence, the entire cDNA was subsequently sequenced. Construction of ADV/mGM-CSF was as described for ADV/mIL-2. The control adenovirus DL-312 with El region deletion was obtained from Dr. Tom Shenk of Princeton University. The viral constructs are summarized in Table I. The viral titer (pfus/ml) was determined by plaque assay in 293 cells. Establishment and Treatment of Hepatic Metastasis Model of Colon Carcinoma. Metastatic colon carcinoma was induced in the liver by intrahepatic implantation of CC36 cells, a chemically induced colon carcinoma line derived from BALB/c mouse (8). Cells (I X 10s) were injected at the tip of the left lateral liver lobe of syngeneic mice. At day 7, various titers of recombinant adenoviral vectors were injected intratumorally in 70 (til of 10 mM Tris-HCl (pH 7.4)/l mM MgCl,/lO% (v/v) glycerol/Polybrene (20 jug/ml). Twelve h after viral injection, the animals were treated i.p. with GCV at K) mg/kg twice daily for 5 consecutive days. Histopathological and Immunocytochemical Analyses of Hepatic Tu mors. Fourteen days after various gene therapy treatments, the animals were sacrificed, and computerized morphometric analysis of the largest crosssectional areas of the relapsed hepatic tumors was performed as described previously (3). For immunocytochemical staining, livers from euthanized animals of various treatment groups were collected at day 7 and cut in the

3758

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research

SUICIDI:

AND CYTOKINK

GENE THERAPY

FOR HEPATIC

METASTASES

Table 1 Summary af variable adenm-iral cim.\triu'tx

vector treatment groups were challenged with tumorigenic

doses (1.5 x IO5)

of parental CC36 tumor cells. The presence of s.c. tumors in animals after andinsertion various gene therapy treatment was observed for 2-3 weeks. The results were ConstructADV/ikADV/mIL-2ADV/mGM-CSFDL-312Deletion siteEIAEIAEIAEIAPromoterRSVRSVRSVcDNAHSV-tkmIL-2mGM-CSFPolyadenylateHSV-tkBGH"BGH analyzed statistically by logistic regression (9).

" BGH. bovine growth hormone.

middle at the site of the original tumor inoculation.

One-half

of the tissue

was fixed in 10% buffered formalin and stained with hematoxylin and eosin for histopathological analysis. The corresponding half was embedded in Tissue Tek OCT embedding medium (Miles. Inc.. Elkhurt. IN) and frozen immediately in 2-methvlhutane in liquid nitrogen for immunocytochemical analyses. The fluorescein-antifluorescein system was used to identify in filtrating inflammatory cells in the liver. Fluorescein-conjugated primary monoclonal antibodies used in the assay were the following: rat antimouse CD4 (L3T4: Life Technologies. Inc.. Grand Island. NY), rat antimouse CD8a (ly-2: Life Technologies. Inc.). and rat antimouse macrophage (F4/ 80; Har-an Bioproducts for Science. Inc., Indianapolis, IN). After reaction with primary antibodies, the sections were incubated with peroxidaseconjugated rabbit anti-FITC (DAKO. Carpinter-a. CA). The slides were then incubated in chromogen solution (3 ml 3,3'-diaminobenzidine. 10 ml PBS. 50 fil We NiClj. 1 /I307r H2O2) and then counterstained with Nuclear Fast Red for observation. Distant Site Challenge in Treated Animals with Parental Tumor Cells. At different time points after primary tumor inoculation, animals in different DL-312
TK+GM TK + IL-2 TK+IL-2+GM Treatment Groups Fig. I. Viable tumor si/e in animals after various gene therapy treatments. Maximal cross-sectional areas of the tumors were measured by compuleri/ed niorphometric anal ysis. The virus dose per mouse in each of the treatment groups of IO animals each was the following: <<i)DL-312 (6 x 10fus); (hi ADV/tk (3 X 10* pfus): (c) ADV/tk (3 X 10* p pfus) + ADV/mGM-CSF (2 X IO7 pfus); (J) ADV/tk (3 x 10" pfus) + ADV/mlL-2 (3 X 10" pfus); and (<! ADV/lk(3 X 10* pfus) + ADV/mIL-2 (3 X I0*pfus) + ADV/ mGM-CSF (2 X IO7 pfus). Columns, mean values for each group: burs. SDs.

