Efficacy of herpes simplex virus thymidine kinase in ... - Nature

© 2000 Nature America, Inc. 0929-1903/00/$15.00/⫹0 www.nature.com/cgt

Efficacy of herpes simplex virus thymidine kinase in combination with cytokine gene therapy in an experimental metastatic breast cancer model Anish Sen Majumdar,1 Alya Zolotorev,1 Sara Samuel,1 Kimvan Tran,1 Beth Vertin,1 Melinda Hall-Meier,2 Beth-Ann Antoni,2 Emmanuelle Adeline,3 Mohan Philip,2 and Ramila Philip1 Departments of 1Oncology and 2Vector Development, Rho ˆne-Poulenc Rorer Gencell, Discovery Research, Hayward, California 94545; and 3Department of Molecular Histopathology and Biodistribution, Centre de Recherche de Vitry-Alfortville, Vitry-Alfortville, France. Immunotherapy in combination with suicide gene therapy for breast cancer was tested using a metastatic animal model. Subcutaneous injection of the nonimmunogenic breast cancer cell line 4T1 in BALB/C mice gave rise to tumors in 100% of mice with both micrometastases and macrometastases in the lung. We used the herpes simplex virus thymidine kinase (HSV-TK) gene along with the cytokine genes granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-2 (IL-2) to determine their effect on tumor regression and inhibition of lung metastasis. Adenoviral (AV) vectors carrying these transgenes, in separate constructs, were used in this study. Intratumoral administration of AV-TK followed by 10 days of ganciclovir treatment resulted in a delay in tumor growth and, in some cases, in a low to moderate reduction in tumor volume. Inclusion of either GM-CSF or IL-2 gene with HSV-TK resulted in a slightly greater reduction in tumor volume, although these data were not significantly different from those obtained for TK treatment alone. However, when both cytokine genes were combined with TK, a substantial reduction in tumor growth was observed compared with HSV-TK alone (P ⬍ .02). Tumor weight data also exhibited superior efficacy of TK/GM-CSF/IL-2 treatment when compared with animals treated with TK gene only (P ⬍ .01). More importantly, TK/GM-CSF/IL-2 combination gene therapy induced a significant reduction in lung metastasis compared with any other treatment groups in the 4T1 model (P ⬍ .001 between TK GM-CSF/IL-2 versus TK only). Surgical excision of primary tumors after TK/GM-CSF/IL-2 plus ganciclovir treatment resulted in anti-metastatic activity that was similar to that observed for animals in which no surgery was performed. Survival analysis showed a significant difference between animals treated with AV-TK/GM-CSF/IL-2 and animals treated with TK only at 35 days after virus injection (P ⬍ .01). Immunophenotypic data suggest infiltration of lymphocytes within the tumor microenvironment in TK- and cytokine gene-treated animals. Thus, the overall data presented here demonstrate that TK gene therapy in combination with GM-CSF and IL-2 gene-mediated immunotherapy strategies have important implications in the treatment of breast cancer. Cancer Gene Therapy (2000) 7, 1086 –1099

Key words: Gene therapy; breast cancer; metastasis; herpes simplex virus thymidine kinase; granulocyte-macrophage colonystimulating factor; interleukin-2.

S

omatic cell gene therapy has become one of the leading ways to treat cancer in a number of experimental models. One of the major treatment modalities of gene therapy is to transfer the herpes simplex virus thymidine kinase (HSV-TK) gene, which is known to confer conditional cytotoxicity to the nucleoside analog ganciclovir (GCV), to a variety of tumor cells in vivo and in vitro.1– 4 The HSV-TK gene product exerts its function by phosphorylating the Received October 22, 1999; accepted April 15, 2000. Address correspondence and reprint requests to Dr. A. Sen Majumdar, Geron Corporation, Department of Cell Biology, 230 Constitution Drive, Menlo Park, CA 94025. E-mail address: [email protected]

1086

nontoxic prodrug GCV; the phosphorylated substrate becomes toxic for dividing cells by acting as a chain terminator for DNA synthesis. This so called “suicide gene” therapy approach has the additional advantage of eliciting a “bystander effect,” in which the neighboring untransduced tumor cells within the tumor microenvironment become susceptible to GCV killing after HSV-TK treatment.2,5–7 Experimental data from murine and human tumor models suggest different but not mutually exclusive mechanisms that explain the bystander effect, including transfer of phosphorylated GCV between tumor cells via gap junctions, transductions of endothelial cells within the tumor, and induction of antitumoral immunity.8 –12

Cancer Gene Therapy, Vol 7, No 7, 2000: pp 1086 –1099

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

Immunotherapy, however, uses the transfer of genes of various cytokines and costimulatory molecules into tumor cells to stimulate an antitumoral immune response in experimental animals.13–15 This has mainly been achieved in two different ways. In the first instance, injection of genetically modified tumor cells with various cytokine genes, such as interleukin-2 (IL-2), IL-4, interferon-␥, granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage-CSF (GM-CSF), IL-6, IL-7 and IL-12, into syngeneic mice resulted in tumor rejection because of induction of an antitumoral immune response in the hosts.16 –25 In some instances, such treatment has resulted in immunity directed against the unmodified tumor cells after vaccination with cytokinesecreting tumor cells. Direct intratumoral (i.t.) transfer of cytokine genes and sustained expression in cells has also been a strategy for augmenting tumor immunogenicity, which ultimately generated strong antitumoral immune response in mice.13 Recently, combination gene therapy with HSV-TK and various cytokines has proven to be useful in a variety of experimental tumor models.26 –29 This combination therapy is advantageous because it induces conditional toxicity after HSV-TK and GCV treatment within the local tumor microenvironment and because it generates a local and systemic immune response due to cytokine administration. Use of recombinant adenoviral (AV) vector-mediated transgene delivery has many advantages over retroviral vectors, which have considerable limitations.15,30,31 Adenoviruses can be produced at high titers and do not need dividing cells for transduction and transgene expression; as a result, these vectors are suitable for direct gene transfer into a variety of cell types irrespective of their cell cycle status at the time of exposure to the viral transgene.32,33 In this report, we have studied the effect of the combination gene therapy approach using HSV-TK and the cytokine genes GM-CSF and IL-2 in a nonimmunogenic, metastatic breast tumor model, 4T1.34,35 IL-2 has long been known for its ability to activate antigen (Ag)-specific T cells both ex vivo and in vivo, resulting in an effective antitumoral immune response.16,36 GM-CSF is one of the most well-characterized cytokines, promoting the maturation and activation of Ag-presenting cells (APC).24 We reasoned that a combination therapy comprising the suicide and the cytokine genes would be more beneficial in our experimental model system to treat both local tumor growth and metastasis at a distant site compared with AV-TK treatment alone. Therefore, we used AV vectors to deliver these transgenes directly into the subcutaneously (s.c.) grown tumors. Here we demonstrate the efficacy of these genes to induce a local and systemic immune response in 4T1 tumor-bearing syngeneic BALB/c mice. MATERIALS AND METHODS

Mice We obtained 6- to 8-week-old BALB/c mice from Harlem Laboratories (Indianapolis, Ind). Mice were treated according

Cancer Gene Therapy, Vol 7, No 7, 2000

1087

to National Institutes of Health guidelines. All protocols requiring animal experimentation received prior approval from the Rho ˆne-Poulenc Rorer Gencell Animal Care and Use Committee.

