Originally published as JCO Early Release 10.1200/JCO.2005.01.6816 on October 31 2005

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Luiten, R. M. Articles by de Gast, G. C. Search for Related Content PubMed PubMed Citation Articles by Luiten, R. M. Articles by de Gast, G. C. Related Articles Related Editorial

Immunogenicity, Including Vitiligo, and Feasibility of Vaccination With Autologous GM-CSF–Transduced Tumor Cells in Metastatic Melanoma Patients

Rosalie M. Luiten, Esther W.M. Kueter, Wolter Mooi, Maarten P.W. Gallee, Elaine M. Rankin, Winald R. Gerritsen, Shirley M. Clift, Willem J. Nooijen, Pauline Weder, Willeke F. van de Kasteele, Johan Sein, Paul C.M. van den Berk, Omgo E. Nieweg, Anton M. Berns, Hergen Spits, Gijsbert C. de Gast

From the Departments of Medical Oncology, Immunology, Pathology, Clinical Chemistry, Surgical Oncology, and Molecular Genetics, The Netherlands Cancer Institute, Amsterdam, the Netherlands; and Cell Genesys, Foster City, CA

Address reprint requests to G.C. de Gast, MD, PhD, Professor of Clinical Immunotherapy, Department of Medical Oncology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands; e-mail: g.d.gast{at}nki.nl.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
PURPOSE: To determine the feasibility, toxicity, and immunologic effects of vaccination with autologous tumor cells retrovirally transduced with the GM-CSF gene, we performed a phase I/II vaccination study in stage IV metastatic melanoma patients.

PATIENTS AND METHODS: Sixty-four patients were randomly assigned to receive three vaccinations of high-dose or low-dose tumor cells at 3-week intervals. Tumor cell vaccine preparation succeeded for 56 patients (88%), but because of progressive disease, the well-tolerated vaccination was completed in only 28 patients. We analyzed the priming of T cells against melanoma antigens, MART-1, tyrosinase, gp100, MAGE-A1, and MAGE-A3 using human leukocyte antigen/peptide tetramers and functional assays.

RESULTS: The high-dose vaccination induced the infiltration of T cells into the tumor tissue. Three of 14 patients receiving the high-dose vaccine showed an increase in MART-1– or gp100-specific T cells in the peripheral blood during vaccination. Six patients experienced disease-free survival for more than 5 years, and two of these patients developed vitiligo at multiple sites after vaccination. MART-1– and gp100-specific T cells were found infiltrating in vitiligo skin. Upon vaccination, the T cells acquired an effector phenotype and produced interferon- on specific antigenic stimulation.

CONCLUSION: We conclude that vaccination with GM-CSF–transduced autologous tumor cells has limited toxicity and can enhance T-cell activation against melanocyte differentiation antigens, which can lead to vitiligo. Whether the induction of autoimmune vitiligo may prolong disease-free survival of metastatic melanoma patients who are surgically rendered as having no evidence of disease before vaccination is worthy of further investigation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
Malignant melanoma is an immunogenic tumor, which has encouraged its use as tumor vaccine. Analyses of spontaneous immune responses in melanoma patients have identified a series of antigens, including melanocyte-differentiation antigens, cancer-testis antigens, and antigens from mutated gene products,¹ which enables monitoring of vaccination-induced immune responses in patients at the level of antigen-specificity of cytotoxic T lymphocytes. Priming of antitumor immunity is improved by appropriate adjuvants, such as the hematopoietic growth factor granulocyte macrophage colony-stimulating factor (GM-CSF).^2-4 Injection of GM-CSF causes local inflammation, enhanced dendritic cell (DC) maturation, migration, and increased human leukocyte antigen (HLA) –class II expression.⁵ We have previously performed a phase I trial of subcutaneous GM-CSF injections combined with low-dose interleukin-2 (IL-2) and interferon- (IFN-), and observed that GM-CSF is well tolerated by patients.⁶ Other studies of GM-CSF injections in melanoma patients have confirmed the low toxicity of GM-CSF and showed occasional tumor regressions or prolonged disease-free survival of stage III or IV melanoma patients after surgical resection of disease.^7-14

To ensure prolonged GM-CSF levels during vaccination, gene-modified tumor vaccines have been developed that locally produce cytokines at the site of injection.¹⁵ In the B16 murine melanoma model, Dranoff et al¹⁶ demonstrated that transduction of tumor cells with GM-CSF increases the immunogenicity of melanoma cells. We performed a phase I/II study in patients with stage IV melanoma using autologous tumor cells transduced with the GM-CSF gene. During the time course of our clinical study, Dranoff's group reported the feasibility and limited toxicity of vaccination with GM-CSF-transduced autologous tumor vaccines in melanoma patients with several indications for the priming of an immune response by vaccination.^17,18

