Cancer Therapy Vol 1, 299-314, 2003.

Dendritic cell based vaccines for immunotherapy of cancer

Review Article

 

Asim Saha, Sunil K. Chatterjee, Kartik Mohanty, Kenneth A. Foon1, Malaya Bhattacharya-Chatterjee*

Department of Internal Medicine and the Barrett Cancer Center, University of Cincinnati, Ohio 45267

1University of Pittsburgh Cancer Institute, Pittsburgh, PA 15232

__________________________________________________________________________________

*Correspondence: Malaya Bhattacharya-Chatterjee, PhD, The Vontz Center for Molecular Studies, Room 1316, University of Cincinnati, 3125 Eden Avenue, Cincinnati, OH 45267-0509. Tel: (513) 558-0425; Fax: (513) 558-1096; e-mail: malaya.chatterjee@uc.edu

Key Words: vaccines, dendritic cells, cancer, immunotherapy

Abbreviations: Dendritic cells (DCs); antigen presenting cells (APCs); ELISA method (ELISPOT); Carcinoembryonic antigen (CEA)

 

Received: 16 October 2003; Revised: 17 December 2003

Accepted: 18 December 2003; electronically published: December 2003

 

Summary

Researchers and clinicians have tried for decades to use specific and non-specific immunotherapy for the fight against cancer. Initial attempts were based on soluble immune mediators such as antibodies or cytotoxic proteins for the therapy of malignancies. Major improvements in our understanding of the induction and regulation of cellular immunity have now made it possible to generate effector cells in cancer patients that can specifically recognize and destroy malignant cells. Human tumors express a number of protein antigens that can be recognized by T cells, thus providing potential targets for cancer immunotherapy. Tumor antigens have to be presented to T cells in order to activate them and drive them into clonal expansion. This is done by antigen presenting cells (APCs). Dendritic cells (DCs) are professional APCs, which have an extraordinary capacity to stimulate na•ve T cells and initiate primary immune responses. This established function of DCs has now offered the hope to apply DC-based immunotherapy for cancers. Pilot clinical trials of DC vaccination have established the safety and feasibility of this approach and have produced encouraging evidence of therapeutic efficacy. Importantly, significant advances in our understanding of the DC biology can be used to support the design of new vaccines in order to elicit effective cellular immune responses for the treatment of cancer. In this review, recent findings of DC immunotherapy for a variety of tumors including colon cancer have been discussed. Also, the development of DC-based vaccines in preclinical models of colorectal cancer as a prelude to clinical trials has been summarized.

 


I. Introduction

Colorectal cancer is one of the most common malignancies in men and women representing 10% of all cancer deaths in the United States with ~ 130,000 new cases diagnosed annually (Landis et al, 1998). Progression of the disease occurs through invasion of the colonic wall, involvement of regional lymph nodes, and distant metastasis. At the time of diagnosis, almost 50% of the patients have tumors invading through the bowel wall with spread to regional lymph nodes; in addition, 10% of the patients present with synchronous metastatic cancer to the liver. There have been important advances in our understanding of the biology and genetics of this disease, and if diagnosed early, colorectal cancer is curable. Surgery is associated with a 90% five-year survival rate in patients with tumors involving only the mucosa or submucosa. However, for majority of the patients having metastasis of the disease in local lymph nodes and adjacent organs, survival is substantially worse (Greenlee et al, 2000). Systemic chemotherapy with 5-flurouracil-based regimens or newer agents such as irinotecan has improved survival of these patients with high-risk disease (Moore and Haller, 1999). Despite this, only 60% of all patients diagnosed with colorectal cancer survive more than ten years. At present there is no therapeutic regime capable of curing unresectable metastatic disease. Clearly, more effective treatments are necessary for this disease.

Vaccines against infectious diseases are the success story of immunology. Smallpox has been eradicated and vaccination strategies have saved countless people from tetanus, polio, measles and hepatitis. Consequently, there is a hope for the generation of effective cancer vaccine(s). However, to date, human anti-tumor vaccination has not delivered on its promises. Reasons for failure include tumor immune escape mechanisms, limited availability of tumor specific antigens, as well as failure to deliver tumor antigens in the right immunological context. Progress in immunology and molecular biology has provided technologies to detect an ever increasing choice of new tumor specific antigens. One of the most important issues is to deliver these tumor antigens in an effective way to induce immune responses in cancer patients. However, recent insights into the role of DCs as the pivotal APCs that initiate immune response may provide the basis for generating more effective antitumor immune responses in patients. Our understanding of the DC biology has opened new ways for the application of these cells for immunotherapy of cancer.

