REVIEW Dendritic Cell-Based Vaccine

REVIEW

Dendritic cell-based vaccine: a promising approach for cancer immunotherapy K Tarte1 and B Klein1,2

1INSERM U 475, Montpellier; and 2Unit For Cellular Therapy, CHU Saint Eloi, Montpellier, France

The unique ability of dendritic cells to pick up antigens and to + + and interstitial DC pick up antigens (Ag) in peripheral tissues activate naive and memory CD4 and CD8 T cells raised the and migrate to lymphoid organs through lymphatic vessels in possibility of using them to trigger a specific anti-tumor 8 immunity. If numerous studies have shown a major interest in response to inflammatory stimuli. Activated mature DC, dendritic cell-based vaccines for cancer immunotherapy in ani- which express CD1a and CD83 antigens, represent the latter mal models, only a few have been carried out in human can- stage of LC and interstitial DC maturation. According to their cers. In this review, we describe recent findings in the biology high expression of MHC class I, MHC class II and costimu- of dendritic cells that are important to generate anti-tumor cyto- latory molecules, they present, efficiently, Ag-derived peptides toxic T cells in vitro and we also detail clinical studies reporting to CD4+ and CD8+ T lymphocytes in lymph node T cell areas the obtention of specific immunity to human cancers in vivo using reinfusion of dendritic cells pulsed with tumor antigens. (Figure 1). Recently, another subset of DC was identified on Keywords: dendritic cells; cancer; immunotherapy the basis of their high expression of the interleukin-3 receptor alpha chain (IL-3R␣hi). These cells are able to migrate to lymphoid organs without any prior activation by inflamma- 9 Introduction tory stimuli. The possibility of generating DC in vitro has helped to Despite progress in chemotherapy regimens, several malig- understand their ontogeny and to further characterize their nancies remain incurable by conventional therapies. It is thus phenotype. As illustrated in Figure 2, all DC subsets originate from bone marrow CD34+ hematopoietic precursor cells fol- necessary to develop alternative strategies, in particular anti- 10–12 tumor immunotherapy. Dendritic cells (DC) play a crucial role lowing two overlapping pathways. Granulocyte–macro- in the initiation of primary immune response and are therefore phage colony-stimulating factor (GM-CSF) and tumor necrosis factor ␣ (TNF␣) are the two main cytokines necessary for DC probably the best adjuvant for anti-tumor immunity. However, 13 their very low frequency in vivo has limited research in this maturation. Stem cell factor (SCF) and Flt3 ligand (Flt3L) enhance the yield of DC generated in vitro from CD34+ field. The possibility of obtaining large numbers of DC in vitro 14 has boosted the understanding of their ontogeny and func- cells, and Flt3L-treated mice show a great increase in the number of mature DC.15 The differentiation of CD34+ cells tions. In this paper, we will briefly review the main recent ␤ features of DC characteristics that are important for cancer into LC requires, in addition, transforming growth factor 1 ␤ ␤ −/− vaccination and discuss the various strategies to vaccinate (TGF 1) since TGF 1 knockout mice are completely devoid 16 ␤ cancer patients with DC. of LC. Moreover, TGF 1 cooperate with Flt3L for the development of CD1a+ Lag+LC in vitro.17 DC can also be generated from peripheral blood CD14+ MO, at least in vitro.18–20 In the Dendritic cells: phenotype and ontogeny presence of macrophage colony-stimulating factor (M-CSF), MO are induced to differentiate into CD14+ M⌽. However, DC constitute a heterogeneous leukocyte population defined in the presence of GM-CSF and interleukin-4 (IL-4) or interleukin-13 (IL-13),21–23 macrophage differentiation is blocked24,25 by morphologic, phenotypic and functional criteria which dis- − + − tinguish them from monocytes (MO) and macrophages (M⌽).1 and MO differentiate into CD14 CD1a Lag immature DC. Finally, the addition of TGF␤1 leads to obtaining CD14− Although there is evidence that a lymphoid precursor can gen- + + 26 + − erate DC, T, B and NK cells,2,3 we will focus only on myeloid CD1a Lag LC. CD1a CD83 immature MO-derived DC are lineage-derived DC which are used both in vitro and in vivo induced into highly efficient Ag presenting cells (APC) (CD1a+CD83+ mature DC) in the presence of TNF␣27,28 or in to enhance specific immune response. Originally, DC were 29,30 characterized by the presence of veils, polarized lamellipodia response to CD40 ligand (CD40L) stimulation which mim- and long spiny processes that were continually being ics interaction with T lymphocytes in T cell areas of lymphoid extended and retracted thus allowing them to be highly organs. Interestingly, the phenotype of these differentiated ⌽ motile.4 As illustrated in Figure 1, three major DC populations cells can be reverted since even fully mature M can be induced into CD1a+CD83− immature DC in the presence of have been identified in vivo: Langerhans cells (LC), interstitial + − DC and activated mature DC. LC are present in skin epidermis IL-4, and vice versa, immature CD1a CD83 DC, unlike + + ⌽ and mucosae and express CD1a, the cutaneous associated mature CD1a CD83 DC, can acquire a M phenotype when 31 antigen (CLA), E-cadherin and the Lag antigen associated with GM-CSF and IL-4 are replaced by M-CSF. the presence of Birbeck granules.5–7 CD1a+ interstitial DC are The phenotype of these various DC populations is now bet- found in various organs especially liver, kidney and heart. LC ter known (Table 1). DC are defined as lineage-negative cells, ie they do not express classical T cell, B cell, NK cell and monocyte (especially CD14) markers.1,32,33 The usual DC Correspondence: B Klein, INSERM U475, 99, rue Puech Villa, 34197 markers are: CD1a which is expressed by LC, MO-derived DC Montpellier, Cedex 5, France; Fax: 33 04 67 04 18–63 and lymph node DC but not by blood DC and which is Received 8 September 1998; accepted 27 January 1999 involved in the presentation of lipid and glycolipid Ag to T Review K Tarte and B Klein 654

Figure 1 Immature DC are localized in peripheral tissues and migrate to lymphoid organs where they activate T lymphocytes.

Figure 2 DC ontogeny and lineage.