,;
V ''l&---'\

TK+GM

TK

TK

TK+IL-2

TK+GM

TK+H--2+GM

m^i ' ^m***.-'

&>'.+..

- - ' '

v"3-,?
. HI

WP j*',

. \J23iar

B
Fig. 2. Im m unocy toc he m iaanalyses of the colon carcinoma 7 days after various treatments. A. H&B; 0. CD4 (L3T4): C CDS (Ly-2): D. macrophage (R/XO). In all images, the I tumor is on ihe left and the liver is tin the r-.if/i/. the samples from animals treated with ik + mIL-2 and Ik + mIL-2 +In mGM-CSK. no viable tumor remains, so only necrolic tissue is seen. Magnification. X2(M).The liver next to the tumor is very inflamed and there is a fatty change of the hepatticues and a dilation of the sinusoids in the cases with total tumor necrosis.

3759

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research

SUICIDE

AND CYTOKINE

GENE THERAPY

TOR HEPATIC

METASTASES

Fig. 3. Cellular immune response in animals after various gene therapy treatments as measured by CTL assays. The splenocytes were isolated from various treatment groups at different days after tumor cell inoculation. There were four animals in each group at days 14 and 25. After 60 days, only two animals in the tk + mIL-2 -t- mGM-CSF treatment group survived.

were increased dramatically in the livers of animals treated with tk + mGM-CSF or tk + mGM-CSF + mIL-2 (Fig. 2D). Immunokinetic Study in Animals Receiving Combination Gene Therapy. To assess the antitumor immune response in the treated animals, a kinetic study of cytolytic activity was performed. There was no significant CTL activity in the splenocytes of the animals treated with tk or tk + GM-CSF. but significant activities were detected at day 14 in the splenocytes of animals treated with tk + mIL-2 or tk + mIL-2 + mGM-CSF (Fig. 3). In the tk + mlL2-treated animals, the cytolytic activity did not persist. After 25 days, the activity had decreased to the background level. However, in the tk + mIL-2 + mGM-CSF-treated group, cytolytic activity was main tained at high levels for up to 60 days in the surviving animals. These result indicated that CTL activity was prolonged significantly in animals treated with both cytokine vectors. To evaluate further whether this CTL activity resulted in prolongation of systemic antitumoral immunity, protection against challenges by s.c. inoculation of tumorigenic doses of parental tumor cells was evaluated at various time points (Fig. 4). When challenged at day 14 after primary hepatic tumor cell implantation, all animals in the tk treatment group devel oped palpable s.c. tumors to an average size of 4 X 3 mm2 at challenge sites after 7 days. Five of five animals treated with tk + mIL-2 and four of five animals treated with tk + mlL2 + mGM-CSF showed protection against parental tumor challenge. However, when the challenges were performed at subsequent time points, there was a significant reduction in protection in the tk + mlL2-treatment group; only 50% were protected when challenged at day 28, and no protection was seen at day 75. Conversely, in the tk + mIL-2 + mGM-CSF-treated group, none of the eight animals formed s.c. tumors at day 28, and protection was also seen at day 75 and even at day 150. A significant difference in protection from tumor rechallenge (P < 0.04) was demonstrated between tk + mIL-2 and tk + mIL-2 + mGM-CSF by logistic regression. Long-Term Survival of Treated Animals. To assess treatment outcome, animals receiving tk. tk + mGM-CSF. tk + mIL-2, and tk + mIL-2 + mGM-CSF were studied for long-term survival (Fig. 5). In group treated with tk. tk + mGM-CSF. or tk + mIL-2, all animals developed relapsed tumor and died at 40-60 days. In the tk + IL-2 + GM-CSF-treated group, animals survived significantly

