Patient-derived renal cell carcinoma cells fused with allogeneic dendritic cells elicit anti-tumor activity: in vitro results and clinical responses

Abstract

Renal cell carcinoma (RCC) has been shown to be susceptible to immunotherapeutic treatment strategies. In the present study, patient-derived tumor cells were fused with allogeneic dendritic cells (DC) to elicit anti-tumor activity against RCC. DC from HLA-A2+ healthy donors were fused with primary RCC cells from ten patients. Phenotype of fusion cells were characterized by flow cytometer and confocal microscopy. In vitro, T cell proliferation, IFN-γ secretion and cytotocic T lymphocytes (CTL) activity elicited by allogeneic DC/RCC fusion cells were assessed. Clinically, ten patients were vaccinated with allogeneic DC/RCC fusion vaccine. The adverse effects and toxicity were observed. The clinical response was evaluated by CT scans. After fusion, the created hybrids expressed both tumor associated antigen and DC-derived molecules and could stimulate the proliferation and IFN-γ secretion of T cells as well as elicit strong CTL activity against RCC cells in vitro. In vivo, no serious adverse effects, toxicity, or signs of autoimmune disease were observed after vaccination therapy. Percentage of T lymphocyte subsets in peripheral blood of patients was increased significantly. One of ten patients exhibited a partial response with regression of lung metastases. Six patients showed stable disease with stabilization of previously progressive disease (follow up 1.5 years). The PR and SD responses, exhibited by 7/10 patients who received the allogeneic DC/RCC fusion vaccine treatment, suggest that this approach is safe and can elicit immunological responses in a significant portion of patients with RCC.

Introduction

Renal cell carcinoma (RCC) is the most lethal carcinoma of genitourinary tumors, and accounts for approximately 3% of the adult malignancies. Currently surgical resection, essentially radical nephrectomy, is the first line of treatment for primary RCC. However, in 25–30% of newly diagnosed patients there are already demonstrated metastases. Metastases also occurred in 20–30% of the cases with radical nephrectomy. Moreover, RCC respond minimally to standard chemotherapeutic agents, hormones and radiotherapy. Hence, the prognosis for RCC patients with metastases is very poor with a 5-year survival rate of less than 10% [1, 2]. Novel effective therapy is urgently needed. RCC are considered to be one of the most immunogenic tumors, and has been demonstrated particularly susceptible to immune-based treatment. Indeed, cytokine-based therapies have lead to clinical responses in a small subset of patients. However, toxicity and adverse effects of these approaches are substantial [3–8]. Perhaps, targeted immunotherapy may represent a promising strategy with minimal toxicity.

Targeted immunotherapy can be achieved by manipulating dendritic cells (DC). DC are potent antigen-presenting cells and are uniquely effective in stimulating naive T lymphocytes to generate primary immune response [9]. Various protocols have been developed to generate DC in vitro from precursor populations in peripheral blood, bone marrow or cord blood [10]. These protocols permit the in vitro manipulation of DC for clinical and laboratory studies [11]. A large amount of strategies have been developed to load tumor antigen into DC, including pulsing or incubation DC with tumor cell lysate, tumor antigen peptides, specific tumor proteins, or delivery of tumor protein gene into DC by way of virus vectors as well as the use of whole tumor cells as a source of antigens [12–19].

An alternative approach is the fusion of DC and tumor cells. In this approach, multiple tumor-associated antigens, including those yet unidentified, can be processed and presented by DC to CD8+ and CD4+ T cells in the context of HLA class I and class II molecules. Clinical trial studies of autologous DC/tumor fusion vaccine have showed moderate immunologic responses. However, very few clinical trials of allogeneic DC/tumor fusion vaccine are reported. In this study, patient-derived tumor cells were fused with allogeneic DC to generate the vaccine for RCC. Here, significant data is presented on the characterization and effectiveness of adoptively transferred allogeneic DC/tumor cell treatments.

