A.E. Pedersen and I.M. Svane contributed equally to this work.
Phenotypic and Functional Characterization of Clinical Grade Dendritic Cells Generated from Patients with Advanced Breast Cancer for Therapeutic Vaccination
Article first published online: 31 JAN 2005
Scandinavian Journal of Immunology
Volume 61, Issue 2, pages 147–156, February 2005
How to Cite
Pedersen, A. E., Thorn, M., Gad, M., Walter, M. R., Johnsen, H. E., Gaarsdal, E., Nikolajsen, K., Buus, S., Claesson, M. H. and Svane, I. M. (2005), Phenotypic and Functional Characterization of Clinical Grade Dendritic Cells Generated from Patients with Advanced Breast Cancer for Therapeutic Vaccination. Scandinavian Journal of Immunology, 61: 147–156. doi: 10.1111/j.0300-9475.2005.01531.x
- Issue published online: 31 JAN 2005
- Article first published online: 31 JAN 2005
- Received 18 August 2004; Accepted in revised form 20 October 2004
Dendritic cells (DC) are promising candidates for cancer immunotherapy. However, it is not known whether in vitro-generated monocyte-derived DC from cancer patients are altered compared with DC from healthy donors. In a clinical phase I/II study, monocyte-derived DC were generated in vitro utilizing granulocyte macrophage colony-stimulating factor and rh-interleukin-4 (IL-4) and used for cancer immunotherapy. In this study, we tested the effect of various maturation cocktails and performed a comparative evaluation of the DC phenotype and functional characteristics. Polyriboinosinic polyribocytidylic acid (Poly I:C) + tumour necrosis factor-alpha (TNF-α) induced significant IL-12 p70 secretion, which was increased after addition of a decoy IL-10 receptor. The lymph node homing chemokine receptor CCR-7 expression was induced by TNF-α + IL-1β + IL-6 + prostaglandin E2 but was not induced by Poly I:C + TNF-α. In general, DC from patients had an intermediate maturity phenotype with a significantly higher expression of CD40 and CD54 compared with healthy donors. In vitro analyses showed an unimpaired capacity of the patient-derived DC for antigen-specific (cytomegalovirus, tetanus and keyhole limpet haemocyanin) T-cell stimulation, whereas the allostimulatory capacity of patient-derived DC was significantly decreased. These data suggest that patient-derived DC are more differentiated but are less sensitive to maturation-inducing agents than DC obtained from healthy individuals.
As a result of extensive research in methods to generate large amounts of dendritic cells (DC) from peripheral blood monocytes for clinical applications , it has become feasible to make use of these cells as adjuvant in cancer immunotherapy. DC are known to be superior antigen-presenting cells (APC) , which express a large number of costimulatory molecules and secrete proinflammatory cytokines, primarily interleukin-12 (IL-12). As a constituent of a cancer vaccine, these features enable the DC to break down tumour tolerance and induce tumour-specific immune responses leading to tumour rejection [3–7]. Up to now, a large number of phase I/II vaccination trials have been published employing a diverse repertoire of DC preparations and tumour antigens, but few studies have tested the qualitative differences between DC generated from cancer patients compared with DC generated from healthy donors [8–10].
Recently, it has been established that mature DC are superior to immature DC in clinical cancer vaccine trials  and that they are more potent inducers of antigen-specific CD8+ T cells [3, 12]. Priming of CD8+ T cells is facilitated by a T-helper type 1 (Th1) response  and maturation of DC might indeed favour this response mainly by secretion of IL-12 . The biological active form of IL-12 is a heterodimer, p70, which is composed of the subunits p35 and p40. However, when analysing IL-12 production, the necessity of measuring the biological active form, p70, is often underestimated . Only a narrow panel of maturation-inducing agents is able to induce IL-12 p70 production . Another important parameter for the clinical outcome of a vaccination trial is the capability of the injected DC to migrate to the draining lymph nodes. Surface expression of the chemokine receptor CCR-7  is known to be necessary for this migration from peripheral tissue into afferent lymphatic vessels and for homing to the T-cell areas of the draining lymph nodes [7, 18–20].