CTL Assay. The CTL assay was performed according to standard proto cols (10). Viable splenocytes were isolated from various animal treatment groups at different time points alter primary hepatic tumor inoculation. In vitro stimulation was performed for 5 days in 24-well plates, each well containing 6 x If)6 splenocytes. recombinant mIL-2 (20 units/ml), and 5 x IO5 CC36 cells that had received 15.000 rads (150 Gy) of radiation. The stimulated effector cells were coincubated with Cr" -labeled parental tumor cells for 4 h at 37Cin different effector and target cell ratios. Data represent the means of triplicate cultures and were analyzed by logistic regression.

RESULTS Regression of Hepatic Colon Tumor in Syngeneic Animals after Combination Gene Therapy in Vivo. An animal model for colon carcinoma in the liver was established by intrahepatic implantation of 1 X Iff CC36 cells. After 7 days, tumors of 5 X 4 mm2 in size were selected for subsequent gene therapy experiments. The animals with hepatic colon carcinoma were divided into five treatment groups, each receiving (a) a control adenoviral vector (DL-312); (/>) ADV/tk; (c) ADV/tk + ADV/mGM-CSF; (d) ADV/tk + ADV/mIL-2; or (e) ADV/tk + ADV/mIL-2 + ADV/mGM-CSF. Twelve h after viral inoculation, all animals received i.p. GCV ( 10 mg/kg twice daily) for 5 consecutive days. Two weeks after viral inoculation, the viable tumor cells in various treatment groups were quantified with comput erized morphometric point-count analysis of the maximal cross-sec tional area of the viable tumors (Fig. 1). Animals treated with ADV/tk vector alone had an apparent 50% reduction of tumor size as com pared with those treated with the control virus DL-312 (P < 0.05). In the animal groups treated with both tk and mIL-2 or mGM-CSF vectors, there was a further significant reduction of the residual tumor size as compared to the animal group treated with the tk vector alone (P < 0.05). There was, however, no significant further reduction of tumor size in animals treated with ADV/tk and both cytokine vector. There was massive infiltration of inflammatory cells surrounding the necrotic tumor area in the animals treated with tk + cytokine but not in those treated with tk alone (Fig. 2A). Immunocytochemical analy ses revealed that the infiltrates were mainly CDS+ lymphocytes in the tumor boundary area of animals treated with tk + mIL-2 or tk + mlL2 + mGM-CSF (Fig. 25). The number of CD4f lymphocytes was approximately equal in the animals treated with tk + mGM-CSF and those treated with tk + mIL-2 + mGM-CSF (Fig. 2C). Macrophages

~
O

100-

u> m

75

1
tk+GM El 0 tk+IL-2 tk+IL-2+GM

DAY
Fig. 4. Systemic antitumoral immunity against parental Iunior cell challenges at distant sites. A lumorigenic dose {1.5 X 10s) of colon tumor cells was inoculated s.c. in Ihe Hank of surviving inoculation. tk + mIL-2. tk + mIL-2 unimals in all Day 14: n = n = 9; tk + + mGM-CSF. treatment groups at different lime points after primary tumor 5 per group. Day 28: Ik. = 5; tk + mGM-CSF. n = 5; mIL-2 + mGM-CSF, n = 8. Day 75: tk + mIL-2, n = 2; n = 2. Day 150: tk + mIL-2 + mGM-CSF, /i = 2.

