- Email: [email protected]
Antigen-specific immunotherapy for human papillomavirus 16 E7-expressing tumors grown in the liver
Journal of Hepatology 2000; 33:91-98 Printed in Denmark • All rights reserved Munksgaard. Copenhagen
Copyright © EuropeanAssociation for the Study of the Liver2000 Journal of Hepatology
ISSN 0168-8278
Antigen-specific immunotherapy for human papillomavirus 16 E7-expressing tumors grown in the liver Chien-Hung Chen 1,6, Kwang Wook Suh 2'7, Hongxiu Ji 3, Michael A. Choti 1'2, Drew M. Pardoll 1 and T.-C. W H 1'3'4'5 Departments o f ~Oncology, 2Surgery, 3Pathology, Obstetrics and 4Gynecology, and 5Molecular Microbiology and Immunology, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA; 6Department o f Internal Medicine, National Taiwan University Hospital, College o f Medicine, National Taiwan University, Taipei, Taiwan; and 7Department o f Surgery, School of Medicine, Ajou University, South Korea
Background~Aims: We have previously reported a recombinant vaccinia-based vaccine (vac-Sig/E7/ L A M P - l ) that demonstrated a significant anti-tumor effect in a subcutaneous tumor challenge model. Since the fiver is one of the most common sites for tumor metastasis and organ microenvironments may modulate tumor cell responses to therapies, the aim of the
present study was to evaluate the potency of vac-Sig/ E7/LAMP-1 in treating E7-expressing tumors grown in the fiver. Methods: For in vivo tumor prevention experiments, mice were vaccinated intraperitoneally with vac-Sig/ E7/LAMP-1 followed by intrahepatic tumor challenge. For in vivo tumor regression experiments, mice were first challenged with tumor cells and then vaccinated with vac-Sig/E7/LAMP-1 intraperitoneally. In addition, enzyme-finked immunospot assays were used to determine the frequency of E7-specific T cell precursors. Results: For in vivo tumor protection experiments, tumor growth was observed in all of the mice vaccinated with wild-type vaccinia and 60% of the mice vaccinated with wild-type E7 vaccinia. All of the
MMUNOTHERAPYrepresents an attractive approach to
I cancer treatment because it has the ability to eradicate systemic tumor in multiple sites in the body and the specificity to discriminate between neoplastic and non-neoplastic cells. Anti-tumor effects of the immune system are mainly mediated by cellular immunity. The cell-mediated component of the immune system is Received 29 July; revised 20 December; accepted 24 December 1999
Correspondence: T.-C. Wu, Department of Pathology, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, Maryland 21205, USA. Tel: 410 614 3899. Fax: 410 614 3548. e-mail: [email protected]
mice vaccinated with vac-Sig/E7/LAMP-1 remained tumor-free 30 days after tumor challenge. For the tumor regression assays, all of the mice vaccinated with vac-Sig/E7/LAMP-1 remained tumor-free 30 days after vaccination. In contrast, all of those mice receiving culture medium, wild-type vaccinia, or wild-type E7 vaccinia developed tumors in the liver. In addition, mice vaccinated with vac-Sig/E7/ LAMP-1 had the highest E7-specific CD8 + T cell precursors. Conclusions: Our data suggest that vac-Sig/E7/ LAMP-1 is an effective vaccine for controlling E7expressing tumors grown in the fiver and our model suggests that antigen-specific immunotherapy may represent a powerful tool for treating fiver tumors with characterized tumor-specific antigens. In addition, our data indicate that the number of E7-specific CD8 + T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7/LAMP-1 vaccinia.
Key words: Antigen-specific; Hepatic tumor; Human papifiomavirus; Immunotherapy; Liver; Lysosome-associated membrane protein-l; Vaccine; Vaccinia.
equipped with multiple effector mechanisms capable of eradicating tumors, and most of these anti-tumor immune responses are regulated by T cells (for review, see (1-3)). T cells also possess the ability to recognize tumor-specific antigens which serve as targets that T cells can use to distinguish neoplastic from non-neoplastic tissues (4) Tumor-specific antigens, when efficiently presented by antigen-presenting cells to both CD8 + cytotoxic T lymphocytes (CTLs) and CD4 + helper T cells, are capable of inducing potent T cell-mediated immunity (5). Activated T cells may function directly as effector cells, providing anti-tumor immunity through the lysis of tu-
91
C-H.
Chen et al.
mor cells or through the release of cytokines capable of interfering with the propagation of tumors. While most of the focus in cancer immunology is on CD8+ cytotoxic T lymphocyte responses, recent evidence indicates that CD4+ T cells are an equally critical component of the anti-tumor immune response. Successful immunity to cancer will therefore require activation of tumor-specific CD4+ T cells (6). Thus, the ideal cancer vaccine would enhance both CD8+ and CD4+ T cell responses by delivering a tumor-specific antigen into both the major histocompatibility complex I (MHC-I) and MHC-II pathways of antigen presentation. We have previously linked the sorting signals of the lysosome associated membrane protein-l (LAMP-l) to human papillomavirus-16 (HPV-16) E7 antigen, creating a chimera, Sig/E7/LAMP-1 (7). LAMP-l is a type 1 transmembrane protein localized predominantly in lysosomes and late endosomes (8). The cytoplasmic domain of LAMP-l protein contains the amino acid sequence, Tyr-Gln-Thr-Ile, that mediates the targeting of LAMP-l into the endosomal and lysosomal compartments (9). We found that expression of this chimera in vitro and in vivo with a recombinant vaccinia virus targets E7 to the endosomal and lysosomal compartments and enhances MHC-II presentation to CD4+ T cells compared to vaccinia virus expressing wild-type E7 (7). In addition, E7-specific CTL responses were augmented as well, possibly as a consequence of enhanced CD4+ T helper cell function (7). Furthermore, the Sig/E7/LAMP-1 vaccinia virus (vacSig/E7/LAMP-1) vaccine has been shown to generate strong anti-tumor immunity against HPV-16 E7-expressing subcutaneous tumors or lung metastases in the murine model (10,l l), while the wild-type E7 vaccinia virus (vat-E7) vaccine has no effect on established tumors (10). These findings unequivocally demonstrated the in vivo efficacy of the vat-Sig!E7/LAMP-1 using extrahepatic sites of tumor challenge. The organ microenvironment may modulate the tumor cell responses to therapies. Therefore, when developing clinical applications for the treatment of common solid malignancies, demonstrating efficacy of therapy within visceral sites such as the liver is important (12). For example, Fidler et al. have shown that subcutaneous tumors are sensitive to doxorubicin, whereas lung or liver metastases are not (13). Fukumura et al. have demonstrated that the liver microenvironment has different effects on some aspects of tumor angiogenesis and microcirculation compared with subcutaneous tissues (14). The density of microvessels in the tumors might influence the degree of immune effector cells infiltrating into the tumors (15). This suggests that immuno92
therapies that are effective against subcutaneous tumors may not be effective against the hepatic tumors of the same cell type. Therefore, the aim of the present study was to evaluate the efficacy of the vat-S&/E71 LAMP-l vaccine in preventing and treating TC-1 tumors grown in the liver.
Materials and Methods Murine tumor cell line The production and maintenance of TC-1 cells have been described previously (10). In brief, HPV-16 E6, E7 and ras oncogene were used to transform primary C57BL/6 mice lung epithelial cells. The cells were grown in RPM1 1640. supplemented with 10% (voVvo1) fetal bovine serum, 50 units/ml penicillin/streptomycin, 2 mM L-glutamine, 1 mM sodium pyruvate, 2 mM nonessential amino acids and 0.4 mg/ ml G418 at 37°C with 5%) COz. On the day of tumor challenge, TC1 cells were harvested by trypsinization, washed twice with 1 X Hanks buffered salt solution and finally resuspended in 1 X Hanks buffered salt solution to the designated concentration for injection. Mice and tumor cell inoculutiott Mice used in this study were purchased as 6- to 8-week-old male C57BL/6 mice from the National Cancer Institute (Frederick, Maryland, USA) and kept in the oncology animal facility of the Johns Hopkins Hospital (Baltimore, Maryland. USA). All animals procedures were performed according to approved protocols and in accordance with recommendations for the proper use and care of laboratory animals. The mice were anesthetized with intraperitoneal injection of 0.1 ml solution containing k&amine hydrochloride (25 mgi ml), xyladne (2.5 mg/ml), and 14.25% ethyl alcohol diluted 1:3 in 0.9% NaCl solutions. A small midline laparotomy incision was made and the left lobe of the liver was exteriorized. TC-1 cells in 10 ,1t1 Hanks buffered salt solution were injected with a 30-gauge needle directly into the liver parenchyma. The localized injection was confirmed by blanching of the liver parenchyma. Gentle compression for 60 s was used to minimize cellular extravasation and promote hemostasis. The liver was then placed back into the peritoneal cavity, and the incision was closed with metal w-ound clips. Mice were sacrificed by cervical dislocation at the preset endpoint. Vaccine preparation and administratiorl The generation of recombinant vaccinia virus containing wild-type HPV-16 E7 (vacE7) or Sig/E7/LAMP-I (vat-SigiE’TILAMP-1) were described as before (7). Wild-type vaccinia virus (vat-wt) was used as a vector control and vaccination with culture medium only was also used as a vaccine control. The viral stock was preserved at -70°C prior to vaccination. Before use, the virus was thawed, sonicated in liquid phase for 30 s, trypsinized with trypsin/EDTA in 37°C water bath for 30 min, and diluted with minimal essential medium containing 2.5% fetal bovine serum to the final concentration of 1 X IO8 PFU/ ml. Each mouse was vaccinated with 10’ PFU of vaccinia (0.1 ml of the diluted vaccine) intraperitoneally. Tumor growth kinetics experiments To determine the minimal tumor dose that can grow tumor in the liver, TC-1 tumor cells were inoculated in the livers in five groups of C57BL16 mice (five mice per group) at various doses, including 5X104, 1X104, 2X10”, 4X10’ and 1X10’ cells/mouse. Mice were monitored twice a week and were sacrificed at day 30 to determine the presence or absence of tumors within the liver. In vivo tumor protection experiments Four groups of mice (five mice per group) were used. For vaccination, 1 X 10’ PFUs of each vaccinia (including vat-wt, vat-E7, vat-Sig/E7/ LAMP-l) were injected intraperitoneally in 3 groups of mice. The fourth group of mice was injected intraperitoneally with culture medium only. Intrahepatic tumor challenges with 1X10’ TC-1 cells/ mouse were done at day 7 after vaccination. Mice were monitored
Antigen specific immunotherapy for liver tumors twice a week and were sacrificed at day 30 after tumor challenge determine the presence or absence of tumors within the liver.