In an ongoing clinical phase II trial, DC are generated from patients with progressive breast cancer utilizing granulocyte macrophage colony-stimulating factor (GM-CSF) and rh-IL-4. Subsequently, the DC are pulsed with human leucocyte antigen (HLA)-A2-binding peptides derived from the tumour-associated protein p53 in combination with a PAN DR reactive epitope (PADRE) . These DC are then used for the vaccination therapy of the patients. We have previously shown that the treatment leads to an increase in p53-specific CD8+ T cells in some of the immunized patients . In the present study, we perform a DC quality assessment including a comparative evaluation of DC yield, immunophenotype and functional characteristics of the generated therapeutic DC preparations. We also test the effect of clinically applicable maturation agents on IL-12 secretion and phenotype including CCR-7 and show that the two different clinically applicable DC maturation cocktails tumour necrosis factor-alpha (TNF-α) + IL-1β + IL-6 + prostaglandin E2 and polyriboinosinic polyribocytidylic acid (Poly I:C) + TNF-α have different effects on these two parameters. In addition, we demonstrate that the presence of a decoy IL-10 receptor can effectively increase the effect of the maturation agent by increasing IL-12 p70 secretion.
Materials and methods
Generation of DC. Peripheral blood was obtained after informed consent from healthy donors (blood sample) or patients (leukapheresis product) with progressive metastatic cancer enrolled in a phase I/II trial with cancer vaccination using wildtype p53 peptide-pulsed DC . Blood from patients was taken before vaccination, and the patients did not receive any systemic treatment such as radiation therapy, chemotherapy or therapy with immunosuppressive drugs within the prior 4 weeks. For further information of included patients see Svane et al. . Peripheral blood mononuclear cells (PBMC) were treated with Orthomune lysing solution (provided by hospital pharmacy) to lyse remaining red blood cells, washed and resuspended in culture medium (X-VIVO15, 2%l-glutamin 200 mm and 1% autologous heat-inactivated plasma) at 7 × 106 cells/ml and separated by 1 h adherence to plastic Nunclon dishes (Nunc, Biotech Line, Slangerup, Denmark). Nonadherent cells were removed, and adherent cells were subsequently cultured for 7 days in culture medium supplemented with 250 IU/ml rh-IL-4 (CellGenix, Freiburg, Germany) and 1000 IU/ml GM-CSF (Leucomax, Schering Plough, Farum, Denmark). On day 7, DC were harvested, frozen in 85% autologous serum, 10% DMSO (Merck, Darmstadt, Germany) and 5% Glucosteril 40% w/v (Fresenius, Albertslund, Denmark) and stored in liquid nitrogen using automated cryopreservation (Planer Freezing Unit; Planer, London, UK). These DC preparations were used for either patient therapy or in vitro studies. Mixed lymphocyte culture reaction and autologous T-cell proliferation were performed with thawed cells (cell viability >90% was invariably found), whereas all other experiments were performed on freshly obtained DC.
Antibodies and flowcytometry. Monoclonal antibodies were used against the following antigens: CD11c (APC), CD33 (APC) and HLA-DR (PerCP) from BD Bioscience (San José, CA, USA); CD40 [fluorescein isothiocyanate (FITC)], CD80 (FITC), CD86 (APC) and CD1a (APC) form Pharmingen (San Diego, CA, USA); CCR-7 (FITC) from R&D Systems (Minneapolis, MN, USA); CD83 (APC) from Caltag Laboratories (Burlingame, CA, USA) and CD54 (FITC) from Immunotech (Marseilles, France). Phycoerythrin-lineage cocktail was prepared for the simultaneous labelling of T cells, B cells and monocytes by antibodies against CD3, CD14 and CD19 (BD Bioscience). Aliquots of the cultured cells were stained simultaneously for 30 min at 4 °C with lineage cocktail and antibodies against three of the relevant DC-surface markers and then washed twice in phosphate-buffered saline (PBS). All antibodies were used in concentrations recommended by the manufacturer. Prior to staining, each sample was blocked with 20 µg/ml polyclonal human immunoglobulin G. Four-colour analysis of DC was performed on a FACSCalibur flow cytometer (BD Bioscience), and data were analysed using CellQuest software (BD Bioscience).