Establishment of metastatic breast tumor model A nonimmunogenic mouse breast tumor cell line, 4T1, was used to establish the tumor model in syngeneic BALB/c mice as described previously.34,35 A total of 2.5 ⫻ 105 4T1 cells were injected s.c. in the right flank. Palpable tumors started to develop after 7 days, at which point tumor volumes were measured; thereafter, tumors were routinely measured twice a week until the experiments were completed. Metastasis to lung was detected in 4T1 tumor-bearing mice from day 18 onwards; at a later stage, liver metastasis was also observed. Tumor volume in mm3 was calculated using the following formula: L ⫻ W ⫻ H/2, where L equals length, W equals width, and H equals height); alternatively, the formula L ⫻ W2/2 was used when the width and height of the tumors were the same.

Construction of AV vectors for gene therapy An AV vector encoding HSV-TK gene was obtained from M. Janicot and J. M. Guillaume (Rhone Poulenc Rorer, Vitry Sur Seine, France). This is a first generation replication-defective vector lacking the E1 and E3 regions. The TK gene is expressed from an expression cassette consisting of the cytomegalovirus (CMV) immediate early promoter and the bovine growth hormone polyadenylation signal. For construction of murine GM-CSF (mGM-CSF)- and human IL-2 (hIL-2)encoding AV vectors, we designed an expression cassette consisting of the CMV immediate early promoter followed by a hybrid intron with ␤-globin splice donor and immunoglobulin G (IgG) splice acceptor (pCI; Promega, Madison, Wis) and the simian virus 40 late polyadenylation signal. Coding sequences for hIL-2 and mGM-CSF were obtained by polymerase chain reaction using primers spanning the complete coding region. These entire expression cassettes were transferred into pGY63 to obtain recombinant shuttle plasmids. These plasmids contain adeno 5 sequences 1–382 followed by the expression cassette and adeno sequence 3446 –3968.37 These two regions will undergo homologous recombination in Escherichia coli with an E1- and E3-deleted AV genome in pXL 2689.37 After recombination, the transgene expression cassette is in the E1 region of AV. The final constructs were digested with PacI and transfected in 293 cells to obtain AV prestocks. AV vectors were amplified in 293 cells and purified by CsCl density gradient centrifugation.

Immunofluorescence analysis for TK expression in 4T1 tumors: In vitro analysis Intracellular TK expression in 4T1 cells was determined by flow cytometry using a modified procedure for measuring intracellular cytokine expression that has been described previously.38 At different timepoints after AV-TK infection, 4T1 cells were trypsinized and washed once in phosphate-buffered saline (PBS) containing 10% fetal calf serum (FCS). Cell pellets were resuspended in the appropriate volume of PBSFCS, and cell counts were performed using a hemocytometer. Aliquots were made of 1.0 ⫻ 106 cells for each staining condition, and cells were centrifuged at 400 ⫻ g for 5 minutes. The supernatants were aspirated, and each cell pellet was resuspended in 500 ␮L of 4% paraformaldehyde. After incubation at room temperature for 15 minutes, cells were washed with 2 mL of PBS. Next, the cell pellets were resuspended in

1088

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

500 ␮L of 0.2% Triton X-100 in PBS, incubated at room temperature for 5 minutes for permeabilization, and subsequently washed with PBS as before. To determine TK expression, 4T1 cell pellets were resuspended in 100 ␮L of power block reagent (Biogenex, San Ramon, Calif) containing anti-TK antibody (Ab) (1/250 dilution of polyclonal rabbit antiserum) or in 1 ␮g of rabbit IgG as a negative control and incubated at room temperature for 1 hour. Cells were washed in PBS, and the cell pellets were resuspended in 100 ␮L of power block reagent containing 1 ␮g of goat anti-rabbit Ig conjugated to fluorescein isothiocyanate (Caltag Laboratories, South San Francisco, Calif) and incubated at room temperature for 30 minutes. Cells were washed as described above, and the pellets were resuspended in 300 mL of 1% paraformaldehyde and analyzed with a FACScan (Becton Dickinson, San Jose, Calif) using the Lysis II program. Mice were sacrificed at different timepoints to determine the in vivo expression of the TK transgene in 4T1 tumors after adenovirus infection. Tumors were taken from euthanized mice by surgery and placed in Dulbecco’s modified Eagle’s medium-FCS before processing. Tumor cell suspensions were made by pressing the tumor in a 40-␮M nylon cell strainer (Becton Dickinson Labware, Franklin Lakes, NJ) using a 10-mL syringe piston. Cells were washed in PBS-FCS, counted, and resuspended at a concentration of 1.0 ⫻ 107 cells/mL. A total of 1 ⫻ 106 cells were taken out of each tumor sample, and TK expression was determined as described above. The in vivo expression of the GM-CSF and IL-2 transgenes in 4T1 tumors after i.t. injection of adenovirus constructs was determined by enzyme-linked immunosorbent assay (ELISA). At 24 hours after AV-TK plus AV-GM-CSF plus AV-IL-2 administration, 4T1 tumors were resected. Tumor volume was determined before making tumor cell suspensions, and the total cell number from each tumor was enumerated. The total cells obtained from each tumor were resuspended in medium and seeded in a 24-well plate. After 3 days, the supernatants were collected and quantitated for GM-CSF and IL-2 by ELISA according to the manufacturer’s protocol (R&D Systems, Minneapolis, Minn). 4T1 tumors injected with AV-TK constructs were used as negative controls.