In this article, we describe the occurrence of vitiligo and T-cell responses against known melanoma antigens in patients after GM-CSF–transduced tumor cell vaccination. To perform a detailed analysis of melanoma-reactive T-cell responses, we developed a set of 16 HLA/peptide tetramers to detect melanoma antigen–specific T cells in patients expressing HLA-A1, -A2, -A3, -A28, or -B7 molecules. We characterized the functional activity of the melanoma-reactive T cells by analysis of the phenotypic activation state, as well as their responsiveness to antigenic stimulation in vitro.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
Eligibility Criteria
Patients were eligible to enter the study if they had progressive stage IV malignant melanoma that was at least partially accessible for resection. Other criteria were Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; normal hepatic and renal function; normal hemoglobin, WBC, and platelet count; no use of systemic steroids; negative tests for hepatitis B and HIV; and magnetic resonance imaging scan showing no evidence of brain metastasis. Prior chemotherapy, immunotherapy, or radiotherapy were allowed if the patient had recovered from all toxic effects and treatment had been completed at least 4 weeks before tumor collection. The trial (M93CSF) protocol was approved by the local medical ethics committee, by the Dutch recombinant DNA advisory committee (Commissie Genetische Modificatie [COGEM]), and by the US Food and Drug Administration. Patients were reassessed after vaccine production and immediately before vaccination, and were allowed to proceed provided that the ECOG performance status was 2. Patients with metastasis to the CNS at that time were eligible for vaccination if they were asymptomatic, there were no signs of neurological impairment, and no tumor had exceeded 1 cm in diameter. Prior radiotherapy or chemotherapy had to be completed 3 weeks before vaccination, and the patient had to have recovered from any toxicity.

Preparation of Vaccine
Tumor for vaccine production was collected from 64 eligible patients by surgery of cutaneous or subcutaneous metastases, or metastases in lymph nodes, lung, liver, or stomach (Table 1). Tumor vaccine production was conducted under current Good Manufacturing Practices. A small portion of the tumor sample was reserved for histologic examination, and the remainder was transported under sterile conditions on wet ice to Somatix Therapy Corp (now Cell Genesys Inc, Alameda, CA). On arrival, samples of the tumor and transport medium were taken for sterility testing and the remaining tumor was mechanically dissociated and digested with collagenase to create a single-cell suspension, which was used to establish a primary culture. Aliquots of nontransduced cells were cryopreserved for delayed-type hypersensitivity (DTH) testing. The retroviral particles used to transduce the primary tumor cells were generated from the MFG-S-human (-hu) GM-CSF producer cell line. This producer cell line was created by the stable introduction, using electroporation, of a plasmid-based retroviral vector containing the human cDNA for GM-CSF into the genome of the -CRIP packaging cell line.¹⁹ In this vector, the Moloney murine leukemia virus (MoMuLV) long terminal repeat sequences are used to generate full-length RNA needed for the generation of viral particles and subgenomic mRNA analogous to the MoMuLV env mRNA, which is responsible for the expression of the GM-CSF transgene. The -CRIP packaging cell line is derived from NIH 3T3, and it provides the viral proteins gag, pol, and env from MoMuLV necessary for the encapsidation of the recombinant vector genome into transducing viral particles. -CRIP was specifically designed and constructed with safety features to prevent the encapsidation and mobilization of RNA molecules encoding viral structural gene products. This allows the generation of stocks of replication-deficient retroviral-transducing particles of MFG-S–huGM-CSF with a decreased frequency for generation of replication-competent retrovirus. The MFG-S–huGM-CSF producer cell line was extensively tested according to US United States Food and Drug Administration guidelines and stored as Retroviral Producer Master and Working Cell Banks. The transducing particles produced from the MFG-S–huGM-CSF Retroviral Producer Working Cell Bank in defined lots used for the transduction of the tumor cells were also tested as required. Transduction of the tumor cell cultures was performed as described previously.^20,21 Briefly, a freshly thawed supernatant containing the viral particles from the producer cell line was layered onto primary tumor cell cultures, seeded at the first passage at a density of approximately 10⁴ cells/cm², and left for 12 to 16 hours. The GM-CSF production of the transduced cultures cells was measured by enzyme-linked immunosorbent assay, and if the production was less than 40 ng/10⁶ cells/24 hours, a second round of transduction was performed. After removal of the viral supernatant, propagation of the tumor cell cultures continued for up to two additional passages to meet the required dose of cells needed for vaccination. Cultures were collected, irradiated with 150 Gy, and cryopreserved in liquid nitrogen. Transduced cells were shown to be free of replication competent retrovirus according to US United States Food and Drug Administration guidelines by bioamplification on Mus dunni cells followed by quantitative measurement of vector mobilization by immunocytochemistry. All cultured tumor cells used in the study were free of endotoxin, bacteria, and mycoplasma before use. Immediately before injection of vaccine and DTH cells, vials of tumor cells were thawed rapidly at 37°C, washed twice with Hanks' Balanced Salt Solution (HBSS; Life Sciences Invitrogen, Breda, the Netherlands), checked for viability by trypan blue exclusion, re-suspended in HBSS, and used within 20 minutes. The viability of transduced cells immediately before injection ranged from 77% to 99%. The final dose for vaccine cells was 5 x 10⁶ cells/mL (low dose) or 5 x 10⁷ cells/mL (high dose), and 5 x 10⁵ cells/mL for the DTH cells. Establishment of a primary tumor cell culture and transduction with GM-CSF was successful for 56 (88%) of 64 tumor samples. Tumor cells obtained from six patients, comprising three subcutaneous/cutaneous metastases, two lymph node metastases, and one soft tissue metastasis, failed to proliferate. For two patients, an insufficient amount of tumor was available.

