Vaccination with dendritic cells pulsed with apoptotic tumors in combination with anti-OX40 and anti-4-1BB monoclonal antibodies induces T cell–mediated protective immunity in Her-2/neu transgenic mice
Her-2/neu is a transmembrane glycoprotein with tyrosine kinase activity whose structure is similar to the epidermal growth factor receptor.1 Amplification of the Her-2/neu gene has been reported in various types of cancer, including ovarian, gastric, colon, prostate and especially breast cancers.2, 3 Her-2/neu receptors play a role in the process of growth signal transduction across the cell membrane.4 Consequently, overexpression of this protein contributes to uncontrolled growth signal transduction and, therefore, cellular transformation.5 Overexpression of Her-2/neu is associated with metastatic disease, poor prognosis and low survival.6, 7 The facts that the level of Her-2/neu expression is a prognostic marker and that there have been reports indicating that cancer patients generate anti-Her-2/neu immune responses support the notion of using this protein as a target for T-cell immunotherapy. However, the efficacy of induction of tumor immunity after active immunization is limited by self-tolerance due to the fact that Her-2/neu is a self-antigen. We have previously demonstrated that Her-2/neu transgenic mice are tolerant and contain only a low-avidity repertoire against Her-2/neu antigens.7 Despite such tolerance, these low-avidity T cells have antitumor activity, indicating that tolerance is not absolute and that the residual repertoire of self-tumor antigens can be exploited for the induction of antitumor responses.
With the identification of TAAs, the use of a common method to induce tumor-specific immune responses is immunizing tumor-bearing hosts with DCs pulsed with TAA-derived epitopes.8, 9 However, peptide-based vaccines are limited by factors such as the identification of allele-specific epitopes and the fact that immunity against a single epitope might not be sufficient for effective induction of tumor immunity. Additionally, tumors may induce tumor antigen escape variants that might not be recognized by CTLs induced with the native peptide.10 Alternatively, the use of DCs pulsed with recombinant proteins or apoptotic tumor cells11, 12 could offer an opportunity to overcome restrictions posed by peptide vaccines. The key advantage of using DCs pulsed with proteins or tumor cells is that they provide a universal and unlimited source of antigens for vaccination, including the stimulation of class I– and class II–restricted T-cell responses without the need for predetermined allele-specific peptides, inducing generation of CTL responses with multiple specificities and broadening the T-cell responses against the tumor. However, use of proteins or tumor cells as a source of antigens has drawbacks, such as the production and purification of proteins or the accessibility to autologous tumor cells.
A productive T-cell response is contingent upon the transduction of several signals.13 Activation of T cells is mediated by the interaction of the TCR with MHC/peptide bimolecular complexes on APCs accompanied by the costimulation signal delivered by the interaction of B7/CD28 molecules.14 However, the activated T cell requires an additional signal from other accessory molecules in order to differentiate into effector cells, proliferate, expand and induce productive immune responses.15 In recent years, several molecules have been identified as having costimulatory functions in T-cell activation. Among these, members of the TNFR superfamily such as CD27, CD30, CD40, 4-1BB and OX40 have gained importance as costimulatory molecules delivering signals that prolong and propagate T-cell responses.16 In this regard, antibodies against 4-1BB and OX40 induce a vast amplification of T cell–mediated immune responses,17, 18 inhibit apoptotic cell death19, 20 and stimulate long-lived T-cell responses.21, 22 Also, anti-4-1BB and anti-OX40 MAbs induce immune responses that significantly reduce the growth of established tumors,23, 24 making these attractive targets for combined therapy with DCs.