to
In vivo tumor regression experiments Four groups of mice (five mice per group) received intrahepatic tumor challenges with 1 X104 TC-1 cells/mouse. Seven days later, 1X10’ PFUs of each vaccinia (including vat-wt, vat-E7, vat-Sig/B7/LAMP1) were intraperitoneally injected in 3 groups of mice. The fourth group of mice was injected intraperitoneally with culture medium only. Mice were monitored twice a week and were sacrificed at day 30 after vaccination to determine the presence or absence of tumors within the liver. Tumor assessment and histologic studies After sacrificing the mice, the livers were removed and examined in their fresh state for the presence of tumors by inspection and palpation. Tumor size was measured using calipers and was calculated according to the formula: AXB2Xn/6, where A is the maximal tumor diameter and B is the diameter perpendicular to the maximal diameter. The samples were then fixed with 10% neutral buffered formalin, embedded in paraffin and cut into 5-pm serial sections. After hematoxylin and eosin staining, microscopic sections were examined using conventional light microscopy. Enzyme-linked immunospot (ELISPOT) assay for IFN-y-secreting cells For ELISPOT assay, four groups of mice (two mice per group) were used. The vaccination dose and route of administration were the same as those used in the tumor protection and regression experiments. At day 10 after vaccination, the mice were sacrificed and the splenocytes were harvested. The ELISPOT assay described by Miyahira et al. and Murali-Krishna et al. was modified to detect HPV-16 E7-specific T cells (16,17). The 96-well filtration plates (Millipore, Bedford, MA, USA) were coated with 10 ,&ml rat anti-mouse interferon-y (IFN-y) antibody (clone R46A2, Pharmingen, San Diego, CA, USA) in 50 pl of phosphate-buffered saline. After overnight incubation at 4°C the wells were washed and blocked with culture medium containing 10% fetal bovine serum. Different concentrations of fresh isolated spleen cells from each vaccinated mice group, starting from 1 X 106/ well, were added in triplicate to the wells along with 15 U/ml interleukin-2. Cells were incubated at 37°C for 24 h either with or without 1 &ml E7 specific H-2Db CTL epitope E7 49-57 (18). After culture, the plate was washed and then followed by incubation with 5 &ml biotinylated IFN-), antibody (clone XMGl.2, Pharmingen) in 50 ~1 in phosphate-buffered saline at 4°C overnight. After washing six times, 1.25 pg/ml avidin-alkaline phosphatase (Sigma, St. Louis, MO, USA) in 50 ~1 phosphate-buffered saline, were added and incubated for 2 h at room temperature. After washing, spots were developed by adding 50 ~1 BCIP/NBT solution (Boehringer Mannheim, Indianapolis, IN, USA) and incubated at room temperature for 1 h. The images of the spots in each well were captured by CCD video camera and the quantitation of spot numbers were analyzed on a Macintosh computer using the NIH Image program 1.6 (developed at the U.S. National Institutes of Health and available on the Internet at http:// rsb.info.nih.gov/nih-image/) according to method used by MuCutcheon et al. (19).
1X 104, 2X lo3 cells/mouse developed tumor growth in the liver; at a dose of 4~ lo2 cell/mouse, 40% (2 of 5) of mice injected with TC-1 tumor cells developed tumors; at a dose of 1X lo2 cell/mouse, none of the mice developed hepatic tumors (Fig. 1). Thus, the minimal tumor dose for TC-1 cells to grow in the liver was 2X lo3 cells/mouse. We used 5 times the minimal tumor dose, i.e. 1X104 cells/mouse, as the tumor challenge dose in the subsequent tumor protection experiments and tumor regression experiments. In this murine model, a solitary tumor was seen in all untreated animals at 30 days. Minimal extrahepatic disease was seen upon gross examination of the peritoneal surfaces, lungs, and other sites. TC-1 tumors grown in the liver had slightly higher microvessel density than subcutaneous ones (data not shown). Vaccination
with vat-SiglE7lLAMP-1
against the growth of E7-expressing
protects
mice
tumors in mouse
livers
To determine whether vaccination with vacSg/E7/ LAMP-l protects mice against challenge with E7-ex-
5 0s
Statistical analysis The statistical significance and the error bars for the ELISPOT assay was determined using the analysis of variance (ANOVA) test, which determines the mean differences between each group. p-values less than 0.05 were considered statistically significant.
2
‘1 0
50000
10000
2000
1 400
100
TC-1 tumor dose Results TC-1 tumor growth in mouse liver
Tumor growth kinetics were determined by injecting various doses of TC-1 cells into the livers of C57BL/6 mice. All the mice injected with TC-1 cells at 5 X 104,
1. Intrahepatic tumor growth kinetics in TC-1. TC-1 tumors cells were injected into the livers of C57BLl6 mice at various doses (.5X1@, 1 X104, 2X103, 4X10’ and 1X102 cellslmouse). The mice were sacri@ed at day 30 to determine the presence or absence of tumors within the liver. Fig.
93
C.-H. Chen et ul.
pressing tumors, mice were vaccinated with 1X lo7 PFUs vaccinia/mouse and then challenged with 1X104 TC-1 cells/mouse 7 days after vaccination. Thirty days following tumor challenge, all mice that received vacSig/E7/LAMP-1 remained tumor-free. In contrast, 60% of the vat-E7-vaccinated mice grew hepatic tumors and all of those mice vaccinated with vat-wt or culture medium developed hepatic tumors (Fig. 2). Thus, vaccination with vat-Sig/E7/LAMP-1 completely protected mice against the growth of E7-expressing tumors in the liver, while vat-E7 provided only partial protection. The average tumor size (&standard deviation) in tumor-bearing mice was 4232241 mm3 in the control mice; 478 + 282 mm3 in vat-wt-vaccinated mice; and 358-+32 mm3 in vat-E7-vaccinated mice. There was no statistically significant difference between average tumor size in the control, vat-wt-vaccinated and vat-E7-vaccinated groups. Microscopic examination
revealed proliferative fibroblasts forming a wide fibrous zone between the tumor and non-tumor portion of the liver in vat-E7-vaccinated mice. Also, profuse inflammatory infiltrates predominantly composed of mononuclear cells were identified in the fibrous zone in vat-E7-vaccinated mice. The fibrous zone and inflammatory cell infiltrate were much less prominent in vat-wt-vaccinated mice. Only scattered mononuclear cells aggregates, without viable TC-1 tumor cells, could be found in the vat-Sig/E7lLAMP-1 -vaccinated mice (data not shown). Vaccination with vat-SiglE7lLAMP-1 previously inoculated E7-expressing mouse livers
To determine the therapeutic potential of vat-Sig/E7/ LAMP-l, mice were inoculated with 1X lo4 TC-1 cells/mouse to the liver and then vaccinated 7 days later with 1x lo7 PFUs vaccinia/mouse. The mean hepatic tumor size at day 7 after TC-1 inoculation was 1 mm in diameter. At the time of sacrifice (30 days post-vaccination), we found all the mice vaccinated with vdc-E7, vat-wt, or culture medium had developed tumor growth in their livers, while mice vaccinated with vat-SiglE7/LAMP-1 remained tumorfree (Fig. 3). Representative gross photographs of the liver tumors are shown in Fig. 4. The average tumor size was 2456% 1030 mm3 in control mice; 158123 14 mm3 in vat-wt-vaccinated mice; 15042690 mm3 in vat-E7-vaccinated mice. Though the tumor size was slightly larger in the control mice than in the vat-E7vaccinated mice, there was no statistically significant difference between the average tumor size in these three groups. Vat-SiglE7lLAMP-1 vaccination specific cellular immune response
Control
Vat-wt
Vat-E7
VacSigIE7ILl
Fig. 2. Vat-SiglE7lLAMP-1 protects mice against intrahepatic challenge with TC-I tumor. CS7BLl6 mice were immunized intraperitoneally with culture medium or vat-wt, vat-E7, vat-SigfE7lLAMP-I at 1 X107 PFUslmouse. Seven days later, mice were challenged intrahepatically with TCI tumor at I X104 cellslmouse. Mice were monitored twice a week and were sacrt$ced at day 30 after tumor challenge to determine the presence or absence of tumors within the liver.
