The development of therapeutic strategies which induce effective cellular antitumor immunity represents an important goal in cancer immunology. Here, we used the unique features of the genetically engineered Hgf-Cdk4R24C mouse model to identify a combination chemoimmunotherapy for melanoma. These mice develop primary cutaneous melanomas which grow progressively and metastasize in the absence of immunogenic foreign proteins such as oncogenes or antigens. Primary and metastatic tumors evade innate and adaptive immune defenses, although they naturally express melanocytic antigens which can be recognized by antigen-specific T cells. We found that primary melanomas continued to grow despite infiltration with adoptively transferred, in vivo–activated, tumor-specific CD8+ T cells. To promote tumor immune defense, we developed a treatment protocol consisting of four complementary components: (a) chemotherapeutic preconditioning prior to (b) adoptive lymphocyte transfer and (c) viral vaccination followed by (d) adjuvant peritumoral injections of immunostimulatory nucleic acids. Lymphocyte ablation and innate antiviral immune stimulation cooperatively enhanced the expansion and the effector cell differentiation of adoptively transferred lymphocytes. The efficacy of the different treatment approaches converged in the tumor microenvironment and induced a strong cytotoxic inflammatory response enabling preferential recognition and destruction of melanoma cells. This combination chemoimmunotherapy caused complete regression of advanced primary melanomas in the skin and metastases in the lung with minimal autoimmune side effects. Our results in a clinically highly relevant experimental model provide a scientific rationale to evaluate similar strategies which unleash the power of innate and adaptive immune defense in future clinical trials. [Cancer Res 2009;69(15):6265–74]

Powerful new strategies for the antigen-specific immunotherapy of cancer have been developed in experiments with transplanted mouse tumor cells. CD8+ T lymphocytes capable of specifically recognizing and destroying tumor cells in vivo can be effectively stimulated using a number of different vaccines including synthetic MHC class I–binding tumor peptides, peptide-pulsed dendritic cells, recombinant viral vectors, and genetically modified tumor cells expressing immunostimulatory cytokines (14). However, despite encouraging results in experimental models, each of these vaccine approaches by itself showed only limited success in clinical trials (511). Subsequent studies in mice using TCR-transgenic mice and antigen-transfected transplantable tumor cell lines revealed that self-tolerance (12), ignorance (13), and tumor-induced immune suppression (1416) could prevent immune cell–mediated recognition and destruction of tumors and likely present barriers for effective cancer immunotherapy in patients.

In order to increase the efficacy of antigen-specific tumor immunotherapy, vaccines have been combined with other treatment modalities such as immunostimulatory monoclonal antibodies and toll-like receptor agonists (1720). An adoptive transfer of tumor-specific T lymphocytes also potently enhances antitumor immunity (21, 22). The identification of appropriate combination treatment protocols which can be successfully translated into the clinic has become one of the major challenges in the field. Experiments with transplanted mouse tumor cells have only been of limited predictive value. This is best illustrated by the fact that tumor vaccines with synthetic peptides and dendritic cells were highly effective against transplanted tumors in mice (1, 2), but not against primary tumors in patients (5, 6). Transplanted mouse tumor cell lines vary greatly in their immunogenicity and interact with the immune system in a different way when compared with primary tumors evolving autochthonously in an immunocompetent host.

To overcome the experimental limitations of tumor transplantation, we established the genetically engineered Hgf-Cdk4R24C mouse model for melanoma, a tumor entity of considerable interest to tumor immunologists (23). This experimental system imitates the expected clinical situation because individual Hgf-Cdk4R24C mice sporadically develop primary cutaneous melanomas due to the cooperative activity of deregulated receptor tyrosine kinase signaling and impaired p16-dependent cell cycle arrest, two characteristic pathogenetic features of human melanoma (2328). Here, we used the unique features of the Hgf-Cdk4R24C mouse melanoma model to develop a highly effective combination chemoimmunotherapy protocol. The sequential application of four complementary therapeutic approaches, i.e., chemotherapeutic preconditioning prior to adoptive lymphocyte transfer and viral vaccination followed by adjuvant peritumoral injections of immunostimulatory nucleic acids, was able to promote strong cytotoxic inflammation in the tumor microenvironment associated with complete regression of advanced primary melanomas in the skin, disappearance of metastases in the lungs, and only minimal autoimmune side effects.

Mice and carcinogen treatment. C57BL/6 mice (H-2b) were purchased from Charles River. Hgf-Cdk4R24C C57Bl/6 mice were bred as described previously (20). TCR-transgenic pmel-1 mice (21) were obtained from The Jackson Laboratory. To enhance melanomagenesis, newborn mice were painted once at day 4 of life with 40 μg of 7,12-dimethylbenz(a)anthracene (DMBA) in 10 μL of acetone. Animal experiments were performed at the Central Animal Facility of the University Hospital Bonn in adherence to the standards of German law for the care and use of laboratory animals.

Analysis of primary melanomas. Mice were inspected and photographed weekly. Growing melanocytic tumors >2 mm in diameter were considered primary melanomas. Their sizes were measured using a vernier calliper and recorded as mean diameter (in millimeters). Mice were sacrificed when progressively growing melanomas exceeded 10 mm or when signs of illness were observed. Autopsy was performed on all mice. Regional lymph nodes and internal organs were inspected for visible metastases and tumors harvested.

Tumor transplantation experiments. B16 cells were maintained in DMEM supplemented with 10% FCS, 2 mmol/L of l-glutamine, 50 μmol/L of 2-mercaptoethanol, 100 IU/mL of penicillin, and 100 μg/mL of streptomycin (Invitrogen). Syngeneic C57BL/6 (H-2b) mice were challenged intracutaneously with 105 B16 melanoma cells. The tumor size was calculated according to the formula, area = (length × width) in mm2. Mice with tumors >100 mm2 were sacrificed. Experiments were performed in groups of five mice and repeated at least twice.

Hgf-Cdk4R24C melanomas were serially transplanted by intracutaneous injections of 2 × 105 viable tumor cells derived from one transplanted melanoma. The tumor size was calculated according to the formula, volume = (length × width2) × 0.5 in mm3. Mice with tumors >2,000 mm3 were sacrificed. Experiments were performed with groups of five mice and repeated twice.