In these studies, we compared the antitumor effect of DCs pulsed with recombinant Her-2/neu proteins and DCs pulsed with apoptotic tumor cells as a strategy for inducing an antitumor immune response in Her-2/neu mice. Given the potential benefit of anti-4-1BB and anti-OX40 MAbs in enhancing the immune responses, we also tested the effect of these costimulatory molecules in combination with DC-based vaccines in the Her-2/neu tumor model. Our results demonstrate that immunization with DCs pulsed with apoptotic tumor cells induced a more effective antitumor response in Her-2/neu mice than immunization with Her-2/neu-recombinant protein. Moreover, there is an additive effect from combining DC vaccination and anti-4-1BB and anti-OX40, resulting in significant tumor delay in Her-2/neu mice. These results suggest that the use of vaccines which stimulate a broad immune response in combination with costimulatory molecules to enhance the antitumor responses is critical for the optimization of vaccination strategies to overcome tolerance.
Her-2/neu transgenic mice (line 202) were commercially obtained from Jackson Laboratories (Bar Harbor, ME) and maintained homozygously. The N202.1A mammary cell line derived from a tumor from Her-2/neu transgenic mice was obtained from Dr. P.-L. Lollini (University of Bologna, Bologna, Italy).25 The N202-H2B-GFP cell line was generated by infecting the N202.1A cell line with H2B-GFP-LXRN vector, as previously described.26 All cell lines were maintained in complete RPMI-1640 supplemented with 10% FCS, 2 mM glutamine, 5 × 10−5 ME and 50 μg/ml gentamicin. All infected cells were selected in the presence of 1 mg/ml of G418. The embryonic 3T3-FVB fibroblast cell line was generated in our laboratory in the same manner as other embryonic 3T3 fibroblast cell lines were previously developed.27 Anti-OX40 (OX86) MAb was obtained from the European Cell Culture Collection (Wiltshire, UK), and the anti-4-1BB (3H3) was obtained from Dr. R. Mittler (Emory University, Atlanta, GA).
Generation of DCs
DCs were derived from bone marrow as described by Inaba et al.28 Briefly, bone marrow cells were depleted of lymphocytes utilizing magnetic beads conjugated with antibodies against CD4, CD8 and B220 (Dynal, Oslo, Norway). The remaining cells were cultured in complete RPMI medium containing 3% GM-CSF (supernatant from J558L cells transfected with murine GM-CSF gene; obtained from Dr. R. Steinman, Rockefeller University, New York, NY). Medium was changed every second day, each time applying fresh complete RPMI medium containing 3% GM-CSF. On day 8, DCs were collected.
Apoptosis induction of tumor cells
N202.1A cells were exposed to UVB irradiation (2 mj/cm2/sec) and incubated for 18 hr in medium without serum at 37°C for the induction of apoptosis. DCs and apoptotic cells were collected and cocultured in complete RPMI medium containing 3% GM-CSF overnight at a ratio of 2 apoptotic cells per 1 DC (5 × 105 DCs/well).
Production and purification of sneu protein
sneu protein was produced utilizing a baculovirus expression system. The extracellular portion of the neu gene (nucleotides 89–1988)29 was amplified by RT-PCR from N202.1A cells. The PCR product was inserted into the pFastBac1 vector (Invitrogen, Carlsbad, CA), which was modified by inserting the honeybee melittin secretion signal followed by a 6× histidine affinity tag sequence and a Factor Xa cleavage site. The pFastBac1-neu vector was used to transform the DH10Bac cell (Invitrogen), and recombinant bacmids were analyzed by PCR. Sf9 insect cells (Invitrogen) were transfected with a positive recombinant neu-bacmid. After 72 hr, posttransfection supernatant was analyzed for the presence of recombinant sneu protein by dot blot with anti-Her-2/neu antibodies against the extracellular portion produced in the laboratory (J.L., unpublished results). Viruses from Sf9 cells producing sneu were then harvested to infect High-Five cells (Invitrogen), to produce significantly higher levels of recombinant protein. Supernatant from High Five cells was collected and passed through an Ni-NTA column to purify the sneu protein. Afterward, the protein was cleaved with Factor Xa, and the sample was collected by gel filtration. DCs were pulsed with 50 μg/ml of sneu protein overnight. Maturation of pulsed DCs was confirmed by FACS analysis, evaluating the levels of class I, class II, CD80 and CD86 molecules. As shown in Figure 1, there were no differences in the level of expression of these molecules after pulsing DCs with apoptotic N202.1A cells or sneu protein.