Immunotherapy is an attractive approach to cancer therapy. The aim of immunotherapy is to induce or increase the ability of the host to mount antitumor immune responses in vivo. Convincing evidence now exists that the effector cells of the immune system (T, B, and NK cells), when appropriately activated, are able to lyse tumor cells through specific recognition of tumor associated antigens (TAAs) (Rosenberg, 1996; Finn and Lotze, 1998). Although a variety of both humoral and cellular antitumor immune responses have been documented, T cells, and in particular CD8+ cytotoxic T lymphocytes, are likely to play an important role in antitumor immunity (Maeurer and Lotze, 1997). Identification of TAAs together with a better understanding of the mechanisms involved in the immune response against cancer, have given investigators tools to manipulate the immune system to induce an efficient immune response in the tumor bearing host (Pardoll, 2000; Van Gool et al, 2000; Borrello and Sotomayor, 2002; Drake and Pardoll, 2002).

 

II. What is a dendritic cell?

DCs were originally discovered as antigen presenting cells critical for the induction of primary T cell dependent immune responses (Steinman, 1991). DCs represent only 0.5% of blood leukocytes. Like other cell types within the immune system, they arise from a common CD34+ progenitor in the bone marrow whose expansion and differentiation is influenced by a variety of cytokine growth factors including stem cell factor, fetal liver tyrosine kinase-3 (Flt-3) ligand, IL-3, granulocyte/macrophage colony stimulating factor (GM-CSF), TNF-a and TGF-b (Shortman and Caux, 1997; Pulendran et al, 2001). Mature DCs have a distinct morphology characterized by the presence of numerous membrane processes that can take the form of dendrites, pseudopods, or veils. Morphologic features of DCs include high concentrations of intracellular structures related to antigen processing such as endosomes, lysosomes, and the Birbeck granules of Langerhans cells of the epidermis.

DCs are also characterized by abundant expression of molecules used for their specialized interactions with T cells. These include the antigen-presentation molecules CD1 and the class I and class II MHC proteins, the adhesion molecules CD11a (LFA-1a), CD11b, CD11c, CD50 (ICAM-2), CD54 (ICAM-1), CD58 (LFA-3), and CD102 (ICAM-3), although all of these markers can also be found on monocytes and macrophages (Hart and Prickett, 1993). Costimulatory molecules such as CD80 (B7.1) and CD86 (B7.2), and molecules regulating costimulation such as CD40 are also expressed on mature myeloid DCs. DCs do not express surface differentiation antigens found on B cells (CD19, CD20), T cells (CD3), monocytes (CD14), and natural killer cells (CD56). Several antibodies have been described that preferentially but not exclusively stain mature DCs. Antibodies reactive against human DCs include anti-CD83 and CMRF-44 (Zhou   et al, 1992; Hock et al, 1994). Antibodies directed at mouse DCs include 33D1, N418 (anti-CD11c), and DEC-205 (Kraal et al, 1986; Metlay et al, 1990).

DCs originate from the bone marrow and their precursors home via the bloodstream to almost all organs, where they can be found as sentinels in an immature state with high endocytic and phagocytic capacity. The current view is that these immature interstitial DCs are the precursors of the mature interdigitating DCs found in the T cell rich area of secondary lymphoid organs. Upon contact with bacterial DNA, LPS, dsRNA, and inflammatory cytokines such as TNF-a or IL-1 the interstitial DCs change their phenotype and function and migrate to the germinal centers of regional lymph nodes, where they present antigens captured in the periphery to resting or na•ve T cells and induce antigen-specific T cell responses.

With DC activation and migration from the tissue, antigen uptake activity and the associated antigen receptors are down regulated, resulting in a switch in APC function from antigen uptake to antigen presentation (Hart and McKenzie, 1988). DCs are capable of processing antigen via classical pathways: endogenous antigens via the proteasome into the MHC class I compartment, and exogenous antigens via endocytic lysosomes into the MHC class II compartment (Lanzavecchia, 1996). DCs also process alternative pathways of antigen processing and can route exogenous antigen into the MHC class I pathway through a mechanism known as cross-priming (Norbury et al, 1997). DCs may also utilize molecular chaperones such as heat shock proteins (hsp96) to deliver antigens via the class I pathway (Arnold-Schild et al, 1999).