33–36 44 ␣ ␤ ⌽ cells; CD83 which is present on some rare circulating ecules (Figure 3). In particular, v 5 is absent on MO/M blood DC34 and which is induced during DC maturation and this could explain why these cells are unable to present (cultured blood DC, mature MO-derived DC, lymph node Ag derived from apoptotic cells, even if they internalize apop- DC);37 CMRF-44 which is less specific than CD83 and is also totic cells very efficiently. Ag capture is optimal for immature expressed by a subset of blood DC and LC. CD83 and CMRF- dendritic cells generated from peripheral blood MO after 5– 44 have unknown functions but they are interesting for DC 7 days of culturing with GM-CSF and IL-4. When DC are positive selection.38 induced to mature in vitro, in particular in the presence of TNF␣ or CD40L stimulation, they lose their ability to internalize soluble Ag and apoptotic cells39,44 (Figure 3). Together Dendritic cell functions in vitro and in vivo with Ag uptake, Ag processing is also modulated during DC maturation. Ag are concentrated in MHC II compartments in Again, due to the possibility of producing DC in vitro, their immature cells and, after treatment with TNF␣, these intra- maturation and function have been extensively characterized. cellular vesicules disappear and surface MHC II expression is Immature DC are able to capture and process Ag extensively upregulated.39,45 It is now well documented that immature DC in the periphery. They can internalize soluble Ag by macropi- can process exogenous soluble proteins captured by macropinocytosis,39 ie uptake of large vesicules (0.5–3 ␮m), as well nocytosis for MHC class I presentation.46–48 This could explain as by endocytosis of small vesicules40 by at least two types of the cross-priming phenomenon: in vivo, MHC I-restricted receptors for Ag uptake: Fc receptor and mannose receptor. tumor Ag are presented mainly, if not exclusively, by pro- CD32 (Fc␥RII), CD64 (Fc␥RI) and mannose receptor are fessional APC (mainly DC) that have taken up Ag released by downregulated during DC maturation in vitro39,41,42 (Figure 3). dying tumor cells, rather than by the tumor cells themselves.49 LC and interstitial DC are phagocytic but to a lesser extent This is of major importance for vaccination strategies for can- than M⌽ and this capacity is lost during DC maturation.42,43 cer patients. The method of Ag capture influences its presen- Immature DC are also able to internalize apoptotic cells by tation. Endocytosis via mannose receptor seems more efficient ␣ ␤ ␣ ␤ binding them through v 5, v 3 and CD36 adhesion mol- for MHC class I presentation than endocytosis via Fc recep- Review K Tarte and B Klein 655 Table 1 Surface molecules express on DC kines, namely macrophage inflammatory protein (MIP)-1␣, MIP-1␤ and RANTES. Induction of DC maturation by CD40 Functions ligand (CD40L), TNF␣ or LPS results in a rapid downregulation of these receptors, allowing DC to leave inflamed Dendritic-specific markers tissues. It also results in the slower induction of CCR7, which CD1 MHC-like molecules is the receptor for MIP-3␤/EBI1 ligand chemokine (ELC) and Presentation of non-peptide Ag to T cells secondary lymphoid tissue chemokine (SLC), two chemokines CD83 Unknown function constitutively produced in lymphoid organs. In the lymph CMRF-44 Ag Unknown function nodes, DC form clusters with T lymphocytes through various CLA E-selectin ligand adhesion molecules, in particular CD54 (ICAM-1), CD102 Skin homing of LC (ICAM-3) and CD58 (LFA-3), that are highly expressed by activated DC.55 The DC/T cell interaction is stabilized by the spe- Ag uptake CD32 (Fc␥RII) Antibody binding cific ligation of the T cell receptor (TCR) by the MHC–peptide CD64 (Fc␥RI) Antibody binding complex, which delivers the first activation signal to T lym- Mannose receptor phocytes. A second costimulatory signal is absolutely necessary to avoid T cell anergy and allow T cell proliferation. This Adhesion molecules signal is mediated in part by the interaction of CD80 (B7–1) CD54 (ICAM-1) CD11a (LFA-1) ligand and CD86 (B7–2), both of which present on DC, with T cell CD58 (LFA-3) CD2 ligand CD102 (ICAM-3) CD11a (LFA-1) ligand CD28. Immature resting DC express few CD80 and CD86 but these molecules are rapidly induced upon differentiation/ MHC molecules activation,56 especially after contact with T lymphocytes MHC I Peptide presentation to CD8+ T which allows CD40–CD40L interaction.30 CD40 is constitut- cells + ively expressed by DC and is upregulated by several cytokines MHC II Peptide presentation to CD4 T such as GM-CSF and TNF␣.57 CD40 stimulation upregulates cells CD80–CD86 expression on DC57 and induces them to pro- 29,58 Costimulatory molecules duce p75 interleukin-12 (IL-12), which plays the key role CD80 (B7-1) Interaction with CD28 (second in the emergence of Th1 cellular response. Recent studies signal) show that in vivo DC initially activate CD4+ T cells (Figure CD86 (B7-2) Interaction with CD28 (second 4). Activated CD4+ T lymphocytes become CD40L+ and signal) deliver, in turn, a stimulation signal to CD40+ DC. These ‘con- CD40 DC differentiation/activation + ditioned’ DC are then able to prime Ag-specific CD8 T cells. (interaction with CD40L on T cell) + OX40L Activation of OX40+ CD4+ T cells This two-step model of T cell activation explains why CD4 DC differentiation/activation help is necessary to get cytotoxic T lymphocytes. However, CD4+ help can be replaced by anti-CD40 antibodies which might be interesting for developing anti-tumor cytotoxic T lymphocytes (CTL) in vitro using DC pulsed with MHC I- tor.50 In addition, engulfed apoptotic cells generate MHC class restricted peptides.59–61 Other costimulatory molecules are I-restricted peptides.44,51,52 likely to be important for T cell activation. Among them, After Ag uptake and processing, DC leave peripheral tissues OX40 ligand (OX40 L) is inducible by CD40 stimulation on and migrate to lymphoid organs in order to present Ag to T DC62 and it might play a major role in T cell priming. First, lymphocytes (Figure 1).8 This migration could be explained OX40 L engagement increases DC maturation and expression by the switch in chemokine receptor expression induced of CD80 and CD86.62 Secondly, OX40 cross-linking on CD4+ through DC maturation.53,54 Immature DC express CCR1 and activated T cells costimulates their proliferation and cytokine CCR5, which are the receptors for some inflammatory chemo- production.63 Expression of high amounts of MHC and

Figure 3 DC differentiation and maturation in vitro: modulation of molecules involved in Ag uptake, Ag presentation and DC migration. Review K Tarte and B Klein 656

Figure 4 Cross-talk of DC with CD4+ and CD8+ T cells. Role of CD40 and other costimulatory molecules.

