Abstract

BackgroundIDH-wildtype glioblastoma (GBM) is the most prevalent primary brain cancer with a 5-year survival rate below 10%. Despite combined treatment through extensive resection and radiochemotherapy, nine out of ten patients develop recurrences. The lack of targeted treatment options and reliable diagnostic markers for recurrent tumors remain major challenges. Methods & AimsIn this study, we present the proteomic characterization of tissue and serum from 55 initial GBM tumors and five matching recurrences, which we investigated for proteomic tumor subtypes and proteomic signatures associated with recurrence. ResultsPrimary tumors revealed four distinct subgroups through hierarchical clustering: a neuronal cluster with elevated mature neuron markers, an innate immunity cluster with increased protease expression, a mixed cluster, and a stem-cell cluster. Neurodevelopmental and inflammatory processes were identified as key factors influencing clustering, with proteolytic activity increasing relative to the degree of inflammation. An analysis comprising proteins with lower coverage confirmed and expanded this pattern. Patients in the neuronal cluster exhibited significantly longer survival compared to those in the stem-cell cluster. In a patient-matched differential expression analysis, five recurrent tumors displayed significantly altered protein expression compared to their primary counterparts, emphasizing the proteomic plasticity of recurrent tumors. Investigation of serum proteomes before and after surgery, using a depletion-based protocol, revealed highly patient-specific and stable proteome compositions, despite a notable increase in inflammation markers post-surgery. However, the levels of circulating proteolytic products matched to the proteolytic activity within the tissue and one fragment of proteolysis activated receptor 2 (PAR2) consistently dropped in abundance after removal of inflamed tumors. ConclusionOverall, we describe a large proteomic GBM cohort. We identified distinct tumor subgroups, molecular patterns of recurrence, and matching proteomic patterns in the bloodstream, which may improve risk prediction for recurrent GBM.