The family of human DC displays considerable heterogeneity and plasticity at the level of phenotype and function (Banchereau et al, 2000). DC may be derived from two potential lineages: myeloid or lymphoid. Myeloid progenitors give rise to two main precursors: CD14+ CD11c+ precursors and CD14- CD11c+ precursors (Caux et al, 1996). CD14+ CD11c+ cells differentiate in the presence of IL-4 and GM-CSF into interstitial DCs that correspond to dermal DCs in vivo (Grassi et al, 1998). In the presence of M-CSF, CD14+ CD11c+ precursor cells acquire characteristics of macrophages. CD11c+ DCs may be also reverted to macrophages under the same conditions. CD14- CD11c+ precursors yield DC of the Langerhans cell type in response to GM-CSF, IL-4 and TGF-b. Immature dermal or Langerhans type DC correspond to tissue resident sentinels in peripheral tissue sites. Upon encounter of T cell derived signals such as CD40L or microbial products such as LPS, they might be further driven along their differentiation pathway to mature DC. Mature DC resides in T cell areas of lymph nodes and marginal zone of spleen with stable phenotype and function. A third major subset of DC are CD14- CD11c- IL-3Ra+ DC precursors (Grouard et al, 1997). These cells depend on IL-3 as survival factor and may be matured through CD40 signaling. Further surface markers include CD45RA and ILT-3 (Cella et al, 1999). They display low phagocytic activity and are the major source of IFN-a production in response to viral infection (Siegal et al, 1999).

 

III. What is the role of dendritic cell for the induction of tumor immunity?

Effector mechanisms against endogenous tumors include both cellular and humoral immunity. The majority of experimental systems clearly demonstrate that tumor immunity is largely provided by CD4+ T lymphocytes (Hung et al, 1998; Dembic et al, 2000; Qin and Blankenstein, 2000), CD8+ T lymphocytes (Celluzzi et al, 1996; Jenne et al, 2000; Terheyden et al, 2000) or NK cells (Fernandez et al, 1999). T cells, and possibly also NK cells, however, require activation by APCs, and in this context DCs are pivotal for this process. To activate na•ve T cells, DCs take up, process and present antigen by their MHC molecules (Banchereau and Steinman, 1998). In addition, T cell activation requires engagement of co-stimulatory receptors on the T cell, adequate types and concentrations of T cell activating cytokines and T cell attracting chemokines, and maintenance of the activation signal over a sufficient period of time. Currently, the list of family of co-stimulatory molecules is increasing dramatically, and it appears likely that DCs can minutely control the outcome of immune activation by means of differential surface receptor expression (Coyle and Gutierrez-Ramos, 2001), and that T cells in turn signal back to modulate the function of DCs (Bennett et al, 1998; Ridge et al, 1998; Schoenberger et al, 1998). Activation of NK cells and of macrophages is less well understood, but an interaction between DCs and these cell types has been demonstrated (Fernandez et al, 1999). Apart from generating a powerful antitumor immune response, DCs may also play an active role in the eradication of tumors themselves, since DCs have been shown to kill tumor cells via expression of death receptor ligands (Fanger et al, 1999), and recent data suggest that DCs activated by pro-inflammatory cytokines or LPS can directly inhibit the growth of tumor cell lines (Chapoval et al, 2000). Thus DCs are at the very center of a developing tumor-specific immune response, and are involved both in the initiation of tumor-specific immunity and the generation of immune effector functions.

Promising results obtained in a variety of murine tumor models using DCs presenting tumor antigens as well as the identification of a growing number of T cell epitopes presented by human malignant cells prompted the rationale for evaluating the efficacy of DC-based vaccines in clinical studies.