Tumor treatment with adoptively transferred lymphocytes and recombinant adenovirus. Pmel-1 TCR-transgenic mice recognize the H-2Db–binding human gp100 (amino acids 25–33) peptide KVPRNQDWL. Effector T cells cross-react with the mouse gp100 (amino acids 25–33) peptide EGSRNQDWL (21). Pmel-1 mice expressed Thy1.1 to facilitate T-cell tracking. For adoptive transfer, lymphocytes were isolated from lymph nodes and spleens. The number of CD8+Vβ13+Thy1.1+ TCR-transgenic (pmel-1) T cells was determined by flow cytometry. Each treated mouse received 2 × 106 pmel-1 T cells i.v. T cells were activated in vivo by one i.p. injection of 5 × 108 plaque-forming unit recombinant adenovirus expressing human gp100. Ad-gp100 was generated as described previously (29), propagated on 293 cells, purified by cesium chloride density gradient centrifugation and subsequent dialysis according to standard protocols.

Chemotherapeutic preconditioning and adjuvant innate immune stimulation. Chemotherapy was performed by injecting mice i.p. with 2 mg (100 mg/kg) of cyclophosphamide. The synthetic phosphothioate-stabilized CpG oligodeoxynucleotide 1826 with the sequence 5′-TCCATGACGTTCCTGACGTT-3′ (“CpG”) was purchased from TIB Molbiol. Polyinosinic-polycytidylic acid (“pIC”) was purchased from Sigma-Aldrich. Treatment of mice was performed by injecting 50 μg of CpG DNA and 50 μg of pIC dissolved in 100 μL of PBS peritumorally every 3 to 4 d.

Histopathology, immunohistochemistry, and immunofluorescence. Tumor samples were immersed in a zinc-based fixative (DAKO), embedded in paraffin, and stained with H&E. Immunostaining was performed with the rabbit pAb Pep7 against TRP1 (a kind gift from Vincent Hearing, NIH, Bethesda, MD) or the rat monoclonal antibodies against CD45 (BD PharMingen) and CD3 (Serotec) followed by appropriate fluorochrome-conjugated or biotin-conjugated secondary antibodies and the LSAB-2 color development system (DAKO). Heavily pigmented tumors were bleached before staining (20 min at 37°C in 30% H2O2 and 0.5% KOH, 20 s in 1% acetic acid, and 5 min in TRIS buffer).

Quantitative reverse transcription-PCR. Tumor samples were immediately snap-frozen in liquid nitrogen. Total RNA was isolated using TRI Reagent (Sigma) and purified using RNeasy columns (Qiagen). RNA was reverse-transcribed using Superscript II and oligo-dT18 primers (Invitrogen). PCR was performed with TaqMan assays [Bact Mm00607939_s1, Si (gp100) Mm00498996_m1, Tgfb1 Mm03024053_m1, Cxcl2 Mm00436450_m1, and Cxcl10 Mm00445235_m1] or with fast SYBR Green Master Mix (ABI) using the following primers (from 5′ to 3′): Ubc forward AGGCAAGACCATCACCTTGGACG, reverse CCATCACACCCAAGAACAAGCACA; Ifng forward CTCTTCTTGGATATCTGGAGG, reverse CCTGATTGTCTTTCAAGACTTC; Gzmb forward CTCCAATGACATCATGCTGC, reverse TGGCTTCACATTGACATTGC; Prf1 forward TGAGAAGACCTATCAGGACC, reverse AAGTCAAGGTGGAGTGGAGG; Tbet forward CAACAACCCCTTTGCCAAAG, reverse TCCCCCAAGCAGTTGACAGT; and Foxp3 forward CCCAGGAAAGACAGCAACCTT, reverse TTCTCACAACCAGGCCACTTG.

Flow cytometry. Tumor tissue, spleens, and lymph nodes were harvested, dissociated mechanically, incubated with 1 mg/mL of collagenase D (Roche) in PBS with 5% fetal bovine serum (Biochrom) for 30 min at 37°C and filtered through 70-μm cell strainers (BD Biosciences). Staining was performed with fluorochrome-conjugated antibodies against CD45, B220, CD8α, CD3, Vβ13, CD90, B220, Gr-1, CD11b, CD11c, H2-Kb (all from BD PharMingen), and IFN-γ (eBioscience). For intracellular staining, Golgistop and Golgiplug (BD Biosciences) were used according to standard protocols. Data were acquired with a FACSCanto flow or a FACS LSRII flow cytometry system (FACS Core Facility, Bonn, Germany) and analyzed with FACSDiva (BD Biosciences) or FlowJo (TreeStar, V8.7.1) software.

Statistical analyses. Statistically significant differences were calculated with the nonparametric Mann-Whitney U test using the SPSS 11 computer program.

Primary Hgf-Cdk4R24C melanomas avoid recognition and destruction by tumor-specific T lymphocytes. We previously reported that a treatment regimen consisting of recombinant adenoviral vaccination against the melanosomal protein TRP2 in combination with adjuvant peritumoral injections of TLR agonists was only able to delay the growth of primary carcinogen-induced melanomas in Hgf-Cdk4R24C mice (20). To further enhance cellular antitumor immunity, we performed adoptive lymphocyte transfer experiments using TCR-transgenic CD8+ T cells derived from pmel-1 mice. Pmel-1 T cells recognize an H2-Db–binding peptide epitope derived from the human melanosomal protein gp100. Effector T cells cross-reacts with the slightly different mouse gp100 epitope, which is naturally expressed by melanocytes and melanoma cells (21). In our experiments, we took advantage of the early and synchronous development of primary melanomas in Hgf-Cdk4R24C mice following epicutaneous application of DMBA shortly after birth. At an age of 70 days, all DMBA-exposed Hgf-Cdk4R24C mice carry between 5 and 15 macroscopically visible primary cutaneous melanomas with the largest tumors measuring between 4 and 6 mm in diameter. Tumor-bearing mice were randomly assigned to treatment and control cohorts. The treatment cohort received an adoptive transfer of pmel-1 T lymphocytes and a simultaneous vaccination with recombinant Ad-gp100 (lymph/vacc, LV). The treatment was repeated in surviving mice at an age of 91 days. Flow cytometric analyses of splenocytes and circulating peripheral blood mononuclear cells in individual LV-treated tumor-bearing mice confirmed expansion and effector cell differentiation of transferred pmel-1 T lymphocytes in vivo (Fig. 1A). Despite the presence of large numbers of circulating tumor-specific effector T lymphocytes in LV-treated mice, we observed that primary Hgf-Cdk4R24C melanomas continuously grew similar to untreated control mice (Fig. 1B).