Flow cytometric analysis of lymphocytes Cell suspensions from spleens and tumors of mice from various treatment groups were prepared as described above. A total of 1 ⫻ 106 cells were incubated with fluorescein isothiocyanate, phycoerythrin, or biotinylated monoclonal Abs for 20 minutes at 4°C. The cells were then washed with PBS and incubated with avidin-tricolor (Caltag Laboratories) for 20 minutes at 4°C. After the final wash, cells were fixed with 1% paraformaldehyde solution. Three-color flow cytometric analysis was performed with Lysis II software on a FACScan (Becton Dickinson). Splenic lymphocytes or 4T1 tumor-infiltrating lymphocytes were phenotyped with monoclonal Abs to the following cell surface markers: CD3, CD4, and CD8 (PharMingen, San Diego, Calif).

In vivo gene therapy Because 4T1 tumors were metastatic within 2–3 weeks after injection, we chose day 8 –12 as a starting point for gene therapy experiments. At this timepoint, primary tumors that were grown s.c. were palpable, tumor volumes usually ranged between 50 and 100 mm3, and there was no evidence of metastasis (data not shown). Tumor volumes were measured before i.t. injection of various AV constructs. In most of the experiments described here, 1 ⫻ 1010 particles of AV-TK,

AV-GM-CSF, or AV-IL-2 virus were used. Equivalent numbers of particles of AV1.0 constructs containing no transgene were used as negative control. The total number of injected AV particles was never ⬎3 ⫻ 1010/mouse. AV vectors were injected in 30 –50 ␮L of 10 mM tris(hydroxymethyl)aminomethane-HCl (pH 8.0), 1 mM ethylenediaminetetraacetic acid, 100 mM NaCl, and 10% (vol/vol) glycerol at three different sites for each tumor. At 18 hours after viral injection, the animals were treated i.p. with GCV (100 ␮L) at 75 mg/kg twice daily for 10 consecutive days. Control groups of animals received saline injections for the same period of time. To analyze the efficacy of HSV-TK gene therapy with or without the combination of cytokine genes, three to five animals were sacrificed at 5–10 days after the completion of GCV treatment. At autopsy, the primary tumor was removed and weighed. To observe any effect of the therapy on prevention of metastasis, lungs were removed, blotted on tissue paper to remove blood clots, and weighed. Tumor and lung samples were fixed in 10% buffered formalin (Sigma, St. Louis, Mo) and stored for histological analysis.

Survival analysis Survival studies were set up with different treatment groups of animals in an identical manner. Study endpoints were death or sacrifice of the animal when it appeared in distress, as evidenced by lethargy, difficulty breathing, ruffled fur, and very high tumor volume (⬃1500 –2000 mm3), Generally animals started to exhibit some of these symptoms between 20 and 30 days after GCV treatment ended. Tumor volumes were determined twice a week throughout the experiment. After autopsy, primary tumor weights were measured. Lung weights were also measured to evaluate metastasis.

Histological and morphometric analysis of lung samples In addition to determining lung weights as a measure of metastasis, we have also performed the following experiment with TK- and cytokine gene-treated animals. BALB/c mice with palpable 4T1 tumors were divided into the following four groups: (a) TK plus GCV, (b) TK plus saline, (c) TK/GM-CSF/ IL-2 plus GCV, and (d) no virus treatment (control). I.t. AV administration was followed by 7 days of GCV treatment. After 3 days, primary tumors were surgically removed from half of the animals in each group; the remaining animals in each group were used as nonsurgery controls. At 2 weeks after surgery, animals were sacrificed from all surgery and nonsurgery groups and lung weights were measured. In nonsurgery groups, tumor volumes were measured twice a week until the animals in these groups were sacrificed. After lung weight determination, lung samples were fixed in neutral buffer containing 3.7% formalin for ⬎24 hours and embedded in paraffin. Serial sections (5 ␮m) were performed for each block, and the sections were stained with hematoxylin and eosin. The overall area of metastasis for each lung was quantified using an Axiohome microscope (Axiohome, Paris, France); the data are presented in ␮m2.

Statistical analysis Analysis of variance was performed on data from experiments on tumor volumes, lung weights, and survival studies using the Microsoft Excel program. The analysis tool in the program was used to test the hypothesis that the means from two or more sets of data are equal. All values on tumor volume and lung weight are means ⫾ SD. For morphometric studies on metas-

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

tasis for total lung area, statistical analysis was performed using Statview software (Abacus Concepts, Cary, NC). After transformations of the data in ordinal values, the Mann-Whitney U test was used to compare between groups (statistical significance when P values were ⬍.05).

RESULTS

HSV-TK gene expression in 4T1 cells To determine the level of HSV-TK gene expression in vitro, 4T1 cells were transduced at an increasing multiplicity of infection (MOI) with AV-TK. Transduced cells were harvested at 24 hours posttransduction, and TK expression was determined by flow analysis (Fig 1A). A linear relationship was observed between the percentage of cells expressing the TK transgene and increasing dose of virus. At a MOI of 500, ⬎50% of infected 4T1 cells expressed the transgene. To test whether 4T1 cells expressing the TK gene were susceptible to GCV toxicity in vitro, transduced cells were treated with either 10 ␮g/mL GCV in medium or with PBS for a period of 4 days. At a MOI of 100, ⬎90% of 4T1 cells were dead, suggesting that these cells are susceptible to GCV toxicity after transduction (data not shown). For in vivo TK gene expression in 4T1 tumors, palpable tumors were infected with AV-TK particles or with AV constructs containing no transgene (AVE); TK expression was determined 3 days later (Fig 1B). At this timepoint, between 15% and 22% of tumor cells were found to be positive for the TK gene product from five different tumors, whereas ⬃5% of cells expressed background staining when the tumors were injected with AVE constructs. In vivo expression of GM-CSF and IL-2 after adenovirus-mediated i.t. delivery was also determined by ELISA after 24 hours. 4T1 tumors injected with AV-TK/ GM-CSF/IL-2 showed an average of 45 pg of GM-CSF/ mm3 of tumor, whereas the average IL-2 expression was determined to be 75 pg/mm3 of tumor. Culture supernatants obtained from tumors injected with AV-TK construct did not show detectable levels of GM-CSF or IL-2. Taken together, these results indicate that all three transgenes were expressed in 4T1 tumors in vivo. At 1 week after the s.c. implantation of 2.5 ⫻ 105 cells in the mice, measurable tumors grew in all 40 mice. From this point onward, tumors were measured twice a week until the experiment was terminated. At day 10, tumor volumes ranged between 30 and 70 mm3. The mice were divided into two groups in such a way that the difference in average tumor volume between each treatment group did not vary by ⬎20 mm3. All mice received 1 ⫻ 1010 AV-TK virus particles i.t. in three different locations within each tumor. After 18 hours, one group of mice received i.p injections of GCV (75 mg/kg, twice daily), whereas the control group of mice received saline injections. The appropriate dose of GCV was calculated from the results of a titration experiment in which three doses (25 mg/kg, 50 mg/kg, and 75 mg/kg, respectively) of GCV were used after AV-TK injection. Tumor regression was best observed in mice that received a 75