 

 

IV. Generation of dendritic cells for cancer immunotherapy

Clinical trials of DC vaccination have been made possible by the development of methods for obtaining large numbers of human DCs. Three general approaches have been exploited for use in clinical trials (Table 1). Myeloid DCs can be directly purified from blood by density-gradient centrifugation procedures (Hsu et al, 1996). The systemic administration of   Flt-3 ligand or G-CSF increases blood DC numbers several fold (Maraskovsky et al, 2000; Pulendran et al, 2000). Alternatively, DCs can be prepared from CD14+ blood monocytes by in vitro culture with GM-CSF and IL-4 for 5-7 days, and differentiation with maturation stimuli (Bender et al, 1996; Romani et al, 1996). Maturation is essential to prevent reversion to monocytes. Autologous monocyte conditioned medium (MCM), CD40L or a cocktail of TNF-a, IL-1b, IL-6 and PGE-2 are available for the maturation of DCs (Reddy et al, 1997; Thurner et al, 1999). CD34+ hematopoietic progenitor cells obtained from bone marrow, umbilical cord blood or peripheral blood following treatment with GM-CSF or G-CSF are also source of DC precursors. Following culture in GM-CSF and TNF-a (Caux et al, 1997), and stem cell factor or Flt-3 ligand, a mixed population of immature DCs with characteristics of both Langerhans cells and interstitial DCs has been obtained. Comparative studies will be required to establish differences between these various sources of DCs.

 

V. Critical parameters for optimal DC vaccination

Apart from choosing the right source of DC, critical issues for successful vaccination involve choice of antigen, antigen loading, route and schedule of administration, as well as immuno-monitoring.

The choice of DC is likely to depend on the type of antigen used. The cellular machinery required for processing antigen differs according to whether it is delivered as a peptide, protein or genetic vaccine. Immature DCs, which are actively endocytic and can internalize exogenous antigens efficiently, may be most suitable for the delivery of protein or complex antigens that require processing by the DC. In contrast, mature DCs, with higher expression of MHC molecules, may be more suitable for peptide-based protocols. Strategies to enhance DC function genetically to improve vaccine delivery and the subsequent induction of a powerful immune response are also being evaluated. These include genetic manipulation of DC to express cytokines or immuno-stimulatory molecules that can potentiate DC-T-cell interactions (Philip et al, 1998).

Wide range of antigenic preparations are available for loading of DC (Table 2). Peptide antigens are well defined antigenic epitopes binding to a defined set of MHC molecules, and easily accessible for immuno-monitoring of a peptide specific T cell response. However, peptide approaches are limited by the requirement of analysis of the MHC background of patients and the

Table 1. DC types available for clinical studies


Peripheral blood DC populations

Directly isolated from blood or expanded in vivo by growth factors, such as Flt-3 ligand or G-CSF

 

DCs derived from CD14+ monocytes

Monocytes enriched by anti-CD14 immunomagnetic beads

or by plastic adherence

 

DCs derived from CD34+ hematopoietic progenitors

From bone marrow, peripheral blood or cord blood

 


knowledge of the sequence of the relevant peptide epitope. The use of whole protein antigens, DNA, RNA or recombinant viruses encoding the antigen of choice allows host HLA molecules to select the appropriate peptide epitope for presentation as peptide-MHC complex on the cell surface. This approach does not require analysis of MHC molecules, although we have to be aware of the fact that spectrum of epitopes seen by effector T cells might be restricted, since certain peptides are not presented by DC due to incomplete processing at the level of the proteasome (Morel et al, 2000). Recently, other approaches have been applied to use the entire antigenic content of a tumor cell for vaccination to present as many tumor antigens as possible to the immune system and minimize the occurrence of immune escape variants. This might be achieved by either pulsing DCs with whole tumor cell lysate (Ashley et al, 1997), tumor derived RNA (Nair et al, 1998), DNA (Philip et al, 1998) or fusion of tumor cells and DCs (Hart and Colaco, 1997). This technique does not require the definition of the TAA or MHC haplotype of the patients and has the potential for broad clinical application. The limitation of this approach is the availability of tissue serving as a source of tumor lysate or tumor derived RNA. A major disadvantage in using whole tumor in the form of lysates, RNA or DC-tumor fusions is that monitoring effector cells functions in vitro and in vivo is difficult to achieve.