costimulatory molecules, and production of IL-12 by mature ate easily the phase of Ag capture and processing and the DC, make them very potent professional APC. phase of Ag presentation since TNF␣ or other maturating stimuli can be added in the course of the culture.27,28 Thus, tumor Ag requiring internalization and processing (soluble proteins, Generation of human DC for vaccination strategies RNA, apoptotic cells) must be added to immature CD1a+ CD83− DC (5–7 days of culture with GM-CSF and IL-4). Con- According to their ontogeny, DC from three sources can be versely, defined 9-mer synthetic peptides which bind directly used to vaccinate patients with cancer. DC can be purified to MHC molecules without processing must be added to from peripheral blood. After a short in vitro culture, their low mature DC, which express high amounts of MHC class I and density allows enrichment by using centrifugation on metriza- II after 2 additional days of culture with TNF␣, CD40L or mon- mide or Nycodenz gradients,32,64 or elutriation.65 Blood DC ocyte-conditioned medium. Noteworthy, the use of apheresis can also be purified after removal of other defined T, B, NK cells collected in very large amounts during hematopoietic and MO populations by using antibodies and magnetic beads precursor mobilization with growth factors greatly enhances or a cell-sorter, on the basis of their HLA DR++ and lineage− the DC yield.19,69 This is explained by the proliferation and phenotype35 or according to their expression of CD83 or differentiation in vitro of some non-CD34+ adherent DC pre- CMRF-44 molecules.38 However, the very low frequency of cursors that are not present in normal non-mobilized periph- DC in accessible body samples, especially blood, limits the eral blood. This proliferation can be increased by adding use of these DC for vaccination protocols. IL-3 and TNF␣,69 indicating that it might involve some It is now far easier to generate DC in vitro. According to IL-3R␣hi cells. To date, apheresis samples are collected the knowledge of DC ontogeny (Figure 2), two major strategies after treatment with G-CSF or GM-CSF. However, it would be are used. The first is based on the ability of CD34+ progenitors interesting to study the effect of other mobilization reagents, isolated from bone marrow,66,67 peripheral blood68 or neo- particulary SCF and Flt3L, on DC precursor mobilization.14 natal cord blood11 to differentiate in vitro within 14 days into The reason why the majority of in vitro or in vivo studies use mature CD1a+ CD83+ HLA DR++ DC in the presence of GM- these MO-derived DC to generate anti-tumor immunity is that CSF and TNF␣. The addition of IL-4 could enhance DC purity they are easy to obtain. but not DC yield, unlike SCF and Flt3L which synergistically For immunization protocols, it is important to minimize the allow the obtention of large numbers of DC. One of the presence of xenogenic or allogenic Ag, especially coming advantages of this methodology is the high rate of cell prolifer- from fetal calf serum, which may be immunodominant, ation which occurs during the first days of cell culture and although these Ag could be interesting adjuvants for in vivo allows gene transfer with retrovirus. However, there are sev- immunization. In addition, serum contains proteases that may eral limitations: these cell populations contain only 50% of alter the antigenicity of MHC I-restricted peptides added in DC, several cytokines are required and it is difficult to clearly vitro on DC preparations.70 Finally, serum contains numerous separate the phase of Ag capture and processing from the proteins, particularly antibodies, which may interfere with the phase of Ag presentation, especially since TNF␣ needs to be method of Ag capture and processing. This has been shown added at the beginning of the culture. for human natural anti-Gal␣(1,3)Gal antibodies that crossreact The second major strategy uses purified peripheral blood with MUC-1 peptides and therefore drive an endocytosis of CD14+ MO or adherent mononuclear cells and makes it poss- mannose receptor-targeted MUC-1 peptides through Fc and ible to obtain a cell population containing 90% of immature complement receptors. This induces a MUC-1 presentation CD1a+ CD83− DC in the presence of only two cytokines (GM- through MHC II molecules and favors antibody production.50 CSF and IL-4 or IL-13).18–20,69 As illustrated in Figure 3 and On the other hand, when these mannose receptor-targeted discussed above, this methodology makes it possible to separ- MUC-1 peptides are endocytosed through mannose receptor, Review K Tarte and B Klein 657 as occurs in mice, which lack anti-Gal␣(1,3)Gal antibodies, a ral, transfer could allow stable expression, processing and CTL response is induced. Several studies have thus focused presentation of TAA epitopes.101 on the generation of DC in FCS-free conditions, starting from There are three major restrictions when using defined tumor CD34+ progenitors or adherent PBMC. They generally use Ag. Firstly, identification of effective TAA is lacking in a large serum-free media alone (especially AIM-V, X-VIVO 15 or X- number of cancers. Secondly, it is now clear that immunother- VIVO 20)17,22,71,72 or supplemented with 1–10% of autologous apy approaches directed against a unique Ag will favor the plasma,23,73,74 autologous serum75,76 or pooled human AB selection of tumor mutants that have selectively lost the ability serum.73,77,78 TGF␤1 can replace serum for obtaining CD1a+ to present efficiently the defined Ag.107 Thirdly, the obtention DC from CD34+ progenitors in the presence of GM-CSF, of Ag-specific CTL in vitro is not a guarantee of the develop- TNF␣, SCF and Flt3L.17 Monocyte-conditioned medium23,73 or ment of a curative immune response in vivo. To bypass these TNF␣ + prostaglandin E2 (PGE2)74,77 can be added, after 5–7 disadvantages, several alternative methodologies using a mix days of culture, to trigger the maturation of MO-derived DC of TAA have been developed. Unfractionated MHC I- obtained with GM-CSF and IL-4. This will enhance their presented peptides can be eluted from tumor cells108 and ability to activate T lymphocytes and will induce the loaded on to DC.109 DC are also able to internalize complete expression of appropriate chemokine receptors, especially cell lysates110 or apoptotic cells44,51,52 and to present derived CCR7, to guide localization of DC into lymphoid organs after Ag in an HLA I-restricted manner. For these strategies, the DC injection in vivo. major concerns are the need for large numbers of primary The choice of culture conditions, especially the cytokine samples or tumor cell lines and the risk of autoimmune reac- combination, will influence DC purity, maturation and function. As discussed recently,111 the use of RNA instead of pro- tionality, and this is a consideration of prime importance tein could constitute a good alternative since it could be before starting DC-based immunotherapy assays. amplified in vitro. In addition, substractive hybridization could allow the enrichment of tumor-specific RNA, thus limit- ing immune response against self Ag.110,112,113 Two other Suppression of endogenous DC function in patients with methods have been explored. The first one is the fusion of cancer murine tumor cell lines with bone marrow-derived DC (Figure 5b). This interesting work leads to the obtention of a hybrid It has been reported that tumor infiltration by DC is a factor displaying functional properties of DC and expressing tumor of good prognosis in several cancers.79–82 However, there is Ag.114,115 However, cell fusion followed by selection requires no evidence that these DC are able to pick up and efficiently pure and proliferating neoplastic cells which is a major techni- present tumor-associated Ag (TAA). In particular, tumor cells cal limitation when applied to human cancers. In this context, produce cytokines that inhibit DC differentiation, recruitment the recent demonstration that cell-contact generated by short- or functions. Interleukin-10 (IL-10), a potent immunosuppres- term coculture of DC and tumor cells without cell fusion sive cytokine, is produced by melanoma, brain neoplasm, B results in a potent immunogen is particularly relevant.116 The lymphoma, myeloma and several carcinoma tumor cells.83–88 second method is specific to the human chronic myeloid leu- In vitro, IL-10 prevents the generation of MO-derived DC kemia (CML) model (Figure 5c). In this disease, tumor cells induced by GM-CSF and IL-4 and favors the development of can be induced to differentiate, in vitro, in the presence of MO.89,90 In vivo, IL-10 production by tumor cells has been GM-CSF, IL-4 and TNF␣, into mature DC able to stimulate shown to prevent DC accumulation and anti-tumor immunity autologous anti-leukemic CTL.117 Finally, tumor Ag-presenting in a murine model.91 In addition, all immunosuppressive treat- DC can be used directly to generate specific CTL in vitro and ments used in cancer, especially high-dose chemotherapy in vivo or to produce exosomes. These secreting vesicules associated with auto- or allograft, could affect DC. express MHC I, MHC II and CD86 molecules and are efficient in priming CTL, at least in vivo in murine models.118 Among all these different methodologies, those which have Preclinical studies of anti-tumor vaccination with DC been successfully used in vitro to obtain human anti-tumor CTL are summarized in Table 2. These studies are encour- Several approaches, depending on the TAA added on DC, aging and they used HLA-restricted peptides,72,94,95,97,98,119 have been used. They are summed up in Figure 5. The objec- cloned cDNA,101 cell lysates,72 irradiated tumor cells76 or tive is to obtain professional TAA presenting cells using tumor tumor RNA113 as tumor Ag. The source of DC was either pur- cells that express TAA but fail to present them, and autologous ified DC, MO-derived DC or CD34+-derived DC. Optimiz- DC that express no TAA. A number of tumor Ag recognized ation of the different steps, ie Ag uptake, Ag presentation and by CD8+ CTL have already been identified92 (Figure 5a). These coculture protocols is now in process. For example, it has defined Ag can simply be added as 9-11 mer HLA I-restricted been demonstrated that treatment of MO-derived DC, peptides to mature DC preparation in the presence of ␤2 obtained in serum-free AIM-V medium, with anti-sense oli- microglobulin.93–97 However, it has been suggested that the gonucleotides to the mRNA of transporter associated with use of longer peptides or complete proteins,98,99 which are not antigen presentation (TAP) or with MG132 proteasome inhibi- directly presented but are endocytosed and processed, is a tor decreases the association of MHC I molecules with endo- better way to generate a fully competent DC vaccine since it geneously generated peptides, thus enhancing exogeneously allows selection by the APC of the optimal 9-mer peptide for added peptide presentation.119 Immature MO-derived DC are presentation.98 After cloning, these tumor Ag can also be used very interesting since their high endocytic ability allows them as cDNA in appropriate vectors to load DC. Transfection of to capture tumor proteins and to internalize tumor RNA with- human DC have been reported using adenovirus for MO- out any transfection reagent being added.113 However, it derived DC100 and retrovirus for CD34+-derived DC101,102 but, should be emphasized that, to date, the great majority of the in contrast to murine models,103–106 the efficiency of trans- studies reporting the in vitro obtention of CTL to human tumor fected DC to stimulate anti-tumor CTL in vivo has not yet been cells were carried out with DC and T cells of healthy individ- demonstrated in human. The use of retroviral, unlike adenovi- uals. Only five studies have used cells obtained from Review K Tarte and B Klein 658

Figure 5 Different methodologies used to obtained efﬁcient tumoral Ag presenting DC: Ag uptake (a), fusion or cell-contact with tumoral cells (b) and CML cells differentiation (c).