Materials and methods

Patients

This study was conducted with approval of the Sun Yat-Sen University Cancer Center institutional ethics committee, and written consent was obtained from each patient. Ten patients with renal cell carcinoma treated at the Cancer Center, Sun Yat-Sen University from October 2004 to June 2006 were enrolled in the study (Table 1). Patients were selected using the following inclusion criteria: histologically proven renal cell carcinoma; stage IV disease; Karnofsky Performance scores at least 70; age between 18 and 70; no chemotherapy, radiotherapy or cytokine treatment before surgery; HLA-A2 positive; bidimensionally measurable metastatic lesions that could not be removed and cultured autologous tumor cells available. Patients were excluded if they had autoimmune disease, active brain metastases, decompensated heart failure, severe psychiatric disease, and active hepatitis A, B, C or HIV.

Table 1.

Patient characteristics

Patient no.	Age/sex (years)	pTNM	Location of metastases	Location of measurable disease
1 (RCC1)	44/M	T3bN1M1	Lungs, lymph nodes	Lungs
2 (RCC2)	43/F	T4N1M1	Lungs, bones, lymph nodes	Lungs, bones
3 (RCC3)	45/F	T3aN1M1	Lungs, lymph nodes	Lungs
4 (RCC4)	52/M	T3aN1M1	Liver, lymph nodes	Liver
5 (RCC5)	58/M	T3cN1M1	Lungs, lymph nodes	Lungs
6 (RCC6)	52/F	T3bN0M1	Lungs	Lungs
7 (RCC7)	55/M	T4NXM1	Lungs	Lungs
8 (RCC8)	53/M	T4N1M1	Lungs, lymph nodes	Lungs
9 (RCC9)	42/F	T4N1M1	Lungs, lymph nodes	Lungs
10 (RCC10)	55/M	T3cN1M1	Lungs, lymph nodes	Lungs

Isolation and culture of patient-derived RCC cells

RCC Cells were isolated from primary tumors obtained within 1 h of surgery. The tumor tissue was minced and digested in RPMI 1640 medium (GIBCO, Carlsbad, CA) containing collagenase IV (1 mg/ml) for 2 h in a shaking water bath (37°C). The cell suspension was obtained by passing through a cell strainer to remove the residue. To obtain purified RCC cells, the fibroblasts, red blood cells, monocytes and T cells were removed using serum column [20]. The RCC cells were cultured in RPMI 1640 medium supplemented with 10% autologous heat-inactivated serum. After expansion of several days, the primary RCC cells were frozen in liquid nitrogen until used as fusion partners and targets for cytotocic T lymphocytes (CTL) assay. For detection of delayed-type hypersensitivity reaction, tumor cells were lysed by five freeze-thaw cycles. Supernatants were passed through a 0.22 μm filter and stored at −80°C until use.

Generation of monocyte-derived allogeneic dendritic cells and autologous T cells

Peripheral blood mononuclear cells (PBMC) were separated from 60 ml peripheral blood of HLA-A2+ healthy donors by Ficoll density-gradient centrifugation (GE Health, Piscataway, USA) and incubated in six-well culture plates at 37°C for 3 h in AIM-V medium (GIBCO, Carlsbad, USA). After incubation, nonadherent cells were removed. The adherent PBMC were cultured in AIM-V containing 1,000 U/ml granulocyte macrophage colony-stimulating factor (GM-CSF) (Immunex, Seattle, USA), and 500 U/ml interleukin-4 (IL-4) (StemCell Technologies, Vancouver, Canada) for 6 days. After 6 days culture, monocyte-derived dendritic cells were harvested from the nonadherent and loosely adherent cells and used for fusion. Autologous T cells were purified from nonadherent PBMC of RCC patients by nylon wool column and cultured at 37°C in RPMI 1640 with 20 U/ml IL-2, 10% autologous, heat-inactivated serum.