94
eradicates tumor cells in
generates potent
E7-
The ELISPOT assay is a sensitive functional assay for IFN-), production by individual cell and can be applied to quantify antigen-specific CD8+ T cells (17). We used the ELISPOT assay to determine the frequencies of E7-specific CD8+ T cells in each vaccination group. We detected a greater number of E7 antigen-specific CD8+ T cells (160 spots/lo6 splenocytes) in vat-Sig/ E7/LAMP- 1-vaccinated mice, compared to only 50/l O6 splenocytes in vat-E7-vaccinated mice, (p-value= 0.0002). We did not detect E7-specific IFN-y-secreting CDS’ T cells in control or wild-type-vaccinated treated mice (Fig. 5). Thus, E7-specific CDS+ T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7lLAMP-1 vaccinia. These ELISPOT numbers are consistent with our previous CTL studies, which demonstrated enhanced specific lysis in
Antigen spectjic immunotherapy for liver tumors
Control Vacbvt
VacLE7 SigIE7ILl
Fig. 3. Vat-SiglE7lLAMP-1 eliminates previously inoculated intrahepatic TC-1 tumors. C57BLl6 mice were injected intrahepatically with TC-I tumor at 1 XI@ cells1 mouse and immunized with culture medium or vat-wt, vacE7, vat-SiglE7lLAMP-1 at I X10’ PFUslmouse 7 days later. Mice were sacrificed 30 days after vaccination to determine the presence or absence of tumors within the liver.
native approach to treat tumors in the liver. Though a number of strategies of cancer gene therapy have been demonstrated (22), the major limitation of most forms of cancer gene therapy is the lack of a systemic effect (23). The ideal cancer gene therapy should be able to generate systemic anti-tumor immunity in addition to a local anti-tumor effect. In this regard, immunotherapy is especially attractive because immunotherapy has the potency to eradicate systemic tumor in multiple sites in the body and the specificity to discriminate between neoplastic and non-neoplastic cells. Various strategies have been applied to experimental cancer immunotherapy (for review, see (24)). Some of these immunotherapy strategies have been applied to treat primary or metastatic liver tumors. Cytokine gene therapy (25-31) and dendritic cell-based therapy (32,33) can lead to either substantial hepatic tumor regression and/or induce an effective systemic anti-tumor immunity in the host and to prolong the median survival time of the treated animals. However, all of the above studies used whole tumor cells or whole cell lysates to present the full spectrum of uncharacterized tumor-specific antigens. In other words, these strategies are not antigen-specific forms of immunotherapy. Antigen-specific immunotherapy represents a more desirable approach for controlling tumors. Antigenspecific immunotherapy is less likely to generate non-
the vac_Sig/E7/LAMP- 1-vaccinated mice compared to vat-E7-vaccinated mice (7).
Discussion In this study, we demonstrated that vaccination with recombinant vaccinia vat-Sig/E7lLAMP-1 induces potent E7-specific anti-tumor immunity against E7-expressing tumors in the liver. Furthermore, the numbers of E7-specific CD8+ T cell precursors detected by ELISPOT directly correlated with the anti-tumor effect generated by vat-Sig/E7/LAMP-1. We focused on the liver in the present study because primary hepatocellular carcinoma is one of the most common fatal cancers (20) and the liver is the most common site of cancer metastases (21). Treatment of primary or metastatic tumors in the liver using conventional therapy, such as surgery, chemotherapy, radiotherapy and transcatheter arterial embolization has met with limited success. Gene therapy can be an alter-
Fig. 4. Representative gross photographs of the tumors in the liver in each vaccinated group. C57BLl6 mice were injected intrahepatically with TC-1 tumor at 1 Xl@ cells1 mouse and immunized with culture medium or vat-wt, vacE7, vat-SiglE7lLAMP-1 at 1X107 PFUslmouse 7 days later. Mice were sacrificed at day 30 after vaccination. Though the liver tumor size shown here is smaller in the vat-E7-vaccinated mice, there is no statistically signtjicance between the liver sizes in culture medium, vat-wt, and vacE7-vaccinated mice. No hepatic tumors were seen in vacSigIE7ILAMP-I-vaccinated mice.
95
C-H.
A
Chen et al.
240 q
without peptide
m with peptide
160
80
Control
Vat-wt
Vat-E7 SiglE7iLl
vat-SiglE7lLAMP-1 Splenocytes were harvested IO days after vaccination. The number of IFN-y-producing E7-speciJic CD8+ T cell precursors was determined using the ELISPOT assay (See texts for the detailed method). Statistical signiJicance and error bars were generated using ANOVA, with a p-value 4.0002. (A). The mean spot numbers of triplicates t SE in each vaccinated group,. (B). Representative ELISPOT pictures in each vaccinated group.
specific autoimmunity, and it provides more flexibility controlling the amount of antigen administered and the methods of antigen presentation to the immune system. In addition, one can correlate the clinical outcome to a specific immune response (34). To date. few studies directly address the issue of using antigen-specific immunotherapy, especially active immunization, for 96
tumors in the liver. The use of antigen-specific immunotherapy for treatment of tumors in the liver has previously been applied in the form of antibody-cytokine fusion proteins. For example, several studies in animals demonstrated that targeting delivery of interleukin-2 with antibody to tumor-specific antigens can elicit an effective cellular immune response against hepatic metastases (35-37). Alternatively, effective tumor vaccines can be generated by in vitro modification of tumor cells with cytokines and bispecific monoclonal antibodies that bind antigen on tumor cells to CD28 on T cells (38). Our current study used a vectorbased vaccine to deliver active antigen-specific immunization in order to control tumors in the liver. Though direct inoculation of cancer cells into the liver is not a true liver metastasis model, our model is an ideal model for investigating antigen-specific immunotherapy for tumors in the liver. We showed that the vacSig/E7/LAMP-1 could prevent hepatic tumor growth and eliminate established hepatic tumors. Our demonstration of the effectiveness of active antigen-specific immunization for the treatment of tumors in the liver is an important finding because several studies indirectly implied that the efficacy of immunotherapy might differ between subcutaneous and hepatic tumors (39). For example. Ganss et al. demonstrated that a strong tumor-specific antigen and the generation of CTL are not sufficient for the rejection of established tumors. The tumor environment, such as tumor microvasculature, is also a critical parameter that influences the effectiveness of activated anti-tumor lymphocytes (39). Furthermore, the metastatic cancer cells in the liver exhibit a higher percentage of Fas ligand expression than their primary counterparts, and thus these metastatic cancers cells may be able to evade immune destruction by inducing apoptosis of activated T lymphocytes (40). Our results are also consistent with the concept that the anti-tumor effects of the immune system are mainly mediated by cellular immunity, especially T cells. Activated T cells may function directly as effector cells, providing anti-tumor immunity through the lysis of tumor cells or through the release of cytokines capable of interfering with the propagation of tumors. Though the chromium release assay is a standard assay for cellular immunity, the ELISPOT assay is more sensitive and permits the determination of the number of antigen-specific T cells (16). Using the ELISPOT assay, we demonstrated that the frequency of E7-specific CDS ‘T cell precursors directly correlated with the anti-tumor effect generated by vat-Sig/E7/LAMP-1. The highest spot numbers were found in the vat-Sig/E7/ LAMP-l -vaccinated group. This finding is consistent
Antigen specific immunotherapy for liver tumors
with the CTL assay (7), in which the highest specific lysis was also found in the vat-Sig/E7/LAMP-l-vaccinated mice. Our previous study demonstrates that the anti-tumor immunity generated by vac_Sig/E7/LAMP1 is dependent on CD4+ and CDS+ T cells, as well as NK cells (10). We also observed that the average tumor size at day 30 after inoculation was larger in the control mice of this tumor regression experiment (2456 mm3) (Fig. 3) compared to those in the tumor protection experiment (423 mm3) (Fig. 2). It should be clarified that the endpoint of the tumor protection experiment is “day 30 after tumor challenge”, while the endpoint of the tumor regression experiment is “day 30 after vaccination”, or 37 days after tumor challenge (vaccination was given 1 week after tumor challenge). Therefore, the additional 7 days may account for the difference we observed between the tumor sizes of the control groups in the in vivo protection and in vivo treatment experiments. Active antigen-specific immunotherapy may have some potential pitfalls. Romieu et al. demonstrated that passive immunotherapy (adoptive transfer of selftumor antigen-specific CTLs), but not active immunization with peptide-based vaccine, is effective treating transgenic mice that harbored a progressive liver tumor associated with the expression of the SV40 large tumor T oncoprotein (41). This suggests that passive therapies targeted to self-tumor antigen may be more suitable than active immunization in the treatment of spontaneous tumors (41). The drawback of adoptive immunotherapy is the lack of reproducible and efficient approaches to the isolation and expansion of antigenspecific T cells (42). Safety issues regarding the use of the vaccinia virus as a vector must be considered. Using vaccinia virus to mediate gene transfer has the advantages of high efficiency of infection and high levels of recombinant gene expression (43). However, the highly attenuated vaccinia is still replication-efficient. Several clinical studies have shown that vaccinia vaccines are safe vectors in clinical trials in immunocompetent patients. A recombinant vaccinia vaccine encoding HPV-16 E6/E7 has been used for a phase I clinical trial in cervical cancer patients (44). Furthermore, a carcinoembryonic antigen-expressing recombinant vaccinia vaccine has been used in patients with advanced colorectal cancer (45). More recently, a Plasmodium falciparum malaria gene-expressing recombinant vaccinia vaccine is in a phase I/II clinical trial (46). No significant complications or environmental spread of vaccinia vaccines was noted in these trials. The recombinant vaccinia vaccines perhaps cannot be multiply administered be-
cause pre-exposure to vaccinia may decrease the immunogenicity of subsequent vaccinations with vaccinia (47). However, the barrier caused by preexisting vaccinia immunity can potentially be overcome by boosting with recombinant avian pox virus (48) or by administering the recombinant vaccinia via a mucosal vaccination route (49). In summary, our data suggest that vat-Sig/E7/ LAMP-l is an effective vaccine for controlling E7-expressing tumors in the liver and the number of E7specific CD8+ T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7/LAMP-1 vaccinia. Though our hepatic tumor models were designed for HPV-16 E7 antigen-containing tumors, similar strategies can potentially be employed to the other cancer systems with known tumor-specific antigens. As more tumor-specific antigens are characterized, due to recent advances in molecular immunology (4), antigenspecific vaccination in treating tumors of the liver becomes more promising.