Two days after the first vaccination (day 3) and 4 days after the third vaccination (day 47), 4-mm punch biopsies were taken from the vaccination and DTH sites. At days 1 (day of vaccination), 2, 3, 8, 15, and 21 after each vaccination, the patient was examined, vital signs were recorded, local and systemic effects were noted, and toxicity was scored using the National Cancer Institute Common Toxicity Criteria. A CBC, differential, and absolute eosinophil count were measured. Serial blood samples were taken for measurement of urea, electrolytes, serum biochemistry, C-reactive protein, and replication-competent retrovirus. Serum GM-CSF levels were measured immediately before and 6, 24, 48, and 120 hours after the first vaccination. Tumor sites were evaluated weekly. Radiologic evaluation was performed before vaccination, 6 and 9 weeks after the first vaccination, and thereafter when indicated. In patients with asymptomatic small brain metastases, the magnetic resonance imaging scan was repeated before each vaccination. New accessible metastases were removed under local anesthesia from 17 patients during and/or after vaccination (Table 2). Blood samples were taken before vaccination and on days 1, 3, 22, 24, 43, 45, and 64 during vaccination. Peripheral blood mononuclear cells (PBMCs) were isolated by density centrifugation on a Ficoll gradient and frozen in aliquots of 5 x 10⁶ to 10 x 10⁶ cells in liquid nitrogen.

Synthesis of HLA/Peptide Tetramers
Synthetic peptides were produced by standard fluorenylmethoxycarbonyl chemistry. Soluble allophycocyanin (APC) -labeled HLA/peptide tetramers were produced as described previously.²² HLA-A28 and -B7 heavy-chain biotinylated complexes were provided by Pierre Coulie, PhD (Ludwig Institute for Cancer Research, Brussels, Belgium), and by Ton Schumacher, PhD (The Netherlands Cancer Institute, Amsterdam, the Netherlands), respectively.

Detection of Tetramer-Binding T Cells Among PBMCs
Frozen PBMCs were thawed and incubated with 10 U/mL DNase (Sigma, Zwijndrecht, the Netherlands) in Iscove's medium supplemented with 8% fetal calf serum (Life Sciences Invitrogen, Breda, the Netherlands) for 2 hour at room temperature. Cells were washed and, per staining, 1 x 10⁶ cells were incubated with APC-conjugated HLA/peptide tetramer in phosphate-buffered saline (PBS) supplemented with 1% bovine serum albumin (PBSA) for 15 minutes at 37°C. After washing, the cells were incubated with fluorescein isothiocyanate (FITC) –conjugated anti-CD5 Ab, phycoerythrin (PE) -conjugated anti-CD4, anti-CD19 and anti–T-cell receptor (TCR) - Abs (all antibodies obtained from Becton Dickinson) in PBSA for 20 minutes on ice. Cells were washed and analyzed on a FACScalibur (Becton Dickinson). Dead cells were excluded from analysis by propidium iodide staining. For phenotypic analysis, PBMCs were stained with FITC-conjugated anti-CD27 or anti-CD45RO Ab, PE-conjugated anti-CD8ß, Cy5-conjugated anti-CD45RA Abs, and APC-conjugated tetramer. Because of limited numbers of PBMCs available for analysis, the reactivity with the A2/gp100(154) tetramer was tested neither in patient 45 and patient 46 at day 43 and day 64, nor in patient 58 at days 22, 43 and 45. In patient 41, A1/tyrosinase and A3/flu tetramer reactivity was not tested at day 22, and the reactivity with the A1/flu tetramer was tested only at days 3 and 43.

Detection of Melanoma-Reactive T Cells in the Vitiligo Perilesional Skin
A 4-mm skin biopsy sampled at the perilesional skin of a vitiligo lesion of patient 46 was cultured for 1 week in Yssels medium²³ supplemented with 1% human serum and 20 U/ml recombinant human (rh) IL-2 (Proleukin, Chiron, Amsterdam, The Netherlands). Cells that migrated out of the tissue were treated with erythrocyte lysis buffer (155 mmol/L NH₄Cl; 10 mmol/L KHCO₃; 0.1 mmol/L EDTA acid, pH 7.4), washed and stained with APC-conjugated A2/gp100(154) tetramer, PE-conjugated A2/MART-1 tetramer and FITC-conjugated anti-CD8 Ab (Becton Becton Dickinson). Cells were analyzed on a FACScalibur and dead cells were excluded from analysis by propidium iodide staining.

