Laheru and Jaffee review the potential role of tumor vaccines in the management of gastrointestinal (GI) malignancies, which represent the leading cause of cancer death and are believed to be poorly immunogenic. The authors carefully review the questions and controversies surrounding currently available immunotherapeutic strategies and describe ongoing clinical protocols using tumor vaccine therapy, a few of which deserve special comment.
Immunologically Hostile Tumor Microenvironment
One of the hurdles that must be overcome in using immunotherapy for GI malignancies is the immunologically hostile tumor microenvironment. The impediments to the success of immunologic therapy strategies may be related to our incomplete understanding of the intimate, dynamic interaction that takes place within the tumor microenvironment. This includes tumor cells, effector lymphocytes, interstitial cells, and professional antigen-presenting cells.
The mechanisms by which tumor cells evade immunologic recognition include (at least):
Failure to express individual major histocompatibility complex (MHC) class I alleles. Global class I expression defects can be found in the setting of beta-2-microglobulin loss or inactivation of the transporter associated with antigen processing (TAP) gene.
Induction of tolerance or anergy because of the absence of appropriate induction molecules during the evolution of the tumor to elicit an immune response. In the absence of appropriate costimulatory molecules, antigen-specific T-cell hyporesponsiveness may develop.
Production of immunosuppressive factors, such as transforming growth factor-beta (TGF-beta), prostaglandin E2 (PGE2), and interleukin-10 (IL-10), among others. Recently, it has been shown that some GI malignancies downregulate or lose the expression of Fas or may express nonfunctional Fas. This enables tumors to escape Fas-mediated apoptosis. In addition, some tumors have been shown to express functional Fas ligand, which induces apoptosis in activated T-lymphocytes.
Inability of immune effectors to cross endothelial barriers into tumor sites. Virtually all GI malignancies express vascular endothelial growth factor (VEGF), which increases permeability for obtaining oxygen and nutrition but not the expression of adhesion molecules of the vascular endothelium. Consequently, the effector lymphocytes cannot adhere to tumor vessels.
Premature apoptosis of immune effectors in the tumor microenvironment, associated with insufficient reciprocal provision of survival factor between dendritic cells and T-cells, or direct induction of apoptosis by tumor-derived factors.
Approaches to Modifying the Tumor Microenvironment
Laheru and Jaffee review some possible solutions to these problems. These include the induction of class I or costimulatory molecules and minimization of the tumor burden by conventional therapeutics.
Another approach to modifying the tumor microenvironment and presumably presenting the tumor antigen to antigen-presenting cells in vivo is peritumoral injection of genetically modified fibroblasts that produce copious amounts of cytokines, such as interleukin-4 (IL-4), interleukin-12 (IL-12), and granulocyte-macrophage colony-stimulating factor (GM-CSF).
Antiangiogenesis treatment is an ideal combination with tumor vaccine therapy, especially when the tumor consists of minimal residual disease. Delivery of dendritic cells to tumor sites using exogenous administration of such cytokines as FLT3 ligand or direct injection of suitably activated dendritic cells would be possible. Protecting dendritic cells from premature apoptotic death in the tumor microenvironment then becomes a worthwhile goal.
Identifying Tumor-Specific Antigens
Another hurdle is the failure to identify the specific antigens present in given tumors. Although the strategies to identify tumor-specific antigens have been successfully developed, they are still time-consuming, laborious, and expensive. Tumors may have mutations or post-translational modification of protein, such as glycosylation. Furthermore, most tumors are heterogeneous.
In the absence of apparent or known targets, the use of whole tumor, preferably that derived from apoptotic or necrotic tumor cells fed to immature dendritic cells, would seem to be a reasonable approach that is supported by available data. Our group and other investigators have demonstrated that apoptotic cells and bodies represent the preferred antigen source for dendritic cells, promoting maturation of these cells, as well as their activation.
Optimal Vaccine Delivery Strategies
A successful vaccine will also require the identification of strategies to deliver the antigen(s) to antigen-presenting cells in a way that results in optimal induction of antitumor immunity. Recently, Henry et al reported that antigen-presenting cells that phagocytose apoptotic tumor cells could be potent tumor vaccines. They showed that injection into tumor-bearing rats of antigen-presenting cells that had phagocytosed apoptotic bodies derived from poorly immunogenic tumor cells produced an 80% cure rate, whereas phagocytic cells exposed to nonapoptotic tumor cell extracts were essentially without effect.
These results suggest a simple protocol involving only the ex vivo mixing of a patients antigen-presenting cells with apoptotic bodies derived from autologous cancer cells. This protocol implies no prior knowledge of how antigens are expressed by tumor cells, and no choice needs to be made concerning the best antigen to target. Thus, each individual would benefit from a tailor-made therapy.
What is the best way to generate immunogenic apoptotic cells? Apoptosis can be induced chemically (by, for example, sodium butyrate and anticancer drugs) and physically (gamma irradiation, ultraviolet irradiation). Recent research suggests the important role played by natural killer cells. Recognition of MHC class I imparts a negative or inactivating signal to the natural killer cell, implying, conversely, that downregulation of MHC class I would make tumor cells more susceptible to natural killer cell attack.
Activated adherent natural killer (A-NK) cells that adhere rapidly to solid surfaces under the influence of interleukin-2 (IL-2) have a higher level of IL-2 receptors and adhesion molecules and are superior to their nonadherent counterparts in entering solid tumors and prolonging survival following adoptive transfer with IL-2 into animal models. Our laboratory is now testing the efficacy of apoptotic tumor bodies generated by A-NK cells in inducing tumor-specific immunity.
Although a substantial amount of work remains, the possibility of designing an effective vaccine approach for GI malignancies has become more realistic, and the hope is that such an approach will be realized in the near future.