Acknowledgements We would like to thank Drs. Keerti V Shah and Robert J. Kurman for helpful discussions. We would also like to thank Drs. Ralph H. Hruban and Richard Roden for critical review of the manuscript and Morris Ling for preparation of the manuscript. This work was supported by NIH 5 pol 34582-01, U19 CA72108-02, ROl CA7263 l-01, the Richard W. TeLinde fund, and the Alexander and Margaret Stewart Trust grant.
References 1. Pardoll DM. Cancer vaccines. Nat Med 1998; 4: 525-31. 2. Greten TF, Jaffee EM. Cancer vaccines. J Clin Oncol 1999; 17: 1047-60. based on the 3. Rosenberg SA. A new era for cancer immunotherapy genes that encode cancer antigens. Immunity 1999; 10: 281-7. 4. Van den Eynde BJ, van der Bruggen F? T ccl1 defined tumor antigens. Curr Opin lmmunol 1997; 9: 68493. SA. Cancer vaccines based on the identification of 5. Rosenberg genes encoding cancer regression antigens. Immunol Today 1997; 18: 175-82. 6. Pardoll DM, Topalian SL. The role of CD4+ T cell responses in anti-tumor immunity. Curr Opin Immunol 1998; 10: 588-94. I. Wu TC, Guarnieri FG, Staveley OCKF, Viscidi RP Levitsky HI, Hedrick L, et al. Engineering an intracellular pathway for major histocompatibility 8.
9.
10
11
C.-H. Chen et al.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21. 22. 23.
24. 25.
26.
27.
28.
29.
30.
98
Antigen-specific immunotherapy for murine lung metastatic tumors expressing human papillomavirus type 16 E7 oncoprotein. Int J Cancer 1998; 78: 41-5. Gilboa E. Murine models for cancer immunotherapy using cytokine gene modified tumor vaccines. In: Forni G, Foa R, Santoni A, Frati L, editors. Cytokine-Induced Tumor Immunogenicity. London: Academic Press; 1994. p. 130-43. Fidler IJ, Wilmanns C, Staroselsky A, Radinsky R, Dong Z, Fau D. Modulation of tumor cell response to chemotherapy by the organ environment. Cancer Metastasis Rev 1994; 13: 209-22. Fukumura D, Yuan F, Monsky WL, Chen Y, Jain RK. Effect of host microenvironment on the microcirculation of human colon adenocarcinoma. Am J Path01 1997; 151: 679-88. Nannmark U, Johansson BR, Bryant JL, Unger ML, Hokland ME, Goldfarb RH, et al. Microvessel origin and distribution in pulmonary metastases of B16 melanoma: implication for adoptive immunotherapy. Cancer Res 1995; 55: 4627-32. Miyahira Y, Murata K. Rodriguez D, Rodriguez JR, Esteban M, Rodrigues MM, et al. Quantification of antigen specific CDI+ T cells using an ELISPOT assay. J Immunol Methods 1995; 181: 45554. Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, Miller JD, et al. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 1998; 8: 177.-87. Feltkamp MC, Smits HL, Vierboom MP, Minnaar RP, de JB, Drijlhout JW, ter SJ, Melief CJ, Kast WM. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur J Immunol 1993; 23: 2242-9. McCutcheon M, Wehner N, Wensky A, Kushner M, Doan S, Hsiao L, et al. A sensitive ELISPOT assay to detect low-frequency human T lymphocytes. J Immunol Methods 1997; 210: 149966. Okuda K. Hepatocellular carcinoma: recent progress. Hepatology 1992; 15: 948-63. Lygidakis NJ, Pearl A. Metastatic liver disease-a review. Hepatogastroenterology 1997; 44: 1484-7. Ruiz J, Qian C, Prieto J. Gene therapy of liver tumors: principles and applications. Digestion 1998; 59 Suppl 2: 92-6. Schuster MJ, Wu GY. Gene therapy for hepatocellular carcinoma: progress but many stones yet unturned! Gastroenterology 1997; 112: 6569. Chen CH, Wu TC. Experimental vaccine strategies for cancer immunotherapy. J Biomed Sci 1998; 5: 231-52. Caruso M, Pham-Nguyen K, Kwong YL, Xu B, Kosai KI, Finegold M, et al. Adenovirus-mediated interleukin-12 gene therapy for metastatic colon carcinoma. Proc Nat1 Acad Sci USA 1996; 93: 11302-6. Shiratori Y, Kanai F, Hikiba Y, Moriyama H, Hamada H, Matsumura M, et al. Gene therapy for hepatic micrometastasis of murine colon carcinoma. J Hepatol 1998; 28: 88695. Siders WM, Wright PW, Hixon JA, Alvord WG, Back TC, Wiltrout RH, et al. T cell- and NK cell-independent inhibition of hepatic metastases by systemic administration of an IL-12-expressing recombinant adenovirus. J Immunoll998; 160: 5465574. Cao G, Kuriyama S, Du P, Sakamoto T, Kong X, Masui K, et al. Complete regression of established murine hepatocellular carcinoma by in viva tumor necrosis factor alpha gene transfer. Gastroenterology 1997; 112: 501-10. Atarashi Y, Yasumura S, Nambu S, Yoshio Y, Murakami J, Takahara T, et al. A novel human tumor necrosis factor alfa mutein, F4614, inhibits in vitro and in viva growth of murine and human hepatoma: implication for immunotherapy of human hepatocellular carcinoma. Hepatology 1998; 28: 57-67. Karpoff HM, D’Angelica M, Blair S, Brownlee MD, Federoff H, Fong Y. Prevention of hepatic tumor metastases in rats with herpes viral vaccines and gamma-interferon. J Clin Invest 1997; 99: 7999804.
31. Yoshida H, Enomoto H, Tagawa M, Takenaga K, Tasaki K, Nakagawara A, et al. Impaired tumorigenicity and decreased liver metastasis of murine neuroblastoma cells engineered to secrete interleukin-2 or granulocyte macrophage colony-stimulating factor. Cancer Gene Ther 1998; 5: 67-73. 32. Peron JM, Esche C, Subbotin VM, Maliszewski C, Lotze MT. Shurin MR. FLT3-ligand administration inhibits liver metastases: role of NK cells. J Immunol 1998; 161: 6164-70. 33. DeMatos P, Abdel-Wahab Z, Vervaert C, Seigler HF. Vaccination with dendritic cells inhibits the growth of hepatic metastases in B6 mice. Cell Immunol 1998; 185: 65-74. 34. Sznol M, Holmlund J. Antigen-specific agents in development [review]. Semin Oncol 1997; 24: 173-86. 35. Becker JC, Pancook JD, Gillies SD, Furukawa K, Reisfeld RA. T cell-mediated eradication of murine metastatic melanoma induced by targeted interleukin 2 therapy. J Exp Med 1996; 183: 2361-6. 36. Xiang R, Lode HN, Dolman CS, Dreier T, Varki NM, Qian X, et al. Elimination of established murine colon carcinoma metastases by antibody-interleukin 2 fusion protein therapy. Cancer Res 1997; 57: 4948855. 37. Lode HN, Xiang R, Dreier T, Varki NM, Gillies SD. Reisfeld RA. Natural killer cell-mediated eradication of neuroblastoma metastases to bone marrow by targeted interleukin-2 therapy. Blood 1998; 91: 170615. 38. Guo YJ, Che XY, Shen F Xie TP Ma J, Wang XN. et al. Effective tumor vaccines generated by in vitro modification of tumor cells with cytokines and bispecific monoclonal antibodies. Nat Med 1997; 3: 451 -5. 39. Ganss R, Hanahan D. Tumor microenvironment can restrict the effectiveness of activated anti-tumor lymphocytes. Cancer Res 1998; 58: 4673381. 40. Shiraki K, Tsuji N, Shioda T, Isselbacher KJ. Takahashi H. Expression of Fas ligand in liver metastases of human colonic adenocarcinomas. Proc Nat1 Acad Sci U S A 1997; 94: 642@5. 41. Romieu R, Baratin M. Kayibanda M, Lacabanne V Ziol M, Guillet JG, et al. Passive but not active CD8+ T cell-based immunotherapy interferes with liver tumor progression in a transgenic mouse model. J Immunol 1998; 161: 5133-7. 42. Yee C, Riddell SR, Greenberg PD. Prospects for adoptive T cell therapy. Curr Opin Immunol 1997; 9: 702--8. 43. Moss B. Genetically engineered poxviruses for recombinant gene expression, vaccination. and safety. Proc Nat1 Acad Sci USA 1996; 93: 11341-8. 44. Borysiewicz LK, Fiander A. Nimako M. Man S, Wilkinson GW, Westmoreland D, et al. A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 1996; 347: 152337. 45. McAneny D, Ryan CA, Beazley RM, Kaufman HL. Results of a phase 1 trial of a recombinant vaccinia virus that expresses carcinoembryonic antigen in patients with advanced colorectal cancer. Ann Surg Oncol 1996; 3: 4955500. 46. Ockenhouse CF, Sun PF Lanar DE, Wellde BT, Hall BT, Kester K, et al. Phase l/Ha safety, immunogenicity. and efficacy trial of NYVAC-Pt7. a pox-vectored. multiantigen, multistage vaccine candidate for Plusmodium ,falcipurrm malaria. J Infect Dis 1998; 177: 1664-73. 47. Hanke T, Blanchard TJ, Schneider J, Hannan CM, Becker M, Gilbert SC, et al. Enhancement of MHC class I-restricted peptide-specific T cell induction by a DNA prime/MVA boost vaccination regime. Vaccine 1998; 16: 439.-45. 48. Hodge JW, McLaughlin JP, Kantor JA, Schlom J. Diversified prime and boost protocols using recombinant vaccinia virus and recombinant non-replicating avian pox virus to enhance T-cell immunity and anti-tumor responses. Vaccine 1997; 15: 759-68. 49. Belyakov IM, Moss B, Strober W, Berzofsky JA. Mucosal vaccination overcomes the barrier to recombinant vaccinia immunization caused by preexisting poxvirus immunity. Proc Nat1 Acad Sci USA 1999: 96: 4512-7.