T-Lymphocyte Culture and Cloning
CD8⁺ T cells were isolated by a magnetically activated cell sorter using anti-CD8 Ab-coated beads (Miltenyi Biotec, Bergisch Gladbach, Germany), and cultured at a ratio of 1:2 with irradiated (40 Gy) autologous CD8^– cells that were preloaded with 100 µmol/L of each peptide in Yssels medium²³ supplemented with 1% human serum and 20 U/mL rhIL-2. Cultures were restimulated weekly with irradiated, peptide-loaded autologous CD8^– cells. Antigen-specific T cells were detected in the cultures by tetramer staining. Limiting dilution of tetramer-positive T cells was performed by fluorescence-activated cell sorting (FACStar Plus, Becton Dickinson) at 100 cells/well with 30, 10, or 3 cells, or 1 cell per well densities in round-bottom, 96-well plates. Sorted cells were stimulated biweekly with 5 x 10⁴/well irradiated allogeneic PBMCs, 0.5 x 10⁴ cells/well JY, 100 ng/mL phytohemagglutinin (HA16, Murex, Dartford, UK) and 20 U/mL rhIL-2 until growing cultures were visible for cloning efficiency determination. Clones were tested for specific tetramer binding and for specific target-cell recognition in ⁵¹Cr release assays, as described previously.²⁴

T-Cell Activation Assays
T-cell activation was determined by the production of IFN- upon stimulation, using the intracellular IFN- staining kit (BD Pharmingen, Heidelberg, Germany). Per stimulation, 1 x 10⁶ PBMCs were stimulated with 1 µmol/L peptide, Brefeldin A (Golgiplug 1:1000 dilution, BD Pharmingen) and 20 U/ml rhIL-2 for 5 hour at 37°C. Positive control incubations for activation were performed in 10 ng/mL phorbol 12-myristate-13-acetate and 2 µmol/L ionomycin. Cells were washed and incubated with APC-conjugated tetramers and PE-conjugated anti-CD8ß Ab (Coulter Immunotech, Marseille, France). Subsequently, the cells were fixed, permeabilized, and stained with anti–IFN- Ab, according to the manufacturer's protocol (BD Pharmingen).

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
Feasibility and Toxicity of the Vaccination
Twenty-eight stage IV metastatic melanoma patients received three vaccinations with autologous tumor cells that were retrovirally transduced with a retroviral vector encoding the human GM-CSF gene in a phase I/II clinical study. Table 1 shows the individual patient demographics, previous therapy, vaccination dose, and GM-CSF production by the vaccine. Vaccine preparation succeeded for the majority (56) of the original 64 eligible patients in the study (88%), and the treatment was well tolerated. Because of the advanced disease stages of melanomas that were included in our study, 39 (61%) of 64 patients entered the vaccination protocol, of whom only 28 patients completed three vaccinations. Therefore, possible selection of slow-growing disease cannot be excluded.

During the vaccination, no significant changes in vital signs or hepatic, renal, or other organ function were observed. Two patients developed a generalized urticarial rash 10 days after the second injection. They had associated facial and periorbital edema, which resolved with antihistamines. One week after a test dose of 0.5 x 10⁶ nontransduced cells, the patients received the third vaccination, with antihistamines, and experienced no further adverse effects. Systemic toxicity included fever of less than grade 2 in 30% of vaccinations and mild to moderate headache in 33%. Fatigue and generalized pruritus occurred in 20% of patients. These toxicities did not correlate with the dose level of the vaccine, nor were the effects cumulative. Pruritus at the site of vaccination was the most common toxicity, observed with 85% of the vaccinations. The occasionally observed edema and local induration at the site of vaccination caused discomfort.

Clinical Responses and Vitiligo Development
Long-term survival of more than 5 years was observed for six patients (patients 8, 11, 29, 46, 50, and 64; Table 1). Three of these patients have remained free of disease (patients 8, 46, and 64) and one patient (patient 29) had a subcutaneous metastasis removed after 6 months and has remained free of disease since. Two patients (patients 48 and 63) had stable disease with a duration of 6 months. Nonassessable disease was noted in 9 patients because all tumor sites either had been removed by surgery for vaccine preparation or had disappeared after previous radiotherapy (Table 1). Relapse occurred in 6 of the 9 patients with nonassessable disease after 4 to 34 months (median, 6 months).

Two patients (patients 46 and 64) of the high-dose tumor-cell vaccine group developed vitiligo lesions, an autoimmune disease directed against the melanocytes, at multiple sites a few months after vaccination. These patients experienced long-term disease-free survival of more than 84 and 67 months, respectively (Table 1).