Table 2 In vitro obtention of anti-tumor CTL using DC in human cancers

Stimulator cells Added antigen Responder Protocol of culture Results Ref. (S) cells (R) S/R Cytokines No. of ratio boosts

CML CML-derived DC None PBMC 1/3 IL-2 2 CTL 117 RPMI-FCS Puriﬁed DC Peptides from CD3+ T cells 1/20 IL-1 + IL-2 1 with DC CTL 98 MO for reboosts bcr-abl junction CD8+ T cells IL-2 + IL-4 5–6 with MO Not HLA-restricted from day 10 lysis

Melanoma MO-derived DC Irradiated T cells 1/10 None 1 T cell proliferation 76 RPMI-autologous melanoma cells serum MO-derived DC gp100 peptides PBMC 1/10 IL-7 + IL-10 5 CTL for 3/5 72 AIM-V cell lysates IL-2 patients MO-derived DC gp100 peptides CD8+ T cells 1/20 IL-7 + IL-10 1 with DC CTL against 94 RPMI-SAB IL-10 + IL-2 5 with MO subdominant MO for reboosts peptides CD34-derived DC MART-1 cDNA CD3+ enriched 1/10 IL-2 2 CTL for 1/3 101 Iscove-SAB Retrovirus T cells patients MO-derived DC Melanoma CD8+ T cells 1/10 IL-7 4 CTL for 3/3 normal 97 Iscove-FCS peptides IL-7 + IL-2 donors

Others MO-derived DC CEA RNA PBMC 1/10 IL-2 + IL-7 2 CTL 113 RPMI-FCS or RNA from CEA+ CD8+ T cell Immature DC do AIM-V cell line from day 12 not need lipids for transfection MO-derived DC EBV, CMV, CEA PBMC 1/10 IL-2 + IL-7 2 CTL 119 AIM-V peptides + TAP-AS CD8+ T cell Increase or MG132 from day 12 exogenously added peptide presentation Enriched DC PAP peptides Enriched T 1/10 IL-2 3–4 CTL for 5/12 95 cells normal donors

TAP-AS, transporter associated with Ag presentation-antisense; PAP, prostatic acid phosphatase. Review K Tarte and B Klein 659 patients.72,94,101,113,117 In addition, a CTL activity against auto- four of them demonstrating a clinical response to treatment logous tumor cells was reported in only two studies72,117 (less (one complete response (CR), two partial responses (PR) and than 10 patients in all): one performed on CML patients and one minor response (MR)). Among the four patients treated one on melanoma patients. with DC pulsed with tumor cell lysates, two developed DTH reactivity toward the lysate in association with clinical response (one CR and one PR). The two patients with CR were Clinical studies of anti-tumor vaccination with DC still tumor-free 15 months later. Finally, the third study was carried out on metastatic renal cancer patients using MO- Only five studies concerning the in vivo use of DC in human derived DC pulsed with KLH and autologous tumor cell lys- cancer have been published to date120–125 (Table 3) but others ate.125 Only two patients completed the treatment and one of have now been started. Three of them report objective clinical them showed a proliferative response of PBMC against cell responses. The first study was carried out on B cell lymphoma lysates in association with a regression of tumor mass of 61%. patients using DC purified from peripheral blood by density The method of administration and the nature of adjuvants are gradient centrifugation, pulsed with idiotype protein (Id) and certainly more important than the absolute number of rein- reinfused.124 All lymphoma patients treated developed a cellu- jected DC since as few as 1 to 2 × 106 DC are able to mount lar response as shown by the proliferation of PBMC in the a specific immune response. On this basis, Nestle et al121 used presence of tumor Id without anti-Id antibody production. a direct intralymphatic injection to minimize DC loss. One Among four patients studied, one had a partial regression of may hypothesize that the knowledge of chemokine receptor some tumor sites, one underwent a complete tumor regression modulation will help to direct reinjected DC to lymphoid and, for one patient with equivocal radiographic findings, organs. In addition, DC were generated in the presence of evaluation of residual disease by PCR showed that bone mar- xenogenic FCS Ag and KLH neo-Ag to favor an adjuvant T row and blood tumor cells had disappeared. The second study helper cell response.121 Another important factor which gov- was carried out on patients with advanced melanoma using erns the efficiency of DC-based vaccines is the immunological MO-derived DC pulsed with a mixture of HLA-restricted status of treated patients. It must be emphasized that all clini- melanoma peptides or with cell lysates to avoid the selection cal assays were performed on advanced stage disease. Immu- of TAA− tumor cells.121 Biological response was evaluated by nocompetent young patients with minimal residual disease delayed-type hypersensitivity (DTH) reactivity. Among the 12 would probably be the best candidates for these approaches. patients treated with DC pulsed with A1 and/or A2-restricted Nevertheless, biological anti-tumor responses were detectable peptides, nine showed a DTH reactivity to DC loaded with in all studies and DC vaccine seems to be well tolerated with these peptides. However, a significant DTH reactivity no significant adverse events despite the high rate of autoanti- (Ͼ10 mm in diameter) was observed in only five patients with bodies reported by Nestle et al121 (three patients with anti-

Table 3 In vivo immunization with TAA-pulsed DC in human cancers

Stimulator cells Added antigen Protocol of immunization Results Ref.

Melanoma DC-like cells MAGE-1 peptide Intradermal injection Good tolerance 120 (Adherent-cells in GM- 5 × 105 to 1 × 107 DC + Biological response in CSF alone) 1 i.v. of 108 cells 2/3 patients (cytotoxic AIM-V TIL) 0 Clinical response MO-derived DC Mix of melanoma Injection in lymph nodes Good tolerance 121 RPMI-FCS peptides or cell 1 × 106 DC + KLH Autoantibodies in 4/16 lysates patients Biological response (DTH) in 11/16 patients 6/16 patients with clinical response: 2 CR, 3PRand1MR

Prostatic MO-derived DC PMSA peptides i.v. Good tolerance 122–123 cancer Optimem-Autologous 2 × 106 to 2 × 107 DC ± 9/33 patients with PR plasma GM-CSF

B cell Puriﬁed DC Idiotypic protein + i.v. Good tolerance 124 lymphoma KLH 2 × 106 to 32 × 106 Biological response idiotype- and KLH- (PBMC proliferation) in pulsed DC 4/4 patients + injection of soluble Ag Clinical response in 3/4 after each DC infusion patients with 1 CR

Renal cancer MO-derived DC Tumor cell lysates + i.v. Biological response 125 RPMI-SAB + HSA KLH 5 × 106 to 10 × 106 DC (PBMC proliferation) in 1/2 patients Clinical response (PR) in 1/2 patients