Fusion of dendritic cells and tumor cells

Allogeneic DC were mixed with primary RCC cells at a ratio of 10:1. After low-speed centrifugation, the mixed cell pellets were gently resuspended in 500 μl of prewarmed (37°C) 50% polyethylene glycol (PEG) solution (Sigma-Aldrich, St Louis, USA) for 3 min. Subsequently, PEG was progressively diluted by the slow addition of serum-free RPMI 1640 medium. The cells were washed free of PEG and cultured at 37°C in RPMI 1640 with 500 U/ml GM-CSF, 10% autologous, heat-inactivated serum for 5 days. The fusion cells were selected by gentle pipetting, and firmly attached tumor cells were discarded. Before vaccination, fusion cells were irradiated for 200 Gy.

Cell surface marker analysis

For FACS analysis, cells were washed with PBS and incubated for 40 min on ice with FITC-conjugated antibodies against MUC-1, HLA-ABC, HLA-DR, CD80, CD83, CD86, and HLA-A2 (BD Pharmingen, San Diego, USA). For analysis of dual expression, cells were incubated with FITC-conjugated anti-MUC1 mAb and PE-conjugated anti-HLA-DR or CD86 mAb (BD Pharmingen, San Diego, USA). All the samples were then washed, fixed with 2% paraformaldehyde and analyzed by flow cytometer (Becton Dickinson, Mountain View, USA).

Analysis of DC/RCC fusion cells by confocal microscopy

To confirm the successful fusion of DC and RCC cells, immunofluorescent staining was performed with FITC-conjugated anti-MUC1 and PE-conjugated anti-CD86 mAb and Hoechst 33342 was used for counter-staining. The cells were assessed by confocal microscopy (LSM 510, Zeiss with Krypton-argon and He-Ne laser, Thornwood, USA) to evaluate the coexpression of MUC-1 and CD86 molecules on DC/RCC fusion cells.

T-cell proliferation assay

Autologous T cells were cocultured with allogeneic DC/RCC fusion cells, DC mixed with tumor cells, DC or tumor cells alone at a ratio of 10:1 in the presence of 20 U/ml IL-2 for 7 days. T-cell proliferation was detected by Cell Titer 96 AQueous One Solution Cell Proliferation Assay Kit according to the protocol (Promega, Madison, USA) [21, 22]. Tetrazolium salt (MTS) solution was added to each well and incubated for 4 h. The absorbance of soluble formazan produced by cellular reduction of the MTS was measured at 490 nm using a Microplate Imaging System. All determinations were conducted in triplicate and expressed as the mean ± SD.

Cytotoxicity assays

Autologous T cells were stimulated with allogeneic DC/RCC fusion cells, DC mixed with tumor cells, DC or tumor cells alone at a ratio of 10:1 in the presence of 20 U/ml IL-2. After 7 days co-culture, T cells were used for cytotoxicity assays. The cytotoxicity assays were conducted using Cyto Tox 96 Non-Radioactive Cytotoxicity Assay Kit according to the protocol (Promega, Madison, USA) [21–24]. The targets included primary RCC cells used for fusion, allogeneic RCC cells and K562 cells. After washing, T cells and targets were resuspended in AIM-V medium at 30:1 effector-to-target ratios in 96-well, V-bottom plates and incubated for 6 h at 37°C. In the antibody-blocking assays, the target cells were incubated with anti-HLA-A2 mAb for 1 h at 37°C before the addition of effector cells. The supernatants were collected and assayed for LDH release in Microplate Imaging System at an absorbance of 490 nm. Spontaneous release of LDH was assessed by incubation of T cells or targets in the absence of another, and maximum release of LDH was determined by incubation of targets in 0.1% Triton X-100. Percentage of specific LDH release was calculated using the following equation: percent specific release = (experimental − effector spontaneous − target spontaneous)/(target maximum − target spontaneous) × 100.

