Engineering T Cell Receptors to Target H3.3G34 Mutant Glioblastoma. (16th November 2020)

• Record Type: Journal Article
• Authors: Wang, Anthony C; Owens, Geoffrey; Lee, Alexander; Sun, Matthew Z; Peeters, Sophie; Kienzler, Jenny C; Orpilla, Joey; Contreras, Erick; Treger, Janet; Odesa, Sylvia K; Everson, Richard G; Holland, Eric C; Becher, Oren; Nathanson, David; Xing, Yi; Liau, Linda M; Prins, Robert M
• Is Part Of: Neurosurgery. Volume 67(2010)Supplement 1
• Publication Date: 2020-11-16
• Subjects: Nervous system -- Surgery -- Periodicals; 617.48005
• DOI: 10.1093/neuros/nyaa447_794
• Languages: English
• ISSNs: 0148-396X

Abstract

Abstract: INTRODUCTION: The two main subtypes of pediatric glioblastoma (GBM) are marked by highly conserved somatic H3-3A gene mutations. The H3.3G34R/V GBM occurs in children and young adults, is typically lobar in location, and frequently harbor concurrent TP53, ATRX/DAXX, and PDGFRA alterations. H3.3G34 alterations appear to dysregulate specific alternative splicing events. We hypothesize that the H3.3G34R/V mutation in GBM results in tumor-specific neoantigens targetable by T cell receptor (TCR)-engineered cytotoxic T lymphocytes. METHODS: We used NetMHC to identify peptide sequences encompassing H3.3G34R/V predicted to show strong human leukocyte antigen (HLA)-major histocompatibility complex (MHC) binding; GATK-Mutect2, to identify neoantigens called from somatic variants; and IRIS to identify neoantigens derived from alternative splicing variants. RESULTS: Clonally expanded populations of T cells demonstrated binding affinity for mutation-derived neoantigen targets, and activation in response to patient-derived H3.3G34R GBM cells in vitro. These TCRs were then cloned in a retroviral expression vector. To isolate TCRs mediating HLA-specific binding of the peptides encompassing the H3.3G34R mutation predicted to bind with strong affinity, HLA-A*03:01 specific oligomeric class I MHC molecules were designed for high-value predicted peptides encompassing the H3.3G34R mutation, according to NetMHC predicted strong binders. We created a patient-derived xenograft cell line from an HLA-A*02:01 patient harboring H3.3G34R GBM. Patient CD3+ T cells were expanded in the presence of synthetic long peptides encompassing predicted neoantigens, and sorted for CD8+ populations. The peptides were synthesized and used in an indirect HLA binding assay. HLA-specific CD8+ T cell binding was tested to confirm T cell binding of identified neoantigen candidates using oligomeric MHC I molecules. Serial oligomer selection was performed to expand the T cell population selectively binding the target peptide sequence. Jurkat cells were transfected to express the neoantigen-specific TCRab chains, and confirmed antigen-specific binding. CONCLUSION: We have confirmed a small number of candidate neoantigens present within H3.3G34R GBM. Re-stimulated T cells expanded against these peptides demonstrated high binding affinity for the two splice variant-derived neoantigen targets, and activation in response to patient-derived H3.3G34R GBM cells in vitro.