Effector-Cell Recruitment
Metastases were resected from 17 patients during or after vaccination, of which all patients who received the high-dose tumor cell vaccine showed a clear infiltration of T cells and plasma cells at the tumor site (Table 2; Figs 1A and 1B). This infiltration was absent in the metastases that were resected before vaccination, except for patient 53, who had minimal T-cell infiltration in one metastasis before vaccination. The presence of necrosis in the resected metastases did not, however, correlate with the presence of infiltrating T cells. Infiltration of T cells into the tumor was found in four of seven patients analyzed who received the high-dose vaccine and in three of 10 patients analyzed who received the low-dose vaccine (Table 2). Peritumoral infiltration was present in three and six patients of each group, respectively. Moreover, T-cell infiltration was found at a higher density in the patients receiving the high-dose vaccine as compared with patients receiving the low-dose vaccine, suggesting a dose effect of the vaccination.

View larger version (114K): [in this window] [in a new window]   Fig 1. Cellular infiltration. Metastases were stained with (A) anti-CD3 antibody (Ab) or (B) hematoxylin and eosin. Sections of biopsies of the vaccination site after the third vaccination at day 47 were stained with (C) anti-CD3 Ab, (D) anti-CD4 Ab, (E) anti-CD8, (F) anti-CD1a, or (G) anti-CD14. Pictures are representative of all biopsies analyzed.

The vaccination induced eosinophilia in the peripheral blood in all patients with peak levels between 1 and 3 days after each vaccination (Fig 2). Patients receiving the high-dose vaccine (Fig 2B) developed more eosinophilia than patients receiving the low-dose vaccine (Fig 2A). GM-CSF was not found in the serum after vaccination.

View larger version (18K): [in this window] [in a new window]   Fig 2. Characteristics of blood eosinophilia on vaccination. Eosinophil counts in the peripheral blood of vaccinated patients were recorded at 1- to 5-day intervals during and after vaccination. Graphs show the average eosinophil counts and standard deviation of the patients receiving (A) the low-dose vaccine, and (B) the high-dose vaccine.

View larger version (22K): [in this window] [in a new window]   Fig 3. Detection of T cells reactive with melanoma antigens. Reactivitity of peripheral blood mononuclear cells of patient 46 with A2/MART-1 tetramer in the CD5-high, CD4–, CD19–, TCR-– population, consisting of > 95% CD8+ T cells (histogram). The numbers in the dot plots indicate the percentage of tetramer-reactive cells of CD5-high, CD4–, CD19–, TCR-– population. No reactivity with tetramers was observed in the CD5–, CD4–, CD19–,TCR– population.

View larger version (29K): [in this window] [in a new window]   Fig 4. Melanoma-reactive T cells in the peripheral blood. T-cell response during and after vaccination of patients 46 and 64 who developed vitiligo after vaccination and of patients 45, 58, 19, and 41, using HLA-A1, A2 or A3 tetramers. (A) Patient 46, HLA-A2; (B) patient 46, HLA-A3; (C) patient 64, HLA-A2; (D) patient 45, HLA-A2; (E) patient 58, HLA-A2; (F) patient 19, HLA-A3; (G) patient 41, HLA-A1; and (H) patient 41, HLA-A3. HLA, human leukocyte antigen; tyr, tyrosinase; flu, influenza virus.

T-Cell Activation and Phenotype
To determine the functionality of the melanoma-reactive T cells in the vaccinated patients, we analyzed the phenotype ex vivo as well as antigen-specific T-cell activation. The A2/MART-1–reactive T cells of patient 46 consisted of CD27⁺ CD45RA⁺-naive cells, CD45RA^–-activated/-memory cells, and CD27^– CD45RA⁺-effector cells (Fig 5A). ²⁶ During vaccination, the number of CD45RO^– CD45RA⁺ cells had increased from 10% at day 1 to 21% at day 45. This accumulation in CD45RA⁺ cells represents an increase in effector-cell population, given that the number of CD27⁺ cells had not changed. The MART-1–reactive T cells from patient 46 produced IFN- on ex vivo stimulation with the A2/MART-1 peptide, and not by control A2/flu-virus peptide stimulation (Fig 5B), indicating MART-1–specific T-cell activation. To evaluate whether these T cells were involved in the vitiligo development, we analyzed the T cells isolated from a biopsy taken from the perilesional vitiligo skin. HLA-A2–/MART-1–reactive T cells were present at a high percentage (3.05%) as well as a low level of A2/gp100(154)-specific T cells (0.44%; Fig 5C), indicating that MART-1– and gp100-reactive T cells had migrated to the skin.

View larger version (34K): [in this window] [in a new window]   Fig 5. Autoimmunity in vitiligo patient 46. (A) Percentage of cells expressing CD45RA, CD27, or CD45RO of A2/MART-1–tetramer+ T cells at days 1 and 45. (B) Intracellular interferon- staining of A2/MART-1–tetramer+ T cells with A2/flu or MART-1 peptide, or PMA/ionomycine. (C) A2/MART-1 and A2/gp100(154) tetramer+, viable (PI–) CD8+ T cells in perilesional vitiligo skin. Flu, influenza virus; PMA, phorbol 12-myristate-13-acetate; PI, propium iodide.