After pulsing with tumor antigen, DCs need to be administered in an effective way to the cancer patients. Subcutaneous, intradermal, intravenous and intranodal approaches to deliver DCs have been evaluated clinically. The intranodal approach (Nestle et al, 1998) bypasses the requirement for vaccine-loaded DCs to migrate to lymphoid tissue and simply relies on their capacity to express effective T cell stimulatory capacity. The intravenous route results in the accumulation of DCs to lung, liver, spleen and bone marrow, but not the lymph nodes or tumor sites (Morse et al, 1999). In contrast, studies using intradermal injection of monocyte derived DCs have demonstrated direct migration of DCs to the draining lymph nodes. However, these particular studies used immature DCs, and only ~1% of DCs migrated to the regional lymph node, the majority remained at the injection site (Thomas et al, 1999). In contrast, monocyte-derived DCs matured in vitro have shown impressive immune responses (Schuler-Thurner et al, 2000). It is unclear whether this effect was due to efficient migration and antigen-presentation induced by in vitro activation of DCs.

Detection of an antigen specific immune response is an important surrogate marker to control for an effective vaccination strategy even though correlation with clinical response will be the most important issue. The classic way to detect CTL activity is measurement of lytic activity against 51Cr labeled target cells. Since precursor frequencies are low, in vitro re-stimulations are often necessary in order to reach a CTL frequency detectable by cytotoxicity assays. Although these techniques are time consuming, still this method is the gold standard since it measures lytic activity of effector cells.


 

Table 2. Delivery of antigens by dendritic cells

Known tumor associated antigens:

Synthetic or eluted peptides

Recombinant or purified protein

Non-peptide antigens, such as carbohydrates (e.g. MUC-1) or glycolipids (e.g. GM2)

Transfection with cDNA or RNA encoding known tumor-associated antigen

Recombinant viruses (adenoviruses, vaccinia, or retroviruses)

 

Approaches when antigens are unknown:

Differentiation of DCs from malignant cells (acute myelogenous leukemia, chronic myeloid leukemia)

Tumor-DC fusions

DC-derived exosomes

Tumor RNA

Apoptotic or necrotic tumor cells

Tumor lysates


These tests quantify antigen-specific T cells in the blood. It is unclear whether the blood is the best place to look for evidence of emerging immunity against vaccination, or whether the tumor site or draining lymph nodes may be more appropriate. Recently introduced methods rely on measurement of release of cytokines by CTL after contact with antigen (Romero et al, 1998). Cytokines may be measured by an ELISA method (ELISPOT) or quantified by intracellular cytokine staining and detection by flow cytometry. Detection of cytokine release does not necessarily correlate with the cytolytic activity of a given cell. Peptide-MHC tetrameric complex is another tool for detection of an antigen specific immune response. However, this technique may not be able to detect low or intermediate affinity T cells, which are important in the context of vaccination against self-antigens. An additional method is delayed type hypersensitivity (DTH) testing for peptide specific immune responses. Peptide DTH testing was demonstrated in humans (Nestle et al, 1998) and this technique is easy to perform even though objective read out might be a problem and is observer dependent. DTH reactions are important since it measures induction of an antigen specific immune response. It is therefore crucial to develop appropriate immunological assays that may predict clinical responses more closely.

 


 

Table 3. Published clinical trials conducted with DC-based vaccines against cancer

 

Disease type

Antigen

  DC type                        

   Route                    

Investigator

Melanoma

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lymphoma 

 

 

 

 

 

 

Myeloma

 

a) Melan-A,

gp100, tyrosinase

b) MAGE-1, MAGE-3, Melan-A, gp100,

tyrosinase   

c) MAGE-3

 

 

d) MAGE-1,

MAGE-3,

Melan-A, gp100,

tyrosinase

e) MART-1, gp100

 

f) Melan-A,

MAGE-3, gp100,

tyrosinase

g) Tulys

 

h) tumor cell-DC fusion

i) Mart-127-35

peptide

j) tumor cells

 

k) acid-eluted

peptides

 

a) idiotype

 

b) idiotype

 

c) Tulys

 

 

a) idiotype

 

b) idiotype

 

c) idiotype

 

 

  mDC

 

  mDC

 

 

 

  mDC

 

 

  CD34-DC

 

 

 

  mDC

 

  CD34-DC

 

 

  mDC

 

  mDC

 

  mDC

 

  mDC

 

  mDC

 

 

  PBDC

 

  PBDC

 

  mDC

 

 

  mDC

 

  PBDC

 

  CD34-DC

 

   i.v.