Proliferation assay. Stimulation of autologous proliferation of T cells was measured by PKH26 labelling . Labelling was carried out as described previously . In brief, PBMC (1 × 107) were pelleted in polypropylene tubes, resuspended in 1 ml diluent C and incubated for 2 min with 1 ml of 3 µm PKH26 dye (PKH26-GL; Sigma-Aldrich Inc., St Louis, MO, USA). Staining reaction was stopped by the addition of 2 ml of heat-inactivated 100% AB serum. Labelled cells were washed and counted.
Thawed DC were plated at 1 × 105 cells/well in a 24-well microculture plate (Nunc) in X-VIVO15 with 10% heat-inactivated AB serum and loaded with the individual antigens. The antigens used were: 25 µg/ml tetanus toxoid (TT; Statens Seruminstitut, Copenhagen, Denmark); 60 µg/ml cyto- megalovirus (CMV) lysate (Virion, Zürich, Switzerland); 100 µg/ml keyhole limpet haemocyanin (KLH; Calbiochem, San Diego, CA, USA) and 40 µg/ml PADRE peptide (Schafer-N, Copenhagen, Denmark) . For negative control, no Ag was added to the cells. After 2 h of incubation, 106 PKH26-labelled T cells were added to each well. The cells were cultured in a final volume of 1 ml, and each condition was set out in duplicates. Plates were incubated for 7 days at 37 °C, 5% CO2. Supernatant was collected for measurement of cytokines. The cells were harvested in staining tubes, washed once in washing buffer and stained for 30 min with CD3 APC, CD8 PerCP and CD4 FITC. Samples were washed and fixed in 1% paraformaldehyde in PBS. Proliferation of T-cell subsets was then measured with flowcytometry as the percentage of cells with decreased PKH26 concentration due to dilution during cell division (the fluorescence intensity of membrane staining halves with each cell division). The response was defined as positive when proliferating CD8+ or CD4+ T cells constituted >2% of the T cells in the subset after subtraction of negative control values.
Stimulation of allogeneic T-cell proliferation was measured by [3H]-thymidine incorporation. In brief, DC were irradiated (3000 rad) before addition to 96-well culture dishes (Nunc) in titrated numbers from 200 to 12,500 DC/well. Then, 105 PBMC were added at a final volume of 200 µl/well, and cells were cultured for 5 days with addition of [3H]-thymidine for the last 18 h.
Maturation of DC. Day 7 immature DC were harvested and plated in 96-well flat-bottomed culture dishes (Nunc) at 2 × 105 cells/well. Maturation was then performed by the addition of combinations of maturation reagents in the following concentrations: 2 µg/ml CD40L (Biocarta Europe, Hamburg, Germany), 50 ng/ml TNF-α, 10 ng/ml IL-1β, 15 ng/ml IL-6 (all from R&D Systems, Minneapolis, MN, USA), 2 µg/ml prostaglandin E2 (Prostine E2; Pharmacia & Upjohn, Paris, France), 12.5 µg/ml Poly I:C (Sigma-Aldrich) or 2.5 µg/ml soluble IL-10R. Generation of the soluble IL-10R (sIL-10R1Q6) was published elsewhere .
Measurement of cytokines. Supernatants were harvested after 48 h of DC maturation and production of IL-10 and IL-12 (p40 or p70) was analysed. IL-10 and IL-12 (p40) were analysed by sandwich enzyme-linked immunosorbent assay (ELISA) using the manufacturer's protocol (Pharmingen) with the following antibodies. For IL-10, rat anti-human IL-10 (#18551D) and biotinylated rat anti-human IL-10 (#18562D) were used and recombinant human IL-10 (#19701V) was used as standard. For IL-12 (p40), mouse anti-human IL-12 (p40) (#20711D) and biotinylated mouse anti-human IL-12 (p40/70) (#20512D) were used and recombinant human IL-12 (p40) (#19931V) was used as a standard. Measurement of IL-12 (p70) was performed with a commercial ELISA kit (Biocarta Europe).
Statistical analysis. Comparison of samples to establish the statistical significance of difference was determined by the two-tailed Student's t-test for independent samples or the Fisher's exact test. Results were considered to be statistical significant when P ≤ 0.05.
