{
  "abstract": "Background Gene fusions are an important class of tumor-specific neoantigens due to their unique expression in cancer cells and absence in normal tissues, offering high specificity for immunotherapeutic targeting approaches. 1 2 The FGFR3-TACC3 fusion, recurrent across several malignancies including glioblastoma and bladder cancer, drives oncogenesis through constitutive FGFR3 kinase activation but remains understudied as a neoantigen source. Here, we present a comprehensive pipeline integrating computational prediction, structural modeling, biochemical validation, and single-cell immune profiling to identify immunogenic peptides derived from the FGFR3-TACC3 fusion.Methods Patient-derived xenograft (PDX) model harboring the FGFR3 exons 1–15 fused to TACC3 exons 9–16 transcript was utilized to design 36 synthetic peptides (9–11mers) spanning the fusion junction and flanking wild-type regions. Peptides were computationally screened for HLA-A*01:01 binding affinity using NetMHCpan 4.1, MixMHCpred 3.0, and BigMHC. Structural peptide-HLA (pHLA) complex modeling and docking were performed via Boltz1 and APE-GEN2.0, with binding free energy consensus ranking been evaluated using the PBEE machine-learning predictor. Cross-reactivity potential was assessed with CrossDome to minimize off target risks. Experimental validation was performed using flow cytometry-based HLA stabilization assay and IFN-γ ELISpot assays on CD8 + T cells from HLA-matched donors’ peripheral blood mononuclear cells (PBMC). Pools of confirmed immunogenic peptides induced robust T cell activation and proliferation, confirmed by single-cell RNA sequencing revealing activation and exhaustion specific transcriptional signatures. T cell receptor (TCR) repertoire analyses identified clonal expansions with conserved CDR3 motifs, and TCR-pHLA interaction modeling using TCRmodel and TCRpMHCmodels-1.0 further confirmed immunogenic potential at the receptor level.Results FGFR3-TACC3 fusions were detected in 5.73% of analyzed samples. Multiple fusion-junction peptides including YTHQSDVMH, QSDVMHGANET, YTHQSDVMHGA, QSDVMHGANE, VYTHQSDVMH, RVYTHQSDVMH and HQSDVMHGA demonstrated high predicted affinity, stable HLA binding, and strong IFN-γ induction. CrossDome analysis predicted negligible off-target binding. Comparative assessment highlighted APE-GEN2.0 and TCRpMHCmodels-1.0 as superior tools for structural and immunogenicity prediction, respectively.Conclusions This study establishes the FGFR3-TACC3 fusion as a rich neoantigen source and validates a robust, integrated neoantigen discovery pipeline combining in silico and in vitro methods. These findings lay the groundwork for peptide and mRNA-based vaccines targeting fusion-derived neoepitopes, advancing precision cancer immunotherapy.Acknowledgements UH Sequencing core for sequencing services, XenoSTART for providing PDX models and Antunes Lab at UH for the guidance in structural bioinformatic analyses.References Nacu S, Yuan W, Kan Z, Bhatt D, Rivers CS, Stinson J, Peters BA, Modrusan Z, Jung K, Seshagiri S, Wu TD. Deep RNA sequencing analysis of readthrough gene fusions in human prostate adenocarcinoma and reference samples. BMC Medical Genomics 2011;4:11–21.Varley KE, Gertz J, Roberts BS, Davis NS, Bowling KM, Kirby MK, Nesmith AS, Oliver PG, Grizzle WE, Forero A, Buchsbaum DJ, Lobuglio AF, Myers RM. Recurrent read-through fusion transcripts in breast cancer. Breast Cancer Research and Treatment 2014;146,287–297.",
  "authors": [
    {
      "affiliations": [
        "University of Houston, Houston, TX, USA"
      ],
      "name": "Shiyanth Thevasagayampillai"
    },
    {
      "affiliations": [
        "University of Houston, Houston, TX, USA"
      ],
      "name": "Preethi H Gunaratne"
    },
    {
      "affiliations": [
        "University of Houston, Houston, TX, USA"
      ],
      "name": "Aaranyah Kandasamy"
    },
    {
      "affiliations": [
        "University of Houston, Houston, TX, USA"
      ],
      "name": "Dilshan Adhikari"
    }
  ],
  "title": "939 Discovery of immunogenic neo-peptides from actionable RNA fusion FGFR3-TACC3 for developing cancer vaccines",
  "uid": "b2a75fe7-a3b6-5990-b02e-dcab18034c3b"
}