 

   intranodal

 

 

 

   s.c., i.v.   

 

  

   i.v.     

 

 

 

   i.v. 

 

   s.c.

  

  

   i.d.

 

   s.c.

 

   i.v., or i.d.

 

   i.d.

 

   i.d.

  

 

   i.v.,

   s.c. boost.

   i.v.

   s.c. boost  

   intranodal

  

  

   i.v. 

 

   i.v.

   s.c. boost

   s.c.,

   s.c. boost

Lotze MT et al

(1997)

Nestle FO et al

(1998)

 

 

Schuler-  

Thurner B et al

(2000)

Mackensen A

et al (2000)

 

 

Panelli MC

et al (2000)

Banchereau J

et al (2001)

 

Chang AE

et al (2002)

Krause SW

et al (2002)

Butterfield LH

et al (2003)

OŐRourke MGE

et al (2003)

Smithers M

et al (2003)

 

Hsu FJ et al

(1996)

Timmerman JM

et al (2002)

Maier T et al

(2003)

 

Lim SH et al

(1999)

Reichardt VL

et al (1999)

Titzer S et al

(2000)

 

Prostate

 

 

 

 

 

 

 

 

 

 

RCC

 

 

 

 

 

 

 

 

 

 

 

 

Liver

 

Various

 

Bladder

 

 

CEA-expressing

malignant

Lung, colon

 

 

Stomach,

esophagus, colon                                         

Colon

 

a) PSMA

 

b) Fusion protein

(PAP+GM-CSF)

c) Fusion protein

(PAP+GM-CSF)

d) PSM-P1,

PSM-P2

e) rmPAP

 

 

a) Tulys

 

b) tumor cell-DC

fusion

c) Tulys

 

 

d) Tulys

 

e) tumor cell-DC

fusion

f) tumor RNA

 

Tulys

 

Tulys

 

MAGE-3

 

 

CAP-1

(CEA peptide)

610D

(CEA peptide)

 

MAGE-3

 

CEA mRNA

 

 mDC

        

 PBDC

 

 PBDC

 

 mDC

 

 PBDC

 

 

 Allogeneic

 mDC

 Allogeneic

 mDC

 mDC

 

 

 mDC

 

 Allogeneic

 mDC

 mDC

 

 mDC

 

 mDC

 

 mDC

 

 

 mDC

 

 FL mobilized,

 density  

 purified DC

 mDC

 

 mDC

  

   i.v.

       

   i.v., s.c.

 

   i.v.

 

   i.v.

 

   i.v., i.d., or

   intranodal

 

   i.v.

 

   s.c.,

   s.c. boost

   i.d.

  

  

   s.c.

  

   i.d.

  

   i.v., i.d.

  

   intranodal

 

   i.d.

 

   s.c.

 

  

   i.v., i.d.

 

   i.v.

  

 

   i.v.

 

   i.v., i.d.

 

Tjoa BA et al

(1999)

Burch PA et al

(2000)

Small EJ et al

(2000)

Lodge PA et al

(2000)

Fong L, Brockstedt D et al (2001)

 

Holtl L et al

(1999)

Kugler A et al

(2000)

Oosterwijk-

Wakka JC et al

(2002)

Marten A et al

(2002)

Marten A et al

(2003)

Su Z et al (2003)

 

Iwashita Y et al

(2003)

Geiger J et al

(2001)

Nishiyama T

et al (2001)

 

Morse MA

et al (1999)

Fong L et al

(2001)

 

Sadanaga N

et al (2001)

Morse MA

et al (2003)

__________________________________________________________________________________________________

mDC, monocyte-derived DCs; CD34-DC, DCs derived from CD34-positive progenitors; PBDC, peripheral blood-derived DCs; i.v. intravenous; s.c., subcutaneous; i.d., intradermal; MAGE, melanoma antigen; PSMA, prostate-specific membrane antigen; PAP, prostatic acid phosphatase; PSM-P, prostate-specific membrane antigen-derived peptide; rm, recombinant mouse; RCC, renal-cell carcinoma; Tulys, tumor lysate; FL, Flt-3 ligand.