Analysis for intracellular cytokines

To determine the stimulation of autologous T cells by allogeneic DC/RCC fusion cells, IFN-γ and IL-4 expression was assessed by intracellular staining. T cells were cocultured with DC/RCC fusion cells, DC mixed with tumor cells, DC or RCC alone for 7 days and then collected for analysis of IFN-γ and IL-4 expression. Briefly, T cells were incubated at 37°C for 6 h in RPMI 1640 medium containing 200 ng/ml anti-CD3 mAb and 1,000 ng/ml anti-CD28 mAb. Brefeldin A (Sigma, St Louis, USA), 10 ng/ml, was added for the final 5 h of incubation to block cytokine secretion [25]. The cells were washed and fixed with 4% paraformaldehyde and permeabilized with PBS/0.5% saponin (Sigma, St Louis, USA) for 10 min. After incubation, cells were labeled with APC-conjugated anti-IFN-γ or anti-IL-4 mAb and further stained with PE-conjugated anti-CD4 and FITC-conjugated anti-CD8 mAb (BD Pharmingen, San Diego, USA) for 30 min on ice and analyzed by flow cytometer.

Patient vaccination

All patients were treated using this vaccination protocol after 2–3 weeks of nephrectomy. Fusion cells (2.5–3.0 × 10⁶) were suspended in 1 ml normal saline and injected intradermally close to regional lymph nodes at three different sites each time. This treatment was repeated every week for at least eight times. Number of vaccination for each patient was shown in Table 2. Patients subsequently underwent CT scan every 2 months to evaluate the clinical response. Clinical response was classified as one of the following four categories: complete response (CR), defined as disappearance of the entire tumor for a minimum of 4 weeks; partial response (PR), defined as a reduction of 50% or more in tumor size for a minimum of 4 weeks; stable disease (SD), defined as either a decrease of less than 50% or an increase of less than 25% in tumor size; and progressive disease (PD), defined as an increase of 25% or more in tumor size or occurrence of new lesions. To assess the adverse effects and toxicity of vaccination, physical examination, full blood count, blood chemistry basics and urine analysis were performed every 2 weeks during the vaccination period. Blood was also taken for surface phenotype testing of peripheral blood lymphocytes with flow cytometer. The adverse effects were assessed according to World Health Organization criteria. For investigating anti-tumor immunity, autologous tumor cell lysate were injected intradermally before and after treatment to examine delayed-type hypersensitivity (DTH). Patients were also monitored for evidence of autoimmunity. An increase in the antinuclear antibody titer of >1 dilution was defined as significant.

Table 2.

Adverse effects and response status

Patient no.	No. of vaccinations	Adverse effects	DTH against tumor lysate		Clinical response
			Before therapy	After second vaccination
1	12	Erythema, induration, chills	Negative	Negative	SD
2	8	Fever, induration	Negative	Positive	PD
3	8	Pain, chills	Negative	Positive	SD
4	12	Fever, induration	Negative	Positive	SD
5	8	Fever, induration, pain	Negative	Positive	SD
6	8	Fever, chills	Negative	Negative	PD
7	8	Induration, erythema	Negative	Positive	SD
8	12	Fever, induration	Negative	Positive	SD
9	12	Fever, induration	Negative	Positive	PR
10	8	No	Negative	Negative	PD

Statistical analysis

The number of T lymphocyte subsets before and after vaccination was recorded as mean ± SD. Paired-samples t test was used to assess the differences. All statistical analyses were performed in the SPSS 15.0 software, and a P value <0.05 was considered statistically significant.

