Rigorous Comparison of Extracellular Vesicle Processing to Enhance Downstream Analysis for Glioblastoma Characterization

Extracellular vesicles (EVs) are highly sought after as a source of biomarkers for disease detection and monitoring. Tumor EV isolation, processing, and evaluation from biofluids is convoluted by EV heterogeneity and biological contaminants and is limited by technical processing efficacy. This study rigorously compares common bulk EV isolation workflows (size exclusion chromatography, SEC; membrane affinity, MA) alongside downstream RNA extraction protocols to investigate molecular analyte recovery. EV integrity and recovery is evaluated using a variety of technologies to quantify total intact EVs, total and surface proteins, and RNA purity and recovery. A comprehensive evaluation of each analyte is performed, with a specific emphasis on maintaining user (n = 2), biological (n = 3), and technical replicates (n & GE;3) under in vitro conditions. Subsequent study of tumor EV spike-in into healthy donor plasma samples is performed to further validate biofluid-derived EV purity and isolation for clinical application. Results show that EV surface integrity is considerably preserved in eluates from SEC-derived EVs, but RNA recovery and purity, as well as bulk protein isolation, is significantly improved in MA-isolated EVs. This study concludes that EV isolation and RNA extraction pipelines govern recovered analyte integrity, necessitating careful selection of processing modality to enhance recovery of the analyte of interest. This work provides rigorous comparison of extracellular vesicle (EV) isolation methods, via size exclusion chromatography and membrane affinity, and RNA extraction protocols. The isolated analytes (intact EVs, EV RNA, and proteins) are assessed using a variety of technologies. The nature of each processing modality dictates the quantity and integrity of the final product, underscoring the importance of process selection for enhanced analyte detection in liquid biopsy applications.image