Cancer Gene Therapy, Vol 7, No 7, 2000

1089

mg/kg dose of GCV; hence, this concentration was used for all subsequent experiments. Data obtained from three experiments were pooled together, and tumor volumes were plotted as a function of time after AV-TK injection (Fig 2A). Although the tumor volume consistently increased in the saline group, we observed a delay in tumor growth in AV-TK plus GCV-treated animals. However, this difference in tumor volume was not statistically significant. Lung weights from these mice were determined as a measure of metastasis when the experiment was terminated (Fig 2B). Although the mean lung weight in the TK plus GCV group of animals was slightly lower than in the TK plus saline-treated animals, the difference was not statistically significant. Moreover, when compared with average lung weights (0.207 g) from six normal age-matched mice, lung weights in both groups were considerably higher. Therefore, it appears that although AV-TK followed by GCV treatment delays tumor progression, it does not reduce lung metastasis formation in the 4T1 model.

Combination of TK gene therapy with immunotherapy for the treatment of established 4T1 tumors Several investigations have revealed the importance of cytokine gene-mediated immunotherapy for the treatment of various local and metastatic tumors with or without the combination of the suicide TK gene.16,24,26 –29 In an attempt to determine whether such a combination would be useful in our model, GM-CSF and IL-2 gene constructs were prepared in AV vectors as described in Materials and Methods. Pre-established 4T1 tumors were injected with different combinations of AV-TK and/or AV-GM-CSF and AV-IL-2 constructs followed by GCV treatment for 10 days. Five groups of animals were injected with the following combinations of AV constructs: (a) AV-TK plus GCV, (b) AV-TK plus saline, (c) AV-TK/GM-CSF plus GCV, (d) AV-TK/IL-2 plus GCV, and (e) AV-TK/GM-CSF/IL-2 plus GCV. The total number of AV particles used was 3.0 ⫻ 1010 per mouse. Data pooled from three such experiments are presented in Figure 3A. As described above, there were no significant differences in pretreatment tumor sizes between the five experimental groups. Similar to the experiment described in Figure 2, we found slower growth of tumors injected with AV-TK followed by GCV administration when compared with the corresponding saline controls (Fig 3A). Mice that received an i.t. injection of AV-TK/IL-2 did not have appreciable difference in tumor growth compared with AV-TK treatment only. Conversely, the combination AV-TK/GM-CSF treatment in this 4T1 model induced a further reduction in the tumor growth rate compared with either treatment with AV-TK alone or AV-TK/IL-2, although this difference is not statistically significant (Fig 3A). However, when both GM-CSF and IL-2 vectors were combined with the TK gene, there was a definite decrease in tumor growth in the majority of mice. The difference in final tumor volume obtained from this group of mice compared with that seen for the AV-TK plus GCV and AV-TK/IL-2 plus GCV groups was found to be statistically significant (P ⬍ .02 and

1090

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

Figure 1. HSV-TK gene expression in 4T1 cells. A: 4T1 cells were transduced with various MOIs of AV-TK and assayed 24 hours later for TK expression by intracellular staining. B: 4T1 tumors in BALB/c mice were injected with either AV-TK or AVE. After 3 days, tumors were excised and a single-cell suspension was made. Tumor cells were stained with polyclonal rabbit anti-TK Ab and analyzed by flow cytometry.

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

1091

Figure 2. Effect of AV-TK plus GCV on 4T1 tumor response. 4T1 tumors were injected with 1 ⫻ 1010 particles followed by 10 days of GCV or saline treatment. A: Tumor volumes were measured two to three times a week until animals were sacrificed. B: Lung weights were measured as a measure of metastasis. The data presented here represent 15–20 animals per group. Days posttreatment indicate time after i.t. injection of AV-TK.

P ⬍ .03 respectively). Moreover, the mean tumor volume obtained from these mice at the termination of the experiment was the lowest compared with any other groups. No tumor regression or delay was observed in 4T1 tumorbearing mice that received AV-TK/GM-CSF/IL-2 followed by saline treatment (data not shown). To determine whether AV-GM-CSF/IL-2 plus GCV administration has

Cancer Gene Therapy, Vol 7, No 7, 2000

any effect on 4T1 tumor growth, a separate experiment was performed with the following treatment groups: AV-TK plus GCV, AV-TK plus saline, AV-TK/GM-CSF/IL-2 plus GCV, AV-TK/GM-CSF/IL-2 plus saline, and AV-GMCSF/IL-2 plus GCV. Although the tumor regression pattern in animals receiving AV-TK/GM-CSF/IL-2 plus GCV and AV-TK plus GCV treatments was quite comparable

1092

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

Figure 3. Effect of i.t. injection of AV-TK in combination with AV-GM-CSF and AV-IL-2 on 4T1 tumor growth. A: Growth of 4T1 as a function of time in various treatment groups. Tumors of 50 –100 mm3 were injected with a viral gene combination, as indicated in the figure, followed by 10 days of GCV or saline administration. Tumor volume was determined two to three times per week. B: Lung weights from mice sacrificed from all treatment groups at the end of the experiment (25 days after AV injection). Average normal lung weight (⫾ SD) was obtained from six mice. The data for tumor volume and lung weight were obtained from 15–20 animals per group. Days posttreatment indicate time after i.t. delivery of AV vectors.

with the data shown in Figure 3A, the corresponding saline control groups did not exhibit tumor growth inhibition. Moreover, no tumor growth inhibition was observed in animals treated with the GM-CSF and IL-2 cytokine gene

combination followed by GCV administration (data not shown). Thus, it appears that cytokine gene therapy by itself is not sufficient to induce tumor growth reduction in the 4T1 breast tumor model.

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

1093

Figure 4. 4T1 tumor weight measurements obtained from AV-TK plus GCV-treated, AV-TK plus saline-treated, and AV-TK/GM-CSF/IL-2 plus GCV-treated groups of animals. BALB/c mice bearing 4T1 tumors (50 –100 mm3) were injected i.t. with different combinations of virus. A total of 5 to 10 mice were sacrificed 25 days after AV administration (35 days after 4T1 injection). Mean tumor weight (⫾ SD) data are presented.