CR, complete response; PR, partial response; MR, minor response; DTH, delayed-type hypersensitivity. Review K Tarte and B Klein 660 TSH receptor antibodies and one patient with antinuclear CSF and TNF-alpha cooperate in the generation of dendritic Lan- antibodies). In patients with multiple myeloma (MM), a late B gerhans cells. Nature 1992; 360: 258–261. cell disorder, the monoclonal immunoglobulin should be an 14 Lyman SD, Jacobsen SE. c-kit ligand and Flt3 ligand: stem/progenitor cell factors with overlapping yet distinct activi- interesting tumor Ag since its genes are somatically hypermut- ties. Blood 1998; 91: 1101–1134. 126 ated and since it can be purified from serum. Accordingly, 15 Maraskovsky E, Brasel K, Teepe M, Roux ER, Lyman SD, Short- one study reports an Id-specific T cell proliferative response man K, Mckenna HJ. Dramatic increase in the numbers of func- after vaccination of only one MM patient with autologous tionally mature dendritic cells in Flt3 ligand-treated mice: mul- MO-derived DC pulsed with monoclonal component. IgM tiple dendritic cell subpopulations identified. J Exp Med 1996; and IgG anti-Id antibodies were detected without any clinical 184: 1953–1962. 127 16 Borkowski TA, Letterio JJ, Farr AG, Udey MC. A role for endogen- response. Several other studies are now in progress in this ous transforming growth factor beta 1 in Langerhans cell biology: disease. the skin of transforming growth factor beta 1 null mice is devoid All these preliminary data are very promising and indicate of epidermal Langerhans cells. J Exp Med 1996; 184: 2417–2422. that DC could be safely used in vivo as immunotherapy adju- 17 Strobl H, Bello-Fernandez C, Riedl E, Pickl WF, Majdic O, Lyman vants in a wide variety of cancers even without identified SD, Knapp W. flt3 ligand in cooperation with transforming TAA. growth factor-beta1 potentiates in vitro development of Langer- hans-type dendritic cells and allows single-cell dendritic cell cluster formation under serum-free conditions. Blood 1997; 90: 1425–1434. Acknowledgements 18 Pickl WF, Majdic O, Kohl P, Sto¨ckl J, Riedl E, Scheinecker C, Bello-Fernandez C, Knapp W. Molecular and functional charac- + This work was supported by grants from LFNC axe ‘Immuno- teristics of dendritic cells generated from highly purified CD14 peripheral blood monocytes. J Immunol 1996; 157: 3850–3859. the´rapie des cancers’ (Paris, France) and AFS (Paris, France). 19 Romani N, Gruner S, Brang D, Kampgen E, Lenz A, Trocken- bacher B, Konwalinka G, Fritsch PO, Steinman RM, Schuler G. Proliferating dendritic cell progenitors in human blood. J Exp References Med 1994; 180: 83–93. 20 Chapuis F, Rosenzwajg M, Yagello M, Ekman M, Biberfeld P, Gluckman JC. Differentiation of human dendritic cells from mon- 1 Hart DNJ. Dendritic cells: unique leukocyte populations which ocytes in vitro. Eur J Immunol 1997; 27: 431–441. control the primary immune response. Blood 1998; 90: 3245– 21 Piemonti L, Bernasconi S, Luini W, Trobonjaca Z, Minty A, Allav- 3287. ena P, Mantovani A IL-13 supports differentiation of dendritic 2 Galy A, Travis M, Cen D, Chen B. Human T, B, natural killer, cells from circulating precursors in concert with GM-CSF. Eur and dendritic cells arise from a common bone marrow progenitor Cytokine Netw 1995; 6: 245–252. cell subset. Immunity 1995; 3: 459–473. 3 Res P, Martinez-Caceres E, Cristina Jaleco A, Staal F, Noteboom 22 Tarte K, Olsen SJ, Lu ZY, Legouffe E, Rossi JF, Chang Y, Klein B. E, Weijer K, Spits H. CD34+CD38dim cells in the human thymus Clinical grade functional dendritic cells from patients with mul- can differentiate into T, natural killer, and dendritic cells but are tiple myeloma are not infected with Kaposi’s sarcoma-associated distinct from pluripotent stem cells. Blood 1996; 87: 5196–5206. herpesvirus. Blood 1998; 91: 1852–1857. 4 Steinman RM. The dendritic cell system and its role in immuno- 23 Romani N, Reider D, Heuer M, Ebner S, Kampgen E, Eibl B, Nie- genicity. Annu Rev Immunol 1991; 9: 271–296. derwieser D, Schuler G. Generation of mature dendritic cells 5 Tang A, Amagai M, Granger LG, Stanley JR, Udey MC. Adhesion from human blood. An improved method with special regard to of epidermal Langerhans cells to keratinocytes mediated by E- clinical applicability. J Immunol Meth 1996; 196: 137–151. cadherin. Nature 1993; 361: 82–85. 24 Lauener RP, Goyert SM, Geha RS, Vercelli D. Interleukin 4 6 Yasaka N, Furue M, Tamaki K. Expression of cutaneous lympho- down-regulates the expression of CD14 in normal human monocyte-associated antigen defined by monoclonal antibody HECA- cytes. Eur J Immunol 1990; 20: 2375–2381. 452 on human Langerhans cells. J Dermatol Sci 1996; 11: 19–27. 25 Jansen JH, Wientjens GJ, Fibbe WE, Willemze R, Kluin-Nelemans 7 Kashihara M, Ueda M, Horiguchi Y, Furukawa F, Hanaoka M, HC. Inhibition of human macrophage colony formation by Imamura S. A monoclonal antibody specifically reactive to interleukin 4. J Exp Med 1989; 170: 577–582. human Langerhans cells. J Invest Dermatol 1986; 87: 602–607. 26 Geissmann F, Prost C, Monnet JP, Dy M, Brousse N, Hermine 8 Roake JA, Rao AS, Morris PJ, Larsen CP, Hankins DF, Austyn O. Transforming growth factor beta 1, in the presence of JM. Dendritic cell loss from nonlymphoid tissues after systemic granulocyte/macrophage colony-stimulating factor and interleu- administration of lipopolysaccharide, tumor necrosis factor, and kin 4, induces differentiation of human peripheral blood mono- interleukin 1. J Exp Med 1995; 181: 2237–2247. cytes into dendritic Langerhans cells. J Exp Med 1998; 187: 9 Olweus J, BitMansour A, Warnke R, Thompson PA, Carballido J, 961–966. Picker LJ, Lund-Johansen F. Dendritic cell ontogeny: a human 27 Sallusto F, Lanzavecchia A. Efficient presentation of soluble anti- dendritic cell lineage of myeloid origin. Proc Natl Acad Sci USA gen by cultured human dendritic cells is maintained by 1997; 94: 12551–12556. granulocyte/macrophage colony-stimulating factor plus interleu- 10 Szabolcs P, Avigan D, Gezelter S, Ciocon DH, Moore MA, Stein- kin 4 and downregulated by tumor necrosis factor alpha. J Exp man RM, Young JW. Dendritic cells and macrophages can Med 1994; 179: 1109–1118. mature independently from a human bone marrow-derived, post- 28 Chen B, Shi Y, Smith JD, Choi D, Geiger JD, Mule JJ. The role colony-forming unit intermediate. Blood 1996; 87: 4520–4530. of tumor necrosis factor alpha in modulating the quantity of per- 11 Caux C, Vanbervliet B, Massacrier C, Dezutter Dambuyant C, de ipheral blood-derived, cytokine-driven human dendritic cells and Saint Vis B, Jacquet C, Yoneda K, Imamura S, Schmitt D, Banch- its role in enhancing the quality of dendritic cell function in + ereau J. CD34+ hematopoietic progenitors from human cord presenting soluble antigens to CD4 T cells in vitro. Blood 1998; blood differentiate along two independent dendritic cell path- 91: 4652–4661. ways in response to GM-CSF + TNF alpha. J Exp Med 1996; 184: 29 Cella M, Scheidegger D, Palmer Lehmann K, Lane P, Lane Lanza- 695–706. vecchia A, Alber G. Ligation of CD40 on dendritic cells triggers 12 Caux C, Massacrier C, Vanbervliet B, Dubois B, Durand I, Cella production of high levels of interleukin-12 and enhances T cell M, Lanzavecchia A, Banchereau J. CD34+ hematopoietic pro- stimulatory capacity: T-T help via APC activation. J Exp Med genitors from human cord blood differentiate along two inde- 1996; 184: 747–752. pendent dendritic cell pathways in response to granulocyte– 30 Caux C, Massacrier C, Vanbervliet B, Dubois B, Van Kooten C, macrophage colony-stimulating factor plus tumor necrosis factor Durand I, Banchereau J. Activation of human dendritic cells alpha: II. Functional analysis. Blood 1997; 90: 1458–1470. through CD40 cross-linking. J Exp Med 1994; 180: 1263–1272. 