Generation of CTL bulk cultures and cytotoxic activity
Her-2/neu mice were immunized s.c. with 106 DCs pulsed with apoptotic N202.1A tumors or with DCs pulsed with sneu. Two weeks later, responder spleen cells from primed animals were restimulated in vitro with DCs pulsed with apoptotic N202.1A cells or DCs pulsed with sneu for 5 days. Cytotoxic activity was measured in a standard 51Cr-release assay. N202.1A and 3T3 FVB target cells were incubated with 150 μ Ci of 51Cr sodium chromate for 1 hr at 37°C. Cells were washed 3 times and resuspended in complete RPMI medium. For the assay, 51Cr-labeled target cells (104) were incubated with varying concentrations of effector cells in a final volume of 200 μl in U-bottomed 96-well microtiter plates. Supernatants were recovered after 6 hr of incubation at 37°C, and the percent of lysis was determined by the following formula: percent specific lysis = 100 × (experimental release – spontaneous release)/(maximum release – spontaneous release).
Analysis of tumor growth inhibition by intravital microscopy
Dorsal skin chambers in the mouse were prepared as previously described.30 Female Her-2/neu mice (25–30 g body weight, around 10–12 weeks old) were anesthetized (7.3 mg ketamine hydrochloride and 2.3 mg xylazine/100 g body weight, i.p.) and placed on a heating pad. Two symmetrical titanium frames were implanted into a dorsal skinfold, to sandwich the extended double layer of skin. The underlying muscle (M. cutaneous max) and s.c. tissues were covered with a glass coverslip incorporated in one of the frames. After a recovery period of 3–4 days, tumor spheroids were carefully placed in the chamber.30 Vaccination with pulsed DC was executed as described above on days 4 and 11. Fluorescence microscopy was performed as previously described.30 The tumor spheroid was analyzed off-line from the captured images using photodensitometric computer software (IMAGE PRO-PLUS, Media-Cybernetics, Silver Springs, MD) to calculate tumor area and relative photometric density of the tumor.
s.c. tumor model
To evaluate the antitumor effect of DCs pulsed with apoptotic tumor or sneu with a larger tumor burden, Her-2/neu mice were implanted s.c. with 106 N202.1A tumor cells. Tumors were allowed to grow for 7 days before treatment was initiated. On day 7 after tumor inoculation, animals were randomly divided into groups of at least 5 mice/group and vaccinated 4 times s.c. with DCs pulsed with apoptotic N202.1A cells or sneu (with intervals of 10 days between each vaccination) in the presence or absence of anti-4-1BB and anti-OX40 MAbs. Anti-4-1BB and anti-OX40 MAbs were injected 2 days after each DC immunization (100 μg/injection). DC vaccinations were performed at the opposite site to where tumors were implanted. Tumor growth was monitored every 5 days, and growth rates were determined by caliper measurements in 2 diameters. Tumor volume was expressed as (minor diameter)2 × major diameter/2. Statistical analysis was determined by Student's t-test. In each experiment, at least 5 animals were included per group.
In vivo depletion of T lymphocytes
To confirm that the antitumor immune responses were T cell–mediated, anti-CD4 (GK1.5) and anti-CD8 (56-6.37) MAbs were used for in vivo depletion of T-cell subsets. Animals were injected i.p. with 300 μg of the respective antibody twice a week during the course of the experiment, starting 1 week prior to immunization with DCs. Depletion of T cells was confirmed by FACS analysis of spleens (Fig. 2).