Figure 1.

Primary Hgf-Cdk4R24C melanomas are accessible to tumor-specific T lymphocytes but avoid recognition and destruction in the tumor microenvironment. A, experimental protocol (left); expansion and effector cell differentiation of pmel-1 TCR transgenic T cells in a LV-treated mouse 7 d after adoptive transfer (right). Flow cytometric dot plot quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate and corresponding histogram for antigen-specific intracellular IFN-γ expression in CD8+Thy1.1+Vβ13+ pmel-1 T cells. B, tumor growth kinetics. Left, tumor diameter of the largest primary melanoma over time in individual Hgf-Cdk4R24C mice of a LV-treated cohort (top) and in a control cohort (bottom). Right, macroscopic pictures of one mouse in each group over time. C, immune cell infiltration in primary tumors. Immunofluorescence picture of a primary tumor from an untreated (left) and an LV-treated mouse (right) 7 d after adoptive T-cell transfer staining for TRP1 (green), CD45 (red), and DAPI (blue). D, T lymphocytes in close contact with tumor cells. Left, corresponding immunohistochemical stain for CD3+ T lymphocytes in a primary tumor from a LV-treated mouse. Right, flow cytometric dot plot showing Thy1.1+Vβ13+ pmel-1 cells in tumor-infiltrating CD8+ T cells.

Figure 1.

Primary Hgf-Cdk4R24C melanomas are accessible to tumor-specific T lymphocytes but avoid recognition and destruction in the tumor microenvironment. A, experimental protocol (left); expansion and effector cell differentiation of pmel-1 TCR transgenic T cells in a LV-treated mouse 7 d after adoptive transfer (right). Flow cytometric dot plot quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate and corresponding histogram for antigen-specific intracellular IFN-γ expression in CD8+Thy1.1+Vβ13+ pmel-1 T cells. B, tumor growth kinetics. Left, tumor diameter of the largest primary melanoma over time in individual Hgf-Cdk4R24C mice of a LV-treated cohort (top) and in a control cohort (bottom). Right, macroscopic pictures of one mouse in each group over time. C, immune cell infiltration in primary tumors. Immunofluorescence picture of a primary tumor from an untreated (left) and an LV-treated mouse (right) 7 d after adoptive T-cell transfer staining for TRP1 (green), CD45 (red), and DAPI (blue). D, T lymphocytes in close contact with tumor cells. Left, corresponding immunohistochemical stain for CD3+ T lymphocytes in a primary tumor from a LV-treated mouse. Right, flow cytometric dot plot showing Thy1.1+Vβ13+ pmel-1 cells in tumor-infiltrating CD8+ T cells.

Close modal

Next, we investigated how primary tumors avoid recognition and destruction by tumor-specific T lymphocytes. We sacrificed Hgf-Cdk4R24C mice bearing primary tumors 10 days after the first LV treatment. Immunofluorescence and flow cytometric analyses of primary tumors from untreated control mice revealed only very few infiltrating immune cells. In contrast, tumors from LV-treated mice showed diffuse infiltration with CD45+ immune cells (Fig. 1C). Histomorphologic, immunohistochemical, and flow cytometric analyses of LV-treated tumors showed large numbers of infiltrating CD3+ T lymphocytes in close contact with melanoma cells, including a significant percentage of CD8+Thy1.1+Vβ13+ (pmel-1) effector cells (Fig. 1D). Thus, primary Hgf-Cdk4R24C melanomas were principally accessible for tumor-specific T lymphocytes but escaped their cytotoxic effector functions in the tumor microenvironment. The majority of immune cells within primary tumors showed a Gr1+CD11b+ phenotype, Gr1CD11b+ or CD4+CD25+ cells could only very rarely be found. This suggested that tumor tolerance may involve the activity of myeloid-derived suppressor cells but not of tumor-associated macrophages or of regulatory T cells (data not shown).

Enhancement of cellular antitumor immunity using combination treatment approaches. Next, we investigated the ability of pmel-1 T cells to recognize primary Hgf-Cdk4R24C melanoma cells in vitro. Primary Hgf-Cdk4R24C melanoma cells are heavily pigmented and reverse transcription-PCR (RT-PCR) analyses show strong expression of the melanocyte-specific protein gp100. Short-term cultures of primary melanoma cells showed both spindle-shaped and epithelioid morphologies which recapitulated their growth pattern in vivo (Fig. 2A). These melanoma cells only expressed very low levels of MHC class I molecules on their cell surface which could be up-regulated following exposure to recombinant IFN-α (Fig. 2B). IFN-γ ELISPOT assays showed that cultured melanoma cells could be recognized by pmel-1 effector T lymphocytes in vitro as effectively as cultured B16 melanoma cells. Tumor cell recognition could be significantly improved following incubation with recombinant IFN-α (Fig. 2C).

Figure 2.

IFN-α increases MHC class I expression on the surface of Hgf-Cdk4R24C melanoma cells and enhances recognition by pmel-1 T lymphocytes. A, microscopic appearance of Hgf-Cdk4R24C melanoma cells in vitro (magnification, ×200). B, MHC class I surface expression on Hgf-Cdk4R24C melanoma cells (left) and influence of recombinant IFN-α (right). C, recognition of Hgf-Cdk4R24C melanoma cells by pmel-1 T lymphocytes and influence of recombinant IFN-α. IFN-γ ELISPOT with memory pmel-1 T cells and Hgf-Cdk4R24C melanoma cells (left), B16 melanoma cells (middle), and synthetic peptide (right; *, P < 0.05).

Figure 2.

IFN-α increases MHC class I expression on the surface of Hgf-Cdk4R24C melanoma cells and enhances recognition by pmel-1 T lymphocytes. A, microscopic appearance of Hgf-Cdk4R24C melanoma cells in vitro (magnification, ×200). B, MHC class I surface expression on Hgf-Cdk4R24C melanoma cells (left) and influence of recombinant IFN-α (right). C, recognition of Hgf-Cdk4R24C melanoma cells by pmel-1 T lymphocytes and influence of recombinant IFN-α. IFN-γ ELISPOT with memory pmel-1 T cells and Hgf-Cdk4R24C melanoma cells (left), B16 melanoma cells (middle), and synthetic peptide (right; *, P < 0.05).