3760

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research

SUICIDE

AND CYTOKINE

GENE THERAPY

FOR HEPATIC

METASTASES

tk
tk+IL-2 tk+GM-CSF

pared with either treatment alone. Because there was no significant CTL activity detected in animals receiving the tk + mGM-CSF treatment, the synergistic effect in hepatic tumor regression was probably due to immune effector cells other than the T lymphocytes (17, 18). Significant CTL activity was observed only when tk was combined with mIL-2 with or without mGM-CSF. However, it was only with the combination tk + mIL-2 + mGM-CSF treatment that

s?
' Ie

tk+IL-2+GM-CSF

<
100
IL-Z

DL-312 80 control

30

60

90

120

1 50 60-

day
Fig. 5. Long-term survival of animals after various gene therapy treatments. The tumor-bearing animals were divided into four treatment groups: () ADV/tk. n = 8; (b) ADV/tk + ADV/mGM-CSF. n = 7: (c) ADV/tk + ADV/mIL-2. n = 7; and (d) ADV/tk -I- ADV/mIL-2 + ADV/mGM-CSF. n = 8. All the animals received GCV treatment 12 h after virus injection and were observed for survival over time. The survival of animals was analyzed by the Wilcoxon test (9). 40-

20-

longer than those in the tk-treated groups (P < 0.005), and 25% of treated animals developed long-term survival. To evaluate whether prolonged expression of cytokines is necessary to achieve better long-term survival, mIL-2 or control adenovirus-transduced parental tumor cells after irradiation were implanted s.c. into the treated animals at days 21, 35, and 49. The results showed that 9 of 16 animals treated originally with tk + mIL-2 + mGM-CSF showed long-term survival (Fig. 6A). The results were also dependent on mIL-2 expression, as control adenovirus-transduced parental tumor cells were ineffective, and the difference is statistically significant (P < 0.05). This beneficial outcome was not observed in the animals treated with tk + mIL-2 (Fig. 60). DISCUSSION Suicide gene therapy and cancer vaccine are promising approaches for cancer gene therapy (11-13). The combination of both approaches has been shown to be more effective than either approach alone, and the synergistic effect on the induction of antitumoral cellular immune response in the recipient animals has been reported previously (3). Although IL-2 has potent effects on tumor-specific T-cell activation and proliferation, there are multiple steps and regulatory cytokines to maximize an immune response n vivo. Simultaneous delivery of the most potent combination of cytokines to achieve maximal antitumoral cellular immunity is therefore critical in the development of an effec tive cancer treatment strategy. Studies of an approach using a com bination of cytokines by recombinant proteins, or ex vivo "cancer vaccine" approach, have been reported for a number of tumor models and have shown some success (14-16), although the details of the underlying immune mechanisms have not been well studied. Recom binant adenoviral vectors are capable of efficient transduction of target cells in vivo. The current study is the first paper on the use of adenoviral vectors to investigate the effect of combination suicide and multiple cytokine gene therapy by direct intratumoral delivery in a pre-established tumor model in vivo. Using a model of hepatic me tastasis of colon carcinoma, we showed that mGM-CSF induced tumor regression more efficiently when combined with tk as com-

20

40

60

80

DAY

100

80-

CO

I (0

DAY
Fig. 6. Long-term survival of combination gene therapy-treated animals with subse quent immunity boosting. A, hepatic tumors were treated with adenovirus vectors tk + raIL-2 + mGM-CSF and followed by GCV treatment for 5 days. Two weeks after virus inoculation, the immunity of treated animals was boosted with irradiated I X IO6 ADV/mIL-2 (n = 16)- or DL-312 (n = l4)-transduced tumor cells at multiplicity of infection 320, or with irradiated tumor cells alone (n = 13). The animals were boosted subsequently three times at 2-week intervals. B. hepatic tumors were treated with adeno virus vectors tk + mIL-2 and received an identical boosting schedule. Mice boosted with ADV/mIL-2-transduced tumors, a = 19; mice boosted with control virus-transduced tumor, n = 16; and mice boosted with irradiated tumor cells alone, n The survival 12. of animals was analyzed statistically by the Wilcoxon test (9).