Phenotype of DC prepared from breast cancer patients and healthy controls
After 7 days of culture, we harvest PBMC-derived cells for clinical use, and DC are defined as large granular cells by microscopy. The mean DC yield from the initial PBMC is 36.4 and 6.3% for patients (n = 16) and healthy controls (n = 9), respectively, a difference which may in part reflect a minor difference in the initial monocyte fraction of the PBMC (12.6 and 9.9%, respectively). The phenotype of the cultured cells was further analysed by flowcytometry (Table 1). Based on forward-side-scatter dot plots, on average 44% of the cultured cells from breast cancer patients compared with 26% from controls were gated as DC (large granular cells) with typical DC morphology. However, within this gate on average only mean 29% of the cells from patients were lineage negative (–CD19, –CD3 and –CD14) contra only 13% from controls. We next analysed the surface-antigen expression on these lineage-negative DC (Table 1). In both patients and controls, we found that the majority of the DC had a myeloid phenotype with >90% expressing CD11c and about 50% expressing CD33. The expression of CD1a was more variable and less pronounced. Several adhesion and costimulatory molecules are essential for the DC to induce an effective cytotoxic T-lymphocyte (CTL) response. DC prepared from breast cancer patients were found to have an intermediate mature phenotype with a high fraction of the DC expressing CD40, CD54 and CD86, and a small fraction expressing CD80 and CD83 in both patients and controls. Furthermore, HLA-DR was also expressed by the majority of cells. Notably, the fractions of positive cells as well as the mean expression level of these important DC molecules were, in general, higher in DC prepared from breast cancer patients than from healthy donors. A difference found to be significant for CD40 and CD54. Freezing and thawing of the DC did not result in any significant changes in the phenotype except for a reduction in CD40-expressing cells from mean 91 to 80%.
|Donors||% (SD)‡||% (SD)‡||% (SD)||% (SD)||% (SD)|
|Mean patients (n = 16)||44 (17)||29 (19)||66 (26)||24 (16)||97 (2)|
|Mean controls (n = 9)||26 (9)||13 (6)||45 (25)||27 (24)||91 (8)‡|
|% (SD)||MFI (SD)||% (SD)‡||MFI (SD)‡||% (SD)‡||MFI (SD)‡|
|Mean patients (n = 16)||69 (22)||205 (79)||84 (20)||262 (73)||80 (18)||239 (67)|
|Mean controls (n = 9)||64 (21)||162 (84)||63 (20)||154 (53)||61 (24)||136 (62)|
|% (SD)||MFI (SD)||% (SD)||MFI (SD)||% (SD)||MFI (SD)|
|Mean patients (n = 16)||66 (25)||267 (138)||20 (15)||64 (26)||10 (9)||17 (28)|
|Mean controls (n = 9)||50 (26)||219 (156)||16 (5)||35 (17)||14 (5)||5 (16)|
As a result of the relative low fraction of lineage-negative cells, supplementary experiments (on three normal donors and nine cancer patients) were performed to evaluate the content of CD14+ cells in the DC gate. Only about 10% of the cells in the DC gate (large granular cells) stained positive for the B- and T-cell lineage markers CD19 and CD3, respectively. Thus, the relatively high fraction of the lineage-positive cells in the DC gate almost exclusively consisted of CD14+ monocytes. Additional evaluation by flowcytometry revealed that the CD14– DC and CD14+ monocyte-derived cells from the DC gate had comparable expression of DC-related surface antigens such as DR, CD86, CD40 and CD54 (data not shown).
Phenotype of DC after maturation
To activate T cells in vivo, DC increase their expression of HLA-DR, CD80 and CD83 molecules in response to a variety of maturation stimuli. Two maturation cocktails consisting of either Poly I:C + TNF-α or TNF-α + IL-1β + IL-6 + prostaglandin E2 were evaluated for the ability to induce phenotypic changes associated with maturation in DC from breast cancer patients (Table 2). Maturation of the DC by either cocktails did not affect the number of cultured cells gated as DC, but both cocktails induced a marked increase in the fraction of these cells being lineage negative. From Table 2, it appears that both cocktails were able to increase the fraction of cells expressing HLA-DR, CD80 and CD83 as well as the median expression (median fluorescence intensity) level of these molecules. However, considerable differences between the two maturation cocktails were observed: significant expression of CCR-7 was only induced by the TNF-α + IL-1β + IL-6 + prostaglandin E2 cocktail (Table 2) and was not influenced further by adding Poly I:C to this cocktail (data not shown). The highest median fluorescence intensity of the analysed surface molecules was, in general, also attained with the TNF-α + IL-1β + IL-6 + prostaglandin E2 cocktail. Comparable results were obtained with DC from healthy controls (data not shown).
