Tumor Regression Induced by Intratumor Therapy with a Disabled Infectious Single Cycle (DISC) Herpes Simplex Virus (HSV) Vector, DISC/HSV/Murine Granulocyte-Macrophage Colony-Stimulating Factor, Correlates with Antigen-Specific Adaptive Immunity¹ Free

Major histocompatibility complex-restricted CTLs (CD8⁺) play an essential role in immunity against cancer, and are the most specific and functionally active effector cells (1, 2, 3, 4). Induction of T cell immunity against preselected T cell epitopes through peptide vaccination can induce protective antitumor immunity, but in some instances may lead to the induction of specific T cell tolerance resulting in progressive tumor growth (5); the presentation of MHC-restricted peptides together with the expression of appropriate costimulatory molecules expressed by dendritic cells (DC)³ can overcome tolerance and promote tumor rejection (6). Furthermore, the mode of tumor cell death may determine whether T cell activation or T cell tolerance occurs, and current evidence suggests that necrotic cell death provides the necessary danger signals for DC activation, whereas apoptotic death is usually passive, and fails to promote T cell immunity (7). Although DC take up apoptotic vesicles via specific receptors (8) and present antigenic peptides derived from them to CD8⁺ or CD4⁺ T cells, apoptotic bodies are poor immune activators (9). High numbers of apoptotic bodies are required for DC activation to occur (10). Necrotic cell death stimulates DC maturation (11); therefore, immunotherapies that preferentially promote or increase tumor cell death by necrosis are more likely to induce Ag-specific tumor immunity and rejection.

Genetically modified whole tumor cell vaccines can induce tumor immunity (12), and we have previously reported that tumor cells infected with a disabled infectious single-cycle HSV (DISC-HSV) encoding cytokines were more likely to die by necrosis than by apoptosis (13, 14). Mice with progressive tumors were vaccinated with irradiated tumor cells infected with DISC-HSV-2 carrying an expression construct for murine GM-CSF (mGM-CSF) or injected with DISC/mGM-CSF intratumorally; these treatments significantly reduced the incidence and growth of tumors (13, 14). Tumor rejection was shown to be dependent on the functionality of both CD4⁺ and CD8⁺ T cells, and by inference immunity would require the appropriate recruitment and activation of DC for the presentation of tumor Ag to T cells. In addition, several reports have demonstrated that the density of DC present within tumors correlates with prognosis (15) and that migration of DC from the vicinity of the tumor to the draining lymph nodes is essential for the induction of immunity. Furthermore, intratumor injection of bone marrow DC transduced with a retrovirus carrying cytokine genes promotes the regression of weakly immunogenic murine tumors (16).

The present study further investigates the immunological mechanism whereby DISC/mGM-CSF promotes tumor rejection; the selection of this virus is based on our previous studies demonstrating the efficacy of this viral vector in tumor immunotherapy (13, 14). We report in this study that tumor regression induced by DISC/mGM-CSF intratumor therapy correlates with a MHC class I-restricted peptide-specific CTL response toward the AH-1 peptide of the gp70 tumor Ag. Mice that fail to respond to therapy (tumor progressor) also fail to generate a prominent CTL response. Intratumor DISC/mGM-CSF therapy was also shown to restrict the growth and development of tumors at distant body sites (secondary s.c. and lung tumors). Tumor rejection was also induced in mice previously exposed to HSV-1 infection, indicating that pre-existing HSV immunity does not compromise immunotherapy. Intratumor therapy with DISC/mGM-CSF virus combined with in vitro-cultured murine bone marrow-derived DC enhanced the therapeutic effect of DISC/mGM-CSF.

Female BALB/c mice were purchased from Harlan Olac (Oxon, U.K.), and maintained in accordance with the Home Office Codes of Practice for the housing and care of animals.

A known murine leukemia virus gp70-derived H2-L^d-restricted peptide AH-1 (SPSYVYHQF, 138-147) and a control β-galactosidase H2-L^d-restricted peptide (TPHPARIGL, 877-88) were synthesized and used for in vitro assays.