13 Caux C, Dezutter Dambuyant C, Schmitt D, Banchereau J. GM- 31 Palucka KA, Taquet N, Sanchez-Chapuis F, Gluckman JC. Den- Review K Tarte and B Klein 661 dritic cells as the terminal stage of monocyte differentiation. J totic bodies yields T cell epitopes. J Immunol 1997; 159: Immunol 1998; 160: 4587–4595. 5391–5399. 32 Freudenthal PS, Steinman RM. The distinct surface of human 52 Albert ML, Sauter B, Bhardwaj N. Dendritic cells acquire antigen blood dendritic cells, as observed after an improved isolation from apoptotic cells and induce class I-restricted CTLs. Nature method. Proc Natl Acad Sci USA 1990; 87: 7698–7702. 1998; 392: 86–89. 33 Thomas R, Davis LS, Lipsky PE. Isolation and characterization of 53 Sozzani S, Allavena P, D’Amico G, Luini W, Bianchi G, Kataura human peripheral blood dendritic cells. J Immunol 1993; 150: M, Imai T, Yoshie O, Bonecchi R, Mantovani A. Differential regu- 821–834. lation of chemokine receptors during dendritic cell maturation: 34 Zhou LJ, Tedder TF. Human blood dendritic cells selectively a model for their trafficking properties. J Immunol 1998; 161: express CD83, a member of the immunoglobulin superfamily. J 1083–1086. Immunol 1995; 154: 3821–3835. 54 Sallusto F, Schaerli P, Loetscher P, Schaniel C, Leing D, Mackay 35 O’Doherty U, Steinman RM, Peng M, Cameron PU, Gezelter S, CR, Qin S, Lanzavecchia A. Rapid and coordinated switch in Kopeloff I, Swiggard WJ, Pope M, Bhardwaj N. Dendritic cells chemokine receptor expression during dendritic cell maturation. freshly isolated from human blood express CD4 and mature into Eur J Immunol 1998; 28: 2760–2769. typical immunostimulatory dendritic cells after culture in mono- 55 Scheeren RA, Koopman G, Van der Baan S, Meijer CJ, Pals ST. cyte-conditioned medium. J Exp Med 1993; 178: 1067–1076. Adhesion receptors involved in clustering of blood dendritic cells 36 Porcelli SA, Segelke BW, Sugita M, Wilson IA, Brenner MB. The and T lymphocytes. Eur J Immunol 1991; 21: 1101–1105. CD1 family of lipid antigen-presenting molecules. Immunol 56 McLellan AD, Starling GC, Williams LA, Hock BD, Hart DN. Today 1998; 19: 362–368. Activation of human peripheral blood dendritic cells induces the 37 Zhou LJ, Tedder TF. CD14+ blood monocytes can differentiate CD86 co-stimulatory molecule. Eur J Immunol 1995; 25: into functionally mature CD83+ dendritic cells. Proc Natl Acad 2064–2068. Sci USA 1996; 93: 2588–2592. 57 McLellan AD, Sorg RV, Williams LA, Hart DN. Human dendritic 38 Fearnley DB, McLellan AD, Mannering SI, Hock BD, Hart DN. cells activate T lymphocytes via a CD40: CD40 ligand-depen- Isolation of human blood dendritic cells using the CMRF-44 dent pathway. Eur J Immunol 1996; 26: 1204–1210. monoclonal antibody: implications for studies on antigen- 58 Heufler C, Koch F, Stanzl U, Topar G, Wysocka M, Trinchieri presenting cell function and immunotherapy. Blood 1997; 89: G, Enk A, Steinman RM, Romani N, Schuler G. Interleukin-12 is 3708–3716. produced by dendritic cells and mediates T helper 1 develop- 39 Sallusto F, Cella M, Danieli C, Lanzavecchia A. Dendritic cells ment as well as interferon-gamma production by T helper 1 cells. use macropinocytosis and the mannose receptor to concentrate Eur J Immunol 1996; 26: 659–668. 59 Ridge JP, Di Rosa F, Matzinger P. A conditioned dendritic cell macromolecules in the major histocompatibility complex class II + compartment: downregulation by cytokines and bacterial pro- can be a temporal bridge between a CD4 T-helper and a T- ducts. J Exp Med 1995; 182: 389–400. killer cell. Nature 1998; 393: 474–478. 40 Steinman RM, Swanson J. The endocytic activity of dendritic 60 Bennett SR, Carbone FR, Karamalis F, Flavell RA, Miller JF, cells. J Exp Med 1995; 182: 283–288. Health WR. Help for cytotoxic-T-cell responses is mediated by 41 Fanger NA, Voigtlaender D, Liu C, Swink S, Wardwell K, Fisher CD40 signalling. Nature 1998; 393: 478–480. J, Graziano RF, Pfefferkorn LC, Guyre PM. Characterization of 61 Schoenberger SP, Toes RE, van der Voort EI, Offringa R, Melief expression, cytokine regulation, and effector function of the high CJ. T-cell help for cytotoxic T lymphocytes is mediated by CD40– affinity lgG receptor Fc gamma RI (CD64) expressed on human CD40L interactions. Nature 1998; 393: 480–483. blood dendritic cells. J Immunol 1997; 158: 3090–3098. 62 Ohshima Y, Tanaka Y, Tozawa H, Takahashi Y, Maliszewski C, 42 Fanger NA, Wardwell K, Shen L, Tedder TF, Guyre PM. Type I Delespesse G. Expression and function of OX40 ligand on human (CD64) and type II (CD32) Fc gamma receptor-mediated phago- dendritic cells. J Immunol 1997; 159: 3838–3848. 63 Godfrey WR, Fagnoni FF, Harara MA, Buck D, Engleman EG. cytosis by human blood dendritic cells. J Immunol 1996; 157: Identification of a human OX-40 ligand, a costimulator of CD4+ 541–548. T cells with homology to tumor necrosis factor. J Exp Med 1994; 43 Reis S, Stahl PD, Austyn JM. Phagocytosis of antigens by Langer- 180: 757–762. hans cells in vitro. J Exp Med 1993; 178: 509–519. 64 McLellan AD, Starling GC, Hart DN. Isolation of human blood 44 Albert ML, Pearce SFA, Francisco LM, Sauter B, Roy P, Silverstein dendritic cells by discontinous Nycodenz gradient centrifugation. RL, Bhardwaj N. Immature dendritic cells phagocytose apoptotic J Immunol Meth 1995; 184: 81–89. cells via alphavbeta5 and CD36, and cross-present antigens to 65 Kabel PJ, de Haan-Meulman M, Voorbji HA, Kleingeld M, Knol cytotoxic T lymphocytes. J Exp Med 1998; 188: 1359–1368. EF, Drexhage HA. Accessory cells with a morphology and marker 45 Nijman HW, Kleijmeer MJ, Ossevoort MA, Oorschot VM, Vier- pattern of dendritic cells can be obtained from elutriator-purified boom MP, van de Keur M, Kenemans P, Kast WM, Geuze HJ, blood monocyte fractions. An enhancing effect of metrizamide Melief CJ. Antigen capture and major histocompatibility class II in this differentiation. Immunobiology 1989; 179: 395–341. compartments of freshly isolated and cultured human blood den- 66 Reid CD, Stackpoole A, Meager A, Tikerpae J. Interactions of dritic cells. J Exp Med 1995; 182: 163–174. tumor necrosis factor with granulocyte–macrophage colony-sti- 46 Brossart P, Bevan MJ. Presentation of exogenous protein antigens mulating factor and other cytokines in the regulation of dendritic on major histocompatibility complex class I molecules by den- cell growth in vitro from early bipotent CD34+ progenitors in dritic cells: pathway of presentation and regulation by cytokines. human bone marrow. J Immunol 1992; 149: 2681–2688. Blood 1997; 90: 1594–1599. 