Comparison of T-cell responses induced by DCs pulsed with apoptotic tumor cells vs. sneu
Previously, we demonstrated that Her-2/neu mice are tolerant and contain only a low-avidity T-cell repertoire to Her-2/neu antigens.7 To establish whether DCs pulsed with apoptotic N202.1A cells or sneu induce a T-cell response and whether vaccine potency differences exist between apoptotic tumor cells and sneu, Her-2/neu mice were immunized s.c. with 106 pulsed DCs. As a control, to assure the induction of a T-cell response, parental FVB-w.t mice were used. Two weeks postvaccination, spleens were excised and stimulated in vitro with DCs pulsed with sneu or N202.1A apoptotic cells. Vaccination with DCs pulsed with apoptotic tumor cells (Fig. 3a) or with sneu (Fig. 3b) induced a CTL response in Her-2/neu mice, albeit at much lower levels than in FVB mice. Cytotoxic activity is stronger in animals vaccinated with DCs pulsed with apoptotic tumor cells compared to sneu vaccination. No killing was detected against 3T3-FVB cells (used as a control for specificity), confirming that stimulated CTLs specifically recognized Her-2/neu or other antigens expressed on tumor cells. Since DCs have the ability to stimulate both CD8 and CD4 T-cell responses, we also evaluated the proliferation activity of CD4+ T cells. CD4+ T cells were enriched from FVB-w.t. or Her-2/neu mice immunized with DCs pulsed with apoptotic tumor cells (Fig. 3c) or with sneu (Fig. 3d). CD4 T-cell responses were measured in a proliferation assay against DCs pulsed with apoptotic N202.1A and apoptotic 3T3-FVB cells (used as a control). Our data demonstrated cytotoxic activity, indicating that immunization with DCs pulsed with N202.1A apoptotic cells (Fig. 3c) provides stronger proliferation than immunization with DCs pulsed with sneu (Fig. 3d). As expected, CD4+ T cells from FVB mice have a stronger proliferation capacity after in vitro stimulation than CD4+ T cells from Her-2/neu mice. Taken together, these results confirmed that the T-cell repertoire that exists in Her-2/neu mice against the tumor or Her-2/neu antigens differs from that in parental mice.
Immunization with pulsed DCs induces protective immunity in Her-2/neu mice
The preceding results demonstrate that although the T-cell repertoire of Her-2/neu mice is hyporesponsive to neu antigens, there was a residual repertoire that could be activated and could recognize N202.1A cells. To evaluate whether DC immunization would induce an immune response resulting in the rejection or delay of tumor growth, we first analyzed the antitumor responses utilizing a dorsal skinfold chamber tumor model. This system resembles a micrometastases tumor model where (i) there is implantation of a low load of tumor cells (5 × 104) and (ii) tumor growth can be evaluated in real time. To visualize the tumor, N202-H2B-GFP spheroids (5 × 104 cells) were implanted into the dorsal skinfold chambers. After 4 days, Her-2/neu mice were immunized s.c. with 106 DCs pulsed with apoptotic 3T3-FVB fibroblast cells, apoptotic N202.1A cells and sneu and boosted on day 11. As illustrated in Figure 4a, immunization with DCs pulsed with apoptotic N202.1A cells or sneu delayed tumor growth. Immunization with DCs pulsed with apoptotic N202.1A cells induced a stronger antitumor response than immunization with DCs pulsed with sneu. These results were in agreement with the cytotoxic activity and proliferation assay, indicating that immunization with DCs pulsed with apoptotic tumor cells was more effective than that with DCs pulsed with sneu. Measurements of tumor growth (Fig. 4b) indicated that there was a significant tumor delay compared to animals immunized with DCs pulsed with apoptotic 3T3-FVB cells, which showed no antitumor effect. These results confirm that the immune response induced by DCs pulsed with apoptotic N202.1A cells or sneu was antigen-specific.