The CT26 cell line is a N-nitroso-N-methylurethane-induced BALB/c murine colon carcinoma, maintained by serial in vitro passage in DMEM tissue culture medium supplemented with 2 mM l-glutamine and 10% FCS. This cell line was provided by I. Hart (Imperial Cancer Research Fund, London, U.K.).

The A20 BALB/c murine B cell lymphoma, RENCA (a BALB/c renal cell carcinoma line), and the NK-sensitive YAC-1 lymphoma were maintained by serial in vitro passage in RPMI 1640 tissue culture medium supplemented with 2 mM l-glutamine, 10% FCS, and 0.05 mM 2-ME.

Subcutaneous CT26 tumors were induced by the injection of 8 × 10⁴–1 × 10⁵ cells on the right flank; tumors were of 0.09–0.36-cm² diameters by 8–10 days. Ten animals per group were injected intratumorally with 1.25–2.5 × 10⁷ PFU of DISC/mGM-CSF or medium in a volume of 50 μl. In some experiments, virus injection was followed 3–5 h later by a further intratumor injection of 5 × 10⁵ DC in a volume of 50 μl serum-free RPMI medium. A second intratumor DISC/mGM-CSF injection was given 2 days later.

To assess the effect of intratumor therapy with DISC/mGM-CSF on distant tumor growth, a group of 10 mice was implanted s.c. with 1 × 10⁵ CT26 cells and injected intratumorally with 2.5 × 10⁷ PFU of DISC/mGM-CSF. These mice and 10 naive (control) mice were challenged immediately s.c. on the contralateral flank with 1 × 10⁴ CT26 cells. Additional experiments were conducted to determine the effect of intratumor therapy with DISC/mGM-CSF on experimental metastasis (metastasis to lungs following i.v. injection of tumor cells). One of the two groups carrying CT26 s.c. tumors was treated intratumorally with 2.5 × 10⁷ PFU of DISC/mGM-CSF; both groups, together with a group of five naive mice, were challenged at the same time i.v. with 1 × 10⁴ CT26 cells.

To evaluate the mRNA expression pattern of gp70 in murine cancer cell lines, total RNA was extracted from CT26, RENCA, and A20 tumor cells using CsCl-guanidine thiocyanate gradient method (Tel-Test, Friendswood, TX). Gene-specific oligonucleotide primers were designed to amplify cDNA segments with the estimated primer-melting temperature of 63°C. RT-PCR was performed using 18 amplification cycles in a thermal cycler (Biometra, Tampa, FL) at an annealing temperature of 58°C, and the products were analyzed by gel electrophoresis and visualized using ethidium bromide. The expression level of gp70 in CT26 cells was 3- and 5-fold higher than that of A20 and RENCA cells, respectively (results not shown).

To examine the effect of HSV-seropositive status on DISC/mGM-CSF therapy, 24 mice were immunized on three separate occasions, 2 wk apart i.p. with 2 × 10⁶ PFU of wild-type HSV-1; 24 mice were given PBS i.p. as controls. Seven days following the last immunization, all mice were bled and their sera assayed for the presence of Abs to HSV using ELISA. All animals were then implanted with 1 × 10⁵ CT26 tumor cells on the right flank, and intratumor therapy with DISC/mGM-CSF was performed, as detailed above. The midpoint titer of HSV-1-immunized mice ranged from 1 to 2 logs; four nonresponders were sacrificed and not included in the experiment.

