Integrating Rapid Evaporative Ionization Mass Spectrometry Classification with Matrix-Assisted Laser Desorption Ionization Mass Spectrometry Imaging and Liquid Chromatography-Tandem Mass Spectrometry to Unveil Glioblastoma Overall Survival Prediction.

Abstract

Glioblastoma multiforme (GBM) is a highly aggressive brain cancer with a median survival of 15 months. Despite advancements in conventional treatment approaches such as surgery and chemotherapy, the prognosis remains poor. This study investigates the use of rapid evaporative ionization mass spectrometry (REIMS) for real-time overall survival time classification of GBM samples and uses matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI) to compare lipidomic differences within GBM tumors. A total of 45 GBM biopsies were analyzed to develop a survival prediction model for IDH-wild type GBM. REIMS patterns from 28 patients were classified with a 97.7% correct classification rate, identifying key discriminators between short-term (0-12 months) and prolonged (>12 months) survivors. Cross-validation with additional samples showed that the model correctly classified short-term and prolonged survival with 66.7 and 69.4% accuracy, respectively. MALDI-MSI was performed to confirm the discriminators derived from REIMS data. Results indicated 42 and 33 discriminating features for short-term and prolonged survival, respectively. Proteomic profiling was performed by isolating tumor regions via laser-capture microdissection (LMD) and liquid chromatography-tandem mass spectrometry (LC-MS/MS). Subsequently, 1387 proteins were identified, of which 79 were significantly altered. In conclusion, this study shows that REIMS rapidly predicts glioblastoma survival times based on lipidomic profiles during electrosurgical dissection. MALDI-MSI confirmed that these differences were specific to the tumor region in the glioblastoma sections. LMD-guided LC-MS/MS-based proteomics revealed significantly altered pathways between short-term and prolonged survival. This research, including the comprehensive predictive survival model for GBM, could guide tumor resection surgeries based on accurate real-time tumor tissue identification as well as provide insights into overall survival mechanisms, possibly related to therapy response.