Poster Presentation 31st Annual Lorne Proteomics Symposium 2026

Multi-dimensional Immunopeptidomic Analysis of Tumour, PBMC, and Plasma in Gliolastoma (#16)

Erwin Tanuwidjaya 1 2 3 , Gwo-Yaw Ho 2 3 4 , Joshua Ooi 2 3 , Janet Chang 2 3 , Adrian Praeger 4 , Justin Moore 4 , Joel R Steele 1 , Ralf B Schittenhelm 1 , Pouya Faridi 1 2 3
  1. Monash Proteomics & Metabolomics Facility, Department of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria, Australia
  2. School of Clinical Sciences, Monash Health, Clayton, Victoria, Australia
  3. Hudson Institute of Medical Research, Clayton, Victoria, Australia
  4. Monash Medical Centre, Monash Health, Clayton, Victoria, Australia

Introduction:

The human leukocyte antigen (HLA) complexes are pivotal in guiding human adaptive immune responses through their presentation of peptide ligands derived from extracellular and intracellular proteins. Characterizing this peptide repertoire, collectively known as the immunopeptidome, is critical in understanding the biology of adaptive immune defence and is central to the development of personalized cancer immunotherapies. Recent developments in mass spectrometry (MS)-based immunopeptidomics, including our SAPrIm 2.0 workflow have enabled studies on smaller input amounts of cells and tissues, and more recently from biofluids. To investigate whether circulating plasma HLA peptidome could represent tumor-derived HLA peptidome repertoire, here we use  our SAPrIm 2.0 workflow to dissect the similarities and differences in the immunopeptidome of tumor, plasma, and peripheral blood mononuclear cells (PBMC) derived from three glioma patient donors under the Brain on Monash Live Bio-Banking (MoLBi) initiative. 

Method:

Immunopeptidomics profiling was performed on tumor, plasma and PBMC derived from six glioma patient donors (1 glioma and 5 glioblastoma patients). These samples were subjected to the SAPrIm2.0 workflow, leveraging the KingFisher instrument for automated peptide-HLA affinity purification. Liquid chromatography/Mass spectrometry (LC/MS) analyses on the enriched immunopeptides were performed using Orbitrap Exploris 480 mass spectrometer operating on data-independent acquisition (DIA). RAW files were analyzed using PEAKS Studio 13 DeepNovo library-free workflow, using 1% false discovery rate (FDR) at peptide level. Peptide identification and bioinformatics analysis was conducted using R, and peptide binding prediction analysis was done on NetMHCpan-4.2.

Result:

Here, we highlight the findings of the Brain on MoLBi initiative on 6 glioma patients. More than 40,000 unique HLA-I immunopeptides are identified from the 6 donors, which are predicted to bind to the donor alleles. A high proportion (~50%) of patient-derived peptidomes are shared between the tumor and the circulating compartments. Additionally, Pearson pairwise analyses on these shared peptides reveal moderate correlation (0.2 < R < 0.7) observed across tissue compartments, suggesting that plasma soluble HLA peptidome provides a good representation of tumor HLA repertoire. Moreover, more than 100 cancer-associated and cancer testis antigenic source genes were identified from these samples, ranging from known cancer drivers such as TP53, SRC as well as multiple MAGE family antigens, underscoring the breadth of tumor-associated antigenic diversity recoverable through the use of  immunopeptidomics profiling. Together, these findings highlight the potential of immunopeptidomics to uncover clinically relevant antigenic biomarkers and therapeutic targets, particularly from minimally invasive circulating sHLA sources.