DC were generated using a method adapted from Inaba et al. (17). Briefly, hind limbs (femurs and tibias) were harvested aseptically from female BALB/c mice and placed in sterile PBS supplemented with 50 IU/ml penicillin, 50 μg/ml streptomycin, and 0.25 μg/ml fungizone. The marrow was collected and gently resuspended to make a single cell suspension. The cells were washed twice in serum-free RPMI medium. The pellet was resuspended in DC medium (RPMI 1640 medium supplemented with 2 mM glutamine, 5% FCS, 10 mM HEPES, 20 mg/ml gentamicin sulfate, 50 μM 2-ME, 50 μg/ml penicillin, 50 μg/ml streptomycin, 0.25 μg/ml fungizone, and 20 ng/ml mGM-CSF) at a concentration of 10⁶ leukocytes/ml and seeded in 24-well plates at 1 ml/well and incubated at 37°C in a 5% CO₂ in air-humidified atmosphere. The nonadherent cells (T cells, B cells, granulocytes) were removed on days 2 and 4, and the remaining cells were cultured in fresh DC medium. Clusters of loosely adherent DC cultured for 7–9 days were dislodged by gently washing medium over each well using a Pasteur pipette. The cells were collected, washed twice, and resuspended in serum-free RPMI at 1 × 10⁷ cells/ml and stored on ice until required.

Spleens were harvested from tumor-bearer mice receiving therapy (test) and nontreated normal mice (control). Cells were flushed from the spleen (lymphocyte fraction) with serum-free RPMI using a 25-g needle and syringe. The remaining spleen tissue was cut into four pieces, digested with 2 ml enzyme cocktail (1.6 mg/ml collagenase and 0.1% DNase in serum-free medium; Sigma-Aldrich, Dorset, U.K.) at 37°C in 5% CO₂ in air-humidified atmosphere, for 60 min. The spleen tissue was then dissociated by gentle pipetting. The cells were collected, washed twice in serum-free medium, mixed with the lymphocyte fraction, suspended in CTL medium, and cultured at a concentration of 5 × 10⁶ cells/2 ml (RPMI 1640 supplemented with 1% glutamine, 10% FCS, 20 mM HEPES, 50 μM 2-ME, 50 U/ml penicillin, 50 μg/ml streptomycin, and 0.25 μg/ml fungizone) with 10 μM of the relevant peptide and added to 24-well plates at 2 ml/well for 5 days.

On day 5 of in vitro stimulation, splenocytes were harvested, washed twice in serum-free medium, resuspended in CTL medium, counted, and used as effector cells. Target cells were harvested by trypsinization, washed, and labeled with chromium-51. A standard 4-h Cr release assay was performed, and the percentage of specific cytotoxicity was determined using the following equation: the percentage of specific cytotoxicity = (experimental release − spontaneous release)/(maximum release − spontaneous release) × 100.

Flow cytometric analysis was performed on freshly isolated (day 0) and 7- and 9-day cultured bone marrow cells. Abs against CD11c (N418 hamster hybridoma; American Type Culture Collection, Manassas, VA), CD80, CD40, CD45R, CD4, CD8, MHC class II, and macrophage/monocyte Ags (Serotec, Oxford, U.K.) were used for staining, and cells were analyzed by flow cytometer. DC were chosen for use in therapy based on their expression of phenotypic markers on days 7–9 of in vitro culture. DC at days 7 and 9 expressed high levels of CD80, CD40, MHC II, and CD11c in comparison with DC at day 0, in which the expression of these markers was negligible (results not shown).

Splenocytes from immune and naive mice were prepared as above. Cells were plated with or without the relevant peptide (10 μM) at a concentration of 5 × 10⁶ cells/2 ml/well in 24-well plates. The plates were incubated at 37°C, in 5% CO₂ in air-humidified atmosphere. A total of 500 μl supernatant was collected from each well on day 5, and stored at −20°C until required for cytokine analysis. The level of IFN-γ and IL-4 was determined using ELISA kits (R&D Systems, Abington, U.K.), performed according to the manufacturer’s instructions.

Splenocytes from naive, CT26 tumor-bearer, regressor, and progressor mice were separated and prepared, as described above. A total of 2 × 10⁴ splenocytes in a volume of 100 μl CTL medium was cultured at 37°C in the presence or absence of 10 μg/ml Con A (Sigma-Aldrich) in 96-well plates for 3 days. Cells were then labeled with 5-bromo-2′-deoxyuridine (BrdU), and proliferation was assessed using a Boehringer Mannheim (Indianapolis, IN) ELISA kit.