Next, we determined the effect of cytokine with TK gene therapy on lung metastasis in the 4T1 model. Three to five animals from each group were sacrificed at the end of the experiment and lung weights were determined as a measurement of tumor metastasis and compared with normal lung weights (average of six animals). The lung weight data obtained from tumor-bearing mice in all treatment groups are presented in Figure 3B. For comparison, normal lung weight data obtained from six mice are also shown in the figure. It is evident from the data that inhibition of metastasis was best observed in mice that received the combination of TK, GM-CSF, and IL-2 constructs; the average lung weights (⬃0.185 g) of these mice were very similar to the normal lung weight average (⬃0.207 g). When the lung weight data obtained from the TK/GM-CSF/IL-2 group were compared with the group that was treated with TK alone, the P value was found to be highly significant (P ⬍ .001). A significant difference in lung metastasis was also observed between AV-TK/GM-CSF and AV-TK/GMCSF/IL-2-treated animals (P ⬍ .02). In contrast, the TK/ GM-CSF and TK/IL-2 groups did not show any statistically significant effect on the inhibition of lung metastasis compared with the group treated with AV-TK only. These data demonstrate that both cytokines are necessary in conjunction with TK to induce inhibition of local tumor growth and lung metastasis in the 4T1 model.

Effect of AV-TK and AV-TK/GM-CSF/IL-2 treatment on tumor weights In addition to measuring tumor volumes, determination of wet tumor weight can be a good indicator of inhibition of tumor growth after i.t. gene therapy. Therefore, in certain

Cancer Gene Therapy, Vol 7, No 7, 2000

experiments, we determined 4T1 tumor weights when the animals were sacrificed at 20 –25 days after adenovirus injections. Pooled data obtained from five to seven mice in the AV-TK plus GCV, TK plus saline, and AV-TK/GMCSF/IL-2 plus GCV groups are presented in Figure 4 for comparison. As expected, the average 4T1 tumor weight was the highest (2.0 ⫾ 0.9) in the TK plus saline group, followed by tumors in animals that received TK plus GCV administration (1.1 ⫾ 0.5). However, the lowest tumor weight average was obtained from mice that received AV-TK in combination with the two cytokines. The mean tumor weights from animals treated with AV-TK/GMCSF/IL-2 (0.5 ⫾ 0.2) was 2-fold less than those for the TK plus saline-treated group and ⬃4-fold less than those for the TK plus saline group. In fact, tumor weight comparison data between mice in groups treated with AV/TK/GMCSF/IL-2 plus GCV were significantly lower than those for mice treated with TK plus GCV (P ⬍ .01). Thus, it appears that both tumor volume and tumor weight data corroborate each other and point to the fact that treatment with TK in combination with the two cytokine genes exhibits a significant antitumoral effect compared with any other groups.

Effect of surgery after AV-TK/GM-CSF/IL-2 treatment on the development of metastasis This experiment was performed to determine whether surgical removal of the primary tumor has any effect on lung metastasis after AV-TK or AV-TK/GM-CSF/IL-2 treatment. Similar to the results described above, 4T1 tumor growth was best controlled in animals that received the AV-TK/GM-CSF/IL-2 gene combination

1094

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

compared with the TK plus GCV, TK plus saline, or no treatment groups (data not shown). Lung metastasis data obtained from the surgery and nonsurgery groups showed striking similarities (Fig 5A). In both groups, inhibition of lung metastasis was most predominant in animals in which the triple gene combination was used; the mean lung weights were comparable with those values obtained from the normal lung controls. However, unlike the results shown in Figure 3B, the difference of average lung weights between the AV-TK plus GCV and AV-TK/GM-CSF/IL-2 plus GCV groups was not statistically significant. This is perhaps due to the small number (three to four) of animals used in this experiment. To further determine whether analysis of micrometastasis would reveal a significant difference, we performed morphometric analysis on paraffin-embedded lung sections. Three of the four lung samples in AV-TK/ GM-CSF/IL-2-treated animals had no micrometastasis or macrometastasis in both the surgery and the nonsurgery groups. The remaining two animals (one each in the surgery and nonsurgery groups), which exhibited lung metastasis by morphometric analysis, had considerably lower values compared with the data generated from animals in other treatment groups (TK plus GCV, TK plus saline, and no treatment) where macrometastasis was observed throughout the lungs (Fig 5B). The differences in lung metastasis observed between the AV-TK/ GM-CSF/IL-2 group and the AV-TK plus GCV group, between the AV-TK/GM-CSF/IL-2 plus GCV group and the AV-TK plus saline group, and between the AV-TK/ GM-CSF/IL-2 plus GCV group and the virus untreated control group were all significant (P ⬍ .05). These data indicate that TK in combination with GM-CSF/IL-2 genes has a significant anti-metastatic activity irrespective of whether or not the primary tumors have been removed in our model system.

Overall survival analysis of 4T1 tumor-bearing mice after different treatment courses Because the AV-TK/GM-CSF/IL-2 gene-treated group of mice exhibited the best response in terms of antitumoral and anti-metastatic activities, we wanted to determine whether such treatment would have any effect on the survival of treated animals. Tumor-bearing mice were divided into three groups: (a) AV-TK plus GCV, (b) AVE (empty, no transgene) plus GCV, and (c) AV-TK/GM-CSF/IL-2 plus GCV. The tumors were injected with the appropriate viral constructs, and GCV treatment was followed for 10 days. The mice were monitored routinely for symptoms of distress such as weight loss, breathing difficulties, hunched backs, and roughed fur. Animal death was scored routinely. The experiment was terminated when all of the mice in the two control groups (AV-TK plus GCV and AVE plus GCV) died; the results are presented in Figure 6. Although no mice survived in the AV-TK or AVE groups, 6 of 20 animals survived from the groups that received the combination treatment of TK/GM-CSF/ IL-2. This difference in survival between the AV-TK/

GM-CSF/IL-2 group and both the AV-TK group and the AVE group was found to be statistically significant (P ⬍ .01). Taking all these data into consideration, it appears that the TK/GM-CSF/IL-2 gene combination had the best therapeutic effect in terms of tumor growth, lung metastasis, and survival in the 4T1 tumor model.

