Gerold Schuler
University Hospital, of Erlangen, Erlangen, Germany

Dendritic Cell-Based Vaccination Against Cancer

Among the many cancer therapeutics being presently developed, antigen-specific immunotherapy is considered a particularly promising approach. Active immunotherapy or vaccination to induce tumor-specific killer and helper T cells directly in the patient is potentially the most powerful and broadly applicable strategy, but has yet to be developed for humans. A novel approach is to take advantage of DCs as “nature’s adjuvant” and to actively immunize cancer patients with a sample of their own DC charged with tumor antigens (reviewed in 1,2). The use of DC as adjuvants is supported by many animal experiments, and also by the trials reported so far for humans. DC vaccination is, however, still in an early stage, and most potentially important variables like type and maturational stage of DCs, type of antigen and loading method, cell dose, route and frequency of injections have still to be systematically adressed in future studies. Despite this need for establishing an optimized DC vaccine even in some of the initial exploratory clinical trials, vaccination with DC loaded with tumor antigens in the form of peptides has induced a) tumor-specific killer and helper T cells (“proof of concept”), and b) occasional regression of metastases even in far-advanced cancer patients. This was reported for DC directly isolated from blood (3) as well as for DC generated ex vivo from either CD34+ (4) or CD14+ precursors (5,6). Most of the studies including informative ones in volunteers (7) have been performed by using DC generated from CD14+ monocytes (so-called Monocyte-derived DC or Mo-DC) which are now considered as a gold standard. These Mo-DC can be reproducibly generated within a few days in large numbers (300-500 million mature DC per apheresis) from precursors in blood without any need for pretreating the patients with cytokines like GM-CSF or Flt3-L (8). Importantly, it is possible to obtain populations of immature DC by exposing monocytes to GM-CSF + IL-4, which can then be transformed into homogenously mature DC by various stimuli such as TLR ligands (e.g. microbial products such as LPS or poly I:C), inflammatory cytokines like TNF alpha, monocyte-conditioned medium or its mimic (IL-1beta + TNF alpha + IL-6 + PGE2), or CD40L. The use of mature DC is likely critical to induce strong immunity as it has become clear that antigen delivered on immature or incompletely matured DC can even induce tolerance (9). Interestingly, it has recently also become evident that in case of the Mo-DC the choice of maturation stimulus is probably critical for success. Specifically, PGE2 has to be part of the maturation stimulus in order to obtain CCR7 expressing Mo-DC that migrate in response to CCL19 and CCL21 that guide DC into lymphoid organs (10,11). It is of note that many trials employing Mo-DC have not used such optimally matured DC while we and some others have favorized both the use of mature DC and monocyte conditioned medium or its mimic as maturation stimulus. Using such DC loaded with tumor peptides their migratory capacity in vivo, the induction of tumor-specific cytotoxic and helper T cells, and the presence of tumor-antigen specific T cells in situ in regressing metastases have been demonstrated. Recently, new approaches to charge DC with antigens have become evident which promote the DC vaccination approach. The observation that DC can take up naked RNA, express antigens encoded by the RNA and induce antigen-specific T cells in vitro as well as in vivo in patients has given an enormous additional momentum to the use of DC as vectors for antigen delivery and cancer vaccination (12). This is particularly relevant as the variable transfection with naked RNA can be substituted by a reproducible electroporation protocol (13). This now allows one to administer to DC both defined antigens, including universal ones such as telomerase or survivin as well as the total antigenic repertoire of a given tumor (as total tumor or PCR amplified RNA). The use of dying tumor cells, notably antibody-coated ones, is an alternative (14). Other preclinical research has shown that the delivery of defined antigens as antigen-antibody complexes to DC enhances crosspresentation and allows for the potent induction of both CD4+ and CD8+ T-cell responses (15-17). Several recent results suggest that DC might even be useful to directly trigger NK (18) and to mobilize the additional power of the innate immune system to attack tumor cells. It appears, therefore, rational and timely to optimize the use of DC as vectors for the delivery of antigens to vaccinate against cancer.

Selected References:

1. Steinman RM, Dhodapkar M.. Int J Cancer. 94: 459, 2001.
2. Banchereau J, Schuler-Thurner B, Palucka AK, Schuler G. Cell. 106: 271, 2001.
3. Fong L et al. Proc Natl Acad Sci U S A. 98: 8809, 2001.
4. Banchereau J et al. Cancer Res. 61: 6451, 2001.
5. Thurner B et al. J Exp Med. 190: 1669, 1999.
6. Schuler-Thurner B et al. J Exp Med. 195: 1279, 2002.
7. Dhodapkar MV et al. J Clin Invest.105: R9-R14, 2000.
8. Feuerstein B et al. J Immunol Methods. 245: 15-29, 2000.
9. Dhodapkar MV et al. J Exp Med. 193: 233, 2001.
10. Luft T et al. Blood. 100: 1362, 2002.
11. Scandella E et al. Blood. 100: 1354, 2002.
12. Heiser A et al. J Clin Invest. 109: 409, 2002.
13. Van Tendeloo VF et al. Blood. 98: 49, 2001.
14. Dhodapkar KM et al. J Exp Med. 195: 125, 2002.
15. Berlyn KA et al. Clin Immunol. 101: 276, 2001.
16. Kalergis AM, Ravetch JV. J Exp Med. 195: 1653, 2002.
17. Rafiq K, Bergtold A, Clynes R. J Clin Invest. 110: 71, 2002.
18. Ferlazzo G et al. J Exp Med. 195: 343, 2002.