Copyright © EuropeanAssociation for the Study of the Liver2000 Journal of Hepatology
ISSN 0168-8278
Antigen-specific immunotherapy for human papillomavirus 16 E7-expressing tumors grown in the liver Chien-Hung Chen 1,6, Kwang Wook Suh 2'7, Hongxiu Ji 3, Michael A. Choti 1'2, Drew M. Pardoll 1 and T.-C. W H 1'3'4'5 Departments o f ~Oncology, 2Surgery, 3Pathology, Obstetrics and 4Gynecology, and 5Molecular Microbiology and Immunology, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA; 6Department o f Internal Medicine, National Taiwan University Hospital, College o f Medicine, National Taiwan University, Taipei, Taiwan; and 7Department o f Surgery, School of Medicine, Ajou University, South Korea
Background~Aims: We have previously reported a recombinant vaccinia-based vaccine (vac-Sig/E7/ L A M P - l ) that demonstrated a significant anti-tumor effect in a subcutaneous tumor challenge model. Since the fiver is one of the most common sites for tumor metastasis and organ microenvironments may modulate tumor cell responses to therapies, the aim of the
present study was to evaluate the potency of vac-Sig/ E7/LAMP-1 in treating E7-expressing tumors grown in the fiver. Methods: For in vivo tumor prevention experiments, mice were vaccinated intraperitoneally with vac-Sig/ E7/LAMP-1 followed by intrahepatic tumor challenge. For in vivo tumor regression experiments, mice were first challenged with tumor cells and then vaccinated with vac-Sig/E7/LAMP-1 intraperitoneally. In addition, enzyme-finked immunospot assays were used to determine the frequency of E7-specific T cell precursors. Results: For in vivo tumor protection experiments, tumor growth was observed in all of the mice vaccinated with wild-type vaccinia and 60% of the mice vaccinated with wild-type E7 vaccinia. All of the
MMUNOTHERAPYrepresents an attractive approach to
I cancer treatment because it has the ability to eradicate systemic tumor in multiple sites in the body and the specificity to discriminate between neoplastic and non-neoplastic cells. Anti-tumor effects of the immune system are mainly mediated by cellular immunity. The cell-mediated component of the immune system is Received 29 July; revised 20 December; accepted 24 December 1999
Correspondence: T.-C. Wu, Department of Pathology, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, Maryland 21205, USA. Tel: 410 614 3899. Fax: 410 614 3548. e-mail: [email protected]
mice vaccinated with vac-Sig/E7/LAMP-1 remained tumor-free 30 days after tumor challenge. For the tumor regression assays, all of the mice vaccinated with vac-Sig/E7/LAMP-1 remained tumor-free 30 days after vaccination. In contrast, all of those mice receiving culture medium, wild-type vaccinia, or wild-type E7 vaccinia developed tumors in the liver. In addition, mice vaccinated with vac-Sig/E7/ LAMP-1 had the highest E7-specific CD8 + T cell precursors. Conclusions: Our data suggest that vac-Sig/E7/ LAMP-1 is an effective vaccine for controlling E7expressing tumors grown in the fiver and our model suggests that antigen-specific immunotherapy may represent a powerful tool for treating fiver tumors with characterized tumor-specific antigens. In addition, our data indicate that the number of E7-specific CD8 + T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7/LAMP-1 vaccinia.
Key words: Antigen-specific; Hepatic tumor; Human papifiomavirus; Immunotherapy; Liver; Lysosome-associated membrane protein-l; Vaccine; Vaccinia.
equipped with multiple effector mechanisms capable of eradicating tumors, and most of these anti-tumor immune responses are regulated by T cells (for review, see (1-3)). T cells also possess the ability to recognize tumor-specific antigens which serve as targets that T cells can use to distinguish neoplastic from non-neoplastic tissues (4) Tumor-specific antigens, when efficiently presented by antigen-presenting cells to both CD8 + cytotoxic T lymphocytes (CTLs) and CD4 + helper T cells, are capable of inducing potent T cell-mediated immunity (5). Activated T cells may function directly as effector cells, providing anti-tumor immunity through the lysis of tu-
91
C-H.
Chen et al.
mor cells or through the release of cytokines capable of interfering with the propagation of tumors. While most of the focus in cancer immunology is on CD8+ cytotoxic T lymphocyte responses, recent evidence indicates that CD4+ T cells are an equally critical component of the anti-tumor immune response. Successful immunity to cancer will therefore require activation of tumor-specific CD4+ T cells (6). Thus, the ideal cancer vaccine would enhance both CD8+ and CD4+ T cell responses by delivering a tumor-specific antigen into both the major histocompatibility complex I (MHC-I) and MHC-II pathways of antigen presentation. We have previously linked the sorting signals of the lysosome associated membrane protein-l (LAMP-l) to human papillomavirus-16 (HPV-16) E7 antigen, creating a chimera, Sig/E7/LAMP-1 (7). LAMP-l is a type 1 transmembrane protein localized predominantly in lysosomes and late endosomes (8). The cytoplasmic domain of LAMP-l protein contains the amino acid sequence, Tyr-Gln-Thr-Ile, that mediates the targeting of LAMP-l into the endosomal and lysosomal compartments (9). We found that expression of this chimera in vitro and in vivo with a recombinant vaccinia virus targets E7 to the endosomal and lysosomal compartments and enhances MHC-II presentation to CD4+ T cells compared to vaccinia virus expressing wild-type E7 (7). In addition, E7-specific CTL responses were augmented as well, possibly as a consequence of enhanced CD4+ T helper cell function (7). Furthermore, the Sig/E7/LAMP-1 vaccinia virus (vacSig/E7/LAMP-1) vaccine has been shown to generate strong anti-tumor immunity against HPV-16 E7-expressing subcutaneous tumors or lung metastases in the murine model (10,l l), while the wild-type E7 vaccinia virus (vat-E7) vaccine has no effect on established tumors (10). These findings unequivocally demonstrated the in vivo efficacy of the vat-Sig!E7/LAMP-1 using extrahepatic sites of tumor challenge. The organ microenvironment may modulate the tumor cell responses to therapies. Therefore, when developing clinical applications for the treatment of common solid malignancies, demonstrating efficacy of therapy within visceral sites such as the liver is important (12). For example, Fidler et al. have shown that subcutaneous tumors are sensitive to doxorubicin, whereas lung or liver metastases are not (13). Fukumura et al. have demonstrated that the liver microenvironment has different effects on some aspects of tumor angiogenesis and microcirculation compared with subcutaneous tissues (14). The density of microvessels in the tumors might influence the degree of immune effector cells infiltrating into the tumors (15). This suggests that immuno92
therapies that are effective against subcutaneous tumors may not be effective against the hepatic tumors of the same cell type. Therefore, the aim of the present study was to evaluate the efficacy of the vat-S&/E71 LAMP-l vaccine in preventing and treating TC-1 tumors grown in the liver.
Materials and Methods Murine tumor cell line The production and maintenance of TC-1 cells have been described previously (10). In brief, HPV-16 E6, E7 and ras oncogene were used to transform primary C57BL/6 mice lung epithelial cells. The cells were grown in RPM1 1640. supplemented with 10% (voVvo1) fetal bovine serum, 50 units/ml penicillin/streptomycin, 2 mM L-glutamine, 1 mM sodium pyruvate, 2 mM nonessential amino acids and 0.4 mg/ ml G418 at 37°C with 5%) COz. On the day of tumor challenge, TC1 cells were harvested by trypsinization, washed twice with 1 X Hanks buffered salt solution and finally resuspended in 1 X Hanks buffered salt solution to the designated concentration for injection. Mice and tumor cell inoculutiott Mice used in this study were purchased as 6- to 8-week-old male C57BL/6 mice from the National Cancer Institute (Frederick, Maryland, USA) and kept in the oncology animal facility of the Johns Hopkins Hospital (Baltimore, Maryland. USA). All animals procedures were performed according to approved protocols and in accordance with recommendations for the proper use and care of laboratory animals. The mice were anesthetized with intraperitoneal injection of 0.1 ml solution containing k&amine hydrochloride (25 mgi ml), xyladne (2.5 mg/ml), and 14.25% ethyl alcohol diluted 1:3 in 0.9% NaCl solutions. A small midline laparotomy incision was made and the left lobe of the liver was exteriorized. TC-1 cells in 10 ,1t1 Hanks buffered salt solution were injected with a 30-gauge needle directly into the liver parenchyma. The localized injection was confirmed by blanching of the liver parenchyma. Gentle compression for 60 s was used to minimize cellular extravasation and promote hemostasis. The liver was then placed back into the peritoneal cavity, and the incision was closed with metal w-ound clips. Mice were sacrificed by cervical dislocation at the preset endpoint. Vaccine preparation and administratiorl The generation of recombinant vaccinia virus containing wild-type HPV-16 E7 (vacE7) or Sig/E7/LAMP-I (vat-SigiE’TILAMP-1) were described as before (7). Wild-type vaccinia virus (vat-wt) was used as a vector control and vaccination with culture medium only was also used as a vaccine control. The viral stock was preserved at -70°C prior to vaccination. Before use, the virus was thawed, sonicated in liquid phase for 30 s, trypsinized with trypsin/EDTA in 37°C water bath for 30 min, and diluted with minimal essential medium containing 2.5% fetal bovine serum to the final concentration of 1 X IO8 PFU/ ml. Each mouse was vaccinated with 10’ PFU of vaccinia (0.1 ml of the diluted vaccine) intraperitoneally. Tumor growth kinetics experiments To determine the minimal tumor dose that can grow tumor in the liver, TC-1 tumor cells were inoculated in the livers in five groups of C57BL16 mice (five mice per group) at various doses, including 5X104, 1X104, 2X10”, 4X10’ and 1X10’ cells/mouse. Mice were monitored twice a week and were sacrificed at day 30 to determine the presence or absence of tumors within the liver. In vivo tumor protection experiments Four groups of mice (five mice per group) were used. For vaccination, 1 X 10’ PFUs of each vaccinia (including vat-wt, vat-E7, vat-Sig/E7/ LAMP-l) were injected intraperitoneally in 3 groups of mice. The fourth group of mice was injected intraperitoneally with culture medium only. Intrahepatic tumor challenges with 1X10’ TC-1 cells/ mouse were done at day 7 after vaccination. Mice were monitored
Antigen specific immunotherapy for liver tumors twice a week and were sacrificed at day 30 after tumor challenge determine the presence or absence of tumors within the liver.