Immunophenotypic analyses in mice receiving TK and cytokine combination gene therapy The HSV-TK gene, when injected i.t., is known to cause inflammation at the tumor site.28 In response to inflammatory signals, APCs migrate to the tumor microenvironment and elicit an immune response. We reasoned that a combination of the cytokine genes GM-CSF and IL-2 would further potentiate the existing antitumoral immune response. To verify our theory, 4T1 tumors from different treatment groups were analyzed by flow cytometry for the presence of a lymphocyte population (Fig 7). We found a substantial increase in CD3⫹ T lymphocytes from tumors that had received TK/GMCSF/IL-2 gene treatment. The increase in the CD3⫹ lymphocytes correlated well with a concomitant increase in the CD4⫹ and CD8⫹ T cells. In contrast, in the TK plus GCV and TK plus saline groups, very few CD4⫹ cells were found, although in both these groups, moderate levels of CD3⫹ and CD8⫹ T cells were detected. It appears from the results that the addition of GM-CSF and IL-2 genes has boosted either the recruitment or the proliferation of T lymphocytes within the tumor microenvironment. DISCUSSION Breast cancer is one of the most frequent cancers in women, and there is a worldwide increase in its incidence. Although conventional therapy has yielded positive results, there is a need for the development of more effective and less toxic therapies that would reduce mortality in these patients. Such a therapy should also be effective in curing metastatic masses in patients, as the majority of the breast cancer patients die from disseminated metastatic disease occurring primarily in the lungs and liver. In this report, we have used a nonimmunogenic, highly metastatic tumor cell line, 4T1, in BALB/c mice.34,35 Although the tumors are injected s.c., spontaneous metastasis to the lung is observed in nearly 100% of the animals 3 weeks after tumor implantation. The suicide gene therapy approach has been used extensively to cure a variety of tumors in different experimental models.2– 4,7,9,10,31 In fact, HSV-TK gene therapy has been shown to be quite efficacious in another syngeneic mouse model of breast cancer induced by MOD cells, and also in nude mice in which a human breast carcinoma cell line was used to develop the tumor.39 In these models, primary tumors were generated in the liver. Our model differs from others in the sense that the primary s.c. tumors were injected with AV vectors carrying the therapeutic genes and was designed to address the question of whether such a

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

1095

Figure 5. Effect of posttreatment surgery on the development of metastasis. 4T1 tumors were injected with AV-TK with or without GM-CSF and IL-2 cytokine genes followed by 7 days of GCV treatment. After 3 days, tumors were surgically resected in half the number of animals in each group. Thereafter, the animals were observed for 15 days for the development of metastasis. Three to six animals were sacrificed from each group (with surgery and no surgery), and lung samples were weighed and subsequently fixed in formalin for histopathological studies. A: Mean lung weights (⫾ SD) are presented from three to six animals from the surgery and nonsurgery groups. B: Morphometric analysis of lung samples (two to four animals per group). The overall area of metastasis for each lung was quantified using an Axiohome microscope. Total lung metastasis area in ␮m2 is shown in the graph for each group.

Cancer Gene Therapy, Vol 7, No 7, 2000

1096

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

Figure 6. Survival of 4T1 tumor-bearing mice after AV-TK/GM-CSF/IL-2 treatment. BALB/c mice bearing 10-day-old 4T1 tumors were divided into three groups to receive the following treatments: AV-TK plus GCV, AVE plus GCV, or AV-TK/GM-CSF/IL-2 plus GCV. After 10 days of GCV treatment, the animals were left for 35 days for survival analysis. No mice survived in the AV-TK and AVE groups, whereas 6 of 20 mice survived in the group that received the triple gene combination.

delivery had any effect on spontaneous metastasis to a distant organ such as the lungs. Immune surveillance, which is normally operative in healthy individuals, is compromised in a tumor-bearing host. Within the intricate circuitry of various immune modulators, cytokines play a very important role in balancing the immune system. Numerous studies have shown antitumoral activities of various cytokines such as tumor necrosis factor-␣, IL-2, IL-6, IL-7, interferon-␥, GM-CSF, and IL-12.16 –18,20,24,40 These cytokines are useful because they exert direct effects on the tumor or recruit endogenous APCs and effector T cells to the tumor site. They are also thought to influence the enhancement of major histocompatibility complex expression on the tumor cells as well as increase the level of cell adhesion molecules and tumor-associated Ags on the cancerous cells.14 In light of these findings, suicide gene therapy and cancer vaccine approaches using the cytokine gene delivery system have become prospective and leading tools for the treatment of cancer. In several experimental models, the combination of both approaches has been shown to be more efficacious than either approach alone and has also been shown to elicit a synergistic antitumoral immune response in the tumor-bearing animals. In a hepatic metastatic model of colon carcinoma, a combined gene therapy approach using HSV-TK and the cytokine genes GM-CSF and IL-2 showed sustained antitumoral immunity and prolonged the survival of treated animals.28 Our results showed that AV-mediated local delivery of TK,

GM-CSF, and IL-2 genes together exhibits strong antitumoral effects locally, and more importantly, at a distant metastatic site, when compared with TK/GMCSF, TK/IL-2, or TK alone. As hypothesized in the publication by Chen et al,28 this effect is probably mediated by multiple cascades of immune responses generated by the delivery of TK and the two cytokine genes. First, treatment of the tumors with TK plus GCV causes inflammation at the tumor site, which draws and potentiates differentiation of various APCs into the tumor microenvironment. Second, the presence of local GM-CSF enhances and activates the APCs to present unknown tumor Ags from necrotic 4T1 tumor cells to the effector lymphocytes. The local expression of the IL-2 by transduced tumor cells then activates and enhances the proliferation of stimulated helper CD4⫹ T cells and cytolytic CD8⫹ T cells to elicit an effective antitumoral immune response. Immunohistochemical studies performed in the colon carcinoma model showed the presence of CD4 and CD8 immune effector cells at the tumor site when the animals were injected with the same gene combination. Thus, it appears that the combination of TK along with the cytokine genes GM-CSF and IL-2 could be effective in reducing tumor progression by increasing the antitumoral specific immune response. 4T1 tumor-bearing animals treated with these gene combinations also survived for a longer period of time in comparison with TK alone. The antitumoral specific immune responses generated by TK and cytokine genes can considerably prolong the long-term survival

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

1097

Figure 7. Phenotypic analysis of 4T1 tumors after treatment. Tumors were taken out from mice from the three different groups 10 days after the completion of GCV administration. A single-cell suspension was prepared and stained with directly conjugated anti-CD4, anti-CD8, and anti-CD3 Abs as described in Materials and Methods. Ab-stained cells were analyzed in a FACScan (Becton Dickinson) using electronic gating to depict the lymphoid population.