3761

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research

SUICIDE

AND OTOKINE

GENE THERAPY

I-OR HEPATIC

METASTASES

antitumoral cellular immune response persisted, particularly with sub sequent boosting with mIL-2-expressing irradiated tumor cells. Regarding the immune mechanisms that are responsible for tumor rejection, we hypothesize the following mechanism. The initial tk + GCV treatment causes tumor death in situ, while local mGMCSF expression enhances the inflammatory response and activates the APCs, thus attracting CD4+ T cells to the local tumor area, as reflected by the results of the immunohistological staining. Locally expressed mIL-2 by transduced tumor cells then act synergistically to activate and enhance the proliferation of cytolytic T cells. Although effective antitumoral immune response can be induced by tk + mIL-2 treatment alone, the effect is only temporary. The capacity of GMCSF-activated professional APCs to promote activation and persist ence of tumor-specific precursor T cells is crucial for prolongation of the CTL response (19, 20). Because the level of tumor antigen and cytokine will decrease with time after primary treatment, prolongation of tumor antigen presentation and cytokine expression are necessary for improvement of long-term animal survival. The "cancer vaccine" approach with IL-2 has been shown to enhance the proliferation of colon tumor-specific CDS ' cytotoxic T cells (21). These educated T cells can also be clonally expanded in the lymphoid organ of a sensiti/.ed animal (22). Thus, in future gene therapy strategies involving tumor imtnunomodulation. it is imperative to focus on the enhancement and prolon gation of cytolytic immune response. Additional investigation to characterize the antitumoral immune regulatory mechanisms en hanced by GM-CSF will also be necessary to refine the future devel opment as a new treatment modality for metastatic colon carcinoma in vivo.

5. 6.

7.

8.

9. 10. 11.

12.

13. 14.

15.

16.

ACKNOWLEDGMENTS
We thank Dr. Susan Rich for helpful discussion and comments. Dr. Tom Shenk for use of the DL-312 vector, and Dr. F. Graham for providing the cloning vector for recombinant adenovirus construction. We also thank X-H. Li Chen and Guiling Zhao for their expert technical assistance.

17.

18. 19.

REFERENCES
1. Drehen. J. A., and Niederhuber. J. E. Colon Cancer. In: 1. E. Niederhuber (ed.). Current Therapy in Oncology, pp. 426-431. St. Louis: B. C. Decker. Inc.. 1993. 2. Lebovic, G. S., and Niederhuber. J. E. Colorectal cancer metastatic to the liver: hepatic arterial infusion. In: J. E. Niederhuber (ed.). Current Therapy in Oncology, pp. 389-395. Niederhuber. J. E., ed. B. C. Decker, St. Louis.. 1993. 3. Chen. S-H., Li Chen. X. H.. Wang. Y.. Kosai, K-i.. Finegold. M. J.. Rich. S. S.. and Woo, S. L. C. Combination gene therapy for liver metastasis of colon carcinoma in vivo. Proc. Nail. Acad. Sci. USA. 92: 2577-2581, 1995. 4. Gansbacher. B., Zier. K.. Daniels. B.. Cronin. K.. Bannerji, R., and Gilboa. E.

20.