to
In vivo tumor regression experiments Four groups of mice (five mice per group) received intrahepatic tumor challenges with 1 X104 TC-1 cells/mouse. Seven days later, 1X10’ PFUs of each vaccinia (including vat-wt, vat-E7, vat-Sig/B7/LAMP1) were intraperitoneally injected in 3 groups of mice. The fourth group of mice was injected intraperitoneally with culture medium only. Mice were monitored twice a week and were sacrificed at day 30 after vaccination to determine the presence or absence of tumors within the liver. Tumor assessment and histologic studies After sacrificing the mice, the livers were removed and examined in their fresh state for the presence of tumors by inspection and palpation. Tumor size was measured using calipers and was calculated according to the formula: AXB2Xn/6, where A is the maximal tumor diameter and B is the diameter perpendicular to the maximal diameter. The samples were then fixed with 10% neutral buffered formalin, embedded in paraffin and cut into 5-pm serial sections. After hematoxylin and eosin staining, microscopic sections were examined using conventional light microscopy. Enzyme-linked immunospot (ELISPOT) assay for IFN-y-secreting cells For ELISPOT assay, four groups of mice (two mice per group) were used. The vaccination dose and route of administration were the same as those used in the tumor protection and regression experiments. At day 10 after vaccination, the mice were sacrificed and the splenocytes were harvested. The ELISPOT assay described by Miyahira et al. and Murali-Krishna et al. was modified to detect HPV-16 E7-specific T cells (16,17). The 96-well filtration plates (Millipore, Bedford, MA, USA) were coated with 10 ,&ml rat anti-mouse interferon-y (IFN-y) antibody (clone R46A2, Pharmingen, San Diego, CA, USA) in 50 pl of phosphate-buffered saline. After overnight incubation at 4°C the wells were washed and blocked with culture medium containing 10% fetal bovine serum. Different concentrations of fresh isolated spleen cells from each vaccinated mice group, starting from 1 X 106/ well, were added in triplicate to the wells along with 15 U/ml interleukin-2. Cells were incubated at 37°C for 24 h either with or without 1 &ml E7 specific H-2Db CTL epitope E7 49-57 (18). After culture, the plate was washed and then followed by incubation with 5 &ml biotinylated IFN-), antibody (clone XMGl.2, Pharmingen) in 50 ~1 in phosphate-buffered saline at 4°C overnight. After washing six times, 1.25 pg/ml avidin-alkaline phosphatase (Sigma, St. Louis, MO, USA) in 50 ~1 phosphate-buffered saline, were added and incubated for 2 h at room temperature. After washing, spots were developed by adding 50 ~1 BCIP/NBT solution (Boehringer Mannheim, Indianapolis, IN, USA) and incubated at room temperature for 1 h. The images of the spots in each well were captured by CCD video camera and the quantitation of spot numbers were analyzed on a Macintosh computer using the NIH Image program 1.6 (developed at the U.S. National Institutes of Health and available on the Internet at http:// rsb.info.nih.gov/nih-image/) according to method used by MuCutcheon et al. (19).
1X 104, 2X lo3 cells/mouse developed tumor growth in the liver; at a dose of 4~ lo2 cell/mouse, 40% (2 of 5) of mice injected with TC-1 tumor cells developed tumors; at a dose of 1X lo2 cell/mouse, none of the mice developed hepatic tumors (Fig. 1). Thus, the minimal tumor dose for TC-1 cells to grow in the liver was 2X lo3 cells/mouse. We used 5 times the minimal tumor dose, i.e. 1X104 cells/mouse, as the tumor challenge dose in the subsequent tumor protection experiments and tumor regression experiments. In this murine model, a solitary tumor was seen in all untreated animals at 30 days. Minimal extrahepatic disease was seen upon gross examination of the peritoneal surfaces, lungs, and other sites. TC-1 tumors grown in the liver had slightly higher microvessel density than subcutaneous ones (data not shown). Vaccination
with vat-SiglE7lLAMP-1
against the growth of E7-expressing
protects
mice
tumors in mouse
livers
To determine whether vaccination with vacSg/E7/ LAMP-l protects mice against challenge with E7-ex-
5 0s
Statistical analysis The statistical significance and the error bars for the ELISPOT assay was determined using the analysis of variance (ANOVA) test, which determines the mean differences between each group. p-values less than 0.05 were considered statistically significant.
2
‘1 0
50000
10000
2000
1 400
100
TC-1 tumor dose Results TC-1 tumor growth in mouse liver
Tumor growth kinetics were determined by injecting various doses of TC-1 cells into the livers of C57BL/6 mice. All the mice injected with TC-1 cells at 5 X 104,
1. Intrahepatic tumor growth kinetics in TC-1. TC-1 tumors cells were injected into the livers of C57BLl6 mice at various doses (.5X1@, 1 X104, 2X103, 4X10’ and 1X102 cellslmouse). The mice were sacri@ed at day 30 to determine the presence or absence of tumors within the liver. Fig.
93
C.-H. Chen et ul.
pressing tumors, mice were vaccinated with 1X lo7 PFUs vaccinia/mouse and then challenged with 1X104 TC-1 cells/mouse 7 days after vaccination. Thirty days following tumor challenge, all mice that received vacSig/E7/LAMP-1 remained tumor-free. In contrast, 60% of the vat-E7-vaccinated mice grew hepatic tumors and all of those mice vaccinated with vat-wt or culture medium developed hepatic tumors (Fig. 2). Thus, vaccination with vat-Sig/E7/LAMP-1 completely protected mice against the growth of E7-expressing tumors in the liver, while vat-E7 provided only partial protection. The average tumor size (&standard deviation) in tumor-bearing mice was 4232241 mm3 in the control mice; 478 + 282 mm3 in vat-wt-vaccinated mice; and 358-+32 mm3 in vat-E7-vaccinated mice. There was no statistically significant difference between average tumor size in the control, vat-wt-vaccinated and vat-E7-vaccinated groups. Microscopic examination
revealed proliferative fibroblasts forming a wide fibrous zone between the tumor and non-tumor portion of the liver in vat-E7-vaccinated mice. Also, profuse inflammatory infiltrates predominantly composed of mononuclear cells were identified in the fibrous zone in vat-E7-vaccinated mice. The fibrous zone and inflammatory cell infiltrate were much less prominent in vat-wt-vaccinated mice. Only scattered mononuclear cells aggregates, without viable TC-1 tumor cells, could be found in the vat-Sig/E7lLAMP-1 -vaccinated mice (data not shown). Vaccination with vat-SiglE7lLAMP-1 previously inoculated E7-expressing mouse livers
To determine the therapeutic potential of vat-Sig/E7/ LAMP-l, mice were inoculated with 1X lo4 TC-1 cells/mouse to the liver and then vaccinated 7 days later with 1x lo7 PFUs vaccinia/mouse. The mean hepatic tumor size at day 7 after TC-1 inoculation was 1 mm in diameter. At the time of sacrifice (30 days post-vaccination), we found all the mice vaccinated with vdc-E7, vat-wt, or culture medium had developed tumor growth in their livers, while mice vaccinated with vat-SiglE7/LAMP-1 remained tumorfree (Fig. 3). Representative gross photographs of the liver tumors are shown in Fig. 4. The average tumor size was 2456% 1030 mm3 in control mice; 158123 14 mm3 in vat-wt-vaccinated mice; 15042690 mm3 in vat-E7-vaccinated mice. Though the tumor size was slightly larger in the control mice than in the vat-E7vaccinated mice, there was no statistically significant difference between the average tumor size in these three groups. Vat-SiglE7lLAMP-1 vaccination specific cellular immune response
Control
Vat-wt
Vat-E7
VacSigIE7ILl
Fig. 2. Vat-SiglE7lLAMP-1 protects mice against intrahepatic challenge with TC-I tumor. CS7BLl6 mice were immunized intraperitoneally with culture medium or vat-wt, vat-E7, vat-SigfE7lLAMP-I at 1 X107 PFUslmouse. Seven days later, mice were challenged intrahepatically with TCI tumor at I X104 cellslmouse. Mice were monitored twice a week and were sacrt$ced at day 30 after tumor challenge to determine the presence or absence of tumors within the liver.