Close modal

Based on our previous experience with adjuvant injections of the TLR agonists, CpG and pIC (20), we reasoned that the stimulation of viral pattern recognition receptors which activate the type I IFN system should promote the antitumor efficacy of adoptively transferred pmel-1 T lymphocytes (30, 31). We initially evaluated this hypothesis in parallel experiments using the transplanted B16 melanoma model in a rigorous 10-day treatment model in which tumors were already macroscopically visible (Fig. 3A). Adoptively transferred pmel-1 T lymphocytes and viral vaccination with Ad-gp100 (LV) do not affect the progressive growth of advanced B16 melanomas in this experimental setting (Fig. 3B). The antitumor efficacy of adoptive lymphocyte transfer and viral vaccination could be boosted by persistent stimulation of innate immunity through adjuvant peritumoral injections of CpG and pIC (LV + I) leading to considerably prolonged survival (Fig. 3B).

Figure 3.

Sequential application of chemotherapy, adoptive lymphocyte transfer, viral vaccination, and immunostimulatory nucleic acids causes the regression of advanced transplanted melanomas. A, experimental protocol. B, tumor growth kinetics. Tumor area over time in individual mice from representative groups treated as indicated. C, experimental protocol. D, tumor growth kinetics. Tumor volume over time in groups of mice treated as indicated (points, mean; bars, SE).

Figure 3.

Sequential application of chemotherapy, adoptive lymphocyte transfer, viral vaccination, and immunostimulatory nucleic acids causes the regression of advanced transplanted melanomas. A, experimental protocol. B, tumor growth kinetics. Tumor area over time in individual mice from representative groups treated as indicated. C, experimental protocol. D, tumor growth kinetics. Tumor volume over time in groups of mice treated as indicated (points, mean; bars, SE).

Close modal

We previously showed that chemotherapeutic preconditioning with cyclophosphamide enhances the efficacy of melanoma vaccines (18). Cyclophosphamide also facilitates the expansion of adoptively transferred T lymphocytes (32, 33). Strikingly, a treatment protocol consisting of cyclophosphamide treatment before adoptive lymphocyte transfer and viral vaccination followed by adjuvant injections of CpG and pIC (C + LV + I) achieved complete regression of large B16 melanomas (Fig. 3B). This combination treatment protocol was much less effective without innate immune stimulation (C + LV) or without adoptive lymphocyte transfer (C + V + I), indicating that all four components were required (Fig. 3B). Mice in the C + LV + I–treated groups remained tumor-free for several months and showed only limited destruction of melanocytes indicating preferential recognition and destruction of malignant over normal melanocytic cells (data not shown).

In subsequent experiments, we treated established transplanted Hgf-Cdk4R24C melanomas generated by serial transplantation of primary tumors (Fig. 3C). Groups of mice with macroscopically visible transplanted Hgf-Cdk4R24C melanomas received an adoptive transfer of pmel-1 T lymphocytes and recombinant adenoviral vaccination (LV) on day 14 or were left untreated. Established transplanted Hgf-Cdk4R24C melanomas were not affected by the LV treatment similar to advanced transplanted B16 melanomas (Fig. 3D). However, treatment according to the C + LV + I protocol also resulted in complete regression and long-term cure, demonstrating unequivocally that pmel-1 T cells are able to recognize and destroy Hgf-Cdk4R24C melanoma cells (Fig. 3D).

Complete regression of primary cutaneous as well as metastatic Hgf-Cdk4R24C melanomas. Having shown the efficacy of our combination chemoimmunotherapy protocol against transplanted Hgf-Cdk4R24C melanoma, we also treated mice with primary Hgf-Cdk4R24C melanomas. DMBA-exposed Hgf-Cdk4R24C mice were again randomly assigned to different cohorts at an age of 70 days at which time they carried multiple, macroscopically visible primary melanomas. Strikingly, two consecutive C + LV + I treatment cycles were able to cause the complete regression of primary cutaneous melanomas (Fig. 4A and B). In a total of 30 mice treated in 8 independent experiments, we observed significantly prolonged survival (Fig. 4C). Control cohorts of tumor-bearing mice received either cyclophosphamide and CpG/pIC (C + I) or adoptive lymphocytes and viral vaccination (LV) only, both of which slightly delayed tumor progression (Fig. 4C).

Figure 4.

Combination chemoimmunotherapy causes the regression of advanced primary and metastatic melanomas. A, experimental protocol. B, macroscopic pictures of a C + LV + I–treated Hgf-Cdk4R24C mouse over time. C, top row, tumor growth kinetics. Tumor diameter of the largest primary melanoma over time in individual Hgf-Cdk4R24C mice treated as indicated. Arrow, start of treatment. Bottom left, Kaplan-Meier graph indicating cumulative survival in cohorts of mice treated as indicated (Ctrl, n = 20; LV, n = 15; C + I, n = 15; C + LV + I, n = 30). Bottom right, survival in days (columns, mean; bars, SE; **, P < 0.01). D, macroscopic pictures of lung metastases in mice treated as indicated.

Figure 4.

Combination chemoimmunotherapy causes the regression of advanced primary and metastatic melanomas. A, experimental protocol. B, macroscopic pictures of a C + LV + I–treated Hgf-Cdk4R24C mouse over time. C, top row, tumor growth kinetics. Tumor diameter of the largest primary melanoma over time in individual Hgf-Cdk4R24C mice treated as indicated. Arrow, start of treatment. Bottom left, Kaplan-Meier graph indicating cumulative survival in cohorts of mice treated as indicated (Ctrl, n = 20; LV, n = 15; C + I, n = 15; C + LV + I, n = 30). Bottom right, survival in days (columns, mean; bars, SE; **, P < 0.01). D, macroscopic pictures of lung metastases in mice treated as indicated.

Close modal

Next, we determined the effect of the C + LV + I treatment protocol on metastatic disease. DMBA-exposed Hgf-Cdk4R24C mice already carried spontaneous lung metastases at an age of 70 days when treatment was started (Fig. 4D). In a cohort of five C + LV + I–treated mice which showed complete regression of their DMBA-induced primary melanomas in the skin at an age of 135 days, we did not find any metastases in the lungs. In contrast, lung metastases were found in almost all LV-treated and untreated mice at the time of death (Fig. 4D). Thus, our combination chemoimmunotherapy not only caused the regression of primary cutaneous melanomas but also of spontaneous metastases in the lungs.

