Improvement of Tumor Vaccines with MiImeotopes of Tumor-associated Antigens.
Research projects in our laboratory combine expertise in molecular biology and immunology to define strategies for induction of effective responses against neuroectodermally-derived tumors including neuroblastoma, melanoma and glioma.
Evidence suggests that among antigens expressed on the neuroectodermally-derived tumors, GD2 ganglioside can be a subject to specific immunological attacks. A number of monoclonal antibodies (mAbs) directed against this ganglioside has previously been shown to generate reproducible clinical responses in patients with melanoma and neuroblastoma. Because GD2 is weakly immunogenic, we investigated whether interconversion of GD2 into a peptide mimic form would induce immune responses capable of inhibiting growth of GD2-expressing primary and metastatic tumors in a syngeneic murine model. Screening of the X15 phage display peptide library with anti-GD2 mAb 14G2a led to isolation of a surrogate peptide of GD2 based on a molecular model (Figure 1) and the ability to inhibit the binding of 14G2a antibody to GD2+ tumor cells. The research projects currently under study include optimization of the GD2 mimetic vaccine delivery system, and development of protocols combining the GD2 mimetic vaccine with cytokines and intratumoral virotherapy to augment antitumor immune responses during adoptive T cell transfer.
The ability of the mimetic peptide to induce protective immune responses in a clinically relevant tumor tolerance model, which could not be effectively obtained by immunization with the GD2 ganglioside vaccine, stresses the importance of peptide mimetic vaccines in cancer-bearing patients.
Future directions include:
- Augmenting an adjuvant property of the mimetic vaccine by expressing the peptide mimic of GD2 ganglioside in the context of IgG2 Fc fusion proteins for delivery of the antigenic cassette to the activating Fc gamma receptors on dendritic cells.
- Elucidation of mechanisms of an antigenic mimicry between the peptide mimic and the carbohydrate residues of GD2 responsible for induction of CD8+ T cell responses that recognize a cross-reactive epitope on tumor cells.
Effect of CD133 inhibition on the self-renewal capacity of tumor stem cells in a SCID mouse model of human glioblastomas. Collaborarion with Drs. Michael Ciesilski and Robert Fenstermaker, Departments of Neurosurgery and Immunology, Roswell Park Comprehensive Cancer Center.
Glioblastoma multiforme (GBM) is the most common primary brain tumor of adults, and is among the most lethal of all cancers. Despite aggressive multimodality management with surgery, radiation, and temozolomide (TMZ)-based chemotherapy, the median survival times of GBM patients still range from 12 to 15 months. Clearly, new approaches to the understanding and treatment of GBM are needed. Molecular and genetic mechanisms that serve to enhance the malignant phenotype of these tumors are becoming better understood. Recently, brain tumor stem cells (BTSCs) have been identified and purified from human gliomas, and showed to possess a marked capacity for proliferation, self-renewal, and differentiation. The increased self-renewal capacity was highest in the most malignant clinical samples of brain tumors. The BTSCs were exclusively isolated with the cell fraction expressing the neuronal stem cell marker CD133. These CD133+ cells could differentiate in culture into tumor cells that phenotypically resembled the tumor from the patient, thus providing a powerful tool to investigate the tumorigenic process in the central nervous system and to develop therapies targeted to the BTSC. The long-term objective of the study is to unravel the molecular mechanism of GBM development and progression and to employ this information for developing novel therapeutic strategies for these tumors. Using shRNA construct specific for CD133, we have downregulated CD133 expression in human U87 glioblastoma cells. The stable U87 transfectant clones are being tested for tumor growth in traditional orthotopic murine models (Figure 2).