Results

Characterization of DC, RCC and fusion cells

After serum column treatment, the RCC cells were purified. As Fig. 1 shows, the fibroblasts and other non-RCC cells were almost completely excluded. To assess fusion cells, the resulting population was subjected to FACS analysis for the expression of tumor antigens and DC-derived molecules. DC expressed HLA-ABC, HLA-DR, CD80, CD83, CD86 and HLA-A2 molecules and was negative for MUC1 tumor associated antigen. By contrast, the RCC cells expressed MUC1, HLA-ABC, HLA-A2, but not HLA-DR, CD80, CD83 or CD86 molecules. Fusion of RCC cells to allogeneic DC resulted in hybrids with expression of both MUC1 tumor-associated antigens and DC-derived molecules (Fig. 2a). To determine the fusion efficiency, fusion cells were double stained with mAb against MUC1, CD86 or HLA-DR, respectively. As shown in Fig. 2b, about 35.1–41.0% of DC/RCC fusion cells expressed both MUC1 and CD86 or HLA-DR. The coexpression of MUC-1 and CD86 on fusion cells was confirmed by immunofluorescent staining (Fig. 2c). These results indicate that the fusion cells possess the phenotype of their parent cells.

Fig. 1.

Primary RCC cells culture. a Before purification. Many fibroblasts can be seen between the RCC cells mass. b After purification

Fig. 2.

Characterization of DC, RCC and DC/RCC fusion cells. a Allogeneic DC from healthy donor, RCC cells from patient one, and DC/RCC1 fusion cells were stained with indicated mAb and analyzed by flow cytometer for expression of the indicated molecules. b To determine the efficiency of DC/RCC fusion cell preparation, DC, RCC2 and DC/RCC2 fusion cells were analyzed by two-color flow cytometry for coexpression of MUC1 and CD86 or HLA-DR. Original histograms are shown with percentage of dual-positive cells. c Analysis of DC/RCC2 fusion cells by confocal microscopy. a FITC-conjugated MUC1 fluorescence is shown in green. b PE-conjugated CD86 is seen as red fluorescence. c Cell nucleus stained with Hoechst 33342 is showed in blue. d Areas of colocalization appear in yellow in the overlapping image. The yellow area in this image proved coexpression of MUC-1 and CD86. Scale bar 20 μm

Proliferation of autologous T cells stimulated by allogeneic DC/RCC fusion cells

The formation of large T cell clusters was observed in T cells cocultured with DC/RCC fusion cells, and minor ones observed in T cells cocultured with DC mixed with tumor cells or DC, but not with tumor cells (Fig. 3a). Significant T-cell proliferation was induced by DC/RCC fusion cells, and to a much lesser extent, by dendritic cells mixed with tumor cells and dendritic cells. In contrast, there was minimal T-cell proliferation when cocultured with RCC alone (Fig. 3b). These result show allogeneic DC/RCC fusion cells stimulate T-cell proliferation the most strongly.

Fig. 3.

Proliferation of T cells stimulated by allogeneic DC/RCC fusion cells. a T cells incubation with DC/RCC3 fusion cells, but not with DC mixed with tumor cells, DC or tumor cells alone, resulted in the formation of large T cell clusters. b T cells were cocultured with DC/RCC fusion cells, DC mixed with RCC, DC or RCC alone at a ratio of 10:1 for 7 days. T-cell proliferation was measured

Induction of tumor-reactive CTL by allogeneic DC/RCC fusion cells

Allogeneic DC/RCC fusion cells were first analyzed for their ability to stimulate autologous naïve T cells towards cytotoxic activity against RCC cells that were used as a fusion partner. As seen in Fig. 4a, the resulting stimulated CTL were effective in lysis of RCC tumor cells targets. By contrast, there was low level of lysis by T cells stimulated by unaltered DC mixed with tumor cells, DC or tumor cells alone. Similar results were obtained from five patients (Fig. 4a).

Fig. 4.

Stimulation of anti-tumor CTL cells by allogeneic DC/RCC fusion cells. a T cells stimulated by DC/RCC fusion cells, DC mixed with tumor cells, DC or tumor cells alone were incubated with autologous RCC at 30:1 ratio. LDH release assay was used to assess the CTL response. b Targets were stained with indicated mAb and analyzed by flow cytometry. c T cells were stimulated with DC/RCC3 (patient three) fusion cells and CTL activity against indicated targets were measured at 30:1 ratios. In antibody-blocking assay, targets were pre-incubated for 1 h at 37°C with anti-HLA-A2 mAb, and then cocultured with T cells stimulated with DC/RCC3 (patient three). Results are expressed as the mean ± SD of three replicates