94
eradicates tumor cells in
generates potent
E7-
The ELISPOT assay is a sensitive functional assay for IFN-), production by individual cell and can be applied to quantify antigen-specific CD8+ T cells (17). We used the ELISPOT assay to determine the frequencies of E7-specific CD8+ T cells in each vaccination group. We detected a greater number of E7 antigen-specific CD8+ T cells (160 spots/lo6 splenocytes) in vat-Sig/ E7/LAMP- 1-vaccinated mice, compared to only 50/l O6 splenocytes in vat-E7-vaccinated mice, (p-value= 0.0002). We did not detect E7-specific IFN-y-secreting CDS’ T cells in control or wild-type-vaccinated treated mice (Fig. 5). Thus, E7-specific CDS+ T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7lLAMP-1 vaccinia. These ELISPOT numbers are consistent with our previous CTL studies, which demonstrated enhanced specific lysis in
Antigen spectjic immunotherapy for liver tumors
Control Vacbvt
VacLE7 SigIE7ILl
Fig. 3. Vat-SiglE7lLAMP-1 eliminates previously inoculated intrahepatic TC-1 tumors. C57BLl6 mice were injected intrahepatically with TC-I tumor at 1 XI@ cells1 mouse and immunized with culture medium or vat-wt, vacE7, vat-SiglE7lLAMP-1 at I X10’ PFUslmouse 7 days later. Mice were sacrificed 30 days after vaccination to determine the presence or absence of tumors within the liver.
native approach to treat tumors in the liver. Though a number of strategies of cancer gene therapy have been demonstrated (22), the major limitation of most forms of cancer gene therapy is the lack of a systemic effect (23). The ideal cancer gene therapy should be able to generate systemic anti-tumor immunity in addition to a local anti-tumor effect. In this regard, immunotherapy is especially attractive because immunotherapy has the potency to eradicate systemic tumor in multiple sites in the body and the specificity to discriminate between neoplastic and non-neoplastic cells. Various strategies have been applied to experimental cancer immunotherapy (for review, see (24)). Some of these immunotherapy strategies have been applied to treat primary or metastatic liver tumors. Cytokine gene therapy (25-31) and dendritic cell-based therapy (32,33) can lead to either substantial hepatic tumor regression and/or induce an effective systemic anti-tumor immunity in the host and to prolong the median survival time of the treated animals. However, all of the above studies used whole tumor cells or whole cell lysates to present the full spectrum of uncharacterized tumor-specific antigens. In other words, these strategies are not antigen-specific forms of immunotherapy. Antigen-specific immunotherapy represents a more desirable approach for controlling tumors. Antigenspecific immunotherapy is less likely to generate non-
the vac_Sig/E7/LAMP- 1-vaccinated mice compared to vat-E7-vaccinated mice (7).
Discussion In this study, we demonstrated that vaccination with recombinant vaccinia vat-Sig/E7lLAMP-1 induces potent E7-specific anti-tumor immunity against E7-expressing tumors in the liver. Furthermore, the numbers of E7-specific CD8+ T cell precursors detected by ELISPOT directly correlated with the anti-tumor effect generated by vat-Sig/E7/LAMP-1. We focused on the liver in the present study because primary hepatocellular carcinoma is one of the most common fatal cancers (20) and the liver is the most common site of cancer metastases (21). Treatment of primary or metastatic tumors in the liver using conventional therapy, such as surgery, chemotherapy, radiotherapy and transcatheter arterial embolization has met with limited success. Gene therapy can be an alter-
Fig. 4. Representative gross photographs of the tumors in the liver in each vaccinated group. C57BLl6 mice were injected intrahepatically with TC-1 tumor at 1 Xl@ cells1 mouse and immunized with culture medium or vat-wt, vacE7, vat-SiglE7lLAMP-1 at 1X107 PFUslmouse 7 days later. Mice were sacrificed at day 30 after vaccination. Though the liver tumor size shown here is smaller in the vat-E7-vaccinated mice, there is no statistically signtjicance between the liver sizes in culture medium, vat-wt, and vacE7-vaccinated mice. No hepatic tumors were seen in vacSigIE7ILAMP-I-vaccinated mice.
95
C-H.
A
Chen et al.
240 q
without peptide
m with peptide
160
80
Control
Vat-wt
Vat-E7 SiglE7iLl
vat-SiglE7lLAMP-1 Splenocytes were harvested IO days after vaccination. The number of IFN-y-producing E7-speciJic CD8+ T cell precursors was determined using the ELISPOT assay (See texts for the detailed method). Statistical signiJicance and error bars were generated using ANOVA, with a p-value 4.0002. (A). The mean spot numbers of triplicates t SE in each vaccinated group,. (B). Representative ELISPOT pictures in each vaccinated group.
specific autoimmunity, and it provides more flexibility controlling the amount of antigen administered and the methods of antigen presentation to the immune system. In addition, one can correlate the clinical outcome to a specific immune response (34). To date. few studies directly address the issue of using antigen-specific immunotherapy, especially active immunization, for 96
tumors in the liver. The use of antigen-specific immunotherapy for treatment of tumors in the liver has previously been applied in the form of antibody-cytokine fusion proteins. For example, several studies in animals demonstrated that targeting delivery of interleukin-2 with antibody to tumor-specific antigens can elicit an effective cellular immune response against hepatic metastases (35-37). Alternatively, effective tumor vaccines can be generated by in vitro modification of tumor cells with cytokines and bispecific monoclonal antibodies that bind antigen on tumor cells to CD28 on T cells (38). Our current study used a vectorbased vaccine to deliver active antigen-specific immunization in order to control tumors in the liver. Though direct inoculation of cancer cells into the liver is not a true liver metastasis model, our model is an ideal model for investigating antigen-specific immunotherapy for tumors in the liver. We showed that the vacSig/E7/LAMP-1 could prevent hepatic tumor growth and eliminate established hepatic tumors. Our demonstration of the effectiveness of active antigen-specific immunization for the treatment of tumors in the liver is an important finding because several studies indirectly implied that the efficacy of immunotherapy might differ between subcutaneous and hepatic tumors (39). For example. Ganss et al. demonstrated that a strong tumor-specific antigen and the generation of CTL are not sufficient for the rejection of established tumors. The tumor environment, such as tumor microvasculature, is also a critical parameter that influences the effectiveness of activated anti-tumor lymphocytes (39). Furthermore, the metastatic cancer cells in the liver exhibit a higher percentage of Fas ligand expression than their primary counterparts, and thus these metastatic cancers cells may be able to evade immune destruction by inducing apoptosis of activated T lymphocytes (40). Our results are also consistent with the concept that the anti-tumor effects of the immune system are mainly mediated by cellular immunity, especially T cells. Activated T cells may function directly as effector cells, providing anti-tumor immunity through the lysis of tumor cells or through the release of cytokines capable of interfering with the propagation of tumors. Though the chromium release assay is a standard assay for cellular immunity, the ELISPOT assay is more sensitive and permits the determination of the number of antigen-specific T cells (16). Using the ELISPOT assay, we demonstrated that the frequency of E7-specific CDS ‘T cell precursors directly correlated with the anti-tumor effect generated by vat-Sig/E7/LAMP-1. The highest spot numbers were found in the vat-Sig/E7/ LAMP-l -vaccinated group. This finding is consistent
Antigen specific immunotherapy for liver tumors
with the CTL assay (7), in which the highest specific lysis was also found in the vat-Sig/E7/LAMP-l-vaccinated mice. Our previous study demonstrates that the anti-tumor immunity generated by vac_Sig/E7/LAMP1 is dependent on CD4+ and CDS+ T cells, as well as NK cells (10). We also observed that the average tumor size at day 30 after inoculation was larger in the control mice of this tumor regression experiment (2456 mm3) (Fig. 3) compared to those in the tumor protection experiment (423 mm3) (Fig. 2). It should be clarified that the endpoint of the tumor protection experiment is “day 30 after tumor challenge”, while the endpoint of the tumor regression experiment is “day 30 after vaccination”, or 37 days after tumor challenge (vaccination was given 1 week after tumor challenge). Therefore, the additional 7 days may account for the difference we observed between the tumor sizes of the control groups in the in vivo protection and in vivo treatment experiments. Active antigen-specific immunotherapy may have some potential pitfalls. Romieu et al. demonstrated that passive immunotherapy (adoptive transfer of selftumor antigen-specific CTLs), but not active immunization with peptide-based vaccine, is effective treating transgenic mice that harbored a progressive liver tumor associated with the expression of the SV40 large tumor T oncoprotein (41). This suggests that passive therapies targeted to self-tumor antigen may be more suitable than active immunization in the treatment of spontaneous tumors (41). The drawback of adoptive immunotherapy is the lack of reproducible and efficient approaches to the isolation and expansion of antigenspecific T cells (42). Safety issues regarding the use of the vaccinia virus as a vector must be considered. Using vaccinia virus to mediate gene transfer has the advantages of high efficiency of infection and high levels of recombinant gene expression (43). However, the highly attenuated vaccinia is still replication-efficient. Several clinical studies have shown that vaccinia vaccines are safe vectors in clinical trials in immunocompetent patients. A recombinant vaccinia vaccine encoding HPV-16 E6/E7 has been used for a phase I clinical trial in cervical cancer patients (44). Furthermore, a carcinoembryonic antigen-expressing recombinant vaccinia vaccine has been used in patients with advanced colorectal cancer (45). More recently, a Plasmodium falciparum malaria gene-expressing recombinant vaccinia vaccine is in a phase I/II clinical trial (46). No significant complications or environmental spread of vaccinia vaccines was noted in these trials. The recombinant vaccinia vaccines perhaps cannot be multiply administered be-
cause pre-exposure to vaccinia may decrease the immunogenicity of subsequent vaccinations with vaccinia (47). However, the barrier caused by preexisting vaccinia immunity can potentially be overcome by boosting with recombinant avian pox virus (48) or by administering the recombinant vaccinia via a mucosal vaccination route (49). In summary, our data suggest that vat-Sig/E7/ LAMP-l is an effective vaccine for controlling E7-expressing tumors in the liver and the number of E7specific CD8+ T cell precursors directly correlated with the anti-tumor effect generated by Sig/E7/LAMP-1 vaccinia. Though our hepatic tumor models were designed for HPV-16 E7 antigen-containing tumors, similar strategies can potentially be employed to the other cancer systems with known tumor-specific antigens. As more tumor-specific antigens are characterized, due to recent advances in molecular immunology (4), antigenspecific vaccination in treating tumors of the liver becomes more promising.
Acknowledgements We would like to thank Drs. Keerti V Shah and Robert J. Kurman for helpful discussions. We would also like to thank Drs. Ralph H. Hruban and Richard Roden for critical review of the manuscript and Morris Ling for preparation of the manuscript. This work was supported by NIH 5 pol 34582-01, U19 CA72108-02, ROl CA7263 l-01, the Richard W. TeLinde fund, and the Alexander and Margaret Stewart Trust grant.