|Maturation cocktail||DC* (%)||Lin gate† (%)||%||MFI||%||MFI||%||MFI||%||MFI||%||MFI||%||MFI||%||MFI|
|IL-1β, TNF-α, IL-6, Pge2||30||58||92||306||73||205||84||201||87||390||92||416||95||441||92||289|
|Poly I:C, TNF-α||28||53||91||250||18||31||88||172||62||181||95||415||98||413||93||277|
Influence of different maturation reagents on IL-12 secretion by healthy donors
IL-12 secreted by mature DC is known to stimulate the induction of Th1 lymphocytes, a process that is pivotal in the induction of antitumour immunity. Initially, DC prepared from healthy donors were investigated for the ability to produce IL-12. As shown in Fig. 1(A), there was only low spontaneous secretion of IL-12 p40 of the DC before maturation. In our clinical trial , DC are not matured but express an intermediate maturity phenotype. These DC are pulsed with major histocompatibility complex (MHC) class I-binding peptides in combination with the pan-MHC class II-binding helper peptide PADRE  before injection into cancer patients. In murine models, PADRE has been shown to promote maturation of DC and generation of CTL through activation of Th cells . We tested in vitro whether coculture of PADRE-pulsed DC and autologous T cells could induce secretion of IL-12 p40 from DC, but this was not the case (Fig. 1A). For comparison, soluble CD40L was added as a positive control (Fig. 1A). However, the addition of PADRE seemed to increase the amount of IL-12 p40 secreted when DC were pulsed with CMV-derived antigens or TT and cocultured with autologous T cells (data not shown).
To further test the effect of DC maturation on IL-12 secretion, several DC maturation reagents were applied to DC prepared from healthy donors to determine any difference in their ability to induce secretion of active p70. As shown in Fig. 1(B), maturation induced by CD40L or TNF-α alone did not stimulate IL-12 p70 secretion, whereas the presence of Poly I:C led to a substantial IL-12 p70 secretion. The two established maturation cocktails were also investigated; Poly I:C + TNF-α was the only maturation cocktail that lead to a significant increase in IL-12 p70 (P < 0.05) compared with immature DC, whereas no p70 secretion was observed after maturation with TNF-α + IL-1β + IL-6 + prostaglandin E2. However, both maturation cocktails lead to the secretion of IL-12 p40 accompanied by a comparable level of IL-10 secretion. When adding Poly I:C to the TNF-α + IL-1β + IL-6 + prostaglandin E2 maturation cocktail, IL-12 p70 secretion by DC was restored in some but not all donors (data not shown).
With the purpose to further optimize DC maturation, the effect of a soluble IL-10 decoy receptor (IL-10R) was investigated. As shown in Fig. 2(A), the addition of this receptor could significantly increase the level of IL-12 p70 (P < 0.05), but not p40, secreted when DC of healthy controls were maturated with Poly I:C + TNF-α (Fig. 2B). A similar effect was observed in a few donors, but not in general, when IL-10R was added to DC during maturation with Poly I:C + TNF-α + IL-1β + IL-6 + prostaglandin E2 (data not shown). The DC from healthy controls were also found to have a significantly increased capacity for allogeneic stimulation after addition of IL-10R (P < 0.05) (Fig. 2C).
IL-12 p70 secretion by DC from patients with progressive breast cancer
To test the ability of patient-derived DC to secrete the active form of IL-12, the maturation of DC from patients with breast cancer and from healthy controls was performed with Poly I:C + TNF-α. IL-12 p70 secretion was measured in DC from only three of eight patients (Table 3), whereas DC from eight of nine healthy donors secreted p70. This difference was significant (P < 0.05, Fischer's exact test). In contrast, the fraction of patients and healthy donors with DC capable of IL-12 p40 secretion did not differ. Because DC from patients and healthy controls were obtained from different PBMC sources, we tested separately whether DC derived from peripheral blood samples or leukapheresis products had a different IL-12 p70 secretion level, but this was not the case (data not shown).