Cytotoxic inflammation in the tumor microenvironment. Next, we investigated the effect of chemotherapeutic preconditioning and adjuvant innate immune stimulation on the number and functional status of adoptively transferred pmel-1 T cells in tumor-bearing mice. Cyclophosphamide treatment before adoptive lymphocyte transfer and viral vaccination (C + LV) enhanced the expansion of adoptively transferred pmel-1 T lymphocytes (Fig. 5A). Additional adjuvant peritumoral injections of CpG/pIC after adoptive lymphocyte transfer and viral vaccination (C + LV + I) promoted their survival and supported their effector cell differentiation as evidenced by an increase in the percentage of lymphocytes capable of producing IFN-γ (Fig. 5A). Experiments with groups of mice bearing transplanted B16 melanoma analyzed at different time points confirmed that chemotherapy and innate immune stimulation cooperatively increased expansion, survival, and effector differentiation of adoptively transferred pmel-1 T lymphocytes in vivo (*, P < 0.05; Fig. 5B).

Figure 5.

Combination chemoimmunotherapy promotes strong cytotoxic inflammation in the tumor microenvironment of primary melanomas. A, T-cell expansion and effector cell differentiation 10 d after adoptive transfer in tumor-bearing Hgf-Cdk4R24C mice. Top row, flow cytometric dot plots quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate of mice treated as indicated. Bottom row, corresponding histograms for antigen-specific intracellular IFN-γ expression in CD8+Thy1.1+Vβ13+ T cells. B, time course of T-cell expansion and effector cell differentiation in mice bearing established B16 melanomas. Left, mean percentage (±SE) of Thy1.1+Vβ13+ splenocytes in the CD8+ gate. Right, percentage of CD8+Thy1.1+Vβ13+ T cells able to produce IFN-γ (points, mean; bars, SE). Data are from six individual mice in three separate experiments for each time point (*, P < 0.05). C, molecular characterization of the microenvironment in treated primary tumors. Results of quantitative RT-PCR for the indicated genes calculated as mean fold-change of expression in five individual samples of mice from the LV-treated and the C + LV + I–treated group relative to the untreated control group (columns, mean; bars, SE; *, P < 0.05; **, P < 0.01). D, cytotoxic immune cell infiltration. Left, H&E stain of a C + LV + I–treated primary melanoma (magnification, ×200). Right, corresponding immunohistochemical stain for CD3+ T lymphocytes (magnification, ×400).

Figure 5.

Combination chemoimmunotherapy promotes strong cytotoxic inflammation in the tumor microenvironment of primary melanomas. A, T-cell expansion and effector cell differentiation 10 d after adoptive transfer in tumor-bearing Hgf-Cdk4R24C mice. Top row, flow cytometric dot plots quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate of mice treated as indicated. Bottom row, corresponding histograms for antigen-specific intracellular IFN-γ expression in CD8+Thy1.1+Vβ13+ T cells. B, time course of T-cell expansion and effector cell differentiation in mice bearing established B16 melanomas. Left, mean percentage (±SE) of Thy1.1+Vβ13+ splenocytes in the CD8+ gate. Right, percentage of CD8+Thy1.1+Vβ13+ T cells able to produce IFN-γ (points, mean; bars, SE). Data are from six individual mice in three separate experiments for each time point (*, P < 0.05). C, molecular characterization of the microenvironment in treated primary tumors. Results of quantitative RT-PCR for the indicated genes calculated as mean fold-change of expression in five individual samples of mice from the LV-treated and the C + LV + I–treated group relative to the untreated control group (columns, mean; bars, SE; *, P < 0.05; **, P < 0.01). D, cytotoxic immune cell infiltration. Left, H&E stain of a C + LV + I–treated primary melanoma (magnification, ×200). Right, corresponding immunohistochemical stain for CD3+ T lymphocytes (magnification, ×400).

Close modal

The combined activities of the C + LV + I protocol converge in the microenvironment of primary tumors. Quantitative RT-PCR analyses 10 days after adoptive lymphocyte transfer in C + LV + I–treated mice showed significantly increased expression of mRNA coding for IFN-γ, CXCL-10, perforin, and granzyme B (*, P < 0.05; Fig. 5C) when compared with LV-treated mice. This correlated with enhanced immune cell infiltration in C + LV + I–treated mice (data not shown). Increased expression of the transcription factor T-bet characteristic for a Th1-biased cytotoxic immune response and decreased expression of the transcription factor Foxp3 characteristic for regulatory T cells was observed in both LV-treated and C + LV + I–treated tumors. Histopathologic and immunofluorescence analyses revealed massive tumor cell destruction associated with dense infiltration of CD3+ T cells (Fig. 5D). Collectively, these results show that the C + LV + I treatment strongly promotes cytotoxic inflammation in primary melanomas.

Development of new primary melanomas in successfully treated Hgf-Cdk4R24C mice. Subsequently, we monitored cohorts of C + LV + I–treated mice which showed complete tumor regression for longer time periods. Starting at an age of 150 days, ∼2 months after the second C + LV + I treatment cycle, these mice developed new pigmented melanomas in their carcinogen-treated skin (Fig. 6A). These new primary melanomas in successfully C + LV + I–treated mice at an age of 200 days histomorphologically resembled those observed in untreated mice at an age of 90 days with very little immune cell infiltrate. Quantitative RT-PCR analyses of tumors revealed increased expression of transforming growth factor-β and CXCL-2 relative to IFN-γ and CXCL-10, as well as decreased expression of perforin, granzyme B, and T-bet (Fig. 6B). Interestingly, we did not observe an increase in the expression of Foxp3 or an increase in CD4+CD25+ T cells in these tumors which would indicate the local activity of regulatory T cells (Fig. 6B and C). Again, most immune cells within tumors showed a Gr1+CD11b+ phenotype which corresponded to enhanced infiltration with granulocytes, frequently in association with superficial ulceration (Fig. 6C; data not shown).

Figure 6.

Development of new primary melanomas in successfully treated Hgf-Cdk4R24C mice. A, tumor growth kinetics. Left, tumor diameter of the largest new primary melanoma over time in individual Hgf-Cdk4R24C mice of a C + LV + I–treated cohort. Right, macroscopic pictures of a mouse over time. B, molecular characterization of the microenvironment in new primary tumors. Results of quantitative RT-PCR for the indicated genes calculated as mean fold change of expression in five individual samples of mice from the C + LV + I–treated group relative to the untreated control group (columns, mean; bars, SE). C, tumor-infiltrating CD4+CD25+ and Gr1+CD11b+. Flow cytometric dot plots for a new primary melanoma of a previously C + LV + I–treated mouse at an age of 200 d (left) compared with a primary melanoma of an untreated control mouse at an age of 90 d (right). D, persistence of pmel-1 memory T cells. Top, flow cytometric dot plots quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate of C + LV + I–treated mice with new primary melanomas. Analysis of a 200-d-old mouse (left) and a 207-d-old mouse 1 wk after a booster vaccination with Ad-gp100 (right). Bottom, corresponding dot plots showing CD8+CD45+ T lymphocytes within the tumor and histograms quantifying the percentage of Thy1.1+ pmel-1 cells.