Interleukin 2 gene transfer into tumor cells abrogates tumorigenicity and induces protective immunity. J. Exp. Med., 172: 1217-1224, 1990. Steinman, R. M. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol.. 9: 271-296. 1991. Dranoff. G.. Jaffee. E. Lazenby. A.. Golumbek. P.. Levitsky. H.. Brose. K., Jackson. V.. Hamada. H.. Pardoll. D.. and Mulligan, R. C. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-niacrophage colony-stimulating factor stimulates potent: specific, and long-lasting anti-tumor immunity. Proc. Nail. Acad. Sci. USA, 90: 3539-3543. 1993. Chen, S. H., Shine, H. D., Goodman, J. C.. Grossman, R. G., and Woo. S. L. C. Gene therapy for brain tumors: regression of experimental gliomas by adenovirus-mediated gene transfer in vim. Proc. Nati. Acad. Sci. USA, 91: 3054-3057. 1994. Corbet!. T. H.. Griswold, D. P.. Roberts. B. J.. Peckham. J. C. and Schabel, F. M., Jr. Tumor induction relationships in development of transplantable cancers of the colon in mice for chemotherapy assays, with a note on carcinogen structure. Cancer Res.. 35: 2434-2439, 1975. Cox, D. R.. and Oakes. D. Analysis of survival data, pp. 91-139. New York: Chapman and Hall, Inc.. 1984. Coligan, J. E.. Kruisbeck. A. M., Marguiles. D. H., Shevach, E. M.. and Straber, W. Current protocols in immunology. New York: Wiley-Interscience, 1991. Ezzeddine. Z. D., Martuza. R. L.. Platika. D.. Short. M. P.. Malick. A., Choi, B., and Breakfield. X. O. Selective killing of glioma cells in culture and in vim by retrovirus transfer of the Herpes simplex virus thymidine kinase gene. New Biol.. .<: 608-614. 1991. Culver. K. W.. Ram, Z.. Walbridge, S.. Ishii. H.. Oldfield, E. H., and Blaese, R. M. In vivo gene transfer with retroviral vector-producer cells for treatment of experi mental brain tumors. Science (Washington DC). 256: 1550-1552, 1992. Pardoll. D. M. Paracrine cylokine adjuvants in cancer immunotherapy. Annu. Rev. Immunol.. 13: 399-415. 1995. Pak, A. S.. Ip. G.. Wright. M. A., and Young. M. R. I. Treating tumor-bearing mice with low-dose y-interferon plus tumor necrosis factor to diminish immune suppressive granulocyte-macrophage progenitor cells increase responsiveness to interlcukin-2 immunotherapy. Cancer Res.. 55: 885-890, 1995. Pappo, I.. Tahara. H.. Rohbins. P. D.. Gately, M. K., Wolf, S. F.. Barhea. A., and Lotze. M. T., Administration of systemic or local interleukin-2 enhances the anti-tumor effects of interleukin-12 gene therapy. J Surg. Res., 58: 218-226. 1995. Tanaka, N., Sivanandham. M.. and Wallack. M. K. Immunotherapy of a vaccinia colon oncolysate prepared with interleukin 2 gene-encoded vaccinia virus and interferon a increases the survival of mice bearing syngenic colon adenocarcinoma. J. Immunother. Emphasis Tumor Immunol.. 16: 283-293. 1994. Schuurman. B.. Heuff, G.. Beelen. R. H.. and Meyer. S. Immunotherapy of human Kupffer cells after activation with human granulocyte/macrophage-colony-stimulating factor and interferon y. Cancer Immunol. Immunother., .9: 179-184. 1994. Jones, T. C. The effects of rhGM-CSF on macrophage function. Eur. J. Cancer. 2ft4(Suppl. 3): S10-S13, 1993. Allison. J. P.. Hurwit*. A. A., and Leaach, D. R. Manipulation of costimulatory signals to enhance antitumor T-cell responses. Curr. Opin. Immunol. 7: 682-686. 1995. Grabbe, S.. Geissent. S.. Schwanz. T.. and Granstein. R. D. Dendritic cells as initiators of tumor immune response: a possible strategy for tumor mmunotherapy1'

Immunol. Today. 16: 117-121, 1995. 21. Fearon, E. R.. Pardoll. D. M., Itaya. T.. Golumbek. P., Levitsky. H. !.. Simons. J. W.. Karasuyama. H.. Vogelstein. B.. and Frost. P. lnterleukin-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell, 60: 397-403. 1990. 22. Schmid!, W., Schweighoffer. T.. Herbst. E., Maass, G.. Berger. M.. Sehilcher, F., Schaffner. G., and Birnstiel, M. L. Cancer vaccines: the interleukin 2 dosage effect. Proc. Nati. Acad. Sci. USA. 92: 4711-4714. 1995.

3762

Downloaded from cancerres.aacrjournals.org on May 21, 2012 Copyright 1996 American Association for Cancer Research