The effect of depleting CD4⁺ and CD8⁺ T cells on the generation of cytotoxicity by in vitro stimulation of splenocytes from regressor and progressor mice with the AH-1 peptide was investigated. Splenocytes from a regressor mouse were harvested and prepared, as outlined above. Depletion using Dynal (Wirral, U.K.) beads was performed according to the manufacturer’s protocol: briefly, 4 × 10⁷ splenocytes were depleted of either CD4⁺ or CD8⁺ T cells using magnetic beads (Dynabeads mouse CD4 (L3T4) and Dynabeads mouse CD8 (Lyt-2) kits) or depleted of both CD4⁺ and CD8⁺ T cells. The remaining splenocytes were then stimulated with the AH-1 peptide in culture for 5 days (as described above). A fraction of the cells was stained with anti-CD4 and anti-CD8 Abs (BD PharMingen, San Diego, CA) and analyzed by flow cytometry. Depletion resulted in greater than 98% reduction in CD4⁺ and CD8⁺ T cells (data not shown).

DISC/HSV/mGM-CSF (dH2B) was constructed by recombination, as previously described (13, 14). The DISC/mGM-CSF virus (dH2B) is thymidine kinase negative, HSV envelop glycoprotein negative, and expresses mGM-CSF following infection of normal or complementing cells.

We have previously reported that direct injection of DISC-HSV into CT26 cutaneous tumors in BALB/c mice inhibited tumor growth and caused the complete regression of tumors in up to 70% of mice (13) (Fig. 2, A–D). In this study, we extend our study to investigate the effect of direct intratumor injection of DISC/mGM-CSF virus on the growth of tumor cells at distant body sites (s.c. tumors or experimental metastasis in the lungs). A total of 1 × 10⁵ CT26 tumor cells was implanted s.c. into the flank of BALB/c mice; 8–9 days later, 10 mice (with tumors of between 0.09- and 0.36-cm² surface area) were randomly selected and injected intratumorally with DISC/mGM-CSF. These mice together with five naive BALB/c mice were challenged s.c. on the contralateral flank with 1 × 10⁴ CT26 cells (×10 tumor dose 50%). Intratumor injection with DISC/mGM-CSF virus completely inhibited the development of CT26 tumors on the contralateral flank (Fig. 1,A), and significantly (p < 0.01) reduced the growth of experimental lung metastasis, initiated following the i.v. injection of 1 × 10⁴ CT26 tumor cells (Fig. 1 B). Four tumor foci were detected in the lungs of one of nine mice injected intratumorally with DISC/mGM-CSF virus, whereas all (eight of eight) mice injected intratumorally with medium showed detectable lung foci (with an average of 11 foci per mouse). All BALB/c mice with progressive CT26 s.c. tumors developed visible lung nodules (with an average of 30 per mouse). These results infer that intratumor injection of DISC/mGM-CSF virus prevents the development of lung foci compared with controls (media), although the intratumor injection of medium and/or the presence of primary tumors also appear to reduce the number of microscopically detectable lung tumor foci compared with naive control mice.

In a further series of experiments, the effect of increasing the number of DC at the primary tumor site was investigated. CT26 s.c. tumors were directly injected either with DISC/mGM-CSF, DISC/mGM-CSF, followed by DC, DC alone, or media. Fig. 2 shows the results obtained in four separate experiments. Tumors injected with 50 μl serum-free medium continued to progress, and by day 23–28 all mice developed 1 cm² or larger tumors and were sacrificed. Intratumor injection of DC alone resulted in a significant (see footnotes to Fig. 2 for statistical analysis of individual experiments) delay in tumor growth compared with control mice, but tumors progressed and by day 28 all the mice were sacrificed. Injection of DISC/mGM-CSF alone delayed tumor growth, and complete regression was observed in up to 70% of treated animals. When DISC/mGM-CSF injection was followed 5 h later by injection of DC, an increased number of mice regressed their tumors; combination therapy resulted in up to 100% of animals showing complete tumor regression (Table I; Fig. 2), and remained tumor free for up to 18 mo following therapy. These mice were resistant to further challenge with live parental tumor cells (data not shown).