Figure 6.

Development of new primary melanomas in successfully treated Hgf-Cdk4R24C mice. A, tumor growth kinetics. Left, tumor diameter of the largest new primary melanoma over time in individual Hgf-Cdk4R24C mice of a C + LV + I–treated cohort. Right, macroscopic pictures of a mouse over time. B, molecular characterization of the microenvironment in new primary tumors. Results of quantitative RT-PCR for the indicated genes calculated as mean fold change of expression in five individual samples of mice from the C + LV + I–treated group relative to the untreated control group (columns, mean; bars, SE). C, tumor-infiltrating CD4+CD25+ and Gr1+CD11b+. Flow cytometric dot plots for a new primary melanoma of a previously C + LV + I–treated mouse at an age of 200 d (left) compared with a primary melanoma of an untreated control mouse at an age of 90 d (right). D, persistence of pmel-1 memory T cells. Top, flow cytometric dot plots quantifying Thy1.1+Vβ13+ splenocytes in the CD8+ gate of C + LV + I–treated mice with new primary melanomas. Analysis of a 200-d-old mouse (left) and a 207-d-old mouse 1 wk after a booster vaccination with Ad-gp100 (right). Bottom, corresponding dot plots showing CD8+CD45+ T lymphocytes within the tumor and histograms quantifying the percentage of Thy1.1+ pmel-1 cells.

Close modal

Importantly, we were able to detect a small but significant population of CD8+Thy1.1+Vβ13+ (pmel-1) T cells not only in the spleens but also in the new pigmented cutaneous melanomas in all investigated mice at an average age of 200 days, i.e., 3 months after the second treatment cycle. These pmel-1 T cells could be expanded to large numbers following a booster vaccination with recombinant Ad-gp100 (Fig. 6D). Thus, adoptively transferred pmel-1 T cells in C + LV + I–treated mice acquire memory status, persist for an extended time, and can be efficiently restimulated in vivo. Future experiments will have to reveal whether these reactivated T cells are again capable of causing tumor regression.

In our work, we identified a highly effective combination chemoimmunotherapy which can reverse immune tolerance in the tumor microenvironment and cause complete regression of advanced primary and metastatic melanomas in a novel genetically engineered tumor model. This model avoids the use of immunogenic foreign proteins as oncogenes or antigens (i.e. SV40 large T antigen or Her-2/neu; refs. 3441). Instead, tumors arise autochthonously in the skin, grow progressively, and metastasize spontaneously to draining lymph nodes and lungs because of two nonimmunogenic genetic changes also observed in patients. Although tumor cells express several melanocyte-specific lineage differentiation antigens which can be potentially recognized by Tlymphocytes, they express low levels of MHC class I molecules and effectively evade innate as well as adaptive immune defense. Using an adoptive T-cell transfer approach, we found that tumors continued to grow even in the presence of large numbers of invivo–activated melanoma-specific T lymphocytes. To overcome tumor immune tolerance, we combined adoptive lymphocyte transfer and viral vaccination with adjuvant peritumoral injections of the immunostimulatory nucleic acids, CpG and pIC, which activate viral pattern recognition receptors of the toll receptor family (23, 30, 42). Persistent innate immune stimulation strongly enhanced the therapeutic efficacy of adoptive lymphocyte transfer and adenoviral vaccination. Chemotherapeutic preconditioning with cyclophosphamide, which has facilitated tumor treatment with adoptively transferred lymphocytes for many years (43), further increased antitumor efficacy leading to complete and long-term regression of advanced, macroscopically visible transplanted as well as primary melanomas. All four complementary components of our treatment protocol were required to achieve complete regression and long-term remission of large primary and transplanted melanomas.

Cyclophosphamide pretreatment and adjuvant injections of immunostimulatory nucleic acids cooperatively enhanced both the expansion as well as the cytotoxic effector cell differentiation of adoptively transferred pmel-1 T lymphocytes and induced a strong cytotoxic inflammation in the tumor microenvironment. Cyclophosphamide has pleiotropic effects including the promotion of T-cell homeostatic expansion, inhibition of regulatory T-cell function, damage to the tumor vasculature, and functional alteration of tumor-infiltrating immune cells (32, 33, 44). CpG and pIC activate the type I IFN system through viral pattern recognition receptors (31). IFNs may be involved in resurrecting the barriers to cancer development on several levels. We show that IFN-α increases MHC class I expression on Hgf-Cdk4R24C melanoma cells and supports recognition by pmel-1 T cells. In addition, IFNs can also promote tumor cell apoptosis, impair angiogenesis, counteract immunosuppression, prolong T-cell survival, and enhance cytotoxic effector functions in the tumor microenvironment (4548).

Of the potential mechanisms for tumor immune tolerance in primary Hgf-Cdk4R24C melanomas, we could rule out the local activity of regulatory T cells or tumor-associated macrophages as these cells were rarely found within primary tumors. Instead, myeloid-derived suppressor cells may be involved because the majority of tumor-infiltrating immune cells show a Gr1+CD11b+ phenotype. In addition, we speculate that the well-known immunosuppressive activity of hepatocyte growth factor and the distinct immunologic properties of melanocytes might participate in inhibiting T-cell effector functions in the tumor microenvironment.

All successfully C + LV + I–treated mice with complete tumor regression eventually developed new primary DMBA-induced melanomas which morphologically resembled those observed in untreated mice. A detailed molecular and cellular analysis of the tumor microenvironment indicated that the balance of cytokines and chemokines had shifted back in favor of tumor progression. Tumor growth occurred despite the presence of melanoma-specific (pmel-1) memory T cells, which could be found both in secondary lymphoid organs as well as in the tumor itself. These potentially tumor-specific T cells could be efficiently reactivated in vivo with a recombinant adenovirus booster immunization. The newly arising primary melanomas apparently evaded the surveillance function by pmel-1 memory T cells, presumably because they do not adequately activate the innate immune system. The effect of the vaccine-induced T-cell memory response on tumor growth will have to be determined in future experiments.

