Main

The discovery of distinct antigenic features of neoplastic cells has suggested that it may be possible to create anti-tumor vaccines1. Vaccine strategies have used tumor-derived cells2, or cellular material3, containing cryptic or unidentified antigens to induce specific immune recognition of these tumor-associated antigens (TAAs). Alternatively, several recently discovered tumor-associated antigens have been tested directly as immunogens in vaccine formulations4,5. To improve antigen presentation, vaccine approaches have also included adjuvant agents to potentiate immune recognition of the TAA. These adjuvants have evolved to improve immune activation, and are the result of many years of research directed towards understanding of the mechanisms of anti-tumor immunity6. Bendandi et al., on page 1171 of this issue7, report a TAA vaccination protocol that, with granuloyte-monocyte colony-stimuting factor (GM-CSF) as an immune adjuvant, achieves an effective anti-tumor response.

A variety of anti-tumor vaccine clinical trials have been recently undertaken. In spite of the large number of these trials, and the plethora of distinct approaches investigated, there has been little evidence of clinical efficacy3,4. Furthermore, precise correlates of clinical effects and immunological responses have been lacking. These limitations reflect several shortcomings in the design of the studies. Studies have often used cohorts of patients with advanced stage disease, for which a full response is unlikely for various biological reasons. Moreover, many of the adjuvants tested lacked the potential to effectively overcome intrinsic immunosuppressive activity of the tumor. Finally, precise assays to monitor clinical response and, most importantly, immune responses have been lacking. Thus, despite a large body of work in this area, many tumor vaccine protocols tested to date have not proven effective or yielded large amounts of useful information.

The study by Bendandi et al. makes a large advance over previous tumor vaccine trials. Idiotypic determinants of the Ig synthesized by a clonal B-cell cancer, such as follicular lymphoma (FL) are unique, and are thus useful as TAAs (Refs 4,9). The vaccine strategy of Bendandi et al. included a new group of follicular lymphoma (FL) patients that were in complete remission after chemotherapy. Most FL patients in remission still have chromosomal translocation-bearing tumor cells, and are therefore at risk for relapse. The authors created a vaccine that combined an FL-associated idiotype IgG with keyhole limpet hemocyanin (KLH) (Fig. 1). The inclusion of KLH provided additional epitopes for T-helper-cell recognition, allowing for optimal CTL activation, and also provided a positive control antigen, allowing the authors to monitor T-cell priming. The vaccine was administered in conjunction with the molecular adjuvant GM-CSF, which is very effective in induction of immune effector cells6,8. The protocol also included highly sensitive assays to carefully monitor detectable translocations in the blood and quantify the tumor-specific T-cell response, the hallmark of true anti-tumor immunization.

Figure 1: Mechanisms of CTL activation after vaccination with a lymphoma-associated idiotype (Id) IgG coupled to keyhole limpet hemocyanin (KLH) and mixed with GM-CSF.
figure 1

GM-CSF induces recruitment of dendritic cell (DC) precursors from the blood to the vaccination site and stimulates local DC maturation. The Ig–KLH complex is internalized and processed by the DCs, and Id-derived epitopes are expressed on the cell surface in the context of MHC-I and II molecules. DCs then migrate to the lymph nodes (oval), where they present the antigen and activate a TAA-specific T-cell response.

Full size image

This study included 20 FL patients, eleven of which possessed the chromosomal translocations characteristic of FL. After six months recovery from chemotherapy, patients were vaccinated with four monthly injections of their own lymphoma Ig idiotypic determinant, along with recombinant GM-CSF. After vaccination, patients were monitored for chromosome translocation-bearing cells by PCR. Anti-tumor immunity was determined by assays to measure cytolysis and cytokine release in response to autologous tumor cells. Of the eleven patients evaluated, eight cleared the residual chromosomal translocation-bearing cells in response to vaccination. Nineteen of the twenty patients tested showed specific CD8+ and CD4+ T-cell reactivity to the tumor. Evidence of anti-tumor immunity correlated with clinical evidence of a reduction of tumor burden.

This trial, along with other recent trials, has benefited from the advances made in our understanding of anti-tumor immunity. Most new immunotherapeutic approaches aim to promote the activation of CD8+ cytotoxic T lymphocytes (CTLs), which have been recognized as the main effector cells in a multitude of murine tumor rejection models. It is also known that tumors use a variety of mechanisms to induce host tolerance and prevent TAA-specific CTL activation6. Dendritic cells (DCs), which express high levels of MHC molecules, co-stimulatory molecules, and T-cell-activating cytokines, may be useful in overcoming these obstacles and activating TAA-specific CTL responses (Fig. 1). Many new vaccines, therefore, also include autologous DCs that have been loaded with TAA (Refs 3,6) in vitro. This study instead uses the DC stimulant GM-CSF.

GM-CSF functions to promote TAA presentation by antigen presenting cells (APCs) and CTL activation by recruiting peripheral DC precursors and inducing DC maturation (Fig. 1) (ref. 6). Recent studies have also confirmed the role of CD40-mediated activation of DCs in activation of the TAA-specific CTL response10. Indeed, CTL-dependent rejection of B-lymphomas in mice was recently reported after treatment with an agonistic antibody against CD40 (ref. 11). Thus, GM-CSF and CD40 ligand may prove to be a winning adjuvant combination in future lymphoma, as well as other tumor vaccines.

The vaccine strategy reported by Bendandi et al. does not represent a particularly new approach, but its validation of therapeutic benefit, well-planned study design, and careful execution allow this study to stand as a hallmark in the field of cancer vaccine development. Consideration of the patients' disease stage cannot be ignored in determining the potential of a vaccine to achieve a beneficial therapeutic outcome. These findings suggest that in cancer therapy, early intervention with vaccine-based approaches is warranted. Furthermore, studies including specific TAAs as vaccines should carefully monitor post-vaccination immune responses. Future studies aimed at identification of TAAs for other tumor types, and additional carefully- monitored tumor vaccine trials, should lead to the development of new, more effective vaccine strategies.