67 Szabolcs P, Moore MAS, Young JW. Expansion of immunostimul- 47 Mitchell DA, Nair SK, Gilboa E. Dendritic cell/macrophage pre- atory dendritic cells among the myeloid progeny of human cursors capture exogenous antigen for MHC class I presentation CD34+ bone marrow precursors cultured with c-kit ligand, gra- by dendritic cells. Eur J Immunol 1998; 28: 1923–1933. nulocyte–macrophage colony-stimulating factor, and TNF-alpha. 48 Shen Z, Reznikoff G, Dranoff G, Rock KL. Cloned dendritic cells J Immunol 1995; 154: 5851–5861. can present exogenous antigens on both MHC class I and class 68 Strunk D, Rappersberger K, Egger C, Strobl H, Kromer E, Elbe A, II molecules. J Immunol 1997; 158: 2723–2730. Maurer D, Stingl G. Generation of human dendritic 49 Huang AY, Golumbek P, Ahmadzadeh M, Jaffee E, Pardoll D, cells/Langerhans cells from circulating CD34+ hematopoietic Levitsky H. Role of bone marrow-derived cells in presenting progenitor cells. Blood 1996; 87: 1292–1302. MHC class I-restricted tumor antigens. Science 1994; 264: 69 Tarte K, Lu ZY, Fiol G, Legouffe E, Rossi JF, Klein B. Generation 961–965. of virtually pure and potentially proliferating dendritic cells from 50 Apostolopoulos V, Osinski C, McKenzie IF. MUC1 cross-reactive non-CD34 apheresis cells from patients with multiple myeloma. Gal alpha(1,3)Gal antibodies in humans switch immune Blood 1997; 90: 3482–3495. responses from cellular to humoral. Nature Med 1998; 4: 315– 70 Falo LD Jr, Colarusso LJ, Benacerraf B, Rock KL. Serum proteases 320. alter the antigenicity of peptides presented by class I major histo- 51 Bellone M, lezzi G, Rovere P, Galati G, Ronchetti A, Protti MP, compatibility complex molecules. Proc Natl Acad Sci USA 1992; Davoust J, Rugarli C, Manfredi AA. Processing of engulfed apop- 89: 8347–8350. Review K Tarte and B Klein 662 71 Luft T, Pang KC, Thomas E, Bradley CJ, Savoia H, Trapani J, with interleukin-4 and granulocyte/macrophage colony-stimulat- Cebon J. A serum-free culture model for studying the differen- ing factor. Eur J Immunol 1997; 27: 756–762. tiation of human dendritic cells from adult CD34+ progenitor 90 Allavena P, Piemonti L, Longoni D, Bernasconi S, Stoppacciaro cells. Exp Hematol 1998; 26: 489–500. A, Ruco L, Mantovani A. IL-10 prevents the differentiation of 72 Abdel-Wahab Z, DeMatos P, Hester D, Dong XD, Seigler HF. monocytes to dendritic cells but promotes their maturation to Human dendritic cells, pulsed with either melanoma tumor cell macrophages. Eur J Immunol 1998; 28: 359–369. lysates or the gp100 peptide(280–288), induce pairs of T-cell cul- 91 Qin Z, Noffz G, Mohaupt M, Blankenstein T. Interleukin-10 pre- tures with similar phenotype and lytic activity. Cell Immunol vents dendritic cell accumulation and vaccination with granulo- 1998; 186: 63–74. cyte–macrophage colony-stimulating factor gene-modified tumor 73 Reddy A, Sapp M, Feldman M, Subklewe M, Bhardwaj N. A cells. J Immunol 1997; 159: 770–776. monocyte conditioned medium is more effective than defined 92 Boon T, Van der Bruggen P. Human tumor antigens recognized cytokines in mediating the terminal maturation of human den- by T lymphocytes. J Exp Med 1996; 183: 725–729. dritic cells. Blood 1997; 90: 3640–3646. 93 Celluzzi CM, Mayordomo JI, Storkus WJ, Lotze MT, Falo LD. 74 Jonuleit H, Kuhn U, Muller G, Steinbrink K, Paragnik L, Schmitt Peptide-pulsed dendritic cells induce antigen-specific, CTL- E, Knop J, Enk AH. Pro-inflammatory cytokines and prosta- mediated protective tumor immunity. J Exp Med 1996; 183: glandins induce maturation of potent immunostimulatory den- 283–287. dritic cells under fetal calf serum-free conditions. Eur J Immunol 94 Tsai V, Southwood S, Sidney J, Sakaguchi K, Kawakami Y, 1997; 27: 3135–3142. Appella E, Sette A, Celis E. Identification of subdominant CTL 75 Siena S, Dinicola M, Bregni M, Mortarini R, Anichini A, Lom- epitopes of the GP100 melanoma-associated tumor antigen by bardi L, Ravagnani F, Parmiani G, Gianni AM. Massive ex vivo primary in vitro immunization with peptide-pulsed dendritic generation of functional dendritic cells from mobilized CD34(+) cells. J Immunol 1997; 158: 1796–1802. blood progenitors for anticancer therapy. Exp Hematol 1995; 23: 95 Peshwa MV, Shi JD, Ruegg C, Laus R, van Schooten WC. Induc- + 1463–1471. tion of prostate tumor-specific CD8 cytotoxic T-lymphocytes in 76 Soruri A, Fayyazi A, Gieseler R, Schlott T, Runger TM, Neumann vitro using antigen-presenting cells pulsed with prostatic acid C, Peters JH. Specific autologous anti-melanoma T cell response phosphatase peptide. Prostate 1998; 36: 129–138. in vitro using monocyte-derived dendritic cells. Immunobiology 96 Mayordomo JI, Loftus DJ, Sakamoto H, De Cesare CM, Appas- 1998; 198: 527–538. amy PM, Lotze MT, Storkus WJ, Appella E, Deleo AB. Therapy of 77 Rieser C, Bock G, Klocker H, Bartsch G, Thurnher M. Prostaglan- murine tumors with p53 wild-type and mutant sequence peptide- din E2 and tumor necrosis factor alpha cooperate to activate based vaccines. J Exp Med 1996; 183: 1357–1365. human dendritic cells: synergistic activation of interleukin 12 97 Bakker AB, Marland G, de Boer AJ, Huijbens RJ, Danen EH, production. J Exp Med 1997; 186: 1603–1608. Adema GJ, Figdor CG. Generation of antimelanoma cytotoxic T 78 Anton D, Dabadghao S, Palucka K, Holm G, Yi Q. Generation lymphocytes from healthy donors after presentation of mela- of dendritic cells from peripheral blood adherent cells in medium noma-associated antigen-derived epitopes by dendritic cells in with human serum. Scand J Immunol 1998; 47: 116–121. vitro. Cancer Res 1995; 55: 5330–5334. 79 Ambe K, Mori M, Enjoji M. S-100 protein-positive dendritic cells 98 Nieda M, Nicol A, Kikuchi A, Kashiwase K, Taylor K, Suzuki K, Tadokoro K, Juji T. Dendritic cells stimulate the expansion of bcr- in colorectal adenocarcinomas. Distribution and relation to the + clinical prognosis. Cancer 1989; 63: 496–503. abl specific CD8 T cells with cytotoxic activity against leukemic 80 Tsujitani S, Kakeji Y, Watanabe A, Kohnoe S, Maehara Y, Sugi- cells from patients with chronic myeloid leukemia. Blood 1998; machi K. Infiltration of dendritic cells in relation to tumor 91: 977–983. invasion and lymph node metastasis in human gastric cancer. 99 Paglia P, Chiodoni C, Rodolfo M, Colombo MP. Murine dendritic Cancer 1990; 66: 2012–2016. cells loaded in vitro with soluble protein prime cytotoxic T lymphocytes against tumor antigen in vivo. J Exp Med 1996; 183: 81 Schroder S, Schwarz W, Rehpenning W, Loning T, Bocker W. 317–322. Dendritic/Langerhans cells and prognosis in patients with papil- 100 Dietz AB, Vuk-Pavlovic S. High efficiency adenovirus-mediated lary thyroid carcinomas. Immunocytochemical study of 106 thy- gene transfer to human dendritic cells. Blood 1998; 91: 392–398. roid neoplasms correlated to follow-up data. Am J Clin Pathol 101 Reeves ME, Royal RE, Lam JS, Rosenberg SA, Hwu P. Retroviral 1988; 89: 295–300. transduction of human dendritic cells with a tumor-associated 82 Bigotti G, Coli A, Castagnola D. Distribution of Langerhans cells antigen gene. Cancer Res 1996; 56: 5672–5677. and HLA class II molecules in prostatic carcinomas of different 102 Henderson RA, Nimgaonkar MT, Watkins SC, Robbins PD, Ball histopathological grade. Prostate 1991; 19: 73–87. ED, Finn OJ. Human dendritic cells genetically engineered to 83 Gu ZJ, Costes V, Lu ZY, Zhang XG, Pitard