An important consideration with regard to the use of viral vectors in the treatment of human disease is the presence of preexisting immunity to viral Ags. Human cancer therapy using DISC-HSV will necessitate inoculation of the vector into patients with immunological memory for natural HSV infection. Therefore, to establish whether previous exposure to HSV infection influenced the efficacy of intratumor therapy with the DISC/mGM-CSF virus s.c. CT26 tumors were established in mice previously exposed to HSV (see Materials and Methods for details), and the resulting tumors were injected intratumorally with DISC/mGM-CSF. Four groups of ten BALB/c mice implanted with CT26 tumors were used in this experiment: two groups of mice had previously been infected with HSV-1 (see Materials and Methods), and blood serum Ab titers to HSV-1 were determined before tumor induction (average titer was 1:6309, as determined by ELISA); the two remaining groups of mice were used as controls. HSV-1-infected and control mice were subsequently inoculated intratumorally with either DISC/mGM-CSF virus or medium. The results demonstrate that previous exposure to HSV-1 did not decrease the efficacy of DISC/mGM-CSF therapy (Fig. 3). A comparable reduction in tumor growth and incidence was observed in HSV-1 Ab-positive and Ab-negative (control) mice inoculated intratumorally with DISC/mGM-CSF, confirming that preexisting immunity to HSV does not reduce the therapeutic potency of DISC/mGM-CSF.

To explore whether CTL were generated in response to DISC/mGM-CSF intratumor therapy, experiments were performed to assess CTL activity to the H2-L^d-restricted AH-1 peptide of the gp70 tumor Ag, endogenously expressed by CT26 tumor cells. Splenocytes were harvested from mice with completely regressed tumors (regressors) and from mice failing to respond to DISC/mGM-CSF therapy (progressors). Splenocytes were restimulated in vitro with the AH-1 peptide (SPSYVYHQF) and assayed for cytolytic activity against CT26 cells, A20 cells pulsed with the AH-1 peptide, RENCA, and YAC-1 cells (see Materials and Methods for details of gp70 expression by cell lines and CTL restimulation). Spleen cells from all mice with regressed tumors (seven of seven) cultured in vitro in the presence of the AH-1 peptide demonstrated significant CTL activity against CT26 target cells (Fig. 4,A), and against A20 tumor cells pulsed with the AH-1 peptide (Fig. 4,A). The cytolytic activity (at 50:1, E:T ratio) of CTLs generated from mice with regressed tumors was 75% against CT26 cells and 86% against A20 target cells pulsed with the AH-1 peptide, respectively (mean of seven experiments; individual values are given in Table II). This contrasted with a reduced frequency and lower level of CTL activity against CT26 targets of spleen cells from mice with progressive tumors (failed to respond to DISC/mGM-CSF therapy, in which only two of seven mice responded to restimulation with the AH-1 peptide (at 50:1, E:T ratio) giving a mean cytolytic activity against CT26 targets of 15% (Table II, Fig. 4, B and C)). Similar results were obtained using A20 peptide-pulsed target cells. Regressor CTL activity was highly statistically significant at all E:T ratios (p < 0.05–0.001) compared with the relatively low level of significant killing attained with progressor splenocytes (see Fig. 4). CTL activity against A20 AH-1 peptide-pulsed target cells was abrogated by anti-H2 Ab and was shown to be peptide specific, because killing of A20 cells pulsed with an irrelevant H2-L^d-restricted peptide was not observed (results not given). In addition, no cytotoxicity was observed against YAC-1 lymphoma cells or an irrelevant BALB/c tumor target cell line (RENCA) not expressing gp70 (data not shown).