Interestingly, treated mice with complete tumor regression showed only limited vitiligo-like fur depigmentation predominantly at the site of tumors, although pmel-1 T cells can principally recognize and kill all cells of melanocytic origin which express the melanosomal protein gp100. This suggests that our combination chemoimmunotherapy promotes preferential recognition and destruction of malignant over normal pigment cells. The relative tumor selectivity may be conferred by combining the antigen-specificity of T cells with the nonspecific action of chemotherapy and innate immune stimulation on both immune and tumor cells. This may be an advantage over the use of high-dose interleukin 2, which promotes the antitumor efficacy of adoptively transferred lymphocytes with significant autoreactivity against pigment cells (49).

Treatment protocols combining chemotherapy, total body irradiation, adoptive lymphocyte transfer and vaccination strategies are currently evaluated in a large number of clinical trials. Our results suggest that adjuvant innate immune stimulation using immunostimulatory nucleic acids can also be a powerful approach to enhance the therapeutic efficacy of adoptive lymphocyte transfer and viral vaccination. The use of TLR agonists may even replace total body irradiation because it has recently been reported that this treatment modality depends on TLR triggering (50). The unique experimental features of the genetically engineered Hgf-Cdk4R24C mouse model, which faithfully portrays the biology of melanoma in man, may help to facilitate the successful clinical translation of effective strategies for combination chemoimmunotherapy against this deadly disease.

No potential conflicts of interest were disclosed.

Note: J. Kohlmeyer and M. Cron contributed equally.

Grant support: Deutsche Forschungsgemeinschaft Tu 90/6-2, SFB832 A12 (T. Tüting) and the BONFOR program of the University of Bonn (J. Kohlmeyer).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Stefan Holtin, Noriaki Nakai, Alexander Sporleder, Nils Schoof, and Jens Wohlmann for expert technical assistance and helpful discussions.