|IL-12 p70 (pg/ml)||IL-12 p40 (ng/ml)|
Capacity of DC to induce autologous T-cell proliferation
DC prepared from nine breast cancer patients and nine healthy donors were assessed for the capacity to stimulate the proliferation of antigen-specific autologous T cells (Table 4). To evaluate the memory T-cell response, TT and CMV lysate antigens were used. KLH was used to assess the primary T-cell response. T-cell proliferation was measured by PKH26 labelling of the cells prior to Ag stimulation. To compute the frequency of viable T cells which had proliferated during the 7 days of activation, T-cell subsets were positively identified among other cell populations by forward-side-scatter signals, CD3, CD4/8 expression and the reduced PKH26 staining.
|Donors||CD3+ T-cell subset||CMV-induced proliferation*||Tetanus-induced proliferation*||KLH-induced proliferation*|
|Patients (n = 9)||CD8+||4/9||5/6||4/9|
|Controls (n = 9)||CD8+||2/9||5/9||1/9|
Recall response to TT with proliferation of CD4+ and CD8+ T cells was detectable in five of six patients and in five of nine controls. Addition of PADRE resulted in an increased tetanus-specific T-cell proliferation in five patients and three controls. Proliferative T-cell responses to CMV were detected in five of nine patients and three of nine controls. In one patient and one control, only CD4+ T-cell proliferation was induced. Addition of PADRE did not influence the magnitude of these responses. In six of nine patients, a low but significant primary CD4+ or CD4+ and CD8+ T-cell proliferative response to KLH was induced. Surprisingly, this was only the case in two of nine controls. However, this difference was not significant (P = 0.07, Fisher's exact test). In many donors, costimulation with PADRE was necessary either to induce KLH response or to increase the existing response. When testing for IL-12 production in the supernatants of these experiments, IL-12 p40 was primarily secreted in response to CMV whereas no significant IL-12 p70 was produced (Table 5). Again, the addition of PADRE increased the amount of IL-12 p40 secreted among the responding donors (data not shown).
|Donors||DC + TC *||DC + TC + CMV *||DC + TC + TT *||DC + TC + KLH *|
|Patients (n = 6)|
|Controls (n = 9)|
DC as allostimulatory cells
To test the ability of DC to induce the proliferation of T lymphocytes in primary allogeneic MLR, the stimulatory capacity of DC from nine patients and nine control donors was analysed. Allogeneic PBMC were exposed to titrated numbers of irradiated DC harvested at day 7 of the culture. As shown in Fig. 3, DC from breast cancer patients had a significantly (P < 0.05, Student's t-test) decreased allostimulatory capacity compared with DC of healthy controls.
Recent technical advances in the culture of large numbers of clinical grade DC from leukapheresis products have opened up the opportunity to test these cells in large cancer vaccination trials. Several protocols to obtain clinically applicable DC have been described  and include maturation procedures that influence the phenotype and functional characteristics of these cells [16, 28]. One aim of the present study was to test the ability of clinically applicable maturation reagents to induce optimal DC maturation with regard to IL-12 p70 secretion and CCR-7 expression. These parameters were chosen because they are necessary for Th1 differentiation  and homing of DC to draining lymph nodes [7, 18–20], respectively. Unfortunately, none of these parameters were induced in the PADRE peptide-pulsed DC used for the clinical cancer vaccine trial  when the DC were not maturated, even when cocultured with autologous T cells. The maturation cocktails TNF-α + IL-1β + IL-6 + prostaglandin E2 and Poly I:C + TNF-α were compared, and it was observed that IL-12 p70 secretion was only induced after maturation with Poly I:C + TNF-α whereas CCR-7 upregulation was only observed after maturation with TNF-α + IL-1β + IL-6 + prostaglandin E2. This selective pattern of IL-12 p70 secretion has been shown by others [14, 16] and confirms that Poly I:C is a potent inducer of IL-12 p70. However, the different ability of the two maturation cocktails to induce CCR-7 upregulation has not been described previously and shows the caveats concerning the choice of appropriate maturation agents for DC in clinical cancer vaccine trials. The combination of the two maturation cocktails was shown to restore both functions.