To analyze the involvement of T cell subsets in the generation of cytotoxic effectors, depletion of CD4⁺ or CD8⁺ or CD4⁺ and CD8⁺ T cells was performed on splenocytes from regressor mice before in vitro stimulation with the AH-1 peptide. Representatives are shown in Fig. 5, demonstrating that the depletion of CD8⁺ or CD4⁺ and CD8⁺ T cells completely abolished the cytotoxic activity. Depletion of CD4⁺ T cells reduced CTL activity by ∼30%, but did not completely abrogate killing. Similar results were obtained using A20 target cells pulsed with the relevant peptide. The failure of DISC/mGM-CSF therapy in congenitally T cell-deficient (Nu⁺/Nu⁺) mice further substantiates the involvement of T cells in promoting tumor rejection (results not shown).

The IFN-γ and IL-4 cytokine response of splenocytes harvested from regressor and progressor mice to in vitro stimulation with the AH-1 peptide was analyzed. Splenocytes, at a concentration of 5 × 10⁶ cells/2 ml/well, were cultured in the presence or absence of 10 μM peptide, and supernatants were collected at day 5 and assayed by ELISA. IFN-γ (range: 283–714 pg/ml) was detected in the supernatants harvested from splenocyte cultures established from mice responding to DISC/mGM-CSF therapy and restimulated with the AH-1 peptide (Table III), the levels of which did not necessarily correlate directly with the cytotoxic activity. In these experiments, progressor splenocytes produced lower levels of IFN-γ (0–131 pg/ml) and no detectable CTL activity. The levels of IFN-γ production by regressor lymphocytes were significantly higher than those of progressor splenocyte cultures (<0.001). Low levels of IL-4 were detected in regressor and progressor mice, reaching a maximum of 75 and 45 pg/ml IL-4, respectively.

To assess the general immune status of the mice given intratumor therapy with DISC/mGM-CSF, splenocytes from regressor and progressor mice were stimulated in vitro for 3 days with 10 μg/ml Con A, and cell proliferation was measured using a BrdU uptake ELISA. Splenocytes from tumor-bearer and naive mice were included as controls. The proliferation of splenocytes from naive mice (average of 10 experiments) cultured for 3 days in the presence of 10 μg/ml Con A was significantly greater (p < 0.01) than background; the OD increased from 0.365 to 0.787, respectively (Fig. 6). Comparable results were obtained with splenocytes from regressor mice (average of 12 experiments); the OD increased from 0.388 to 0.790 for cells cultured in medium and Con A, respectively. No significant proliferation in response to Con A stimulation above background levels was shown for splenocytes of tumor-bearer mice not receiving therapy (average of 9 experiments) or mice injected intratumorally with DISC/mGM-CSF, but developing progressive tumors (average of 8 experiments).

A variety of strategies to develop therapeutic cancer vaccines have been adopted, including immunization with peptides, recombinant proteins, DC pulsed with peptides or tumor cell lysates, and tumor cells transduced with recombinant viruses or plasmid DNAs encoding immune response genes (18, 19). Several viruses, including retroviruses, adenoviruses, adeno-associated virus, vaccinia, and HSV, have been genetically modified to express cytokines and costimulatory molecules and used as vectors in cancer therapy (13, 20, 21, 22). We previously described the construction of a HSV envelop glycoprotein-deleted HSV-2, which we termed DISC, to be used as a vaccine for the prevention of HSV-induced diseases (23). Further work has demonstrated the efficacy of DISC-HSV virus as a vector for gene transfection therapy of tumors in preclinical murine models (13, 14). This virus has many advantages over other vector systems, including the ability to express therapeutic genes for up to 72 h following infection and the efficient delivery of appropriate cytokines genes to tumor cells. Only a fraction (7%) of cells die by apoptosis following DISC-HSV infection (13, 14), suggesting that the majority of cells die by an alternative necrotic pathway, which is known to induce DC maturation, the expression of costimulatory molecules, and activation of both CD4⁺ and CD8⁺ T cells (24).

In the present study, injection of DISC/mGM-CSF virus into established tumors was shown to inhibit primary tumor growth, inducing complete tumor regression in up to 70% of treated mice, and to stem the development of distant lung metastasis and s.c. tumors; mice that had undergone tumor regression following DISC/mGM-CSF therapy remained immune to a rechallenge with live tumor cells (results not shown). Moreover, other researchers have shown that intratumor injection of the HSV vector G207 induces a prominent oncolytic antitumor effect in mice harboring N18 brain or s.c. tumors and a systemic antitumor immune response (25).