1
Mandelboim O, Vadai E, Fridkin M, et al. Regression of established murine carcinoma metastases following vaccination with tumour-associated antigen peptides.
Nat Med
1995
;
1
:
1179
–83.
2
Mayordomo JI, Zorina T, Storkus WJ, et al. Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumour immunity.
Nat Med
1995
;
1
:
1297
–302.
3
Overwijk WW, Lee DS, Surman DR, et al. Vaccination with a recombinant vaccinia virus encoding a “self” antigen induces autoimmune vitiligo and tumor cell destruction in mice: requirement for CD4(+) T lymphocytes.
Proc Natl Acad Sci U S A
1999
;
96
:
2982
–7.
4
Dranoff G, Jaffee E, Lazenby A, et al. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity.
Proc Natl Acad Sci U S A
1993
;
90
:
3539
–43.
5
Rosenberg SA, Yang JC, Schwartzentruber DJ, et al. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma.
Nat Med
1998
;
4
:
321
–7.
6
Thurner B, Haendle I, Roder C, et al. Vaccination with mage-3A1 peptide-pulsed mature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma.
J Exp Med
1999
;
190
:
1669
–78.
7
Rosenberg SA, Zhai Y, Yang JC, et al. Immunizing patients with metastatic melanoma using recombinant adenoviruses encoding MART-1 or gp100 melanoma antigens.
J Natl Cancer Inst
1998
;
90
:
1894
–900.
8
Soiffer R, Lynch T, Mihm M, et al. Vaccination with irradiated autologous melanoma cells engineered to secrete human granulocyte-macrophage colony-stimulating factor generates potent antitumor immunity in patients with metastatic melanoma.
Proc Natl Acad Sci U S A
1998
;
95
:
13141
–6.
9
van BN, Bonnet MC, Dreno B, et al. Tumoral and immunologic response after vaccination of melanoma patients with an ALVAC virus encoding MAGE antigens recognized by T cells.
J Clin Oncol
2005
;
23
:
9008
–21.
10
Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: moving beyond current vaccines.
Nat Med
2004
;
10
:
909
–15.
11
Boon T, Coulie PG, Van den Eynde BJ, van der BP. Human T cell responses against melanoma.
Annu Rev Immunol
2006
;
24
:
175
–208.
12
Staveley-O'Carroll K, Sotomayor E, Montgomery J, et al. Induction of antigen-specific T cell anergy: an early event in the course of tumor progression.
Proc Natl Acad Sci U S A
1998
;
95
:
1178
–83.
13
Ochsenbein AF, Sierro S, Odermatt B, et al. Roles of tumour localization, second signals and cross priming in cytotoxic T-cell induction.
Nature
2001
;
411
:
1058
–64.
14
Gorelik L, Flavell RA. Immune-mediated eradication of tumors through the blockade of transforming growth factor-β signaling in T cells.
Nat Med
2001
;
7
:
1118
–22.
15
Pardoll D. Does the immune system see tumors as foreign or self?
Annu Rev Immunol
2003
;
21
:
807
–39.
16
Wang T, Niu G, Kortylewski M, et al. Regulation of the innate and adaptive immune responses by Stat-3 signaling in tumor cells.
Nat Med
2004
;
10
:
48
–54.
17
Sotomayor EM, Borrello I, Tubb E, et al. Conversion of tumor-specific CD4+ T-cell tolerance to T-cell priming through in vivo ligation of CD40.
Nat Med
1999
;
5
:
780
–7.
18
Steitz J, Brück J, Lenz J, Knop J, Tüting T. Depletion of CD25(+) CD4(+) T cells and treatment with tyrosinase-related protein 2-transduced dendritic cells enhance the interferon α-induced, CD8(+) T-cell-dependent immune defense of B16 melanoma.
Cancer Res
2001
;
61
:
8643
–6.
19
Yang Y, Huang CT, Huang X, Pardoll DM. Persistent Toll-like receptor signals are required for reversal of regulatory T cell-mediated CD8 tolerance.
Nat Immunol
2004
;
5
:
508
–15.
20
Tormo D, Ferrer A, Bosch P, et al. Therapeutic efficacy of antigen-specific vaccination and toll-like receptor stimulation against established transplanted and autochthonous melanoma in mice.
Cancer Res
2006
;
66
:
5427
–35.
21
Overwijk WW, Theoret MR, Finkelstein SE, et al. Tumor regression and autoimmunity after reversal of a functionally tolerant state of self-reactive CD8+ T cells.
J Exp Med
2003
;
198
:
569
–80.
22
Teague RM, Sather BD, Sacks JA, et al. Interleukin-15 rescues tolerant CD8+ T cells for use in adoptive immunotherapy of established tumors.
Nat Med
2006
;
12
:
335
–41.
23
Tormo D, Ferrer A, Gaffal E, et al. Rapid growth of invasive metastatic melanoma in carcinogen-treated hepatocyte growth factor/scatter factor-transgenic mice carrying an oncogenic CDK4 mutation.
Am J Pathol
2006
;
169
:
665
–72.
24
Otsuka T, Takayama H, Sharp R, et al. c-Met autocrine activation induces development of malignant melanoma and acquisition of the metastatic phenotype.
Cancer Res
1998
;
58
:
5157
–67.
25
Satyamoorthy K, Li G, Gerrero MR, et al. Constitutive mitogen-activated protein kinase activation in melanoma is mediated by both BRAF mutations and autocrine growth factor stimulation.
Cancer Res
2003
;
63
:
756
–9.
26
Wölfel T, Hauer M, Schneider J, et al. A p16INK4a-insensitive CDK4 mutant targeted by cytolytic T lymphocytes in a human melanoma.
Science
1995
;
269
:
1281
–4.
27
Sotillo R, Garcia JF, Ortega S, et al. Invasive melanoma in Cdk4-targeted mice.
Proc Natl Acad Sci U S A
2001
;
98
:
13312
–7.
28
Miller AJ, Mihm MC, Jr. Melanoma.
N Engl J Med
2006
;
355
:
51
–65.
29
Jonuleit H, Tüting T, Steitz J, et al. Efficient transduction of mature CD83+ dendritic cells using recombinant adenovirus suppressed T cell stimulatory capacity.
Gene Ther
2000
;
7
:
249
–54.
30
Whitmore MM, deVeer MJ, Edling A, et al. Synergistic activation of innate immunity by double-stranded RNA and CpG DNA promotes enhanced antitumor activity.
Cancer Res
2004
;
64
:
5850
–60.
31
Kawai T, Akira S. Innate immune recognition of viral infection.
Nat Immunol
2006
;
7
:
131
–7.
32
Ercolini AM, Ladle BH, Manning EA, et al. Recruitment of latent pools of high-avidity CD8(+) T cells to the antitumor immune response.
J Exp Med
2005
;
201
:
1591
–602.
33
Bracci L, Moschella F, Sestili P, et al. Cyclophosphamide enhances the antitumor efficacy of adoptively transferred immune cells through the induction of cytokine expression, B-cell and T-cell homeostatic proliferation, and specific tumor infiltration.
Clin Cancer Res
2007
;
13
:
644
–53.
34
Speiser DE, Miranda R, Zakarian A, et al. Self antigens expressed by solid tumors do not efficiently stimulate naive or activated T cells: implications for immunotherapy.
J Exp Med
1997
;
186
:
645
–53.
35
Garbi N, Arnold B, Gordon S, Hammerling GJ, Ganss R. CpG motifs as proinflammatory factors render autochthonous tumors permissive for infiltration and destruction.
J Immunol
2004
;
172
:
5861
–9.
36
Willimsky G, Blankenstein T. Sporadic immunogenic tumours avoid destruction by inducing T-cell tolerance.
Nature
2005
;
437
:
141
–6.
37
Lollini PL, Cavallo F, Nanni P, Forni G. Vaccines for tumour prevention.
Nat Rev Cancer
2006
;
6
:
204
–16.
38
Bai A, Higham E, Eisen HN, Wittrup KD, Chen J. Rapid tolerization of virus-activated tumor-specific CD8+ T cells in prostate tumors of TRAMP mice.
Proc Natl Acad Sci U S A
2008
;
105
:
13003
–8.
39
Savage PA, Vosseller K, Kang C, et al. Recognition of a ubiquitous self antigen by prostate cancer-infiltrating CD8+ T lymphocytes.
Science
2008
;
319
:
215
–20.
40
Verdeil G, Marquardt K, Surh CD, Sherman LA. Adjuvants targeting innate and adaptive immunity synergize to enhance tumor immunotherapy.
Proc Natl Acad Sci U S A
2008
;
105
:
16683
–8.
41
Boon T, van der Bruggen P. Human tumor antigens recognized by T lymphocytes.
J Exp Med
1996
;
183
:
725
–9.
42
Heckelsmiller K, Rall K, Beck S, et al. Peritumoral CpG DNA elicits a coordinated response of CD8 T cells and innate effectors to cure established tumors in a murine colon carcinoma model.
J Immunol
2002
;
169
:
3892
–9.
43
Greenberg PD, Cheever MA, Fefer A. Eradication of disseminated murine leukemia by chemoimmunotherapy with cyclophosphamide and adoptively transferred immune syngeneic Lyt-1+2- lymphocytes.
J Exp Med
1981
;
154
:
952
–63.
44
Ibe S, Qin Z, Schuler T, Preiss S, Blankenstein T. Tumor rejection by disturbing tumor stroma cell interactions.
J Exp Med
2001
;
194
:
1549
–59.
45
Hiroishi K, Tüting T, Lotze MT. IFN-α-expressing tumor cells enhance generation and promote survival of tumor-specific CTLs.
J Immunol
2000
;
164
:
567
–72.
46
Dunn GP, Bruce AT, Sheehan KC, et al. A critical function for type I interferons in cancer immunoediting.
Nat Immunol
2005
;
6
:
722
–9.
47
Qin Z, Schwartzkopff J, Pradera F, et al. A critical requirement of interferon γ-mediated angiostasis for tumor rejection by CD8+ T cells.
Cancer Res
2003
;
63
:
4095
–100.
48
Ganss R, Arnold B, Hammerling GJ. Mini-review: overcoming tumor-intrinsic resistance to immune effector function.
Eur J Immunol
2004
;
34
:
2635
–41.
49
Palmer DC, Chan CC, Gattinoni L, et al. Effective tumor treatment targeting a melanoma/melanocyte-associated antigen triggers severe ocular autoimmunity.
Proc Natl Acad Sci U S A
2008
;
105
:
8061
–6.
50
Paulos CM, Wrzesinski C, Kaiser A, et al. Microbial translocation augments the function of adoptively transferred self/tumor-specific CD8+ T cells via TLR4 signaling.
J Clin Invest
2007
;
117
:
2197
–204.