Comprehensive Antigen Discovery in DIPG for the Development of Novel Immunotherapies

Diffuse intrinsic pontine glioma (DIPG) is a universally fatal pediatric brain tumor with median survival under 12 months. Current therapies offer only temporary benefit. A critical barrier to immunotherapy in DIPG is the absence of well-defined, tumor-specific antigens. Canonical, mutation-derived antigens are rare, but emerging evidence suggests that structural variation (SV), transcriptional dysregulation, and aberrant transposable element (TE) expression generate abundant non-canonical antigens that may serve as novel, immunogenic targets.

Hypothesis:
We hypothesize that DIPG harbors a diverse and immunogenic repertoire of non-canonical antigens arising from SVs, tumor-specific transcripts, aberrant splicing, and TE activation, and that these can be discovered and validated using long-read multi-omic and immunopeptidomic approaches to enable development of new immunotherapies.

Specific Aims:
• Aim 1: Identify immunogenic antigens arising from pediatric glioma-specific genomic structural variants.
• Aim 2: Elucidate antigens deriving from tumor-specific transcripts and aberrant alternative splicing.
• Aim 3: Discover antigens arising from transposable element alterations specific to pediatric gliomas.

Scientific Approach:
The project will proceed in two parallel phases for each aim:
1. Discovery: We will integrate short- and long-read genomic and transcriptomic sequencing with immunopeptidomic profiling in at least 50 DIPG patient samples. Our computational pipelines will nominate the top recurrent candidate antigens per class (SVs, splicing isoforms, novel transcripts, TE-derived).
2. Validation: Using our established in vitro T-cell expansion platform, we will confirm HLA presentation and immunogenicity of nominated antigens with healthy donor and DIPG patient PBMCs. Flow cytometry and functional assays will define antigen-specific CD4+ and CD8+ T-cell responses. Cytotoxicity will be tested against HLA-matched patient-derived xenograft (PDX) models. Single-cell RNA/TCR sequencing will yield cognate TCRs for therapeutic development.

Preliminary Data:
• Long-read sequencing expands the antigenic landscape. Using long-read sequencing in conjunction with immunopeptidomics, we identified over 100-fold more human leukocyte antigen (HLA)-presented non-canonical antigens compared with conventional mutation-based pipelines, uncovering 100–500 novel antigens per tumor across multiple cancer types.
• Non-canonical antigens dominate in diffuse midline glioma (DIPG). In three DIPG cell lines, we discovered more than 140 novel HLA-presented non-canonical antigens that were exclusively tumor- associated. By contrast, fewer than two mutation-derived antigens were identified per line, strongly validating the relevance of our approach in this disease context.
• T-cell assays confirm immunogenicity of shared non-canonical antigens. We established a robust peripheral blood mononuclear cell (PBMC)-based assay capable of rapidly expanding endogenous antigen-specific T cells. Using this platform, we demonstrated immunogenicity and cytotoxic activity against shared non-canonical antigens in pancreatic cancer, with responses observed in >90% of cases.

Significance:
This work directly addresses the central barrier to immunotherapy in DIPG: the lack of validated tumor-specific targets. By comprehensively discovering and validating novel antigen classes, this project will generate a prioritized catalog of DIPG-restricted antigens, define their cognate TCRs, and provide translational proof-of- concept for vaccine and TCR-based therapies. These outcomes will establish the foundation for first-in-human immunotherapy trials in DIPG and transform the therapeutic landscape for this universally fatal pediatric cancer.