Isolation of extracellular vesicle with different precipitation-based methods exerts a tremendous impact on the biomarker analysis for clinical plasma samples

Abstract

Extracellular vesicles(EVs) is an emerging approach of cancer liquid biopsy. Although the precipitation-based method with commercial kits has gained popularity as the second most commonly used technique, these protocols vary tremendously with many included reagents still unknown to the community. In this study, we assigned each of the 3 clinical plasma samples into 6 aliquots to assess five commercial EV isolation kits, in comparison with ultracentrifugation(UC). We implemented a standardized EV preparation and transcriptome analysis workflow except the EV isolation methods used. The metrics of EVs and its RNA cargo (evRNA) were compared to assess the technical variations versus the biological variations in the clinical setting. Although the size range of the isolated EVs demonstrated a similar distribution, we found significant technical variability among these methods, in terms of EV amount, purity, subpopulations and RNA integrity. Such variabilities were further relayed to a drastic divergence of evRNA expression on a transcriptome-wide fashion. Our study demonstrated a highly variable result from polymeric precipitation-based EV isolation methods, making EVs based biomarker analysis difficult to interpret and reproduce. We highlighted the importance of benchmarking and transparent reporting of the precipitation-based protocols in the liquid biopsy research.