Recognition of HLA-A2 restricted tumor associated antigen by CTL elicited by allogeneic DC/RCC fusion cells

To determine the antigen-specific and HLA-restricted tumor lysis by CTL, multiple tumor targets were used (Fig. 4b). CTL stimulated by DC/RCC3 fusion cells showed high CTL activity against RCC3 (HLA-A2+, MUC-1+). The killing activity was also observed against RCC1, RCC2, RCC4, RCC5 (all HLA-A2+, MUC-1+). In contrast, there was minimal CTL activity against K562 (HLA-A2−, MUC-1−) (Fig. 4c). Moreover, the lysis of RCC1, RCC2, RCC3, RCC4, RCC5 was significantly reduced by preincubation of the tumor cells with anti-HLA-A2 mAb (Fig. 4c). These results suggest allogeneic DC/RCC fusion cells can elicit vigorous antigen-specific, HLA class I-restricted CTL.

Cytokine production of CD4+ and CD8+ T cells

Cytokine production of T cells was measured by intracellular staining to determine T-cell activation and effector status. A high level of IFN-γ expression was detected in both CD4+ and CD8+ T cells stimulated by allogeneic DC/RCC fusion cells (Fig. 5). The effectiveness of T-cell responses to allogeneic DC/RCC fusion cells was confirmed by the lack of IFN-γ production from T cells stimulated by DC mixed with RCC, DC or RCC alone (Fig. 5). Minimal expression of IL-4 was detected in both CD4+ and CD8+ T cells in this study. Similar results were obtained from five patients (patients 1–5).

Fig. 5.

Cytokine production in CD4+ and CD8+ T cells stimulated by allogeneic DC/RCC fusion cells. T cells were cocultured with DC/RCC1 fusion cells (patient one), DC mixed with RCC1, DC or RCC1 alone for 7 days. T cells were collected, purified by nylon wool, and accessed for intracellular expression of IFN-γ and IL-4 in CD4+ and CD8+ T cells using flow cytometry. Original histograms are shown with percentage of dual-positive cells

Clinical results of allogeneic DC/RCC fusion cell vaccination

All ten patients had bidimensionally measurable metastatic lesions at the time of entering our protocol (Table 1). Clinical outcome was based mainly on comparison of CT scans before and after treatment. Patient no. 9 experienced a PR. The female patient was found to have a large lung metastatic node by chest CT scan (Fig. 6a, left arrows). After 12 times of allogeneic DC/RCC treatment, the lung metastases were significantly regressed (Fig. 6b, right arrows, image taken 4 months after the first vaccination). The remaining node (right) was confirmed as metastatic renal clear cell carcinoma by pathological examination after thoracoscopic resection. Six patients (no. 1, 3, 4, 5, 7, 8) showed SD with stabilization of previously progressive disease (follow up 1.5 year). However, three patients remained with PD. Patient no. 2 and 6 were found to have increased in tumor mass. Patient no. 10 had bone metastases.

Fig. 6.

Clinical response to allogeneic DC/RCC vaccination. CT scans of patient no. 9, showing lung metastases. a Arrow, sites of lesions before vaccination, b arrow, sites of lesions 4 months after allogeneic DC/RCC vaccination

No serious adverse effects were observed. In seven of the ten patients, indurations were observed at the injection site after the second immunization. Six patients had transient fever for not more than 1 day. Two patients were found to have erythema near the injection site. Three patients felt pain after vaccination and released after a short time. And three patients experienced a short period of chills after injection. The side effects in the majority of the cases did not exceed grade I and did not require medical intervention (Table 2). No substantial changes in the results of physical examination, full blood count, blood chemistry basics and urine analysis were observed. Also, no signs of autoimmune disease were detected. For DTH test, none of the patients showed a positive reactivity before the vaccination therapy. But seven patients showed reactivity after vaccination. One of them was the patient with the partial response (Table 2).