View larger version (29K): [in this window] [in a new window]   Fig 6. Activation of melanoma-reactive T cells during vaccination. (A) Phenotype of A3/flu- (upper panel) or A3/MAGE-A1–tetramer+ T cells (lower panel) of patient 19 at day 3 (above) and 45 (below), representative of all 6 patients analyzed. B. Interferon- production by A3/flu- or A3/MAGE-A1–tetramer+ T cells of patient 19 stimulated with A3/flu peptide, A3/MAGE-A1 peptide, or without peptide. Flu, influenza virus.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

We conclude that the vaccination can enhance melanoma immunity on the basis of the following findings. The vaccination induced infiltration of T cells into metastases that arose during treatment. The vaccination sites were infiltrated by T cells, macrophages, DCs, and eosinophils, and increased levels of melanoma antigen–specific T cells were found in the peripheral blood. Two patients developed vitiligo after vaccination, indicating that GM-CSF–transduced tumor-cell vaccination may even break tolerance to self-antigens expressed on melanoma cells and melanocytes. Because spontaneous vitiligo occurs in less than 0.1% of stage IV melanoma patients (G.C. de Gast, unpublished observations), the observed incidence of vitiligo in two (14%) of 14 patients in the high-dose vaccine group can be considered a vaccination effect. We found T cells that are specific for melanocyte-differentiation antigens, shared between normal melanocytes and melanoma cells, in vitiligo skin and in the peripheral blood. These T cells became more activated toward effector cells during vaccination and produced IFN- on specific peptide stimulation. The tumor of vitiligo patient 46 homogenously expressed MART-1 and gp100 at the time of excision for vaccine preparation (Table 4), which suggests that the in vivo effective T-cell response, as judged by the autoimmunity, played a role in preventing the relapse of the melanoma (Table 1). These results are in agreement with earlier observations of a possible correlation between autoimmunity and relapse-free survival of advanced melanoma patients.²⁷

In other studies of GM-CSF–transduced tumor-cell vaccination in melanoma patients, no vitiligo development was observed.^17,18 Whether vitiligo develops during immunotherapy depends on the antigen specificity of the primed T cells. Melanoma regression in combination with vitiligo development was observed on adoptive transfer of MART-1– and gp100-specific T cells,^28,29 or on vaccination with DCs pulsed with gp100, MART-1, or tyrosinase peptides.^30,31 Furthermore, blocking T-cell regulation with antibodies against CTLA-4 during gp100-peptide vaccination enhanced breakage of tolerance to the gp100 antigen, resulting in melanoma regression and vitiligo.³² In the GM-CSF–transduced melanoma vaccination study of Soiffer et al,^17,33 the antigen specificity of the T-cell responses was characterized to be against the melanoma inhibitor of apoptosis protein (ML-IAP). ML-IAP is not expressed in normal melanocytes³⁴ and, therefore, will not cause autoimmunity towards melanocytes and depigmentation. We have investigated the T-cell responses against known melanoma antigens, and found that the development of vitiligo and disease-free survival after vaccination coincided with increased levels of activated MART-1– and gp100-specific T cells in the periphery and infiltrating in vitiligo. These results suggest that the GM-CSF–transduced tumor cell vaccination can activate T cells to exert lytic activity against normal melanocytes as well as against residual melanoma cells in vivo.

Antitumor immune responses were mostly found in patients treated with the high-dose vaccine, which indicates that the high-dose vaccine was more effective than the low-dose vaccine. In patients who did not respond to the vaccination, we observed that the melanoma antigen–specific T cells were not activated to grow or produce cytokine on ex vivo antigenic stimulation, which may be due to the advanced disease stage. Nonresponsiveness of T cells toward tumors has been observed frequently in a variety of cancer patients^35,36 and should be overcome to achieve effective antitumor immunity. Several mechanisms that explain this lack of responsiveness have been described. For example, most tumor antigens are self-antigens to which tolerance exists. Moreover, tumor antigens are not presented in an inflammatory environment, which may lead to the induction of T-cell tolerance. GM-CSF has a major effect on the activation and function of DCs, and may provide the required stimulus for the DCs to prime T cells against tumor antigens. On the other hand, anergy to self-antigens may persist as a result of the continuous level of self-antigen presentation. Vitiligo development was observed in two of the three patients in the high-dose vaccine group with nonassessable disease during vaccination. In these patients, the melanocyte-differentiation antigens (shared among tumor cells and melanocytes) were present at a lower level because of the absence of tumor tissue. This temporary lower level of antigen may have facilitated the T cells' regaining their responsiveness to these self-antigens, as was described for CD4⁺ T cells in a TCR-transgenic murine model.³⁷ These results suggest that the tumor cell vaccinations performed here are most effective in breaking tolerance to self-antigens and inducing autoimmunity when applied in a minimal residual disease setting. To achieve activation of melanoma-reactive T cells in the nonresponding patients of this study, in whom tumor tissue was present during vaccination, may, therefore, require more than three vaccinations. Indeed, the patients who responded to GM-CSF–transduced melanoma vaccines described by Soiffer et al received 6 or more weekly or biweekly vaccinations.^17,18 They did not observe melanoma immunity in patients who were vaccinated with only three doses every 28 days, which vaccination schedule is most comparable to our study. Likewise, the clinical and immunologic effects in other clinical trials of cytokine-secreting tumor cell vaccines were observed only in patients who were subjected to frequent vaccinations at short time intervals,^18,21,38-44 or following adoptive transfer of primed lymph node cells.⁴⁵

