SYST-07 PLASMA MEMBRANE PROFILING REVEALS A TARGETABLE CELL-SURFACE PHENOTYPE OF WNT-MEDULLOBLASTOMA

WNT-medulloblastoma has an excellent prognosis, with an overall survival rate of 90% among children receiving standard-of-care surgical resection, radiotherapy, and chemotherapy. While curative, this treatment is associated with major, long-term, debilitating side effects. Therefore, it is crucial we develop effective, less toxic therapies for these children. The Gilbertson lab has previously shown that WNT-medulloblastomas are devoid of a Blood-Brain-Barrier (BBB). This disruption allows large molecules such as monoclonal antibodies (mAbs) to penetrate WNT-medulloblastomas. To discover WNT-medulloblastoma specific antigens for mAb targeting, we have first generated bulk and single-cell RNA sequencing profiles of the wild-type and WNT-medulloblastoma-prone mossy-fibre progenitor lineages in the lower rhombic lip, as well as human medulloblastoma patient derived orthotopic xenograft (PDOX) tumours. We next generated plasma membrane proteome profiles of hindbrain germinal zones and WNT-medulloblastomas. Plasma Membrane Profiling (PMP) deploys an isobaric chemical labelling approach -Tandem Mass Tags (TMT) - that allows multiplexing of up to sixteen variables per mass spectrometry experiment and ex vivo TMT-labelling of primary cells. Using this approach, we have mapped the cell surface proteome of our medulloblastoma models. Interrogation of these developmental and tumour data has unmasked WNT-medulloblastoma associated antigens that are specific for prenatal brain and WNT-medulloblastoma, but that are not expressed on critical postnatal tissues. Our workflow integrates multi-omic data sets to uncover a targetable landscape on the surface of WNT-medulloblastoma cells that we have cross-validated with a large cohort of patient RNA sequencing profiles and histological samples of both mouse and human tumours. To translate this dataset to real patient benefit, we are generating high affinity antibodies of these validated cell-surface proteins for assessment in our pre-clinical mouse models. Through our comparative analyses we have additionally identified non-WNT-specific targets and thus aim to expand this approach to identifying targets in other more agressive CNS tumours and brain metastases.