In addition, the phenotype of peripheral blood lymphocytes was investigated using flow cytometer before and after immunotherapy. A significant increase in percentage of CD3, CD4, CD8 and CD56 positive cells were observed after treatment (P < 0.05, respectively, Table 3).

Table 3.

T lymphocytes subsets proportions before and after vaccination

Time	CD3+ (%)	CD4+ (%)	CD8+ (%)	CD56+ (%)
Pre-treatment	72.62 ± 3.61	41.85 ± 3.81	22.14 ± 3.97	12.53 ± 2.99
Post-treatment	76.45 ± 5.53	45.86 ± 5.19	24.98 ± 6.57	14.79 ± 3.93
P value	0.015	0.008	0.044	0.032

Discussion

Immune system has been demonstrated to possess the ability to recognize and attack tumors [26]. To date, several approaches have been used to load DC with target antigen to elicit tumor specific immune response. However, human tumor immunotherapy based on DC has met with only limited success. The reasons for this may be related to: (1) the limited availability of known tumor-associated antigens, (2) insufficient stimulation to T cells due to protein degradation and major histocompatibility complex molecule cycling, (3) the inefficient antigen loading, (4) the resident tolerizing environment of the tumor, and (5) generation of insufficient numbers of functional DC from patients with advanced stage of disease due to a variety of reasons including immunosuppression by long-term chemotherapy [27].

Dendritic cell/tumor fusion cells have been proved to be an effective vaccine to induce CTL responses against tumor. In animal models, vaccination with fusions of DC and tumor cells resulted in protection from tumor challenge, regression of established metastatic disease and reversing immunologic unresponsiveness to tumor-associated antigens [28–32]. In preclinical studies, fusions generated with tumor cells and DC potently induced CTL that are able to lyse tumor targets [20, 33–36]. However, sometimes it is difficult to obtain sufficient numbers of functional DC from peripheral blood of patient. Allogeneic dendritic cell/autologous tumor fusion cell vaccine may be an ideal alternative approach because large amount of functional DC can be generated in vitro from healthy donors. And it has been suggested that allogeneic dendritic cells might cause nonspecific systemic immune stimulation and create a potent immunostimulatory environment at the injection site and therefore favor the generation of CD8 cytotoxic T-cell responses to tumor peptides presented by fusion cells.

In some studies, tumor cell lines were used as fusion partner. The advantage of this strategy is allogeneic tumor cell lines can grow well in vitro. Thus, infinite tumor cells can be obtained. But tumor cell lines are different from primary cultured tumor cells. There are some known tumor associated antigens shared between tumor cell lines and primary tumor cells, but there must be some unknown tumor associated antigens not shared between them. In addition, the antigen spectrum of tumor cell lines may have changed after long-term culture. Therefore patient-derived tumor cells should be more specific source of tumor cells for fusion.

Avigan et al. [37] generated hybrids of autologous RCC cells and allogeneic dendritic cells by electrofusion techniques. After vaccination, two patients demonstrated a partial clinical response and eight patients had stabilization of their disease. Marten et al. [38] also did a clinical trial of allogeneic DC fused with tumor cells in patients with metastatic RCC. They observed no adverse effects and no difference in clinical outcome between the allogeneic and the autologous treatment. DTH reaction using tumor cells was positive after treatment in 7 of 12 patients, 2 of whom were found to have stable disease. An increase in the reactivity against recall antigens was seen in most patients.

However, tumor cells used in the above studies were not purified. In our study, serum column was used to purify RCC cells from tumor tissue. This method can improve the success rate of primary culture of RCC. Also, the high percentage of purified RCC cells could exclude the impact of healthy tissue cells.