We describe here the long-term survival of six (21%) of 28 vaccinated stage IV melanoma patients for more than 5 years. Other vaccination trials of stage III or IV melanoma patients with various types of vaccines, including peptides, protein, DCs, tumor cell lysates, autologous tumor cells, or cytokine-transduced tumor cells, have reported 3-year survival rates ranging from 5% to 30%.^{17,18,38,46-49} Chapman et al⁵⁰ observed an even higher 3-year survival rate of 71% after vaccination with anti-GD3 ganglioside antibody. However, many clinical trials have not yet reached a long-term follow-up,^30,39,51-55 and studies in which advanced melanoma patients are followed for more than 5 years are, therefore, less abundant. Dinitrophenyl-haptenated tumor cell vaccination resulted in 30% survival after 5 years.⁵⁶ A vaccination study with tumor cells mixed with GM-CSF revealed a 5-year survival of 23%,⁵⁷ which is comparable to our study. It is, however, intriguing to note that in our study all patients (three of three) in the high-dose vaccination group that entered the vaccination with nonassessable disease experienced prolonged survival, as well as half of the patients (three of six) with nonassessable disease receiving the low-dose vaccine (Table 1). This observation compares beneficially to the reported the 5-year survival of 21% of patients with resected stage III or IV melanoma without vaccination, or 41% after allogeneic tumor–cell vaccination.⁵⁸ Moreover, Spitler et al¹⁰ have reported that GM-CSF administration can prolong survival by 4 years in 50% of resected stage III or IV melanoma patients. Although the numbers in our study are too low for statistical conclusions, our results suggest that GM-CSF–transduced tumor cell vaccination may be most effective in prolonging survival in an adjuvant setting.

Together, the results of our study show that the GM-CSF–transduced autologous tumor cell vaccination can activate melanoma-specific T-cell responses and autoimmunity in stage IV metastatic melanoma patients. Whether the induction of autoimmune vitiligo may prolong disease-free survival of metastatic melanoma patients who are surgically rendered as having no evidence of disease before vaccination is worthy of further investigation.

	Authors' Disclosures of Potential Conflicts of Interest

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
The authors indicated no potential conflicts of interest.

	Acknowledgment

 
We thank Pierre Coulie, PhD, and Pierre van der Bruggen, PhD, for kindly providing the HLA-A28 monomers, HLA-A28/MUM-3 tetramers, and the MAGE-specific T cell clones. We thank Ton Schumacher, PhD, for kindly providing HLA-B7 monomers, and Donnee Majoor and Irene Urlus for excellent assistance.

	NOTES

 
Supported by Cell Genesys, Foster City, CA. R.M.L. and E.W.M.K. were supported by Dutch Cancer Society Grant No. NKI99-2048.

Terms in blue are defined in the glossary, found at the end of this issue and online at www.jco.org.

Authors' disclosures of potential conflicts of interest are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Authors' Disclosures of... REFERENCES

 
1. Van den Eynde BJ, van der Bruggen P: T cell defined tumor antigens. Curr Opin Immunol 9:684-693, 1997[CrossRef][Medline]

2. Disis ML, Bernhard H, Shiota FM, et al: Granulocyte-macrophage colony-stimulating factor: An effective adjuvant for protein and peptide-based vaccines. Blood 88:202-210, 1996[Abstract/Free Full Text]

3. Kwak LW, Young HA, Pennington RW, et al: Vaccination with syngeneic, lymphoma-derived immunoglobulin idiotype combined with granulocyte/macrophage colony-stimulating factor primes mice for a protective T-cell response. Proc Natl Acad Sci U S A 93:10972-10977, 1996[Abstract/Free Full Text]

4. Mellstedt H, Fagerberg J, Frodin JE, et al: Augmentation of the immune response with granulocyte-macrophage colony-stimulating factor and other hematopoietic growth factors. Curr Opin Hematol 6:169-175, 1999[CrossRef][Medline]

5. Jones T, Stern A, Lin R: Potential role of granulocyte-macrophage colony-stimulating factor as vaccine adjuvant. Eur J Clin Microbiol Infect Dis 13:47-53, 1994 (suppl 2)[CrossRef]

6. de Gast GC, Klumpen HJ, Vyth-Dreese FA, et al: Phase I trial of combined immunotherapy with subcutaneous granulocyte macrophage colony-stimulating factor, low-dose interleukin 2, and interferon alpha in progressive metastatic melanoma and renal cell carcinoma. Clin Cancer Res 6:1267-1272, 2000[Abstract/Free Full Text]