Having demonstrated that vaccination with DCs pulsed with apoptotic N202.1A cells or sneu induced an antitumor response in the micrometastases model and substantially inhibited tumor growth, it was important to evaluate whether the same immunotherapeutic intervention was effective with a larger tumor burden. We compared the effectiveness of DCs pulsed with apoptotic tumors or sneu to induce an antitumor response. Her-2/neu mice were implanted with 106 N202.1A cells, and the tumor was allowed to grow for 7 days. Animals were immunized a total of 4 times, each time with 106 pulsed DCs, with an interval of 10 days between each immunization. Immunizations consisted of DCs pulsed with apoptotic 3T3-FVB fibroblasts, apoptotic N202.1A cells or sneu. For comparison, we included a group of animals immunized only with apoptotic-N202.1A cells. As shown in Figure 5, animals immunized with DCs pulsed with apoptotic tumors showed a stronger anti-tumor response (approx. 30% tumor growth inhibition) compared to animals immunized with DCs pulsed with sneu (approx. 18% tumor growth inhibition). These results confirmed that immunization with apoptotic tumor cells induced a more efficient antitumor immune response in Her-2/neu mice than immunization with Her-2/neu antigens. Mice immunized with apoptotic N202.1A cells showed minimal protection (approx. 8% tumor growth inhibition), while animals immunized with DCs pulsed with apoptotic 3T3-FVB cells showed no protection. We confirmed that the antitumor immune responses were dependent on CD4+ and CD8+ T cells since treatment with anti-CD4 or anti-CD8 MAb abrogated the immune protection. It is important to note that the antitumor effect after DC vaccination in the micrometastases model (Fig. 4) with a lower tumor load (5 × 104) is more effective compared to DC vaccination in the s.c. model (Fig. 5) with a larger tumor load (1 × 106). These results suggest that the tumor burden could be a critical factor in determining the effectiveness of the antitumor immune response in tolerant hosts.
Immunization with apoptotic tumors induces an immune response to neu and other antigenic determinants
Our results indicate that DCs pulsed with apoptotic tumors induce a stronger immune response than DCs pulsed with Her-2/nneu antigens. The difference in responses between sneu and apoptotic tumors could be attributed to several factors. For instance, sneu is composed of the extracellular portion only, while in the N202.1A cells there is expression of the full length of the Her-2/neu protein. Additionally, N202.1A cells might express antigens other than Her-2/neu that could stimulate an antitumor immune response. To better understand the immune response induced by apoptotic N202.1A cells, we isolated a mutant cell line from N202.1A cells that lost expression of Her-2/neu (N202.1A-neu−) (Fig. 6a). Mice were immunized with DCs pulsed with apoptotic N202.1A or N202.1A-neu− cells, and stimulated cultures were evaluated in a proliferation assay against DCs pulsed with sneu, apoptotic N202.1A, N202.1A-neu− and 3T3-FVB cells. Analysis of the immune responses reveals that T cells from animals immunized with DCs pulsed with apoptotic N202.1A cells recognize DCs pulsed with sneu (Fig. 6b) in contrast to animals immunized with DCs pulsed with apoptotic N202.1A-neu− cells (Fig. 6c). T cells from animals immunized with DCs pulsed with apoptotic N202.1A or N202.1A-neu− cells recognized DCs pulsed with N202.1A or N202.1A-neu− cells but not DCs pulsed with apoptotic 3T3-FVB cells. Taken together, these results indicate that immunization with DCs pulsed with apoptotic N202.1A tumors induce an immune response to neu antigens and to other antigenic determinants expressed on tumor cells.
Combination of DC-based vaccine and anti-OX40/anti-4-1BB enhance the antitumor response in Her-2/neu mice
The preceding results demonstrate that immunization with DCs pulsed with apoptotic tumors was more effective at activating an antitumor response than DCs pulsed with Her-2/neu antigens. However, this response was not sufficient for the complete control of tumor growth. Previous studies have demonstrated that treatment of tumor-bearing mice with MAbs directed against OX40 or 4-1BB promote effective tumor immunity.23, 24 We next analyzed whether the combination of DC immunization and anti-OX40/anti-4-1BB would improve the immunotherapeutic efficacy, resulting in a more effective antitumor response. We tested the combination of DC immunization with each of the antibodies alone or with the combination of both anti-OX40 and anti-4-1BB MAbs. Antitumor responses were analyzed in both the micrometastases and s.c. tumor models. Her-2/neu mice with implanted dorsal skinfold chamber were inoculated with N202-H2B-GFP cells (5 × 104). On days 4 and 11, animals were immunized with DCs pulsed with apoptotic N202.1A cells. As a control for the specificity of the vaccine, we included animals immunized with DCs pulsed with apoptotic 3T3-FVB cells. Anti-OX40, anti-4-1BB or anti-OX40 plus anti-4-1BB MAb was applied 2 days after each immunization. As illustrated in the photomicrographs (Fig. 7a) and measurement of tumor growth (Fig. 7b), DC immunization plus anti-OX40 or anti-4-1BB MAb further enhanced the antitumor response compared to animals that received vaccination with DCs pulsed with apoptotic N202.1A cells alone. Interestingly, DC immunization plus anti-OX40/anti-4-1BB MAb showed a significant antitumor response that completely eradicated tumors in the micrometastases model. Minimal antitumor response was observed in animals immunized with DCs pulsed with apoptotic 3T3-FVB cells plus anti-OX40 and anti-4-1BB, indicating that the specific responses induced by DC apoptotic N202.1A cells was specific and significantly enhanced in the presence of these costimulatory antibodies.