In the previous studies of Poly I:C + TNF-α- or TNF-α + IL-1β + IL-6 + prostaglandin E2-mediated DC maturation [16, 23], IL-10 secretion of the DC was not investigated. In our study, we found IL-10 secretion from DC after maturation with both cocktails. According to the traditional classification of IL-10 as an anti-inflammatory cytokine , IL-10 might inhibit the ability of DC to activate peptide-specific CTL especially when not accompanied by IL-12 p70 secretion. Thus, TNF-α + IL-1β + IL-6 + prostaglandin E2 might induce an unfavourable cytokine profile when used as a DC maturation cocktail. However, when response rates in clinical trials were evaluated, TNF-α + IL-1β + IL-6 + prostaglandin E2-maturated DC did not seem to be less effective compared with DC maturated otherwise . In addition, it has become evident that IL-10 might have proinflammatory properties when secreted by DC early in the priming of an immune response , because of its activation of natural killer cells [31, 32]. By using activated natural killer cells as a bridge to adaptive immunity, TNF-α + IL-1β + IL-6 + prostaglandin E2-maturated DC might lead to the activation of CD4+ and CD8+ T cells with a Th1 cytokine profile . In addition, our present data suggest that high IL-12 p70 secretion is likely to be accompanied by IL-10 secretion.
In the present study, the addition of a soluble decoy IL-10 receptor during DC maturation was able to increase the secretion of IL-12 p70, especially when combined with maturation with Poly I:C + TNF-α. Likewise, the capacity of DC to induce allogeneic T-cell stimulation was also increased. Others have shown a similar effect of inhibiting endogenously produced IL-10 during lipopolysaccharide maturation  using a monoclonal antibody against IL-10. Thus, IL-10 blocking seems to be a feasible strategy in several kinds of maturation regimens. Taking advantage of a soluble decoy receptor might overcome potential induction of neutralizing antibodies, which would be expected if anti-IL-10 antibodies were used in this context during the injection of DC into patients.
The demonstrated lack of CCR-7 upregulation during maturation with Poly I:C + TNF-α could reflect the fact that prostaglandin E2 is a major inducer of this receptor . Although important for DC homing to regional lymph nodes, CCR-7 upregulation in the context of DC immunotherapy might not have functional significance without the simultaneous presence of its ligand CCL21 at the site of vaccination. CCL21 expression by the endothelial cells of the lymphatic vessels is induced by inflammatory stimuli [35–37], and the treatment of the skin at the site of injection with an additional adjuvant  might therefore be of importance to increase the migration of CCR-7-expressing DC into the afferent vessels of the draining lymph nodes.
In conclusion, none of the DC-differentiation agents compared above are, unless in combination, sufficient to generate differentiated, proinflammatory and migratory DC. As soon as such agents become available through good manufacturing practice technology, we suggest that they should be evaluated for optimizing DC adjuvant in anticancer vaccines.
A second aim of the present investigation was to perform a quality assessment of DC prepared from breast cancer patients for therapeutic use in a vaccination trial including test of yield, phenotype and stimulatory capacity. Our data demonstrate that patients have a preserved capacity as DC source; actually, a higher fraction of DC was attained from breast cancer patients with disseminated disease than from healthy donors. However, a relatively high content of CD14+ cells found in the DC gate from both patients and normal donors emphasizes the necessity for further optimization of the generation of clinical grade DC. Increased purity can be obtained by isolation of DC from the final cell population or as shown by the addition of maturation reagents. Otherwise, it is possible that heterogeneity of the DC population can cause variation in the clinical and immunological outcome of the vaccinated patients due to a negative effect of the contaminating cells. Therefore, the maturation of DC is a part of our future clinical protocols. However, the majority of CD14+ cells in the DC gate did express relevant DC markers and can probably function as APC as well. In addition to clinical effect, it is also possible that the heterogeneity of the DC population has in part affected the results of the functional and cytokine secretion assays discussed below.