7. Hoeller C, Jansen B, Heere-Ress E, et al: Perilesional injection of r-GM-CSF in patients with cutaneous melanoma metastases. J Invest Dermatol 117:371-374, 2001[CrossRef][Medline]

8. Ridolfi L, Ridolfi R, Ascari-Raccagni A, et al: Intralesional granulocyte-monocyte colony-stimulating factor followed by subcutaneous interleukin-2 in metastatic melanoma: A pilot study in elderly patients. J Eur Acad Dermatol Venereol 15:218-223, 2001[CrossRef][Medline]

9. Si Z, Hersey P, Coates AS: Clinical responses and lymphoid infiltrates in metastatic melanoma following treatment with intralesional GM-CSF. Melanoma Res 6:247-255, 1996[CrossRef][Medline]

10. Spitler LE, Grossbard ML, Ernstoff MS, et al: Adjuvant therapy of stage III and IV malignant melanoma using granulocyte-macrophage colony-stimulating factor. J Clin Oncol 18:1614-1621, 2000[Abstract/Free Full Text]

11. Hunger RE, Brand CU, Streit M, et al: Successful induction of immune responses against mutant ras in melanoma patients using intradermal injection of peptides and GM-CSF as adjuvant. Exp Dermatol 10:161-167, 2001[CrossRef][Medline]

12. Jager E, Ringhoffer M, Dienes HP, et al: Granulocyte-macrophage-colony-stimulating factor enhances immune responses to melanoma-associated peptides in vivo. Int J Cancer 67:54-62, 1996[CrossRef][Medline]

13. Leong SP, Enders-Zohr P, Zhou YM, et al: Recombinant human granulocyte macrophage-colony stimulating factor (rhGM-CSF) and autologous melanoma vaccine mediate tumor regression in patients with metastatic melanoma. J Immunother 22:166-174, 1999

14. Scheibenbogen C, Schmittel A, Keilholz U, et al: Phase 2 trial of vaccination with tyrosinase peptides and granulocyte-macrophage colony-stimulating factor in patients with metastatic melanoma. J Immunother 23:275-281, 2000

15. Dranoff G: GM-CSF-secreting melanoma vaccines. Oncogene 22:3188-3192, 2003[CrossRef][Medline]

16. 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 90:3539-3543, 1993[Abstract/Free Full Text]

17. 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 95:13141-13146, 1998[Abstract/Free Full Text]

18. Soiffer R, Hodi FS, Haluska F, et al: Vaccination with irradiated, autologous melanoma cells engineered to secrete granulocyte-macrophage colony-stimulating factor by adenoviral-mediated gene transfer augments antitumor immunity in patients with metastatic melanoma. J Clin Oncol 21:3343-3350, 2003[Abstract/Free Full Text]

19. Danos O, Mulligan RC: Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc Natl Acad Sci U S A 85:6460-6464, 1988[Abstract/Free Full Text]

20. Sanda MG, Ayyagari SR, Jaffee EM, et al: Demonstration of a rational strategy for human prostate cancer gene therapy. J Urol 151:622-628, 1994[Medline]

21. Simons JW, Jaffee EM, Weber CE, et al: Bioactivity of autologous irradiated renal cell carcinoma vaccines generated by ex vivo granulocyte-macrophage colony-stimulating factor gene transfer. Cancer Res 57:1537-1546, 1997[Abstract/Free Full Text]

22. Altman JD, Moss PA, Goulder PJ, et al: Phenotypic analysis of antigen-specific T lymphocytes. Science 274:94-96, 1996[Abstract/Free Full Text]

23. Yssel H, De Vries JE, Koken M, et al: Serum-free medium for generation and propagation of functional human cytotoxic and helper T cell clones. J Immunol Methods 72:219-227, 1984[CrossRef][Medline]

24. Hooijberg E, Ruizendaal JJ, Snijders PJ, et al: Immortalization of human CD8+ T cell clones by ectopic expression of telomerase reverse transcriptase. J Immunol 165:4239-4245, 2000[Abstract/Free Full Text]

25. Denkberg G, Cohen CJ, Reiter Y: Critical role for CD8 in binding of MHC tetramers to TCR: CD8 antibodies block specific binding of human tumor-specific MHC-peptide tetramers to TCR. J Immunol 167:270-276, 2001[Abstract/Free Full Text]

26. Hamann D, Baars PA, Rep MH, et al: Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med 186:1407-1418, 1997[Abstract/Free Full Text]

27. Sanderson K, Scotland R, Lee P, et al: Autoimmunity in a phase I trial of a fully human anti-cytotoxic T-lymphocyte antigen-4 monoclonal antibody with multiple melanoma peptides and Montanide ISA 51 for patients with resected stages III and IV melanoma. J Clin Oncol 23:741-750, 2005[Abstract/Free Full Text]

28. Dudley ME, Wunderlich JR, Robbins PF, et al: Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science 298:850-854, 2002[Abstract/Free Full Text]

29.