396. Development of GD2-Specific Immunoliposomes for Immunotherapy of Neuroblastoma

Abstract

Tumor microenvironment (TME) impairs immune effector cell localization, persistence, or the execution of their anti-tumor function. This represents a major and as yet unsolved critical challenge to the development of effective adoptive immunotherapy of solid tumors. To selectively target TME in neuroblastoma and make it permissible for survival and function of tumor-specific effector cells, we have developed a novel nanoparticle (NP) delivery platform which consists of 150 nm immunoliposomes rendered specific for neuroblastoma cells using a single-chain variable fragment (scFv) from the anti-GD2 14g2a mAb. GD2-specific but not control NPs could specifically bind GD2-positive but not GD2-negative neuroblastoma cells in vitro as determined by FACS. To examine the in vivo biodistribution of NPs, DiR-labeled GD2-specific or non-specific NPs were injected to NSG mice implanted with human neuroblastoma xenografts. Tumor tissues and normal organs were analyzed after 72 hours using ex vivo fluorescence imaging. Up to 58% of GD2-specific NPs accumulated at the tumor sites. NPs loaded with human IL-15 effectively delivered cytokine to the tumor site. A combined immunotherapy using human NKT cells expressing a GD2-specific chimeric antigen receptor (CAR) and IL-15-containing NPs resulted in potent in vivo expansion of CAR NKT cells and tumor eradication. In contrast, soluble IL-15 failed to support long-term CAR NKT cell persistence that resulted in tumor recurrence. These results inform design of cancer immunotherapy in combination with TME-modifying NPs.