V, Moreau JF, Bataille express high levels of the human epithelial tumor antigen mucin R, Wijdenes J, Rossi JF, Klein B. Interleukin-10 is a growth factor (MUC-1). Cancer Res 1996; 56: 3763–3770. for human myeloma cells by induction of an oncostatin M 103 Brossart P, Goldrath AW, Butz EA, Martin S, Bevan MJ. Virus- autocrine loop. Blood 1996; 88: 3972–3986. mediated delivery of antigenic epitopes into dendritic cells as a 84 Pisa P, Halapi E, Pisa EK, Gerdin E, Hising C, Bucht A, Gerdin means to induce CTL. J Immunol 1997; 158: 3270–3276. B, Kiessling R. Selective expression of interleukin 10, interferon 104 Song W, Kong HL, Carpenter H, Torii H, Granstein R, Rafii S, gamma, and granulocyte–macrophage colony-stimulating factor Moore MA, Crystal RG. Dendritic cells genetically modified with in ovarian cancer biopsies. Proc Natl Acad Sci USA 1992; 89: an adenovirus vector encoding the cDNA for a model antigen 7708–7712. induce protective and therapeutic antitumor immunity. J Exp 85 Gastl GA, Abrams JS, Nanus DM, Oosterkamp R, Silver J, Liu F, Med 1997; 186: 1247–1256. Chen M, Albino AP, Bander NH. Interleukin-10 production by 105 Specht JM, Wang G, Do MT, Lam JS, Royal RE, Reeves ME, human carcinoma cell lines and its relationship to interleukin-6 Rosenberg SA, Hwu P. Dendritic cells retrovirally transduced expression. Int J Cancer 1993; 55: 96–101. with a model antigen gene are therapeutically effective against 86 Kruger-Krasagakes S, Krasagakis K, Garbe C, Schmitt E, Huls C, established pulmonary metastases. J Exp Med 1997; 186: Blankenstein T, Diamantstein T. Expression of interleukin 10 in 1213–1221. human melanoma. Br J Cancer 1994; 70: 1182–1185. 106 Gong J, Chen L, Chen D, Kashiwaba M, Manome Y, Tanaka T, 87 Smith DR, Kunkel SL, Burdick MD, Wilke CA, Orringer MB, Kufe D. Induction of antigen-specific antitumor immunity with Whyte RI, Strieter RM. Production of interleukin-10 by human adenovirus-transduced dendritic cells. Gene Therapy 1997; 4: bronchogenic carcinoma. Am J Pathol 1994; 145: 18–25. 1023–1028. 88 Khatri VP, Caligiuri MA. A review of the association between 107 Ikeda H, Lethe B, Lehmann F, van Baren N, Baurain JF, de Smet interleukin-10 and human B-cell malignancies. Cancer Immunol C, Chambost H, Vitale M, Moretta A, Boon T, Coulie PG. Charac- Immunother 1998; 46: 239–244. terization of an antigen that is recognized on a melanoma show- 89 Buelens C, Verhasselt V, De Groote D, Thielemans K, Goldman ing partial HLA loss by CTL expressing an NK inhibitory receptor. M, Willems F. Interleukin-10 prevents the generation of dendritic Immunity 1997; 6: 199–208. cells from human peripheral blood mononuclear cells cultured 108 Storkus WJ, Zeh HJ, Salter RD, Lotze MT. Identification of T-cell Review K Tarte and B Klein 663 epitopes: rapid isolation of class I-presented peptides from viable established murine tumors using a novel cell-free vaccine: den- cells by mild acid elution. J Immunother 1993; 14: 94–103. dritic cell-derived exosomes. Nature Med 1998; 4: 594–600. 109 Zitvogel L, Mayordomo JI, Tjandrawan T, Deleo AB, Clarke MR, 119 Wong C, Morse M, Nair SK. Induction of primary, human anti- Lotze MT, Storkus WJ. Therapy of murine tumors with tumor pep- gen-specific cytotoxic T lymphocytes in vitro using dendritic tide-pulsed dendritic cells: dependence on T cells, B7 constimu- cells pulsed with peptides. J Immunother 1998; 21: 32–40. lation, and T helper cell 1-associated cytokines. J Exp Med 1996; 120 Mukherji B, Chakraborty NG, Yamasaki S, Okino T, Yamase H, 183: 87–97. Sporn JR, Kurtzman SK, Ergin MT, Ozols J, Meehan J. Induction 110 Ashley DM, Faiola B, Nair S, Hale LP, Bigner DD, Gilboa E. Bone of antigen-specific cytolytic T cells in situ in human melanoma marrow-generated dendritic cells pulsed with tumor extracts or by immunization with synthetic peptide-pulsed autologous anti- tumor RNA induce antitumor immunity against central nervous gen presenting cells. Proc Natl Acad Sci USA 1995; 92: 8078– system tumors. J Exp Med 1997; 186: 1177–1182. 8082. 111 Gilboa E, Nair SK, Lyerly HK. Immunotherapy of cancer with 121 Nestle FO, Alijagic S, Gilliet M, Sun Y, Grabbe S, Dummer R, dendritic-cell-based vaccines. Cancer Immunol Immunother Burg G, Schadendorf D. Vaccination of melanoma patients with 1998; 46: 82–87. peptide- or tumor lysate-pulsed dendritic cells. Nature Med 112 Boczkowski D, Nair SK, Snyder D, Gilboa E. Dendritic cells 1998; 4: 328–332. pulsed with RNA are potent antigen-presenting cells in vitro and 122 Murphy G, Tjoa B, Ragde H, Kenny G, Boynton A. Phase I clini- in vivo. J Exp Med 1996; 184: 465–472. cal trial: T-cell therapy for prostate cancer using autologous den- 113 Nair SK, Boczkowski D, Morse M, Cumming RI, Lyerly HK, Gil- dritic cells pulsed with HLA-A0201-specific peptides from pros- boa E. Induction of primary carcinoembryonic antigen (CEA)-spe- tate-specific membrane antigen. Prostate 1996; 29: 371–380. cific cytotoxic T lymphocytes in vitro using human dendritic cells 123 Tjoa BA, Simmons SJ, Bowes VA, Ragde H, Rogers M, Elgamal transfected with RNA. Nat Biotechnol 1998; 16: 364–369. A, Kenny GM, Cobb OE, Ireton RC, Troychak MJ, Salgaller ML, Boynton AL, Murphy GP. Evaluation of phase I/II clinical trials in 114 Lespagnard L, Mettens P, Verheyden AM, Tasiaux N, Thielemans prostate cancer with dendritic cells and PSMA peptides. Prostate K, van Meirvenne S, Geldhof A, De Baetselier P, Urbain J, Leo 1998; 36: 39–44. O, Moser M. Dendritic cells fused with mastocytoma cells elicit 124 Hsu FJ, Benike C, Fagnoni F, Liles TM, Czerwinski D, Taidi B, therapeutic antitumor immunity. Int J Cancer 1998; 76: 250–258. Engleman EG, Levy R. Vaccination of patients with B-cell lym- 115 Gong J, Chen D, Kashiwaba M, Kufe D. Induction of antitumor phoma using autologous antigen-pulsed dendritic cells. Nature activity by immunization with fusions of dendritic and carcinoma Med 1996; 2: 52–58. cells. Nature Med 1997; 3: 558–561. 125 Holtl L, Rieser C, Papesh C, Ramoner R, Bartsch G, Thurnher M. 116 Celluzzi CM, Falo LD Jr. Physical interaction between dendritic CD83+ blood dendritic cells as a vaccine for immunotherapy of cells and tumor cells results in an immunogen that induces pro- metastatic renal-cell cancer (letter). Lancet 1998; 352: 1358. tective and therapeutic tumor rejection. J Immunol 1998; 160: 126 Vescio RA, Cao J, Hong CH, Lee JC, Wu CH, Der Danielian M, 3081–3085. Wu V, Newman R, Lichtenstein AK, Berenson JR. Myeloma lg 117 Choudhury A, Gajewski JL, Liang JC, Popat U, Claxton DF, heavy chain V region sequences reveal prior antigenic selection Kliche KO, Andreeff M, Champlin RE. Use of leukemic dendritic and marked somatic mutation but no intraclonal diversity. J cells for the generation of antileukemic cellular cytotoxicity Immunol 1995; 155: 2487–2497. against Philadelphia chromosome-positive chronic myelogenous 127 Wen YJ, Ling M, Bailey-Wood R, Lim SH. Idiotypic protein- leukemia. Blood 1997; 89: 1133–1142. pulsed adherent peripheral blood mononuclear cell-derived den- 118 Zitvogel L, Regnault A, Lozier A, Wolfers J, Flament C, Tenza D, dritic cells prime immune system in multiple myeloma. Clin Can- Ricciardi-Castagnoli P, Raposo G, Amigorena S. Eradication of cer Res 1998; 4: 957–962.