Characterization of the humoral and cellular immune responses to cell-based neuroblastoma vaccines.

摘要

Neuroblastoma is the most common extra-cranial tumor in pediatrics, and is responsible for approximately 15% of all childhood cancer deaths. Although advances have been made in the treatment of neuroblastoma, the prognoses for advanced stage disease remains bleak. Immunotherapy holds promise as a novel avenue of treatment in neuroblastoma as clinical observations from bone marrow transplant recipients suggest that the immune system is capable of recognizing and eradicating neuroblastoma. Clinical trials have provided evidence that immune responses can be generated against neuroblastoma, however obstacles such as immunological tolerance, a lack of known tumor antigens and improper activation of T cells must be overcome if these strategies are to reach their full potential.;Immunological tolerance prevents the activation of the adaptive immune system to the molecules that make up an organism (self-antigens). While tolerance is an important part of immunological homeostasis, it is a hindrance to the induction of therapeutic immunity against malignancies, which are composed predominantly of self-antigens. Therefore, breaking immunological tolerance is thought to be a requisite for the development of a successful immunotherapeutic regimen.;The mouse tumor cell line, AGN2a, is used as a preclinical model of neuroblastoma. While inoculation with live AGN2a will cause the host to succumb to the tumor, genetically engineering the tumor cell line with the appropriate expression vectors transforms this lethal tumor into a cell-based vaccine. In our lab, we have found that specific cell surface molecules, termed immune co-stimulatory molecules, are sufficient to transform a tumor cell into a vaccine. When we examined the serum of vaccinated mice, we found broken tolerance to both the AGN2a cell line and to self-antigens, as these mice generated tumor- and self-reactive antibodies. While the production of tumor-reactive antibodies correlated with tumor immunity, they were not essential in protecting the mice from tumor challenge.;Nevertheless, these tumor-reactive antibodies are a unique tool for the discovery of tumor-antigens. It has been well documented that cancer patients produce antibodies against their malignancies, and these antibodies have been used to identify tumor antigens. It has been questioned, however, whether antigens identified in patients which are tolerant of their tumors are ideal targets for immunotherapy. Using antibodies from a model of broken tolerance and tumor immunity, we developed Reverse-SEREX, a novel strategy for the identification of tumor antigens. Reverse-SEREX allowed us to combine gene expression data collected from the tumors of forty-three neuroblastoma patients with the antibody response of our murine model of tumor immunity. Through this, we identified Nek2, a serine/threonine kinase involved in cell cycle progression, as a neuroblastoma antigen. We found the antibody response to Nek2 was affected by tolerance, as the antibody response in vaccinated (tumor-immune) mice was up to 45 times greater than that of tumor-bearing (tumor-tolerant) mice.;We have demonstrated that treatment with our cell-based vaccine elicits a CD8+ T cell response which protects mice from a lethal challenge of AGN2a. In a more clinically relevant setting, however, we have shown that vaccination cannot serve as a therapeutic intervention in mice with established tumors. After vaccination, the CD8+ T cells of tumor-bearing mice were found to have a diminished interferon-gamma response to tumor. Gene expression profiling of CD8+ T cells after vaccination revealed that the expression of stem cell-related genes correlated with the production of tumor immunity in tumor-free mice. This genetic signature was lost in tumor-bearing mice after vaccination, suggesting the expression of these genes is required for successful tumor immunity. This is the first report of stem cell-related genes being expressed early in T cell activation and we hypothesize that these genes are involved in memory formation in the CD8+ T cell compartment.;The data presented within this dissertation establishes a new methodology for the identification of tumor-antigens that utilizes clinical gene expression data in a murine model of tumor immunity. Moreover we have identified a genetic signature of early T cell activation that predicts CD8+ T cell responses and productive tumor immunity after vaccination. Finally, we present a novel hypothesis for the mechanism of memory formation in CD8+ T cells early after activation and the influence of established tumor on these early events.