ID #920 Combination XPO1 inhibition and cellular immunotherapy for the treatment of pediatric high-grade gliomas
Charlotte Feddersen, Gabrielle Link, Eric Hoffmeyer, John Desisto, Stephanie McGrath, Steven Dow, Michael Verneris, Adam GreenAbstract
Background
Pediatric high-grade glioma (PHGG) is the most common cause of pediatric cancer death, and current treatments provide only short-term benefit. Two promising treatments are inhibition of nuclear transporter XPO1 and B7-H3 CAR-T therapy. XPO1 is the sole nuclear exporter of many tumor suppressor proteins, transporting them out of the nucleus to encourage tumor growth. The XPO1 inhibitor selinexor (SLX) was well tolerated in phase I/II trials for PHGG and led to prolonged stable disease in several patients. B7-H3 is an immune-regulatory checkpoint protein with low expression in normal tissue but high expression in PHGG. Data from the diffuse intrinsic pontine glioma arm of the Seattle Children’s B7-H3 CAR-T trial demonstrates the safety and promising effects in a subset of patients. Additionally, XPO1 inhibitors combined with CD19 CAR-T cells for multiple myeloma improves treatment, potentially through SLX’s ability to modify the tumor immune microenvironment (TIME). Based on these findings, we hypothesized that sequential XPO1 inhibition and B7-H3 CAR-T cell immunotherapy would improve treatment of PHGG.
Methods
Patient-derived PHGG neurospheres were co-cultured with B7-H3+CXCR2 CAR-T (BC2-CAR-T) in vitro and in orthotopic mouse models to understand the antitumor effect of SLX and CAR-T. The CXCR2 (IL-8 receptor) was added to the CAR-T to improve CAR-T homing to IL-8 expressing tumors like PHGG. To study the TIME, complex co-cultures were created with macrophages, CAR-T cells, and neurospheres.
Results and Discussion
BC2-CAR-T is effective against PHGG in vitro. Concurrent SLX leads to increased CAR-T PD1, decreased cytokines, and reduced tumor clearance as compared to CARs not treated with SLX, potentially due to the effect of XPO1 inhibition directly on the CARs. However, sequential SLX then CAR-T improves tumor clearance and has no effect on CAR-T exhaustion or cytokines. Further projects with tri-cellular co-culture and orthotopic mouse models will focus on exploring the effect of SLX on macrophages, based on the hypothesis that SLX decreases the immunosuppressive macrophage phenotype and improves CAR-T function.