Next, we tested the combination of DC immunization and anti-OX40 and anti-4-1BB MAbs with a larger tumor burden utilizing the s.c tumor model. Her-2/neu mice were inoculated on day 0 with 106 N202.1A cells, and on day 7 animals received the first immunization with DCs pulsed with apoptotic N202.1A cells. Animals were immunized every 10 days for a total of 4 times. Anti-OX40, anti-4-1BB and anti-OX40/anti-4-1BB were injected 2 days after each immunization. As shown in Figure 8a, animals immunized with DCs pulsed with apoptotic tumor cells alone or plus control antibody showed approximately 25% tumor growth inhibition. Animals immunized with DCs plus apoptotic N202.1A cells plus anti-OX40 or anti-4-1BB showed approximately 45–50% tumor growth inhibition, while animals that received DC immunization plus both antibodies demonstrated approximately 70% tumor growth inhibition. Animals immunized with DCs pulsed with sneu demonstrated approximately 15–18% tumor growth inhibition, while in the presence of anti-OX40 or anti-4-1BB MAb a stronger antitumor response was observed, resulting in approximately 25–32% tumor growth inhibition. DC sneu immunizations plus dual costimulation significantly enhanced antitumor responses, resulting in approximately 40% tumor growth inhibition (Fig. 8b). Mice immunized with DCs plus apoptotic 3T3-FVB cells plus both antibodies demonstrated 15% tumor growth inhibition. Taken together, these results demonstrated that specific DC plus apoptotic N202.1A cell or DC plus sneu immunizations plus costimulation yielded a cooperative effect, enhancing antitumor immune responses in Her-2/neu mice.
Anti-OX40 and anti-4-1BB enhance T-cell responses in Her-2/neu mice
To determine how anti-OX40 and anti-4-1BB MAbs influence antitumor responses, we evaluated the effect that these antibodies have on cytotoxic and proliferative activities. Her-2/neu mice were immunized s.c. with 106 DCs with apoptotic N202.1A cells in the presence or absence of anti-OX40 and anti-4-1BB MAbs. Two weeks postvaccination, spleens were excised and stimulated in vitro with DCs pulsed with N202.1A apoptotic cells. As expected, animals immunized in the presence of the costimulatory antibodies demonstrated higher cytotoxic (Fig. 9a) and proliferative (Fig. 9b) activities against N202.1A cells than animals that did not receive antibodies. The lytic or proliferative activity was higher in animals that received both anti-OX40 and anti-41BB MAbs than in animals treated with either antibody alone (Fig. 9). Taken together, these results suggested that anti-OX40 and anti-4-1BB modulated the immune responses in Her-2/neu mice by stimulating and enhancing both CD4+ and CD8+ T cells.