In general, although viral vectors are effective for gene transfer into tissues in vivo, their therapeutic application has several limitations. Expression of viral proteins may lead to irreversible tissue damage (26), and immunization with recombinant viruses usually elicits a strong humoral response and the production of neutralizing Abs that compromises the subsequent use of the virus (27, 28). In this study, the presence of neutralizing Abs to HSV-1 in mice receiving DISC/mGM-CSF therapy did not reduce the therapeutic effect, which was comparable with that observed in seronegative mice.

DC-based cancer vaccines, formulated by loading DC with antigenic peptides or tumor lysates or by genetically modifying DC before in vivo administration, represent a potentially powerful strategy for cancer immunotherapy. Genetically modified or peptide-pulsed DC can generate immunity to established tumors or tumor challenge (29, 30). In this study, we report the effect of combined intratumor injection of DISC/mGM-CSF and syngeneic bone marrow-derived DC on the growth of s.c. CT26 tumors. It was shown that direct injection of DISC/mGM-CSF virus into CT26 s.c. tumors, followed 4–6 h later by intratumor injection of bone marrow-derived DC significantly enhanced the therapeutic effect of DISC/mGM-CSF. Other studies have shown that the oncolytic activity of the replication-restricted HSV-1 virus can inhibit Lewis lung carcinoma growth in vivo in the absence of an adaptive immune response (31), and that DC injected intratumorally can cause the regression of MT-901 murine breast carcinomas (32), in which the level of tumor cell apoptosis was shown to be a determining factor in promoting tumor rejection.

Mechler et al. (33) have shown that the adoptive transfer of DC in both syngeneic and allogeneic murine tumor models generates effective systemic antitumor immunity, correlated with the intrinsic immunogenicity of tumor. It has also been demonstrated that host DC capture, process, and present tumor-associated Ags to naive T cells within the draining lymph nodes by cross-priming (34). The results of the present research suggest that providing additional mature DC at the tumor site is an effective way of enhancing antitumor immunity and rejection. Because APCs are often functionally impaired in patients with cancer (35), the adoptive transfer of ex vivo cultured, functionally mature DC represents a promising clinical approach. Melero et al. (36), using intratumor injection of DC engineered to secrete IL-12, have shown that tumor regression is associated with a detectable CTL response directed against tumor-specific Ags; in this model, successful therapy was strictly dependent on IL-12 expression by DC. The essential role of cytokines in modulating immune responses against tumors has been demonstrated both in vitro and in vivo (37), especially with regard to the preferential generation of either a Th1 or Th2 response. The results presented in this work for CT26 immunotherapy suggest that the response is driven toward Th1 cells.

Progressive tumor growth also compromises general immunocompetence. Only splenocytes from naive mice or mice responding to DISC/mGM-CSF therapy (regressor mice) were able to respond to in vitro stimulation with Con A, whereas splenocytes from progressor or tumor-bearer mice failed to respond. Evidence suggests a direct relationship between cancer progression and immune dysfunction (38, 39), which initially may be localized in or around the tumor, but which eventually becomes systemic (40). Unresponsiveness to Con A was clearly demonstrated in tumor-bearing mice, and defects in T cell and NK cell cytotoxic and proliferative responses and reduced levels of Th1 cytokine production do occur in mice with progressive primary tumors (41). Histological examination of regressor tumors showed that the rejection of CT26 tumors is accompanied by the infiltration of inflammatory cells (results not given), although CD8⁺ and CD4⁺ lymphocytes are essential for tumor rejection to occur (13). In addition to acting as an efficient vector for gene delivery, DISC-HSV also acts as an immune adjuvant, promoting Ag-specific antitumor immunity (13) and up-regulation of IL-12 gene expression (42). This induces a Th1 response and CTL effectors, breaking tolerance and/or energy.

We acknowledge Glenda Kill for preparing this manuscript, and Steve Reeder and Robert Davy for technical assistance.