of tumor-bearing mice by decreasing tumor growth and by inhibiting spontaneous metastasis to the lung. One of the major focuses of our study was to evaluate whether TK plus cytokine gene therapy has any effect on the spontaneous metastasis that occurs in 4T1 model in BALB/c mice. I.t. injection of the TK gene exhibited very little effect on tumor metastasis to the lungs. However, TK gene delivery in combination with GM-CSF and IL-2 made a significant difference in lung metastasis when compared with any other group of animals (Fig 3B). The mechanism by which lung metastasis is inhibited in these animals is not clear at this time. There may be multiple possibilities to explain this observation. Immune effector cells, activated by the cytokine genes, may be directly cytotoxic to tumor cells with metastatic potential. Alternatively, activated cytotoxic T lymphocytes may migrate through systemic routes and elicit cytotoxic activity against tumor cells at a distant site. Because 4T1 tumors are highly vascularized, such a scenario may be envisioned. In a recent publication using the 4T1 model, Pulaski and Ostrand-Rosenberg41 have shown that treatment of mice with established primary and metastatic disease with major histocompatibility complex class II and B7.1-transfected tumor cells reduces or eliminates established spontaneous metastasis. However, this therapy had no effect on primary tumor growth. In a related series of experiments, Coveney et al35 have demonstrated a significant re-

Cancer Gene Therapy, Vol 7, No 7, 2000

duction in pulmonary metastasis in the lungs of mice receiving IL-2 gene-modified 4T1 cells in the hind limb amputation model in BALB/c mice. Taken together, these data and the results shown in this manuscript suggest that immunotherapy treatment using cytokine genes or genes coding for costimulatory molecules can have significant therapeutic potential for the treatment of metastasis. In addition to the measurement of lung weights as an indication of metastasis, we also examined the histopathology of these lung samples. Morphometric analysis of total lung area showed that in both the surgery group and nonsurgery group of animals, AV-TK/GM-CSF/IL2-treated mice showed a significant inhibition of metastasis (Fig 5B). These data indicate that the TK immunotherapy strategy may possibly stimulate the immune system to control the development of metastasis. This may have important implications in treating patients with combination gene therapy before surgical removal of the primary tumor to elicit an impact on the metastases at later timepoints. In a recent publication, Bonnekoh et al42 have described the efficacy of i.t. administration of AV-TK/GM-CSF/IL-2 followed by GCV treatment at inducing a systemic antitumoral immune response in a rechallenge model of B16 melanoma. However, inclusion of the cytokine genes along with the TK-suicide gene therapy did not show inhibition of primary tumor growth during the 7-day period. Although the immunotherapy treatment in combination

1098

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

with the TK gene induced a significant reduction in challenged tumor volume compared with the TK only control group, its role in inhibiting lung metastasis has not been addressed.42 In summary, it is evident that our present investigation conclusively reports a local antitumoral response, as well as inhibition of metastasis at a distal site. The overall data presented here demonstrate that AV constructs with TK in combination with GM-CSF and IL-2 genes have important implications in the treatment of breast cancer. ACKNOWLEDGMENTS

11.

12.

13.

We thank Drs. M. Janicot and J.-M. Guillaume for providing AV-TK prestock. We are also grateful to D. Cosmatos and K. Miller for performing some of the statistical analyses on the data presented in this publication. We also gratefully acknowledge the excellent technical help with all animal experiments provided by T. Ramirez, G. Wilkerson, and J. Sudduth. Finally, we thank Drs. F. Meyer, J. Lebkowski, and Y. Chiang for their helpful suggestions and discussion throughout the course of the study.

14.

REFERENCES

17.

1. Moolten FL. Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control study. Cancer Res. 1986;46: 5276 –5281. 2. Moolten FL. Drug sensitivity (suicide) genes for selective cancer chemotherapy. Cancer Gene Ther. 1994;1:279 –287. 3. Freeman SM, Abboud CN, Whartenby KA, et al. The “bystander effect”: tumor regression when a fraction of tumor mass is genetically modified. Cancer Res. 1993;53: 5274 –5283. 4. Chen SH, Shine HD, Goodman JC, Grossman RG, Woo SL. Gene therapy for brain tumors: regression of experimental gliomas by adenovirus-mediated transfer in vivo. Proc Natl Acad Sci USA. 1994;91:3054 –3057. 5. Hamel W, Magnelli L, Chiarugi VP, Israel MA. Herpes simplex virus thymidine kinase/ganciclovir-mediated apoptotic death of bystander cells. Cancer Res. 1996;56:2697– 2702. 6. Elshami AA, Saavedra A, Zhang H, et al. Gap junctions play a role in the “bystander effect” of the herpes simplex virus thymidine kinase/ganciclovir system in vitro. Gene Ther. 1996;3:85–92. 7. Kianmanesh AR, Perrin H, Panis Y, et al. A “distant” bystander effect of suicide gene therapy: regression of nontransduced tumors together with a distant transduced tumor. Hum Gene Ther. 1997;8:1807–1814. 8. Vrionis FD, Wu JK, Qi P, Waltzman M, Cherington V, Spray DC. The bystander effect exerted by tumor cells expressing the herpes simplex virus thymidine kinase (HSVtk) gene is dependent on connexin expression and cell communication via gap junctions. Gene Ther. 1997;4:577– 585. 9. Ram Z, Walbridge S, Shaewr T, Culver KW, Blaese RM, Oldfield EH. The effect of thymidine kinase transduction and ganciclovir therapy on tumor vasculature and growth of 9L gliomas in rats. J Neurosurg. 1994;81:256 –260. 10. Barba D, Hardin J, Sadelain M, Gage FH. Development of anti-tumor immunity following thymidine kinase-mediated

15. 16.

18.

19. 20.

21.

22.

23. 24.