In a previous study, we demonstrated that Her-2/neu mice have a residual low-avidity T-cell repertoire against Her-2/neu and that these CTLs could recognize and kill Her-2/neu+ tumors, albeit with a much lower intensity compared to nontransgenic mice.7 In the present study, we tested the antitumor potential of this residual T-cell repertoire to determine if it was possible to induce tumor protective immunity in Her-2/neu mice. We compared immunization with DCs pulsed with a soluble Her-2/neu protein and DCs pulsed with apoptotic tumors. The results demonstrate that immunization with both DCs pulsed with sneu and DCs pulsed with apoptotic tumor induced CTL responses in neu mice, albeit with significantly lower activity than in FVB mice. Having established that immunization with DCs pulsed with apoptotic tumor cells or sneu induced an immune response, we next evaluated its antitumor potential. Analysis of the antitumor responses in the micrometastases model indicated that immunization with DCs pulsed with sneu inhibited tumor growth by approximately 35%, while immunization with DCs pulsed with apoptotic cells further delayed tumor growth, resulting in approximately 50–60% inhibition of tumor growth. Similar results were observed in the s.c. tumor model. These results indicated that immunization with Her-2/neu antigens could induce protective immunity in Her-2/neu mice; however, immunization with DCs pulsed with apoptotic tumors was a more potent antigenic antitumor stimulant. It is important to note that the difference in the immune response between DCs pulsed with apoptotic tumor and DCs pulsed with sneu is not due to the maturation status of the DCs (Fig. 1). This difference in responses between sneu and apoptotic tumors could be attributed to several factors. For instance, sneu is composed of the extracellular portion only, while in the N202.1A cells there is expression of the full length of the Her-2/neu protein, in which epitopes from the intracellular portion could stimulate a T-cell response. Additionally, N202.1A cells express antigens other than Her-2/neu, including CD4 and CD8 antigenic determinants that are capable of stimulating an immune response directed against the tumor. Thus, immunization with DCs pulsed with apoptotic tumor cells might broaden the immune response against multiple antigens present on the tumor cells.
Previously, we demonstrated that to induce an immune response that retards tumor growth in Her-2/neu mice it is necessary to apply multiple immunizations.7, 30 In the s.c. tumor model, immunization 4 times with DCs pulsed with apoptotic tumor cells results in 30% tumor growth inhibition, while immunization with DCs pulsed with sneu results in 18% tumor growth inhibition. In addition, cumulative evidence has indicated that targeting costimulatory molecules from the TNFR family could enhance and augment the antitumor response.31, 32 We then tested whether DC immunization in the presence of anti-OX40 and anti-4-1BB would induce a stronger antitumor response. Although previous studies have demonstrated that, in C57BL/6 or DBA/2 tumor-bearing mice, treatment with anti-OX40 or anti-4-1BB alone is sufficient to eradicate the tumor, in sharp contrast, our results indicated that treatment of Her-2/neu-bearing tumor mice with these antibodies alone has a more modest effect. It has been hypothesized that the mechanism by which these antibodies induced rejection of the tumor was that inoculation of tumor cells alone induced a weak immune response that was strongly amplified in the presence of these costimulatory antibodies.33 We know that tolerance is not absolute and that different levels of tolerance may exist in the host.34 We demonstrated that Her-2/neu mice are already tolerant to Her-2/neu antigens prior to tumor onset. It is possible that the immune repertoire of Her-2/neu mice could be affected more profoundly by tolerance mechanisms than that of C57BL/6 or DBA/2 mice against tumor antigen(s) expressed on the tumors. Therefore, injection of N202.1A cells induced a very weak or no immune response, whereby this weak immune response could not be amplified effectively by OX40 or anti-4-1BB MAb. From our data, it appears that the interaction of APCs and T cells is necessary to promote priming and activation of T cells. Subsequently, the induced immune response is significantly enhanced by costimulation with anti-OX40 and anti-4-1BB MAbs. The therapeutic effect of the immunotherapy appeared to be limited by the tumor burden and the help provided by the costimulatory antibodies. In the micrometastases model, where a low load of tumor cells is implanted, DC immunization plus anti-OX40 and anti-4-1BB MAbs led to complete rejection of tumor. In the s.c. tumor model, immunization with