For reasons above, in our present studies, purified primary RCC cells derived from patients were fused with allogeneic DC to elicit antigen-specific CTL against autologous RCC cells. We demonstrate that fusion of allogeneic DC with tumor cells resulted in hybrids possessing the phenotype of their parent cells and the function of DC. The hybrids could stimulate both CD4+ and CD8+ T cells, both of which are important in the antitumor immunity. Human antitumor immune responses are mainly dependent on cellular immunity. CD4+ T cell play a central role in orchestrating the immune response to cancer. They provide help in recruiting CD8+ T cells and activating tumoricidal macrophages through cytokine production. CD8+ CTL are believed to be critical effectors of antitumor immune responses. Also, CD4+ T cells are important sources of proinflammatory cytokines such as IL-2 and IFN-γ which play an important role in the induction and maintenance of CTL. High levels of IFN-γ expression with minimal IL-4 secretion by CD4+ and CD8+ T cells stimulated by the fusion cells suggests a Th1 immune response, which is favorable for antitumor immunity. CD56+ cells were named “natural killer” because of the initial notion that they do not require activation in order to kill tumor cells and they are non-MHC-restricted. They target tumor cells and protect against a wide variety of infectious microbes. In our study, we found that NK cells could increase obviously after fusion cells vaccinations, which may enhance the antitumor immunity.

Most importantly, these fusion cells could prime and propagate tumor antigen-specific CTL in an efficient manner. Allogeneic DC/RCC fusion cells can induce CTL activity against autologous RCC cells and other HLA-A2+ RCC targets with shared MUC-1 antigen. But there is minimal CTL activity against HLA-A2−, MUC-I− K562 cells. In addition, the CTL activity is inhibited by preincubation target cells with anti-HLA-A2 mAb. These results indicate that the cytotoxicity is antigen-specific and HLA class I restricted. The result also showed that the cytotoxicity against autologous RCC cells is much higher than against other allogeneic HLA-A2+ RCC targets, even they also express MUC-1 antigen. It is possible that there are some other not shared tumor antigens presented by fusion cells. Taken together, these data indicate that allogeneic DC/RCC fusion cell is a promising vaccine to stimulate specific-CTL response against autologous RCC cells.

In the clinical study, 10 patients with metastatic RCC were vaccinated with allogeneic DC/RCC fusion vaccine. The clinical responses were one PR and six SD. In patient no. 9, vaccination resulted in the near complete regression of lung metastasis mass and the patient remained without evidence of disease progression for 4 months. In this trial, a complete absence of DTH responses to tumor cell lysate was observed before vaccination, indicating immune system severely compromised to tumor. Nevertheless, we observed DTH responses in seven out of ten patients after vaccination, including the PR patient (no. 9). Vaccination was well tolerated, and there was no evidence of clinically significant autoimmunity. The adverse effects were generally minor indicating low toxicity and some signs of restricted therapy induced auto-immune reactions. All of the side effects including erythema, indurations and pain at the site of injection, low grade fever, and chills are indications of, in some cases vigorous, cellular immune reactions and, thus, part of the intended response. No other signs of vaccination associated auto-immune reaction were seen. The side effects in the majority of the cases did not exceed grade I and did not require medical intervention.

We conclude from our data that vaccination with allogeneic DC/RCC fusion vaccine is safe and can induce immunological and clinical responses in patients with late stage of renal cell carcinoma. Despite the small number of patients treated, the immunological and clinical responses observed are encouraging and warrant further studies of tumor vaccination on more patients, particularly in patients with limited disease, since patients with minimal disease would probably benefit more from vaccination therapy.

This study also provides a foundation for future studies which incorporate additional improvements in fusion cell generation and delivery techniques. There are many possibilities for improving this basic protocol. First, the DC:RCC ratio can be optimized. Second, the fusion cells can be further pretreated with additional important cytokines (e.g., IL-12) before injection to enhance DC phenotype and function. Third, the fusion cells might be direcly injected into tumor masses in addition to lymph node delivery. Fourth, the fusion cells might be injected along with parallel cultured T cells. Fifth, the Fusion cells might be coinjected with immunogenic adjuvants (e.g., LPS) to further enhance immune responsiveness.