25. 26. 27.

killing of experimental brain tumors. Proc Natl Acad Sci USA. 1994;91:4348 – 4352. Gangadeep S, Brew R, Green B, et al. Prodrug-activated gene therapy: involvement of an immunological component in the bystander effect. Cancer Gene Ther. 1996;3:83– 88. Vile RG, Castleden S, Marshall J, Camplejohn R, Upton C, Chong H. Generation of an anti-tumor immune response in a non-immunogenic tumor: HSVtk killing in vivo stimulates a mononuclear cell infiltrate and a TH1-like profile of intratumoral cytokine expression. Int J Cancer. 1997;71:267–274. Colombo MP, Forni G. Cytokine gene transfer in tumor inhibition and tentative tumor therapy: where are we now? Immunol Today. 1994;15:48 –51. Porgador A, Feldman M, Eisenbach L. Immunotherapy of tumor metastases via gene therapy. Nat Immun. 1994;13: 113–130. Roth JA, Cristiano RJ. Gene therapy for cancer: what have we done and where are we going? J Natl Cancer Inst. 1997;89:21–39. Gansbacher B, Zier K, Daniels B, Cronin K, Bannerji R, Gilboa E. Interleukin-2 gene transfer into tumor cells abrogates tumorigenicity and induces protective immunity. J Exp Med. 1990;172:1217–1224. Cavallo F, Di Pierro F, Giovarelli M, et al. Protective and curative potential of vaccination with IL-2 gene-transfected cells from a spontaneous mouse mammary adenocarcinoma. Cancer Res. 1993;53:5067–5070. Stoppacciaro A, Melani C, Parenza M, et al. Regression of an established tumor genetically modified to release granulocyte colony-stimulating factor requires granulocyte-T cell cooperation and T-cell-produced interferon-␥. J Exp Med. 1993;178:151–161. Tepper RI, Pattengale PK, Leder P. Murine interleukin-4 displays potent antitumor activity in vivo. Cell. 1989;57: 503–512. Ganbacher B, Bannerji R, Daniels B, Zier K, Cronin K, Gilboa E. Retroviral vector-mediated interferon-␥ gene transfer into tumor cells generates potent and long-lasting antitumor immunity. Cancer Res. 1990;50:7820 –7825. Colombo MP, Ferrari G, Stoppacciaro A, et al. Granulocyte colony-stimulating factor gene transfer suppresses tumorigenicity of a murine adenocarcinoma in vivo. J Exp Med. 1991;173:889 – 897. Porgador A, Tzehoval E, Katz A, et al. Interleukin-6 gene transfection into Lewis lung carcinoma tumor cells suppresses the malignant phenotype and confers immunotherapeutic competence against parental metastatic cells. Cancer Res. 1992;52:3679 –3686. Hock H, Dorsch M, Diamantstein T, Blankenstein T. Interleukin-7 induces CD4⫹ T cell-dependent tumor rejection. J Exp Med. 1991;174:1291–1298. Dranoff G, Jaffee E, Lazenby A, et al. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA. 1993;90:3539 –3543. Caruso M, Pham-Nguyen K, Kwong YL, et al. Adenovirusmediated interleukin-12 gene therapy for metastatic colon carcinoma. Proc Natl Acad Sci USA. 1996;93:11302–11306. Chen SH, Chen XH, Wang Y, et al. Combination gene therapy for liver metastasis of colon carcinoma in vivo. Proc Natl Acad Sci USA. 1995;92:2577–2581. O’Malley BW Jr, Cope KA, Chen SH, Li D, Schwarta MR,

Cancer Gene Therapy, Vol 7, No 7, 2000

MAJUMDAR, ZOLOTOREV, SAMUEL, ET AL: HSV-TK AND CYTOKINE GENE THERAPY FOR BREAST CANCER

28.

29.

30. 31. 32.

33.

34.

35.

Woo SL. Combination gene therapy for oral cancer in a murine model. Cancer Res. 1996;56:1737–1741. Chen SH, Kosai K, Xu B, et al. Combination suicide and cytokine gene therapy for hepatic metastases of colon carcinoma: sustained antitumor immunity prolongs animal survival. Cancer Res. 1996;56:3758 –3762. Santodonanto L, D’Agostino G, Santini SM, et al. Local and systemic antitumor response after combined therapy of mouse metastatic tumors with tumor cells expressing IFN-␣ and HSVtk: perspectives for the generation of cancer vaccines. Gene Ther. 1997;4:1246 –1255. Kremer EJ, Perricaudet M. Adenovirus and adeno-associated virus-mediated gene transfer. Br Med Bull. 1995;51: 31– 44. Descamps V, Duffour MT, Mathieu MC, et al. Strategies for cancer gene therapy using adenoviral vectors. J Mol Med. 1996;74:183–189. Mittal SK, Bett AJ, Prevec L, Graham FL. Foreign gene expression by human adenovirus type 5-based vectors studied using firefly luciferase and bacterial ␤ galactosidase genes as reporters. Virology. 1995;210:226 –230. Engelhardt JF, Ye X, Doranz B, Wilson JM. Ablation of E2A in recombinant adenoviruses improves transgene persistence and decreases inflammatory response in mouse liver. Proc Natl Acad Sci USA. 1994;91:6196 – 6200. Aslakson CJ, Miller FR. Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor. Cancer Res. 1992;52:1399 –1405. Coveney E, Clary B, Iacobucci M, Philip R, Lyerly K.

Cancer Gene Therapy, Vol 7, No 7, 2000

36. 37. 38. 39.

40. 41.

42.

1099

Active immunotherapy with transiently transfected cytokine-secreting tumor cells inhibits breast cancer metastases in tumor-bearing animals. Surgery. 1996;120: 265–272. Arai K, Lee F, Miyajima A, Miyatake S, Arai N, Yokota T. Cytokines: coordinators of immune and inflammatory responses. Annu Rev Biochem. 1990;59:783– 836. Crouzet J, Naudin L, Orsini C, et al. Recombinational construction in Escherichia coli of infectious adenoviral genomes. Proc Natl Acad Sci USA. 1997;94:1414 –1419. Jung R, Schauer U, Heusser C, Neumann C, Rieger C. Detection of intracellular cytokines by flow cytometry. J Immunol Methods. 1993;159:197–207. Kwong YL, Chen SH, Kosai K, Finegold MJ, Woo SL. Adenoviral-mediated suicide gene therapy for hepatic metastases of breast cancer. Cancer Gene Ther. 1996;3: 339 –344. Blankenstein T, Qin ZH, Uberla K, et al. Tumor suppression after tumor cell-targeted tumor necrosis factor-␣ gene transfer. J Exp Med. 1991;173:1047–1052. Pulaski BA, Ostrand-Rosenberg S. Reduction of established spontaneous mammary carcinoma metastases following immunotherapy with major histocompatibility complex class II and B7.1 cell-based tumor vaccines. Cancer Res. 1998;58:1486 –1493. Bonnekoh B, Greenhalgh DA, Chen SH, et al. Ex vivo and in vivo adenovirus-mediated gene therapy strategies induce a systemic anti-tumor immune defense in the B16 melanoma model. J Invest Dermatol. 1998;110:867– 871.