DCs pulsed with apoptotic tumors and dual costimulation significantly enhanced the antitumor response, resulting in approximately 70% tumor growth inhibition, while immunization with DCs pulsed with sneu plus the antibodies resulted in approximately 40% tumor growth inhibition. Although these results indicate that it is possible to enhance the antitumor immune responses against self tumor antigens, they confirm that immunization with DCs pulsed with apoptotic tumors induce a stronger antitumor immune response. It is highly probable that immunotherapy alone will not be sufficient to completely eradicate tumors with a large tumor load in neu mice. However, if the immune repertoire for a self tumor antigen is properly stimulated and used in combination with another form of therapy, it could result in tumor eradication. In this regard, we previously evaluated the combination of immunotherapy and antiangiogenic therapy. Our results demonstrate that only when these therapies were combined did a superior antitumor response in Her-2/neu mice result.30 Furthermore, immunohistologic analysis of neu mice after immunotherapeutic intervention reveals no damage or accumulation of T-cell infiltrates in normal tissues such as kidneys or intestines, where low levels of Her-2/neu are expressed (data not shown). Thus, the observation that the low-affinity T cells that persist in a tolerant host do not induce an overt autoimmune response and recognize and kill tumor cells indicates that these cells could be an important component for the antitumor response.
Our data indicate that, for an effective antitumor response, the participation of CD4+ and CD8+ T cells was necessary. These results are in agreement with observations made by Melero et al.,35 which demonstrated that eradication of established tumors in vivo after treatment with anti-4-1BB MAb was mediated by both CD4+ and CD8+. However, these results were not in agreement with reports indicating that CD4+ T cells are not necessary for an antitumor response.36 It has been demonstrated that CD4+ T cells are not required for the generation of the initial CD8+ T-cell immune response but are critical in the generation of the CD8+ T-cell memory response. Cordaro et al.37 demonstrated that induction of an effective antitumor response against a self-antigen is achieved only when a memory low-avidity T-cell response is generated. We showed that to generate an immune response in Her-2/neu mice, it was necessary to apply multiple immunizations, which in turn might result in the generation of a memory response. By depleting the CD4+ T cells, we might prevent the generation of a low-avidity memory response, resulting in a diminished antitumor response. In addition, we cannot discard the possibility that low-avidity CD8+ T cells from Her-2/neu mice are more dependent on the help provided by CD4+ T cells to effectively initiate an immune response. Our results demonstrate that immunization with DCs pulsed with apoptotic N202.1A cells or with sneu alone induce a weak antitumor immune response in neu mice. However, immune responses are significantly enhanced in the presence of anti-OX40 and anti-4-1BB MAbs. Furthermore, dual costimulation with both anti-OX40 and anti-4-1BB MAbs further enhanced CD4+ and CD8+ T-cell responses, and a significantly greater effect with the use of both antibodies is observed in the antitumor immune responses. The effect of utilizing multiple DC immunizations and injections of both anti-OX40 and anti-4-1BB is that it expanded and prolonged the survival of CD4+ and CD8+ specific T-cell responses, resulting in a more effective antitumor response.7 In addition, signaling through OX40 can result in enhancement of therapeutic antitumor responses due to in vivo priming of Th1-type T-cell responses. These results were in agreement with our observation that the antitumor response induced in Her-2/neu mice was dependent on both CD4+ and CD8+ T cells.
Our results clearly indicate that the low-avidity T-cell repertoire from Her-2/neu mice is capable of mounting an antitumor response. Immunization with DCs pulsed with apoptotic tumor cells demonstrates this to be a powerful source for tumor antigens resulting in a more efficient antitumor response. However, the antitumor response in Her-2/neu was effective only as long as sufficient costimulation was provided. The combination of DC immunization and anti-OX40/anti-4-1BB led to complete tumor elimination in the micrometastases model and prolonged the survival of animals in the s.c. tumor model. Taken together, our data show that DCs pulsed with a larger source of antigens such as apoptotic tumors in combination with costimulatory signals may be a valuable strategy for the treatment of solid tumors in tolerant hosts.