 


VI. Dendritic cell-based vaccines in clinical trials

Numerous trials are currently ongoing or planned in the very fast moving field of DC immunotherapy trials. We will only discuss published trials in DC vaccination that includes malignant melanoma, non-hodgkin lymphoma, multiple myeloma, prostate cancer, renal-cell carcinoma, liver cancer, pediatric solid tumor, bladder cancer and colorectal cancer (Table 3).

 

A. Melanoma

Many immunologically relevant melanoma antigens including differentiation antigens Melan-A, gp100, and tyrosinase, as well as cancer-testis antigens, such as those of the melanoma antigen (MAGE) family are currently being investigated in DC-based clinical studies.

Lotze and co-workers (Lotze et al, 1997; Lotze et al, 1998) used monocyte-derived DCs to treat HLA-A2 positive patients with metastatic melanoma. DCs were pulsed with peptides derived from Melan-A, gp100, or tyrosinase. Twenty-eight patients received weekly intravenous and subcutaneous infusions of peptide-pulsed DCs for four weeks. Two patients achieved a complete response and 1 patient responded partially, although, one of the complete responders later developed overt rheumatoid arthritis.

Nestle and colleagues (Nestle et al, 1998; Nestle, 2000) used monocyte-derived DCs pulsed with MAGE-1 and MAGE-3 peptides (for patients expressing HLA-A1), Melan-A, gp100 and tyrosinase peptides (for patients expressing HLA-A2), or MAGE-3 and tyrosinase peptides (for patients expressing HLA-B44). Patients with metastatic melanoma received weekly intranodal vaccinations for four weeks with a fifth injection in week 6 and then patients had monthly injections for up to 10 months, depending on clinical response. Vaccinations were well tolerated and 8 of 30 patients had clinical responses, with 3 complete and 5 partial remissions.

Thurner and colleagues (Thurner et al, 1999) used monocyte-derived DCs pulsed with MAGE-3 peptides to treat HLA-A1 positive patients with metastatic melanoma. Five vaccinations (three subcutaneous followed by two intravenous) were given on every 2 weeks. Six of 11 patients had mixed responses. Eight patients showed an increase in MAGE-3-specific CTL responses. A follow-up study (Schuler-Thurner et al, 2000) involving 12 patients with stage IV melanoma used the same procedure as described by Thurner and colleagues. Three to five vaccinations with mature, monocyte derived DCs loaded with HLA-A2 restricted peptides for MAGE-3 and for influenza matrix generated vigorous immune responses as detected by in vitro assays in all eight vaccinated patients. However, no significant clinical responses were observed after final vaccination. Four patients died early in the treatment period due to disease progression. Only 1 patient had stable disease, but disease progressed in all remaining patients.

Mackensen and colleagues (Mackensen et al, 2000) did perform a phase I study in 14 patients with advanced melanoma using CD34 derived DCs matured with TNF-a. DCs were pulsed with MAGE-1 and MAGE-3 peptides (for patients expressing HLA-A1) or Melan-A, gp100 and tyrosinase peptides (for patients expressing HLA-A2). Patients received at least four intravenous vaccinations, biweekly. Patients with stable or responding disease continued to receive vaccination every four weeks until disease progression. The vaccines were tolerated. One patient had a mixed response and six patients had stable disease for 3 to 8 months. Peptide-specific DTH response was observed in 4 patients and expansion of peptide specific CTL response was observed in 1 patient. A similar study was conducted (Banchereau et al, 2001) in 18 HLA-A*0201+ patients with stage IV melanoma using CD34 derived DCs. Patients were immunized with DCs pulsed with peptides derived from four melanoma antigens (MelAgs) Melan-A/MART-1, tyrosinase, MAGE-3, and gp 100 subcutaneously every two weeks for a total of four vaccinations. DC injections were well tolerated except for two patients. DC vaccination resulted enhanced immunity to ³ 1 MelAgs in 16 of 18 patients. Ten of 14 evaluated patients developed DTH to at least one peptide after repeated DC vaccination. The development of T cell response to multiple tumor antigens on peptide-pulsed DCs in this study was associated with a favorable early clinical outcome. Seven of 17 evaluable patients experienced tumor progression. The remaining 10 patients did not progress at this time point (10 weeks from study entry). Among these, four patients had regression of tumor metastases at one or more disease sites and three patients cleared any evidence of disease.

Recently, Butterfield and colleagues (Butterfield et al, 2003) have conducted a phase I study in 18 HLA-A*