References 1. Pardoll DM. Cancer vaccines. Nat Med 1998; 4: 525-31. 2. Greten TF, Jaffee EM. Cancer vaccines. J Clin Oncol 1999; 17: 1047-60. based on the 3. Rosenberg SA. A new era for cancer immunotherapy genes that encode cancer antigens. Immunity 1999; 10: 281-7. 4. Van den Eynde BJ, van der Bruggen F? T ccl1 defined tumor antigens. Curr Opin lmmunol 1997; 9: 68493. SA. Cancer vaccines based on the identification of 5. Rosenberg genes encoding cancer regression antigens. Immunol Today 1997; 18: 175-82. 6. Pardoll DM, Topalian SL. The role of CD4+ T cell responses in anti-tumor immunity. Curr Opin Immunol 1998; 10: 588-94. I. Wu TC, Guarnieri FG, Staveley OCKF, Viscidi RP Levitsky HI, Hedrick L, et al. Engineering an intracellular pathway for major histocompatibility 8.
9.
10
11
C.-H. Chen et al.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21. 22. 23.
24. 25.
26.
27.
28.
29.
30.
98
Antigen-specific immunotherapy for murine lung metastatic tumors expressing human papillomavirus type 16 E7 oncoprotein. Int J Cancer 1998; 78: 41-5. Gilboa E. Murine models for cancer immunotherapy using cytokine gene modified tumor vaccines. In: Forni G, Foa R, Santoni A, Frati L, editors. Cytokine-Induced Tumor Immunogenicity. London: Academic Press; 1994. p. 130-43. Fidler IJ, Wilmanns C, Staroselsky A, Radinsky R, Dong Z, Fau D. Modulation of tumor cell response to chemotherapy by the organ environment. Cancer Metastasis Rev 1994; 13: 209-22. Fukumura D, Yuan F, Monsky WL, Chen Y, Jain RK. Effect of host microenvironment on the microcirculation of human colon adenocarcinoma. Am J Path01 1997; 151: 679-88. Nannmark U, Johansson BR, Bryant JL, Unger ML, Hokland ME, Goldfarb RH, et al. Microvessel origin and distribution in pulmonary metastases of B16 melanoma: implication for adoptive immunotherapy. Cancer Res 1995; 55: 4627-32. Miyahira Y, Murata K. Rodriguez D, Rodriguez JR, Esteban M, Rodrigues MM, et al. Quantification of antigen specific CDI+ T cells using an ELISPOT assay. J Immunol Methods 1995; 181: 45554. Murali-Krishna K, Altman JD, Suresh M, Sourdive DJ, Zajac AJ, Miller JD, et al. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 1998; 8: 177.-87. Feltkamp MC, Smits HL, Vierboom MP, Minnaar RP, de JB, Drijlhout JW, ter SJ, Melief CJ, Kast WM. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur J Immunol 1993; 23: 2242-9. McCutcheon M, Wehner N, Wensky A, Kushner M, Doan S, Hsiao L, et al. A sensitive ELISPOT assay to detect low-frequency human T lymphocytes. J Immunol Methods 1997; 210: 149966. Okuda K. Hepatocellular carcinoma: recent progress. Hepatology 1992; 15: 948-63. Lygidakis NJ, Pearl A. Metastatic liver disease-a review. Hepatogastroenterology 1997; 44: 1484-7. Ruiz J, Qian C, Prieto J. Gene therapy of liver tumors: principles and applications. Digestion 1998; 59 Suppl 2: 92-6. Schuster MJ, Wu GY. Gene therapy for hepatocellular carcinoma: progress but many stones yet unturned! Gastroenterology 1997; 112: 6569. Chen CH, Wu TC. Experimental vaccine strategies for cancer immunotherapy. J Biomed Sci 1998; 5: 231-52. Caruso M, Pham-Nguyen K, Kwong YL, Xu B, Kosai KI, Finegold M, et al. Adenovirus-mediated interleukin-12 gene therapy for metastatic colon carcinoma. Proc Nat1 Acad Sci USA 1996; 93: 11302-6. Shiratori Y, Kanai F, Hikiba Y, Moriyama H, Hamada H, Matsumura M, et al. Gene therapy for hepatic micrometastasis of murine colon carcinoma. J Hepatol 1998; 28: 88695. Siders WM, Wright PW, Hixon JA, Alvord WG, Back TC, Wiltrout RH, et al. T cell- and NK cell-independent inhibition of hepatic metastases by systemic administration of an IL-12-expressing recombinant adenovirus. J Immunoll998; 160: 5465574. Cao G, Kuriyama S, Du P, Sakamoto T, Kong X, Masui K, et al. Complete regression of established murine hepatocellular carcinoma by in viva tumor necrosis factor alpha gene transfer. Gastroenterology 1997; 112: 501-10. Atarashi Y, Yasumura S, Nambu S, Yoshio Y, Murakami J, Takahara T, et al. A novel human tumor necrosis factor alfa mutein, F4614, inhibits in vitro and in viva growth of murine and human hepatoma: implication for immunotherapy of human hepatocellular carcinoma. Hepatology 1998; 28: 57-67. Karpoff HM, D’Angelica M, Blair S, Brownlee MD, Federoff H, Fong Y. Prevention of hepatic tumor metastases in rats with herpes viral vaccines and gamma-interferon. J Clin Invest 1997; 99: 7999804.
31. Yoshida H, Enomoto H, Tagawa M, Takenaga K, Tasaki K, Nakagawara A, et al. Impaired tumorigenicity and decreased liver metastasis of murine neuroblastoma cells engineered to secrete interleukin-2 or granulocyte macrophage colony-stimulating factor. Cancer Gene Ther 1998; 5: 67-73. 32. Peron JM, Esche C, Subbotin VM, Maliszewski C, Lotze MT. Shurin MR. FLT3-ligand administration inhibits liver metastases: role of NK cells. J Immunol 1998; 161: 6164-70. 33. DeMatos P, Abdel-Wahab Z, Vervaert C, Seigler HF. Vaccination with dendritic cells inhibits the growth of hepatic metastases in B6 mice. Cell Immunol 1998; 185: 65-74. 34. Sznol M, Holmlund J. Antigen-specific agents in development [review]. Semin Oncol 1997; 24: 173-86. 35. Becker JC, Pancook JD, Gillies SD, Furukawa K, Reisfeld RA. T cell-mediated eradication of murine metastatic melanoma induced by targeted interleukin 2 therapy. J Exp Med 1996; 183: 2361-6. 36. Xiang R, Lode HN, Dolman CS, Dreier T, Varki NM, Qian X, et al. Elimination of established murine colon carcinoma metastases by antibody-interleukin 2 fusion protein therapy. Cancer Res 1997; 57: 4948855. 37. Lode HN, Xiang R, Dreier T, Varki NM, Gillies SD. Reisfeld RA. Natural killer cell-mediated eradication of neuroblastoma metastases to bone marrow by targeted interleukin-2 therapy. Blood 1998; 91: 170615. 38. Guo YJ, Che XY, Shen F Xie TP Ma J, Wang XN. et al. Effective tumor vaccines generated by in vitro modification of tumor cells with cytokines and bispecific monoclonal antibodies. Nat Med 1997; 3: 451 -5. 39. Ganss R, Hanahan D. Tumor microenvironment can restrict the effectiveness of activated anti-tumor lymphocytes. Cancer Res 1998; 58: 4673381. 40. Shiraki K, Tsuji N, Shioda T, Isselbacher KJ. Takahashi H. Expression of Fas ligand in liver metastases of human colonic adenocarcinomas. Proc Nat1 Acad Sci U S A 1997; 94: 642@5. 41. Romieu R, Baratin M. Kayibanda M, Lacabanne V Ziol M, Guillet JG, et al. Passive but not active CD8+ T cell-based immunotherapy interferes with liver tumor progression in a transgenic mouse model. J Immunol 1998; 161: 5133-7. 42. Yee C, Riddell SR, Greenberg PD. Prospects for adoptive T cell therapy. Curr Opin Immunol 1997; 9: 702--8. 43. Moss B. Genetically engineered poxviruses for recombinant gene expression, vaccination. and safety. Proc Nat1 Acad Sci USA 1996; 93: 11341-8. 44. Borysiewicz LK, Fiander A. Nimako M. Man S, Wilkinson GW, Westmoreland D, et al. A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 1996; 347: 152337. 45. McAneny D, Ryan CA, Beazley RM, Kaufman HL. Results of a phase 1 trial of a recombinant vaccinia virus that expresses carcinoembryonic antigen in patients with advanced colorectal cancer. Ann Surg Oncol 1996; 3: 4955500. 46. Ockenhouse CF, Sun PF Lanar DE, Wellde BT, Hall BT, Kester K, et al. Phase l/Ha safety, immunogenicity. and efficacy trial of NYVAC-Pt7. a pox-vectored. multiantigen, multistage vaccine candidate for Plusmodium ,falcipurrm malaria. J Infect Dis 1998; 177: 1664-73. 47. Hanke T, Blanchard TJ, Schneider J, Hannan CM, Becker M, Gilbert SC, et al. Enhancement of MHC class I-restricted peptide-specific T cell induction by a DNA prime/MVA boost vaccination regime. Vaccine 1998; 16: 439.-45. 48. Hodge JW, McLaughlin JP, Kantor JA, Schlom J. Diversified prime and boost protocols using recombinant vaccinia virus and recombinant non-replicating avian pox virus to enhance T-cell immunity and anti-tumor responses. Vaccine 1997; 15: 759-68. 49. Belyakov IM, Moss B, Strober W, Berzofsky JA. Mucosal vaccination overcomes the barrier to recombinant vaccinia immunization caused by preexisting poxvirus immunity. Proc Nat1 Acad Sci USA 1999: 96: 4512-7.