Immune responses are generated in a complex network of cellular and humoral factors. The complexity of this system makes it difficult to generate subsets of cells in vivo that are most effective against cancer cells. The goal of vaccine strategies is to redirect the immune system against cancer cells primarily by generating specific T-cell responses which would be the most effective anti-tumor effector cells.
The immune response against cancer cells is significantly impaired by deficiencies in tumor antigen presentation, and by the lack of Th1-like cytokine network responses.
Furthermore, while tumor vaccines might be able to generate specific T-cell responses, the local tumor cell microenvironment can inhibit these activated cells by secreting immunosuppressive factors or by not allowing sufficient tumor penetration due to lack of vascular supply in tumor tissue. Although in vitro generated lymphokine activated killer (LAK) cells have been found to have difficulties homing to the tumor site in vivo, it is possible that these cells stimulated by vaccines in vivo possess a greater ability to recognize tumor cells at distant sites. Thus, a higher concentration of effector cells at the tumor site might elicit clinical anti-tumor effects superior to those observed in clinical studies employing ex vivo expanded lymphokine activated killer cells.
In this issue of Oncology, Gurski and Steller have provided an excellent summary of therapeutic and prophylactic vaccine strategies currently under investigation for gynecologic cancers. As they indicated, immune responses generated by tumor cell vaccines should ideally target specific tumor-associated antigens. However, no specific antigen has yet been defined for gynecologic malignancies. The most promising candidates for immunization against tumor antigens are Her2/neu, which is overexpressed in approximately 30% of ovarian cancers, or the HPV16 and HPV18 proteins in cervical cancer. MUC-1, CD44, and CA125 might also be useful as target antigens.
Dendritic cells are very potent antigen-presenting cells. They have been used in a number of animal models to stimulate tumor-associated antigens leading to significant anti-tumor effects. Dendritic cells can be purified from peripheral blood and bone marrow, but they are difficult to manipulate. Stimulation and expansion in vitro must be performed in the presence of different mixtures of growth factors but the best methodology has yet to be determined. For example, it is unclear when during in vitro expansion, dendritic cells possess the greatest immune stimulatory capacity. Furthermore, it is possible that other dendritic-like cells exist that have even greater immunostimulatory capacitya candidate cell type that meets this criteria has recently been described and is called a fibrocyte.
The use of a prophylactic vaccine in cervical cancer having immunodominant HPV16 related peptides is described. It remains to be seen whether the observations made in vitro and in animal models that describe the generation of anti HPV-specific CTL (Cytotoxic Lymphocytes) responses will translate into humans. Furthermore, although the correlation between cervical cancer and HPV infection has been described extensively, the molecular event leading from a normal cervical epithelial cell to cervical intraepithelial neoplasia (CIN) and finally cervical cancer is still unknown. Recent evidence suggests that HPV might facilitate the uptake of foreign DNA into the host cell genome. It is therefore possible that HPV is not the transforming factor, but rather promotes the integration of foreign DNA derived from for example oncogenic viruses with transforming activity. HPV based tumor cell vaccines may prevent infection with HPV, but would not be able to abrogate the oncogenicity of other particles. Another possible problem might arise when different mutants of a specific HPV serotype escape the effect of immunization.
Most immunotherapeutic studies of gynecologic malignancies have been performed in ovarian cancer. Tumor-infiltrating lymphocytes and different cytokines including interleukin-2 have been injected into the peritoneal cavity of patients with ovarian cancer with some clinical effect. In immunogene therapy, the in vivo expression of immunostimulatory cytokine genes in autologous tumor cells has generated anti-tumor responses. This approach presents an interesting alternative particularly since the continuous production of cytokines obviates the necessity of high cytokine doses and significantly reduces side effects. However, the ovarian cancer microenvironment contains tumor-cell derived immunosuppressive factors such as TGF-beta, that might interfere with the effect of vaccine-stimulated effector cells. A more profound understanding of how tumor-associated factors inhibit anti-tumor responses is needed to increase the clinical effect of vaccine approaches.
Site of Vaccine Injection
An important question that remains to be answered pertains to the site of vaccine injection. It is possible that intraperitoneal vaccination in ovarian cancer could enhance clinical effects due to the presence of large numbers of peritoneal macrophages that are able to migrate to local lymph nodes and effectively present antigen to the cellular immune system. Intradermal injection might be able to target Langerhans cells, located primarily in this compartment and functioning as potent immunostimulatory dendritic cells. Additionally, lymph nodes adjacent to immunization sites might be harvested and the lymphocytes isolated from these lymph nodes could be used for adoptive cell transfer. Preliminary observations from clinical tumor cell vaccine studies in other cancers have shown some interesting anti-tumor effects with this approach.
Tumor cell vaccines have promising clinical potential for use in gynecologic malignancies. The vaccines will likely be most effective in a prophylactic or adjuvant setting. It remains to be seen whether tumor cell vaccines can prevent tumor growth in patients that are genetically predisposed to develop specific diseases like ovarian cancer. Germ-line mutations found in ovarian cancer may lead to the development of malignancies through overexpression of oncogenes or dysfunction of tumor-suppressor genes. These mechanisms could override anti-tumor immune responses generated by tumor cell vaccines. Therefore, it might be necessary to combine tumor cell vaccines with other antiproliferative molecular interventions.