The ability of tumour tissue to compromise T-cell function by the production of immunosuppressive factors [39, 40], apoptosis induction [41, 42] and interference with frequencies of different T-cell subsets  has previously been described. In contrast, less is known about the effect on DC and most studies have focused on alterations in phenotype and function of DC isolated directly from peripheral blood. In contrast, monocyte-derived in vitro-generated DC are most often used for vaccine trial, but only few studies have described tumour-associated changes of these subsets of DC. In this study, in vitro-generated monocyte-derived DC from breast cancer patients were compared to similar DC from healthy controls. Patient-derived DC exhibited a more mature phenotype when compared with DC from healthy controls, particularly when comparing levels of CD40 and CD54 expression, and this confirms findings for other kinds of DC preparations in cancer patients [8, 10]. Because mature DC do not take up antigen, the intermediate maturity of patient-derived DC might inhibit their uptake of antigen. Indeed, this is the mechanism by which supernatants from tumours have been shown to inhibit the uptake of antigen in monocyte-derived DC . In contrast, other tumour-derived agents, e.g. vascular endothelial growth factor, may inhibit the maturation of DC derived from CD34+ bone marrow cells [44, 45]. Thus, tumours take the advantage of several different strategies to interfere with DC maturation and function.
Very few DC preparations from breast cancer patients responded with IL-12 p70 secretion upon maturation. However, the IL-12 p70 levels obtained from healthy controls were also rather low, probably due to the fact that only a minority of cells in the heterogeneous cell population are in fact DC. Therefore, even though the ratio of patient-derived DC that did not secrete IL-12 p70 was significantly lower compared with healthy controls, caution should be taken when interpreting this result. However, others have found impaired IL-12 production of DC isolated directly from peripheral blood of breast cancer patients and it was suggested that the tumour-secreted immunoactive factor, spermine, was responsible for this effect . Consequently, DC from cancer patients may be expected to have impaired capacity for T-cell stimulation. In agreement, purified DC obtained from peripheral blood or lymph nodes from patients with operable breast cancer were shown to have decreased capacity for both allogeneic T-cell stimulation and autologous T-cell stimulation against purified protein derivate , with a concurrent decrease in HLA-DR, CD86, CD40 expression and low IL-12 secretion. In our study, the monocyte-derived DC preparations generated from patients showed a reduced capacity to induce allogeneic T-cell proliferation compared with DC from healthy controls. On the other hand, our study could not document a defective function of DC to induce autologous CD8 and CD4 T-cell proliferation against the recall antigens TT and CMV lysate and autologous priming against KLH. This difference in ability to stimulate autologous and allogeneic T-cell reactivity might reflect differences in the antigenic nature and cytokines involved in the T-cell activation. However, it is also possible that these diverging results could be a result of the heterogeneity of the DC population which cause variation between donors in the actual number of stimulatory cells. In addition, we used two different assays to measure autologous and allogeneic T-cell stimulation with variation in irradiation procedure and other parameters; hence, these results should be interpreted with caution. However, in support of our data, Gabrilovich et al.  found DC from peripheral blood with impaired T-cell stimulatory capacity in breast cancer patients, although the defective T-cell stimulatory function could not be retrieved in DC generated in vitro from precursor cells.
Suppression of DC can be a result of treatment rather than a result of the cancer disease itself. However, it is not likely that the partial suppression of the DC in our study is a result of chemotherapy or other immunosuppressive therapy, as this was abrogated more than 4 weeks before blood for DC generation was obtained.
Taken together, the demonstrated low ability of IL-12 production and the low capacity for allogeneic stimulation are two factors pointing towards a reduced functionality of the patient-derived DC. The predictive value of these factors for the clinical applicability of the DC preparations is, however, uncertain. The presence of surface molecules relevant for T-cell interaction and activation as well as an unimpaired ability for autologous antigen presentation, on the other hand, indicate retained DC functionality and are important findings as the DC preparations in the clinical study are used for autologous induction of antigen-specific T-cell immunity. In addition, by introducing an optimal combination of maturation factors some of the observed DC deficiencies seem to be restored, and the fraction of CD14– DC increased.
This work was supported by grants from The Danish Cancer Research Foundation, The Danish Cancer Society, The Aase and Ejnar Danielsens Foundation, The Dagmar Marshalls Foundation, Fabrikant Einar Willumsens Mindelegat, The Family Hede Nielsens Foundation, The Else and Mogens Wedell-Wedellsborg Foundation and The Michaelsen Foundation.
- 41Decreased theta expression and apoptosis in CD3+ peripheral blood T lymphocytes of patients with melanoma. Clin Cancer Res 2001;7 (Suppl.):947–55., , et al.